Synthesis and applications of difluorobenzothiadiazole based conjugated polymers for organic photovoltaics

Article abstract of DOI:10.1039/c0jm04166a

A new electron deficient building block, difluorobenzothiadiazole (FBT) was designed and synthesized for the first time. Two low band gap polymers containing this unit or benzothiadiazole (BT) and an electron rich unit, benzo[1,2-b:4,5-b′]dithiophene (BDT

Full text of DOI:10.1039/c0jm04166a

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2011, 21, 3226
www.rsc.org/materials
PAPER
Synthesis and applications of difluorobenzothiadiazole based conjugated
polymers for organic photovoltaics†
a
Zhao Li, Jianping Lu, Shing-Chi Tse, Jiayun Zhou, Xiaomei Du, Ye Tao and Jianfu Ding
b
b
b
a
b
a
*
*
Received 30th November 2010, Accepted 7th January 2011
DOI: 10.1039/c0jm04166a
A new electron deficient building block, difluorobenzothiadiazole (FBT) was designed and synthesized
for the first time. Two low band gap polymers containing this unit or benzothiadiazole (BT) and an
electron rich unit, benzo[1,2-b:4,5-b0]dithiophene (BDT) were synthesized by Stille coupling reaction. It
was found that fluorine substitution significantly lowered the energy levels but did not affect the UV
absorption properties of the polymer. Comparatively, the fluorinated polymer showed a higher melting
point and a reduced solubility. The electrical and photovoltaic properties of these polymers were
studied by fabricating thin film transistors and polymer solar cells. The solar cell based on FBT polymer
and PC71BM gives power conversion efficiency up to 3.4% under AM 1.5 G illumination (100 mW
cmꢀ2).
with fullerene to facilitate efficient exciton separation and charge
transfer.
1. Introduction
Compared with inorganic silicon based solar cells, organic solar
cells show huge commercial potentials for flexible and large area
applications, because the devices could be manufactured by
existing low-cost industrial technology, such as ink-jet and roll to
roll printing. Another important advantage of organic solar cells
is the diversity of the molecular structures in organic chemistry,
where numerous combinations of various building blocks are
available. So the properties of the resulting materials, such as
solubility, charge mobility and energy levels for photovoltaic
applications can be easily tuned.1
To satisfy these requirements, people usually build alternating
copolymers comprised of electron rich and electron deficient
units. On one hand, such a structure can induce intra-chain
charge transfer and dramatically lower the band gap of resulting
polymers. So the polymers often show strong absorption in the
visible region. On the other hand, their highest occupied
molecular orbital (HOMO) and lowest unoccupied molecular
orbital (LUMO) energy levels can be easily adjusted by choosing
different combinations of electron rich and deficient units.
Among the vast variety of the developed electron deficient
blocks, 2,1,3-benzothiadiazole (BT) has attracted much research
attentions, and has been copolymerized with many electron rich
units, such as fluorene,3 4H-cyclopenta[2,1-b;3,4-b0]dithiophene,4
carbazole,5 benzo[1,2-b:4,5-b0]dithiophene (BDT),6 dithieno[3,2-
b:20,30-d]pyrrole,7 dithienosilole,8 indolo[3,2-b]carbazole,9 and
indenofluorene etc.10 High performance solar cell devices with
power conversion efficiency (PCE) over 5% have been reported
by several groups based on these materials.4,8,11 Furthermore,
the 5 and 6 positions of BT are available to attach with
a functional group to modify properties of the final polymers.
For example, Bo etc. introduced two alkyloxy chains there to
improve the solubility of the resulting polymers.12 Yamashita
etc. fused other electron deficient rings to these positions to
obtain [1,2,5]thiadiazolo[3,4-g]quinoxaline and benzo[1,2-c;3,4-
c0]bis[1,2,5]thiadiazole,13 which demonstrated very strong elec-
tron withdrawing capability and resulted in ultra-small band
gap polymers.14
Up to now, the most efficient and commonly used architecture
for the active layer of organic solar cells is the bulk hetero-
junction (BHJ) structure of two components:2 a conducting
polymer as an electron donor and a fullerene derivative as an
electron acceptor. For the conducting polymers, several crucial
properties are required to achieve a highly efficient solar cell.1
First, it should have a good solubility and can be easily fabricated
into a thin film from solution. Second, the polymer should show
high absorption toward sun light since fullerene derivatives only
has very limited absorption in the solar spectrum. Third, the
polymer should have a good hole mobility, at least comparable
to the electron mobility of fullerene derivatives, so the separated
charge carrier can be easily transported to the corresponding
electrodes. Forth, the energy levels of this polymer should match
aInstitute for Chemical Process and Environmental Technology (ICPET),
National Research Council of Canada (NRC), 1200 Montreal Road,
Ottawa, ON, Canada K1A 0R6. E-mail: zhao.li@nrc-cnrc.gc.ca; jianfu.
ding@nrc-cnrc.gc.ca
bInstitute for Microstructural Sciences (IMS), National Research Council
of Canada (NRC), 1200 Montreal Road, Ottawa, ON, Canada K1A 0R6
† NRC Publication Number: NRCC 52825.
In this work, we attempted to introduce two fluorine atoms
onto these two positions to produce a new electron deficient unit,
4,7-di(thien-2-yl)-5,6-difluoro-2,1,3-benzothiadiazole
(FBT).
This unit was then copolymerized with an electron rich BDT unit
3226 | J. Mater. Chem., 2011, 21, 3226–3233
This journal is ª The Royal Society of Chemistry 2011

View Online
to afford
a
new copolymer, poly(4,8-di(3-pentyl-unde-
thionylaniline and trimethylsilyl chloride in pyridine at 80 C.19
ꢁ
cyl)benzo[1,2-b:4,5-b0]dithiophene-alt-4,7-bis(4-octyl-2-thienyl)-
5,6-difluoro-2,1,3-benzothiadiazole) (P2). For comparison, its
non-fluorinated counterpart, poly(4,8-di(3-pentyl-undecyl)benzo-
[1,2-b:4,5-b0]dithiophene-alt-4,7-bis(4-octyl-2-thienyl)-2,1,3-
benzothiadiazole) (P1) was also prepared. Their optical,
electrochemical, thermal properties, and their devices perfor-
mance in thin film transistors (TFT) and BHJ solar cells have
been characterized. We demonstrated that the fluorine substitu-
tion did not change the UV absorption but pushed down the
HOMO energy level, thus improved the environmental stability
of corresponding polymer. The organic solar cells based on FBT
polymer also showed a higher power conversion efficiency (PCE)
than BT polymer.
Dibromo-monomer 5 was obtained by bromination of 4 with
NBS in chloroform using silica gel as catalyst. We introduced
octyl side chains on each thiophene unit to improve the solubility
of the resulting polymers.9,20
Recently, benzo[1,2-b:4,5-b0]dithiophene (BDT) based poly-
mers have shown very promising properties in organic electronic
devices, especially in polymer solar cells.17,21 During the prepa-
ration of this manuscript, You etc. reported a BDT based
polymer with branched alkyl groups at the 4- and 8- position of
BDT.22 However, detailed synthetic procedure, especially the
synthesis of the key starting compound (branched alkyl chain
substituted acetylene) was not reported. We independently
synthesized a similar BDT monomer and the procedure is shown
in Scheme 2. Compound 6 was synthesized by a chain extension
reaction of 1-octyne.23 It is interesting that two equivalent of n-
butyl lithium will react with 1-octyne to form a 1,3-dilithiooctyne
intermediate, which will then react with 1-bromooctane on the
more reactive 3-position to give compound 6 after acid treatment
and purification. Alkynylmagnesium chloride was generated
in situ from 6 by reacting with isopropylmagnesium chlorideꢁ. It
reacted with benzo[1,2-b:4,5-b0]dithiophene-4,8-dione at 55 C.
After reduction by SnCl2 in hydrochloride solution, dialkynyl
substituted intermediate 7 was obtained in 66% yield. Hydroge-
nation converted the alkynyl side chain to alkyl and trimethyl-
stannyl monomer 9 was obtained by reacting 8 with n-butyl
lithium followed with trimethyltin chloride.21 It is worth to note
that this monomer has a very similar structure as another BDT
monomer reported by Yu0s group, where a branched alkoxy
group are used instead.17 It has been reported that the replace-
ment of alkoxy with alkyl group will reduce the electron donating
effect and lower the HOMO energy level of the resulted
polymers.21
2. Results and discussion
2.1. Material design and synthesis
Fluorinated conjugated polymers show very interesting
features.15 Fluorine atoms can lower both the HOMO and
LUMO energy levels, so the materials will display a greater
resistance against the degradative oxidation. A lower HOMO
energy level also means a higher open circuit voltage (Voc) for
polymer solar cells since Voc is usually determined by the
difference between the HOMO of the polymer and the LUMO of
the fullerene derivative.16 These features are highly desired in
polymer materials for photovoltaic applications. Indeed, high
Voc and very high PCE were obtained by introducing fluorine
into organic photovoltaic polymers very recently.17
Based on the well studied BT structure, we designed and
synthesized fluorinated BT as shown in Scheme 1. This synthesis
started from nitration of 1,4-dibromo-2,3-difluorobenzene,
which was prepared using a reported two-step reaction in a high
yield (90%).18 The crude product can be purified by column
chromatograph to give a pure product 1 in a moderate yield
(31%), which was then reacted with 2-tributyl stannyl-4-octyl
thiophene to give compound 2 by palladium catalyzed Stille
coupling reaction. Two nitro groups on 2 were reduced to
diamine 3 using iron powder in acetic acid solution. Heterocycle
4 was prepared in 94% yield by reacting diamine 3 with N-
Polymers P1 and P2 were synthesized via Stille coupling
reaction as shown in Scheme 3, where BT or FBT were used as
electron deficient units and BDT was used as electron rich unit.
Both polymers were carefully purified by extracting with acetone
Scheme 1 Synthesis of FBT monomer.
Scheme 2 Synthesis of BDT monomer.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 3226–3233 | 3227

View Online
Fig. 1 UV absorption spectra of P1 and P2 in dilute chlorobenzene
solution (ꢂ1.0 ꢃ 10ꢀ5 Mꢀ1cmꢀ1) at 50 ꢁC or as a film at RT.
Scheme 3 Preparation of conjugated polymers P1 and P2.
can substantially lower the HOMO energy level of the polymer
by 0.18 eV. This fluorine effect was less significant in another
fluorinated polymer where only one fluorine atom was attached
to each repeating unit of the polymer backbone.17 Since the first
reduction wave of the polymers was hard to determine, the
LUMO energy levels were then estimated from the optical band
gap and the relative data were summarized in Table 1. It is
interesting that the fluorine substitution does not apparently
change the UV absorption and band gap, but lowers HOMO
energy level. This will lead to an improved Voc for the polymer
solar cell devices and better environmental stability of the
material.
and then hexanes to remove oligmers and other small molecular
part, and GPC analysis showed high molecular weight polymers
were obtained (Table 1). P1 can dissolve in chlorobenzene and
o-dichlorobenzene at room temperature while P2 can only
dissolve in these solvents at elevated temperature.
2.2. Optical and electrochemical properties
The normalized UV-visible absorption spectra of the chloro-
benzene solution and the thin film of polymers P1 and P2 are
shown in Fig. 1. Both polymers show very similar absorption
bands in chlorobenzene solution with a major absorption peak at
558 nm for P1 and 554 nm for P2. This band can be assigned to
the intra molecular charge transfer interaction between electron
rich and deficient units. The other band in short-wavelength near
400 nm can be assigned to 4,7-dithien-2-yl-2,1,3-benzothiadia-
zole (DTBT) units, which was often observed in DTBT
containing conjugated polymers.3,5 In the solid state, the main
absorption peak becomes broader and the maximum shifts
toward longer wavelength at 615 nm for P1 and 622 nm for P2.
This large red shift of ꢂ60 nm from solution to solid state means
more coplanar structure and stronger inter chain p–p stacking in
the solid state. It is interesting that a shoulder peak appeared
near 680 nm for P1 but not for P2.
2.3. Thermal properties and X-ray diffraction (XRD)
The thermal properties of these polymers were studied by
differential scanning colorimetry (DSC) and thermal gravimetric
analysis (TGA) (Fig. 3). Both P1 and P2 showed a broad and
strong melting peak at 247 and 284 ꢁC respectively, but no
discernible glass transition. A corresponding crystallization peak
was also observed for both polymers in the cooling scan with the
crystallizing temperature very close to the melting temperature,
indicating both polymers have a strong crystallization capability.
The thermal decomposition temperatures (2% weight loss) of P1
and P2 are 403 ꢁC and 408 ꢁC respectively, showing good
thermal stability for optoelectronic devices.
The HOMO and LUMO energy levels of these two polymers
were investigated by cyclic voltammetry (Fig. 2). The onset
potentials of the first oxidation wave of P1 and P2 appeared at
0.92 and 1.10 V, corresponding to HOMO energy levels of 5.30
and 5.48 eV respectively. As expected, this indicated that the
introduction of two fluorine atoms on the conjugated backbone
To characterize the crystal structures of these polymers in thin
film, XRD studies were carried out and the curves are shown in
Fig. 4. The results indicate a lamellar crystal structure for both
polymers.24 A broad peak appeared at 2q of 6.95ꢁ and 6.53ꢁ for
P1 and P2, corresponding to an inter-layer spacing of 14.8 and
ꢀ
15.7 A respectively. P1 showed one broad p–p stacking peak at
Table 1 Polymer characterization
PDIa
Soln lmaxb abs (nm)
Film lmax abs (nm)
Egoptc (eV)
Eox d(V)
onset
LUMOe(eV)
HOMOf(eV)
a
Mn (kDa)
P1
P2
16.2
27.8
1.58
2.57
399, 558
400, 554
424, 615
424, 622
1.60
1.56
0.92
1.10
3.70
3.92
5.30
5.48
a
b
Number average molecular weight and polydispersity index (GPC vs. polystyrene standards in chlorobenzene). Solution absorption in
c
chlorobenzene. Optical energy gap estimated from the onset of UV curve measured in thin film. CV measurements of thin films in 0.1 M
d
Bu4NPF6/CH3CN solution vs. Ag. e Estimated from LUMO ¼ HOMO ꢀ Eg
.
f
opt
on
Estimated from HOMO ¼ Eox + 4.38 eV.
3228 | J. Mater. Chem., 2011, 21, 3226–3233
This journal is ª The Royal Society of Chemistry 2011

View Online
of these polymers. Fig. 5 shows representative I–V characteristics
of these devices fabricated from P1 and P2. Both polymers
displayed a well-defined saturation region with p-channel
transport (h+) mobility of 2.12 ꢃ 10ꢀ4 cm2/Vs for P1 and 4.88 ꢃ
10ꢀ5 cm2/Vs for P2 respectively. To our surprise, P2 has a lower
hole mobility, although it has a higher crystallization capability
than P1. We do not have a clear explanation for this phenom-
enon. It might be related to the orientation of the crystal struc-
ture relative to the substrate. In addition, the relative poor film
quality of polymer P2 due to its low solubility may also influence
the charge transfer. This issue could be improved by adjusting
the processing conditions such as the solvent, temperature, spin
speed, annealing and so on.
Polymer solar cells were fabricated from P1 or P2 and (6,6)-
phenyl-C71-butyric acid methyl ester (PC71BM) with a general
device structure of ITO/PEDOT:PSS/Polymer: PC71BM/LiF/Al.
Different polymer to fullerene weight ratio, solvents and
processing temperature are tested to optimize the device perfor-
mance. Fig. 6a showed typical current density-voltage (J-V)
curves and the devices performance data were summarized in
Table 2. Device fabricated from a blend of P1 with PC71BM at
1/2 weight ratio gave a Voc of 0.69 V, a short-circuit current
density (Jsc) of 6.15 mA cmꢀ2, a modest fill factor (FF) of 44.3%
and a PCE of 1.88%. The device from P2 with PC71BM at 1/1
weight ratio gives a better result with a similar Voc of 0.69 V,
a higher Jsc of 8.89 mA cmꢀ2 and an improved FF of 55.4%. The
PCE thus reached 3.4%. It should be noted that the best device
for P1 is fabricated from o-dichlorobenzene solution with 2.5%
DIO as an additive at 70 ꢁC while the blend of P2 and PC71BM
has to be spin-coated from 1,2,4-trichlorobenzene solution at 90
ꢁC. It is important that different solvents and processing
temperature should be used to get optimized performance. We
noted that the devices from P1 and P2 show similar Voc although
there is a big difference between their HOMO energy levels
(0.18 eV).16 This is tentatively explained by the decreased surface
energy of fluorinated polymer. The lower surface energy of the
fluorinated polymer enhanced the tendency of the polymer to
Fig. 2 Cyclic voltammogram of P1 and P2 films coated on a platinum
electrode in 0.1 M Bu4NPF6/CH3CN solution at RT.
Fig. 3 (a) DSC and (b) TGA curve of P1 and P2 with a heating rate of 10
ꢁC minꢀ1 in nitrogen.
Fig. 4 XRD spectra of P1 and P2 thin films.
ꢁ
ꢁ
ꢀ
2q of 25.2 (4.1 A), while P2 showed a broad ꢁpeak at 23.9 (4.32
ꢀ
A) and a more strong and narrow peak at 28.3 , corresponding to
ꢀ
a very small p–p stacking distance of 3.66 A. We propose that
there is stronger interchain interaction in the fluorinated polymer
P2 than in P1, as P2 shows poorer solubility, higher melting
point and closer stacking distance.
2.4. Field effect transistor characterization and photovoltaic
performance
As the performance of organic optoelectronic materials for
transistor or photovoltaic applications is closely related to their
charge carrier mobility, we fabricated bottom-contact organic
thin-film transistors (OTFT) to measure the field-effect mobility
Fig. 5 Output characteristics of p-type OFET devices of P1 (a, b) and P2
(c, d) tested in glove box.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 3226–3233 | 3229

View Online
Fig. 6 (a) J-V curve of P1:PC71BM (weight ratio 1 : 2) and P2:PC71BM
(weight ratio 1 : 1) BHJ solar cells under illumination of AM 1.5G, 100
mW cmꢀ2, (b) EQE spectra of these BHJ solar cells.
migrate to the air/film interface during spin coating,25 resulting in
the relative enrichment of the polymer at the top of the active
layer. This kind of structure is not beneficial for charge transport
to corresponding electrodes and could lower the Voc of the
devices because of the accumulation of space charge. However,
this fluorinated polymer may be used in an inverted structure
where high concentration of polymer on top layer will benefit.25
As shown in Fig. 6b, external quantum efficiency (EQE) curves
of these devices were tested which cover 300 nm to 700 nm range.
Devices of P2 showed a higher efficiency (ꢂ45%) than P1
(ꢂ30%), and the integrated Jsc from EQE data agrees well with
the data from J-V curves (see Table 2).
Fig. 7 AFM topography (a, b) and phase (c, d) images of P1:PC71BM
(1 : 2, w/w) (a, c) and P2:PC71BM (1 : 1, w/w) (b, d) films. The scan size of
the image is 1mm ꢃ 1mm.
spectrum. The absorption spectra of these thin films were
monitored over time while keeping the films under ambient
condition. It was reported that oxygen and water can diffuse into
the active layer of a polymer solar cell and cause photo-oxidative
bleaching under ambient condition and continuous illumina-
tion.27 The degree of this oxidative reaction for the donor poly-
mer is closely related to the position of its HOMO energy level.
Fig. 8a showed that the maximum absorption of P1 film
decreased by 27% and blue shift from 615 nm to 586 nm after
exposure to air and ambient light for 312 h, meaning dramatic
decreasing of the number of original polymer chains and short-
ening of the conjugation length. However, after the same expo-
sure, the maximum absorption peak of polymer P2 only
decreased by 17% and the position retained at 620 nm. This result
demonstrates the improved stability of this series of polymer by
fluorine substitution and agrees well with the lowered HOMO
energy level from CV data.
2.5. Atomic force microscopy (AFM) images and stability
characterization
We then studied the surface morphological structure of the
polymer/PC71BM blend film using tapping mode AFM and the
images are shown in Fig. 7. The phase image of the film from P1
with PC71BM (Fig. 7c) showed obscure domains without clear
phase segregation, this may explain the modest FF of 44.3%
obtained from this device. However, the film from P2 and
PC71BM has a more clear structure which shows bright domains
with the size of ꢂ20 nm. This domain size is desired for better
charge separation, and thus agrees with the improved FF of 55%.
It should be noted that the phase image scale for P2 (0–100ꢁ) is
much larger than the one for P1 (0–30ꢁ). The bright spots in
Fig. 7d can be assigned to the crystalline domains of fluorinated
polymer P2 due to decreased surface energy,26 and the detection
of polymer P2 on the top layer of the film also partially support
our previous assumption on the accumulation of the fluorinated
polymer on air/film interface.
3. Conclusions
A new fluorinated benzothiadiazole monomer, FBT is designed
and synthesized. Two alternating conjugated polymers P1 and
P2 are prepared based on electron deficient BT or FBT and
electron rich BDT monomers. The optical, electrochemical and
thermal studies of these two polymers show that the fluorine
substitution will not change the UV absorption, but will signifi-
cantly lower the HOMO energy level. However, this does not
result in a higher Voc in the polymer solar cell devices. This
phenomenon is believed to be associated with the enrichment of
P2 on the top surface of the active layer due to the low surface
energy of the polymer as observed by AFM. TFT characteriza-
tion shows P2 has lower hole mobility, but this drawback may be
compensated by the better domain structure formed in the device
with enhanced charge separation and transfer, thus better FF
and higher PCE are observed from the P2 devices. Finally, the
improved environmental stability of the fluorinated polymer is
Polymer thin films (ꢂ40 nm) were prepared by spin coating o-
dichlorobenzene solution on glass substrate and the stability of
the polymer thin films in air was investigated by UV-vis
Table 2 Summary of device performance
Ratioa Solvent Jsc (mA cmꢀ2
)
Voc (V) FF (%) PCEd (%)
d
P1 1 : 2
P2 1 : 1
DCBb
TCBc
6.15 (5.89)
8.89 (8.83)
0.69
0.69
44.3
55.4
1.88 (1.80)
3.40 (3.38)
a
b
Polymer/PC71BM weight raio. Containing 1.8-diiodooctane (2.5%
c
d
volume ratio) as additives. TCB ¼ 1,2,4-trichlorobenzene. Data
calculated from EQE spectrum are shown in bracket.
3230 | J. Mater. Chem., 2011, 21, 3226–3233
This journal is ª The Royal Society of Chemistry 2011

View Online
anhydrous CH3CN as the supporting electrolyte. The polymers
were coated on the platinum-working electrode. The CV curves
were recorded referenced to an Ag quasi-reference electrode,
which was calibrated using a ferrocene/ferrocenium (Fc/Fc+)
redox couple (4.8 eV below the vacuum level) as an external
standard. The E1/2 of the Fc/Fc+ redox couple was found to be
0.42 V vs. the Ag quasi-reference electrode. Therefore, the
HOMO energy level of the copolymers can be estimated using the
empirical equation EHOMO ¼ Eoxon + 4.38 eV, where Eoxon stands
for the onset potentials for oxidation relative to the Ag quasi-
reference electrode. X-ray diffraction (XRD) spectrum was
obtained with a Bruker AXS D8 Advance instrument with Co
1
ꢀ
Ka radiation (l ¼ 1.789 A).
4.2. Synthesis
Materials. 1,4-dibromo-2,3-difluorobenzene,18 4,7-bis(5-bromo-
4-dioctyl-2-thienyl)-2,1,3-benzothiadiazole,9 and benzo [1,2-
b:4,5-b0] dithiophene-4,8-dione28 were synthesized according to
literature.
1,4-Dibromo-2,3-difluoro-5,6-dinitrobenzene (1). Fuming nitric
acid (100 mL, 90%) was added slowly to 100 mL of concentrated
sulfuric acid solution at 5 ꢁC in an ice water bath. Then 1,4-
dibromo-2,3-difluorobenzene (20 g, 74 mmol) was added drop
wise. After stirring at room temperature (RT) for 4 h, the solu-
tion was stirred at 67 ꢁC for 14 h. The resulting mixture was
poured into 1400 mL of ice water and the formed yellow
precipitate was collected by filtration. Then the crude product
was purified by column chromatography on silica gel with
hexanes/dichloromethane (4/1 to 3/1 volume ratio). The third
fraction provided 8.3 g white solid (yield 31%) as product, 19F
NMR (CDCl3) d: ꢀ114.3 (s, 2F); 13C NMR (CDCl3) d: 151.2 (d),
148.6 (d), 141.4, 105.7. Mp 121.2 ꢁC. HRMS EI calcd for
C6Br2F2N2O4 359.81929, found 359.81862.
Fig. 8 Change of the UV absorption spectra of thin films of P1 (a) and
P2 (b) with time in ambient conditions.
confirmed by the UV study of polymer thin films which are aged
in air under ambient light.
4. Experimental
4.1. General characterization methods
NMR spectra were recorded in CDCl3 or o-dichlorobenzene-d4
using a Varian Unity Inova spectrometer at a resonance
frequency of 399.96 MHz for 1H, 376.29 MHz for 19F, and 100.58
1
MHz for 13C. H and 19F NMR spectra were obtained using a 5
1,4-Bis(4-octyl-2-thienyl)-2,3-difluoro-5,6-dinitrobenzene (2).
mm indirect detection probe. A 5 mm broadband probe was used
for acquiring 13C NMR spectra. CFCl3 was used as an internal
standard (0 ppm) for the 19F NMR measurements. Gel perme-
ation chromatography (GPC) (Waters Breeze HPLC system with
1525 Binary HPLC Pump and 2414 Differential Refractometer)
was used for measuring the molecular weight and polydispersity
index at 40 ꢁC. Chlorobenzene was used as eluent and
commercial polystyrenes were used as standard. UV-vis spectra
were measured using a Varian Cary 5000 Spectrometer. The
differential scanning calorimetry (DSC) analysis was performed
under a nitrogen atmosphere (50 mL minꢀ1) using a TA Instru-
ments DSC 2920 at a heating rate of 10 ꢁC minꢀ1, calibrated with
the melting transition of indium. The thermal gravimetric anal-
ysis (TGA) was performed using a TA Instruments TGA 2950 at
Compound 1 (0.362 g, 1.00 mmol), 2-tributylstannyl-4-octylth-
iophene
(0.970
g,
2.20
mmol)
and
dichlorobis-
(triphenylphosphine) palladium (0.028 g, 0.040 mmol) were put
in a flask, which was purged with argon under vacuum for three
times before 12 mL of dry THF was added. The solution was
refluxed under argon for 54 h. Then the solvent was removed and
the residue was purified by column chromatograph with hexanes/
dichloromethane (4/1 volume ratio) to give 0.35 g yellow powder
(Yield: 60%). 1H NMR (CDCl3) d: 7.20 (d, 2H), 7.03 (d, 2H), 2.62
(t, 4H), 1.61 (m, 4H), 1.28 (t, 20H), 0.88 (t, 6H); 19F NMR
(CDCl3) d: ꢀ126.6; 13C NMR (CDCl3) d: 150.2 (d), 147.6 (d),
144.3, 139.3, 131.8, 125.1, 124.7, 119.8, 31.8, 30.3, 30.2, 29.4,
29.2, 29.1, 22.7, 14.1. MS (m/z): observed: 593.2, calculated:
593.2 for [M ꢀ H]+.
a heating rate of 10 C minꢀ1 under a nitrogen atmosphere (60
ꢁ
mL minꢀ1). MS data were obtained using a Prince capillary
electrophoresis system coupled to an API3000 mass spectrometer
via a microspray interface. A sheath solution of 1 mL minꢀ1
2-propanol/methanol (2 : 1) was used, with 30 mM ammonium
acetate dichloromethane–methanol (3 : 1) as the running buffer.
Cyclic voltammetry (CV) measurements were carried out under
argon in a three-electrode cell using 0.1 M Bu4NPF6 in
1,4-Bis(4-octyl-2-thienyl)-2,3-difluoro-5,6-diaminobenzene (3).
To a mixture of compound 2 (0.35 g, 0.59 mmol) and iron
powder (0.40 g, 7.1 mmol) was added with acetic acid (12 mL).
The mixture was stirred at 45 ꢁC for 4 h before poured into 25 mL
of cold 5% NaOH solution. The solution was extracted with
ether and the organic phase was washed with 5% NaHCO3
solution for 5 times. After dried over MgSO4, the solvent was
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 3226–3233 | 3231

View Online
removed and the product was purified by column chromato-
graph with 2/3 hexanes/dichloromethane to give pale solid (yield:
73%). 1H NMR (CDCl3) d: 7.07 (d, 2H), 6.98 (d, 2H), 3.62
give white powder (0.94 g, yield: 66%). 1H NMR (CDCl3) d: 7.55
(d, 2H), 7.48 (d, 2H), 2.74 (m, 2H), 1.2–1.8 (m, 44H), 0.9
(m, 12H).
(s, 4H), 2.64 (t, 4H), 1.65 (m, 4H), 1.32 (m, 20H), 0.88 (t, 6H); 19
F
NMR (CDCl3) d: ꢀ150.3; 13C NMR (CDCl3) d: 143.8, 143.4 (d),
141.0 (d), 131.1, 130.3, 129.3, 121.8, 111.2, 31.9, 30.5, 30.4, 29.4,
29.3, 29.2, 22.7, 14.1. MS (m/z): observed: 532.9, calculated:
533.2 for [M ꢀ H]+.
4,8-Di(3-pentylundecyl)benzo[1,2-b:4,5-b0]dithiophene
(8).
Compound 7 (2.0 g, 3.17 mmol) and Pd/C (0.44 g) was added into
a 250 mL flask. After purge with argon, dry THF (80 mL) was
added. The mixture was purged with H2 at a speed of 5 mL minꢀ1
for 16 h. The product was purified by column chromatograph
with 2% dichloromethane in hexanes to give white solid product
(1.62 g, yield: 80%). 1H NMR (CDCl3) d: 7.44 (s, 4H), 3.12
(m, 4H), 1.73 (m, 4H), 1.5 (m, 2H), 1.2–1.4 (m, 44H), 0.9
(m, 12H).
5,6-Difluoro-4,7-bis(4-octyl-2-thienyl)-2,1,3-benzothiadiazole
(4). To a solution of compound 3 (2.13 g, 4.0 mmol) in dry
pyridine (40 mL), N-thionylaniline (0.90 g, 8.0 mmol) and tri-
methylsilyl chloride (0.91 g, 7.2 mmol) were added under argon.
Then the solution was stirred at 80 ꢁC for 16 h before was poured
into ice water, the resulted solid was filtered and washed with
ethanol to give 2.1 g yellow solid (yield: 94%). 1H NMR (CDCl3)
d: 8.11 (d, 2H), 7.20 (d, 2H), 2.71 (t, 4H), 1.68 (m, 4H), 1.32
(m, 20H), 0.88 (t, 6H); 19F NMR (CDCl3) d: ꢀ128.7; 13C NMR
(CDCl3) d: 151.0 (d), 148.9, 148.5 (d), 132.2, 131.2, 123.9, 111.7,
31.9, 30.6, 30.5, 29.4, 29.3, 29.2, 22.7, 14.1. Mp 94.4 ꢁC. MS
(m/z): observed: 560.9, calculated: 561.2 for [M ꢀ H]+.
1,5-Bis(trimethylstannyl)-4,8-di(3-pentylundecyl)benzo-[1,2-
b:4,5-b0]dithiophene (9). Compound 8 (3.1 g, 4.8 mmol) was
dissolved in dry THF (45 mL) in an argon purged flask. The
solution was cooled to ꢀ78 ꢁC before n-BuLi (2.5 M in THF, 4.7
mL, 11.7 mmol) was added dropwise. After stirring at ꢀ78 ꢁC for
30 min, the temperature of the solution wasꢁraised to RT for 30
min. Then the solution was cooled to ꢀ78 C again and trime-
thyltin chloride (1 M in hexane, 13.6 mL, 13.6 mmol) was added.
The solution was allowed to warm to RT overnight before added
with hexane and washed with water. The product was purified by
recrystallization in 2-propanol to get white needles (3.3 g, yield:
71%). 1H NMR (CDCl3) d: 7.48 (s, 2H), 3.14 (m, 4H), 1.75
(m, 4H), 1.50 (m, 2H), 1.2–1.4 (m, 44H), 0.88 (m, 12H). 13C NMR
(Acetone-d6) d: 206.0, 142.3, 141.5, 137.8, 130.5, 128.3, 38.3, 34.4,
33.6, 33.1, 32.7, 30.9, 30.8, 30.6, 30.2, 27.6, 27.3, 23.5, 23.4, 14.5,
14.4, ꢀ8.3. Mp 42.5 ꢁC.
4,7-Bis(5-bromo-4-octyl-2-thienyl)-5,6-difluoro-2,1,3-benzo-
thiadiazole (5). Compound 4 (1.88 g, 3.35 mmol) and NBS (1.19
g, 6.70 mmol) were dissolved in 70 mL chloroform before 0.17 g
silica gel was added. After stirring at RT for 3.5 h, the solvent
was removed and residue was purified by column chromatograph
with hexanes/chloroform (1/1 volume ratio) to give 2.37 g yellow
solid (yield: 98%). 1H NMR (CDCl3) d: 7.96 (s, 2H), 2.65 (t, 4H),
1.66 (m, 4H), 1.32 (m, 20H), 0.87 (t, 6H); 19F NMR (CDCl3) d:
ꢀ128.6; 13C NMR (CDCl3) d: 151.0 (d), 148.4, 148.2 (d), 142.6,
131.7, 131.2, 114.4, 111.0, 31.9, 29.8, 29.6, 29.4, 29.3, 22.7, 14.1.
Mp 96.4 ꢁC. MS (m/z): observed: 718.6, calculated: 718.0 for [M
ꢀ H]+.
General procedure for Stille reaction. The BT monomer,
4,7-bis(5-bromo-4-dioctyl-2-thienyl)-2,1,3-benzothiadiazole (0.205
g, 0.300 mmol) or monomer 5 (0.216 g, 0.300 mmol) and
compound 9 (0.289 g, 0.300 mmol) were added into a 25 mL flask
equipped with a condenser. The system was purged with argon
before Pd(PPh3)4 (0.028 g, 0.040 mmol) was added in a dry box.
Toluene (8 mL) was added and the solution was refluxed for 48 h.
The polymer solution was precipitated into methanol, and the
obtained powder was Soxhlet extracted with acetone and then
hexanes to remove oligmers and small molecular part. Then the
product was collected with toluene to give deep purple shining
3-Pentyl undec-1-yne (6). In a 500 mL flask purged with argon,
dry hexane (150 mL) and 1-octyne (16.5 g, 0.15 mol) was added.
Then n-BuLi (2.5M in hexane, 132 mL, 0.33 mol) wasꢁdropwise
added in a dry ice/acetone bath. After stirring at ꢀ78 C for 30
min, the solution was warmed up to ꢀ42 ꢁC using an acetonitrile/
dry ice bath for 1 h. Freshly distilled 1-bromooctane (29.0 g, 0.15
mol) was added and the mixture was allowed to slowly warm up
to RT overnight. HCl (6 M, 150 mL) was slowly added in an ice
water bath. The organic phase was separated and washed with
water, dried over MgSO4. The product was purified by vacuum
1
film. P1 (0.20 g, yield: 57%). H NMR (1,2-dichlorobenzene-d4,
60 ꢁC) d: 8.26 (s, 2H), 7.81 (s, 2H), 7.79 (s, 2H), 3.30 (br, 4H), 3.15
(br, 4H), 1.96, (br, 8H), 1.2–1.8 (br, 66H), 0.90 (br, 18H). GPC
analysis: Mn ¼ 16200 g molꢀ1, Mw ¼ 25600 g molꢀ1, PDI ¼ 1.58.
P2 (0.13 g, yield: 36%). 1H NMR (1,2-dichlorobenzene-d4, 60 ꢁC)
d: 8.41 (s, 2H), 7.82 (s, 2H), 3.32 (br, 4H) 3.17 (br, 4H), 1.96
(br, 8H), 1.2–1.8 (br, 66H), 0.90 (br, 18H); 19F NMR (o-DCB-d4,
1
distillation to give slightly yellowish oil (15 g, yield: 45%). H
NMR (CDCl3) d: 2.30 (m, 1H), 2.03 (d, 2H), 1.2–1.5 (m, 22H),
0.9 (m, 6H). HRMS EI calcd for C16H34 226.2661, found
226.2650 (M + 4H).
60 ꢁC) d: ꢀ127.8. GPC analysis: Mn ¼ 27800 g molꢀ1, Mw
¼
4,8-Di(3-pentylundecyn)benzo[1,2-b:4,5-b0]dithiophene (7). To
a solution of compound 6 (3.5 g, 13.6 mmol) in THF (4 mL),
isopropylmagnesium chloride (2 M in THF, 5.1 mL, 12.0 mmol)
was dropwise added. The solution was stirred for 90 min at 55 ꢁC
before benzo [1,2-b:4,5-b0] dithiophene-4,8-dione (0.5 g, 2.27
mmol) was added. After 60 min, SnCl2 solution (3.5 g in 8 mL
10% HCl) was slowly added. The solution was stirred for another
60 min at 65 ꢁC before was poured into water and extracted with
hexane. The product was purified by column chromatograph to
71400 g molꢀ1, PDI ¼ 2.57.
4.3. Device fabrication and testing
Bottom-contact thin film transistors were fabricated by spin-
coating P1 and P2 in o-dichlorobenzene solution at 60 ꢁC on
a heavily doped n-Si wafer with an overlayer of SiO2 (230 nm, Ci
¼ 15 nF/cm2). Then gold source and drain electrodes were
sputtered on the substrate prior to the deposition of polymer
3232 | J. Mater. Chem., 2011, 21, 3226–3233
This journal is ª The Royal Society of Chemistry 2011

View Online
5 N. Blouin, A. Michaud and M. Leclerc, Adv. Mater., 2007, 19, 2295;
N. Blouin, A. Michaud, D. Gendron, S. Wakim, E. Blair, R. Neagu-
^
Plesu, M. Belletete, G. Durocher, Y. Tao and M. Leclerc, J. Am.
Chem. Soc., 2008, 130, 732.
6 J. Hou, M.-H. Park, S. Zhang, Y. Yao, L.-M. Chen, J.-H. Li and
Y. Yang, Macromolecules, 2008, 41, 6012.
7 E. Zhou, M. Nakamura, T. Nishizawa, Y. Zhang, Q. Wei, K. Tajima,
C. Yang and K. Hashimoto, Macromolecules, 2008, 41, 8302.
8 J. Hou, H.-Y. Chen, S. Zhang, G. Li and Y. Yang, J. Am. Chem. Soc.,
2008, 130, 16144.
9 J. Lu, F. Liang, N. Drolet, J. Ding, Y. Tao and R. Movileanu, Chem.
Commun., 2008, 5315.
10 Q. Zheng, B. J. Sun, J. Jung and H. E. Katz, J. Am. Chem. Soc., 2010,
132, 5394.
11 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.
film. The transistor channel length and width are 20 mm and 10
mm, respectively. The current–voltage (J-V) characteristics were
measured with a computer-controlled semiconductor parameter
analyzer (HP4145A) in a N2 glove box. The hole mobility was
deduced from the saturation regime of the J-V characteristics.
Polymer solar cells were fabricated with a general structure of
ITO/PEDOT:PSS/Polymer:PC71BM/LiF/Al. Patterned ITO
glass substrates were cleaned with detergent before sonicated in
CMOS grade acetone and isopropanol for 15 min. The organic
residue was further removed by treating with UV-ozone for 10
min. Then a thin layer of PEDOT:PSS (Clevios P, H. C. Starck,
45 nm) was spin-coated and dried for 1 h at 120 ꢁC. Polymer P1
or P2 and PC71BM (ADS) (1 : 2 weight ratio) were dissolved in
o-dichlorobenzene containing 2.5% (v/v) diiodooctane or 1,2,4-
trichlorobenzene at 80 ꢁC to form a mixture solution. The
solution was filtered and spin-coated on the top of PEDOT:PSS
layer. The border of the PEDOT:PSS layer and active layer was
mechanically removed before 1.0 nm of LiF and 100 nm Al layer
were thermally evaporated through a shadow mask at a pressure
of 5 ꢃ 10ꢀ7 mbar in a Boc Edwards Auto 500 System. The active
area is 49 mm2.
12 R. Qin, W. Li, C. Li, C. Du, C. Veit, H.-F. Schleiermacher,
€
M. Andersson, Z. Bo, Z. Liu, O. Inganas, U. Wuerfel and
F. Zhang, J. Am. Chem. Soc., 2009, 131, 14612.
13 C. Kitamura, S. Tanaka and Y. Yamashita, Chem. Mater., 1996, 8,
570.
14 E. Bundgaard and F. C. Krebs, Macromolecules, 2006, 39, 2823;
A. P. Zoombelt, M. Fonrodona, M. M. Wienk, A. B. Sieval,
J. C. Hummelen and R. A. J. Janssen, Org. Lett., 2009, 11, 903.
15 F. Babudri, G. M. Farinola, F. Naso and R. Ragni, Chem. Commun.,
2007, 1003.
The current–voltage (J-V) characteristics were measured with
a Keithley 2400 digital source meter under simulated air mass
(AM) 1.5 solar irradiation of 100 mW cmꢀ2 (Sciencetech Inc.,
SF150). The light intensity was calibrated with a power meter
(Gentec Solo PE Laser Power & Energy Meter). The external
quantum efficiency (EQE) was performed using a Jobin-Yvon
Triax 180 spectrometer, a Jobin-Yvon xenon light source,
a Merlin lock-in amplifier, a calibrated Si UV detector, and an
SR 570 low noise current amplifier.
€
16 M. C. Scharber, D. Muhlbacher, M. Koppe, P. Denk, C. Waldauf,
A. J. Heeger and C. J. Brabec, Adv. Mater., 2006, 18, 789.
17 Y. Liang, D. Feng, Y. Wu, S.-T. Tsai, G. Li, C. Ray and L. Yu, J. Am.
Chem. Soc., 2009, 131, 7792; H.-Y. Chen, J. Hou, S. Zhang, Y. Liang,
G. Yang, Y. Yang, L. Yu, Y. Wu and G. Li, Nat. Photonics, 2009, 3,
650.
18 J. P. Dunn, T. R. Elworthy, D. Stefanidis and Z. K. Sweeney, WO
010545 A1, 2006.
19 Y. Yamashita, K. Ono, M. Tomura and S. Tanaka, Tetrahedron,
1997, 53, 10169.
20 Z. Li, J. Ding, N. Song, J. Lu and Y. Tao, J. Am. Chem. Soc., 2010,
132, 13160.
21 H. Pan, Y. Li, Y. Wu, P. Liu, B. S. Ong, S. Zhu and G. Xu, J. Am.
Chem. Soc., 2007, 129, 4112; H. Pan, Y. Li, Y. Wu, P. Liu,
B. S. Ong, S. Zhu and G. Xu, Chem. Mater., 2006, 18, 3237;
J. Hou, H.-Y. Chen, S. Zhang, R. I. Chen, Y. Yang, Y. Wu and
G. Li, J. Am. Chem. Soc., 2009, 131, 15586.
22 S. C. Price, A. C. Stuart and W. You, Macromolecules, 2010, 43,
4609.
23 S. Bhanu and F. Scheinmann, J. Chem. Soc., Perkin Trans. 1, 1979,
1218.
24 C. Piliego, T. W. Holcombe, J. D. Douglas, C. H. Woo,
Notes and references
ꢁ
1 B. C. Thompson and J. M. J. Frechet, Angew. Chem., Int. Ed., 2008,
47, 58; G. D. Dennler, M. C. Scharber and C. J. Brabec, Adv. Mater.,
2009, 21, 1323; Y.-J. Cheng, S.-H. Yang and C.-S. Hsu, Chem. Rev.,
€
2009, 109, 5868; S. Gunes, H. Neugebauer and N. S. Sariciftci, Chem.
Rev., 2007, 107, 1324.
2 G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. J. Heeger, Science,
1995, 270, 1789; J. J. M. Halls, C. A. Walsh, N. C. Greenham,
E. A. Marseglia, R. H. Friend, S. C. Moratti and A. B. Holmes,
Nature, 1995, 376, 498.
ꢁ
P. M. Beaujuge and J. M. J. Frechet, J. Am. Chem. Soc., 2009, 132,
7595.
25 J. S. Kim, Y. Lee, J. H. Lee, J. H. Park, J. K. Kim and K. Cho, Adv.
Mater., 2010, 22, 1355.
26 L.-M. Chen, Z. Hong, G. Li and Y. Yang, Adv. Mater., 2009, 21,
1434.
3 M. Svensson, F. Zhang, S. C. Veenstra, W. J. H. Verhees,
€
J. C. Hummelen, J. M. Kroon, O. Inganas and M. R. Andersson,
Adv. Mater., 2003, 15, 988; M.-H. Chen, J. Hou, Z. Hong, G. Yang,
S. Sista, L.-M. Chen and Y. Yang, Adv. Mater., 2009, 21, 4238.
€
27 M. Jørgensen, K. Norrman and F. C. Krebs, Sol. Energy Mater. Sol.
Cells, 2008, 92, 686.
28 P. Beinilirig and G. Kobmehl, Chem. Ber., 1986, 119, 3198.
4 D. Muhlbacher, M. Scharber, M. Morana, Z. Zhu, D. Waller,
R. Gaudiana and C. Brabec, Adv. Mater., 2006, 18, 2884;
R. C. Coffin,J. Peet,J.RogersandG.C.Bazan, Nat. Chem., 2009,1, 657.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 3226–3233 | 3233