Impact of the alkyl side chains on the optoelectronic properties of a series of photovoltaic low-band-gap copolymers

Article abstract of DOI:10.1021/ma102164c

The design of novel low-band-gap conjugated polymers with appropriate frontier orbital energy levels and good charge transport is needed to improve the conversion efficiency of organic photovoltaic devices. In this article, we describe the synthesis and structure-property relationships of a series of photovoltaic copolymers with a common conjugated backbone and differing solubilizing side chains. The copolymer optoelectronic properties and the related photovoltaic device performances are reported. Our results clearly show that the side chains have a major impact on the material and device properties. The electronic band gap can be varied by more than 0.3 eV, the charge mobilities by orders of magnitude, and the optimized fullerene content of photovoltaic devices by a factor of 4 by barely changing the side-chain positioning and/or by switching from linear to branched alkyl chains. A power conversion efficiency of 2.7% could be achieved with devices using the most promising polymer.

Full text of DOI:10.1021/ma102164c

Macromolecules 2010, 43, 9779–9786 9779
DOI: 10.1021/ma102164c
Impact of the Alkyl Side Chains on the Optoelectronic Properties of a
Series of Photovoltaic Low-Band-Gap Copolymers
Laure Biniek,† Sadiara Fall,‡ Christos L. Chochos,† Denis V. Anokhin,§ Dimitri A. Ivanov,§
,†
Nicolas Leclerc,* Patrick Leveque,* and Thomas Heiser
,‡
‡
ꢀ ^
†
ꢀ
ꢁ
ꢀ
Laboratoire d’Ingenierie des Polymeres pour les Hautes Technologies, Universite de Strasbourg, Ecole
‡
ꢀ
ꢁ
ꢀ
Europeenne de Chimie, Polymeres et Materiaux, 25 rue Becquerel, 67087 Strasbourg, France, Institut
d’Electronique du Solide et des Systeꢁmes, Laboratoire Commun UDS-CNRS, UMR 7163, 23 rue du Loess,
67037 Strasbourg, France, and §Institut de Sciences des Mateꢀriaux de Mulhouse (CNRS LRC 7228), 15 rue
Jean Starcky, B.P. 2488, 68057 Mulhouse Cedex, France
Received September 17, 2010; Revised Manuscript Received October 25, 2010
ABSTRACT: The design of novel low-band-gap conjugated polymers with appropriate frontier orbital
energy levels and good charge transport is needed to improve the conversion efficiency of organic
photovoltaic devices. In this article, we describe the synthesis and structure-property relationships of a
series of photovoltaic copolymers with a common conjugated backbone and differing solubilizing side chains.
The copolymer optoelectronic properties and the related photovoltaic device performances are reported. Our
results clearly show that the side chains have a major impact on the material and device properties. The
electronic band gap can be varied by more than 0.3 eV, the charge mobilities by orders of magnitude, and the
optimized fullerene content of photovoltaic devices by a factor of 4 by barely changing the side-chain
positioning and/or by switching from linear to branched alkyl chains. A power conversion efficiency of 2.7%
could be achieved with devices using the most promising polymer.
Introduction
optoelectronic properties.4,5 Moreover, in polymer/fullerene
blends, the side chains participate in the molecular interactions
Organic photovoltaic (OPV) devices based on blends of
π-conjugated polymers and fullerene derivatives have been the
focus of intense research activities over the past decade. The
device performances increase continuously, and energy conver-
sion efficiencies above 10% seem within reach for the near
future.1 Yet, the persistence of this progress depends crucially
on the development of new materials with improved optoelec-
tronic and physicochemical properties. In particular, the material
solar photon harvesting capacity, charge collection efficiency,
and related device open-circuit voltage could still be enhanced.
Since these properties are strongly correlated to the polymer
π-electron system, they can be controlled, at least to some extent,
by macromolecular engineering. The design of alternate copoly-
mers from specific conjugated monomers is a possible, well-
recognized route that can be followed to engineer the polymer
frontier orbitals and the related optoelectronic properties.2 Re-
cently, Lianget al.1 succeeded for instance toincreasethe polymer
ionization potential, while keeping constant the polymer band-
gap, by combining benzo[1,2-b:4,5-b0]dithiophene and thieno-
[3,4-b]thiophene units substituted with electron-withdrawing
functional groups. This allowed them to increase the open-circuit
voltage (Voc) and to reach energy conversion efficiencies above
7.4%. Another illustrative example is given by the insertion of
highly planar thieno[3,2-b]thiophene moieties into polythio-
phenes. Such a combination has led to the development of
polymers with record hole mobilities.3 The solubilizing side chains
on a given conjugated monomer may as well have a considerable
impact on the resulting material properties. The chain nature
(linear or ramified) and position not only fix the material solubility
in common solvents. They may also change the molecular
between the blend components and contribute to the resulting
blend morphology, an essential feature for charge generation and
transport.6 For instance, Mayer et al. observed recently that in
polymer/fullerene OPV devices fullerene intercalation among the
polymer side chains affects the optimal fullerene content (i.e., the
content necessary for achieving the highest energy conversion
efficiency).7 It is thus obvious that for a given conjugated polymer
backbone the nature and positioning of the side chains are crucial
to the polymer properties and need to be investigated thoroughly.
In a recent publication, we described the structure-property
relationships of a new copolymer family based on 2,1,3-ben-
zothiadiazole electron acceptor units and electron-rich thiophene
and thieno[3,2-b]thiophene units.8 The alternation of electron
donor (D) and acceptor (A) groups is known to lower the
polymer energy band gap and is therefore of interest for photo-
voltaic applications. The presence of thieno[3,2-b]thiophene
within the molecular backbone is believed to stabilize the HOMO
level and to planarize the molecule, leading to strong π-stacking
interactions.3,8,9 Photovoltaic devices using first members of this
copolymer family as electron donor and [6,6]-phenyl-C61-butyric
acid methyl ester (PCBM) as electron acceptor led to a promising
Voc of 0.74 V with a long wavelength absorption edge (related to
the polymer band gap) comparable to that of poly(3-hexy-
lthiophene) (P3HT). However, in spite of the increase in open-
circuit voltage, the energy conversion efficiency remained lower
than that of P3HT:PCBM-based reference devices. Low hole
mobilities and high optimal fullerene contents (1:4 copolymer/
fullerene weight ratio) are possible causes for this discrepancy. As
pointed out above, both properties are linked to the polymer side
chains in a nontrivial way. This led us to investigate further the
impact of the side-chain nature and positioning on the material
properties for this copolymer family.
*To whom correspondence should be addressed. E-mail: leclercn@
unistra.fr (N.L.); patrick.leveque@iness.c-strasbourg.fr (P.L.).
r 2010 American Chemical Society
Published on Web 11/11/2010
pubs.acs.org/Macromolecules

9780 Macromolecules, Vol. 43, No. 23, 2010
Biniek et al.
Scheme 1. Chemical Structures of the Different Copolymers Studied in
the Present Work
(300 MHz, CDCl3, ppm): δ = 7.26 (dd, 1H, 4J = 4.9 Hz, 3J = 3.0
Hz), 6.93 (m, 2H, 4J = 4.9 Hz), 2.59 (d, 2H, 3J = 6.8 Hz), 1.58 (m,
1H, 3J = 6.1 Hz), 1.29 (m, 10H), 0.90 (t, 6H, 3J = 4.9 Hz). 13C (75
MHz, CDCl3, ppm): δ = 143.40, 128.42, 125.15, 119.88, 32.04,
30.72, 30.44, 29.59, 29.51, 29.42, 22.82, 14.25.
2-(Trimethylstannyl)-4-alkylthiophene (2). Under an argon
atmosphere, 3-alkylthiophene (35.6 mmol) was dissolved in dry
THF (240 mL) and cooled to -78 °C. In another flask, under
argon, diisopropylamine (39.2 mmol) was solubilized in dry THF
(20 mL) cooled to -78 °C. Then nBuLi (37.4 mmol, 2.5 M in
hexane) was added dropwise, and the solution was kept at -78 °C
for 10 min, heated for 10 min at -20 °C, and then cooled again
to -78 °C. The lithium diisopropylamide so synthesized was
added slowly to the alkylthiophene solution, keeping the tempera-
ture below -60 °C. The cooling bath was removed, allowing the
temperature to reach 0 °C (ice bath). The solution was then cooled
back to -78 °C before the addition of trimethyltin chloride
(41 mmol, 1 M in THF). The mixture was stirred overnight and
allowed to reach room temperature. The mixture was quenched
with water and extracted with diethyl ether, and the organic phase
washed with water and dried over sodium sulfate. The solvent was
removed under reduced pressure and was further dried under high
vacuum, providing an oil containing a non-negligible amount of
nonreacted 3-alkylthiophene. This monostannylated compound
was used without further purification (yield: 80%).
In this work, we describe the synthesis and structure-property
relationships of a series ofcopolymers witha common conjugated
backbone but different solubilizing side chains. The copolymer
optoelectronic properties, in solution and in thin films, as well as
the related photovoltaic device performances are reported. Our
results clearly show that the side chains have a major impact on
the material and the related device properties. In particular, the
electronic band gap can be varied by more than 0.3 eV, the charge
mobilities by orders of magnitude, and the fullerene content of
optimized photovoltaic devices by a factor of 4 by barely
changing the side-chain positioning on the thiophene units and/
or by switching from linear to branched alkyl chains. A power
conversion efficiency of 2.7% could be achieved with devices
using the most promising polymer as electron donor material.
2-(Trimethylstannyl)-4-dodecylthiophene (2a). 1H NMR
(300 MHz, CDCl3, ppm):δ=7.22 (s, 1H), 7.04 (s, 1H), 2.67(t, 2 H,
3J = 7.4 Hz), 1.66 (m, 2H), 1.30 (m, 18H), 0.91(t, 3H, 3J = 6.3 Hz),
0.39 (s, 9H).
Experimental Section
2-(Trimethylstannyl)-4-(2-ethylhexyl)thiophene (2b). 1H NMR
(300 MHz, CDCl3, ppm): δ = 7.19 (s, 1H), 6.99 (s, 1H), 2.61 (d,
2 H, 3J = 6.8 Hz), 1.59 (m, 1H), 1.30 (m, 8H), 0.90 (t, 6H, 3J = 7.4
Hz), 0.38 (s, 9H).
Materials. The molecular structure, synthesis methodology,
and nomenclature of the investigated copolymers are presented
in Scheme 1. The acronym “PTBzT2” is used for all the polymers
and refers to a polymer (P) which is formed by the alternation of
a trimer (T), derived from a central 2,1,3-benzothiadiazole (Bz)
unit surrounded by two alkylthiophene units, and a thieno[3,2-
b]thiophene (T2) unit. The acronym extension highlights the
side-chain nature and positioning: C8, C12, and CEH corre-
spond to octyl, dodecyl, and 2-ethylhexyl alkyl chains respec-
tively, whereas the R and β indexes designate chains that are
either in the third position (R) or the fourth position (β).
The synthesis of 3-dodecylthiophene, 3-octylthiophene, 4,7-
dibromo-2,1,3-benzothiadiazole,10 and 2,5-bis(trimethylstannyl)-
thieno[3,2-b]thiophene3 have been done following procedures
reported in the literature. The 3-dodecylthiophene derivative and
the 4,7-bis(5-bromo-3-alkylthiophen-2-yl)-2,1,3-benzothiadiazole
synthesis have been described in our previous work.8 Tetrahydro-
furan (ACS grade), toluene (ACS grade), and diethyl ether (ACS
grade) were distilled over sodium while dichloromethane (ACS
grade) and acetonitrile were distilled over calcium hydride, prior to
use. All other chemicals were purchased from Aldrich and used
without further purification.
4,7-Bis(4-alkylthiophen-2-yl)-2,1,3-benzothiadiazole (3). General
procedure: 5-(trimethylstannyl)-3-alkylthiophene (21 mmol) and
4,7-dibromobenzo-2,1,3-thiadiazole (7 mmol) were dissolved in dry
toluene (100 mL). Then, Pd(PPh3)4 (0.5 mmol) was added, and the
reaction mixture was stirred at 110 °C for 24 h under an argon
atmosphere. Then, the reaction mixture was filtered through a pad
of Celite, and the toluene solution was evaporated and dried under
high vacuum. The crude product was purified by column chroma-
tography on silica gel with cyclohexane/toluene 9/1 as eluent to give
an orange compound (93%).
4,7-Bis(4-dodecylthiophen-2-yl)-2,1,3-benzothiadiazole (3a). 1H
NMR (300 MHz, CDCl3, ppm): δ = 7.99 (s, 1H), 7.85 (s, 1H),
7.06 (s, 1H), 2.71 (t, 2H, 3J = 7.8 Hz), 1.72 (m, 2H, 3J = 7.6 Hz),
3
1.28 (m, 18H), 0.9 (t, 3H, J = 6.8 Hz). 13C NMR (75 MHz,
CDCl3, ppm): δ = 152.65, 144.38, 139.00, 129.01, 126.04, 125.54,
121.53, 33.13, 31.92, 30.66, 30.51, 29.65, 29.62, 29.49, 29.36,
22.69, 14.12.
4,7-Bis(4-(2-ethylhexyl)thiophen-2-yl)-2,1,3-benzothiadiazole
(3b). 1H NMR (300 MHz, CDCl3, ppm): δ = 7.97 (s, 1H), 7.86(s,
1H), 7.04 (s, 1H), 2.66 (d, 2H, 3J = 6.9 Hz), 1.68 (m, 1H), 1.35 (m,
6H), 0.92 (t, 6H, 3J = 7.4 Hz). 13C NMR (75 MHz, CDCl3, ppm):
δ = 152.67, 143.06, 138.82, 129.48, 126.05, 125.49, 122.43, 40.33,
34.70, 32.53, 28.91, 25.68, 23.07, 14.14, 10.86.
NMR Characterization. 1H and 13C NMR spectra were
recorded on a Bruker 300 UltrashieldTM 300 MHz NMR spectro-
meter and a Bruker 400 UltrashieldTM 400 MHz NMR spectro-
meter, with an internal lock on the 2H signal of the solvent
(CDCl3).
Synthesis. 3-(2-Ethylhexyl)thiophene (1). To magnesium
turnings (5.81 g, 0.24 mol) in dry THF (5 mL) was added
2-ethylhexyl bromide (39 mL, 0.22 mol) in THF (55 mL) at a rate
sufficient to maintain reflux. After addition, the reflux was main-
tained for 2 h. The Grignard reagent was then added dropwise
through a canula to a solution of 3-bromothiophene (30.00 g,
0.18 mol) and Ni(dppp)Cl2 (1.84 mmol) in THF (110 mL). The
mixture was refluxed under argon and stirred overnight. Next, the
mixture was hydrolyzed by careful addition of water followed by
38% aqueous HCl and extracted with CH2Cl2. The brown organic
phase was washed with saturated sodium chloride and then with
water and dried over sodium sulfate. After removing the solvent
under reduced pressure, the crude product was distilled to provide
a colorless oil (bp 30 °C at 1.3 mbar) (26 g, yield: 73%). 1H NMR
4,7-Bis(5-bromo-4-alkylthiophen-2-yl)-2,1,3-benzothiadiazole
(4). General procedure: compound 3 (3.59 mmol) was solubi-
lized in DMF (60 mL) under argon in the dark. N-Bromosucci-
nimide (NBS) (7.35 mmol) was added portionwise. The resulting
solution was stirred at room temperature under argon over-
night. Water and diethyl ether were added, and the resulting
solution was stirred for 2 h. The organic phase was separated
from the water phase and extracted with brine (3 ꢀ 100 mL).
The organic phase was dried with sodium sulfate and filtered,
and the solvent evaporated under reduced pressure and further
dried under high vacuum. The crude product was purified by
column chromatography on silica gel with cyclohexane as
eluent, providing the compound 4 as an orange coumpound
(yield: 82%).

Article
Macromolecules, Vol. 43, No. 23, 2010 9781
Orange Solid C12β (4a). 1H NMR (300 MHz, CDCl3, ppm):
DSC. Differential scanning calorimetry analyses were per-
formed on a DSC 2910 apparatus from TA Instruments. The
δ = 7.78 (d, 2H, 5J = 5.5 Hz), 2.65 (t, 2H, 3J = 7.4 Hz), 1.69 (m,
3
2H), 1.28 (m, 18H), 0.89 (t, 3H, J = 6.9 Hz). 13C NMR (100
analyses were carried out under air at a heating rate of 10 °Cmin-1
.
MHz, CDCl3, ppm): δ = 152.27, 143.09, 138.51, 128.11, 125.36,
124.89, 111.61, 31.92, 29.76, 29.68, 29.59, 29.43, 29.35, 29.27,
22.68, 14.11. Elemental analysis results: C, 57.07; H, 6.85; S,
12.10; Br, 20.11; N, 3.52. Calcd for C38H54S3Br2N2: C, 57.42; H,
6.85; S, 12.10; Br, 20.11; N, 3.52.
Orange Oil EHβ (4b). 1H NMR (300 MHz, CDCl3, ppm): δ =
7.77 (d, 2H, 5J = 4.6 Hz), 2.60 (t, 2H, 3J = 7.2 Hz), 1.72 (m, 1H),
1.35 (m, 8H), 0.93 (t, 6H, 3J = 7.4 Hz). 13C NMR (100 MHz,
CDCl3, ppm): δ = 152.18, 142.22, 138.21, 128.58, 125.24,
124.75, 112.27, 40.01, 33.91, 32.52, 28.80, 25.74, 23.09, 14.14,
10.87. Elemental analysis results: C, 52.85; H, 5.73; S, 13.43; Br,
22.95; N, 4.06. Calcd for C30H38S3Br2N2: C, 52.78; H, 5.61; S,
14.1; Br, 23.41; N, 4.10.
TGA. Thermogravimetric analyses were performed on a Hi-Res
TGA 2950 apparatus from TA Instruments. The analyses were
carried out under a mixture of 75% N2 and 25% O2 at a heating rate
of 20 °C min-1 for PTBzT2-C12R and under air at a heating rate of
10 °C/min for PTBzT2-C12β and PTBzT2-CEHβ.
XRD. X-ray diffraction experiments were carried out on the
BM26B beamline14 at the European Synchrotron Radiation
Facility (ESRF) in Grenoble, France, with a Frelon CCD
camera. The films obtained directly from synthesis procedure
were measured both in normal and transverse directions. The
modulus of the scattering vector s (s = 2 sin θ/λ, where θ is the
Bragg angle and λ the wavelength) was calibrated using several
diffraction orders of silver behenate.
Polymer Synthesis General Procedure. 2-Trimethylstannylthieno-
[3,2-b]thiophene (1.36 mmol) and compound 4 (1.36 mmol) were
dissolved in dry toluene (70 mL). Then, tetrakis(triphenylphosphine)-
palladium(0) (Pd(PPh3)4) (0.02 equiv) was added, and the reaction
mixture was stirred at 110 °C under an argon atmosphere for 17 h.
The polymer was then precipitated in a solution of methanol,
filtered, and washed on a Soxhlet apparatus with methanol, cyclo-
hexane, and chloroform. The chloroform fraction was evaporated
under reduced pressure, and the polymer was precipitated in
methanol, filtered, and finally dried under high vacuum, providing
a film with a metallic shine.
Field-Effect-Mobility Measurements. Bottom contact field-
effect transistors (FETs) were elaborated on commercially
available prepatterned test structures whose source and drain
contacts were composed of 30 nm thick gold and 10 nm thick
indium tin oxide (ITO) bilayers. A 230 nm thick silicon oxide
was used as gate dielectric and n-doped (3 ꢀ 1017/cm3) silicon
crystal as gate electrode. The channel length and channel width
were 20 μm and 10 mm, respectively. The test structures were
cleaned in acetone and isopropyl alcohol and subsequently for
15 min in an ultraviolet ozone system. Then, hexamethyldisila-
zane (HMDS) was spin-coated (500 rpm for 5 s and then 4000
rpm for 50 s) under nitrogen ambient and followed by an
annealing step at 130 °C for 5 min. Finally, 4 mg/mL anhydrous
o-dichlorobenzene (o-DCB) polymer solutions were spin-coated
(1250 rpm for 60 s and 2000 rpm for 120 s) to complete the FET
devices. The samples were then left overnight under vacuum
(<10-6 mbar) to remove residual solvent traces. Both the FET
elaboration and characterizations were performed in nitrogen
ambient.
Preparation of Photovoltaic Devices. Bulk heterojunction
devices were elaborated using the PTBzT2 polymers as electron
donor and C60-PCBM as electron acceptor. The standard
device structure was the following: ITO/PEDOT:PSS(∼40
nm)/polymer:PCBM(∼100 nm)/Al(∼120 nm). Indium tin
oxide-coated glass with a surface resistance lower than 20 Ω/
sq was used as transparent substrate. ITO was cleaned sequen-
tially by ultrasonic treatments in acetone, isopropyl alcohol, and
deionized water. After an additional cleaning for 45 min under
ultraviolet generated ozone, a highly conductive polyethylene
dioxythiophene:polystyrenesulfonate PEDOT:PSS was spin-
coated (1500 rpm: 40 nm) from an aqueous solution and dried
for 30 min at 120 °C under vacuum before being transferred to
the nitrogen-filled glovebox. The dichlorobenzene polymer:
PCBM solutions were stirred at 70 °C for 48 h before spin-
coating. Unless otherwise stated, the solution had a total con-
centration of 40 mg/mL with a polymer:fullerene weight frac-
tion ranging from 1:1 to 1:4. Finally, a 120 nm thick aluminum
layer was thermally evaporated and used as cathode. The device
active area was 9 mm2, while each sample included four inde-
pendent diodes. With the most promising polymer, an extra
series of devices were elaborated with specific thermal annealing
conditions (130 °C for 15 min prior to cathode deposition) and
an additional LiF (0.5 nm) layer between the active layer and the
aluminum cathode. Current versus voltage (J-V) characteristics
were measured under darkness and under AM1.5 (100 mW/cm2)
illumination using an Oriel 150 W solar simulator.
PTBzT2-C12β. 1H NMR (300 MHz, CDCl3, ppm): δ = 7.92
(m, 1H), 7.80 (m, 1H), 7.33 (m, 1H), 2.76 (m, 2H), 1.62 (m, 2H)
1.27 (m, 18H), 0.87 (m, 3H).
PTBzT2-CEHβ. 1H NMR (300 MHz, CDCl3, ppm): δ = 7.97
(m, 1H), 7.83 (m, 1H), 7.39 (m, 1H), 2.85 (m, 2H), 1.81 (m, 1H),
1.34(m, 8H), 0.92 (m, 6H).
Computational Study. In order to get an additional insight
into the influence of the alkyl side chain nature and positioning
on the copolymer optoelectronic properties, density functional
theory (using Gaussian 03 package11 at the B3LYP/6-311þG*
level of theory in vacuum)12 was utilized to model the structural
and electronic properties of relevant molecular structures. In
particular, the HOMO and LUMO level positions and related
electron distributions were calculated on three repeating units.13
To keep the computational time within a reasonable range, the
alkyl chains were replaced by methyl groups. It is thereby
assumed that the electronic coupling between alkyl chains and
the π-electron system is negligible. CH3 groups were placed at
both ends, and the dihedral angle between the last carbon atom
and the methyl group was kept fixed in order to mimic the
rigidity of the actual polymer.
Materials Characterizations. UV-vis. UV-visible absorp-
tion spectroscopy measurements were done using a Shimadzu
UV-2101 spectrophotometer. The absorption spectra were mea-
sured on both o-DCB polymer solutions and polymer thin films
spin-coated on glass substrates.
GPC. Gel permeation chromatography measurements were
performed in trichlorobenzene (TCB) with a Waters Alliance
GPCV 2000 instrument (two PL-gel 5 mm mixed-C, a 5 mm
˚
100 A columns or a PLgel Olexis column); two detectors
(viscosimeter and refractometer) in trichlorobenzene (flow rate:
1 mL/min) at 150 °C. The system was calibrated with poly-
styrene standards using universal calibration.
Voltammetry. Cyclic voltammetry analyses were carried out
with a BioLogic VSP potentiostat using platinum electrodes at a
scan rate of 20 mV/s. The measurements were performed on
polymer thin films drop-casted from o-DCB solutions onto a
platinum working electrode. A Pt wire was used as counter
electrode and Ag/Agþ as reference electrode in a 0.1 mol L-1
solution of tetrabutylammonium perchlorate in acetonitrile.
Ferrocene was used as internal standard to convert the values
obtained with Ag/Agþ reference to the saturated calomel elec-
trode scale (SCE).
Results and Discussion
Materials Synthesis. Two new copolymers with an identical
conjugated backbone, composed of alternating thieno[3,2-b]-
thiophene and 4,7-bisthiophene-2,1,3-benzothiadiazole segments
functionalized with alkyl side chains of various types (linear and
branched) on the β position of the thiophene units, have been

9782 Macromolecules, Vol. 43, No. 23, 2010
Scheme 2. Synthesis of Monomers and Polymers
Biniek et al.
synthesized by Stille cross-coupling polycondensation (Scheme 2).
The syntheses of PTBzT2-C8R and PTBzT2-C12R (see
Scheme 1), where linear side chains of various lengths are
introduced on the R position of the thiophene rings, have been
published recently by us.8 In order to obtain the 4,7-bis(4-
alkylthiophene)-2,1,3-benzothiadiazole derivatives (3a, 3b)
( β position), selectively deprotonation of the 3-alkylthiophene
in the 5-position has been performed by the addition of 1 equiv
of the lithium diisopropylamide (LDA) base. The high basicity
and the steric hindrance of the LDA provide selectively the
2-(trimethylstannyl)-4-alkylthiophene derivatives (2a, 2b) after
addition of Me3SnCl. This synthetic approach is different than
the one followed for the synthesis of 2-(trimethylstannyl)-3-
alkylthiophene derivatives, which have been obtained by a
selective nucleophilic substitution of 3-alkylthiophene in the
2-position by n-BuLi.8 Then, the trimer-based derivatives (3a,
3b) were effectively synthesized by the Stille coupling between
either the 2a, 2b, and 4,7-dibromo-2,1,3-benzothiadiazole. Fi-
nally, the desired functional monomers 4a and 4b were provided
upon subsequently dibromination of 3a and 3b using N-bro-
mosuccinimide (NBS) as the bromine agent.
Table 1. Polymer Characterization
polymer
Mw (g/mol)a
Mn (g/mol)a
PDIa
Td (°C)b
PTBzT2-C12R
PTBzT2-C8R
PTBzT2-C12β
PTBzT2-CEHβ
23 000
46 000
55 000
32 000
16 000
30 000
14 000
12 000
1.4
1.5
3.9
2.7
352
304
277
307
a Determined by GPC with polystyrene as standard and TCB as
eluent. b Temperature with 5% weight loss determined by TGA at a scan
rate of 10 °C/min under air.
Molecular Characterizations. Because of the high tendency
of these conjugated copolymers to aggregate in room tempera-
ture solutions, their molecular weights had to be determined by
GPC experiments carried out at 150 °C in trichlorobenzene
solutions (see Table 1). Most polymers have a comparable
molecular weight distribution with a number-average molec-
ular weight (Mn) close to 15 kg/mol except for the PTBzT2-
C8R, whose Mn is twice as high.
Figure 1. Normalized UV-vis absorption spectra for all polymers in
o-DCB solution. The continuous line corresponds to PTBzT2-C8R, the
open squares to PTBzT2-C12R, the filled circles to PTBzT2-C12β, and the
crosses to PTBzT2-CEHβ.
All the polymers exhibit good thermal stability with 5%
weight-loss temperatures higher than 275 °C.
The DSC measurements were featureless up to 300 °C
(highest achievable temperature with our setup) for all the
polymers. It is likely that the pronounced backbone rigidity
as well as the strong interchain interactions induced by the
thieno[3,2-b]thiophene units may lead to melting tempera-
tures above this limit. A similar observation on thieno[3,2-b]-
thiophene containing copolymers has been reported before
by Liang et al.15
Properties of Polymers in Solution. The normalized
UV-vis absorption spectra of the copolymers in o-DCB
are presented in Figure 1, and the corresponding optoelec-
tronic properties are summarized in Table 3. Each polymer is
characterized by two major absorption peaks, a feature
which is commonly observed for alternating donor-acceptor
copolymers. The low-wavelength peak can be attributed to a
π-π* transition while the high-wavelength transition is
believed to be related to an intramolecular D-A charge
transfer.16 Increasing the chain length, from C8 to C12,
leaves constant the polymer absorption spectra. Also, repla-
cing C12 by CEH does not significantly perturb the low-
energy peak position. However, the high-energy peak is blue-
shifted by 26 nm. This shift may indicate a larger twist
between the thiophene and thieno[3,2-b]thiophene units
induced by to the more bulky CEH side chains. When changing
the side chains from the R to the β position, independent of
the side-chain nature, the low-energy peak is significantly
red-shifted while the high-energy peak remains almost un-
changed. These results suggest that the β position induces
less sterical twisting on the conjugated backbone and thereby
leads to a more pronounced coupling between the D and A
units, enhancing charge transfer and reducing the HOMO-
LUMO gap. The slightly higher red shift observed for
PTBzT2-C12β, in comparison to PTBzT2-CEHβ, supports

Article
Macromolecules, Vol. 43, No. 23, 2010 9783
Figure 2. Calculated HOMO (c, d) and LUMO (a, b) distributions for alkyl side chains in R (left: a, c) or β (right: b, d) position.
Scheme 3. Definition of the Calculated Dihedral Angles
Table 2. Calculated Parameters
alkyl
chain
position
R1
(deg)
R2
(deg)
R3
(deg)
EHOMO
(eV)
ELUMO
(eV)
R
β
50
1-2
50
1-2
14
14
-4.9
-4.6
-2.6
-2.8
Figure 3. Normalized UV-vis absorption spectra for spin-coated
films. The open squares correspond to PTBzT2-C12R, the filled circles
to PTBzT2-C12β, and the crosses to PTBzT2-CEHβ.
the assertion that the steric hindrance is lower for linear
chains. The pronounced high wavelength shoulder seen only
for PTBzT2-C12β may reveal the presence of small aggre-
gates in solution and indicates a lower polymer solubility in
comparison to PTBzT2-CEHβ.
in solution (2.1 and 1.7 eV for the R and β position,
respectively). This concurrence strengthens the consistency
of our model molecular structure.
Our density functional theory investigations done on
single molecules in vacuum corroborate these conclusions.
The molecular structures together with the calculated fron-
tier orbital electronic distributions are displayed in Figure 2.
The results can be considered as representative of the molecular
conformation in dilute solutions, where intermolecular π-π
interactions are negligible. The theoretical dihedral angles (see
Scheme 3) as well as the HOMO and LUMO levels are reported
in Table 2.
A highly twisted molecular backbone, with R1 and R2
values as high as 50°, are found with the alkyl chains on the R
position, while these angles are negligible for the β position.
Note however that, due to the longer alkyl chain synthesized
polymers, the actual dihedral angles are possibly somewhat
higher. The model also highlights the influence on the side-
chain positioning on the frontier orbital distributions. The
highest delocalization is found in β position molecules. The
resulting theoretical estimations of the HOMO-LUMO
energy gap (2.3 eV for the R position and 1.8 eV for the
β position) are close to the optical band gaps measured
Polymer Thin Film Optoelectronic Properties. Because of
the poor film quality of PTBzT2-C8R, only PTBzT2-C12R,
PTBzT2-C12β, and PTBzT2-CEHβ thin film UV-vis
absorption spectra were investigated. The results are shown
in Figure 3, and the corresponding peak positions and
optical band gaps (deduced from the long wavelength ab-
sorption edge) are reported in Table 3. Similarly to what has
been observed in solution, the absorption spectra are mostly
affected by the side-chain positioning, rather than by the
chain chemical structure (compare PTBzT2-C12R with
PTBzT2-C12β and PTBzT2-C12β with PTBzT2-CEHβ). All
the peaks are red-shifted when going from solution to the
solid state, with the most pronounced changes occurring for
the low-energy peak. Such a bathochromic shift is generally
attributed to π-stacking interactions and in our case points
out that these interactions are stronger for the polymers with
alkyl side chains in the β-position polymers. This is consis-
tent with the more planar molecular conformation of these
polymers deduced from the solution absorption data and
from our theoretical modeling (Table 2). Interestingly, the

9784 Macromolecules, Vol. 43, No. 23, 2010
Biniek et al.
Table 3. Absorption Maxima and Optical Band Gap on Thin Films
Table 5. Field-Effect Mobility Values
h (cm2/
(V s))
annealing
conditions
μ
h (cm2/
(V s))
polymer
λ
max sol (nm)
λ
max film (nm)
Eogpt (eV)
annealing
conditions
μ
polymer
PTBzT2-C8R
PTBzT2-C12R
PTBzT2-C12β
PTBzT2-CEHβ
405/513
407/510
419/577
393/561
PTBzT2-C12R
PTBzT2-C12β
PTBzT2-CEHβ
as cast
as cast
as cast
5 ꢀ 10-4
7 ꢀ 10-6
1 ꢀ 10-3
130 °C, 15 min
180 °C, 15 min
130 °C, 15 min
1 ꢀ 10-4
1 ꢀ 10-5
1 ꢀ 10-3
416/540
424/614
428/617
1.88
1.56
1.56
Table 4. Electrochemical Results
Finally, the polymer hole mobilities (μh) have been inves-
tigated in bottom-contact field-effect-transistors (FET) on
as-deposited as well as annealed thin film devices. The results
obtained in the saturation regime are summarized in Table 5.
Mobility values measured in the linear regime were in the
same range.
ox
red
onset
polymer
E
/HOMO (eV)
E
/LUMO (eV) Eegc (eV)
onset
PTBzT2-C12R
PTBzT2-C12β
PTBzT2-CEHβ
0.9/5.3
0.6/5.0
0.6/5.0
-1.2/3.2
-1.2/3.2
-1.3/3.1
2.1
1.8
1.9
Note that the hole mobility measured on PTBzT2-C12R is
one order of magnitude above our previously published
results which had been measured in air and on top-contact
FETs without an HMDS surface pretreatment. We may attri-
bute the improvement in mobility reported here to the better
film quality at the dielectric interface when using HMDS.8
The FET results clearly show that the side-chain position
can have a major influence on the charge transport, leading
to orders of magnitude variations in the field-effect mobility.
This result is consistent with the structural properties (i.e.,
π-stacking distances) discussed above. The 2 orders of magni-
tude difference in μh observed for both β-position polymers is
more surprising since both UV-vis and X-ray investigations led
to comparable π-stacking interactions and conjugation lengths.
However, the X-ray data obtained on the bulk samples (thick
films) show more regular ordering of PTBzT2-CEHβ chains
compared to polymers bearing linear side groups. Moreover,
the calculated crystalline density of PTBzT2-CEHβ (1.59 g/cm3)
is significantly higher than the density of PTBzT2-C12R and
PTBzT2-C12β crystal (1.33 and 1.26 g/cm3, respectively). It is
documented that the charge mobility is crucially dependent not
only on the π-stacking distance but also on the intermolecular
distance in the direction normal to the π-stacking.18 Therefore,
the smallest π-stacking distance and the highest lateral chain
packing could explain why the mobility in the PTBzT2-CEHβ
sample is so much different from that of the other two polymers.
Nevertheless, in such analysis one should also take into account
the morphology of the films used in the studies and not only
the structure of the bulk material since they can be very
different.19,20
Photovoltaic Devices. Polymer-fullerene photovoltaic de-
vices have been elaborated using the newly synthesized polymers
as electron donor. Different polymer:PCBM weight ratios have
been investigated as well as different annealing conditions.
For PTBzT2-CEHβ devices we used a low molecular mass
fraction to improve the film homogeneity. For PTBzT2-
CEHβ and PTBzT2-C12β, the thickness of the active layer
was kept close to 110 nm. The best results were obtained after
a 15 min postelaboration heat treatment at 200 and 130 °C
for PTBzT2-C12β and PTBzT2-CEHβ, respectively. The
properties of PTBzT2-C12R-based devices have already been
published before, and the results obtained for devices with a
comparable thickness of 135 nm are reported in Table 6. The
measured average photovoltaic results obtained on devices
with optimal fullerene content and for similar active layer
thicknesses are also summarized in Table 6 as well as the
parameters measured on a series of optimized PTBzT2-
CEHβ-based devices.
high-energy peak positions of both β-position-based poly-
mers almost coincide in solid state, canceling the blue shift
(with respect to the R-position-based polymers, PTBzT2-
C12R and PTBzT2-C8R) observed in solution. This points
out that in this case (alkyl side chains in β-position) the
intermolecular interactions of the conjugated backbones are
strong enough to outdo the influence of the side chains on the
π-π* transitions.
Changing alkyl side chains from R to β position leads, as
expected, to a significant reduction (0.32 eV) of the polymer
energy band gap. The resulting 1.56 eV optical band gap of
both PTBzT2-C12β and PTBzT2-CEHβ is close to the opti-
mal value for single-junction photovoltaic devices17and
makes these polymers highly promising for photovoltaic
applications.
Quasi-reversible oxidation and reduction waves could be
observed by cyclic voltammetry on all polymer thin films (see
Supporting Information). The corresponding oxidation and
reduction onsets are summarized in Table 4.
As can be seen from the above results, the side-chain
positioning drastically influences the HOMO levels of the
polymers, whereas the LUMO levels remain almost unaltered.
Interestingly, the LUMO level is close to that of a structurally
related polymer, PPBzT2-C12R(3.2 eV), reported earlier, whose
donor segment includes two additional thiophene rings.9
These results, along with the density functional theory investi-
gations, point out that the LUMO level is localized on the 2,1,
3-benzothiadiazole.
On the opposite, by passing from PTBzT2-C12R to
PTBzT2-C12β an 0.34 eV higher-lying HOMO level is ob-
tained. A small drop in the open-circuit voltage (Voc) of the
corresponding polymer-fullerene OPV devices should thus
be expected when using the PTBzT2-C12β or PTBzT2-CEHβ
instead of PTBzT2-C12R as electron donor materials (see
Photovoltaic Devices section). Nevertheless, for both posi-
tions, the frontier orbital energy levels should allow an
efficient photoinduced charge transfer to PCBM (whose
LUMO and HOMO levels are estimated to 4.3 and 6 eV,
respectively8).
Preliminary X-ray scattering measurements have been
performed on bulk R- and β-position polymer samples in
order to evaluate the influence of the side-chain positioning
on the π-stacking interactions. A more in-depth structural
analysis of our polymer thin films is out of scope of the
present article and will be reported elsewhere. For all poly-
mers, well-defined diffraction peaks could be observed,
indicating the semicrystalline nature of the materials (see
Supporting Information). The corresponding π-stacking
distances for PTBzT2-C12R, PTBzT2-C12β, and PTBzT2-
CEHβ could be estimated to 0.41, 0.37, and 0.35 nm,
respectively. These results thus further substantiate the
assertion that the π-stacking interactions are most pro-
nounced for the β-position polymers.
As expected, the polymer-fullerene weight ratio has a
major impact on the device performances. As shown in
Figure 4, the normalized power conversion efficiency
(PCE) varies by more than a factor of 3, when the weight
ratio changes from 1:1 to 1:4. Interestingly, for the linear

Article
Macromolecules, Vol. 43, No. 23, 2010 9785
polymer-fullerene bulk heterojunctions reported by Scharber
et al. leads to a Voc of 0.4 V, well below our experimental
findings.23 The actual Voc is close to the value typical for
P3HT:PCBM devices (∼0.6 V). A significant increase in PCE
(from 1.5 to 2.7%) is achieved on PTBzT2-CEHβ devices when
including a LiF layer and an annealing step prior to cathode
deposition (Supporting Information). This result highlights
the potential of this polymer and opens the way to further
optimization.
Conclusion
The synthesis and characterization of a series of donor-acceptor
alternate low-band-gap copolymers with a common conjugated
backbone, including 2,1,3-benzathiadiazole, thiophene, and thieno-
[3,2-b]thiophene units, decorated by different solubilizing side
chains allowed us to investigate quantitatively the relationships
between the polymer optoelectronic and photovoltaic properties
and the positioning and molecular structure of the side chains.
The results show that the polymer side chains can have a
profound impact on the target polymer functionalities, such as
the light harvesting capacity, ionization potential and electron
affinity, charge carrier mobilities, or intermolecular interactions
in polymer:fullerene blends.
Figure 4. Normalized average PCE obtained on polymer:PCBM
photovoltaic devices using PTBzT2-C12R (filled triangles), PTBzT2-
C12β (open squares), and PTBzT2-CEHβ (crosses). The normalization
was performed on the (1:4) polymer: C60-PCBM weight ratio and the
dotted line are used to guide the eye.
Table 6. Average and Best (Indicated by Asterisk) Photovoltaic
Parameters
ratio thickness (nm) Voc (V) Jsc (mA/cm2) FF (%) PCE (%)
The spectral and electrochemical measurements, as well as the
density functional calculations of model oligomers, point out that
these variations are related to the degree of twisting along the
molecular backbone. The side chains on the fourth position of
thiophene rings (β position) induce less steric hindrance and thus
a more planar molecular structure which enhances the π-electron
delocalization and favors intermolecular interactions. Moreover,
using ethylhexyl side chains instead of linear alkyl side chains
modifies the interaction between the polymer and the fullerene
molecules, leading to opposite dependences of the photovoltaic
performances on the fullerene content. The latter observation
supports the idea that fullerene intercalation between the side
chains is at the origin of the frequently observed high optimal
fullerene content in polymer:fullerene bulk heterojunction devices.
PTBzT2-C12R (pristine)
1:4
1:4
135
PTBzT2-C12β (annealed 15 min at 200 °C)
110 0.55 1.88 41
PTBzT2-CEHβ (annealed 15 min at 130 °C)
0.67
2.17
32
0.46
0.42
1:1
1:1*
120
120
0.55
0.67
5.67
7.80
40
52
1.24
2.71
alkyl chain polymers, the PCE increases with fullerene
content, while the opposite is observed for the branched side
chain. The high PCBM loading required to achieve the
optimal efficiency is distinctive for a large variety of poly-
mers reported in the literature21,22 and has been tentatively
attributed by McCulloch et al. to the intercalation of the
fullerene between the polymer side chains.7 In this case, the
high fullerene content is believed to achieve a continuous
electron pathway necessary to support electron transport
efficiently. Accordingly, our results support the idea that
linear alkyl side chains, in both R and β position, make these
polymers prone to fullerene intercalation. Furthermore, the
more bulky 2-ethylhexyl chains could hamper fullerene
intercalation, allowing efficient electron transport to take
place at a lower fullerene content. Additional charge trans-
port and structural analyses on the polymer-fullerene
blends are however necessary to corroborate this assertion.
The highest PCE value on LiF free devices (see Supporting
Information) is achieved with PTBzT2-CEHβ and correlates
with the large short-circuit-current (Jsc) obtained with this
polymer. The higher Jsc value in turn results from the lower
optimal polymer:fullerene ratio (1:1 instead of 1:4 for both
linear dodecyl based polymers), the lower optical band gap (in
comparison to PTBzT2-C12R), and the higher hole mobility
(in comparison to PTBzT2-C12β). The open-circuit voltages
(Voc) of both PTBzT2-CEHβ- and PTBzT2-C12β-based
devices are comparable and slightly lower (by 0.12 V) than
the Voc of the PTBzT2-C12R-based device. This voltage drop
can be linked to the less stable HOMO level of the β-position
polymers with respect to the R-position polymer. Note
however that the value estimated from the empirical linear
relationship between Voc and the polymer HOMO level in
Acknowledgment. Dimitri A. Ivanov and Denis V. Anokhin
acknowledge the financial support from the Agence Nationale de
la Recherche (France) provided in the frame of the SPIRWIND
project (HABISOL program) and T2T project (ANR blanc
program). We acknowledge the financial support from the
ꢀ
Region Alsace and thank the European Synchrotron Radiation
Facility for providing the excellent environment for X-ray mea-
surements. We are grateful to W. Bras and G. Portale (BM26,
ESRF) for assistance in the diffraction experiments and fruitful
discussions, to N. Zimmermann for his contribution to the device
elaboration, and to G. Hadziioannou for his critical comments.
Supporting Information Available: GPC chromatograms,
cyclic voltammogram scans, X-ray diffraction spectrum, OFET
characteristics, and I/V curves. This material is available free of
charge via the Internet at http://pubs.acs.org.
References and Notes
(1) Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.-T.; Wu, Y.; Li, G.; Ray, C.;
Yu, L. Adv. Mater. 2010, 22, E135–E138.
(2) Kroon, R.; Lenes, M.; Hummelen, J. C.; Blom, P. W. M.; De Boer,
B. Polym. Rev. 2008, 48, 531–582.
(3) McCulloch, I.; Heeney, M.; Bailey, C.; Genevicius, K.; MacDonald,
I.; Shkunov, M.; Sparrowe, D.; Tierney, S.; Wagner, R.; Zhang, W.;
Chabinyc, M. L.; Kline, R. J.; McGehee, M. D.; Toney, M. F. Nature
Mater. 2006, 5, 328–333.
€
(4) Nguyen, L. H.; Gunes, S.; Neugebauer, H.; Sariciftci, N. S.;
Colladet, K.; Fourier, S.; Cleij, T. J.; Lutsen, L.; Gelan, J.;
Vanderzande, D. E. P. J. Appl. Phys. 2006, 36, 219–224.

9786 Macromolecules, Vol. 43, No. 23, 2010
Biniek et al.
(5) Li, W.; Qin, R.; Zhou, Y.; Andersson, M.; Li, F.; Zhang, C.; Li, B.;
Liu, Z.; Bo, Z.; Zhang, F. Polymer 2010, 51, 3031–3038.
€
(6) Muller, C.; M. Ferenczi, T. A.; Campoy-Quiles, M.; Frost, J. M.;
Bradley, D. D. C.; Smith, P.; Stingelin-Stutzmann, N.; Nelson, J.
D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong,
M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision B.03;
Gaussian, Inc.: Pittsburgh, PA, 2003.
Adv. Mater. 2008, 20, 3510–3515.
(12) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652. (b) Lee, C. T.;
Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789.
(13) Zhang, L.; Zhang, O.; Ren, H.; Yan, H.; Zhang, J.; Zhang, H.;
Gu, J. Sol. Energy Mater. Sol. Cells 2008, 92, 581–597.
(14) Bras, W.; Dolbnya, I. P.; Detollenaere, D.; van Tol, R.; Malfois,
M.; Greaves, G. N.; Ryan, A. J.; Heeley, E. J. Appl. Crystallogr.
2003, 36, 791–794.
(7) Mayer, A. C.; Toney, M. F.; Scully, S. R.; Rivnay, J.; Brabec, C. J.;
Scharber, M.; Koppe, M.; Heeney, M.; McCulloch, I.; McGehee,
M. D. Adv. Funct. Mater. 2009, 19, 1173–1179.
(8) Biniek, L.; Chochos, C. L.; Leclerc, N.; Hadziioannou, G.; Kallitsis,
ꢀ ^
J. K.; Bechara, R.; Leveque, P.; Heiser, T. J. Mater. Chem. 2009, 19,
4946–4951.
(9) Biniek, L.; Chochos, C. L.; Hadziioannou, G.; Leclerc, N.; Leveque,
P.; Heiser, T. Macromol. Rapid Commun. 2010, 31, 651–656.
(10) Pilgram, K.; Zupan, M.; Skile, R. J. Heterocyl. Chem. 1970, 6, 629–
633.
ꢀ ^
(15) Liang, F.; Lu, J.; Ding, J.; Movileanu, R.; Tao, Y. Macromolecules
2009, 42, 6107–6114.
(16) Roquet, S.; Cravino, A.; Leriche, P.; Aleveque, O.; Frere, P.;
ꢀ ^
Roncali, J. J. Am. Chem. Soc. 2006, 128, 3459–3466.
ꢁ
^
(11) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A.; Jr., Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi,
J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G.A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.;
Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian,
H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli,
C.; Ochterski, J. W.; Ayala, P.Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.;
Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck,
A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.;
Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox,
(17) Wurfel, P. In Physics of Solar Cells; Wiley-VCH: Weinheim, 2005.
(18) Park, Y. D.; Kim, D. H.; Jang, Y.; Cho, J. H.; Hwang, M.; Lee,
H. S.; Lim, J. A.; Cho, K. Org. Electron. 2006, 7, 514–520.
(19) Zhu, X.; Mourran, A.; Moller, M.; Beginn, U.; Anokhin, D. V.;
€
Ivanov, D. A. Phys. Chem. Chem. Phys. 2010, 12, 1444–1452.
(20) Gospodinova, N.; Ivanov, D. A.; Anokhin, D. V.; Mihai, I.; Vidal,
L.; Brun, S.; Romanova, J.; Tadjer, A. Macromol. Rapid Commun.
2009, 30, 29–33.
(21) Hoppe, H.; Sariciftci, N. S. J. Mater. Chem. 2006, 16, 45–61.
(22) Shi, C.; Yao, Y.; Yang, Y.; Pei, Q. J. Am. Chem. Soc. 2006, 128,
8980–8986.
€
(23) Scharber, M. C.; Muhlbacher, D.; Koppe, M.; Denk, P.; Waldauf,
C.; Heeger, A. J.; Brabec, C. J. Adv. Mater. 2006, 18, 789–794.
(24) Clarke, T. M.; Durrant, J. R. Chem. Rev. 2010, ASAP, DOI:
10.1021/cr900271s.