Band structure engineering for low band gap polymers containing thienopyrazine

Article abstract of DOI:10.1039/c2jm12312f

In this research, we demonstrated that the energy levels, including highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO), and the optical absorptions of low band gap conjugated copolymers, consisting of 3-hexylthiophene (3HT) as electron-donating units and 2,3-diethylthieno[3,4-b]pyrazine (ETP) as electron-accepting units, could be systematically tuned by adjusting the composition and the geometric structure of the copolymers. Four new copolymers comprising of 3HT and ETP in different molar ratios (from 1:1 to 4:1), named P1-P4, were designed and synthesized via Suzuki coupling. The positions of the hexyl side chains on 3HT were varied to adjust the co-planarity of the copolymers. The LUMO, ranging from -2.94 eV to -3.11 eV, was lowered monotonically with increasing ETP content, and the break of co-planarity along the main chain showed a trivial effect on the LUMO. By contrast, the HOMO (-4.74 eV to -4.88 eV) was controlled by both the composition and the geometric structure of the copolymer. P2, having a twisted geometric structure, possessed a lower HOMO and a larger band gap compared to the planar P3. Bimodal optical absorptions with a relatively stronger absorption at long wavelengths were observed for all polymers which are due to intramolecular charge transfer. Optical band gaps in solution, ranging from 1.27 eV to 1.76 eV, decreased with increasing ETP content in the copolymer except P2 and the trend was consistent with the HOMO-LUMO gaps. The optical absorptions and the energy levels are further confirmed by theoretical calculations. Good co-planarity was also found to benefit the electrical conductivity of p-doped thin films. Our results suggested that tuning the composition and the geometric structure would be an effective molecular design strategy toward desired band structure for low band gap conjugated polymers based on thiophene and thienopyrazine derivatives.

Full text of DOI:10.1039/c2jm12312f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 7331
www.rsc.org/materials
PAPER
Band structure engineering for low band gap polymers containing
thienopyrazine†
a
Chi-Yang Chao, Chung-Hsiang Chao,a Lung-Pin Chen,a Ying-Chieh Hung,b Shiang-Tai Lin,b Wei-Fang Sua
*
and Ching-Fuh Linc
Received 24th May 2011, Accepted 6th February 2012
DOI: 10.1039/c2jm12312f
In this research, we demonstrated that the energy levels, including highest occupied molecular orbital
(HOMO) and lowest unoccupied molecular orbital (LUMO), and the optical absorptions of low band
gap conjugated copolymers, consisting of 3-hexylthiophene (3HT) as electron-donating units and 2,3-
diethylthieno[3,4-b]pyrazine (ETP) as electron-accepting units, could be systematically tuned by
adjusting the composition and the geometric structure of the copolymers. Four new copolymers
comprising of 3HT and ETP in different molar ratios (from 1 : 1 to 4 : 1), named P1–P4, were designed
and synthesized via Suzuki coupling. The positions of the hexyl side chains on 3HT were varied to
adjust the co-planarity of the copolymers. The LUMO, ranging from ꢀ2.94 eV to ꢀ3.11 eV, was
lowered monotonically with increasing ETP content, and the break of co-planarity along the main
chain showed a trivial effect on the LUMO. By contrast, the HOMO (ꢀ4.74 eV to ꢀ4.88 eV) was
controlled by both the composition and the geometric structure of the copolymer. P2, having a twisted
geometric structure, possessed a lower HOMO and a larger band gap compared to the planar P3.
Bimodal optical absorptions with a relatively stronger absorption at long wavelengths were observed
for all polymers which are due to intramolecular charge transfer. Optical band gaps in solution, ranging
from 1.27 eV to 1.76 eV, decreased with increasing ETP content in the copolymer except P2 and the
trend was consistent with the HOMO–LUMO gaps. The optical absorptions and the energy levels are
further confirmed by theoretical calculations. Good co-planarity was also found to benefit the electrical
conductivity of p-doped thin films. Our results suggested that tuning the composition and the geometric
structure would be an effective molecular design strategy toward desired band structure for low band
gap conjugated polymers based on thiophene and thienopyrazine derivatives.
transportation in the photoactive layer. The amount of solar
1. Introduction
photons absorbed is governed by the optical band gap of the
Conjugated polymers have been widely used for various appli-
polymer, generally dominated by the difference between the
cations including solar cells, light-emitting diodes (LEDs), and
highest occupied molecular orbital (HOMO) and the lowest
organic field-effect transistors. Bulk heterojunction solar cells
unoccupied molecular orbital (LUMO); therefore, low band gap
using these polymers as photoactive materials have attracted
conjugated polymers are inevitably needed to enhance light
extensive attention because of their potential to fabricate large-
harvesting. In addition, proper positioning of energy levels with
area, ultrathin, flexible devices at low cost through wet process-
respect to electron acceptors such as fullerene1 and titanium
ing. So far, the power conversion efficiency of polymer solar cells
dioxide2,3 is critical to the performance of the solar cell. The
is still below satisfactory owing to insufficient photon absorption
LUMO of the polymer is suggested to be 0.3–0.4 eV above the
as well as sub-optimal exciton dissociation and charge
LUMO of the acceptors to ensure effective charge separation.4,5
On the other hand, the low-lying HOMO of the polymer would
enlarge the discrepancy with respect to the LUMO of the electron
aDepartment of Materials Science and Engineering, National Taiwan
acceptor to increase the open circuit voltage (Voc) of the photo-
University, Taipei 10617, Taiwan. E-mail: cychao138@ntu.edu.tw; Fax:
voltaic devices.6,7 Therefore, the chemical structure of a success-
+886-2-23634562; Tel: +886-2-33663898
bDepartment of Chemical Engineering, National Taiwan University, Taipei
10617, Taiwan
cGraduate Institute of Photonics and Optoelectroics, National Taiwan
University, Taipei 10617, Taiwan
ful low band gap polymer must be carefully designed to achieve
optimized band structure.
One of the extensively studied strategies to achieve low band
gap is to employ electron-rich and electron-deficient units in
a polymer as the donor–acceptor (D–A) interaction could
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c2jm12312f
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 7331–7341 | 7331

View Article Online
increase intramolecular charge transfer (ICT) and decrease the
bond length alternation. In principle, both HOMO and LUMO
of the acceptor unit are lower than the corresponding levels of the
donor. The alternating D–A copolymer would exhibit a HOMO
governed by the donor and a LUMO dominated by the acceptor,
resulting in a lowered band gap compared to the homopolymer
of either the donor or the acceptor. While the band structure of
a D–A copolymer could be feasibly adjusted by incorporating
various donors and acceptors,8–25 there is a lack of systematic
design strategies for a polymer with desired energy levels due to
the vast structure diversity of the repeating units and essentially
unlimited number of possible D–A combinations.
In this study, a series of conjugated copolymers consisting of
3-hexylthiophene (3HT) as electron-donating units and 2,3-
diethylthieno[3,4-b]pyrazine (ETP) as electron-accepting units in
different ratios were synthesized via Suzuki coupling to adjust the
composition and to evaluate the corresponding effect on the
energy levels. Each thiophene ring contains a hexyl side chain to
ensure good solubility, which also allows for the acquirement of
electrochemical and absorption properties of the copolymers as
an individual molecule for systematic comparison. The positions
of the hexyl groups on 3HT units were intentionally varied to
alter the geometric structure. The chemical structures, molecular
weights as well as the optical and electrochemical properties were
characterized for the synthesized copolymers. Quantum chemical
calculations were performed to obtain the simulated optical and
energy levels. The results are applied to investigate the signifi-
cance of the chemical composition and the geometric structure
affecting the band structures of these copolymers. This study
demonstrates a feasible and systematic molecular design strategy
for TP-based low band gap polymers toward optimal HOMO,
LUMO and optical absorption.
Another class of non-typical D–A systems consists of one unit
whose HOMO is lower and LUMO is higher than the corre-
sponding levels of the other unit. One unique feature of such
non-typical D–A systems is that the energy levels can be
systematically tuned by adjusting the composition of the D/A
repeating units. One outstanding example is the copolymers
containing thienopyrazine (TP) or its derivatives. TP derivatives
are among the mostly used electron accepting units in D–A
systems for their strong electron withdrawing feature and
capability to stabilize the quinoid form to significantly lower the
LUMO to suppress the band gap.26–33 The band gaps of the D–
A copolymers comprising of TP derivatives are larger than that
of the TP homopolymer but smaller than that of the corre-
sponding homopolymer of donors such as thiophene and fluo-
rene. Therefore, optimization of the band structure for suitable
energy levels and band gap could be accomplished by adjusting
the composition of the TP based copolymers. Cheng et al.
synthesized a series of copolymers comprising thiophene (T) as
electron donor and di-alkyl substituted thienopyrazine (TP) as
electron acceptor in different molar ratios ranging from 1 : 1 to
4 : 1.26 The band gap was as low as 1.1 eV for the copolymer
with T : TP ¼ 1 : 1. However, the lack of side substituent on
thiophene resulted in low molecular weight and insolubility of
the copolymers with T : TP greater than 3 : 1. Jenekhe et al. also
prepared copolymers based on 3-dodecylthiophene and TP
without side substituent; and the optical band gap was ꢁ1.3 eV
for the copolymer with T : TP ¼ 2 : 1. Nevertheless, the
copolymer with T : TP ¼ 3 : 1 was also found to be insoluble.27
Although the combination of TP derivatives and thiophene (T)
(or alkylthiophene (AT)) was demonstrated to exhibit low
optical band gaps, which could be varied by changing the molar
ratio between T (or AT) and TP units, the insolubility prevented
further understanding of the band structures and applicability
to photovoltaic devices. Janssen et al. prepared copolymers with
T : TP ¼ 4 : 1 or 5 : 1 by alternatively coupling TP units and
either quaterthiophene or quinquethiophene segments having
two alkyl side chains in different positions.28 The optical band
gaps of these copolymers in solution were 1.5 eV to 1.6 eV,
larger than those containing fewer thiophene units.29–31 The
position of side chains was also shown to affect the band gap
and the current density generated in the photovoltaic devices.
According to these reports, the band gap of a conjugated
polymer could be affected by its chemical composition and
geometric structure. Therefore, the influence of these factors on
the energy levels and optical absorption of the polymer should
be critical to band gap engineering toward optimal solar cell
performance.
2. Experimental
2.1 Materials
Thiophene, N-bromosuccinimide (NBS), 1-bromohexane, 1-bro-
mooctane, 3-bromothiophene, and Aliquat336 were purchased
from Acros Organics. 4,40-Di-tert-butyl-2,20-dipyridyl (dtbpy), 2-
isopropoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (ip-Bpin),
pinacolborane (HBPin), [1,3-bis(diphenylphosphino)propane]
dichloronickel (Ni(dppp)Cl2), and tetrakis(triphenylphosphine)
palladium(Pd(PPh3)4) wereobtained fromAldrichChemicals. Di-
m-methoxobis(1,5-cyclooctadiene) diiridium ([Ir(OMe)(COD)]2)
was purchased from Strem Chemicals. Tetrahydrofuran (THF)
was dried over sodium/benzophenone and freshly distilled prior to
use. Other materials were used as received. 3-Hexylthiophene
(3HT) was synthesized according to the literature.34
2.2 Measurement and characterization
All new compounds were characterized by an Electron Impact-
1
Mass Spectrophotometer (EI-MS, Finnigan TSQ 700) and H
NMR using a Bruker AVANCE 400 MHz spectrometer with d-
chloroform as solvent. Molecular weights and polydispersity
indices (PDI) of the polymers were determined by gel permeation
chromatography (GPC) using a Viscotek module-350 system
with polystyrene as standard and THF or 1,2,4-trichlorobenzene
(TCB) as eluent. TGA measurement was performed on a Perkin-
Elmer TGA-7 apparatus at a heating rate of 20 C minꢀ1 under
ꢂ
nitrogen atmosphere. UV-vis absorption spectra were recorded
on a JASCO V570 spectrometer at ambient temperature. TCB
solutions with polymer concentrations less than 10ꢀ5 M and thin
films spin-coated from the TCB solutions were used for UV-vis
absorption measurements. Electrochemical cyclic voltammetry
(CV) was performed on polymer thin films coated on the working
electrode using a CH Instruments 611B potentiostat/galvanostat
system with Pt disk, Pt plate, and Ag/Ag+ electrode as the
working electrode, counter electrode, and reference electrode,
respectively, in a 0.1 mol Lꢀ1 tetrabutylammonium hexa-
fluorophosphate (Bu4NPF6) acetonitrile solution. The electrical
7332 | J. Mater. Chem., 2012, 22, 7331–7341
This journal is ª The Royal Society of Chemistry 2012

View Article Online
conductivity of pristine and doped polymers was obtained from
four-probe in-plane resistivity measurement on the spin-coated
thin films using a Mitsubishi MCP-T410 electrometer. The
doping process was performed by immersing the spin-coated thin
films in a 1 M KI3 solution for 24 h and the doped thin films were
washed thoroughly with copious DI water and dried under
vacuum before measurement. Polymer solar cells (PSCs) with an
inverted structure were fabricated using ITO glass as the positive
electrode and Ag as the negative electrode. The ITO substrate
was first covered with ZnO, then the photoactive layer was
deposited by spin coating a TCB solution of polymer and
PC71BM with designated weight ratio. NiO was then deposited
on top of the photoactive layer and followed by the deposition of
Ag by vacuum evaporation at 3 ꢃ 10ꢀ4 Pa. The current–voltage
measurement of the devices was conducted on a computer-
controlled Keithley 2400 Sourcemeter unit. A xenon lamp with
an AM 1.5 filter was used as the white-light source, and the
diaphanous compound, M1, was obtained by vacuum distillation
under 1 ꢃ 10ꢀ3 torr. Yield ¼ 75%.
1H NMR (CDCl3, ppm): d 7.49 (s, 1H), 2.85 (t, J ¼ 1.9 Hz,
2H), 1.58 (m, 2H), 1.31 (m, 30H), 0.88 (t, J ¼ 1.7 Hz, 3H).
EI-MS: m/z ¼ 420.3 (M+), calculated formula weight ¼ 420.22.
EA: calculated for C, 62.88; H, 9.11; S, 7.63. Found for C,
62.78; H, 9.40; S, 7.04%.
3-Hexyl-2-(3-hexylthiophen-2-yl)thiophene (B3HT). Grignard
reagent was prepared by adding 2-bromo-3-hexylthiophene (1.0
g, 4.05 mmol) to a stirred suspension of magnesium powder
(0.295 g, 12.1 mmol) in THF under argon. The resulting mixture
was refluxed for 12 h. The Grignard reagent was then added via
a cannula to a mixture of 2-bromo-3-hexylthiophene (1.5 g, 6.07
mmol) and Ni(dppp)Cl2 (44 mg, 0.081 mmol) in anhydrous
THF and the reaction mixture was refluxed for 24 h under argon.
The crude product was purified by column chromatography
using hexane as eluent and then by vacuum distillation at 200 ꢂC
under 1 ꢃ 10ꢀ3 torr to give the diaphanous product of B3HT.
Yield ¼ 30%.
optical power at the sample was 100 mW cmꢀ2
.
2.3 Synthesis of monomers
1H NMR (CDCl3, ppm): d 7.27 (d, J ¼ 2.1 Hz, 2H), 6.97 (d,
J ¼ 2.1 Hz, 2H), 2.49 (t, J ¼ 1.9 Hz, 4H), 1.55 (m, 4H), 1.25
(m, 12H), 0.85 (t, J ¼ 1.7 Hz, 6H).
The synthetic routes of monomers are outlined in Scheme 1.
2-(3-Hexyl-5-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)thi-
ophen-2-yl)-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (M1). In a 25
mL three-neck flask, [Ir(OMe)(COD)]2 (0.059 g, 0.089 mmol),
HBPin (1.52 g, 11.88 mmol) were mixed in 10 mL dry n-hexane
and stirred at room temperature. To the mixture, dtbpy (0.048 g,
0.178 mmol) dissolved in dry n-hexane was then added drop wise.
Afterward, 3-hexylthiophene (1 g, 5.94 mmol) was added, and
the resulting mixture was refluxed for 24 h. The crude product
was purified by column chromatography using a co-solvent
2-(4-Hexyl-5-(3-hexyl-5-(4,4,5,5-tetramethyl-1,3,2-dioxabor-
olan-2-yl)-thiophene-2-yl)thiophen-2-yl)-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane (M2). [Ir(OMe)(COD)]2 (0.03 g, 0.045 mmol) and
HBPin (0.765 g, 5.98 mmol) were mixed in 5 mL dry n-hexane
and stirred at room temperature. To the above mixture, dtbpy
(0.024 g, 0.089 mmol) dissolved in dry n-hexane was added drop
wise. Afterward, 3-hexyl-2-(3-hexylthiophen-2-yl)thiophene (1 g,
2.99 mmol) was added, and the resulting mixture was refluxed for
24 h. The crude product was purified by column chromatography
(ethyl acetate/hexane
¼
1/10 in v/v) and subsequently
Scheme 1 Syntheses of monomers.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 7331–7341 | 7333

View Article Online
using a co-solvent (ethyl acetate/hexane ¼ 1/10) and precipitation
in methanol. The precipitates were collected by filtration and
then vacuum dried to afford M2 as white powders. Yield ¼ 80%.
1H NMR (CDCl3, ppm): d 7.50 (s, 2H), 2.49 (t, J ¼ 2.0 Hz,
4H), 1.55 (m, 4H), 1.34 (s, 24H), 1.22 (m, 12H), 0.83 (t, J ¼ 1.7
Hz, 6H).
EI-MS: m/z ¼ 586.6 (M+), calculated formula weight ¼ 586.5.
EA: calculated for C, 65.33; H, 8.94; S, 10.93. Found for C,
65.52; H, 8.88; S, 10.26%.
comprising of anhydrous toluene and 2 M aqueous potassium
carbonate in a 2/1 (v/v) ratio. Aliquat336 (2 mol%) and cata-
lyst Pd(PPh3)4 (1 mol%) were added to the above mixture
under argon and the resulting mixture was stirred for 72 h at
80 ꢂC. The reaction mixture was then cooled to room
temperature and slowly added to a vigorously stirred mixture
of MeOH/H2O (v/v ¼ 1/1) to afford precipitates as crude
product. The solid was further purified by Soxhlet extraction
with acetone and diethyl ether to remove oligomers and cata-
lyst residues and was dried under reduced pressure at room
temperature.
5,7-Dibromo-2,3-diethylthieno[3,4-b]pyrazine (M3). 2,3-Dieth-
ylthieno[3,4-b]pyrazine (ETP) was prepared according to the
literature.35 The bromination of ETP by NBS afforded M3.
Yield ¼ 87%.
P1 was synthesized by reacting M1 and M3 to form greenish
1
black solid. H NMR (CDCl3, ppm): d 6.90–7.0 (m, 1H), 2.85–
3.02 (m, 6H), 1.90–2.0 (m, 2H), 1.15–1.40 (m, 12H), 0.80–0.90
(m, 3H). P2, a purplish black solid, was obtained by reacting M2
and M3. 1H NMR (d-THF, ppm): d 6.95–7.03 (m, 2H), 2.90–3.0
(m, 4H), 2.50–2.70 (m, 4H), 1.20–1.55 (m, 22H), 0.75–0.90 (m,
6H). P3, a murrey polymer, was synthesized by reacting M1 and
M4. 1H NMR (d-THF, ppm): d 7.03–7.18 (m, 3H), 2.85–3.10 (m,
8H), 2.58–2.63 (m, 2H), 1.74–1.82 (m, 4H), 1.20–1.60 (m, 26H),
0.80–0.92 (m, 9H). Dark red P4 was synthesized using M2 and
M4. 1H NMR (d-THF, ppm): d 7.05–7.23 (m, 4H), 2.88–3.10 (m,
8H), 2.55–2.63 (m, 4H), 1.76–1.83 (m, 4H), 1.20–1.60 (m, 34H),
0.80–0.95 (m, 12H).
1H NMR (CDCl3, ppm): d 2.96, 2.94 (q, J ¼ 1.8 Hz, 4H), 1.39
(t, J ¼ 1.8 Hz, 6H).
EI-MS: m/z ¼ 350.1 (M+), theoretical formula weight ¼
350.07.
EA: calculated for C, 34.31; H, 2.88; S, 9.16. N, 8.00. Found
for C, 34.65; H, 2.95; S, 10.05, N, 8.22%.
2,3-Diethyl-5,7-bis(3-hexylthiophen-2-yl)thieno[3,4-b]pyrazine
(3HT-ETP-3HT). The Grignard reagent of 2-bromo-3-hexylth-
iophene was prepared by adding 2-bromo-3-hexylthiophene
(5.65 g, 22.9 mmol) to a stirred suspension of magnesium
powder (0.83 g, 34.3 mmol) in THF under argon. The resulting
mixture was refluxed for 12 h and then added via a cannula to
a mixture of Br-ETP-Br (2 g, 5.71 mmol) and [1,3-bis(diphe-
nylphosphino)propane]dichloronickel (Ni(dppp)Cl2) (774 mg,
1.43 mmol) in anhydrous THF. The reaction was carried out at
room temperature for 1 day under argon. The crude product
was purified by column chromatography using a co-solvent
(ethyl acetate/hexane ¼ 1/15) and subsequently by recrystalli-
zation from methanol/methylene chloride gave dark red needles.
Yield ¼ 25%.
2.5 Theoretical calculation
The energy levels and optical properties of the polymers are also
determined based on quantum chemical calculations using
commercial package Gaussian 03.36 The ground state structures
were optimized by using density functional theory (DFT) with
the B3LYP functional and the 6-31G* basis set.37 It has been
shown that B3LYP/6-31G* gives decent ground state structures
of conjugated polymers. HOMO/LUMO energies and band gap
are determined from single point energy calculations performed
at B3LYP/6-31 + G* level. The calculations were performed for
oligomers with the degree of polymerization of 1 to 4. The
properties of the corresponding polymer were determined by
linear extrapolation (using the inverse of the degree of poly-
merization) either to the degree of polymerization measured in
experiment or to infinite. Optical absorption spectra of each
polymer were represented by Gaussian line-broadening using the
equation38
1H NMR (CDCl3, ppm): d 7.36 (d, J ¼ 1.3 Hz, 2H), 7.01 (d,
J ¼ 1.3 Hz, 2H), 2.95 (m, 8H), 1.72 (m, 4H), 1.25–1.5 (m, 18H),
0.88 (t, J ¼ 1.8 Hz, 6H).
EI-MS: m/z ¼ 524.3 (M+), calculated formula weight ¼ 524.85.
5,7-Bis(5-bromo-3-hexylthiophen-2-yl)-2,3-diethylthieno[3,4-b]
pyrazine (M4). To a THF solution of 3HT-ETP-3HT (0.96 g,
1.77 mmol), NBS (0.66 g, 3.71 mmol) in THF was added drop
wise in the dark and the resulting solution was stirred at room
temperature for 3 h. The reddish product was purified by column
2
X
1
ꢀðwn ꢀ wÞ
pffiffiffiffiffiffi
AðwÞ ¼
hn exp
chromatography and recrystallization following
a similar
s2
s
2p
n
procedure as for the purification of 3HT-ETP-3HT and murrey
needles were obtained. Yield ¼ 78%.
where s is FWHM (full width at half maximum) of the UV-vis
absorption peak and un and hn are the absorption length and
oscillator strength of the transition obtained from semiempirical
ZINDO39 calculations in which only singlet states were consid-
ered. The value of s ¼ 300 is chosen to spread the calculated
ZINDO UV-vis spectrum. To investigate the effect of geometric
structure on the energy levels of the polymers, we design two
categories of copolymer (without and with side chain on thio-
phene) with different ratios of T : TP. For saving computational
times, we replaced the hexyl substituent on 3-hexylthiophene
with a methyl substituent and neglected the ethyl substituents
on TP.
1H NMR (CDCl3, ppm): d 6.95 (s, 2H), 2.98, 2.96 (q, J ¼ 1.8
Hz, 4H), 2.89 (t, J ¼ 2.0 Hz, 4H), 1.70 (p, J ¼ 1.8 Hz, 4H), 1.25–
1.5 (m, 18H), 0.89 (t, J ¼ 1.8 Hz, 6H).
EI-MS: m/z ¼ 682.4 (M+), calculated formula weight ¼ 682.64.
EA: calculated for C, 52.78; H, 5.61; S, 14.09; N, 4.1. Found
for C, 53.09; H, 5.67; S, 13.20, N, 4.01%.
2.4 General procedure for Suzuki coupling
Equivalent amounts of dibrominated monomer (M3 or M4)
and diborated monomer (M1 or M2) were added to a mixture
7334 | J. Mater. Chem., 2012, 22, 7331–7341
This journal is ª The Royal Society of Chemistry 2012

View Article Online
catalyst was successful in boration of both 2,5-position of
3-hexylthiophene in high yield (75%) due to very distinctive
reactivity in 2- and 5-positions.
3. Results and discussion
3.1 Design and synthesis of copolymers
All the polymers synthesized could be dissolved in 1,2,4-tri-
chlorobenznene (TCB). With increasing 3HT content, the
copolymers, such as P3 and P4, exhibited good solubility in
chloroform and tetrahydrofuran, indicating that the hexyl side
chains on thiophene would substantially enhance the solubility.
Table 1 lists the weight average molecular weight (Mw) and the
polydispersity index (PDI) of the copolymers obtained from high
temperature gel permeation chromatography (GPC) operated at
130 ꢂC using TCB as eluent to ensure thorough dissolving of the
polymers. All the synthetic polymers exhibited moderate
molecular weights (<10 000), suggesting that Suzuki coupling
using thiophene-based borated monomers (M1 and M2) might
not be very effective. It was found that P1 and P2 synthesized
from M3 with Br on ETP exhibited relative low molecular
weight, while P3 and P4 using M4 with Br on 3HT showed
enlarged numbers of repeating units. This might be attributed to
weak reactivity of Br on ETP in M3 due to the presence of strong
electron withdrawing nitrogen. Stille coupling, employing tri-
methyl-tin decorated 3HT and B3HT to replace borated M1 and
M2, was performed as reference studies. The resulting P1 and P2
still exhibited relative low molecular weights (<2500), while
apparently high molecular weights (>10 000) were observed for
the corresponding P3 and P4. These observations further
A series of copolymers containing 3HT and ETP (P1–P4) in
different ratios were designed and synthesized in this work to
systematically vary the compositions and the geometric struc-
tures. Thus, their energy levels and optical absorptions can be
adjusted. Hexyl side chain is located on every thiophene unit and
di-ethyl side groups are present on TP to ensure good solubility
of the copolymers. The synthetic routes of monomers are out-
lined in Scheme 1. The syntheses and the chemical structures of
the synthetic copolymers as well as the chemical structures of the
analogous polymers for theoretical calculations are illustrated in
Scheme 2. Note that the polymers named as P1M–P4M and
PMT indicated that the 3-hexylthiophene units are replaced with
3-methylthiophene (MT) in the calculations, while all the side
chains are removed for P1X–P4X and PT.
Different from the syntheses of most thiophene based copol-
ymers via Stille coupling, Suzuki coupling was employed for
environmental benignity by avoiding the use of tin-based
reagents. The chemical structures and purity of all monomers
1
were verified by H NMR, Electron Impact-Mass Spectropho-
tometer (EI-MS) and elemental analysis (EA). To the best of our
knowledge, di-borated 3-alkylthiophene (M1) was first synthe-
sized in this work. Although many ligands and catalysts for
boration have been reported,31,40–42 only the iridium-based
Scheme 2 Syntheses of copolymers and the analogous copolymers for theoretical calculations.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 7331–7341 | 7335

View Article Online
Table 1 Molecular weights and thermal properties of the synthetic
polymers
and P4 which adopted M2 as a co-monomer should have
a twisted main chain. In addition, P4 would exhibit a better co-
planarity than P2 since the significant twist along the main chain
occurs with every 5 rings for P4 while that occurs with every 3
rings for P2.
3HT : ETP Number of Number
Polymers (molar ratio) repeat units of rings Mw
PDI Td/ꢂC
P1
P2
P3
P4
1 : 1
2 : 1
3 : 1
4 : 1
5
7
8
9
10
21
32
45
1600 1.51 318
3400 1.75 321
5500 2.21 370
7800 2.69 446
3.3 Optical properties
The UV-Vis absorption spectra of copolymers in TCB solution
and in solid state are shown in Fig. 2(a) and (b) respectively. All
the copolymers show bimodal absorptions, where ETP or 3HT-
ETP-3HT units correspond to the short wavelength absorptions
while the intramolecular charge transfer (ICT) contributed by
the donor–acceptor effect is responsible for the longer wave-
length absorptions. The lmax and the optical band gaps of
synthesized copolymers are summarized in Table 2. All the
copolymers show lower optical band gaps compared to pristine
P3HT.
confirmed the low reactivity of M3 and which should be
responsible for the low molecular weights of P1 and P2.
Although the molecular weights of the obtained polymers were
moderate, the optical absorptions and the energy levels of these
polymers should be applicable to estimate optical absorption
properties and energy levels of the corresponding polymers
possessing high molecular weight. The general tendencies for
HOMO and LUMO levels of the polymers with lower molecular
weights and those of the polymers with infinite degree of poly-
merization are in accordance, as shown in the theoretical calcu-
lation discussed later.
As seen in Fig. 2(a) illustrating the UV-Vis spectra in TCB
solution, the increment in ETP content leads to a red shift in
lmax, except P2 showing the smallest lmax. In addition, P1, P3
and P4 exhibit stronger absorptions at long wavelengths
whereas P2 possesses a more pronounced absorption at short
wavelengths. Also shown in Table 2, increasing the ETP
content in the copolymers reduces the energy gap despite that
Eg,opt of P3 is smaller than that of P2. In general, increasing
the ETP content is expected to enhance the ICT to lower the
band gap and to promote the absorptions at long wavelengths.
Therefore, the unexpected larger band gap and weaker red
absorptions of P2 are believed to be associated with the
significant break of co-planarity in P2, resulting in the
suppression of ICT.
All the copolymers exhibited a weight loss less than 5% upon
heating to 300 ꢂC as retrieved from TGA curves shown in
Fig. S1†, suggesting good thermal stability. In general, the
increment in 3HT content in the copolymer led to higher thermal
decomposition temperature Td, which might be attributed to the
high co-planarity of M4 and larger molecular weights resulting in
better conjugation along the main chain to improve the thermal
stability of P3 and P4.
3.2 Molecular modeling for geometric structures of copolymers
To evaluate the influence of the geometric structure on the energy
levels and optical absorptions of the synthetic copolymers P1–
P4, molecular modeling of the analogous P1M–P4M was per-
formed to retrieve the information of co-planarity of the main
chain. MT, BMT and MT-TP-MT were adopted for the
analogues of 3HT, B3HT and 3HT-ETP-3HT as the building
units respectively; while T, BT and T-TP-T were used for the
counterparts without side substitutes for comparison. The
optimized geometric structures of the repeat units (MT-TP)2,
(BMT-TP)1, (MT-(MT-TP-MT))1 and (BMT-(MT-TP-MT))1 of
P1M–P4M, as well as the corresponding units without methyl
group on thiophene of P1X–P4X are shown in Fig. 1. As seen in
Fig. 1(a) and (b), both (MT-TP)2 and (T-TP)2 show prefect
planar structures, indicating that the presence of a methyl group
in P1M should not affect the co-planarity. By contrast, owing to
the steric hindrance of the two methyl groups on BMT, the
co-planarity of (BMT-TP)1 between the two thiophene rings in
BMT is broken to result in the dihedral angle f2 ¼ 124.9ꢂ, as
shown in Fig. 1(c). Nevertheless, without the methyl group, the
calculated dihedral angle f2 of (BT-TP)1 is 179.9ꢂ (Fig. 1(d)).
Similar to (MT-TP)2 and (T-TP)2, the break of co-planarity is
neither observed in (MT-(MT-TP-MT))1 (Fig.1(e)) nor in (T-(T-
TP-T))1 (Fig. 1(f)). Meanwhile, (BMT-(MT-TP-MT))1 exhibits
a remarkable twist in BMT (Fig. 1(g)), whereas (BT-(T-TP-T))1
(Fig. 1(h)) retains a good co-planarity. From the geometric
structure studies, we could expect P1 and P3 employing M1 as
co-monomer to exhibit a strong co-planarity. By contrast, P2
Regarding the absorptions of thin films depicted in Fig. 2(b),
intermolecular energy transfer results in broader and red-shifted
absorption peaks, as compared with the absorptions in solution.
The red shifts of lmax are more provoked for P1 and P3 than
those for P2 and P4, suggesting that the former two might exhibit
more organized packing in the thin films than the latter two.
Therefore, we might infer that the good co-planarity in P1 and
P3 should lead to more ordered arrangement in solid state and
the irregular orientation of hexyl side chains would not severely
interrupt the inter-chain organization; while the break of co-
planarity in P2 and P4 originated from the B3HT unit could
result in less ordered packing despite the better control on the
orientation of hexyl side chains.
Fig. 3 shows the calculated UV-vis spectra by ZINDO calcu-
lations. Since the intensity of the UV-vis spectrum depends on
the chain length, the calculations were performed using polymers
with the same number of rings. We choose 20 rings for all the
polymers except P2M and P2X (21 rings). The calculated spectra
are generally in good agreement with experimental results
regarding the optical band gaps. In Fig. 3(a), increasing TP
content in P1X–P4X without methyl substituent on thiophene
results in a monotonic red shift, in agreement with the afore-
mentioned expectation. By contrast, the ICT absorption for
P2M with methyl substituent on thiophene shows a blue shift
compared to P3M in Fig. 3(b), in consistency with the measured
spectra for P2 and P3. The study of simulated spectra further
supports the previous argument stating that the particular large
7336 | J. Mater. Chem., 2012, 22, 7331–7341
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Fig. 1 Optimized structures at B3LYP/6-31G* of (a) (MT-TP)2, (b) (T-TP)2, (c) (BMT-TP)1, (d) (BT-TP)1, (e) (MT-(MT-TP-MT))1, (f) (T-(T-TP-T))1,
(g) (BMT-(MT-TP-MT))1, (h) (BT-(T-TP-T))1.
optical band gap of P2 should originate from the severe break of
co-planarity in P2.
systematically larger than Eg,opt probably due to the barriers for
charge transfer between the electrodes, the polymer films and the
supporting electrolyte, resulting in lags between voltage input
and current measurement.
3.4 Electrochemical properties and energy levels
The increment in ETP content would lower the LUMO
because of strong electron accepting feature of ETP enabling
feasible entrance of electrons into LUMO, which is in accordance
with the general agreement that LUMO should be mainly
determined by electron acceptor in a D–A copolymer.8,24,43–47
However, the HOMO is elevated with increasing ETP in the
copolymer, contradictory to the common statement declaring
that the HOMO should be governed by the electron donating
unit. Recently, Janssen et al. suggested that TP derivatives
should be both good electron acceptors and electron donors in
comparison with thiophene, allowing the particular small band
gap of thiophene–TP copolymers as a result of lowered LUMO
and lifted HOMO.48 It is also noticed that HOMO of P2 is lower
than HOMO of P3, as a consequence of the twisted main chain of
P2. Hence, we suggest that the position of HOMO should be
determined by both the TP content and the geometric co-
planarity whereas the level of LUMO is governed by the TP
content, which will be further confirmed by the theoretical
calculation discussed later. It is worthy to point out that the
The electrochemical properties of the polymers, investigated by
cyclic voltammetry (CV), are listed in Table 2. All copolymers
exhibited partial reversibility in both n-doping and p-doping
processes as shown in Fig. 4. From the values of Eox,onset and
Ered,onset, HOMO and LUMO as well as the electrochemical
band gap (Eg,cv) of the polymers could be obtained according to
the following equations:
HOMO ¼ ꢀ4.8 ꢀ Eox,onset
LUMO ¼ ꢀ4.8 ꢀ Ered,onset
Eg,cv ¼ LUMO–HOMO
The band gaps retrieved from the electrochemical method
exhibit a trend similar to the optical band gaps with respect to the
composition of the copolymer; nevertheless, the Eg,cv is
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 7331–7341 | 7337

View Article Online
Fig. 2 Normalized optical absorption spectra of copolymers (a) in TCB
solution and (b) in solid state.
Fig. 3 Simulated UV-vis absorptions for copolymers with the number of
rings ¼ 20 (21) by ZINDO calculation: (a) P1X–P4X and PT (represented
by T(20)) without methyl substituents on thiophene and (b) P1M–P4M
and PMT (represented by T(20))with methyl substituents on thiophene.
reduction in LUMO is more notable than the elevation of
HOMO; in addition, the small content of ETP (20 mol%) in P4
could significantly lower the LUMO but slightly lift the HOMO.
Table 3 summarizes the calculated properties of P1M–P4M
and P1X–P4X, including HOMO, LUMO, and band gaps (H–L
and ZINDO) based on the degree of polymerization of the cor-
responding synthetic polymers P1–P4. The general trends of the
calculated energy levels are in good agreement with the experi-
mental observations. The calculated HOMO–LUMO gaps are
consistently lower than the experimental band gaps from cyclic
voltammetry measurement by about 0.3 eV (0.34, 0.42, 0.33, 0.34
for P1M–P4M (MT : TP ¼ 4 : 1 to 1 : 1) respectively). The band
gaps obtained from ZINDO calculations are in general lower
than the calculated HOMO–LUMO gaps by about 0.2 eV (0.27,
0.21, 0.20, 0.24 for P1M–P4M (MT : TP ¼ 4 : 1 to 1 : 1)
respectively).
Fig. 5(a) shows the energy levels obtained from the theoretical
calculations for P1M–P4M and PMT as well as those from
experimental results of P1–P4 and P3HT. The degree of poly-
merization (DPn) of P1M–P4M and PMT used for calculation
are equal to the DPn of corresponding P1–P4 and P3HT, and
these DPn are denoted as experimental DPn. The LUMO levels of
P1M–P4M are found to decrease with the decrement of MT
content; while the HOMO levels show an irregular variation with
respect to the MT content. The changes of energy levels are in
concurrence with the experimental observation for P1–P4.
Fig. 5(b) shows the calculated energy levels of P1M–P4M and
Table 2 Optical and electrochemical properties of P1–P4
Optical absorption
Electrochemical property
In solution
Thin film
Oxidation
Reduction
Polymer
lmax/nm
Eg,opt/eV
lmax/nm
Eg,opt/eV
Eonset/V
HOMO/eV
Eonset/V
LUMO/eV
Eg,cv/eV
P1
P2
P3
P4
380, 690
381, 565
422, 635
425, 595
1.27
1.44
1.37
1.75
375, 730
387, 580
420, 690
423, 610
1.14
1.24
1.33
1.65
ꢀ0.06
0.02
ꢀ4.74
ꢀ4.82
ꢀ4.76
ꢀ4.88
ꢀ1.69
ꢀ1.74
ꢀ1.78
ꢀ1.86
ꢀ3.11
ꢀ3.06
ꢀ3.02
ꢀ2.94
1.63
1.76
1.74
1.94
ꢀ0.04
0.08
7338 | J. Mater. Chem., 2012, 22, 7331–7341
This journal is ª The Royal Society of Chemistry 2012

View Article Online
compared to P1X and P3X respectively while P2M and P4M
show HOMO similar to those of P2X and P4X despite that
HOMO of P1M–P4M are expected to be systematically higher
than those of P1X–P4X. In addition, the TP containing P4M
would even lower the HOMO as compared to PMT, while the
HOMO of P4X was higher than that of PT. Hence, the HOMO is
suggested to be elevated by increasing the TP content but it could
be significantly lowered by the break of co-planarity, resulting in
irregular variation of HOMO levels with respect to the 3HT (or
MT) contents. The effect of geometric structure could be also
applicable to explain the particular large band gap of P2 and
P2M as seen in Fig. 5(a).
In order to understand the origins for the significant influence
of break of co-planarity on HOMO, we calculated the frontier
orbitals of the copolymers to extract the distribution of p-elec-
trons at HOMO and LUMO. Fig. 6 illustrates the frontier
orbitals of P2M [(BMT-TP)7] and P3M [(MT-(MT-TP-MT))5].
For both molecules, the frontier orbitals of HOMO spread over
all the MT units and the thiophene ring of the TP uniformly, but
the charge density is more localized on the pyrazine ring of TP
for the LUMO orbitals. The localization of p-electrons on TP
would make the bonding character between two neighboring MT
the weakest among all the inter-ring bondings when the polymer
is at LUMO; whereas, the uniform distribution of p-electrons
along the main chain should lead to similar bonding characters in
all inter-ring bondings when the polymer is at HOMO. There-
fore, the deviation from co-planarity might strongly affect the
inter-ring conjugation to influence the HOMO; by contrast, the
change in the inter-ring dihedral angles should have little impact
on the LUMO. As a summary, the energy levels of the thio-
phene–TP system could be affected by both the composition and
the geometric structure, and the latter would exhibit a more
pronounced effect on the HOMO to enlarge the band gap.
Fig. 4 Cyclic voltammograms of the polymer thin films on platinum
electrode in a 0.1 mol Lꢀ1 Bu4NPF6 acetonitrile solution.
PMT with experimental DPn as well as those of P1M–P4M and
PMT with infinite DPn. The variations in HOMO and LUMO
with respect to the content of MT for the samples having relative
low molecular weight and those for high molecular weight
counterparts are in good consistency, and the discrepancy orig-
inating from the differences in chain lengths is small for P3M and
P4M. This observation suggested that the optical and electro-
chemical properties of P1–P4 with moderate molecular weights
should be applicable to describe the energy levels of the analo-
gous polymers with high molecular weight.
3.5 Electrical conductivity and photovoltaic performances
No electrical conductivity could be detected for all pristine
polymer thin films before doping, which should be reasonable
since these polymers are intrinsic semiconductors. The conduc-
tivities of the doped films were 1.76 ꢃ 10ꢀ5, 5.28 ꢃ 10ꢀ5, 9.6 ꢃ
10ꢀ4 and 8.9 ꢃ 10ꢀ4 S cmꢀ1 for P1, P2, P3 and P4 respectively.
The low conductivity of P1 and P2 could be attributed to the
relative low molecular weight, leading to short conjugation
lengths and poor p–p stacking as suggested by the limited red
shift of optical absorptions in solid state. Dramatic increment in
electrical conductivity was observed for P3 and P4 as compared
to P1 and P2 due to the increment in molecular weight. It was
noticed that the conductivity of P3 was higher than that of P4
despite that the chain length of P4 was longer, suggesting that the
good co-planarity in P3 should benefit the charge transportation
because of the elongation of conjugation and the improvement
in interchain packing according to the noticeable red shift in lmax
of P3.
Fig. 5(c) illustrates the calculated energy levels for P1M–P4M
and PMT with experimental DPn as well as those for P1X–P4X
and PT with experimental DPn. For P1X–P4X, both HOMO and
LUMO levels vary monotonically and the band gap decreases
with the increment of TP content. It is noticed that the polymers
containing methyl groups (P1M–P4M) possess LUMO system-
atically higher than the counterparts without methyl groups
(P1X–P4X) because of the electron-donating nature of methyl
group. Nevertheless, only P1M and P3M exhibit higher HOMO
Table 3 Calculated HOMO, LUMO and band gaps (H–L and ZINDO)
HOMO/ LUMO/ Eg (H–L)/ Eg (ZIN)/
Polymer
eV
eV
eV
eV
P1M [(MT-TP)5]
P1X [(T-TP)5]
P2M [(BMT-TP)7]
P2X [(BT-TP)7]
ꢀ4.57
ꢀ4.70
ꢀ4.63
ꢀ4.64
ꢀ3.28
ꢀ3.43
ꢀ3.20
ꢀ3.37
ꢀ3.12
ꢀ3.33
ꢀ3.02
ꢀ3.3
1.29
1.27
1.43
1.27
1.32
1.33
1.60
1.38
1.91
1.86
1.02
1.03
1.23
1.10
1.13
1.19
1.34
1.24
1.67
1.66
Bulk heterojunction (BHJ) photovoltaic cells based on P4 and
PC71BM in different weight ratios were fabricated and evaluated.
P4 was adopted for their low-lying HOMO and relative high
molecular weight among P1–P4. The devices were constructed in
an ITO/ZnO/polymer:PC71BM/NiO/Ag assembly as an inverted
structure for better stability. Open circuit voltage (Voc), short
P3M [(MT-(MT-TP-MT))8] ꢀ4.44
P3X [(T-(T-TP-T))8]
ꢀ4.66
P4M [(BMT-(MT-TP-MT))9] ꢀ4.62
P4X [(BT-(T-TP-T))9]
PMT [(MT)120
PT [(T)120
ꢀ4.69
ꢀ4.52
ꢀ4.76
]
ꢀ2.61
ꢀ2.90
]
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 7331–7341 | 7339

View Article Online
Fig. 5 Comparison of HOMO and LUMO energy levels for (a) P1–P4 and P3HT obtained from cyclic voltammograms as well as P1M–P4M and PMT
with experimental DPn from theoretical calculations; (b) P1M–P4M and PMT with different DPn from calculations; and (c) P1M–P4M and PMT with
experimental DPn as well as P1X–P4X and PT with experimental DPn from calculations.
accepting unit in different molar ratios ranging from 1 : 1 to 4 : 1
were synthesized via Suzuki coupling. The introduction of alkyl
side chains on each ring allowed all the polymers to exhibit
moderate to good solubility in common organic solvents.
Increasing 3HT content in the copolymer led to improvements in
molecular weights, solubility and thermal stability. Varying the
position of the hexyl side chain on 3HT would alter the geometric
structure of the copolymer and thus significantly affect the
HOMO and the inter-chain packing. The HOMO, ranging from
ꢀ4.74 eV to ꢀ4.88 eV for P1–P4, was elevated by increasing the
ETP content but was lowered by the break of the co-planarity.
The LUMO, ranging from ꢀ3.11.eV to ꢀ2.94 eV of P1–P4, was
elevated monotonically with increasing 3HT content. In
comparison with pristine P3HT, with only 20 mol% ETP incor-
porated in P4, a pronounced reduction (0.39 eV) in LUMO and
a slight elevation in HOMO (0.13 eV) could be achieved.
Bimodal UV-vis absorptions with a relative strong absorption at
Fig. 6 Frontier orbitals of P2M and P3M.
long wavelengths were observed for P1, P3 and P4 while the ICT
absorptions were weaker in P2. The optical band gap in solution,
ranging from 1.27 eV to 1.76 eV, increased with the increment of
3HT content in the copolymer except P2, which was consistent
with the variation of the HOMO–LUMO gaps. The particular
large band gap and the weaker ICT of P2 could be attributed to
the break of co-planarity in the geometric structure. Theoretical
energy levels and optical absorptions were calculated based
on density function theory with the B3LYP functional and the
6-31G* basis set were in good agreement with the experimental
results, further confirming that the deviation from co-planarity
along the main chain would significantly affect the HOMO but
would have trivial effect on the LUMO. Therefore, the energy
levels and the optical absorptions of the T-TP copolymers could
be systematically tuned by adjusting the composition and the
geometric structure of the copolymer.
Table 4 Photovoltaic properties of the cells using the blends of P4 and
PC71BM for active layer under the illumination of AM 1.5, 100 mW cmꢀ2
Weight ratio
of P4 : PC71BM
JSC/mA
cmꢀ2
VOC/V
FF (%)
PCE (%)
1 : 1
1 : 4
0.49
0.45
2.08
4.15
32.9
37.4
0.34
0.70
circuit current (Jsc), fill factor (FF), and power conversion effi-
ciencies (PCE) are summarized in Table 4. Voc for these P4 based
cells was 0.45–0.49 V, reasonably close to the energy difference
(ꢁ0.6 eV) between the HOMO of P4 and the LUMO of PC71BM.
Increasing the weight fraction of PC71BM in the photoactive
blend led to a slight decrease in Voc, while Jsc was significantly
elevated and a PCE of 0.7% was achieved. The unsatisfactory
performance might associate with the moderate molecular weight
of P4, resulting in low conductivity and poor film quality and
thus leading to low Jsc and FF.
Acknowledgements
The authors would like to thank National Science Council of
Taiwan for the financial support from NSC98-2218-E-002-002
and NSC99-2120-M002-011.
4. Conclusions
Four new low band gap conjugated copolymers P1–P4
comprising of 3-hexylthiophene (3HT) as the electron-donating
unit and 2,3-diethylthieno[3,4-b]pyrazine (ETP) as the electron-
References
1 Y. J. Cheng, S. H. Yang and C.-S. Hsu, Chem. Rev., 2009, 109, 5686.
7340 | J. Mater. Chem., 2012, 22, 7331–7341
This journal is ª The Royal Society of Chemistry 2012

View Article Online
2 Y. C. Huang, J. H. Hsu, Y. C. Liao, W. C. Yen, S. S. Li, S. T. Lin,
C. W. Chen and W. F. Su, J. Mater. Chem., 2011, 21, 4450.
3 T. W. Zeng, Y. Y. Lin, H. H. Lo, C. W. Chen, C. H. Chen, S. C. Liou,
H. Y. Huang and W. F. Su, Nanotechnology, 2006, 17, 5387.
4 L. J. A. Koster, V. D. Mihailetchi and P. W. M. Blom, Appl. Phys.
Lett., 2006, 88, 093511.
5 M. C. Scharber, D. Muhlbacher, M. Koppe, P. Denk, C. Waldauf,
A. J. Heeger and C. L. Brabec, Adv. Mater., 2006, 18, 789.
6 C. J. Brabec, A. Cravino, D. Meissner, N. S. Sariciftci, T. Fromherz,
M. T. Rispens, L. Sanchez and J. C. Hummelen, Adv. Funct. Mater.,
2001, 11, 374.
25 Y. Lee, T. P. Russell and W. H. Jo, Org. Electron., 2010, 11, 846.
26 K. F. Cheng, C. L. Liu and W. C. Chen, J. Polym. Sci., Part A: Polym.
Chem., 2007, 45, 5872.
27 Y. Zhu, R. D. Champion and S. A. Jenekhe, Macromolecules, 2006,
39, 8712.
28 A. P. Zoombelt, J. Gilot, M. A. Wienk and R. A. J. Janssen, Chem.
Mater., 2009, 21, 1663.
29 M. M. Wienk, M. G. R. Turbiez, M. P. Struijk, M. Fonrodona and
R. A. J. Janssen, Appl. Phys. Lett., 2006, 88, 153511.
30 A. P. Zoombelt, M. Fonrodona, M. G. R. Turbiez, M. M. Wienk and
R. A. J. Janssen, J. Mater. Chem., 2009, 19, 5336.
31 A. P. Zoombelt, M. A. M. Leenen, M. Fonrodona, Y. Nicolas,
M. M. Wienk and R. A. J. Janssen, Polymer, 2009, 50, 4564.
32 R. Mondal, N. Miyaki, H. A. Becerril, J. E. Norton, J. Parmer and
A. C. Mayer, et al., Chem. Mater., 2009, 21, 3618.
33 R. Mondal, H. A. Becerril, E. Verploegen, D. Kim, J. E. Norton,
S. Ko, N. Miyaki, S. Lee, M. F. Toney, J. L. Bredas,
M. D. McGehee and Z. N. Bao, J. Mater. Chem., 2010, 20, 5823.
34 B. Pal, W. C. Yen, J. S. Yang and W. F. Su, Macromolecules, 2007, 40,
8189.
35 D. D. Kenning, K. A. Mitchell, T. R. Calhoun, M. R. Funfar,
D. J. Sattler and S. C. Rasmussen, J. Org. Chem., 2002, 67, 9073.
36 M. J. T. Frisch, G. W. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, et al., Gaussian 03, Revision
C.02, Gaussian Inc, Wallingford CT, 2004.
37 B. C. Lin, C. P. Cheng and Z. P. M. Lao, J. Phys. Chem. A, 2003, 107,
5241.
38 S. Kilina, S. Ivanov and S. Tretiak, J. Am. Chem. Soc., 2009, 131,
7717.
7 A. Gadisa, M. Svensson, M. R. Andersson and O. Inganas, Appl.
Phys. Lett., 2004, 84, 1609.
8 R. Kroon, M. Lenes, J. C. Hummelen, P. W. M. Blom and B. De
Boer, Polym. Rev., 2008, 48, 531.
9 A. Dhanabalan, J. K. J. Van Duren, P. A. Van Hal, J. L. J. Van
Dongen and R. A. J. Janssen, Adv. Funct. Mater., 2001, 11, 255.
10 X. Wang, E. Perzon, J. L. Delgado, P. de la Cruz, F. Zhang, F. Langa,
M. Andersson and O. Inganas, Appl. Phys. Lett., 2004, 85, 5081.
11 O. Inganas, M. Svensson, F. Zhang, A. Gadisa, N. K. Persson and
X. Wang, et al., Appl. Phys. A: Mater. Sci. Process., 2004, 79, 31.
12 L. M. Campos, A. Tontcheva, S. Guenes, G. Sonmez, H. Neugebauer,
N. S. Sariciftci and F. Wudl, Chem. Mater., 2005, 17, 4031.
13 F. Zhang, W. Mammo, L. M. Andersson, S. Admassie,
M. R. Andersson and O. Inganas, Adv. Mater., 2006, 18, 2169.
14 D. Muhlbacher, M. Scharber, M. Morana, Z. Zhu, D. Waller,
R. Gaudiana and C. Brabec, Adv. Mater., 2006, 18, 2884.
15 M. H. Petersen, O. Hagemann, K. T. Nielsen, M. Jorgensen and
F. C. Krebs, Sol. Energy Mater. Sol. Cells, 2007, 91, 996.
16 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.
39 J. B. Foresman, M. Headgordon, J. A. Pople and M. J. Frisch, J.
Phys. Chem., 1992, 96, 135.
17 M. M. Wienk, M. Turbiez, J. Gilot and R. A. J. Janssen, Adv. Mater.,
2008, 20, 2556.
18 J. Hou, H. Y. Chen, S. Zhang, G. Li and Y. Yang, J. Am. Chem. Soc.,
2008, 130, 16144.
19 E. Zhou, M. Nakamura, T. Nishizawa, Y. Zhang, Q. Wei and
K. Tajima, et al., Macromolecules, 2008, 41, 8302.
20 Y. Liang, D. Feng, J. Guo, J. M. Szarko, C. Ray, L. X. Chen and
L. P. Yu, Macromolecules, 2009, 42, 1091.
40 E. Lim, Y. M. Kim, J. I. Lee, B. J. Jung, N. S. Cho, J. Lee, L. M. Do
and H. K. Shim, J. Polym. Sci., Part A: Polym. Chem., 2006, 44,
4709.
41 G. Tu, S. Massip, P. M. Oberhumer, X. He, R. H. Friend,
N. C. Greenham and W. T. S. Huck, J. Mater. Chem., 2010, 20, 9231.
42 J. Mei, N. C. Heston, S. V. Vasilyeva and J. R. Reynolds,
Macromolecules, 2009, 42, 1482.
43 A. Ajayaghosh, Chem. Soc. Rev., 2003, 32, 181.
44 H. A. M. van Mullekom, J. A. J. M. Vekemans, E. E. Havinga and
E. W. Meijer, Mater. Sci. Eng., R, 2001, 32, 1.
21 Y. Liang, Y. Wu, D. Feng, S. T. Tsai, H. J. Son and G. Li, et al., J.
Am. Chem. Soc., 2009, 131, 56.
22 E. Zhou, J. Cong, K. Tajima and K. Hashimoto, Chem. Mater., 2010,
22, 4890.
23 E. Zhou, J. Cong, S. Yamakawa, Q. Wei, M. Nakamura and
K. Tajima, et al., Macromolecules, 2010, 43, 2873.
45 W. C. Wu, C. L. Liu and W. C. Chen, Polymer, 2006, 47, 527.
46 E. Bundgaard and F. C. Krebs, Macromolecules, 2006, 39, 2823.
47 E. Bundgaard and F. C. Krebs, Sol. Energy Mater. Sol. Cells, 2007,
91, 1019.
24 C. H. Chen, C. H. Hsieh, M. Dubosc, Y. J. Cheng and C. S. Hsu,
Macromolecules, 2010, 43, 697.
48 B. P. Karsten, L. Viani, J. Gierschner, J. Cornil and R. A. J. Janssen,
J. Phys. Chem. A, 2009, 113, 10343.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 7331–7341 | 7341