New low band gap thieno[3,4-b]thiophene-based polymers with deep HOMO levels for organic solar cells

Article abstract of DOI:10.1039/c1jm11229e

Two new soluble alternating alkyl-substituted benzo[1,2-b:4,5-b′] dithiophene and ketone-substituted thieno[3,4-b]thiophene copolymers were synthesized and characterized. We found that grafting 3-butyloctyl side chains to the benzo[1,2-b:4,5-b′]dithiophene unit at C4 and C8 afforded the resulting polymer (P1) a high hole mobility (～10^-2 cm² Vs^-1) and a low-lying HOMO energy level (5.22 eV). Preliminary experiments in bulk heterojunction solar cells using P1 as the electron donor demonstrated a high power conversion efficiency of 4.8% even with PC ₆₁BM as the electron acceptor. The introduction of an electron-withdrawing fluorine atom into the thieno[3,4-b]thiophene unit at the C3 position (P2) lowers the HOMO energy level and consequently improves the open circuit voltage from 0.78 to 0.86 V. These values are about 0.1 V higher than those reported for their analogues based on alkoxy-substituted benzo[1,2-b:4,5-b′]dithiophene. This work demonstrates that the replacement of the alkoxy chains on the benzo[1,2-b:4,5-b′]dithiophene unit with less electron-donating alkyl chains is able to lower the HOMO energy levels of this class of polymers without increasing their band gap energy.

Full text of DOI:10.1039/c1jm11229e

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, 10920
www.rsc.org/materials
PAPER
New low band gap thieno[3,4-b]thiophene-based polymers with deep HOMO
levels for organic solar cells
a
a
b
a
a
a
Salem Wakim, Salima Alem, Zhao Li, Yanguang Zhang, Shing-Chi Tse, Jianping Lu, Jianfu Dingb
*
a
*
and Ye Tao
Received 22nd March 2011, Accepted 3rd May 2011
DOI: 10.1039/c1jm11229e
Two new soluble alternating alkyl-substituted benzo[1,2-b:4,5-b0]dithiophene and ketone-substituted
thieno[3,4-b]thiophene copolymers were synthesized and characterized. We found that grafting
3-butyloctyl side chains to the benzo[1,2-b:4,5-b0]dithiophene unit at C4 and C8 afforded the resulting
polymer (P1) a high hole mobility (ꢀ10ꢁ2 cm2 Vsꢁ1) and a low-lying HOMO energy level (5.22 eV).
Preliminary experiments in bulk heterojunction solar cells using P1 as the electron donor demonstrated
a high power conversion efficiency of 4.8% even with PC61BM as the electron acceptor. The
introduction of an electron-withdrawing fluorine atom into the thieno[3,4-b]thiophene unit at the C3
position (P2) lowers the HOMO energy level and consequently improves the open circuit voltage from
0.78 to 0.86 V. These values are about 0.1 V higher than those reported for their analogues based on
alkoxy-substituted benzo[1,2-b:4,5-b0]dithiophene. This work demonstrates that the replacement of the
alkoxy chains on the benzo[1,2-b:4,5-b0]dithiophene unit with less electron-donating alkyl chains is able
to lower the HOMO energy levels of this class of polymers without increasing their band gap energy.
thieno[3,4-b]thiophene unit further lowered both the HOMO and
Introduction
LUMO energy levels of the resulting polymers.11 This structure–
property relation allows us to increase the open circuit voltage
(Voc) but without increasing the band gap and reducing the short
circuit current (Jsc). The polymer based on 3-fluorothieno[3,4-b]
thiophene with ketone side chain showed the lowest LUMO and
HOMO energy levels (3.45 and 5.22 eV respectively) and the best
Voc (0.76 V) recorded for this class of materials.11 We noticed
that a large energy offset (ꢀ0.9 eV) exists between the LUMO
levels of this series of polymers and the commonly used electron
acceptor materials ([6,6]-phenyl C61-butyric acid methyl ester
(PC61BM) or PC71BM). This energy offset is overly generous for
efficient exciton dissociation;13 a slight reduction in this energy
offset should not affect the donor–acceptor charge transfer. This
gives us an opportunity to further lower the LUMO and HOMO
energy levels of this series of polymers to increase the Voc without
losing the Jsc. One effective method to lower the HOMO energy
level is to replace the alkoxy side chains on the benzo[1,2-b:4,5-b0]
dithiophene unit with less electron-donating alkyl chains.14,15 The
combination of such monomer with the ketone-substituted
thieno[3,4-b]thiophene monomers will offer new low band gap
copolymers with deep HOMO energy levels. In order to keep the
copolymers soluble in organic solvents for solution processing,
new branched alkyl side chains have to be designed. In this paper,
we report the synthesis of new alternating benzo[1,2-b:4,5-b0]
dithiophene and thieno[3,4-b]thiophene copolymers, P1 and P2,
and their photovoltaic performance in BHJ solar cells using
PC61BM as the electron acceptor. P1 and P2 showed Voc of 0.77
and 0.86 V, respectively, and a PCE up to 4.8%.
Organic solar cells based on conjugated polymers and soluble
fullerene derivatives have the potential to offer low-cost solar
electricity by using spin-coating, ink-jet printing, or roll-to-roll
printing techniques to fabricate large-area photovoltaic devices
on flexible and light-weight substrates.1–4 The power conversion
efficiencies (PCEs) of solution-processed bulk heterojunction
(BHJ) solar cells5 have recently exceeded 6%, primarily due to the
development of low band gap p-type materials6,7 and better
control of the interpenetrating electron donor/acceptor network
by using processing additives.8–10 Among these materials, alter-
nating thieno[3,4-b]thiophene and benzo[1,2-b:4,5-b0]dithio-
phene copolymers show the best performance with a PCE up to
7.4% when [6,6]-phenyl C71-butyric acid methyl ester (PC71BM)
was used as an electron acceptor.11,12 These materials are very
interesting not only because of their small band gaps but also due
to the possibility of tuning the HOMO and LUMO energy levels
simultaneously without affecting the optical properties or the
band gap energies of the polymers. Chen et al. showed that
substitution of the ester group with a ketone group or/and the
introduction of an electron-withdrawing fluorine atom into the
aInstitute for Microstructural Sciences, National Research Council of
Canada, 1200 Montreal Road, Ottawa, ON, K1A 0R6, Canada. E-mail:
jianping.lu@nrc-cnrc.gc; caye.tao@nrc-cnrc.gc.ca
bInstitute for Chemical Process and Environmental Technology, National
Research Council of Canada, 1200 Montreal Road, Ottawa, ON, K1A
0R6, Canada
10920 | J. Mater. Chem., 2011, 21, 10920–10928
This journal is ª The Royal Society of Chemistry 2011

View Online
chloride, provided compound 12 as white needles in a 70% yield.
As described in Scheme 2, P1 and P2 were synthesized by Stille
coupling reaction between monomer 12 and the corresponding
thieno[3,4-b]thiophene derivative in refluxing toluene/DMF
(10/1) using Pd(PPh3)4 as the catalyst. P1 and P2 were easily
dissolved in chlorinated solvents such as chloroform, chloro-
benzene and dichlorobenzene. P1 and P2 have relatively high
number-average molecular weights (Mn) of 27 and 20 kDa and
polydispersity indexes (PDI) of 2.0 and 1.75, respectively.16 We
note here that when the two 4-butyloctyl side chains of P2 were
replaced by the shorter and more symmetric 3-propylhexyl, the
resulting polymer had a lower solubility and precipitated out of
the solution during the polymerization. Therefore, the soluble
fraction extracted with refluxing chlorobenzene has a lower
molecular weight (Mn ¼ 17 kDa; PDI ¼ 1.60) than P2 and forms
aggregates quickly after cooling to room temperature. Interest-
ingly, using our synthetic approach, the substitution pattern and
length of the branched alkyl chain of the benzo[1,2-b:4,5-b0]
dithiophene unit could be easily tuned to obtain a balance
between good solubility and efficient interchain packing of the
resulting polymer.19
Results and discussion
Material synthesis
Scheme 1 shows the synthesis of monomers 3 and 8. First,
compound 1 was prepared in three steps according to the
procedure reported for its analogue with a branched alkyl
chain.16 Then, compound 1 was converted to the fully aromatic
derivative 2 in an 81% yield. The latter was brominated with N-
bromosuccinimide to give monomer 3 as a light pink solid in an
83% yield. The monomer 8 was already published in the litera-
ture11 but no information was given about its preparation.
Therefore, we have developed independently a synthetic pathway
for the preparation of monomer 8 in five steps starting from 1. As
shown in Scheme 1, the treatment of 1 with ethylene glycol in
refluxing benzene in the presence of a catalytic amount of
p-toluenesulfonic acid offered the desired ketal 4 in a 95% yield.
Lithiation of 4 with n-BuLi in THF and subsequent treatment
with N-fluorobenzenesulfonimide furnished the fluorinated
compound 5 in a 37% yield. Deprotection of 5 in THF in the
presence of 3 N HCl solution provided compound 6 in an 89%
yield. Finally, the conversion of 6 to the fully aromatic derivative
7 followed by the bromination of the latter gave the desired
monomer 8 as an orange-pink solid in excellent yields.
Optical and electrochemical properties
The new branched alkyl substituted benzo[1,2-b:4,5-b0]dithio-
phene 12 was prepared in three steps according to the procedure
reported by Ong et al.17 using benzo[1,2-b:4,5-b0]dithiophene-4,8-
dione and alkyne as starting materials (Scheme 2). In our case,
the 3-butyl-1-octyne 9 was prepared from 1-octyne. The latter
was first treated with 2 equiv. of n-BuLi in hexane. Then, the
regioselective alkylation at C-3 position18 with bromobutane
gave compound 9 in a 36% yield. Compound 9 was treated with
isopropylmagnesium chloride to give the corresponding alky-
nylmagnesium chloride, which then reacted with benzo[1,2-b:4,5-
b0]dithiophene-4,8-dione. After reduction with SnCl2, compound
10 was obtained in a 50% yield. Hydrogenation of 10 furnished
compound 11 in a 70% yield. Finally, the treatment of 11 with
n-BuLi in THF, followed by the addition of trimethyltin
The optical properties of P1 and P2 thin films were characterized
by UV-vis absorption spectroscopy. As shown in Fig. 1, P1 and
P2 showed absorption maxima at 687 and 676 nm, respectively.
These values coincide well with the maximum photon flux region
in the solar spectrum (ꢀ700 nm).2 The optical band gaps of P1
and P2 calculated from the onset of the films absorption are 1.63
and 1.65 eV, respectively. These values are consistent with their
analogues based on alkoxy-substituted benzo[1,2-b:4,5-b0]
dithiophene,11 which means the replacement of the alkoxy chains
by alkyl chains almost has little influence on the band gaps of this
class of polymers. The HOMO and LUMO energy levels of P1
and P2 were evaluated by cyclic voltammetry (CV) measure-
ment. Both polymers underwent reversible cathodic reduction
Scheme 1 Synthetic route for thieno[3,4-b]thiophene monomers.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 10920–10928 | 10921

View Online
Scheme 2 Synthetic route for benzo[1,2-b:4,5-b0]dithiophene monomer and corresponding polymers.
Fig. 2 Cyclic voltammograms of P1 and P2 films cast on a platinum
Fig. 1 Absorption spectra of P1 and P2 in thin films.
electrode in 0.1 M Bu4NPF6/CH3CN solution.
substituted benzo[1,2-b:4,5-b0]dithiophene.11 This could be
related to the interface charge injection barrier due to the pres-
ence of insulating side chains between the polymer film and the
electrode surface.20,21
and anodic oxidation (Fig. 2). The onset potentials of the
oxidation wave of P1 and P2 appeared at 0.82 and 0.94 V, cor-
responding to HOMO energy levels of 5.22 and 5.34 eV,
respectively. Based on the onset potentials of the first reduction
wave, the LUMO energy levels of P1 and P2 were estimated to be
at 3.41 and 3.53 eV, respectively. Both HOMO and LUMO
energy levels of P1 and P2 are about 0.1 eV lower than the values
obtained from their analogues based on alkoxy-substituted
benzo[1,2-b:4,5-b0]dithiophene,11 confirming that the replace-
ment of the alkoxy chains on the benzo[1,2-b:4,5-b0]dithiophene
unit with less electron-donating alkyl chains lowers both the
HOMO and LUMO energy levels of this class of polymers and
meanwhile keeps their band gap energy almost unchanged. It is
worth noting that the electrochemical band gaps of P1 and P2
are higher than the optically measured ones, as already observed
by Chen et al. for the similar polymers based on alkoxy-
Thermal properties
Differential scanning calorimetry (DSC) analysis showed that
both polymers P1 and P2 are crystalline materials with melting
points at 203.2 and 208.5 ꢂC, respectively (Fig. 3). However, P1
has a higher crystallinity and its melting enthalpy reaches 7.14 J
gꢁ1. In contrast, the melting enthalpy of P2 is only 3.53 J gꢁ1. In
addition, P1 tends to form crystalline structures when it is cooled
from the melted state. Therefore, no obvious glass transition is
observed during the second DSC heating scan. P2 shows
a different thermal behavior. Its second DSC scan reveals a glass
10922 | J. Mater. Chem., 2011, 21, 10920–10928
This journal is ª The Royal Society of Chemistry 2011

View Online
structures in the solid state, leading to considerable improvement
in their charge transport properties.23–25
Photovoltaic response
Photovoltaic properties of P1 and P2 were investigated in BHJ
solar cells, using a device structure of ITO/PEDOT:PSS/polymer:
PC61BM/LiF/Al with a device active area of 1.0 cm2. Due to
the good solubility of P1 and P2 in both chloroform (CF) and
1,2-dichlorobenzene (ODCB) at room temperature, these two
solvents were used to prepare the polymer:PC61BM composite
films in order to compare their effect on the device performance.
The photovoltaic performances of P1 and P2 were evaluated
using a polymer : PC61BM weight ratio of 1 : 2 and an active
layer thickness of ꢀ100 nm. The devices were tested under AM
1.5G irradiation of 100 mW cmꢁ2 calibrated with a KG5 filter
covered silicon photovoltaic solar cell. The device performance
data are summarized in Table 1, where the PCEs were calculated
using the Jsc obtained from the integration of the external
quantum efficiency and the AM 1.5G solar spectrum.
Fig. 3 DSC curves of P1 and P2 collected under N2 at a heating rate of
10 ꢂC minꢁ1
.
transition at 143.7 ꢂC, and a broad melting process above 185 ꢂC.
This means that the substitution of the F atom on the polymer
backbone slightly increases the melting point but retards the
recrystallization process probably due to the asymmetric
molecular structure of the new polymer (P2). In fact, it has been
shown with some poly(2,7-carbazole) derivatives that symmetric
polymers have higher structural organization in the solid state
than their corresponding asymmetric polymers.22
As we can see in Table 1, using only chloroform as the solvent,
P1 and P2 exhibited a similar PCE of 2.3% with relatively low Jsc
of 6.1 and 5.4 mA cmꢁ2, respectively. In both cases, the AFM
images of the polymer:PC61BM composites show large domains
with a diameter of about 150–200 nm (Fig. 5). Such big domain
size favors the exciton recombination and consequently reduces
the charge separation efficiency.1,26 As a result, the photocurrent
density values are low. We note here that when 1,2-dichloro-
benzene is used as the processing solvent for P1 and P2, the PCE
values and films morphology (Fig. 5) are very similar to those
obtained with the chloroform. According to these results, one
can suggest that the observed large domains are not related to the
solubility of P1 and P2 in chloroform or 1,2-dichlorobenzene but
to the low miscibility of these polymers with PC61BM.
Field-effect transistor mobility
The hole mobilities of P1 and P2 were measured using an organic
field-effect transistor (OFET) technique. Bottom contact OFETs
of these two copolymers were fabricated on SiO2/Si substrates as
described in the Experimental section. P1 and P2 were found to
exhibit typical p-type organic semiconductor characteristics
The use of processing additives, such as 1,8-diiodooctane
(DIO), has been proven an effective method to modify and
control the nano-scale morphology in polymer solar cells.8–10
Therefore, new P1 and P2 based devices have been prepared
using chloroform/DIO (97/3, v/v) mixture as the solvent. The
(Fig. 4) with hole mobilities of 1 ꢃ 10ꢁ2 and 4 ꢃ 10ꢁ4 cm2 Vꢁ1 sꢁ1
,
respectively. The higher hole mobility of P1 is probably related to
its superior crystallinity, as observed by the DSC analysis. It is
well known that crystalline polymers produce organized
Fig. 4 Source–drain current (IDS) versus source–drain voltage (VDS) at different gate voltages (VG) and transfer characteristics in the saturation regime
at constant source–drain voltage (VDS ¼ ꢁ100 V) for OFET using spin-coated P1 on SiO2/Si substrate as the active channel.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 10920–10928 | 10923

View Online
Table 1 Performance parameters of polymer solar cells using PC61BM as an electron acceptor
Polymer
Solvent
Voc/V
FF
PCEa (%)
a
Jsc /mA cmꢁ2
P1
P1
P1
P1
P2
P2
P2
P2
CHCl3
CHCl3 + 3% DIO
ODCB
ODCB + 3% DIO
CHCl3
CHCl3 + 3% DIO
ODCB
ODCB + 3% DIO
0.82
0.78
0.81
0.77
0.90
0.86
0.89
0.84
6.1
9.8
5.6
9.5
5.4
7.3
5.0
6.7
0.47
0.63
0.49
0.64
0.48
0.62
0.50
0.54
2.3
4.8
2.2
4.7
2.3
3.9
2.2
3.1
a
The PCEs were calculated using the Jsc obtained from the external quantum efficiency (EQE) measurements.
current density–voltage characteristics (J–V) of these devices are
shown in Fig. 6a and the data are summarized in Table 1. One
can see that using DIO as an additive considerably increases the
Jsc and fill factor (FF) for both polymers. In the case of P1, the
device exhibited a Jsc of 9.8 mA cmꢁ2, a Voc of 0.78 V, and a FF
of 63% with an overall PCE of 4.8%. As expected, the fluorinated
polymer P2 showed a significantly enhanced Voc of 0.86 V,
as compared to that of P1. These Voc values are about 0.1 V
higher than those obtained with their analogues based on
alkoxy-substituted benzo[1,2-b:4,5-b0]dithiophene11 indicating,
as expected from HOMO energy levels, that replacement of the
alkoxy chains by alkyl chains is an effective method to increase
the Voc of the devices based on this class of polymers. The 1,2-
dichlorobenzene/DIO (97/3, v/v) mixture has also been tested as
the processing solvent. As shown in Table 1, adding DIO to the
1,2-dichlorobenzene significantly improves the device perfor-
mance but without exceeding those obtained with the chloro-
form/DIO (97/3, v/v) mixture. The significant enhancement in the
Jsc and FF, when DIO is used, could be explained by the change
in the morphology of the blend, as we can see from the AFM
images (Fig. 5). In the case of P1, no large domains are present
after adding DIO to the chloroform or 1,2-dichlorobenzene, and
the domain structure becomes much smaller, indicating a better
nano-scale phase separation between the polymers and PC61BM.
We believe there is still more room for further optimizations to
realize optimum morphology of the active layer, particularly for
P2. It is worth noting that there was a slight drop in the Voc when
DIO was used as the processing additive. This phenomenon
was typically observed in the literature.9 The mechanism is not
very clear at this moment. We think it may be related to the
formation of a very thin PC61BM-enriched layer on PEDOT-PSS
as DIO selectively dissolves PC61BM and has a high density
(1.84 g mLꢁ1).
The EQE spectra (Fig. 6b) show that P1 and P2 based solar
cells harvest light very efficiently between 600 and 700 nm, as
expected from the UV-vis spectra of P1 and P2 (Fig. 1).
However, their absorption below 550 nm is weak, resulting in low
EQE values in this spectral region. The use of PC71BM as an
Fig. 5 AFM phase images (surface scan area: 1 ꢃ 1 mm2) of the P1:PC61BM [(a), (b), (e) and (f)] and P2:PC61BM [(c), (d), (g) and (h)] composite films
prepared using the chloroform [(a) and (c)], the 1,2-dichlorobenzene [(b) and (d)], chloroform/DIO (97/3, v/v) mixture [(e) and (g)] and 1,2-dichloro-
benzene/DIO (97/3, v/v) mixture [(f) and (h)].
10924 | J. Mater. Chem., 2011, 21, 10920–10928
This journal is ª The Royal Society of Chemistry 2011

View Online
reported for this class of materials. The results presented here
show that the replacement of the alkoxy chains on the benzo[1,2-
b:4,5-b0]dithiophene unit with less electron-donating alkyl chains
allows us to increase the Voc of these polymers by 0.1 V but
without increasing their band gap energy which remains around
1.64 eV. Further improvement in device performance is expected
after the device optimization, such as replacing PC61BM with
PC71BM and optimization of the active layer morphology by
using new fabrication processes.
Experimental section
Materials
All starting materials, unless otherwise specified, were purchased
from Aldrich Co. and used without further purification. The
THF used in the reactions was freshly distilled over sodium
under argon. Column chromatography was carried out on silica
ꢀ
gel (size 40–63 mm, pore size 60 A, Silicycle). 1-(4,6-Dihy-
drothieno[3,4-b]thiophen-2-yl)octane-1-one (1) was synthesized
according to the literature methods.16 All other compounds were
synthesized following the procedures described below.
1-(Thieno[3,4-b]thiophen-2-yl)octan-1-one (2)
In a 100 mL flask, 1.66 g (6.19 mmol) of 1 was dissolved in 21 mL
ꢂ
of chloroform and cooled down to ꢁ40 C. Then, 1.52 g (6.19
mmol) of 3-chloroperoxybenzoic acid (77%) dissolved in 10 mL
of chloroform was added dropwisely and the resulting mixture
was stirred for 30 min at ꢁ40 ꢂC. The mixture was warmed up to
room temperature and stirred for 30 min. The chloroform was
removed under reduced pressure, and 6.2 mL of acetic acid
anhydride was added and the mixture was refluxed for 20 min.
The volatile substance was removed under reduced pressure and
the oily crude material residue was purified by column chroma-
tography (silica gel, 5% ethyl acetate in hexanes as eluent) to
provide 1.33 g of the title product as a white solid (81% yield). 1H
NMR (400 MHz, CDCl3, ppm) d 7.63 (d, 1H, J ¼ 2.8 Hz), 7.59 (s,
1H), 7.27 (d, 1H, J ¼ 2.8 Hz), 2.90 (t, 2H, J ¼ 7.4 Hz), 1.74 (m,
2H), 1.32 (m, 8H), 0.88 (t, 3H, J ¼ 6.8 Hz). HRMS (EI) calcu-
lated for C14H18OS2 266.0799, found 266.0803.
Fig. 6 Current density–voltage curves measured under AM 1.5G irra-
diation of 100 mW cmꢁ2 (a) and EQE spectra (b) of polymer:PC61BM BHJ
solar cells prepared from mixed solvents chloroform/DIO (97/3, v/v).
electron acceptor could potentially overcome this problem since
PC71BM has stronger absorption between 400 and 600 nm. This
work is presently in progress. In addition, the molecular weight
of P2 (Mn ¼ 20 kDa) may be not high enough. We are optimizing
our synthetic procedure to further increase its molecular weight.
1-(4,6-Dibromothieno[3,4-b]thiophen-2-yl)octan-1-one (3)
In a 25 mL flask, 319 mg (1.2 mmol) of 2 was dissolved in 3 mL of
DMF under argon. Then, 534 mg (3 mmol) of N-bromosucci-
nimide dissolved in 3 mL of DMF was added, and the mixture
was stirred for 25 min at room temperature in the absence of
light. The mixture was quenched with 20 mL of 5% sodium
thiosulfate solution in an ice bath and extracted three times with
diethyl ether. The combined organic fractions were dried over
magnesium sulfate and the solvent was removed under reduced
pressure. The crude material was purified by column chroma-
tography (silica gel, 4% ethyl acetate in hexanes as eluent) to
provide 420 mg of the title product as a light pink-orange solid
Conclusions
In conclusion, we report the synthesis of new alternating
alkyl-substituted benzo[1,2-b:4,5-b0]dithiophene and ketone-
substituted thieno[3,4-b]thiophene copolymers. When branched
3-butyloctyl side chains were introduced to the benzo[1,2-b:4,5-
b0]dithiophene unit at C4 and C8 positions, the resulting polymer
P1 showed excellent solubility, high hole mobility (ꢀ10ꢁ2 cm2
Vsꢁ1) and low-lying HOMO energy level (5.22 eV). P1 blended
with PC61BM, demonstrated a Voc of 0.78 V and a high PCE of
4.8%. P2 with fluorinated thieno[3,4-b]thiophene exhibited
a significantly improved Voc of 0.86 V, as compared to that of P1.
To the best of our knowledge, these are the highest Voc values
1
(83% yield). H NMR (400 MHz, benzene-d6, ppm) d 6.86 (s,
1H), 2.27 (t, 2H, J ¼ 7.2 Hz), 1.57 (m, 2H), 1.26 (m, 2H), 1.18 (m,
6H), 0.90 (t, 3H, J ¼ 7.0 Hz). HRMS (EI) calculated for
C14H16Br2OS2 421.9409, found 421.9016.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 10920–10928 | 10925

View Online
2-(4,6-Dihydrothieno[3,4-b]thiophen-2-yl)-2-heptyl-1,3-
dioxolane (4)
(m, 2H), 1.32 (m, 8H), 0.88 (t, 3H, J ¼ 7.0 Hz). HRMS (EI)
calculated for C14H19FOS2 286.0861, found 286.0854.
A mixture of 2.0 g (7.5 mmol) of compound 3, 4.65 g (75 mmol)
of ethylene glycol, 52 mL of benzene and 0.14 g (0.75 mmol) of
p-toluenesulfonic acid monohydrate was heated at reflux under
argon in a 100 mL flask equipped with a Dean–Stark apparatus.
After 72 h, the reaction mixture was cooled to room temperature
and 125 mL of ethyl acetate and 48 mL of water were added. The
mixture was extracted three times with ethyl acetate. The
combined organic fractions were washed with 80 mL of 5%
NaHCO3 solution, dried over magnesium sulfate and the
solvents were removed under reduced pressure. The oily crude
material was purified by column chromatography (silica gel, 50%
dichloromethane in hexanes as eluent) to provide 2.2 g of the title
1-(3-Fluorothieno[3,4-b]thiophen-2-yl)octane-1-one (7)
In a 25 mL flask, 0.48 g (1.67 mmol) of 6 was dissolved in 7 mL of
chloroform and cooled down to ꢁ40 ꢂC. Then, 0.38 g (1.67
mmol) of 3-chloroperoxybenzoic acid (77%) dissolved in 3 mL of
chloroform was added dropwise and the mixture was stirred for
30 min at ꢁ40 ꢂC. The mixture was warmed up to room
temperature and stirred for 75 min. The chloroform was removed
under reduced pressure, and 2 mL of acetic acid anhydride was
added and the mixture was refluxed for 20 min. The volatile
substance was removed under reduced pressure and the crude
material was purified by column chromatography (silica gel, 4%
ethyl acetate in hexanes as eluent) to provide 0.41 g of the title
product as a white solid (86% yield). 1H NMR (400 MHz,
1
product as a pale yellow oil (95% yield): H NMR (400 MHz,
acetone-d6, ppm) d 6.74 (s, 1H), 4.12 (m, 2H), 3.99 (m, 4H), 3.91
(m, 2H), 1.91 (m, 2H), 1.38 (m, 2H), 1.27 (m, 8H), 0.87 (t, 3H,
J ¼ 7.2 Hz).
acetone-d6, ppm) d 8.08 (d, 1H, J ¼ 2.8 Hz), 7.70 (t, 1H, J(H–H)
¼
2.6 Hz and J(H–F) ¼ 2.6 Hz), 2.99 (td, 2H, J(H–H) ¼ 7.2 Hz and
J(H–F) ¼ 2.4 Hz), 1.71 (m, 2H), 1.30–140 (m, 8H), 0.89 (t, 3H, J ¼
7.0 Hz). HRMS (EI) calculated for C14H17FOS2 284.0705, found
284.0698.
2-(3-Fluoro-4,6-dihydrothieno[3,4-b]thiophen-2-yl)-2-heptyl-1,3-
dioxolane (5)
1-(3-Fluoro-4,6-dibromothieno[3,4-b]thiophen-2-yl)octan-1-one (8)
In a 100 mL flame-dried flask, 1.51 g (4.85 mmol) of 4 was dis-
solved in 42 mL of dry THF. The solution was cooled at ꢁ78 ꢂC
and 2.50 mL of n-BuLi solution (2.5 M in hexanes, 6.31 mmol)
was added dropwise under argon. After addition, the mixture
In a 25 mL flask, 200 mg (0.70 mmol) of 7 was dissolved under
argon in 2 mL of DMF. Then, 313 mg (1.76 mmol) of N-bro-
mosuccinimide dissolved in 2 mL of DMF was added, and the
mixture was stirred for 75 min at room temperature in the
absence of light. The mixture was quenched with 10 mL of 5%
sodium thiosulfate solution cooled with ice bath and extracted
three times with diethyl ether. The combined organic fractions
were dried over magnesium sulfate and the solvent was removed
under reduced pressure. The crude material was purified by
column chromatography (silica gel, 3% ethyl acetate in hexanes
as eluent) to provide 250 mg of the title product as a light pink-
orange solid (81% yield). 1H NMR (400 MHz, benzene-d6, ppm)
d 2.56 (td, 2H, J(H–H) ¼ 7.2 Hz and J(H–F) ¼ 2.4 Hz), 1.62
(m, 2H), 1.18–1.30 (m, 8H), 0.89 (t, 3H, J ¼ 7.2 Hz). HRMS (EI)
calculated for C14H15Br2FOS2 439.9315, found 439.8898.
ꢂ
was stirred at ꢁ78 C for 2 h. A solution of 1.99 g (6.31 mmol)
N-fluorobenzenesulfonimide in 12 mL of THF was added
dropwise at ꢁ78 ꢂC. After 1.5 h, the cold bath was removed and
the mixture was stirred at room temperature for 2.5 h. The
reaction was quenched with cold NH4Cl solution and extracted
three times with ethyl acetate. The combined organic fractions
were washed with H2O, dried over magnesium sulfate and the
solvents were removed under reduced pressure. The oily crude
material was purified by column chromatography (silica gel, 50%
dichloromethane in hexanes as eluent) to provide 0.64 g of the
1
title product as a pale yellow oil (37% yield, purity by H NMR
92%). Compound 5 was not stable and was used immediately. 1H
NMR of compound 5 (400 MHz, acetone-d6, ppm) d 4.13
(m, 2H), 4.00 (m, 2H), 3.96 (m, 4H), 1.98 (m, 2H), 1.40 (m, 2H),
1.27 (m, 8H), 0.87 (t, 3H, J ¼ 6.8 Hz).
3-Butyl-1-octyne (9)
In a 500 mL oven-dried flask, 10.0 g of 1-octyne (90.7 mmol) and
80 mL of anhydrous hexanes were added under argon. After the
solution was cooled to ꢁ78 ꢂC in a dry ice/acetone bath, 79 mL of
n-butyl lithium (2.5 M in hexanes) was added via a syringe with
vigorous stirring. A lot of white solids precipitated out of the
solution in the beginning. With more n-butyl lithium added, the
white solids dissolved to form a clear but very viscous yellow
solution. The resulting solution was further stirred at ꢁ42 ꢂC in
a dry ice/acetonitrile bath for another 1 h. Then, 12.3 g of freshly
distilled 1-bromobutane (90 mmol) was added. The solution was
allowed to warm to room temperature overnight. 40 mL of 6 M
HCl aqueous solution was dropwise added at 0 ꢂC. The organic
layer was separated and washed with water. After removal of
organic solvents on a rotary evaporator, the target product was
separated from byproducts by fractional distillation under
vacuum (65–75 ꢂC/0.2 mm Hg). Totally, 5.40 g of clear colorless
oil was obtained in 36% yield. 1H NMR (400 MHz, benzene-d6,
1-(3-Fluoro-4,6-dihydrothieno[3,4-b]thiophen-2-yl)octane-1-one (6)
In a 25 mL flask, 0.62 g (1.88 mmol) of 5 was dissolved in 7 mL of
ꢂ
THF. The solution was cooled at 0 C, and 2.4 mL of 3 N HCl
solution was added dropwise. After addition, the mixture was
stirred at room temperature for 4 h. 20 mL of ethyl acetate and
100 mL of water were added. The organic layer was separated,
and the aqueous layer was extracted three times with ethyl
acetate. The combined organic fractions were washed with H2O,
5% NaHCO3 solution, dried over magnesium sulfate and the
solvents were removed under reduced pressure. The crude solid
material was purified by column chromatography (silica gel, 50%
dichloromethane in hexanes as eluent) to provide 0.48 g of the
1
title product as a white solid (89% yield). H NMR (400 MHz,
acetone-d6, ppm) d 4.24 (t, 2H, J ¼ 3.0 Hz), 4.04 (t, 2H, J ¼ 3.2
Hz), 2.85 (td, 2H, J(H–H) ¼ 7.2 Hz and J(H–F) ¼ 2.0 Hz), 1.67
10926 | J. Mater. Chem., 2011, 21, 10920–10928
This journal is ª The Royal Society of Chemistry 2011

View Online
ppm) d 2.23 (m, 1H), 1.89 (d, 1H, J ¼ 2.4 Hz), 1.60–1.15 (m,
14H), 0.92–0.85 (m, 6H). MS (ESI) m/z 204.92 [(M + K)+
calculated for C12H22 205.26].
mmol) of Pd(PPh3)4 and a condenser were added under argon
atmosphere. 7 mL of degassed toluene/DMF (10 : 1, v/v) was
added and the reaction mixture was stirred at 110 ꢂC under argon
atmosphere. After 41 h of reaction time, 110 mg of bromo-
benzene (0.7 mmol) in 1 mL of degassed toluene/DMF (10 : 1,
v/v) was added to the reaction and the reaction was kept at 110
ꢂC for an additional 5 hours. The polymerization solution was
cooled to room temperature and precipitated into methanol. The
resultant polymer was collected by filtration, dried, and extracted
with hexanes and dichloromethane using a Soxhlet extraction
apparatus. The remaining solid was extracted with 250 mL of
toluene. After evaporation of the toluene under reduced pres-
sure, methanol was added and 240 mg of the polymer was
collected by filtration (85% yield). Molecular weight (by GPC
using chlorobenzene as eluent and monodispersed polystyrene as
4,8-Bis(3-butyloct-1-ynyl)benzo[1,2-b:4,5-b0]dithiophene (10)
Thiscompound was synthesized following a literature procedure17
except that 1-dodecyne was replaced with 3-butyl-1-octyne.
Briefly, 4.53 g (27.3 mmol) of 3-butyl-1-octyne in 7 mL of dry THF
was first treated with 13 mL of isopropylmagnesium chloride (2 M
in THF, 26 mmol). The resulting 3-butyloct-1-ynylmagnesium
chloride directly reacted with 1.00 g of benzo[1,2-b:4,5-b0]dithio-
phene-4,8-dione (4.54 mmol) in anhydrous THF, and the inter-
mediate product was reduced with SnCl2/HCl solution (7 g in 18
mL 10% HCl) to give crude 4,8-bis(3-butyloct-1-ynyl)benzo[1,2-
b:4,5-b0]dithiophene (10), which was purified by column chro-
matography (silica gel, 10% dichloromethane in hexanes as
eluent) to give 1.20 g of bright yellow solids (50% yield). 1H NMR
(400 MHz, CDCl3, ppm) d 7.55 (d, 2H, J ¼ 5.6 Hz), 7.49 (d, 2H,
J ¼ 5.6 Hz), 2.74 (m, 2H), 1.75–1.32 (m, 28H), 1.00–0.85 (m, 12H).
1
standard): Mn ¼ 27 kDa and PDI ¼ 2.0. H NMR (400 MHz,
1,2-dichlorobenzene-d4, 140 ꢂC, ppm) d 7.87 (br, 1H), 7.60–7.40
(br, 2H), 3.15 (br, 6H), 2.00–1.35 (m, 44H), 0.95–1.25 (m, 15H).
Polymer P2
4,8-Bis(3-butyloctyl)benzo[1,2-b:4,5-b0]dithiophene (11)
Polymer P2 was synthesized according to the procedure for the
synthesis of polymer P1. Equimolar monomer 8 (153 mg, 0.346
mmol) and monomer 12 (295 mg, 0.346 mmol) were used. After
62 h of reaction time, 12 mg of trimethylphenyltin (0.05 mmol) in
0.5 mL of degassed toluene/DMF (10 : 1, v/v) was added, and the
reaction was kept at 110 ꢂC for an additional 5 hours. 50 mg of
bromobenzene (0.32 mmol) in 1 mL of degassed toluene/DMF
(10 : 1, v/v) was then added, and the reaction was kept at 110 ꢂC
for an additional 6 hours to complete the end-capping reaction.
The polymerization solution was cooled to room temperature
and precipitated into methanol. The resultant polymer was
collected by filtration, dried, and extracted with hexanes and
dichloromethane using a Soxhlet extraction apparatus. The
remaining solid was extracted with 250 mL of toluene. After
evaporation of the toluene under reduced pressure, methanol was
added and 202 mg of the polymer was collected by filtration (70%
yield). Molecular weight (by GPC using chlorobenzene as eluent
and monodispersed polystyrene as standard): Mn ¼ 20 kDa and
This compound was synthesized following a literature proce-
dure.17 Briefly, 1.15 g of compound 10 (2.22 mmol) was reduced
by hydrogen reduction using 10% Pd/C (0.24 g) as the catalyst in
40 mL of anhydrous THF overnight. After purification by
column chromatography using hexanes as an eluent, the target
1
product was obtained in 70% yield as a white solid (0.82 g). H
NMR (400 MHz, benzene-d6, ppm) d 7.37 (d, 2H, J ¼ 5.6 Hz),
7.35 (d, 2H, J ¼ 5.6 Hz), 3.22–3.15 (m, 4H), 1.85–1.78 (m, 4H),
1.52–1.22 (m, 30H), 0.98–0.85 (m, 12H).
2,6-Bis(trimethyltin)-4,8-bis(3-butyloctyl)benzo[1,2-b:4,5-b0]
dithiophene (12)
0.81 g of compound 11 (1.53 mmol) was dissolved in 20 mL of
anhydrous THF, and cooled to ꢁ78 ꢂC in a dry ice/acetone bath.
1.5 mL of n-butyl lithium (2.5 M in hexanes) was added via
a syringe with stirring. After stirring at ꢁ78 ꢂC for 0.5 h, the cold
bath was removed, and the reaction mixture was further stirred
at room temperature for 2 h. A lot of white solids formed. The
1
PDI ¼ 1.75. H NMR (400 MHz, 1,2-dichlorobenzene-d4, 140
ꢂC, ppm) d 7.82 (br, 1H), 7.46 (br, 1H), 3.22 (br, 6H), 2.00–1.35
ꢂ
(m, 44H), 0.95–1.25 (m, 15H).
mixture was cooled down to ꢁ78 C again, and 4.4 mL of tri-
methyltin chloride (1 M) was added in one potion. After stirring
ꢂ
at ꢁ78 C for 0.5 h, the cold bath was removed, and the clear
Instrumentation
reaction solution was stirred at room temperature for another 2
h. 50 mL of water was added, and the mixture was extracted with
hexanes. The organic layer was separated, dried over MgSO4,
and concentrated under reduced pressure. Recrystallization of
the residue from isopropanol two times yielded the product as
white needles (0.92 g, 70% yield). 1H NMR (400 MHz, acetone-
d6, ppm) d 7.65 (s, 2H), 3.25–3.18 (m, 4H), 1.80–1.65 (m, 4H),
1.55–1.25 (m, 30H), 0.98–0.85 (m, 12H), 0.46 (m, 18H). MS (ESI)
m/z 889.62 [(M + K)+ calculated for C40H70S2Sn2 889.39].
1H and 13C NMR spectra were recorded on a 400 MHz Varian
Unity Inova spectrometer. Chemical shifts were reported as
d values (ppm) relative to internal tetramethylsilane. The
number-average (Mn) molecular weight and polydispersity index
(PDI) of the polymers were determined by gel permeation
chromatography (GPC) using a Waters Breeze HPLC system
with a 1525 binary HPLC pump and a 2414 differential refrac-
tometer. Chlorobenzene was used as eluent and commercial
polystyrenes were used as standards. UV-visible absorption
spectra were recorded on a Varian Cary 50 UV-Vis spectro-
photometer. The differential scanning calorimetry (DSC) anal-
ysis was performed on a TA Instruments DSC 2920. DSC traces
were measured at a scanning rate of 10 ꢂC minꢁ1, under
a nitrogen flow (50 mL minꢁ1). Cyclic voltammetry (CV)
Polymer P1
In a 50 mL oven-dried one-necked flask, 151 mg (0.356 mmol) of
monomer 3 and 303.5 mg (0.356 mmol) of monomer 12 were
added. The flask was purged 3 times with argon, then 25 mg (0.02
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 10920–10928 | 10927

View Online
measurements were carried out under argon in a three-electrode
cell using 0.1 M Bu4NPF6 in anhydrous CH3CN as the sup-
porting electrolyte. The polymers were coated on a 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 0.40 V versus the Ag quasi-refer-
ence electrode. Therefore, the HOMO and LUMO energy levels
Keithley 2400 source meter. The external quantum efficiency
(EQE) was performed using a Jobin-Yvon Triax spectrometer,
a Jobin-Yvon xenon light source, a Merlin lock-in amplifier,
a calibrated Si UV detector, and an SR570 low noise current
amplifier.
Acknowledgements
This work was financially supported by Sustainable Development
Technology Canada (SDTC). The authors wish to thank Ms R.
Movileanu, Mr E. Estwick, Mr H. Fukutani, Mr G. Robertson,
and Mr G. J. Gardner at NRC for their technical support.
of the polymers can be estimated using the empirical EHOMO
onset
¼
onset
ꢁ(Eox
+ 4.40) eV and ELUMO ¼ ꢁ(Ered
and Ered
+ 4.40) eV,
onset
onset
respectively, where Eox
are the onset potentials
for oxidation and reduction relative to the Ag quasi-reference
electrode, respectively.
References
Device fabrication and characterization
€
1 S. Gunes, H. Neugebauer and N. S. Sariciftci, Chem. Rev., 2007, 107,
1324.
The hole mobilities of P1 and P2 were measured using organic
ꢁ
2 B. C. Thompson and J. M. J. Frechet, Angew. Chem., Int. Ed., 2008,
47, 58.
field-effect transistor (OFET) measurements. A heavily doped
n-Si wafer with an overlayer of SiO2 (230 nm, Ci ¼ 15 nF cmꢁ2
)
3 G. Dennler, M. C. Scharber and C. Brabec, Adv. Mater., 2009, 21, 1323.
4 C. J. Brabec, S. Gowrisanker, J. J. M. Halls, D. Laird, S. Jia and
S. P. Williams, Adv. Mater., 2010, 22, 3839.
5 G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. J. Heeger, Science,
1995, 270, 1789.
was used as the substrate. With bottom contact configuration,
sputtered gold source and drain electrodes were used to contact
the polymer film. An interdigitated layout was used for the
OFET where the channel length and width are 20 mm and 1 mm,
respectively. A chloroform or 1,2-dichlorobenzene solution of
the desired polymer was spin cast to cover the source and drain
electrodes. The OFETs were characterized inside a dark box at
room temperature under nitrogen protection with a semi-
conductor parameter analyser (HP4145A). The hole mobilities
were calculated in the saturation regime at VDS ¼ ꢁ100 V using
the following equation IDS ¼ (W/2L)mCi(VG ꢁ VT),2 where VDS
is the source–drain voltage, IDS is the source–drain current,
W and L are, respectively, the channel width and length, m is the
field-effect mobility, Ci is the capacitance per unit area of the
SiO2 layer, and VG and VT are, respectively, the gate voltage and
threshold voltage.27,28
ꢁ
6 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.
7 C. Piliego, T. W. Holcombe, J. D. Douglas, C. H. Woo, P. M. Beaujuge
ꢁ
and J. M. J. Frechet, J. Am. Chem. Soc., 2010, 132, 7595.
8 J. Peet, J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses, A. J. Heeger
and G. C. Bazan, Nat. Mater., 2007, 6, 497.
9 J. K. Lee, W. L. Ma, C. J. Brabec, J. Yuen, J. S. Moon, J. Y. Kim,
K. Lee, G. C. Bazan and A. J. Heeger, J. Am. Chem. Soc., 2008,
130, 3619.
10 C. V. Hoven, X. D. Dang, R. C. Coffin, J. Peet, T. Q. Nguyen and
G. C. Bazan, Adv. Mater., 2010, 22, E63.
11 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.
12 Y. Liang, Z. Xu, J. Xia, S.-T. Tsai, Y. Wu, G. Li, C. Ray and L. Yu,
Adv. Mater., 2010, 22, E1.
€
13 M. C. Scharber, D. Muhlbacher, M. Koppe, P. Denk, C. Waldauf,
The BHJ solar cells were prepared on commercial glass slides
coated with patterned ITO. The thickness and sheet resistance of
the ITO are 150 nm and 12 U per square, respectively. The active
area of each solar cell device is 1.0 cm2 with a length : width ratio
of 4 : 1. The substrates were cleaned by sonicating sequentially in
detergent, DI water, acetone, and isopropanol. Immediately
prior to device fabrication, the substrates were treated in a
UV-ozone oven for 15 min. First, a poly(3,4-ethylenedioxy-
thiophene)–poly(styrene sulfonate) (PEDOT–PSS) thin film (30
nm) was spin-coated and then baked at 140 ꢂC for 15 min.
Secondly, an active layer was spin-coated on top of the PEDOT–
PSS from a chloroform or 1,2-dichlorobenzene solution of
polymer:PC61BM with a weight ratio of 1 : 2. In the case of the
devices using an additive, 1,8-diiodooctane with a 3% volume
ratio was added to the solutions before use. Finally, 2 nm of LiF
and 120 nm of Al were deposited on the top of the active layer in
a vacuum of 5 ꢃ 10ꢁ7 Torr to form the top electrode. The solar
cells (with no protective encapsulation) were then tested in air
under AM 1.5G irradiation of 100 mW cmꢁ2 (ScienceTech
Inc., SS 500W solar simulator) calibrated with a KG5 filter
covered silicon photovoltaic solar cell traced to the National
Renewable Energy Laboratory (NREL). Current–voltage (I–V)
A. J. Heeger and C. J. Brabec, Adv. Mater., 2006, 18, 789.
14 C. Shi, Y. Yao, Y. Yang and Q. Pei, J. Am. Chem. Soc., 2006, 128,
8980.
15 G. Daoust and M. Leclerc, Macromolecules, 1991, 24, 455.
16 J. Hou, H.-Y. Chen, S. Zhang, R. I. Chen, Y. Yang, Y. Wu and G. Li,
J. Am. Chem. Soc., 2009, 131, 15586.
17 H. Pan, Y. Li, Y. Wu, P. Liu, B. Ong, S. Zhu and G. Xu, Chem.
Mater., 2006, 18, 3237.
18 S. Bhanu and F. Scheinmann, J. Chem. Soc., Perkin Trans. 1, 1979,
1218.
19 G. Koeckelberghs, L. De Cremer, A. Persoons and T. Vrbiest,
Macromolecules, 2007, 40, 4173.
€
20 D. A. M. Egbe, L. H. Nguyen, H. Hoppe, D. Muhlbacher and
N. S. Sariciftci, Macromol. Rapid Commun., 2005, 26, 1389.
21 Z. K. Chen, W. Huang, L. H. Wang, E. T. Kang, B. J. Chen, C. S. Lee
and S. T. Lee, Macromolecules, 2000, 33, 9015.
22 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.
23 S. Wakim, B. R. Aich, Y. Tao and M. Leclerc, Polym. Rev., 2008, 48,
432.
24 H. Pan, Y. Li, Y. Wu, P. Liu, B. S. Ong, S. Zhu and G. Xu, J. Am.
Chem. Soc., 2007, 129, 4112.
25 I. Osaka, T. Abe, S. Shinamura, E. Miyazaki and K. Takimiya, J. Am.
Chem. Soc., 2010, 132, 5000.
26 K. M. Coakley and M. D. McGehee, Chem. Mater., 2004, 16, 4533.
27 V. Coropceanu, J. Cornil, D. A. Silva Filho, Y. Olivier, R. Silbey and
ꢁ
J. L. Bredas, Chem. Rev., 2007, 107, 926.
ꢁ
28 A. R. Murphy and J. M. J. Frechet, Chem. Rev., 2007, 107, 1066.
characteristics were recorded using
a
computer-controlled
10928 | J. Mater. Chem., 2011, 21, 10920–10928
This journal is ª The Royal Society of Chemistry 2011