Morphological stabilization by in situ polymerization of fullerene derivatives leading to efficient, thermally stable organic photovoltaics

Article abstract of DOI:10.1002/adfm.201002502

The successful design and synthesis of two styryl-functionalized fullerene derivatives, [6,6]-phenyl-C₆₁ -butyric acid styryl dendron ester (PCBSD) and [6,6]-phenyl-C₆₁ -butyric acid styryl ester (PCBS) is presented. The polymerizabl

Full text of DOI:10.1002/adfm.201002502

www.afm-journal.de
www.MaterialsViews.com
Morphological Stabilization by In Situ Polymerization
of Fullerene Derivatives Leading to Efficient, Thermally
Stable Organic Photovoltaics
Yen-Ju Cheng,* Chao-Hsiang Hsieh, Pei-Jung Li, and Chain-Shu Hsu*
and quantum efficiency. PSCs based on
the concept of bulk heterojunction (BHJ)
strategy have provided the most straight-
forward solutions to maximize internal
D−A interfacial area for efficient charge
separation thus far, but tailoring the mor-
phology of the blend in a BHJ device
toward optimized performance is of crit-
ical importance and remains challenging.
Until now, the combination of poly(3-hexy-
lthiophene) (P3HT) as electron donor and
[6,6]-phenyl-C61-butyric acid methyl ester
(PCBM) as electron acceptor in the active
layer represents one of the most efficient
The successful design and synthesis of two styryl-functionalized fullerene
derivatives, [6,6]-phenyl-C61-butyric acid styryl dendron ester (PCBSD) and
[6,6]-phenyl-C61-butyric acid styryl ester (PCBS) is presented. The polymeriz-
able PCBS or PCBSD materials are incorporated into a poly(3-hexylthiophene)
(P3HT):[6,6]-phenyl-C61-butyric acid methyl ester (PCBM) blend to form
an active layer of ternary blend. The blending systems are first thermally
annealed at 110 C for 10 min to induce optimal morphology, followed by
heating at 150 C for 10 min to trigger the in situ polymerization of styrene
groups. Through chemical crosslinking of PCBSD, the initial morphology
of the blend (P3HT:PCBM:PCBSD = 6:5:1 in weight) can be effectively fixed
and stably preserved. The device based on this blend shows extremely stable
device characteristics, delivering an average power conversion efficiency
(PCE) of 3.7% during long-term thermal treatment. By molecular engineering
to reduce the insulating portion, PCBS with higher C60 content (71 wt%) pos-
sesses better electron-transport properties than PCBSD (58 wt%). Encourag-
ingly, at a low doping concentration of PCBS in the blend (P3HT:PCBM:PCBS
= 6:5:1 in weight), linear-polymerized PCBS can stabilize the morphology
against thermal heating. This device exhibits more balanced charge mobility
to achieve an average PCE of 3.8% over 25 h heating at 150 °C.
BHJ solar cells, with power-conversion
efficiencies approaching 5%.[2] The success
of the P3HT:PCBM system mostly relies
on carefully controlling the phase separa-
tion between two components in the bulk
to reach an ideal morphology with D−A
domain size around 10 nm, which is com-
parable to the diffusion length of excitons.
The morphological development of the
P3HT:PCBM blend is governed not only
by intrinsic molecular properties but also
by the extrinsic processing conditions of
devices. For example, the regioregularity,[3] molecular weight,[4]
and polydispersity[5] of P3HT are important structural factors
that greatly influence its crystallinity; thus, variation of these
factors can result in different morphologies of polymer blends.
More importantly, the kinetics of crystallization and segregation
of the P3HT:PCBM components can be manipulated by many
parameters, including casting solvent, blending ratio,[6] external
treatment of solvent annealing, and thermal annealing.[7]
Thermal annealing of P3HT:PCBM composite provides external
energy, which drives the P3HT to reorganize and self assemble
to become crystalline P3HT. Through this morphological evolu-
tion, the featureless and homogeneous blend gradually reaches
a bicontinuous interpenetrating D−A network with optimal
D−A domain size.[2a] As a result, maximum interfacial area for
efficient charge generation and a higher degree of nanoscale
P3HT crystallinity can be achieved for better charge transport.[8]
Unfortunately, this optimal morphology is a kinetically trapped
intermediate that readily moves toward a more thermodynami-
cally stable state if thermal annealing at elevated temperature is
applied constantly.[9] The P3HT domain continues to undergo
further crystallization, while the spherical PCBM, with its high
1. Introduction
Polymer solar cells (PSCs) are a promising alternative for clean
and renewable energy due to their potential to be fabricated
onto large area, light-weight flexible substrates by solution
processing at lower cost. The general working principle in such
solar cells first involves photoexcitation of the donor material by
absorption of light energy to generate excitons. This Coulomb-
correlated electron−hole pair diffuses to the donor−acceptor
(D−A) interface, where exciton dissociation occurs. Because their
limited lifetimes only allow excitons to diffuse a short distance
(5−14 nm),[1] donor excitons created far away from the hetero-
junction interface decay to the ground state before they reach
the D−A interfaces, which leads to the loss of absorbed photons
Prof. Y.-J. Cheng, C.-H. Hsieh, P.-J. Li, Prof. C.-S. Hsu
Department of Applied Chemistry
National Chiao Tung University
1001 Ta Hsueh Road Hsin-Chu, 30010, Taiwan
E-mail: yjcheng@mail.nctu.edu.tw; cshsu@mail.nctu.edu.tw
DOI: 10.1002/adfm.201002502
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2011, 21, 1723–1732
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1723

www.afm-journal.de
www.MaterialsViews.com
Figure 1. a) Morphological evolution of the P3HT:PCBM blend under thermal treatment. b) Two-stage thermal-annealing strategy of the P3HT:PCBM:P-
PCBM blend to chemically lock the morphology and preserve the morphological stability (T1 < T2).
molecular mobility, tends to diffuse out of the polymer matrix
and aggregate into larger clusters or single crystals.[9d,9e,10] Such
a progressive phase segregation between P3HT and PCBM
eventually leads to micronsized D−A domains with concomitant
reduction of the D−A interfacial area, as shown in Figure 1a. Con-
sidering that a photovoltaic device must be exposed to long-term
sunlight irradiation, the accumulated heat may raise the opera-
charge-transport properties.[15] Morphological studies have
shown that PCBM molecules are capable of diffusing rapidly
into crystalline P3HT upon thermal treatment; this ability is
even more pronounced in an amorphous polymer matrix such
as MDMO–PPV.[9c,d] We envisaged that direct crosslinking
between highly mobile PCBM molecules to prevent them from
severe immigration could be a more effective way to maintain
optimal morphology. Moreover, compared to the numerous
p-type conjugated polymers with excellent properties rapidly
being developed, n-type materials are still exclusively dominated
by PCBM-based derivatives. It would be more economical and
practical, from the standpoint of synthetic feasibility, to develop a
single polymerizable PCBM-based material that is widely appli-
cable to all kinds of p-type polymers, regardless of whether they
are amorphous or crystalline, to avoid large-scale phase separa-
tion. A PCBM derivative functionalized with an epoxy group as a
polymerization group has been demonstrated.[14d] However, this
system requires the addition of a catalytic amount of ionic pho-
toinitiator, which turns out to be detrimental to the device due to
the electron-trapping effect.
Styrene is known to be a superb thermally curable group
because it undergoes efficient and rapid polymerization in the
solid state to form polystyrene without the use of any initiators.[16]
Herein, we have designed and synthesized two styryl-function-
alized fullerene derivatives, [6,6]-phenyl-C61-butyric acid styryl
dendron ester (PCBSD)[17] and [6,6]-phenyl-C61-butyric acid styryl
ester (PCBS), to fine-tune the crosslinking and electron-transport
properties (Scheme 1). The C60 weight contents in PCBSD and
PCBS are 58% and 71%, respectively. PCBSD, containing a small
dendron to attach two styrene groups, is more able than PCBS to
efficiently lock morphology when blending with a p-type material,
because the two styrene groups in PCBSD allow a 3D crosslinking
reaction, whereas one styrene group in PCBS only allows linear
polymerization. However, the high content of the insulating den-
dron in PCBSD may inevitably deteriorate its electron-transport
properties and thereby decrease device performance. There-
fore, PCBS should have better electron-transport properties than
tional temperature above the glass transition temperature (T ) of
g
polymers, thus destroying the optimal morphology, and deterio-
rating the device performance.[11] Consequently, morphological
instability of PSCs is considered to be one of the major obstacles
that hinder its path toward commercialization. The development
of new strategies to preserve optimal morphology is urgently
needed but rather challenging. Fréchet and co-workers found
that a slight decrease of the regioregularity of P3HT to weaken
the crystallization-driven phase separation not only retains the
device efficiency but also enhances the thermal stability of mor-
phology in the solar cells.[12] Utilization of secondary interactions
between donors and acceptors to fix morphology has also been
explored.[13] Fréchet et al. used a diblock copolymer containing
fullerene and oligothiophene pendant groups to serve as a com-
patibilizer to improve adhesion between the P3HT and PCBM
boundaries.[13a] A more straightforward method is to lock mor-
phology by covalent bonding.[14] In situ chemical crosslinking in
BHJ films right after the formation of the optimal morphology
is an ideal strategy to develop the desired morphology while
improving morphological stability.[14] For this purpose, func-
tionalities for crosslinking reactions should be incorporated into
either p-type polymers[14a] or n-type fullerene derivatives.[14d,e]
In the former case, photocrosslinkable P3HT copolymers used
to blend with PCBM were developed to stabilize the BHJ film
morphology.[14a] Because the polymeric 3D networks limit the
conformational dynamics of the polymers and indirectly obstruct
the diffusion pathway of PCBM molecules, efficient and stable
photovoltaics could be realized. One potential drawback for this
design is that the attachment of crosslinkers on the side chain
of P3HT may disturb the π−π stacking of P3HT and alter the
©
1724 wileyonlinelibrary.com
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2011, 21, 1723–1732

www.afm-journal.de
www.MaterialsViews.com
Scheme 1. The synthetic route for PCBS, the structure of PCBSD, and their thermal polymerization.
PCBSD. The core structure of PCBSD and PCBS is based on the
most widely used n-type material, PCBM, so they are expected
to inherit all the excellent electrical properties of PCBM. In addi-
tion to tailoring the materials’ properties at molecular level, we
can further adjust the macroscopic properties by varying the
composition in the active layer. The polymerizable PCBM-based
materials (P-PCBM), PCBS, or PCBSD, can be doped into the
P3HT:PCBM system to form a ternary blend system P3HT/
PCBM/P-PCBM. The relative content of P-PCBM to PCBM
can be fine tuned for the purpose of morphological fixation without
affecting the electron-transport properties in the active layer.
The temperature window required to drive the styrene
groups to polymerize in these materials is just above the
annealing temperature used to optimize the morphology of
the P3HT:PCBM system, which allows us to accomplish the
morphological fixation through a distinct two-stage process,
as depicted in Figure 1b. Without the interference of chemical
reactions, the optimized morphology under thermal annealing
at T1 °C is developed at the first stage, followed by sequen-
tially triggering the chemical crosslinking at T2 °C to preserve
the morphology at the second stage, thereby leading to high-
performance solar cells with excellent morphological stability.
The thermal-transition properties of PCBS and PCBSD
monomers were investigated by the first scan of differential
scanning calorimetry (DSC). It showed a broad exothermic peak
with a Tmax of around 150 °C, which clearly indicates thermal
polymerization of styrene groups. In addition, the vibrational
stretching of the vinyl groups at 1622 cm−1 in PCBS and
PCBSD completely disappears in the IR spectrum after thermal
treatment, which provides further evidence for polymeriza-
tion. After the initial heating cycle, a high Tg of ca. 142 °C was
observed during the second DSC scan, which indicates that the
polymerized materials are highly amorphous. Thus, the polym-
erized materials can reduce the tendency toward crystallization
and maintain long-term morphological stability.
2.2. Performances and Thermal Stability of BHJ Solar Cells
In this study, we use the conventional device configuration ITO/
PEDOT:PSS/active layer/Ca/Al, and implement two-stage thermal
annealing as the standard procedure for all of the blending sys-
tems. The spin-coated active-layer blend is first thermally annealed
at 110 °C for 10 min to develop suitable morphology, followed by
thermal annealing at 150 °C for another 10 min to trigger in situ
polymerization. For comparison, we first fabricated a P3HT:PCBM-
based (1:1 in wt%) reference device which exhibited a PCE value
of 4.08% under AM 1.5G illumination (Figure 2a, and Table S1 in
the Supporting Information). We then fabricated a device using
P3HT:PCBSD as the active layer where the PCBM is completely
replaced by the crosslinkable PCBSD. In this case, the weight
ratio of the P3HT:PCBM:PCBSD blend is 1:0:1. For simplicity, we
denote this blend as PCBSD101. Under AM 1.5G illumination
at 100 mW/cm2, device PCBSD101 showed very poor device
performance with a power conversion efficiency (PCE) of 0.04%
(Figure 2a). An obvious kink effect in the J–V curve was observed,
which implies that a high degree of crosslinking and a high
2. Results and Discussion
2.1. Synthesis and Thermal Properties of the
Styryl-Functionalized PCBM Derivatives
The synthesis of PCBSD has been previously reported.[17] PCBS
was prepared by the esterification of PCBA with 4-vinylbenzyl
alcohol in the presence of 1-ethyl-3-(3-dimethylaminopropyl)
carbodiimide (EDC) and 4-dimethylaminopyridine (DMAP) in
69% yield.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2011, 21, 1723–1732
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1725

www.afm-journal.de
www.MaterialsViews.com
a)
b)
2
2
0
P3HT:PCBM
PCBSD651
PCBSD211
PCBSD101
PCBS651
PCBS211
PCBS101
0
-2
-2
-4
-4
-6
-6
-8
-8
-10
-12
-10
-12
-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Voltage (V)
Voltage (V)
Figure 2. J–V characteristics of PSCs based on a) the P3HT:PCBM, PCBSD651, PCBSD211, and PCBSD101 blends, and b) PCBS651, PCBS211, and
PCBS101 blends under AM 1.5 irradiation at 100 mW cm−2
.
content of insulating polystyrene in the n-type material domain
exert a detrimental effect on the electron transport.
To systematically evaluate the ability of PCBSD to stabilize mor-
phology and preserve device characteristics under the influence
of thermal treatment, we fabricated a series of devices identical
in all aspects except that the P3HT:PCBM:PCBSD (6:5:1 in wt%)
thin films, prepared by the standard procedure, were further iso-
thermally heated to 150 °C for various times prior to deposition
of the top electrode. The corresponding reference devices based
on the P3HT:PCBM system were also fabricated and character-
ized to test thermal stability. The J–V curves of the P3HT:PCBM
reference and PCBSD651 devices are shown in Figure 3a and
b, respectively, and their corresponding photovoltaic parameters
(PCE, Voc, Jsc and fill factor (FF)) as functions of heating time at
150 °C were plotted in Figure 4. It can be seen that the perform-
ance of the P3HT:PCBM reference device decreases as the heating
To improve the electron-transporter content in the com-
posite, we incorporated PCBM into the P3HT:PCBSD blend to
dilute the PCBSD content and form a ternary system. We fixed
the blending weight ratio of P3HT to the total n-type materials
(i.e., PCBM plus PCBSD) at 1:1 and adjusted the relative con-
tent between the PCBM and PCBSD. Devices PCBSD211 and
PCBSD651 were fabricated using the P3HT:PCBM:PCBSD (2:1:1
in wt%) blend and the P3HT:PCBM:PCBSD (6:5:1 in wt%) blend,
respectively. The PCE of device PCBSD211 was 0.23% better than
that of device PCBSD101 (Figure 2a). By further decreasing the
PCBSD content in the blend, the PCE of device PCBSD651 was
dramatically improved to a reasonable value of 3.32% (Figure 2a).
a)
P3HT:PCBM
PCBS211
c)
2
0
2
0
Pristine
2 hr
Pristine
2 hr
5 hr
5 hr
-2
-2
10 hr
15 hr
20 hr
25 hr
10 hr
15 hr
20 hr
25 hr
-4
-4
-6
-6
-8
-8
-10
-12
-10
-12
-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Voltage (V)
Voltage (V)
PCBSD651
PCBS651
b
)
d)
2
2
Pristine
2 hr
5 hr
10 hr
15 hr
20 hr
25 hr
Pristine
2 hr
5 hr
10 hr
15 hr
20 hr
25 hr
0
0
-2
-4
-6
-8
-2
-4
-6
-8
-10
-10
-12
-12
-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Voltage (V)
Voltage (V)
Figure 3. J–V characteristics of PSCs based on a) P3HT:PCBM, b) PCBSD561, c) PCBSD211, and d) PCBS651 blends before and after isothermal
heating at 150 °C for various times.
©
1726 wileyonlinelibrary.com
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2011, 21, 1723–1732

www.afm-journal.de
www.MaterialsViews.com
a)
c)
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
10
8
6
P3HT:PCBM
PCBS651
PCBS211
P3HT:PCBM
PCBS651
PCBS211
4
PCBSD651
PCBSD651
2
0
5
10
15
20
25
0
5
10
15
20
25
Time (h)
Time (h)
b)
d)
0.65
80
70
60
50
40
30
0.60
0.55
0.50
0.45
0.40
P3HT:PCBM
PCBS651
PCBS211
P3HT:PCBM
PCBS651
PCBS211
PCBSD651
PCBSD651
0
5
10
15
20
25
0
5
10
15
20
25
Time (h)
Time (h)
Figure 4. Photovoltaic parameters of PSCs based on the P3HT:PCBM, PCBSD651, PCBS211, and PCBS651 blends as functions of heating time
at 150 °C. a) PCE, b) open-circuit voltage (Voc), c) short-circuit current (Jsc), and d) fill factor (FF).
time increases. The PCE of the device dropped dramatically from
4.08% to 0.69% after 25 h isothermal heating (Figure 4a). As
shown in Figure 4c, the decrease in efficiency is mainly a result
of the decrease in Jsc values. In sharp contrast, the PCBSD651
device exhibited very stable device characteristics and did not
show any degradation after 25 h heating (Figure 3b). We asso-
ciate the improved stability with several factors. The crosslinked
PCBSD, with its highly amorphous nature, will reduce the crys-
tallinity of the fullerene materials.[18] Furthermore, the intercon-
nected PCBSD network in the blend will sterically restrict the
mobile PCBM from fast immigration upon thermal heating.
Therefore, the large-scale phase separation in the active layer due
to PCBM aggregation can be greatly suppressed. Surprisingly, the
PCE of PCBSD651 was even increased upon thermal heating,
and reached its highest value—4.01%—after 10 h heating. This
value is already comparable to the initial performance of the
P3HT:PCBM reference device. The thermal-annealing condi-
tions to obtain optimal morphology for the binary P3HT:PCBM
and the ternary P3HT:PCBM:PCBSD blends should be very dif-
ferent. Incorporation of the crosslinked PCBSD network into the
P3HT:PCBM blend will lead to an altered morphology that is not
optimized at the beginning (110 °C for 10 min and 150 °C for
10 min). The continuous thermal annealing of this ternary blend
will gradually induce the optimized morphology that results in the
improved performance. In view of the fact that only 8% weight
content of the PCBSD material in the ternary blend already
sufficiently preserves the morphology, the electron-transporter
content may be further increased by removing the dendritic
moiety in the PCBSD. PCBS, which consists of a simple 4-vinyl-
benzyl group, was designed to improve the active-C60 electron-
transporter content relative to PCBSD (71 wt% versus 58 wt%).
Again, under the same processing conditions, we fabricated three
devices—PCBS101, PCBS211, and PCBS651—based on the three
corresponding blend systems—P3HT:PCBM:PCBS (1:0:1 in
wt%), P3HT:PCBM:PCBS (2:1:1 in wt%), and P3HT:PCBM:PCBS
(6:5:1 in wt%). The J–V characteristics of these devices are
shown in Figure 2b. Device PCBS101 showed an initial PCE of
ca. 2%, which already outperforms the dendron-containing ana-
logue PCBSD101, and even device PCBSD211. Notably, a com-
plete reduction of the kink effects in the J–V curves of devices
PCBS101 and PCBS211 suggests that increasing the active-C60
content indeed greatly improves electron-transport properties.
By further decreasing the PCBS content of the blends, devices
PCBS211 and PCBS651 achieved the enhanced PCEs of 3.02%
and 3.84%, respectively. In a similar manner, the thermal stability
of the PCBS211 and PCBS651 devices upon isothermal heating
at 150 °C were also evaluated. Encouragingly, both PCBS211
and PCBS651 devices also exhibited very stable device perform-
ances against thermal treatment (Figure 3c–d and Figure 4). Con-
sidering that the content of active C60 in PCBS (71 wt%) is very
close to that of PCBM (79 wt%), and the structure of PCBS is
very similar to that of PCBM, the active layers containing PCBS
as dopant can easily achieve optimal morphology with better
electron-transport properties, which leads to improved device
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2011, 21, 1723–1732
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1727

www.afm-journal.de
www.MaterialsViews.com
performance. Most significantly, the low concentration of the
polymerized PCBS in the PCBS651 blend (8 wt%) can already fix
the optimal morphology to preserve device stability against long-
term heating.
2.3. Optical Microscopy (OM) and Scanning Electron
Microscopy (SEM)
OM was used to investigate morphological alteration of the
blends under the influence of heating (Figure 5). Thermal
annealing of the P3HT:PCBM blend at 150 °C for 25 h
induced a severe macromorphological alteration, giving rise
to needle-shaped PCBM crystals of hundreds micrometers in
length.[9a,9b,14a] In contrast, the morphology of the PCBSD651
blend remains almost unchanged before and after thermal
annealing, which demonstrates the excellent ability of PCBSD
to chemically fix the morphology by 3D crosslinking. The
polymerizable PCBS, however, shows less ability to effectively
lock the morphology. It seems reasonable that the OM images
of PCBS651 show some slight local macrographic alteration.
In the PCBS211 blend, on the other hand, alteration was com-
pletely suppressed due to the higher content of PCBS.
To gain more insight into the microstructure of the active
layers on the nanoscale, we also measured the SEM cross-
sectional images of the three blends before and after thermal
heating (Figure 6). The SEM image of the P3HT:PCBM blend
showed severe phase segregation after isothermal heating at
150 °C for 25 h. In contrast, the PCBSD651 and PCBS651
blends maintained well-defined nanomorphologies, which
again demonstrates that polymerization in
Figure 5. OM images of the blends before and after isothermal heating at
150 °C for 25 h. a) P3HT:PCBM blend, b) PCBSD651 blend, c) PCBS651
blend, and d) PCBS211 blend.
the blends is capable of freezing the initially
formed morphologies.
2.4. Crystallinity of P3HT in the Blends
The crystallinity of P3HT in the active layer
plays an important role in the hole-transport
properties. X-ray diffraction (XRD) was
employed to detect P3HT crystallinity vari-
ations in the blends containing different
amounts of PCBSD and PCBS (Figure 7).
After thermal annealing at 110 °C for
10 min and 150 °C for 10 min, all of the blends
exhibited a sharp 2θ peak centered at ca 5.2o
which is the typical signature of well-organ-
ized planar P3HT stacks (100) oriented along
the axis perpendicular to the substrate.[2a]
This means that incorporation of poly-
merizable n-type materials does not appreci-
ably affect the P3HT crystallinity in the blend.
In addition, the first stage of thermal treat-
ment at 110 °C is sufficient for the formation
of P3HT crystallites prior to triggering the
polymerization at an elevated temperature
of 150 °C. Furthermore, the length scales of
Figure 6. SEM cross-sectional images of the glass/ITO/PEDOT:PSS blends before and after iso-
thermal heating at 150 °C for 25 h. a) P3HT:PCBM blend, b) PCBSD651 blend, and c) PCBS651
blend. The thickness of the photoactive layers is approximately 250 nm.
the P3HT crystallites at the (100) reflection in
these blends can be quantified by estimating
©
1728 wileyonlinelibrary.com
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2011, 21, 1723–1732

www.afm-journal.de
www.MaterialsViews.com
P3HT:PCBM
P3HT:PCBM (25 h)
PCBS651
PCBS651 (25 h)
PCBS211
PCBS211 (25 h)
PCBSD651
2.4
2.0
1.6
1.2
0.8
0.4
0.0
PCBSD101
PCBSD211
PCBSD651
PCBSD651 (25 h)
PCBS101
PCBS211
PCBS651
P3HT:PCBM
4
6
8
10
12
14
16
18
20
300
400
500
600
700
2θ / degree
Wavelength (nm)
Figure 7. XRD patterns of the P3HT:PCBM, PCBS651, PCBS211, PCBS101,
PCBSD561, PCBSD211, and PCBSD101 blends.
Figure 8. Absorption spectra of the films of the blends before and after
25 h isothermal heating at 150 °C. (a.u. = arbitrary units)
the mean size of the P3HT crystalline domain (τ100) using
P3HT, thereby reducing the crystallization of P3HT to some
extent. The same trend can be also observed in the case of the
PCBS651 (18.9 nm), PCBS211 (17.7 nm), and PCBS101 (17.3 nm)
blends. However, it can be seen that the structural impact of
PCBS on the crystallinity of P3HT is less than that of PCBSD.
The optical and photoluminescent properties of the
PCBM:P3HT, PCBSD651, PCBS651, and PCBS211 blends
were studied before and after thermal heating at 150 °C for
25 h. The absorption and emission spectra of the PCBSD651,
PCBS651, and PCBS211 blends are essentially unchanged after
thermal heating (Figure 8 and Figure 9); they exhibit absorp-
tion peaks at 515, 550, and 610 nm due to the extensive π−π
stacking along the P3HT backbone (Figure 8).[20] These results
suggest that the polymerization between the PCBSD or PCBS
molecules in the blends effectively preserves the morphology of
the stacking of P3HT and thus maintains the P3HT absorption
and emission properties. However, after 25 h heating, the inten-
sity of the P3HT:PCBM blend in the P3HT absorption region is
enhanced, which is again indicative of a higher degree of P3HT
crystallization.[9e] Moreover, the P3HT emission intensity in
Scherrer’s equation,[19]
K8
100 cos2
J100
≈
$
(1)
(2)
$
= Δ22 B 180°
/
×
100
where K is the shape factor (typically 0.9 for the dimensionless
shape factor), and β100 is the FWHM intensity of the peak. The
mean size of the P3HT crystallite in the P3HT:PCBM blend
was determined to be 20.4 nm (see Table 1). With increasing
PCBSD content in the blend, in conjunction with decreasing
PCBM content, the P3HT crystallite size decreases gradually to
18.5 nm for PCBSD651, 17.7 nm for PCBSD211, and 16.2 nm
for PCBSD101. In comparison with PCBM, the extra dendritic
moiety used for crosslinking in PCBSD will sterically hinder the
crystalline growth of P3HT by interfering with the packing order
of P3HT. In addition, the amorphous nature of PCBSD before
and after polymerization will also enhance its miscibility with
25
P3HT:PCBM
P3HT:PCBM (25 h)
PCBS651
PCBS651 (25 h)
PCBS211
PCBS211 (25 h)
PCBSD651
PCBSD651 (25 h)
T a b le 1 . 2 θ angle, the full width at half maximum (FWHM) of the
P3HT (100) peaks, and the mean size of P3HT crystallites in the
blend films.
20
15
10
5
2θ [°]a)
5.22
5.22
5.22
5.23
5.23
5.22
5.22
Δ2θ [°]b)
0.39
τ [nm]c)
20.4
18.5
17.7
16.2
18.9
17.7
17.3
Blend
P3HT:PCBM
PCBSD651
PCBSD211
PCBSD101
PCBS651
0.43
0.45
0.49
0.42
PCBS211
0.45
0
600
650
700
750
800
PCBS101
0.46
Wavelength (nm)
a)2θ is the angle between the incident and scattered X-ray wavevectors; b)Δ2θ is the
FWHM of the (100) peak; c)τ is the mean size of the P3HT crystallites determined
by Sherrer’s equation.
Figure 9. Photoluminescence spectra of the blends before and after iso-
thermal heating at 150 °C for 25 h.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2011, 21, 1723–1732
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1729

www.afm-journal.de
www.MaterialsViews.com
b)
a)
5
0
5
0
PCBS651
PCBS211
PCBS101
PCBSD651
PCBSD211
PCBSD101
P3HT:PCBM
PCBS651
PCBS211
PCBS101
PCBSD651
PCBSD211
PCBSD101
P3HT:PCBM
-5
-5
0.6
0.8
1.0
1.2
1.4
0.4
0.6
0.8
1.0
1.2
lnV (V)
lnV (V)
Figure 10. Natural logarithm of J–V characteristics of a) the hole-only devices and b) the electron-only devices of PCBS651, PCBS211, PCBS101,
PCBSD651, PCBSD211, PCBSD101, and P3HT:PCBM.
the P3HT:PCBM blend is also enhanced 1.6 fold after heating
(Figure 9). This implies that the efficiency of exciton dissocia-
tion from the P3HT to the PCBM in the blend is reduced.[20b]
The prolonged heating destroys the optimal morphology as a
result of continuing phase segregation between P3HT and
PCBM to a macroscale domain larger than the exciton-diffusion
length. The excess P3HT excitons, which cannot reach the D−A
interface, will decay through radiative relaxation. The dramatic
decrease in the short-circuit current (Jsc) (i.e., from 10.26 to
2.27 mA cm−2 ) of the P3HT:PCBM-based device is therefore
strongly associated with the diminution of the interfacial area
between the donor and acceptor.
In comparison with the P3HT:PCBM blend, the hole mobility
decreases as the content of PCBS or PCBSD increases in the
blends. Because the P3HT content is constantly kept at 50 wt%
in each blend, variations in the n-type materials do not sig-
nificantly affect the hole mobility. The slight reduction in hole
mobility is also in good agreement with the slightly reduced
mean size of the P3HT crystallites in the presence of PCBSD
or PCBS. In contrast, the electron mobility decreases dramati-
cally with increasing PCBSD content. The 3D and insulating
crosslinking network of PCBSD is an obstacle to the electron
transport. On the other hand, PCBS, which contains fewer
insulating moieties, exerts much less pronounced effects on
the electron transport. The carrier mobilities of the blends are
summarized in Table 2. It should be noted that the ratio of hole
to electron mobility (μh/μe) increases with PCBSD content. A
change of μh/μe values of more than three orders of magnitude
is observed for the PCBSD101 (2525) and PCBSD211 (1107)
blends. The less-mobile electrons left behind in the bulk will
form a space–charge region thus affecting the J–V characteris-
tics of the blend. The low FF values for the PCBSD211 (13%)
and PCBSD101 (12%) devices could be attributed to the space–
charge buildup caused by the extreme imbalance in carrier
mobility. In contrast, the μh/μe values are all smaller than ten for
the other blends; i.e., 1.84 for PCBS651, 4.25 for PCBS211, and
3.43 for PCBSD651. The much more balanced carrier mobility
is responsible for the reduction of kink effects in the J–V charac-
teristics under AM 1.5.
2.5. Carrier Mobilities in the Blends
To quantify the influence of the polymerizable PCBSD and
PCBS on the charge-transport characteristics in the active
layers, we fabricated a series of unipolar devices to evaluate
the hole (Figure 10a) and electron (Figure 10b) mobilites of the
P3HT:PCBM, PCBSD651, PCBSD211, PCBSD101, PCBS651,
PCBS211, and PCBS101 blends by the space–charge lim-
ited current (SCLC) technique according to the Mott–Gurney
equation:[20a,21]
ꢀ
ꢁ
2
V
L
J = 9gg0 μ0
(3)
3
8
T a b le 2 . Carrier mobilities of the blends extracted by using the space–
charge limited current method.
3. Conclusions
Morphological instability is the primary concern of PSCs. Con-
tinuous phase segregation upon heating eventually leads to
micron-sized donor–acceptor domains with concomitant reduc-
tion of interfacial area. Spherical PCBM with high molecular
mobility tends to diffuse out of the polymer matrix and aggre-
gate into larger clusters or single crystals. We have developed a
straightforward method to effectively maintain the morphology
by in situ polymerization between highly mobile fullerene
derivatives to prevent them from severe immigration. Styryl-
functionalized polymerizable fullerene derivatives, PCBSD and
PCBS, are designed and synthesized for this purpose. Various
amounts of PCBSD or PCBS are doped into the P3HT:PCBM
Blend
Hole mobility, μh
[cm2s−1 V−1
Electron mobility, μe
[cm2s−1 V−1
Ratio μh/ μe
]
]
3.10 × 10−4
2.45 × 10−4
2.54 × 10−4
1.76 × 10−4
2.98 × 10−4
9.54 × 10−5
7.35 × 10−5
1.25 × 10−4
1.33 × 10−4
5.96 × 10−5
4.82 × 10−5
8.70 × 10−5
8.62 × 10−8
2.91 × 10−8
P3HT:PCBM
PCBS651
2.48
1.84
4.25
3.65
3.43
1107
2525
PCBS211
PCBS101
PCBSD651
PCBSD211
PCBSD101
©
1730 wileyonlinelibrary.com
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2011, 21, 1723–1732

www.afm-journal.de
www.MaterialsViews.com
AI-4083) was spuncast at 2000 rpm for 30 s to produce a 30 nm
thick layer, followed by baking at 150 °C for 30 min. Seven different
solutions of P3HT (20 mg) and fullerene derivatives (totally 20 mg)
in 1 g of o-DCB were prepared separately with fixed blend-weight
ratio of P3HT to the total n-type materials (i.e., PCBM, PCBSD,
PCBS) equal to 1:1, and the relative content was adjusted between
either PCBM and PCBSD or PCBM and PCBS (as shown in Table 1
in the Supporting Information). The solutions were stirred at 70 °C
overnight under a nitrogen atmosphere. Prior to use, the solutions
were filtered by passing through a 0.45 μm Teflon syringe filter. For
BHJ solar-cell devices and unipolar devices for SCLC measurement,
the blend solutions containing polymer/fullerene (1:1, w/w) were spun
at 700 rpm for 30 s to form a 250 nm film on top of the PEDOT:PSS
layer. Films were dried in covered Petri dishes for 20 min to perform
solvent-assisted annealing. The films were then subjected to two-
stage thermal annealing, firstly at 110 °C for 10 min and subsequently
at 150 °C. Finally, the top electrode, made of Ca (10 nm)/Al (100 nm),
was thermally evaporated at a pressure below 10−6 torr to complete
the BHJ solar-cell devices. To perform the accelerated performance
test, BHJ solar-cell devices were subjected to sustained heating at
150 °C for various times prior to the cathode electrode deposition.
All the devices were measured at room temperature under a nitrogen
atmosphere with a Xenon lamp coupled to an AM 1.5G solar filter
(SAN-EI XES-301S solar simulator). J–V characteristics were recorded
with a Keitheley 2400 Source Measurement Unit. The active area was
0.04 cm2.
blend to systematically investigate the morphological stability
and electron-transport properties. The blending systems were
first thermally annealed at 110 °C for 10 min to induce optimal
morphology, followed by heating at 150 °C for 10 min to trigger
the in situ polymerization of styrene groups. It is found that by
doping small amounts of PCBSD into the P3HT:PCBM blend
(P3HT:PCBM:PCBSD = 6:5:1 in weight), the initial morphology
of the blend can be fixed and preserved effectively after polym-
erization. The device based on this blend showed highly stable
device characteristics, delivering an average PCE of 3.7% during
25 h isothermal heating at 150 °C. In sharp contrast, under the
same conditions, the P3HT:PCBM blend undergoes severe
phase separation due to the crystallization of P3HT and the
aggregation of PCBM. Therefore, the PCE of the P3HT:PCBM-
based device dropped dramatically from 4.08% to 0.69%. By
removing one styrene group, PCBS, with its higher C60 content
(71 wt%), possesses better electron-transport properties than
PCBSD (58 wt%). Importantly, at a low doping concentration of
PCBS in the blend (P3HT:PCBM:PCBS = 6:5:1 in weight), lin-
early polymerized PCBS can already stabilize the morphology
against the heating. This device exhibited more balanced charge
mobility to achieve an average PCE of 3.8% during 25 h heating
at 150 °C. We envisage that this novel and simple strategy can
be widely applied to any other BHJ systems that contain various
low-bandgap polymers to achieve highly efficient and morpho-
logically stable polymer solar cells.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
4. Experimental Section
Synthesis of PCBS: A 100 ml, two-necked round-bottomed flask was
charged with PCBA (26 mg, 0.03 mmol), anhydrous 1,2-dichlorobenzene
(20 ml), 4-vinylbenzyl alcohol (3.9 mg, 0.03 mmol), and DMAP (4.25 mg,
0.035 mmol). The solution was cooled to 0 °C with an ice bath, and EDC
(5.4 mg, 0.035 mmol) was added in one portion. After stirring at 0 °C
for 3 h, the ice bath was removed, and the dark brown reaction mixture
was stirred at room temperature for 12 h. After removal of the solvent
under reduced pressure, the residue was purified by silica-gel column
chromatography with toluene/hexane (1:1 v/v, then pure toluene) as
the eluent. The product was redissolved in toluene, and precipitated by
adding the solution to methanol. The solid was filtered off and washed
with hexane, and then dried under vacuum to yield PCBS as a brown
solid (34.6 mg, 69%). 1H NMR (300 MHz, CDCl3, δ): 2.17–2.22 (m,
2 H), 2.5–2.59 (m, 2 H), 2.88–2.93 (m, 2 H), 5.1 (s, 2 H), 5.26 (d, J =
11.4, 1 H), 5.75 (d, J = 17.7, 1 H), 6.66–6.76 (t, 1 H), 7.31–7.41 (m, 4 H),
7.47–7.56 (m, 4 H), 7.91(d, J = 7.5, 1 H); 13C NMR (125 MHz, CDCl3, δ):
22.4, 33.6, 34.1, 51.8, 66.1, 79.8, 114.4, 126.4, 128.2, 128.4, 128.5, 132.1,
135.3, 136.3, 136.7, 137.6, 138.0, 140.7, 140.9, 142.12, 142.17, 142.20,
142.92, 142.98, 143.0 143.1, 143.8, 144.0, 144.4, 144.5, 144.7, 145.03,
145.07, 145.13, 145.18, 145.8, 147.8, 148.8, 172.9.
XRD Measurements: To produce samples with identical film thickness
for XRD measurement, blend films were spun on top of the PEDOT:PSS
layer. The film thickness of ca 250 nm was confirmed by alpha-step
profilometer and cross-sectional SEM. XRD spectra of the samples were
obtained using a Bruker D8 Advance spectrophotometer with CuKα
radiation.
SEM Measurements: Cross-sectional SEM images were measured by
using a field-emission SEM equipped with a cold field-emission-type gun
(JEOL JSM-7401F, Japan).
Acknowledgements
This work was supported by the National Science Council and “ATU
Plan” of the National Chiao Tung University and Ministry of Education,
Taiwan.
Received: November 26, 2010
Published online: March 17, 2011
[1] a) J. J. M. Halls, K. Pichler, R. H. Friend, S. C. Moratti, A. B. Holmes,
Appl. Phys. Lett. 1996, 68, 3120; b) M. Theander, A. Yartsev,
D. Zigmantas, V. Sundström, W. Mammo, M. R. Andersson,
O. Inganäs, Phys. Rev. B 2000, 61, 12957; c) A. Haugeneder,
M. Neges, C. Kallinger, W. Spirkl, U. Lemmer, J. Feldmann,
U. Scherf, E. Harth, A. Gügel, K. Müllen, Phys. Rev. B 1999, 59,
15346; d) T. Stübinger, W. Brütting, J. Appl. Phys. 2001, 90, 3632;
e) D. E. Markov, E. Amsterdam, P. W. M. Blom, A. B. Sieval,
J. C. Hummelen, J. Phys. Chem. A 2005, 109, 5266; f) Y.-J. Cheng,
S.-H. Yang, C.-S. Hsu, Chem. Rev. 2009, 109, 5868.
[2] a) W. Ma, C. Yang, X. Gong, K. Lee, A. J. Heeger, Adv. Funct.
Mater. 2005, 15, 1617; b) A. C. Mayer, S. R. Scully, B. E. Hardin,
M. W. Rowell, M. D. McGehee, Mater. Today 2007, 10, 28;
c)M.Reyes-Reyes,K.Kim,D.L.Carroll,Appl.Phys.Lett.2005,87,083506;
d) C. H. Woo, B. C. Thompson, B. J. Kim, M. F. Toney, J. M. J. Fréchet,
J. Am. Chem. Soc. 2008, 130, 16324.
[3] Y. Kim, S. Cook, S. M. Tuladhar, S. A. Choulis, J. Nelson,
J. R. Durrant, D. D. C. Bradley, M. Giles, I. McCulloch, C.-S. Ha,
M. Ree, Nat. Mater. 2006, 5, 197.
[4] M. Koppe, C. J. Brabec, S. Heiml, A. Schausberger, W. Duffy,
M. Heeney, I. McCulloch, Macromolecules 2009, 42, 4661.
Device Fabrication and Measurements: ITO-coated glass substrates
(provided by RiTdisplay corp.) were ultrasonicated in deionized (DI)
water with 2% detergent, acetone, and isopropyl alcohol for 10 min,
respectively, and subsequently dried under a stream of nitrogen. A
filtered dispersion of PEDOT:PSS in aqueous solution (Baytron PVP
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2011, 21, 1723–1732
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1731

www.afm-journal.de
www.MaterialsViews.com
[5] R. C. Hiorns, R. de Bettignies, J. Leroy, S. Bailly, M. Firon,
C. Sentein, A. Khoukh, H. Rreudhomme, C. Dagron-Lartigau, Adv.
Funct. Mater. 2006, 16, 2263.
[6] a) H. Hoppe, M. Niggenmann, C. Winder, J. Kraut, R. Hiesgen,
A. Hinsch, D. Meissner, N. S. Sariciftci, Adv. Funct. Mater. 2004, 14,
1005; b) D. Chirvase, J. Parisi, J. C. Hummelen, V. Dyakonov, Nano-
technology 2004, 15, 1317.
[7] a) G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery,
Y. Yang, Nat. Mater. 2005, 4, 864; b) F. Padinger, R. S. Rittberger,
N. S. Sariciftci, Adv. Funct. Mater. 2003, 13, 85; c) L. H. Nguyen,
H. Hoppe, T. Erb, S. Günes, G. Gobsh, N. S. Sariciftci, Adv. Funct.
Mater. 2007. 17, 1071; d) M. Reyes-Reyes, K. Kim, J. Dewald,
R. López-Sandoval, A. Avadhanula, S. Curran, D. L. Carroll, Org.
Lett. 2005, 7, 5749; e) H. Hoppe, T. Glatzel, M. Niggemann,
A. Hinsch, M. C. Lux-Steiner, N. S. Sariciftci, Nano Lett.,
2005, 5, 269; f) G. Li, Y. Yao, H. Yang, V. Shrotriya, G. Yang,
Y. Yang, Adv. Funct. Mater. 2007, 17, 1636; g) A. L. Ayzner,
D. D. Wanger, C. J. Tasson, S. H. Tolbert, B. J. Schwartz, J. Phys.
Chem. C 2008, 112, 18711; h) L. H. Nguyen, H. Hoppe, T. Erb,
S. Günes, G. Gobsch, N. S. Sariciftci, Adv. Funct. Mater. 2007, 17, 1071.
[8] H. Hoppe, N. S. Sariciftci, J. Mater. Chem. 2006, 16, 45 and refer-
ences therein.
[12] K. Sivula, C. K. Luscombe, B. C. Thompson, J. M. J Fréchet, J. Am.
Chem. Soc. 2006, 128, 13988.
[13] a) K. Sivula, Z. T. Ball, N. Watanabe, J. M. J. Fréchet, Adv. Mater.
2006, 18, 206; b) Z. Zhou, X. Chen, S. Holdcroft, J. Am. Chem. Soc.
2008, 130, 11711; c) J. B. Kim, K. Allen, S. J. Oh, S. Lee, M. F. Toney,
Y. S. Kim, C. R. Kagan, C. Nuckolls, Y.-L. Loo, Chem. Mater. 2010,
22, 5762; d) N. Camaioni, M. Catellani, S. Luzzati and A. Migliori,
Thin Solid Films, 2002, 403, 489.
[14] a) B. J. Kim, Y. Miyamoto, B. Ma, J. M. J. Fréchet, Adv. Funct.
Mater. 2009, 19, 2273; b) B. Gholamkhass, S. Holdcroft, Chem.
Mater. 2010, 22, 5371; c) Z. Zhu, S. Hadjikyriacou, D. Wallker,
R. Gaudiana, J. Macromol. Sci., Part A: Pure Appl. Chem. 2004,
41, 1467; d) M. Drees, H. Hoppe, C. Winder, H. Neugebauer,
N. S. Sariciftci, W. Schwinger, F. Schäffler, C. Topf, M. C. Scharber,
Z. Zhu, R. Gaudiana, J. Mater. Chem. 2005, 15, 5158;
e) J.-F. Nierengarten, S. Setayesh, New J. Chem. 2006, 30, 313.
[15] E. Zhou, Z. Tan, C. Yang, Y. Li, Macromol. Rapid Commun. 2006, 27,
793.
[16] a) Y.-J. Cheng, M. S. Liu, Y. Zhang, Y. Niu, F. Huang, J.-W. Ka,
H.-L. Yip, Y. Tian, A. K.-Y. Jen, Chem. Mater. 2008, 20, 413;
b) F. Huang, Y.-J. Cheng, Y. Zhang, M. S. Liu, A. K.-Y. Jen, J. Mater.
Chem. 2008, 18, 4495.
[9] a) E. Klimov, W. Li, X. Yang, G. G. Hoffmann, J. Loos, Macromolecules
2006, 39, 4493; b) Y.-C. Huang, S.-Y. Chuang, M.-C. Wu, H.-L. Chen,
C.-W. Chen, W.-F. Su, J. Appl. Phys. 2009, 106, 034506; c) H. Zhong,
X. Yang, B. deWith, J. Loos, Macromolecules 2006, 39, 218;
d) X. Yang, J. K. J. van Duren, R. A. J. Janssen, M. A. J. Michels,
J. Loos, Macromolecules 2004, 37, 2151; e) A. Swinnen,
I. Haeldermans, M. vande Ven, J. D’Haen, G. Vanhoyland, S. Aresu,
M. D’Olieslaeger, J. Manca, Adv. Funct. Mater. 2006, 16, 760;
f) C. Muller, T. A. M. Ferenczi, M. Campoy-Quiles, J. M. Frost,
D. D. C. Bradley, P. Smith, N. Stingelin-Stutzmann, J. Nelson, Adv.
Mater. 2008, 20, 3510.
[17] C.-H. Hsieh, Y.-J. Cheng, P.-J. Li, C.-H. Chen, M. Dubosc,
R.-M. Liang, and C.-S. Hsu, J. Am. Chem. Soc. 2010, 132,
4887.
[18] Y. Zhang, H.-L. Yip, O. Acton, S. K. Hau, F. Huang, A. K.-Y. Jen,
Chem. Mater. 2009, 21, 2598.
[19] a) M.-Y. Chiu, U-S. Jeng, C.-H. Su, K. S. Liang, K.-H. Wei, Adv. Mater.
2008, 20, 2573; b) T. Erb, U. Zhokhavets, G. Gobsch, S. Raleva,
B. Stühn, P. Schilinsky, C. Waldauf, C. J. Brabec, Adv. Funct. Mater.
2005, 15, 1193.
[20] a) G. Zhao, Y. He, Z. Xu, J. Hou, M. Zhang, J. Min, H.-Y. Chen,
M. Ye, Z. Hong, Y. Yang, Y. Li, Adv. Funct. Mater. 2010, 20, 1480;
b) L.-M. Chen, Z. Hong, W. L. Kwan, C.-H. Lu, Y.-F. Lai, B. Lei,
C.-P. Liu, Y. Yang, ACS nano 2010, 4, 4744.
[10] X. Yang, J. K. J. van Duren, M. T. Rispens, J. C. Hummelen,
R. A. J. Janssen, M. A. J. Michels, J. Loos, Adv. Mater. 2004, 16, 802.
[11] M. Jørgensen, K. Norrman, F. C. Krebs, Sol. Energy Mater. Sol.Cells
2008, 92, 686.
[21] C. Goh, R. J. Kline, M. D. McGehee, E. N. Kadnikova, J. M. J. Fréchet,
Appl. Phys. Lett. 2005, 86, 122110.
©
1732 wileyonlinelibrary.com
2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2011, 21, 1723–1732