Effects of solubilizing group modification in fullerene bis-adducts on normal and inverted type polymer solar cells

Article abstract of DOI:10.1021/cm3010369

Structural control of solubilizing side groups in fullerene-based electron acceptors is critically important to optimize their performance in bulk heterojunction (BHJ)-type polymer solar cell (PSC) devices. The structural changes of fullerene derivatives

Full text of DOI:10.1021/cm3010369

Article
pubs.acs.org/cm
Effects of Solubilizing Group Modification in Fullerene Bis-Adducts
on Normal and Inverted Type Polymer Solar Cells
Ki-Hyun Kim,§,† Hyunbum Kang, Hyeong Jun Kim, Pan Seok Kim, Sung Cheol Yoon,*
§,†
†
‡
,‡
,
†
and Bumjoon J. Kim*
†Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon
3
05-701, Korea
‡
Advanced Materials Division, Korea Research Institute of Chemical Technology, Daejeon 305-600, Korea
S Supporting Information
*
ABSTRACT: Structural control of solubilizing side groups in fullerene-based
electron acceptors is critically important to optimize their performance in bulk
heterojunction (BHJ)-type polymer solar cell (PSC) devices. The structural
changes of fullerene derivatives affect not only their optical and electrochemical
properties but also their solubility and miscibility with electron donor polymers.
Herein, we synthesized a series of o-xylenyl C bis-adduct (OXCBA) derivatives
6
0
with different solubilizing side groups to systematically investigate the effects of
fullerene derivative structures on the photovoltaic properties of PSCs. The
xylenyl side groups on the OXCBA were modified to produce several different
OXCBA derivatives in which the xylenyl groups were functionalized with
fluorine (FXCBA), nitro (NXCBA), methoxy and bromine (BMXCBA), and
phenyl groups (ACBA). End group modifications of OXCBA dramatically affect
photovoltaic performance in blend films with poly(3-hexylthiophene) (P3HT),
resulting in power conversion efficiencies (PCEs) ranging from 1.7 to 5.3%. We found that this large range in PCE values is
mainly due to differences in the blend morphology and interfacial area of the P3HT:OXCBA derivative films caused by changes
in the hydrophobicity of the OXCBA derivatives and their interaction with P3HT. The trend in photovoltaic performance of the
different OXCBA derivatives agrees well with those of the interfacial tension, PL quenching, and exciton dissociation probability,
which suggests that changes in the interaction with P3HT are largely responsible for their photovoltaic performances. Finally, the
OXCBA derivatives were applied in inverted type PSC devices. We note that P3HT:OXCBA blend devices exhibited more than
5
% PCE with excellent air stability, which is one of the best inverted type devices based on the P3HT polymer in a simple device
architecture without any extra interlayers.
KEYWORDS: electron acceptors, solubilizing groups, fullerene bis-adducts, interfacial tensions, polymer solar cells
1
. INTRODUCTION
The phase separation behavior in the conjugated polymer
donor and fullerene acceptor blend has been intensively studied
The current state-of-the-art polymer solar cells (PSCs) are of a
type known as bulk heterojunction (BHJ), which consists of a
simple mixture of the electron donating conjugated polymers
in terms of structural modification of the conjugated polymers,
6
,11−17
such as changing the number and length of side chains
and the substituents on the polymer backbone.
18−23
1
−8
Even small
and electron accepting soluble fullerene derivatives.
In BHJ-
changes in a conjugated polymer’s structure can change its
interaction with electron acceptors, resulting in dramatic
type PSCs, an interpenetrating donor−acceptor blend network
with a small domain length-scale in the active layer is a critical
requirement for enhancing device efficiency because separation
of photogenerated excitons can only occur at the donor/
differences in blend morphology and solar cell perform-
24,25
ance.
For example, Kim et al. reported that end group
modification of P3HT greatly affected the various photovoltaic
characteristics of a P3HT/PCBM blend and produced power
2
,9,10
acceptor interface.
Thus, a nanometer-scale bicontinuous
morphology suitable for enhancing exciton separation efficiency
is essential for high device performance in PSCs; however,
morphology control is not trivial because it is dependent on
various parameters affecting both thermodynamic interactions
between donor, acceptor, and solvent and morphology
development kinetics. In particular, the interaction between
electron donor and acceptor is the primary determinant of the
phase separation in the blend.
conversion efficiencies (PCEs) ranging from 2.1% (−OH end
24
group) to 4.5% (−CF end group)).
3
Most research to date has focused on structural modification
of electron donor polymers. While tuning the chemical
Received: April 3, 2012
Revised: May 23, 2012
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/cm3010369 | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
structure of fullerene derivatives could affect the solubility and
measurements of the surface tension of OXCBA derivatives
and their corresponding interfacial tensions with P3HT. The
results were supported by atomic force microscopy (AFM) and
photoluminescence (PL) quenching experiments. In addition,
the probability of exciton dissociation in the P3HT:OXCBA
derivative PSCs was calculated to monitor exciton dissociation
and charge recombination at the P3HT−OXCBA derivative
interface, which are in turn significantly influenced by the blend
morphology. Finally, we also fabricated inverted type
P3HT:OXCBA derivative PSCs, of which the P3HT:OXC-
BA-based devices exhibited greater than 5% in PCE with
excellent stability, which is among the best simple inverted type
P3HT-based PSC PCEs reported to date with basic device
structure.
26−29
miscibility with donor polymers,
systematic study of the
effects of structural modification of fullerene derivative electron
acceptors in PSCs has been very limited. Moreover, small-
molecule fullerene derivatives retain a much higher molecular
mobility, resulting in the tendency to diffuse out of the polymer
30−34
matrix, thus seriously affecting the BHJ blend morphology.
Prior investigations have observed that slight modifications of
6,6]-phenyl-C -butyric acid methyl ester (PCBM) by
[
6
1
3
5
replacement of the phenyl group with a thienyl group,
3
6
36
triphenyl amine, or fluorene group led to significant changes
in its solubility and morphology in blend films, resulting in large
variations in PCE. Additionally, modification of the aliphatic
3
7,38
chain and ester group
of PCBM can also influence the
lowest unoccupied molecular orbital (LUMO) level and thus
the open circuit voltage (VOC) and PCE. However, the
underlying reasons for such changes in blend morphology
and PCE upon the modification of a fullerene derivative’s
structure have not yet been clearly demonstrated. Therefore,
careful and systematic investigation of the molecular structure−
device function relationship, especially with respect to the
structure of fullerene derivatives and its effect on their
interaction with electron donor polymers, is essential to
establish a design rule for fullerene-based electron acceptors
in PSCs.
2
. RESULTS AND DISCUSSION
2.1. Synthesis. The synthetic scheme and molecular
structures of the five different OXCBA derivatives are shown
in Scheme 1 (see the Supporting Information for details). 2,3-
Scheme 1. (a) Synthetic Scheme and (b) Molecular
Structures of OXCBA Derivatives
Of the various types of fullerene-based electron acceptors, we
have chosen o-xylenyl C60 bis−adducts (OXCBA) as a model
system in which to introduce various end-functionalizations and
thereby study the effect of structural changes in the fullerene
acceptor. Previous work reported that fullerene bis-adduct
electron acceptors effectively can enhance the Voc and PCE
39−41
because of their higher LUMO levels than that of PCBM.
42
43
Recently, our group and another group have reported the
use of α,α′-dibromo-o-xylene as a building block for the facile
preparation of OXCBA. Because OXCBA has a higher LUMO
level than PCBM (by 0.17 eV), P3HT:OXCBA devices showed
PCE of 5.31% with a VOC of 0.83 V, nearly 50% higher than
those of P3HT:PCBM control devices (3.68% with VOC of 0.59
V). In addition to high efficiency, the OXCBA model system
has the strong advantages of excellent reaction control and mild
reaction conditions, allowing variation in the OXCBA structure
Dimethylnaphthalene and three o-xylene derivatives function-
alized at the 4-position of phenyl ring with fluorine, methoxy,
and nitro groups were each reacted with N-bromosuccimide
(NBS) in the presence of benzoyl peroxide (BPO) in 1,2-
dichloroethane at reflux to form building blocks for reaction
with C . After purification by silica column chromatography,
60
1
(
i.e., end group modification of the solubilizing group) by
the synthesized building blocks were verified with H-NMR
spectroscopy (see Figure S1 in the Supporting Information).
The various α,α′-dibromo-o-xylene derivatives were reacted
with C60 using a Diels−Alder reaction procedure previously
simply tuning the structure of the α,α′-dibromo-o-xylene
precursor. The synthetic simplicity and tunability of OXCBA
makes it very attractive as a model system for exploration of
molecular structure−device function correlations. Furthermore,
the effect of end group modification in fullerene bis-adducts, to
the best of our knowledge, has not been reported elsewhere.
Herein, we introduce various substituents onto the xylenyl
end group of OXCBA to systematically examine the effects of
fullerene bis-adduct structural changes on interactions with
electron donor polymers and their photovoltaic properties. The
structure of OXCBA was carefully modified on the xylenyl
group with fluorine, bromine, methoxy, phenyl, and nitro
functional groups. Insertion of such functional groups changes
the interactions of the fullerene bis-adducts with the P3HT
electron donor polymer and thus the P3HT:OXCBA derivative
blend morphology, resulting in a range of normal type BHJ-
PSC PCEs from 1.7 to 5.3%. We found that the differences
were largely caused by changes in the hydrophobicity of the
OXCBA derivatives and thus the phase segregation behavior
between OXCBA derivatives and P3HT. Direct evidence of
changes in the interfacial interaction was provided by
4
2
reported for OXCBA synthesis. The desired OXCBA
derivative products were purified using flash silica column
chromatography (hexane/toluene gradient eluent system). The
successfully synthesized regioisomeric OXCBA derivatives, 4-
fluoro-1,2-xylenyl C60 bis-adduct, dihydroanthracyl-based C60
bis-adduct, 1-bromo-2-methoxy-4,5-xylenyl C60 bis-adduct, and
4-nitro-1,2-xylenyl C bis-adduct, were confirmed to have over
60
1
98% purity by H−NMR spectroscopy, MALDI−TOF mass
spectrometry, and elemental analysis (see Figure S1 and S2 in
the Supporting Information). For convenience, these four
OXCBA derivatives are labeled FXCBA, ACBA, BMXCBA, and
NXCBA, as shown in Scheme 1b.
2.2. Electrochemical and Optical Properties of OXCBA
Derivatives in Solution. The LUMO levels of fullerene
derivatives as electron acceptor are one of the most critical
3
9,44,45
factors in determining the VOC values in PSCs.
The
electrochemical properties of OXCBA derivatives in o−
dichlorobenzene were therefore carefully measured via cyclic
B
dx.doi.org/10.1021/cm3010369 | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
voltammetry (CV). The electrochemical properties of these
fullerene derivatives are summarized in Table 1 and their CV
LiF/Al) type PSC from blends of five different OXCBA
derivatives with P3HT, and their performances were measured.
The same thermal annealing conditions (5 min at 150 °C) were
applied to all samples to optimize the device performance.
Figure 1a shows the current density−voltage curves of the
Table 1. Reduction Potentials and LUMO Energy Levels of
the Five OXCBA Derivatives
electron acceptors
E1 (V)
E2 (V)
Eredon (V)
LUMO (eV)
OXCBA
FXCBA
ACBA
−1.32
−1.25
−1.34
−1.26
−1.26
−1.70
−1.64
−1.72
−1.64
−1.76
−1.15
−1.10
−1.19
−1.12
−1.02
−3.65
−3.70
−3.61
−3.68
−3.78
BMXCBA
NXCBA
curves are shown in Figure S3 in the Supporting Information.
All OXCBA derivatives exhibit three quasi−reversible reduction
waves in the negative potential range between 0 to −2.5 V. As
shown in Table 1, FXCBA and BMXCBA exhibited very similar
first (E ) and second (E ) reduction potential peaks at
1
2
approximately −1.25 and −1.64 V, respectively; the reduction
potential peaks of OXCBA and ACBA were shifted to relatively
more negative values when compared with FXCBA and
BMXCBA.
The LUMO energy levels of the OXCBA derivatives were
on
estimated from their reduction onset potentials (E ). As
red
on
shown in Table 1, the Ered values of OXCBA, FXCBA, ACBA,
BMXCBA, and NXCBA are −1.15, −1.10, −1.19, −1.12, and
−
1.02 V, respectively, versus a ferrocene standard. Although
NXCBA has a LUMO level of −3.78 eV, similar to the LUMO
levels of fullerene mono−adducts previously reported (e.g.,
PCBM and o-xylenyl C monoadduct at −3.83 eV), FXCBA
42
6
0
and BMXCBA have LUMO values, −3.70 and −3.68 eV,
respectively, more similar to that of OXCBA (−3.65 eV). And
ACBA had a slightly higher LUMO level (−3.61 eV) than
OXCBA. Hummelen et al. previously reported that the LUMO
level of PCBM could be influenced by placing electron-
donating and electron-withdrawing substituents on its phenyl
46
ring, respectively increasing and decreasing its LUMO level.
Considering these results, the influence of the nitro (NO2)
group of NXCBA, a strong electron−withdrawing group,
considerably decreasing the LUMO level from −3.65
Figure 1. Current density−voltage characteristics of BHJ PSCs.
P3HT:OXCBA (red), P3HT:FXCBA (blue), P3HT:ACBA (green)
P3HT:BMXCBA (orange), and P3HT:NXCBA (black) under AM 1.5
(
OXCBA) to −3.78 eV is easily understandable. Similarly, the
slightly decreased LUMO level of FXCBA (−3.70 eV)
compared to OXCBA was due to the weak electron−
withdrawing characteristics of the fluorine atom. In contrast,
the resonance contributed by the additional phenyl ring of
ACBA increased the LUMO level (−3.61 eV) from that of
OXCBA. Finally, the effect of the methoxy group (a weak
electron−donating group) was counterbalanced by that of the
bromine atom (a weak electron−withdrawing group), leading
to the only slightly decreased LUMO level of BMXCBA (−3.68
eV).
The light absorption ability of the electron acceptors is also
of great importance in characterizing their contribution to JSC in
PSCs. The visible region absorption spectra (300−800 nm) of
OXCBA, FXCBA, ACBA, BMXCBA and NXCBA solutions in
o-dichlorobenzene are shown in Figure S4 in the Supporting
Information. The light absorption properties of the OXCBA
derivatives in solution differed very little.
−2
illumination at 100 mW cm ; (a) normal type device architecture and
(b) inverted type device architecture.
normal type BHJ solar cells under AM 1.5G illumination with
−2
an intensity of 100 mW cm . Each of the devices was prepared
under the same conditions, including the P3HT:OXCBA
derivative blend ratio (1:0.6 w/w) and solvent concentration.
The thicknesses of the P3HT:OXCBA derivative films showed
a similar value of approximately 100 nm. The characteristics of
the PSC devices are summarized in Table 2 and Table S1 in the
Supporting Information. The P3HT:OXCBA device exhibited a
PCE of 5.29%, which is comparable to that of our previous
4
2
report; however, the P3HT:OXCBA derivative devices
showed widely varying PCE values ranging from 1.7 to 5.3%.
The performance of the P3HT:OXCBA derivative devices
increased in the order NXCBA (PCE = 1.75%), BMXCBA
(PCE = 1.84%), ACBA (PCE = 4.27%), FXCBA (PCE =
4.62%), and OXCBA.
2.3. Photovoltaic Properties of OXCBA Derivatives in
Normal and Inverted Type PSCs. To elucidate the
relationship between the molecular structure of OXCBA
derivatives and the function of PSC devices, we fabricated
each normal (ITO/PEDOT:PSS/P3HT:OXCBA derivative/
Inverted type PSCs present advantages over those with a
normal geometry; for example, the inverted type PSCs is
required to be able to use roll-to-roll casting to deposit all layers
C
dx.doi.org/10.1021/cm3010369 | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Table 2. Normal and Inverted Type BHJ PSC Characteristics
Based on P3HT:OXCBA Derivatives (1:0.6 weight ratio, AM
surface tensions of five different OXCBA derivatives and P3HT
were obtained by contact angle measurements. The contact
angles of two different solvents (water and glycerol) on the
OXCBA derivative and P3HT films were measured and
compared (Table 3 and Figure S5 in the Supporting
1
.5G conditions)
PCE
(%)
VOC (V) JSC (mA cm−2
active layer (w/w)
)
FF
normal type
OXCBA
FXCBA
ACBA
0.84
0.77
0.86
0.74
0.55
0.82
0.74
0.82
0.66
0.57
10.53
10.52
9.64
5.88
7.35
9.78
9.05
8.65
4.30
4.12
0.60
0.57
0.52
0.42
0.43
0.63
0.59
0.55
0.53
0.60
5.29
4.62
4.27
1.84
1.75
5.05
3.95
3.89
1.51
1.41
Table 3. Contact Angle of Water and Glycerol and Surface
Tension of OXCBA Derivatives and P3HT
BMXCBA
NXCBA
OXCBA
FXCBA
ACBA
surface tension (mN m−1
electron acceptors θwater (deg) θGL (deg)
)
P3HT
98.2
91.1
88.6
88.2
87.4
86.1
93.2
77.8
76.7
72.6
71.7
69.3
20.1
27.8
28.1
31.1
31.6
33.2
inverted type
OXCBA
FXCBA
ACBA
BMXCBA
NXCBA
BMXCBA
NXCBA
Information). From contact angle data, the surface tension of
onto flexible substrates by solution processing, enabling
47
different OXCBA derivatives can be calculated using the Wu
commercial high-speed mass production. In addition, because
of the relatively lower surface tension of polymer donor (i.e.,
P3HT) compared with fullerene acceptors, the vertically
phase−separated morphology in inverted type devices is
generally more favorable to the transportation of free charge
carriers to their respective electrodes than in normal type
53
model.
d
d
p
p
4
d
γ
γ
4γ
γ
water
water
γwater(1 + cos θwater) =
+
d
p
p
γ
+ γ
γ
+ γ
water
water
(1)
4
8
d
d
p
p
devices. For these reasons, inverted (ITO/ZnO/P3HT:OXC-
BA derivative/PEDOT:PSS/Ag) type PSCs were also fab-
ricated. The device characteristics are shown in Figure 1b and
Table 2. Similar to the normal type PSCs, the inverted type
PSCs showed large variations in the PCE values, which ranged
from 1.4 to 5.1%. Interestingly, the performance of inverted
type P3HT:OXCBA derivative devices increased in the same
order that the normal type PSC device did. The performance of
inverted type device increased in the order NXCBA (PCE =
4γ γ
4γ γ
GL
GL
γ (1 + cos θ ) =
+
GL
GL
d
d
p
p
γ
+ γ
γ
+ γ
GL
GL
(2)
(3)
total
d
p
γ
= γ + γ
total
where γ is the total surface tension of P3HT and an OXCBA
derivative; γ and γ are the dispersion and polar components of
d
p
total
γ
; γ is the total surface tension of the i material (i = water or
i
d
p
glycerol); γ and γ are the dispersion and polar components
i
i
1
.41%), BMXCBA (PCE = 1.51%), ACBA (PCE = 3.89%),
of γ ; and θ is the droplet contact angle (water or glycerol) on
the P3HT and OXCBA derivative film.
i
FXCBA (PCE = 3.95%), and OXCBA (PCE = 5.05%). The
electrical, optical and morphological differences of OXCBA,
FXCBA, ACBA, BMXCBA, and NXCBA in blends with P3HT
will be discussed in the following section to elucidate the reason
for the changes in the photovoltaic parameters in each device.
Of particular interest is the PCE greater than 5% shown by the
P3HT:OXCBA inverted type PSC. To the best of our
knowledge, this is the highest PCE reported to date for
polythiophene:fullerene inverted PSCs with a basic architecture
Table 3 summarizes the measured water and glycerol contact
angles on the each OXCBA derivative and pristine P3HT films.
Each OXCBA derivative film shows a different water contact
angle (OXCBA (91.1°), FXCBA (88.6°), ACBA (88.2°),
BMXCBA (87.4°), and NXCBA (86.1°)) and a corresponding
increase in hydrophobicity in the order NXCBA, BMXCBA,
ACBA, FXCBA, and OXCBA. The same trend was observed
for glycerol contact angles; the surface tensions of the OXCBA
derivatives were also calculated and are compared in Table 3.
The surface tensions of OXCBA derivatives decreased in the
order NXCBA, BMXCBA, ACBA, FXCBA, and OXCBA,
(
ITO/electron transporting layer/active layer/hole transport-
49−51
ing layer/anode) without any further modifications.
2.3.1. Effects of OXCBA Derivatives on the Fill Factor. As
−1
shown in Table 2, both normal and inverted type
P3HT:OXCBA derivative devices showed similar trends in
their VOC and JSC values, but not the FF values. While an
increase in FF values was observed for all P3HT:OXCBA
derivative inverted geometry compared with those in the
normal geometry, the FF values of P3HT:BMXCBA and
P3HT:NXCBA devices were the most significantly improved,
from 0.42 to 0.53 and from 0.43 to 0.60, respectively. It is well-
known that the difference in surface tensions between P3HT
and PCBM causes a vertical concentration gradient in normal
geometry PSCs varying from PCBM-rich near the bottom
ranging from 33.2 to 27.8 mN m . Among the five OXCBA
derivatives, the surface tension of BMXCBA and NXCBA (31.6
−1
and 33.2 mN m , respectively) are higher than those of
−
1
OXCBA and FXCBA (27.8 and 28.1 mN m , respectively).
For comparison, the surface tension of a P3HT film was also
−1
determined, 20.1 mN m , which agrees well with the literature
5
4,55
value.
Because the surface tension of P3HT is lower than
that of all of the OXCBA derivatives, the P3HT and OXCBA
derivative materials are expected to preferentially segregate
toward the top air surface and the bottom of the active layer,
respectively. This vertical phase separation leads to higher FF
values in the inverted type geometry for all P3HT:OXCBA
derivatives. In addition, the larger surface tension difference
between P3HT and OXCBA derivatives in P3HT:BMXCBA
and P3HT:NXCBA films provides a clear explanation for the
significantly larger enhancement in the FF values of the
inverted type PSCs. (from 0.42 to 0.53 for P3HT:BMXCBA
52
PEDOT:PSS interface to P3HT-rich near the top air surface.
The preformed PCBM aggregates and resulting P3HT
crystallites thus form percolation pathways for both carriers
with a vertical phase-separated morphology favorable in
particular to the inverted structure. Therefore, to gain insight
into the increasing FF numbers in the inverted type PSCs, the
48
D
dx.doi.org/10.1021/cm3010369 | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
and from 0.43 to 0.60 for P3HT:NXCBA devices) For example,
a stronger vertical segregation of P3HT and NXCBA can be
expected with NXCBA accumulated at the bottom (ZnO)
surface and P3HT at the air surface, where the existence of a
fullerene packing layer near the ZnO surface prevents charge
carrier recombination and facilitates electron transport, thus
leading to a higher FF. To find direct evidence of the vertical
composition profile of P3HT:OXCBA derivative films, depth
profile measurements using dynamic secondary ion mass
spectrometry is under investigation.
to those used for the optimized BHJ devices. These images
reveal significant differences in morphology upon changes in
the OXCBA derivative end groups. Panels a and b in Figure 2
clearly show that both the P3HT:OXCBA and P3HT:FXCBA
films display relatively better-mixed morphologies at a finer
length scale of 20−30 nm. We believe that these morphological
advantages in P3HT:OXCBA and P3HT:FXCBA blends
contributed to the increase in JSC and FF in BHJ PSC devices,
producing better overall photovoltaic performance. In contrast,
both P3HT:BMXCBA and P3HT:NXCBA films (Figure 2d, e)
−1
2
.3.2. Effects of OXCBA Derivatives on the Blend
with higher γ values of 6.3 and 7.3 mN m , respectively,
display poor morphologies with a larger domain size. In
addition, they exhibited relatively very rough surfaces when
compared with P3HT:OXCBA and P3HT:FXCBA films. rms
surface roughnesses of the P3HT:OXCBA and P3HT:FXCBA
films were 0.98 ± 0.11 and 1.33 ± 0.25 nm, respectively,
whereas those of P3HT:BMXCBA and P3HT:NXCBA films
were 6.37 ± 1.04 and 7.72 ± 0.88 nm, respectively. From these
results, we note that the trend in the rms roughness measured
via AFM in P3HT:OXCBA derivatives is consistent with the
trend in calculated interfacial tension values. Consequently, it is
believed that the limited interfacial area in the P3HT:BMXCBA
and P3HT:NXCBA films results in reduced photovoltaic
performance. In contrast, the small interfacial tension in
P3HT:OXCBA and P3HT:FXCBA films produced a relatively
well-mixed donor/acceptor blend, which facilitates efficient
exciton diffusion and charge transfer. Thus, morphological
differences within each P3HT:OXCBA derivative film can affect
charge separation and charge transfer ability between P3HT
Morphology with P3HT. The interfacial tensions (γ) between
the P3HT and the OXCBA derivatives provide direct
information about the miscibility of the electron donor and
acceptor components and their morphological behavior in
5
6
blend films. In this regard, the γ values between P3HT and
each OXCBA derivative were carefully calculated using the
57
following equation:
d
d
p
p
4
γ γ
4γ γ
1
2
1
2
γ = γ + γ −
−
1
2
1
2
d
d
p
p
γ + γ
γ + γ
(4)
1
2
1 2
where γ is the interfacial tension between P3HT (1) and each
12
OXCBA derivative (2); γ is the surface tension of the i material
i
d
p
(
i = 1 or 2); and γi and γi are the dispersion and polar
components of γ calculated using the contact angle with water
i
and glycerol (see Figure S5 in the Supporting Information).
The change of end group structure in the OXCBA derivatives
derives the significant change in the γ value between P3HT and
the OXCBA derivatives. In particular, two materials with a low
γ value could minimize the energy penalty required to create a
well−mixed interpenetrating network of donors and acceptors
with a large interfacial area. As shown in Table 4, the interfacial
58−60
and OXCBA derivatives.
To support the morphological trends of P3HT:OXCBA
derivative blend films, steady-state photoluminescence (PL)
measurements were performed, which provide useful informa-
tion on the charge (i.e., electron) transfer from P3HT to the
OXCBA derivatives. Figure 3 presents the room temperature
PL spectra obtained with an excitation wavelength of 460 nm
and Table 5 summarizes the PL quenching efficiency of each
P3HT:OXCBA derivative blend film. The P3HT:OXCBA
derivative blend samples were prepared by spin−coating on
glass substrates under optimized device conditions. The
fluorescence spectrum of a pristine P3HT film was measured
as a control sample. The PL quenching efficiency of different
P3HT:OXCBA derivative films was determined by comparing
the PL intensity from each P3HT:OXCBA derivative blend film
to the PL intensity of the pristine P3HT film. The PL spectra in
Figure 3 indicate dramatic PL quenching in P3HT:OXCBA
when compared with the P3HT:FXCBA, P3HT:ACBA,
P3HT:BMXCBA, and P3HT:NXCBA films. P3HT:OXCBA
emission was quenched by 82% relative to pristine P3HT but
P3HT:FXCBA and P3HT:ACBA were quenched slightly less,
by 76% and 75%, respectively. Furthermore, the PL quenching
efficiencies of P3HT:BMXCBA and P3HT:NXCBA were
significantly reduced, to 55% and 66%, respectively. These
dramatically decreased PL quenching abilities indicate that very
poor charge transfer takes place through the interface in the
Table 4. Calculated Interfacial Tension (γ) between P3HT
and Each OXCBA Derivative
electron acceptors
γ (mN m−1)
OXCBA
FXCBA
ACBA
4.1
3.0
6.2
6.3
7.3
BMXCBA
NXCBA
tensions of OXCBA and FXCBA in blends with P3HT (γ = 4.1
−1
and 3 mN m ) are much smaller than those of BMXCBA and
−1
NXCBA in blends with P3HT (γ = 6.3 and 7.3 mN m ). The
miscibility and blend morphology of P3HT and each OXCBA
derivative should thus be distinguishable from one another. For
instance, even under the same device processing conditions, the
P3HT:BMXCBA and P3HT:NXCBA blends (those with
higher γ values) should have a less homogeneous morphology
with smaller interfacial area, which would suppress donor−
acceptor charge transfer and exciton dissociation at the
interface. In contrast, P3HT:OXCBA and P3HT:FXCBA
(
those with lower γ values) could have potential to form a
2
4,61−63
relatively better-mixed donor/acceptor morphology.
P3HT:BMXCBA and P3HT:NXCBA films.
This PL
To gain direct evidence of change in the interfacial tension
for different P3HT/OXCBA derivative blend films, we
investigated the surface morphology of the active layer by
tapping−mode atomic force microscopy (AFM). Figure 2a−e
show P3HT blended with OXCBA, FXCBA, ACBA, BMXCBA,
and NXCBA, respectively, prepared under conditions identical
quenching efficiency trend between P3HT and OXCBA
derivatives is consistent with the trends in the blend
morphology and γ values. The intermixed blend morphology
with a smaller length scale produces a greater interfacial area,
which allows for efficient charge transfer between the donor
and acceptor domains.
E
dx.doi.org/10.1021/cm3010369 | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 2. AFM height images (top line) and phase images (bottom line) of the blend films of P3HT with each OXCBA derivative: (a) OXCBA, (b)
FXCBA, (c) ACBA, (d) BMXCBA, and (e) NXCBA. The scale bar is 500 nm. rms roughness: (a) 0.98 ± 0.11 nm, (b) 1.33 ± 0.25 nm, (c) 4.74 ±
0.16 nm, (d) 6.37 ± 1.04 nm, and (e) 7.72 ± 0.88 nm.
Figure 3. Static PL spectra for P3HT (black), P3HT:OXCBA (red),
P3HT:FXCBA (blue), P3HT:ACBA (green), P3HT:BMXCBA
(
orange), and P3HT:NXCBA (purple) films.
Table 5. Calculated PL Quenching Efficiency for
P3HT:OXCBA Derivatives
electron acceptors
Ifilm/IP3HT
1− (Ifilm/IP3HT
)
quenching eff. (%)
OXCBA
FXCBA
ACBA
0.184
0.235
0.250
0.451
0.336
0.816
0.765
0.750
0.549
0.664
81.6
76.5
75.0
54.9
66.4
BMXCBA
NXCBA
To further explore the effects of the OXCBA derivative end
group modification on the BHJ PSCs, the maximum exciton
generation rate (Gmax) and exciton dissociation probabilities
Figure 4. (a) Photocurrent density (J ) and (b) exciton dissociation
ph
probability (P(E,T)) plotted as a function of effective bias (V ) in
eff
P3HT:OXCBA derivative devices.
[
P(E,T)] were measured and compared for different
P3HT:OXCBA derivative devices. Each device was biased by
sweeping from +1 to −10 V and the saturation photocurrent
density (J ), G , and P(E,T) were successfully calculated
Jph = J − JD
L
(5)
6)
sat
max
Veff = V − V
(
61
0
a
using a published literature method. Figure 4a reveals the
dependence of the photocurrent density (J ) on the effective
where J and J are the current densities under illumination and
L D
ph
−2
voltage (V ), recorded under illumination at 100 mW cm , for
in the dark, respectively, V is the voltage when J = 0, and Va
eff
0 ph
our P3HT:OXCBA derivative devices. The values of J and V
can be determined by the following equations:
is the applied voltage. In the high Veff regime, Jph becomes
saturated and independent of the bias and temperature and
ph
eff
F
dx.doi.org/10.1021/cm3010369 | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
determines J . Assuming that all of the excitons become
dissociated and contribute to the current in the saturated
region,
in the Supporting Information), the UV−vis absorption spectra
of each P3HT:OXCBA derivative blend thin film exhibited
different 610 nm peak intensities, which originates from highly
ordered P3HT chains. Therefore, we suggest that the end
group modification of OXCBA was responsible in large part for
changes in the degree of P3HT packing in P3HT:OXCBA
derivative blend films because differences in the molecular size
of OXCBA derivatives and their corresponding intermolecular
packing structure could affect the P3HT packing and light
absorption properties of the blend film. To compare the
molecular size of the OXCBA derivatives, molecular size
calculations were performed using the Spartan’08 program and
the simulation parameters are shown in Table S2 in the
Supporting Information. Because of the existence of various
sat
J
= qGmaxL
(7)
sat
where q is the electronic charge and L is the thickness of the
active layer. Because only a portion of the excitons can be
dissociated into free carriers under applied voltage, Jph can be
expressed using P(E,T)
Jph = qGmaxP(E, T)L
(8)
As summarized in Table 6, the Gmax of different P3HT:OXCBA
derivative devices showed similar trends to the JSC values. For
64
regioisomers in fullerene bis-adducts, the absolute size
difference between OXCBA derivatives cannot be estimated.
Instead, we compared the size of the solubilizing group of the
different OXCBA derivatives and found that as the molecular
size of the OXCBA derivatives increased, the vibronic peaks
exhibited lower intensity. For example, addition of the relatively
bulkier BMXCBA and NXCBA reduced the peak intensities at
550 and 610 nm.
Table 6. Maximum Exciton Generation Rate (Gmax) and
P(E,T) for P3HT:OXCBA Derivative Devices
electron
acceptors
Jsat
G
P(E,T) under SC condition
max
(A m−2)
(m s−1
−3
)
(%)
2
7
7
7
7
7
OXCBA
FXCBA
ACBA
149
145
120
117
104
7.75 × 10
7.54 × 10
6.24 × 10
6.09 × 10
5.41 × 10
80.3
78.0
75.9
54.9
69.4
2
2
2
2
BMXCBA
NXCBA
In addition, the JSC value of PSC devices is influenced by not
only the Gmax value but also the exciton dissociation probability
[
P(E,T)]. The exciton dissociation probability can be obtained
example, the Gmax of the P3HT:BMXCBA and P3HT:NXCBA
from the normalized photocurrent density (J /J ). Of
particular importance is the P(E,T) value under short circuit
conditions (V = 0 V), which is directly proportional to J in
the solar cell device. Figure 4b shows the curve of P(E,T)
versus V (V) in P3HT:OXCBA derivative devices and Table 6
ph sat
2
7
27 −3 −1
devices (6.09 × 10 and 5.41 × 10 m s , respectively) with
lower JSC values exhibited relatively lower Gmax values compared
with the P3HT:OXCBA and P3HT:FXCBA devices (7.75 ×
a
SC
27
27
−3 −1
10
and 7.54 × 10
m
s , respectively) having higher JSC
eff
values. Because the Gmax value is a measure of the maximum
number of absorbed photons, the light absorption ability of
P3HT:OXCBA derivative films can be dependent upon the
structures of the OXCBA derivatives. The thin-film UV−vis
absorption spectra of the P3HT:OXCBA derivative blend films
fabricated under optimal device conditions were therefore
measured.
demonstrates that the values of P(E,T) under short−circuit
conditions (V = 0 V) provided a calculated exciton dissociation
a
probability decrease from 80.3% for the P3HT:OXCBA
reference device to 78.0, 75.9, 54.9, and 69.4% for the
P3HT:FXCBA, P3HT:ACBA, P3HT:BMXCBA, and
P3HT:NXCBA devices, respectively. It is worth noting that
the trend in exciton dissociation probability for different
P3HT:OXCBA derivative films agrees with those of PL
As shown in Figure 5, the intensity of the peaks at 550 and
6
10 nm in the UV−vis absorption spectra was gradually
quenching efficiency and J . Exciton dissociation and charge
SC
transfer between donor and acceptor occurs through the
interface area, which is mainly determined by the γ value
between the donor and acceptor materials. Our result therefore
shows that the trend of exciton dissociation probability and PL
quenching efficiency matching well with that of γ value is to be
expected. In addition, the difference in the dipole moment (see
Table S2 in the Supporting Information) of OXCBA derivatives
caused by end groups with different dipole moments and
polarities may contribute to J , specifically due to exciton
SC
18
dissociation and charge recombination. For example, the
charge separation would be more difficult for blends of P3HT
with BMXCBA and NXCBA, which have end groups
containing strong negative charges, because the dissociation
of excitons at the interface to produce free electrons can be
suppressed by the presence of strongly electronegative end
groups in BMXCBA and NXCBA near the interface.
Conversely, the end group of OXCBA does not have a
negative charge, leading to more efficient exciton dissociation at
the P3HT:OXCBA interface.
Figure 5. UV−vis absorption spectra of the blended thin films of
OXCBA, FXCBA, ACBA, BMXCBA, and NXCBA with P3HT.
decreased in the order P3HT:OXCBA, P3HT:FXCBA,
P3HT:ACBA, P3HT:BMXCBA, and P3HT:NXCBA. The
trend agrees well with those of Gmax and JSC for P3HT:OXCBA
derivative films. Interestingly, despite the similar light
absorption ability of five OXCBA derivatives (see Figure S4
2.3.3. Effects of OXCBA Derivatives on the Open Circuit
Voltage. It is well-known that VOC in BHJ devices is
proportional to the difference between the LUMO level of
the electron acceptor and the highest occupied molecular
44
orbital (HOMO) level of the electron donor. For this reason,
G
dx.doi.org/10.1021/cm3010369 | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
changes in the LUMO levels of OXCBA derivatives caused by
changes of the end group structure could explain the VOC
differences in P3HT:OXCBA derivative devices. In the case of
NXCBA, which has a very strong electron−withdrawing group,
the LUMO level (−3.78 eV) was significantly lower than that of
OXCBA (−3.65 eV). This lower LUMO level resulted in a
decrease in the difference between the LUMO level of the
fullerene derivative and the HOMO level of P3HT, and thus a
significant decrease in the VOC of the P3HT:NXCBA device
in OXCBA derivatives influence their structural and electrical
properties. More importantly, by simply varying the end
groups, the hydrophobicity of OXCBA derivatives can be easily
tuned; this allows systematic control of the miscibility and
interfacial interaction with P3HT, which were analyzed by
interfacial tension and AFM measurements. As the γ value
between P3HT and the OXCBA derivatives decreased, a
significant increase in the interfacial area was observed, leading
to changes in the PCE ranging from 1.7 to 5.3%. Such control
in the interfacial interaction due to end group modification
results in increased exciton dissociation and suppressed charge
recombination, as shown by increased PL quenching and
calculated exciton dissociation probability, producing higher JSC
values. In addition, changes in the molecular size of OXCBA
derivatives may have some contribution to the P3HT and
fullerene packing and thus the light absorption of the blend
films. End group modification of OXCBA significantly changes
the surface tension of each of the OXCBA derivatives, which
affects the vertical phase−separated morphology in the active
layer. Because the high surface tensions of BMXCBA and
NXCBA lead to accumulation at the bottom surface, the
P3HT:BMXCBA and P3HT:NXCBA inverted type devices
showed significantly increased FF values compared with their
normal type device. Finally, inverted type P3HT:OXCBA PSC
devices exhibited over 5% PCE with excellent air stability. Our
work provides a model system for fundamental understanding
of the molecular structure−device function relationship,
especially with respect to the structure of bis-adduct fullerenes
and their interactions with electron donor polymers, and
suggests a design rule for fullerene-based electron acceptors in
PSCs.
(
0.55 V). Similarly, the introduction of a weak electron
withdrawing group (fluorine group) in FXCBA resulted in a
slightly reduced LUMO energy level (−3.68 eV) compared to
that of OXCBA; accordingly, the P3HT:FXCBA device showed
a slightly decreased VOC (0.77 V) when compared with that of
the P3HT:OXCBA device (0.84 V). In contrast, the LUMO
level of ACBA, which has an additional resonance group
(
phenyl ring), increased to −3.61 eV, which results in the
slightly higher V (0.86 V) of the P3HT:ACBA device. For
OC
BMXCBA, the competing effects of the electron donating
methoxy and the electron withdrawing bromine groups leads to
a slight decrease in the LUMO level (−3.68 eV) that explains
the decrease in the V value.
OC
2
.4. Air Stability Measurement in Normal and
Inverted Type PSCs. The air stability of PSCs is an important
issue for commercialization. In general, inverted geometry
PSCs are more stable in air than the normal type structure
because their high-work-function metal anodes are resistant to
47
oxidation in air. Because of its similar excellent device
performances in both normal and inverted geometries, we
chose the P3HT:OXCBA blend to compare the air stability of
inverted type PSCs with that of normal type ones. Figure 6
ASSOCIATED CONTENT
■
*
S
Supporting Information
Materials and methods, detail experimental procedures, and
additional data (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
E-mail: bumjoonkim@kaist.ac.kr (B.J.K.); yoonsch@krict.re.kr
S.C.Y.).
*
(
Author Contributions
§
These authors contributed equally to this work.
Notes
Figure 6. Air stability tests of normal and inverted type PSCs based on
The authors declare no competing financial interest.
P3HT:OXCBA under ambient condition.
ACKNOWLEDGMENTS
shows the air stability of normal and inverted type PSCs based
on P3HT:OXCBA. The initial PCEs of normal and inverted
type devices exhibited similar values above 5%. Under ambient
atmosphere without any encapsulation, the inverted type PSC
exhibited excellent air stability, showing a PCE decrease of less
than 0.2% (greater than 95% of the initial PCE value) after 3
weeks of exposure to the air. In contrast, the PCE of the normal
type devices dropped to below 50% of the initial value within 2
days and to ∼0.5% within 5 days.
■
This research was supported by the Korea Research
Foundation Grant funded by the Korean Government
(2011−0030387, 2011-0027240), and the EEWS Research
Project of the Office of the KAIST EEWS Initiative (EEWS−
2012−N01120012). This research was also supported by the
New & Renewable Energy KETEP Grant (2011−T100100587)
and the Fundamental R&D Program Grant for Core
Technology of Materials funded by the Ministry of Knowledge
Economy, Republic of Korea.
3. CONCLUSION
REFERENCES
In summary, we synthesized a series of electron acceptors based
on OXCBA with various end groups such as fluorine, bromine,
methoxy, phenyl, and nitro groups. The changes of end groups
■
(1) Li, G.; Shrotriya, V.; Huang, J.; Yao, Y.; Moriarty, T.; Emery, K.;
Yang, Y. Nat. Mater. 2005, 4, 864.
H
dx.doi.org/10.1021/cm3010369 | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
(
2) Ma, W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. Adv. Funct.
Mater. 2005, 15, 1617.
3) Chu, T.-Y.; Lu, J.; Beaupre,
(34) Kim, H. J.; Han, A. R.; Cho, C.-H.; Kang, H.; Cho, H.-H.; Lee,
M. Y.; Frec
215.
́
het, J. M. J.; Oh, J. H.; Kim, B. J. Chem. Mater. 2011, 24,
(
́
S.; Zhang, Y.; Pouliot, J.-R. m.;
Wakim, S.; Zhou, J.; Leclerc, M.; Li, Z.; Ding, J.; Tao, Y. J. Am. Chem.
Soc. 2011, 133, 4250.
(35) Choi, J. H.; Son, K.-I.; Kim, T.; Kim, K.; Ohkubo, K.; Fukuzumi,
S. J. Mater. Chem. 2010, 20, 475.
(
36) Zhang, Y.; Yip, H.-L.; Acton, O.; Hau, S. K.; Huang, F.; Jen, A.
K. Y. Chem. Mater. 2009, 21, 2598.
37) Yang, C.; Kim, J. Y.; Cho, S.; Lee, J. K.; Heeger, A. J.; Wudl, F. J.
Am. Chem. Soc. 2008, 130, 6444.
38) Zheng, L.; Zhou, Q.; Deng, X.; Yuan, M.; Yu, G.; Cao, Y. J. Phys.
Chem. B 2004, 108, 11921.
39) He, Y. J.; Chen, H. Y.; Hou, J. H.; Li, Y. F. J. Am. Chem. Soc.
010, 132, 1377.
40) He, Y. J.; Zhao, G. J.; Peng, B.; Li, Y. F. Adv. Funct. Mater. 2010,
0, 3383.
41) Lenes, M.; Wetzelaer, G.; Kooistra, F. B.; Veenstra, S. C.;
Hummelen, J. C.; Blom, P. W. M. adv. Mater. 2008, 20, 2116.
42) Kim, K.-H.; Kang, H.; Nam, S. Y.; Jung, J.; Kim, P. S.; Cho, C.-
H.; Lee, C.; Yoon, S. C.; Kim, B. J. Chem. Mater. 2011, 23, 5090.
43) Voroshazi, E.; Vasseur, K.; Aernouts, T.; Heremans, P.;
(
4) Price, S. C.; Stuart, A. C.; Yang, L.; Zhou, H.; You, W. J. Am.
Chem. Soc. 2011, 133, 4625.
5) Zhou, H.; Yang, L.; Stuart, A. C.; Price, S. C.; Liu, S.; You, W.
Angew. Chem., Int. Ed. 2011, 50, 2995.
6) Liang, Y.; Feng, D.; Wu, Y.; Tsai, S.-T.; Li, G.; Ray, C.; Yu, L. J.
Am. Chem. Soc. 2009, 131, 7792.
(
(
(
(
(
2
(
2
(
(
7) Li, Y. Acc. Chem. Res. 2012, 45, 723.
8) Tan, Z. a.; Zhang, W.; Zhang, Z.; Qian, D.; Huang, Y.; Hou, J.; Li,
Y. Adv. Mater. 2012, 24, 1476.
9) Thompson, B. C.; Frechet, J. M. J. Angew. Chem., Int. Ed. 2008,
7, 58.
10) Kozub, D. R.; Vakhshouri, K.; Orme, L. M.; Wang, C.; Hexemer,
A.; Gomez, E. D. Macromolecules 2011, 44, 5722.
11) Cho, C.-H.; Kang, H.; Kang, T. E.; Cho, H.-H.; Yoon, S. C.;
Jeon, M.-K.; Kim, B. J. Chem. Commun. 2011, 47, 3577.
12) Thompson, B. C.; Kim, B. J.; Kavulak, D. F.; Sivula, K.; Mauldin,
C.; Frechet, J. M. J. Macromolecules 2007, 40, 7425.
13) Friedel, B.; McNeill, C. R.; Greenham, N. C. Chem. Mater. 2010,
2, 3389.
14) Ren, G.; Wu, P.-T.; Jenekhe, S. A. Chem. Mater. 2010, 22, 2020.
15) Nguyen, L. H.; Hoppe, H.; Erb, T.; Gunes, S.; Gobsch, G.;
Sariciftci, N. S. Adv. Funct. Mater. 2007, 17, 1071.
16) Ko, S.; Verploegen, E.; Hong, S.; Mondal, R.; Hoke, E. T.;
Toney, M. F.; McGehee, M. D.; Bao, Z. J. Am. Chem. Soc. 2011, 133,
6722.
17) Ko, S.; Hoke, E. T.; Pandey, L.; Hong, S.; Mondal, R.; Risko, C.;
Yi, Y.; Noriega, R.; McGehee, M. D.; Bredas, J.-L.; Salleo, A.; Bao, Z. J.
Am. Chem. Soc. 2012, 134, 5222.
18) Carsten, B.; Szarko, J. M.; Son, H. J.; Wang, W.; Lu, L.; He, F.;
Rolczynski, B. S.; Lou, S. J.; Chen, L. X.; Yu, L. J. Am. Chem. Soc. 2011,
33, 20468.
19) Cheng, Y.-J.; Chen, C.-H.; Lin, Y.-S.; Chang, C.-Y.; Hsu, C.-S.
Chem. Mater. 2011, 23, 5068.
20) Kim, B. J.; Miyamoto, Y.; Ma, B.; Frec
Mater. 2009, 19, 2273.
21) Griffini, G.; Douglas, J. D.; Piliego, C.; Holcombe, T. W.; Turri,
S.; Frechet, J. M. J.; Mynar, J. L. Adv. Mater. 2011, 23, 1660.
22) Son, H. J.; Wang, W.; Xu, T.; Liang, Y.; Wu, Y.; Li, G.; Yu, L. J.
Am. Chem. Soc. 2011, 133, 1885.
23) Osaka, I.; Shimawaki, M.; Mori, H.; Doi, I.; Miyazaki, E.;
Koganezawa, T.; Takimiya, K. J. Am. Chem. Soc. 2012, 134, 3498.
24) Kim, J. S.; Lee, Y.; Lee, J. H.; Park, J. H.; Kim, J. K.; Cho, K. Adv.
Mater. 2010, 22, 1355.
25) Park, J. K.; Jo, J.; Seo, J. H.; Moon, J. S.; Park, Y. D.; Lee, K.;
Heeger, A. J.; Bazan, G. C. Adv. Mater. 2011, 23, 2430.
26) Backer, S. A.; Sivula, K.; Kavulak, D. F.; Frechet, J. M. J. Chem.
Mater. 2007, 19, 2927.
27) Matsumoto, K.; Hashimoto, K.; Kamo, M.; Uetani, Y.; Hayase,
S.; Kawatsura, M.; Itoh, T. J. Mater. Chem. 2010, 20, 9226.
28) Shin, W. S.; Lee, J.-C.; Kim, J.-R.; Lee, H. Y.; Lee, S. K.; Yoon, S.
C.; Moon, S.-J. J. Mater. Chem. 2011, 21, 960.
29) Zhao, G.; He, Y.; Xu, Z.; Hou, J.; Zhang, M.; Min, J.; Chen, H.-
Y.; Ye, M.; Hong, Z.; Yang, Y.; Li, Y. Adv. Funct. Mater. 2010, 20, 1480.
30) Cheng, Y.-J.; Hsieh, C.-H.; Li, P.-J.; Hsu, C.-S. Adv. Funct. Mater.
011, 21, 1723.
31) Swinnen, A.; Haeldermans, I.; vande Ven, M.; D’Haen, J.;
(
4
(
́
(
(
(
(
Baumann, A.; Deibel, C.; Xue, X.; Herring, A. J.; Athans, A. J.; Lada,
T. A.; Richter, H.; Rand, B. P. J. Mater. Chem. 2011, 21, 17345.
(
́
(44) Brabec, C. J.; Cravino, A.; Meissner, D.; Sariciftci, N. S.;
(
2
(
(
Fromherz, T.; Rispens, M. T.; Sanchez, L.; Hummelen, J. C. Adv.
Funct. Mater. 2001, 11, 374.
(45) He, Y.; Li, Y. Phys. Chem. Chem. Phys. 2011, 13, 1970.
(46) Kooistra, F. B.; Knol, J.; Kastenberg, F.; Popescu, L. M.;
Verhees, W. J. H.; Kroon, J. M.; Hummelen, J. C. Org. Lett. 2007, 9,
551.
(47) Hau, S. K.; Yip, H.-L.; Jen, A. K. Y. Polym. Rev. 2010, 50, 474.
(48) Xu, Z.; Chen, L.-M.; Yang, G.; Huang, C.-H.; Hou, J.; Wu, Y.;
Li, G.; Hsu, C.-S.; Yang, Y. Adv. Funct. Mater. 2009, 19, 1227.
̈
(
1
(
(
49) Hsieh, C.-H.; Cheng, Y.-J.; Li, P.-J.; Chen, C.-H.; Dubosc, M.;
Liang, R.-M.; Hsu, C.-S. J. Am. Chem. Soc. 2010, 132, 4887.
50) Lin, Y.-H.; Yang, P.-C.; Huang, J.-S.; Huang, G.-D.; Wang, I.-J.;
Wu, W.-H.; Lin, M.-Y.; Su, W.-F.; Lin, C.-F. Sol. Energy Mater. Sol. Cells
011, 95, 2511.
51) Cheng, Y.-J.; Hsieh, C.-H.; He, Y.; Hsu, C.-S.; Li, Y. J. Am.
Chem. Soc. 2010, 132, 17381.
52) Campoy-Quiles, M.; Ferenczi, T.; Agostinelli, T.; Etchegoin, P.
́
(
(
2
(
1
(
(
(
́
het, J. M. J. Adv. Funct.
G.; Kim, Y.; Anthopoulos, T. D.; Stavrinou, P. N.; Bradley, D. D. C.;
Nelson, J. Nat. Mater. 2008, 7, 158.
(
(
(
53) Comyn, J. Int. J. Adhes. Adhes 1992, 12, 145.
54) Oh, J. Y.; Jang, W. S.; Lee, T. I.; Myoung, J.-M.; Baik, H. K. Appl.
́
(
Phys. Lett. 2011, 98, 023303.
55) Kim, J. B.; Allen, K.; Oh, S. J.; Lee, S.; Toney, M. F.; Kim, Y. S.;
Kagan, C. R.; Nuckolls, C.; Loo, Y.-L. Chem. Mater. 2010, 22, 5762.
56) Rosen, M. J. In Surfactants and Interfacial Phenomena; John
(
(
(
(
Wiley & Sons: New York, 2004, p i.
(57) Wu, S. J. Polym. Sci., Part C: Polym. Symp. 1971, 34, 19.
(58) Ma, W.; Gopinathan, A.; Heeger, A. J. Adv. Mater. 2007, 19,
3656.
(59) Sun, Z.; Xiao, K.; Keum, J. K.; Yu, X.; Hong, K.; Browning, J.;
Ivanov, I. N.; Chen, J.; Alonzo, J.; Li, D.; Sumpter, B. G.; Payzant, E.
A.; Rouleau, C. M.; Geohegan, D. B. Adv. Mater. 2011, 23, 5529.
(
(
́
(
(60) Maturova,
Kemerink, M. Adv. Funct. Mater. 2011, 21, 261.
61) Wu, J.-L.; Chen, F.-C.; Hsiao, Y.-S.; Chien, F.-C.; Chen, P.; Kuo,
C.-H.; Huang, M. H.; Hsu, C.-S. ACS Nano 2011, 5, 959.
́
K.; van Bavel, S. S.; Wienk, M. M.; Janssen, R. A. J.;
(
(
(
(62) Alam, M. M.; Jenekhe, S. A. Chem. Mater. 2004, 16, 4647.
(63) Dittmer, J. J.; Marseglia, E. A.; Friend, R. H. Adv. Mater. 2000,
(
2
(
1
(
2, 1270.
64) Kordatos, K.; Bosi, S.; Da Ros, T.; Zambon, A.; Lucchini, V.;
Prato, M. J. Org. Chem. 2001, 66, 2802.
Vanhoyland, G.; Aresu, S.; D’Olieslaeger, M.; Manca, J. Adv. Funct.
Mater. 2006, 16, 760.
(
32) Yang, X.; van Duren, J. K. J.; Rispens, M. T.; Hummelen, J. C.;
Janssen, R. A. J.; Michels, M. A. J.; Loos, J. Adv. Mater. 2004, 16, 802.
33) Yang, X.; van Duren, J. K. J.; Janssen, R. A. J.; Michels, M. A. J.;
Loos, J. Macromolecules 2004, 37, 2151.
(
I
dx.doi.org/10.1021/cm3010369 | Chem. Mater. XXXX, XXX, XXX−XXX