Influence of different copolymer sequences in low band gap polymers on their performance in organic solar cells

Article abstract of DOI:10.1002/pola.25933

The chemical design of a polymer can be tailored by a random or a block sequence of the comonomers in order to influence the properties of the final material. In this work, two sequences, PCPDTBT and F8BT (F8), were polymerized to form a block or a random copolymer. Differences between the various polymers were examined by exploring the surface topography and charge carrier mobility. A distinct surface texture and a higher charge carrier mobility was found for the block copolymer with respect to the other materials. Solar cells were prepared with polymer:PC₇₁BM blend active layers and the best performance of up to 2% was found for the block copolymer, which was a direct result of the fill factor. Overall, the sequences of different copolymers for solar cell applications were varied and a positive impact on efficiency was found when the block copolymer structure was utilized.

Full text of DOI:10.1002/pola.25933

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Influence of Different Copolymer Sequences in Low Band Gap Polymers
on Their Performance in Organic Solar Cells
Alexander Lange,1 Hartmut Krueger,1 Bernhard Ecker,2 Ali Veysel Tunc,2
Elizabeth von Hauff,2 Mauro Morana3
1Fraunhofer Institute for Applied Polymer Research, Geiselbergstrasse 69, 14476 Potsdam-Golm, Germany
2Energy and Semiconductor Research Laboratory, Institute of Physics, Carl von Ossietzky University, 26111 Oldenburg, Germany
3Konarka Austria Forschungs und Enwicklungs GmbH, Altenbergerstr. 69, 4040 Linz, Austria
Correspondence to: H. Krueger (E-mail: krueger@iap.fhg.de)
Received 5 September 2011; accepted 4 January 2012; published online 5 February 2012
DOI: 10.1002/pola.25933
ABSTRACT: The chemical design of a polymer can be tailored
by a random or a block sequence of the comonomers in order
to influence the properties of the final material. In this work,
two sequences, PCPDTBT and F8BT (F8), were polymerized to
form a block or a random copolymer. Differences between the
various polymers were examined by exploring the surface to-
pography and charge carrier mobility. A distinct surface texture
and a higher charge carrier mobility was found for the block
copolymer with respect to the other materials. Solar cells were
prepared with polymer:PC71BM blend active layers and the
best performance of up to 2% was found for the block copoly-
mer, which was a direct result of the fill factor. Overall, the
sequences of different copolymers for solar cell applications
were varied and a positive impact on efficiency was found
C
V
when the block copolymer structure was utilized. 2012 Wiley
Periodicals, Inc. J Polym Sci Part A: Polym Chem 50: 1622–
1635, 2012
KEYWORDS: diblock copolymers; morphology; organic solar
cells; phase behavior; random copolymers
much light as possible from the sun’s spectrum in order to
increase device performance.6–10 In comparison to P3HT,
which absorbs only 46% of the incident light from the sun,
PCPDTBT absorbs close to 64%11 Due to its good solubility
and good matching of fullerene’s electronic levels, PCPDTBT
is an interesting candidate for high-efficiency solar cells.
However, an unfavorable, highly intermixed nanomorphology
observed in as-cast blends with PCBM seems to reduce the
possible applications of this polymer.6 Therefore, other low
band gap polymers have been synthesized and developed
and device efficiencies of over 7% have been reported which
demonstrates the potential of novel materials for solar
cells.12
INTRODUCTION Due to increasing energy demands around
the world and the need for green energy resources, research
activities in the field of organic photovoltaics (OPV) have
steadily grown during the past 10 years.1–3 The most suc-
cessful type of OPV device utilizes a mixture of an electron
donating and an electron accepting species in a bulk hetero-
junction. This device concept utilizes a large interfacial area
between both components to facilitate the splitting of exci-
tons into free charges at phase boundaries, providing proper
energy level alignment between the electron donor and
acceptor.4 Electrons and holes are subsequently transported
to the electrodes via pure phases of the respective compo-
nents. Many studies have explored solar cells with the stand-
ard materials system, semicrystalline poly(3-hexylthiophene)
(P3HT), and phenyl-C61-butyric acid methyl ester (PC61BM).
Power conversion efficiencies (gPCE) of up to 5% have been
reported for devices based on P3HT:PC61BM.5 In addition to
As previously reported, thermal annealing has a major
impact on the properties of P3HT:PC61BM thin films where a
correlation between the treatment temperature and degree
of crystallinity was found.13 Correspondingly, thermal treat-
ments were shown to directly impact the performance of
P3HT:PC61BM solar cells as well as their stability.5,14,15 In
addition, the degree of organization in P3HT:PC61BM films
can be influenced with non-thermal treatments such as
solvent annealing where films are allowed to dry for an
extended period of time after spin coating in a solvent-satu-
rated atmosphere.15 Nonetheless, the degree of organization
a
large emphasis on P3HT:PC61BM, the synthesis of
novel materials has emerged and the focus has partially
shifted to investigate polymers, such as poly[2,1,3-benzothia-
diazole-4,7-diyl[4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-
b0]dithiophene-2,6-diyl]] (PCPDTBT), which have absorbance
maximums at longer wavelengths with respect to P3HT. The
main goal of using low band gap materials is to harvest as
Additional Supporting Information may be found in the online version of this article.
C
V
2012 Wiley Periodicals, Inc.
1622
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
in P3HT:PC61BM films is an important aspect when prepar-
ing high performance devices. In contrast to P3HT, the prop-
erties, such as absorbance, of PCPDTBT do not improve upon
annealing.12 Other methods are then employed to induce
morphology changes in PCPDTBT films such as processing
with solvent additives, i.e., diiodo-octane (DIO) or octane
dithiol (ODT).7,16,17 Different requirements must be fulfilled
when using an additive in order to achieve good results. First,
the additive should have a higher boiling point than the com-
mon solvent.18 Second, only the fullerene species and not the
polymer should be soluble in the additive.17 Due to the boiling
point and solubility properties of the additive, the polymer
precipitates prior to the fullerene species from the common
solution during processing. Correspondingly, the polymer can
phase separate from the fullerene, forming an enhanced net-
work which can lead to better device performance. When
comparing the three major ways to influence the morphology
of polymer–fullerene thin films, the use of solvent additives
was shown to be the most useful for low band gap materi-
als.6,19 The use of additives may however complicate solution
processing and does not always offer a favorable way to coat
polymer–fullerene blends with roll-to-roll methods.
FIGURE 1 The solar cell structure used in this report.
structure of the electron donating species on solar cell per-
formance, active layers were prepared with blends of the
previously mentioned polymers with PC71BM. For better
understanding and characterizing the impact of the copoly-
mers on performance, ternary blends of PCPDTBT, F8BT and
PC71BM were also examined in solar cells with the device
structure shown in Figure 1. Here, we analyze the impact of
different copolymer sequences on device performance and
surface topography. The novel PCPDTBT-block-F8 shows the
best solar cell performance of all the blends investigated.
Electron accepting polymers such as poly(9,9-dioctylfluor-
ene-alt-benzothiadiazole) (F8BT) can be used to create poly-
mer–polymer donor–acceptor systems. F8BT’s highest occu-
pied molecular orbital (HOMO) and lowest unoccupied
molecular orbital (LUMO) levels are well positioned with
those of P3HT in addition to poly(vinyl carbazole) (PVK) for
efficient charge transfer as studies on blends of P3HT:F8BT
and PVK:F8BT have shown.20,21 For the PVK:F8BT devices,
the weight ratio of PVK to F8BT was found to directly impact
performance and the surface topography of the device active
layer.21 However, polymer–polymer solar cells typically suffer
from low short circuit current densities (Jsc) when compared
to devices with PC61BM.22 Despite the low performance of
polymer–polymer devices with respect to solar cells based
on polymers mixed with PC61BM or PC71BM, the chemical
properties of F8BT can be exploited for solar cell applica-
tions. Specifically, the electron donating and electron accept-
ing potentials of different monomers which are combined
and polymerized have a direct impact on the properties of
the resulting polymer.23 This concept has been explored in
the literature for polymers with different donor–acceptor
(DA) or donor–acceptor donor (DAD) units for solar cell
applications where the band gap was found to vary with the
strength of the electron accepting unit.23
RESULTS AND DISCUSSION
Chemistry
For synthesizing the PCPDTBT homopolymer, random
PCPDTBT-co-F8BT copolymer and PCPDTBT-block-F8 copoly-
mer new synthesis procedures were developed. The PCPDTBT
homopolymer and the random PCPDTBT-co-F8BT copolymer
were polymerized via Suzuki-coupling. PCPDTBT-block-F8 was
polymerized in two steps and a polymer analogous reaction.
All polymers were analyzed with differential scanning calorim-
etry to examine their thermal behavior. For all polymers no
thermal transitions and no degradation effects were found up
to 250 ꢀC. Glass transitions could not be determined using this
method. In the following, a detailed description will be given.
Monomers
The 2,6-bis(4,4,5,5-tetramethyl)-[1,3,2]dioxaborolane-4,4-bis
(2-ethylhexyl)-4H-cyclopenta-[2,1-b:3,4-b0]dithiophene mono-
mer was synthesized following an eight-step procedure. The
first six steps were in accordance with literature.24 The sev-
enth and eighth steps were in accordance with general or-
ganic reactions of bromination and boronic ester synthesis.
In this work, we synthesized three amorphous, low band gap
polymers and used a solvent additive (DIO) to process thin
films for organic solar cells. In preparing fullerene blend
films, we used DIO as a solvent additive instead of ODT
because it was found to improve film quality. In addition,
other works have reported similar or even better device effi-
ciencies for films prepared with DIO as compared to ODT.17
The polymer species examined here are pure PCPDTBT, a
block copolymer of PCPDTBT with F8 (PCPDTBT-block-F8)
and a random copolymer of PCPDTBT with F8BT (PCPDTBT-
co-F8BT). In order to determine the impact of the chemical
The 5,8-dibromo-2,1,3-benzothiadiazole was obtained by bro-
mination of 2,1,3-benzothiadiazole using bromine in hydro-
bromic acid.
PCPDTBT Homopolymer
For the PCPDTBT homopolymer, a Stille coupling procedure
is typically applied in the literature.24 Pisula et al. used
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635
1623

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Overview of the Number Average (Mn) and Weight
Average (Mw) Molecular Weights, Polydispersity Indexes (PDI)
and Compositions of the Polymers Synthesized in this Work
weights of the thiophene- and the 5-bromothiophene-termi-
nated F8-blocks are shown in Table 2.
The thiophene end groups were identified by the corre-
sponding NMR signals at 7.40, 7.30, and 7.12 ppm for the
three protons of the thiophene. The success of thiophene
end group bromination was also determined with 1H NMR
by the two remaining signals at 7.13 and 7.06 ppm. By com-
paring the molecular weights (Table 2), it can be seen that
no significant change took place during bromination; thus,
degradation can be excluded.
Molar ratio
F8BT
103 g/mole
Polymer
PCPDTBT
(or F8)
Mn
Mw
PDI
PCPDTBT
1
0
11.7
9.0
31.0
22.5
34.3
8.4
2.65
2.5
PCPDTBT-co-F8BT
PCPDTBT-block-F8
F8BT
0.66
0.94
0
0.34
0.06
1
17.8
3.7
1.93
2.27
The obtained 5-bromothiophene-terminated F8-block was
introduced in the second step Suzuki-polymerization. There-
fore, the 2,6-bis(4,4,5,5-tetramethyl)-[1,3,2]dioxaborolane-
4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene
(n þ z) was used as a diboronic acid comonomer in a 24/1
molar ratio to the 5-bromothiophene-terminated F8-block
(z ¼ 1) with an equimolar amount of 5,8-dibromo-2,1,3-
benzothiadiazole (n ¼ 23) (Fig. 5).
The obtained PCPDTBT-block-F8 shows comparable molecu-
lar weights as the homopolymer PCPDTBT. Because of the
chosen molar monomer ratio, the PCPDTBT-block is found
to exceed 94 mole% in the block copolymer during
polymerization.
Suzuki coupling as the polymerization technique.25 They
used diboronic ester of 5,8-dibromo-2,1,3-benzothiadiazole
and 2,6-dibromo-4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,
4-b0]dithiophene as monomers. We used a modified Suzuki
coupling (Fig. 2) for this step by changing the dibromo and
the diboronic ester monomers. Therefore, we introduced a
novel dicyclohexylphosphino-20,60-dimethoxybiphenyl (dhpdb)
ligand for the Pd(II) acetate catalyst. Satisfying molecular
weights were obtained only with this ligand after treatment
under reflux conditions (Table 1). Nevertheless, the Mn of our
Suzuki-coupled PCPDTBT is still somewhat lower than what is
obtained for the Stille-coupling polymerization.
Energy Levels of the Polymers
The energy levels of the different polymers were estimated
by cyclovoltammetric measurements in films on glassy car-
bon electrodes. The results are summarized in Table 3.
Random PCPDTBT-co-F8BT
The synthesis of the PCPDTBT-co-F8BT was carried out by
Suzuki copolymerization using equimolar moles of 2,6-
bis(4,4,5,5-tetramethyl)-[1,3,2]dioxaborolane-4,4-bis(2-ethyl-
hexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene and 2,7-bis
(4,4,5,5-tetramethyl-[1,3,2]dioxaborolane)-9,9-dioctylfluorene
(Fig. 3, with n ¼ m). An enrichment of the dithiophene
monomer in the copolymer (0.66–0.34 molar ratio) was
found. The molecular weights were in the same range as for
the PCPDTBT homopolymer.
In the PCPDTBT-containing polymers, the HOMO- and the
LUMO-levels were determined by the PCPDTBT-sequence,
resulting in similar values. The block-copolymer exhibits
exactly the same levels as the homopolymer. The random co-
polymer shows just a little wider electrochemical band gap
which agrees with the lower absorption maximum compared
to the homo- and the block-copolymers. Finally, F8BT has a
significantly wider band gap with a considerably lower
HOMO level and a slightly higher LUMO level. The HOMO
levels of the different polymers in Table 3 align well with
the work function of poly(3,4-ethylenedioxythiophene): poly
(styrenesulfonate) (PEDOT:PSS) (ꢁ5 eV) which eliminates
potential hole extraction barriers.21 In addition, the energy
difference between the LUMO levels of the polymers and
PC71BM is large enough (0.9–1.0 eV) to effectively split exci-
tons. Values for PC71BM were taken from ref. 23 which
found values of ꢂ4.1 eV and ꢂ6.5 eV for the HOMO and
LUMO, respectively. Overall, the energy levels of PCPDTBT,
PCPDTBT-block-F8
The block-copolymer synthesis was carried out in a two-step
procedure (Fig. 4). First, a Yamamoto polymerization was
chosen to generate the F8-block. The polymerization was
done in the presence of a 2-bromo-thiophene end-capping
agent. By adjusting the molar ratio of dibromo-F8 and 2-
bromo-thiophene, the molecular weight can be controlled.
The obtained thiophene-terminated F8-block was brominated
in the five-position of the thiophene end groups using NBS
in a polymer analogous reaction. The experimental molecular
FIGURE 2 Synthesis of PCPDTBT homopolymer by SUZUKI-coupling.
1624
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 3 Synthesis of random PCPDTBT-co-F8BT by SUZUKI-coupling.
PCPDTBT-co-F8BT, and PCPDTBT-block-F8 are well suited for
solar cell applications.
UV-Vis spectroscopy was used as shown in Figure 7. When
the spectra of the pure polymers shown in Figure 6 are com-
pared to those of the blends presented in Figure 7(a), the
peak from the polymer absorbance at longer wavelength is
seen in addition to the broad absorbance of PC71BM within
the visible region. The intensities of the peaks around a 750-
nm mirror those of the pure polymers where PCPDTBT-
block-F8 has the highest intensity followed by PCPDTBT and
PCPDTBT-co-F8BT. However, the intensities of the peaks are
also influenced by small layer thickness differences. A close-
up of the absorption spectra between 575 and 900 nm is
shown in Figure 7(b). Interestingly, no distinct peak is seen
at longer wavelengths for PCPDTBT-co-F8BT:PC71BM.
Instead, the spectrum from PC71BM overlaps with the peak
from the polymer because of its considerable blue shift with
respect to the other polymers. Upon addition of F8BT to
PCPDTBT (1:1), a peak at 462 nm becomes visible, which
corresponds to the absorbance of pure F8BT. For a homo-
polymer blend of PCPDTBT:F8BT with less F8BT (1:0.5),
the peak for F8BT overlaps with the spectrum of PC71BM.
Optical Properties of the Pure Polymers and Polymer:
PC71BM Films
The absorption behavior of the different polymers with com-
parable film thicknesses are summarized in Figure 6. F8BT
shows the expected spectrum with two maxima at 331 and
465 nm.26 The optical band gap of PCPDTBT is shifted to
longer wavelengths with respect to F8BT where the absorb-
ance maximum and edge are found at 730 and 900 nm,
respectively. The measured peaks for pure PCPDTBT agree
well with reports in literature.27 For the random PCPDTBT-
co-F8BT copolymer, absorbance ends already at 810 nm and
the maximum is shifted to 664 nm. The considerable blue
shift for this material can be explained by random arrange-
ment of the PCPDTBT and F8BT sequences. More specifically,
the PCPDTBT sequences are diluted with the F8BT sequences
and the resulting conjugation length is shorter with respect
to the homopolymer. In case of the block copolymer, nearly
the same peaks are found as for the PCPDTBT homopolymer.
However, the appearance of a weak shoulder in the spectrum
of PCPDTBT-block-F8 indicates an enhanced organization
within the thin film which corresponds well to the AFM anal-
ysis which will be discussed in detail later. Shoulders in the
absorbance spectra of PCPDTBT have been found upon proc-
essing with additives which resulted in more organized
films.28 In general, a decrease in the low band gap absorb-
ance was found with increasing F8-content in the polymers
examined here.
Photoluminescence
Photoluminescence (PL) measurements are an important
characterization method for electron donor and acceptor
films. Ideally, the majority of the photoluminescence from an
electron donor should be fully quenched upon addition of an
electron acceptor like PC71BM, due to the charge transfer
from donor to acceptor sites. PL measurements were per-
formed on PCPDTBT, PCPDTBT-block-F8, and PCPDTBT-co-
F8BT with and without PC71BM as shown in Figure 8(a,b).
Excitation wavelengths of 420 and 550 nm were used for
Figure 8(a,b), respectively.
In order to understand the optical properties of thin films
consisting of the different polymers blended with PC71BM,
FIGURE 4 Synthesis of 5-bromothiophene-terminated F8-block for block copolymer approach.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635
1625

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 2 Experimental Molecular Weights of the
absorb light in addition to PC71BM (Fig. 7, UV-Vis from
blends). Therefore, a direct comparison of the blend and
pure polymer PL spectra upon excitation at 420 nm is
difficult.
Thiophene- and the 5-Bromothiophene-Terminated F8-Blocks
g/mole
Founda Mn
Mw
PDI
b
b
Solar Cell Characteristics
In order to describe the influence of the different polymer
structures on the charge mobilities in the polymer films,
OFETs were fabricated in a bottom-gate structure. From the
current-voltage characteristics, hole mobilities were calcu-
lated. They are summarized in Table 3. It can be clearly seen
that for the polymers with larger PCPDTBT sequences
(PCPDTBT homo- and block copolymer), a hole mobility in
the range of 10ꢂ3 cm2/Vs was found. In contrast, the ran-
dom copolymer (PCPDTBT-co-F8BT) exhibits a hole mobility
of one order of magnitude lower. Additionally, the block co-
polymer shows a hole mobility two times higher than the
PCPDTBT homopolymer which should result in a better
charge transport ability in solar cells with the block copoly-
mer/PC71BM-blend films.
Thiophene-terminated F8-block 2265
1137 2896 2.55
1386 2636 1.90
5-bromothiophene-terminated
F8-block
2422
a
Calculated from elemental analysis, S-content.
Obtained by GPC using polystyrene calibration.
b
Upon excitation at 550 nm, PL maxima between 790 and
800 nm were found for PCPDTBT, PCPDTBT-block-F8, and
PCPDTBT-co-F8BT [Fig. 8(b)]. As expected, the PL maxima
are red shifted with respect to the absorbance maxima for
the pure polymers. For the polymer blends with PC71BM, the
PL maxima between 790 and 800 nm are almost completely
quenched. The insert in Figure 8(b) shows minor differences
between the three blends, all showing a quenching ratio of
over two orders of magnitude at the polymer maximum
emission, indicating rather efficient exciton collection and
charge transfer. From other works, the maximum PL inten-
sity for PCPDTBT was reported between 872 and ꢁ950
nm.6,11,29–31 Additionally, the PCPDTBT absorbance maximum
was measured between 730 and ꢁ775 nm.6,11,29,30 We pro-
pose that the lower values for the absorbance and PL peaks
for our polymers could be related to their relatively low mo-
lecular weights as compared to other works.6,11
Solar cells were prepared with the different polymers com-
bined with PC71BM and measured under illumination as
shown in Figure 9(a). Dark current-voltage characteristics
are shown in the Supporting Information. The devices were
further characterized with external quantum efficiency (EQE)
measurements. A summary of the data from Figure 9(a,b) is
given in Table 4.
As shown in Figure 9(a) and Table 4, devices with PCPDTBT-
block-F8 had the highest gPCE values of ꢁ2.0% followed by
PCPDTBT and PCPDTBT-co-F8BT. This result corresponds
well to the UV-Vis spectra for the block-copolymer and the
higher OFET mobility with respect to the other polymers.
Low efficiencies for PCPDTBT-co-F8BT can partially be
explained by the small amount of light which this polymer
absorbs at longer wavelengths in addition to the reduced
hole mobility. The lower light absorption for the random co-
polymer at longer wavelengths is a result of the random dis-
tribution of the copolymer sequences. The absorbance spec-
trum showed that the majority of light which this system
absorbs is due to PC71BM and not the polymer. Interestingly,
slightly larger Voc values were found for devices with
PCPDTBT-co-F8BT which is related to the polymer’s rela-
tively high HOMO level. This result corresponds to previously
reported results where the Voc of organic solar cells is
As seen in the absorbance spectra of the pure polymers, an
additional maximum at shorter wavelengths around ꢁ400
nm was also found. Upon excitation of the pure polymer
films with 420 nm, a second PL maximum for each polymer
was found in Figure 8(a). To our knowledge, this PL peak
has not been examined in detail in the literature. In this sit-
uation, the PL peaks at 420 nm are not completely quenched
upon addition of PC71BM. The exact cause of this additional
PL peak at shorter wavelengths is unclear. Different proc-
esses occur upon exciting the pure polymers or the blends at
420 nm. For the pure polymers, the F8BT units are most
likely excited and fluoresce. However, a similar trend is also
seen for PCPDTBT where no F8BT is present. In the blend
systems, both the F8BT units from the different polymers
FIGURE 5 Synthesis of PCPDTBT-block-F8.
1626
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 3 Estimated Energy Levels from CV and Calculated OFET Hole Mobilities of the Synthesized Polymers
Molar ratio eV
F8BT (or F8) LUMO
10ꢂ3 cm2/Vs
Polymer
PCPDTBT
HOMO
Band Gap (CV)
lhole
PCPDTBT
1
0
ꢂ5.1
ꢂ5.2
ꢂ5.1
ꢂ5.9
ꢂ3.2
ꢂ3.1
ꢂ3.2
ꢂ3.0
1.9
2.1
1.9
2.9
1.4
0.11
2.3
0.1
PCPDTBT-co-F8BT
PCPDTBT-block-F8
F8BT
0.66
0.94
0
0.34
0.06
1
directly related to the HOMO and LUMO levels of the donor
and acceptor, respectively.23
A summary of the performance of the different blends for
the systems examined in this report is shown in Figure 10.
The best performance was found for the block copolymer
consisting of PCPDTBT and F8 units. It is clearly seen that
the ternary blends perform worse than the copolymers.
In order to better understand how the chemical composition
of the electron-donating polymer species impacts device per-
formance, solar cells were prepared with ternary blends of
PCPDTBT, F8BT, and PC71BM. Upon addition of F8BT to
PCPDTBT, device performance decreased as shown in Table
4 where considerably smaller Jsc and FF values were found.
As shown in the absorbance spectra in Figure 7(b), a distinct
blue shift is seen for the PCPDTBT:F8BT blend ratio of 1:1
which indicates a more unorganized system with respect to
the 1:0.5 PCPDTBT:F8BT system. Overall, both ternary
blends are blue shifted with respect to PCPDTBT:PC71BM.
Overall, combination of two homopolymers, F8BT and
PCPDTBT, results in reduced solar cell performance.
In comparison to what has been reported in the literature
for solar cells with PCPDTBT:PC71BM, the gPCE values for our
devices are lower. It has been reported that gPCE values of
between 3 and 6% are achievable for PCPDTBT:PC71BM
when processed from chlorobenzene with additives.6,7 The
From the EQE data in Figure 9(b), the sensitivity of the devi-
ces at wavelengths greater than 700 nm in comparison to a
traditional P3HT:PC61BM solar cell is demonstrated. The
highest EQE was observed for PCPDTBT-block-F8 followed by
PCPDTBT and PCPDTBT-co-F8BT which mirrors the Jsc values
in Table 4. For comparison purposes, a device was prepared
with F8BT:PC71BM. Despite the necessary HOMO and LUMO
level offset between F8BT and PC71BM (Table 3), small EQE
values were measured which may be a result hindered hole
transport. Thus, the enhanced EQE of PCPDTBT-block-F8 is a
result of the novel copolymer structure consisting of both
PCPDTBT and F8 blocks.
FIGURE 7 Absolute absorbance versus wavelength for PCPDTBT,
PCPDTBT-block-F8, PCPDTBT-co-F8BT and PCPDTBT:F8BT blended
with PC71BM (a) and normalized absorbance between 575 and 900
nm (b). Please note that the ratio of the polymers to PC71BM was 1:3
unless otherwise stated.
FIGURE 6 Absolute absorbance versus wavelength for
PCPDTBT, PCPDTBT-block-F8, PCPDTBT-co-F8BT and F8BT.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635
1627

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 8 Photoluminescence versus wavelength for pure
PCPDTBT, PCPDTBT-block-F8 and PCPDTBT-co-F8BT and for
the polymers blended with PC71BM upon excitation at 420 nm
(a) or 550 nm (b). The insert in (b) shows the quenching data
on a semilog scale for comparison purposes.
FIGURE 9 Current density versus voltage for devices with PCDTBT,
PCPDTBT-block-F8, PCPDTBT-co-F8BT, or PCPDTBTþF8BT blended
with PC71BM (a) and external quantum efficiency versus wave-
length for devices with PCDTBT, PCPDTBT-block-F8, PCPDTBT-co-
F8BT, or F8BT blended with PC71BM (1:3) (b). A P3HT:PC61BM de-
vice is shown in (b) for comparison purposes.
cause of the lower performance for our PCPDTBT which was
obtained via SUZUKI-polymerization is mainly related to the
molecular weight. Commercially available PCPDTBT from 1-
Material with a Mn of 19,900 g/mole resulted in solar cells
with an increased Jsc (9 mA/cm2) and FF (>50%). This trend
agrees with other works which showed that the Jsc of
PCPDTBT:PC71BM solar cells increases with Mn.27
roughness (Rrms) values for the different films, where 0.7,
0.9, 0.5, and 0.7 nm were found for PCPDTBT, PCPDTBT-
block-F8, PCPDTBT-co-F8BT, and F8BT, respectively. However,
the differences in the surface profiles of the different poly-
mers are most apparent upon visual inspection of the images
in Figure 11 and not by direct comparison of the roughness
AFM
Examination of the surface topography of thin films used for
solar cells is useful because it can give insight into phase
separation and device performance as previously
shown.15,17,20,21,32–35 To begin, films from pure PCPDTBT,
PCPDTBT-block-F8, PCPDTBT-co-F8BT, and F8BT without
PC71BM were examined with AFM as shown in Figure 11.
TABLE 4 Solar Cell Performance Summary for Devices with
Different Polymer:PC71BM Active Layers
Thickness Voc
Jsc
FF
gPCE
Polymer(s)
(nm)
(V)
(mA/cm2) (%) (%)
A distinct topography was found for PCPDTBT-block-F8 [Fig.
11(b)] in comparison to a smooth texture for PCPDTBT-co-
F8BT [Fig. 11(c)]. The surface topography of PCPDTBT [Fig.
11(a)] was more distinct than that of PCPDTBT-co-F8BT but
not as apparent as for PCPDTBT-block-F8. The different sur-
face topographies are partially reflected in root mean square
PCPDTBT
116
112
120
0.620 6.14
0.610 6.38
0.665 4.94
0.555 3.91
0.580 1.06
38.0 1.45
49.9 1.94
36.8 1.21
28.3 0.61
28.8 0.18
PCPDTBT-block-F8
PCPDTBT-co-F8BT
PCPDTBT:F8BT (1:0.5) 114
PCPDTBT:F8BT (1:1) 179
1628
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
the block copolymer are able to pack together and form dis-
tinct groups. This is not the case for the random copolymer
where sequences of PCPDTBT are diluted with F8BT and
vice versa.
AFM was also preformed on polymer blends with PC71BM as
shown in Figure 12. The topography images of the blends
in Figure 12 show smooth and similar surface features
for PCPDTBT, PCPDTBT-block-F8, PCPDTBT-co-F8BT, and
PCPDTBT:F8BT (1:1) as seen in Figure 12(a–c,e). Rrms values
of 0.5, 0.5, 0.3, and 0.7 nm were calculated, respectively. The
topography was much rougher for PCPDTBT:F8BT (1:0.5)
despite similar preparation conditions for all solutions where
Rrms was 8.6 nm. Upon further addition of F8BT to PCPDTBT,
a smooth topography was found with an Rrms of 0.7 nm. The
exact cause of the rough topography for PCPDTBT:F8BT
(1:0.5) is unclear. The phase contrast images show a more
distinct phase signal for PCPDTBT, PCPDTBT-block-F8, and
PCPDTBT:F8BT (1:1) [Figs. 12(f,g,j), respectively]. For
PCPDBT:F8BT (1:0.5), the distinct phase contrast features
are seen which is due to the rough film topography in Figure
12(d). In correlation with the images from the pure polymer
films, the phase contrast image for PCPDTBT-co-
F8BT:PC71BM (1:3) shows little phase information. Again,
the chemical nature of the two different copolymers has a
direct impact on the thin film topography. Overall, the trends
in the AFM images agree well with the other experimental
results. Distinct phases are found for the block copolymer
which also shows a red-shifted absorbance maximum (Fig.
6). The enhanced organization in this material also results in
the highest OFET mobility and the highest gPCE value which
is mainly a result of the enhanced fill factor for the materials
under consideration. In contrast, the random sequence in
FIGURE 10 gPCE as a function of the amount of F8BT in solar
cells with active layers based on PCPDTBT:F8BT:PC71BM (1:X:3)
in comparison to devices with PCPDTBT-block-F8:PC71BM (1:3)
(circles) or PCPDTBT-co-F8BT:PC71BM (1:3) (triangles) active
layers.
values. The trend in the surface topographies is also mir-
rored in the phase contrast images where the largest and
most distinct contrast was seen for PCPDTBT-block-F8 [Fig.
11(f)]. Interestingly, both of the homo polymers used in this
work (PCPDTBT and F8BT) have distinct surface features
in both the topography and phase contrast images [Fig. 11
a,d,e,h)]. However, this is not the case for random copolymer
where smooth and featureless images were generated. This
suggests that
a block copolymer structure based on
PCPDTBT and F8 units results in phase separation which can
be seen on a 2.5-lm scale. The regions of F8 or PCPDTBT in
FIGURE 11 , AFM images of spin-coated PCPDTBT (a,e), PCPDTBT-block-F8 (b,f), PCPDTBT-co-F8BT (c,g) and F8BT (d,e) where a,
b, c, and d are topography images (10.8 nm scale) and e, f, g, and h show phase contrast (2ꢀ scale).
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635
1629

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 12 AFM images of spin-coated blends of PCPDTBT (a,f), PCPDTBT-block-F8 (b,g), PCPDTBT-co-F8BT (c,h), PCPDTBT:F8BT
(1:0.5) (d,i), and PCPDTBT:F8BT (1:1) (e,j) all combined with three parts of PC71BM, where a, b, c, d, and e are topography images
(10.8 nm scale, except for i which has a 40-nm scale) and f, g, h, i, and j show phase contrast (2ꢀ scale).
PCPDTBT-co-F8BT results in a blue-shifted absorbance maxi-
mum (Fig. 6) which directly impacted Jsc. The mobility of the
random copolymer was considerably lower than that of the
block copolymer which corresponds to the feature-less sur-
face topography seen in Figures 11 and 12. The trend seen
here corresponds well to what has been found in the litera-
ture where rougher surface topographies which were deter-
mined with AFM resulted in devices with higher efficien-
cies.33–35
F8, distinct features are seen for the pure films, whereas
feature-less surfaces were found for the blend films. In the
block copolymer case, due to the difference between the
surface energies of the two blocks and the fullerene, it is
possible that some vertical phase separation between the
block copolymer occurs at the PEDOT:PSS interface, as well
as the formation of
gradient.
a vertical fullerene compositional
Vertical phase separation has been documented in
P3HT:PC61BM films where the surface energies of both blend
components and the substrate play an important role.36 The
Upon comparison of the surface topographies of the blend
and pure polymer films for PCPDTBT and PCPDTBT-block-
1630
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
surface roughness change for PCPDTBT and PCPDTBT-block-
F8 may indeed suggest a different material composition at
the film/air interface, which could consist of PC71BM or
polymer:PC71BM.
filtered and diluted in diethyl ether. The organic phase was
dried with Mg2SO4 and filtered. The filtrate was evaporated.
The raw product (33.11 g, 75.1% yield) was directly used in
the next step without any further purification and characteri-
zation because it was unstable.
EXPERIMENTAL
4H-Cyclopenta[2,1-b:3,4-b0]dithiophene-4-carbonic Acid
33.11 g (137.8 mmol) of dithenyl-hydroxy-acetic acid was
charged into a reaction vessel which was flushed with argon.
800 mL of benzene was added. The solution was cooled to 5
ꢀC and 57.6 g AlCl3 was added. Afterwards the reaction mix-
ture was vigorously stirred under reflux conditions for 30
min. The reaction mixture was cooled to room temperature
and 30 mL of water and 200 mL of 4n hydrochloric acid
were purged to the reaction mixture. The product was
extracted with diethyl ether and the extract was washed
with water. After drying and filtration the solvent was evapo-
rated. The product (21.31 g, 68.9% yield) was used without
any further purification. Melting point: 176 ꢀC. Elemental
analysis: Found—C: 54.23%, H: 2.55%, S: 28.97%. Calcd.—C:
54.03%, H: 2.72%, S: 28.85%.
Synthesis
All chemicals, reagents, and solvents were purchased from
Sigma–Aldrich and Acros and were used as received unless
further purification steps are described in this section. 2,7-
Bis(4,4,5,5-tetramethyl-[1,3,2]dioxaborolane)-9,9-dioctylfluor-
ene was used after twofold recrystallization from hexane.
Monomer Synthesis
2,6-Bis(4,4,5,5-tetramethyl)-[1,3,2]dioxaborolane-4,4-bis(2-eth-
ylhexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene. The synthesis
of 2,6-bis(4,4,5,5-tetramethyl)-[1,3,2]dioxaborolane-4,4-bis(2-
ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophen was mostly
in accordance with literature procedures carried out in eight
steps.24
2-Hydroxy-1,2-dithiophene-3-yl-ethanon
3
1H NMR (500 MHz, CDCl3): d [ppm]: 7.21 (dd, 4H, Ar-H, J ¼
An argon-flooded reaction vessel was charged with 50 g
(445.8 mmol) of thiophene-3-carbaldehyde, 5.96 g (22.1
mmol) of 3-benzyl-5-(2-hydroxyethyl)-4-methyl-3-thiazolium-
chloride, 140 mL of ethanol, and 18.9 mL of triethylamine
under a slightly reduced flow of argon. The reaction mixture
was stirred under reflux for 3 h. It was then cooled down
and poured into 1 L of ice water. The precipitate was fil-
tered, washed with water, and dried. The product was
recrystallized from ethanol. The yield was 25.51 g (51.0%).
Melting point: 116 ꢀC. Elemental analysis: Found—C:
53.42%, H: 3.55%, S: 28.35%. Calcd.—C: 53.55%, H: 3.60%,
S: 28.59%.
5.1 Hz), 4.63 (s, 1H, ACHA).
4H-Cyclopenta[2,1-b:3,4-b0]dithiophene
21.27 g (95.7 mmol) of 4H-cyclopenta[2,1-b:3,4-b0]dithiophene-
4-carbonic acid and 4.3 g of Cu powder were placed into a
reaction vessel, evacuated, and refilled with argon three
times. Afterwards, 150 mL of freshly distilled quinoline was
added and the reaction mixture was heated to reflux at about
237 ꢀC for 40 min. It was allowed to cool down to room
temperature and then it was poured into a mixture of ice
and hydrochloric acid. The product was extracted with diethyl
ether. The organic layer was extracted with 2n hydrochloric
acid followed by saturated Na2CO3 solution. The organic solu-
tion was washed with water to bring it to a neutral state.
The solution was dried with Na2SO4 and the solvent was
evaporated. The raw product was further purified by column
chromatography using hexane/ethyl acetate (10:1) as the elu-
ent and by sublimation. 16.46 g of pure product was isolated
(94.3% yield). Melting point: 74.5 ꢀC. Elemental analysis:
Found—C: 60.22%, H: 3.61%, S: 35.88%. Calcd.—C: 60.64%,
H: 3.39%, S: 35.97%.
1,2-Dithiophene-3-yl-ethan-1,2-dion
129.43 g (518.7 mmol) of CuSO4*5H2O was diluted in 84 mL
of water and 170 mL of pyridine. The solution was heated to
ꢀ
70 C and 51.67 g (0.23 mole) of 2-hydroxy-1,2-dithiophene-
3-yl-ethanon was added. The reaction mixture was stirred
for 1 h at 80 C. Afterwards it was cooled to room tempera-
ture and combined with 250 mL of hydrochloric acid (10%).
The organic components were extracted with diethyl ether.
The organic phase was dried over MgSO4, filtered, and
evaporated. The raw product was recrystallized from isopro-
panol. The yield was 41.13 g (80.32%). Melting point: 74 C.
Elemental analysis: Found—C: 54.30%, H: 2.65%, S: 28.82%.
Calcd.—C: 54.03%, H: 2.72%, S: 28.85%.
ꢀ
3
1H NMR (500 MHz, CDCl3): d [ppm]: 7.17 (d, 2H, Ar-H, J ¼
ꢀ
4.9 Hz ), 7.08 (d, 2H, Ar-H), 3.53 (s, 2H, ACH2A).
4,4-Bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b0]
Dithenyl-hydroxy-acetic Acid
dithiophene
41.13 g (0.185 mole) of 1,2-dithiophene-3-yl-ethan-1,2-dion
was diluted in a solution of 35.06 g KOH in ethanol/water
and stirred under reflux for 10 min. The reaction mixture
was cooled to 0 ꢀC and was adjusted to a pH value of 1 by
adding concentrated hydrochloric acid. Afterwards the etha-
nol was evaporated and the reaction mixture was diluted
with diethyl ether. The organic phase was extracted with sat-
urated Na2CO3 solution. The Na2CO3 phase was decolorized
with charcoal and afterwards hydrochloric acid (10%) was
added to precipitate the product. The white precipitate was
7.64 g (42.6 mmol) of 4H-cyclopenta[2,1-b:3,4-b0] dithiophene
was placed into a reaction vessel, evacuated, and refilled with
argon three times. 425 mL of tetrahydrofurane was injected
for dilution. The solution was cooled to 0 ꢀC, and 29.8 mL
(47.8 mmol) of butyllithium solution (1.6 molar in hexane)
was placed into the reaction vessel. The reaction was heated
to room temperature and stirred for 2 h. It was then cooled
again to 0 ꢀC and 8.3 mL (47.8 mmol) 2-ethylhexylbromide
was added. Afterwards it was stirred at 30 ꢀC for 2 h. The
same procedure with butyllithium and 2-ethylhexylbromide
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635
1631

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
addition was repeated with the same quantities and mixing
times. After the second cycle, the reaction vessel was stirred
overnight at room temperature. The raw material was poured
out in 300 mL of saturated NaCl solution. The product was
extracted twice with diethyl ether. The extracts were washed
with deionized water and dried. The solvent was evaporated
and the residue was purified by column chromatography with
hexane as the mobile phase. Finally, 16.01 g of the pure prod-
uct was isolated (93.3% yield).
1H NMR (500 MHz, CDCl3): d [ppm]: 7.44 (m, 2H, Ar-H),
1.86 (m, 4H, ACH2A), 1.34 (m, 24H, ACH3), 1.00–0.80 (m,
18H), 0.73 (t, 6H, ACH3), 0.59 (t, 6H, ACH3).
5,8-Dibromo-2,1,3-benzothiadiazole
25 g (0.183 mole) of 2,1,3-benzothiadiazole was placed in a
reaction vessel and then 125 mL of hydrobromic acid was
added. 30 mL of bromine was dropped slowly into the solu-
tion. Within 1 h after complete bromine addition, the tem-
ꢀ
perature was increased to 95 C for 4 h. The precipitate was
Elemental analysis: Found—C: 74.37%, H: 9.69%, S: 15.79%.
Calcd.—C: 74.56%, H: 9.51%, S: 15.92%.
filtered using a glass frit and washed with water to neutral-
ize. The raw product was recrystallized from methanol/tolu-
ene (95/5).
3
1H NMR (500 MHz, CDCl3): d [ppm]: 7.11 (d, 2H, Ar-H, J ¼
5.0 Hz), 6.92 (m, 2H, Ar-H), 1.86 (m, 4H, ACH2A), 1.10–0.80
(m, 18H), 0.75 (t, 6H, ACH3), 0.59 (t, 6H, ACH3).
1H NMR (500 MHz, CDCl3): d [ppm]: 7.72 (2, 2H, Ar-H).
13C NMR (125 MHz, CDCl3): d [ppm]: 153.0, 132.3, 113.9.
2,6-Dibromo-4,4-bis(2-ethylhexyl)-4H-cyclopenta
[2,1-b:3,4-b0]dithiophene
Polymerization
PCPDTBT Homopolymer
6.78 g (17.39 mmol) of benzyltrimethylammonium chloride
and 2.61 g (24.48 mmol) of ZnCl2 were added to a reaction
vessel which was evacuated and refilled with argon three
times. Afterwards 3.34 g (8.29 mmol) of 4,4-bis(2-ethyl-
hexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene was diluted in
95 mL dimethylformamide and the solution was transferred
to the reaction vessel. The mixture was stirred at room tem-
perature for 90 min and then the reaction was stopped by
adding 20 mL of water. The raw product was extracted with
hexane. The organic phase was dried over MgSO4, filtered,
and evaporated. The purification was done by flush chroma-
tography using hexane as the eluent. The yield was 4.49 g
(96.3%).
1.17 g (1.79 mmol) of 2,6-bis(4,4,5,5-tetramethyl)-[1,3,2]diox-
aborolane-4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b0] di-
thiophene, 0.525 g (1.79 mmol) of 5,8-dibromo-2,1,3-benzo-
thiadiazole, 2.28 g (10.74 mmol) K3PO4, 16.16 mg (0.072
mmol) Pd(II)-acetate and 29.56 mg (0.072 mmol) of dicyclo-
hexylphosphino-20,60-dimethoxybiphenyl were placed in a thor-
oughly pre-dried reaction vessel. The reaction vessel was evac-
uated and refilled three times with argon and then 40 mL of
tetrahydrofurane was added. The solution was stirred for 24 h
at room temperature followed by 24 h under reflux conditions.
Afterwards the polymer was first end-capped using bromoben-
zene followed by phenylboronic acid. The polymer was precipi-
tated in methanol/2n hydrochloric acid (3:1), filtered, and
dried at 40 ꢀC. The polymer was diluted in chloroform and
extracted with diluted ammonia. The organic phase was sepa-
rated, washed to neutralize, precipitated, filtered, and dried.
Further purification was done by soxhlet extraction using sol-
vents with increasing dissolving power to separate low molecu-
lar fractions. The yield was 640 mg.
Elemental analysis: Found—C: 53.47%, H: 6.69%, Br:
27.84%, S: 11.56%. Calcd.—C: 53.57%, H: 6.47%, Br:
28.51%, S: 11.44%.
1H NMR (500 MHz, CDCl3): d [ppm]: 6.93 (m, 2H, Ar-H),
1.80 (m, 4H, ACH2A), 1.10–0.80 (m, 18H), 0.78 (t, 6H,
ACH3), 0.62 (t, 6H, ACH3).
Elemental analysis: Found—C: 69.20%, H: 7.19%, N: 5.34%,
S: 17.35%.
2,6-Bis(4,4,5,5-tetramethyl)-[1,3,2]dioxaborolane-4,4-bis
(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene
4.8 g (8.56 mmol) of 2,6-dibromo-4,4-bis(2-ethylhexyl)-4H-
cyclopenta[2,1-b:3,4-b0]dithiophene was placed in a reaction
vessel which was evacuated and refilled with argon three
times. 80 mL of tetrahydrofurane was added and the solu-
tion was cooled to ꢂ80ꢀC. The reaction mixture was kept
below ꢂ20 ꢀC for 3 h and cooled to ꢂ80 ꢀC again. 5.03 g
(27.03 mmol) of 2-isopropoxy-4,4,5,5-tetramethyl-1,3,2-diox-
aborolane was added and the reaction mixture was stirred
overnight at room temperature. The reaction mixture was
treated with distilled water and the raw product was
extracted with diethyl ether. The organic phase was washed
with water, dried, filtered, and evaporated. The purification
was done by multiple flush chromatography using hexane as
the eluent. The yield was 2.88 g (51.3%) of pure product.
Elemental analysis: Found—C: 67.72%, H: 9.48%, S: 9.91%.
Calcd.—C: 67.89%, H: 9.24%, S: 9.80%.
1H NMR (500 MHz, CDCl3): d [ppm]: 8.14 (s, 2H, Ar-H), 7.88
(s, 2H, Ar-H), 2.08 (m, 4H, aACH2A), 1.59 (m, 2H, ACHA),
1.02 (m, 16H, ACH2A), 0.68 (m, 12H, ACH3).
PCPDTBT-co-F8BT Polymer
0.736
g
(1.125 mmol) of 2,6-bis(4,4,5,5-tetramethyl)-
[1,3,2]dioxaborolane-4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-
b:3,4-b0]dithiophene, 0.209 g (0.375 mmol) of 2,7-bis(4,4,5,5-
tetramethyl-[1,3,2]dioxaborolane)-9,9-dioctylfluorene, 0.441 g
(1.5 mmol) of 5,8-dibromo-2,1,3-benzothiadiazole, 1.91 g
(9.00 mmol) K3PO4, 13.15 mg (0.06 mmol) Pd(II)-acetate,
and 24.6 mg (0.06 mmol) of dicyclohexylphosphino-20,60-
dimethoxybiphenyl were placed in a thoroughly pre-dried
reaction vessel. The reaction vessel was evacuated and
refilled three times with argon and then 40 mL of tetrahy-
drofurane was added. The solution was stirred for 24 h at
room temperature followed by 24 h under reflux conditions.
Afterwards the polymer was first end-capped with
1632
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
bromobenzene followed by phenylboronic acid. The polymer
was precipitated in methanol/2n hydrochloric acid (3:1), fil-
tered, and dried at 40 C. The polymer was diluted in chloro-
form and extracted with diluted ammonia. The organic phase
was separated, washed to neutralize, precipitated, filtered,
and dried. Further purification was done by soxhlet extrac-
tion using solvents with increasing dissolving power to sepa-
rate low molecular fractions. The yield was 440 mg.
PCPDTBT-block-F8 Polymer. 1.025 g (1.566 mmol) of 2,6-
bis(4,4,5,5-tetramethyl)-[1,3,2]dioxaborolane-4,4-bis(2-ethyl-
hexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene, 0.200 g (0.066
mmol) of the 5-bromothiophene-terminated F8-block, 0.441
g (1.5 mmol) of 5,8-dibromo-2,1,3-benzothiadiazole, 1.99 g
(9.396 mmol) K3PO4, 14.09 mg (0.0626 mmol) Pd(II)-ace-
tate, and 25.68 mg (0.0626 mmol) of dicyclohexylphosphino-
20,60-dimethoxybiphenyl were placed in a thoroughly pre-
dried reaction vessel. The reaction vessel was evacuated and
refilled three times with argon followed by 42 mL of tetrahy-
drofurane. The solution was stirred for 24 h at room temper-
ature and then for 24 h under reflux conditions. Afterwards
the polymer was end-capped by first using bromobenzene
followed by phenylboronic acid. The polymer was precipi-
tated in methanol/2n hydrochloric acid (3:1), filtered, and
dried at 40 ꢀC. The polymer was diluted in chloroform and
extracted with diluted ammonia. The organic phase was sep-
arated, washed to neutralize, precipitated, filtered, and dried.
Further purification was done by soxhlet extraction using
solvents with increasing dissolving power to separate low
molecular fractions. The yield was 750 mg.
ꢀ
Elemental analysis: Found—C: 72.06%, H: 7.44%, N: 5.43%,
S: 14.00%.
1H NMR (500 MHz, CDCl3): d [ppm]: 8.14 (m, Ar-H), 8.02 (m,
Ar-H), 7.88 (m, Ar-H), 2.08 (m, 4H, a-CH2A), 1.55 (m,
ACHA), 1.02 (m, ACH2A), 0.68 (m, ACH3).
PCPDTBT-block-F8 Polymer
Thiophene-Terminated F8-block. 2.14 g (7.78 mmol) of Ni-
1,5-cyclooctadiene was placed in a thoroughly pre-dried reac-
tion vessel. The reaction vessel was evacuated and refilled
three times with argon and then 1.22 g (7.78 mmol) of 2,20-
bipyridyl dissolved in 75 mL of toluene/25 mL of dimethylfor-
mamide was added. After briefly stirring, 0.439 g (4.06 mmol)
of 1,5-cyclooctadieneꢀ was added to the mixture which was
stirred for 1 h at 85 C. Afterwards a mixture of 1.86 g (3.39
mmol) of 2,7-bis(4,4,5,5-tetramethyl-[1,3,2]dioxaborolane)-
9,9-dioctylfluorene and 88 mg (0.54 mmol) of 2-bromothio-
phene dissolved in 20 mL of toluene was added. After stir-
ring for 4 days at 85 ꢀC, the reaction was cooled to room
temperature, dissolved in chloroform, and extracted with 2n
hydrochloric acid. The organic phase was extracted with
0.1m EDTA-solution, followed by saturated NaHCO3-solution.
The chloroform-solution was concentrated by evaporation.
The polymer was obtained by precipitation in methanol, fil-
Elemental analysis: Found—C: 71.15%, H: 7.54%, N: 4.91%,
S: 17.09%.
1H NMR (500 MHz, CDCl3): d [ppm]: 8.14 (s, 2H, Ar-H), 7.88
(s, 2H, Ar-H), 2.08 (m, 4H, a-CH2A), 1.55 (m, 2H, ACHA),
1.02 (m, 16H, ACH2A), 0.68 (m, 12H, ACH3).
F8BT Polymer. 2.249 g (3.5 mmol) of 2,7-bis(4,4,5,5-tetra-
methyl-[1,3,2]dioxaborolane)-9,9-dioctylfluorene and 1.029 g
(3.5 mmol) of 5,8-dibromo-2,1,3-benzothiadiazole were
placed in a thoroughly pre-dried reaction vessel. The reac-
tion vessel was evacuated and refilled three times with ar-
gon and then 63 mL of tetrahydrofurane and 7 mL of tolu-
ene were added to dissolve the monomers. Afterwards 87.5
mL of a 2m Na2CO3 solution and 3.5 mL of a catalyst stock
solution, consisting of 3.93 mg (0.0175 mmol) Pd(II)-acetate
and 31.96 mg (0.105 mmol) of tris-(o-tolyl)phosphine dis-
solved in tetrahydrofurane/ethanol (3:1), were added. The
solution was stirred for 48 h at 80 ꢀC. Afterwards the poly-
mer was end-capped by first using bromobenzene followed
by phenylboronic acid. The polymer was precipitated in
methanol/2n hydrochloric acid (3:1), filtered, and dried at
ꢀ
tration and drying in vacuum at 40 C. The yield was 1.13 g
(81.0 %).
Elemental analysis: Found—C: 79.66%, H: 9.53%, S: 2.83%.
1H NMR (500 MHz, CDCl3): d [ppm]: 7.67 (m, 6H, Ar-H),
7.40 (m, Th-H), 7.30 (m, Th-H), 7.12 (m, Th-H), 2.05 (m, 4H,
a-CH2A), 1.12 (m 20H, ACH2A), 0.80 (m, 10H, ACH2ACH3).
5-Bromothiophene-Terminated F8-block. 1.13 g of thio-
phene-terminated F8-block was placed in a reaction vessel
which was filled with argon. The polymer was dissolved in
100 mL of tetrahydrofurane and the solution was cooled to
ꢀ
40 C. The polymer was diluted in chloroform and extracted
with diluted ammonia. The organic phase was separated,
washed to neutralize, precipitated, filtered, and dried. Fur-
ther purification was done by soxhlet extraction using sol-
vents with increasing dissolving power to separate the low
molecular fractions. The yield was 1.25 g (68.1%).
0
ꢀC. 0.461 g (2.589 mmol) of N-bromosuccinimide was
added and the reaction mixture was stirred for 3 h at 0 ꢀC.
The reaction was stopped with distilled water. Chloroform
was added to extract the organic components. The organic
phase was washed with water, dried, and evaporated. Fur-
ther purification was achieved with column chromatography
by using ethyl acetate/hexane (1:10) as the mobile phase
and a short silica gel column as solid phase.
1H NMR (500 MHz, CDCl3): d [ppm]: 8.10 (m, 2H, Ar-H),
8.03 (m, 2H, Ar-H), 7.95 (m, 4H, Ar-H), 2.15 (m, 4H, a-
CH2A), 1.11 (m, 20H, ACH2A), 0.81 (m, 10H, ACH2ACH3).
OFET Fabrication
Bottom gate OFET configurations were used in conjunction
with heavily doped Si-wafers as the substrates and common
Elemental analysis: Found—C: 74.41%, H: 8.93%, S: 2.64%.
1H NMR (500 MHz, CDCl3): d [ppm]: 7.67 (m, 6H, Ar-H),
7.13 (m, Th-H), 7.06 (m, Th-H), 2.05 (m, 4H, a-CH2A), 1.12
(m 20H, ACH2A), 0.80 (m, 10H, ACH2ACH3).
gate electrode. The gate dielectric was thermally grown SiO
2
(230 nm) with capacitance 14.6 nF/cm2. On top of SiO ,
2
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635
1633

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
source and drain electrodes (Au) were deposited using pho-
tolithography (L ¼ 10 lm, W ¼ 195 mm). The wafers were
cleaned with several solvents, dried under nitrogen flow, and
treated with HMDS. Finally, the polymer active layers were
spin-coated from chloroform solutions with concentrations
of 7–10 mg/mL under inert conditions. The final film thick-
ness was about 70 nm. Before electrical characterization, the
samples were annealed in inert atmosphere. No sample
encapsulation was used. The OFET characteristics were
measured in dark under glove box conditions using two
Source-Measure Units 236 combined with a Trigger-Control
Unit 2361 and Metrics Software (all three items were from
Keithley Instruments). Usually about 10 transistors were
investigated for each polymer.
contrast mode. Absorbance properties of the polymer films
were measured with an ultra violet-visible (UV-Vis) spec-
trometer from Perkin Elmer (Lambda 950).
External Quantum Efficiency (EQE)
External quantum efficiencies were measured in the range
from 300 to 950 nm. The solar cells were probed with
chopped monochromatic light from a xenon short arc lamp
or a halogen lamp. The resulting current was recorded using
a lock-in amplifier. The photon flux incident on the sample
was measured using a calibrated silicon-diode.
Photoluminescence
Photoluminescence measurements were carried out in a
spectrofluorometer (Fluorolog-3, Horiba Jobin Yvon) at room
temperature. The excitation wavelength was set from a dou-
ble monochromator equipped with a xenon lamp. The emis-
sion from the sample was guided through a single mono-
chromator and recorded by a photomultiplier.
Cyclovoltammetry
Voltammograms were obtained on an EG and G Parc model
273 potentiostat. A three-electrode configuration was applied
in an undivided cell consisting of a glassy carbon electrode
(area 0.5 cm2) in which the polymer film was deposited, a
platinum mesh was used as the counter electrode and Ag/
AgCl (3 M NaCl and sat. AgCl) was employed as the refer-
ence electrode.
CONCLUSIONS
In this work, we presented the results of the synthesis of
novel donor–acceptor copolymers and their use in photovol-
taic cells. PCPDTBT and F8BT (F8) were chosen as comono-
mer units and they were arranged in different copolymer
sequences. We compared the feasibility of homopolymer mix-
tures, a random copolymer and a block copolymer, as the
electron donor in blends with PC71BM for solar cells. We
were able to assess the impact of molecular arrangement of
the monomer sequences in the copolymers on device per-
formance. The block copolymer had a slightly higher OFET
mobility than what was measured for the PCPDTBT homo-
polymer and for the random copolymer. From surface topog-
raphy studies we infer no macroscopic differences between
the phase separation modes in blends with the fullerene.
This is also confirmed by PL measurements on poly-
mer:PC71BM films, showing fairly good exciton collection and
charge transfer from the donor to the acceptor. The best
photovoltaic performance of about 2%, found for the
PCPDTBT-block-F8 polymer, is a direct result of an enhanced
fill factor with respect to the other materials investigated.
The results suggest a significant compatibility between the
two building blocks PCPDTBT and F8BT, allowing a relatively
fine tuning of the optical and electrical properties of the low
band gap PCPDTBT, by selecting the copolymerization strat-
egy and by tuning the dilution with the wide-band gap F8BT
component.
Bu4NBF4 (0.1 M) in acetonitrile was used as electrolyte and
prior to each measurement the electrochemical cell was
deoxygenated with nitrogen. The electrochemical cell was
calibrated with a ferrocene standard and the ferrocene half-
wave potential was estimated to be 435 mV for this assem-
bly. Polymer solutions (1 wt %) in CHCl3 were prepared and
5 lL was deposited on the glassy carbon electrode. The pre-
pared electrodes were kept under vacuum and dried at 60
ꢀC for 2 h.
Photoelectron Spectroscopy
Photoelectron spectroscopy was applied for the estimation of
the HOMO energy levels using a Riken Keiki AC-2. The poly-
mers were measured as powders placed in stainless steel
crucibles.
Solar Cell Fabrication and Characterization
Glass substrates coated with indium tin oxide (ITO) were
obtained commercially and rinsed with isopropanol before
spin coating. PEDOT:PSS (AI 4083, H.C. Stark) was spun in
air and annealed at 180 ꢀC for 15 min in N2. The polymers
used in this work were blended with PC71BM in chloroben-
zene with 3 vol % DIO and stirred overnight. The ratio of
polymer to PC71BM was 1:3 and the total solution concentra-
tion was 2.4 wt %. Polymer–fullerene active layers were
spun at 500, 1000, or 2000 rpm in N2. The cathode struc-
ture was thermally evaporated at 10ꢂ6 mbar and the result-
ing device areas were 16 or 59.5 mm2. The solar cell struc-
ture used in this report is shown in Figure 1. Current
density–voltage characteristics were measured under illumi-
nation with a light source from K.H. Steuernagel (100 mW/
cm2). The intensity of the light source was calibrated with a
silicon reference diode from Fraunhofer ISE. Individual de-
vice characteristics were not corrected for spectral mis-
match. Surface topographies were analyzed with an atomic
force microscope (AFM) from Nanosurf in noncontact/phase
A. Lange, H. Krueger, and M. Morana gratefully acknowledge
funding from the EC-project ‘‘OLAtronics,’’ (grant agreement
number: 216211). The authors especially thank Eileen Kathol-
ing and Lica Pabel (Fraunhofer IAP) for their assistance in
preparing the polymer samples. A. Lange acknowledges
Dieter Neher (University of Potsdam) and Michael Wegener
(Fraunhofer IAP) for fruitful discussions. B. Ecker and E. von
Hauff acknowledge funding of the EWE-Nachwuchsgruppe
‘‘Du¨nnschichtphotovoltaik’’ by the EWE AG Oldenburg and the
Federal Ministry of Education and Research (BMBF) (EOS).
1634
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
20 Shin, M.; Kim, H.; Kim, Y. Mater. Sci. Eng. B 2010, 176,
REFERENCES AND NOTES
382–386.
1 Brabec, C. J.; Sariciftci, N. S.; Hummelen, J. C. Adv. Funct.
Mater. 2001, 11, 15–26.
21 Chen, L.; Yang, L.; Shim, M.; Chen, H. Sol. Energy Mater.
Sol. Cells 2010 94, 2244–2250.
2 Brabec, C. J.; Gowrisanker, S.; Halls, J. J. M.; Laird, D.; Jia,
S.; Williams, S. P. Adv. Mater. 2010, 22, 3839–3856.
22 McNeill, C.R.; Greenham, N.C. Adv. Mater. 2009, 21, 3840–3850.
€
23 Inganas, O.; Zhang, F.; Tvingstedt, K.; Andersson, L. M.;
3 Bredas, J. L.; Norton, J. E.; Cornil, J.; Coropceanu, V. Acc.
Chem. Res. 2009, 42, 1691–1699.
€
Hellstrom, S.; Andersson, M. Adv. Energy. Mater. 2009, 22,
E100–E116.
4 Sariciftci, N. S.; Smilowitz, L.; Heeger, A. J.; Wudl, F. Science
1992, 258, 1474–1476.
24 Zhu, Z.; Waller D.; Gaudiana, R.; Morana, M.; Muhlbacher,
D.; Scharber, M.; Brabec, C. 2007, 40, 1981–1986.
¨
5 Ma, W.; Yang, C.; Xiong, G.; Lee, K.; Heeger, A. Adv. Funct.
Mater. 2005, 15, 1617–1622.
25 Tsao, H.N.; Cho, D.; Wenzel Andreasen, J.; Rouhanipour, A.;
Breiby, D.W.; Pisula, W.; Mullen, K., Adv. Mater. 2009, 21,
209–212.
¨
6 Muhlbacher, D.; Scharber, M.; Morana, M.; Zhu, Z.; Waller,
¨
D.; Gaudiana, R.; Brabec, C. Adv. Mater. 2006, 18, 2884–2889.
26 Winfield, J.M.; Van Vooren, A.; Park, M.-J.; Hwang, D.-H.; Comil,
J.; Kim, J.-S.; Friend, R.H. J. Chem. Phys. 2009, 131, 035104 -1–5.
7 Peet, J.; Kim, J.; Coates, N.; Ma, W.; Moses, D.; Heeger, A.;
Bazan, G. Nat. Mater. 2007, 6, 497–500.
27 Coffin, R.; Peet, J.; Rogers, J.; Bazan, G. Nat. Chem. 2009, 1,
657–661.
8 Boland, P.; Lee, K.; Gon Namkoong, G. Sol. Energy Mater.
Sol. Cells 2010, 94, 915–920.
28 Di Nuzzo, D.; Aguirre, A.; Shahid, M.; Gevaerts, V.; Meskers,
S.; Janssen, R. Adv. Mater. 2010, 22, 4321–4324.
9 Chen, H.-Y.; Hou, J.; Hayden, A.; Yang, H.; Houk, K.; Yang, Y.
Adv. Mater. 2010, 22, 371–375.
29 Scharber, M.; Koppe, M.; Gao, J.; Cordella, F.; Loi, M.; Denk,
P.; Morana, M.; Egelhaaf, H.-J.; Forberich, K.; Dennler, G.;
Gaudiana, R.; Waller, D.; Zhu, Z.; Shi, X.; Brabec, C. Adv. Mater.
2010, 22, 367–370.
10 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.
11 Soci, C.; Hwang, I.-W.; Moses, D.; Zhu, Z.; Waller, D.; Gaudi-
ana, R.; Brabec, C.; Heeger, A. Adv. Funct. Mater. 2007, 17,
632–636.
30 Jarzab, D.; Cordella, F.; Gao, J.; Scharber, M.; Egelhaaf,
H.-J.; Loi, M. Adv. Energy Mater. 2011, 1, 604–609.
12 Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.-T.; Wu, Y.; Li, G.; Ray, C.;
Yu, L. Adv. Energy Mater. 2010, 22, E135–E138.
31 Morana, M.; Azimi, H.; Dennler, G.; Egelhaar, H.J.; Scharber,
M.; Forberich, K.; Hauch, J.; Gaudiana, R.; Waller, D.; Zhu, Z.;
Hingerl, K.; van Bavel, S.S.; Loos, J.; Brabec, C.J. Adv. Funct.
Mater. 2010, 20, 1180–1188.
13 Zhokhavets, U.; Erb, T.; Gobsch, G.; Alibrahim, M.;
Ambacher, O. Chem. Phys. Lett. 2006, 418, 347–350.
14 Kim, C. S.; Kim, D.-Y.; Lee, S.; Kim, Y. S.; Loo, Y.-L Org.
Electron 2009, 10, 1483–1488.
32 Teichler, A.; Eckardt, R.; Hoeppener, S.; Friebe, C.; Perelaer,
J.; Senes, A.; Morana, M.; Brabec, C.; Schubert, U. Adv. Energy
Mater. 2011, 1, 105–114.
15 Li, G.; Shrotriya, V.; Huang, J.; Yao, Y.; Moriarty, T.; Emery,
K.; Yang, Y. Nat. Mater. 2005, 4, 864–868.
33 Chen, F.C.; Tseng, H.C.; Ko, C.J. Appl. Phys. Lett. 2008, 92,
16 Moet, D.; Lenes, M.; Morana, M.; Azimi, H.; Brabec, C.;
Blom, P. Appl. Phys. Lett. 2010, 96, 213506–1-3.
103316–1-3.
34 Li, G.; Shrotriya, V.; Huang, J.; Yao, Y.; Moriarty, T.; Emery,
K.; Yang, Y. Nat. Mater. 2005, 4, 864–868.
17 Lee, J. K.; Ma, W.; Brabec, C.; Yuen, J.; Moon, J. S.; Kim, J.
Y.; Lee, K.; Bazan, G.; Heeger, A. J. Am. Chem. Soc. 2008, 130,
3619–3623.
35 Yoa, Y.; Hou, J.; Xu, Z.; Li, G.; Yang, Y. Adv. Funct. Mater.
2008, 18, 1783–1789.
18 Moule, A.; Meerholz, K. Adv. Mater. 2008, 20, 240–245.
36 Campoy-Quiles, M.; Ferenczi, T.; Agostinelli, T.; Etchegoin,
P.G.; Kim, Y. Anthopoulos, T.D.; Stavrinou, P.N.; Bradley,
D.D.C.; Nelson, J. Nat. Mater. 2008, 7, 158–164.
19 Chen, H; Hou, J; Hayden, A; Yang, H; Houk, K; Yang, Y.
Adv. Mater. 2010, 22, 371–375.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1622–1635
1635