Methyl and Dimethyl o-Xylenyl-Substituted Fullerene Acceptors for Polymer Solar Cells

Article abstract of DOI:10.1002/ijch.201400207

A series of fullerene derivatives based on methyl- and dimethyl-substituted o-xylenyl groups were synthesized via Diels-Alder reaction from fullerene (C₆₀) and dienes derived from α,α′-dibromo-o-xylene derivatives. The synthesized 1-methyl-4,5-

Full text of DOI:10.1002/ijch.201400207

Full Paper
DOI: 10.1002/ijch.201400207
Methyl and Dimethyl o-Xylenyl-Substituted Fullerene
Acceptors for Polymer Solar Cells
Hee Un Kim,[a] Jong Baek Park,[a] Andrew C. Grimsdale,[b] and Do-Hoon Hwang*[a]
Abstract: A series of fullerene derivatives based on methyl-
and dimethyl-substituted o-xylenyl groups were synthesized
via DielsÀAlder reaction from fullerene (C60) and dienes de-
rived from a,a’-dibromo-o-xylene derivatives. The synthe-
sized 1-methyl-4,5-xylenyl- and 1,2-dimethyl-4,5-xylenyl-sub-
stituted fullerene mono-, bis-, and trisadducts, i.e.,
and 4.10”103 MÀ1, respectively. These values correlate with
the binding affinities of the fluorophore and quencher. The
lowest unoccupied molecular orbital energy levels increased
with increasing number of adducts on the fullerene because
of their decreasing number of sp2-hybridized carbons. Pho-
tovoltaic devices were fabricated using P3HT and the
mono-, bis-, and trisadducts as electron donors and accept-
ors, respectively. Among these, the cells with an ITO/PE-
DOT:PSS/P3HT:DM_OXCBA/LiF/Al structure had the high-
est power conversion efficiency of 4.77% with an open-cir-
cuit voltage of 0.88 V, a short-circuit current density of
10.52 mAcmÀ2, and a fill factor of 0.51 under AM 1.5G
(100 mWcmÀ2) illumination.
M_OXCMA,
M_OXCBA,
M_OXCTA,
DM_OXCMA,
DM_OXCBA, and DM_OXCTA, were used as electron-ac-
ceptor materials for polymer solar cells and have good solu-
bility in common organic solvents. The SternÀVolmer
quenching constants for poly(3-hexylthiophene) (P3HT) as
a fluorophore with M_OXCMA, M_OXCBA, M_OXCTA,
DM_OXCMA, DM_OXCBA, and DM_OXCTA as quenchers
were 1.35”104, 7.67”103, 3.72”103, 1.38”104, 7.95”103,
Keywords: electron acceptors · fullerenes · photovoltaic devices · polymer solar cells · renewable resources
1. Introduction
LUMO of the electron acceptor and the highest occupied
Solution-processed polymer solar cells (PSCs) have at-
tracted considerable interest over the past few years as
a promising source of renewable energy because of their
advantages such as enabling the fabrication of inexpen-
sive and lightweight solar panels through roll-to-roll
printing technology.[1,2] In particular, solution-processed
bulk heterojunction (BHJ) PSC architectures have em-
ployed electron-donating (p-type) polymers, such as
poly(3-hexylthiophene) (P3HT) and low band-gap
donorÀacceptor copolymers, and electron-accepting (n-
type) materials, such as [6,6]-phenyl-C61-butyric acid
methyl ester (PC61BM).[3] Currently, fullerenes such as
PC61BM are the most widely employed electron-acceptor
materials in solution-processed PSCs because of their
high electron affinities and good charge-transporting abil-
ities.[4,5]
In recent years, various derivatives of fullerene electron
acceptors have been developed via structural modifica-
tion of the substituents in order to improve the perform-
ances of PSCs.[6,7] In particular, fullerene derivatives with
high lowest unoccupied molecular orbitals (LUMOs)
have been designed and synthesized to increase the open-
circuit voltage (Voc) values of PSCs, because the Voc of
a PSC is proportional to the energy gap between the
molecular orbital (HOMO) of the electron donor.[8]
Recently, P3HT-based PSCs with high Voc values have
been reported; these involve fullerene bisadducts, which
have high LUMO energy levels.[9,10] For example, Li and
Kim et al. synthesized indene-C60 bisadduct (IC60BA),[9a–c]
o-xylenyl-C60 bisadduct (OXCBA),[10a,b] and dihydro-
naphthyl-based C60 bisadduct (NC60BA),[10c] among
others.[11a,b] The parent fullerene (C60) contains 60 sp2
carbon atoms, while the fullerene adducts contain fewer
[a] H. U. Kim, J. B. Park, D.-H. Hwang
Department of Chemistry and Chemistry Institute for Functional
Materials
Pusan National University
Busan 609-735 (Republic of Korea)
Phone: (+82) 51-510-2232
Fax: (+82) 51-516-7421
e-mail: dohoonhwang@pusan.ac.kr
[b] A. C. Grimsdale
School of Materials Science and Engineering
Nanyang Technological University
50 Nanyang Avenue
Singapore 639798 (Singapore)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/ijch.201400207.
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1017

Full Paper
sp2 carbon atoms: fullerene mono-, bis-, and trisadducts
contain 58, 56, and 54 sp2 carbon atoms, respectively. The
substituents disrupt the p-conjugated system of the fuller-
ene and thus increase the gap between the HOMO and
LUMO. Interestingly, the HOMO energy levels of fuller-
ene multi-adducts usually do not change significantly
whereas the LUMO energy levels are higher than those
of the parent fullerene. Because the Voc of a solar cell is
related to the energy difference between the LUMO of
the acceptor and the HOMO of the donor, the increased
LUMO energy levels of fullerene multi-adducts could in-
crease the Voc values of PSCs if they are used as accept-
ors.
419, 404, 405, 397, and 3928, respectively (Figure S3 in the
Supporting Information). The Td values of both series of
fullerene derivatives decrease with increasing number of
substituents.
In this paper, we report the synthesis and investigation
of the properties of methyl and dimethyl o-xylenyl-substi-
tuted mono-, bis-, and trisadduct fullerenes (i.e., M/DM_
OXCMA, M/DM_OXCBA, and M/DM_OXCTA); in
these compounds, the methyl groups on the o-xylenyl
moieties are expected to improve the solubility of the
multi-adduct fullerenes. In particular, the effect of the
number of o-xylenyl adducts on the photoluminescence
(PL) quenching abilities and electrochemical, charge-
transporting, and photovoltaic properties of the fullerenes
are investigated. The syntheses and chemical structures of
the new fullerene derivatives are shown in Scheme 1. Fig-
ures S1 and S2 in the Supporting Information show the
13C NMR and MALDI-TOF mass spectra of the synthe-
sized fullerene derivatives.
Figure 1. UVÀvisible absorption spectra of (a) methyl- and dimeth-
yl-substituted fullerene derivatives at a 10À5 M concentration in
1,2-dichlorobenzene and (b) P3HT:acceptor blend films.
Scheme 1. Syntheses and chemical structures of methyl- and di-
methyl-substituted o-xylenyl-based fullerene derivatives.
Table 1. Molar absorption coefficients of the synthesized fullerene
derivatives and PC61BM.
Acceptor
Optical (10À5 mol/L)
lmax, abs (nm)
emax (MÀ1 cmÀ1
)
2. Results and Discussion
M_OXCMA
M_OXCBA
M_OXCTA
DM_OXCMA
DM_OXCBA
DM_OXCTA
PC61BM
320
320
320
320
320
320
330
36,300
47,200
61,900
43,600
51,800
70,300
39,500
2.1. Thermal Properties
The thermal properties of the synthesized fullerene deriv-
atives were investigated by thermogravimetric analysis
(TGA). The thrmal stabilities of the synthesized fullerene
derivatives were evaluated by measuring the temperature
at 5% weight loss (Td). The decomposition temperatures
of M_OXCMA, M_OXCBA, M_OXCTA, DM_
OXCMA, DM_OXCBA, and DM_OXCTA were 427,
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.ijc.wiley-vch.de
1018

Full Paper
2.2. Optical Properties
the fullerene derivatives, the absorption spectra are domi-
nated by a band around 310–330 nm, which originates
from the fullerene moiety.[12] All the synthesized fullerene
derivatives except M_OXCMA exhibit higher molar ab-
sorption coefficients than PC61BM. As can be seen from
the data summarized in Table 1, the molar absorption co-
efficients of the fullerene derivatives at 320 nm increase
as the number of substituents on the fullerene increases,
due to the pÀp* transition of the tetraalkyl-substituted
benzene units.
As shown in Figure 1a, the UVÀvisible absorption spectra
of the methyl- and dimethyl-substituted o-xylenyl fuller-
ene derivatives in 1,2-dichlorobenzene solution (10À5 M)
show similar absorption bands in the range of 300–
450 nm with absorption maxima (lmax) at 320 nm. For all
Figure 1b shows the absorption spectra of the
P3HT:acceptor blend films. All of the blend films
showed similar absorption bands; the vibronic peak at
around 600 nm is indicative of the ordered structure of
P3HT.[9c] The spectra of the P3HT:mono- and bisadduct
fullerene (i.e., M/DM_OXCMA and M/DM_OXCBA)
blend films are similar to that of the P3HT:PC61BM
blend film with a lmax of 510 nm. However, the absorption
maximum of the P3HT:trisadduct fullerene (i.e., M/DM_
OXCTA) blend film is blue shifted by about 14 nm from
that of the P3HT:PC61BM blend film. This is because the
bulky M_OXCTA and DM_OXCTA molecules disrupt
the two-dimensional packing of the P3HT chains, and
thus their crystallization.[13] To estimate the photolumines-
cence quenching abilities of the synthesized fullerene de-
rivatives in solution, we measured the PL intensity of
a P3HT solution (fluorophore), while increasing the con-
centrations of the synthesized fullerene derivatives and
PC61BM (quenchers). Pure P3HT solution showed an
emission maximum at 576 nm under excitation at 450 nm;
this changed only slightly upon the addition of the fuller-
ene derivatives (Figure S4 in the Supporting Informa-
tion). To compare the quenching efficiencies of the fuller-
ene derivatives, we used the SternÀVolmer equation, as
follows:[14a]
Iꢀ=I ¼ 1 þ Ksv ½QŠ
ð1Þ
Figure 2. (a) SternÀVolmer plots of P3HT quenching by M_
OXCMA, M_OXCBA, M_OXCTA, DM_OXCMA, DM_OXCBA, DM_
OXCTA, and PC61BM in solution (10À5 M) and (b) PL spectra of pure
P3HT and P3HT:acceptor blend films.
where I8 and I are the measured PL intensities in the ab-
sence and presence of the quencher, respectively, Ksv is
the SternÀVolmer quenching constant, and [Q] is the con-
Table 2. SternÀVolmer quenching constants (Ksv) of P3HT quenching by M_OXCMA, M_OXCBA, M_OXCTA, DM_OXCMA, DM_OXCBA,
DM_OXCTA, and PC61BM in solution (10À5 M), and calculated PL quenching efficiencies for the P3HT:acceptor blend films.
Acceptor
10À5 mol/L
Ksv [MÀ1
PL quenching (thin film)
[b]
[b]
]
IAcceptor[a]/IP3HT
1À(IAcceptor[a]/IP3HT
)
Quench. eff. (%)
M_OXCMA
M_OXCBA
M_OXCTA
DM_OXCMA
DM_OXCBA
DM_OXCTA
PC61BM
1.35”104
7.67”103
3.72”103
1.38”104
7.95”103
4.10”103
5.85”103
0.157
0.242
0.314
0.176
0.211
0.292
0.047
0.843
0.758
0.686
0.824
0.789
0.708
0.953
84.3
75.8
68.6
82.4
78.9
70.8
95.3
[a] PL intensity in the presence of the acceptor. [b] PL intensity of pure P3HT.
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.ijc.wiley-vch.de
1019

Full Paper
2.3. Electrochemical Properties
centration of the quencher. The SternÀVolmer equation
provides useful information regarding the electron-trans-
fer process. The SternÀVolmer quenching plots of the
P3HT solutions with the synthesized fullerene derivatives
as quenchers are shown in Figure 2a. All of the SternÀ
Volmer plots are linear in the low quencher concentration
regime. From these plots, the relative quenching efficien-
cies of the fullerene derivatives were determined: the cal-
culated Ksv values for M_OXCMA, M_OXCBA,
M_OXCTA, DM_OXCMA, DM_OXCBA, DM_OXC-
TA, and PC61BM were 1.35”104, 7.67”103, 3.72”103,
1.38”104, 7.95”103, 4.10”103, and 5.85”103, respectively.
The mono- and bisadducts (i.e., M/DM_OXCMA and M/
DM_OXCBA) have stronger quenching behavior than
PC61BM, whereas the trisadducts (i.e., M/DM_OXCTA)
show relatively weak quenching abilities. The bis- and tri-
sadduct fullerenes, such as M/DM_OXCBA and M/
DM_OXCTA, contain fewer sp2 carbons, which reduces
their electron affinity. A higher Ksv value is correlated
with a higher binding affinity between the fluorophore
and quencher,[14b] thus indicating that fullerenes with
fewer adducts have higher binding affinities toward
P3HT.
To estimate the LUMO energy levels of the synthesized
fullerene derivatives, we studied their electrochemical
properties using cyclic voltammetry (CV) in 1,2-dichloro-
benzene solution. As shown in Figure 3, the CV plots fea-
ture three quasi-reversible reduction waves for all the
methyl- and dimethyl-substituted o-xylenyl fullerene de-
rivatives, and the measured reduction potentials are sum-
marized in Table 3. It can be seen that the onset reduc-
tion potentials (Eonsetred) of all the synthesized fullerene
derivatives are negatively shifted compared to that of
PC61BM. The Eonset values of M_OXCMA, M_OXCBA,
red
M_OXCTA,
DM_OXCMA,
DM_OXCBA,
DM_
OXCTA, and PC61BM are À0.89, À1.01, À1.06, À0.88,
À1.00, À1.03, and À0.84 eV, respectively. To determine
the LUMO energies of the fullerene derivatives, the ref-
erence electrode was calibrated with ferrocene/ferroceni-
um (Fc/Fc+), which has a redox potential with an abso-
lute energy level of À4.80 eV in vacuum; the potential of
this external standard under the experimental conditions
is 0.26 eV. The LUMO energy values of the fullerene de-
rivatives were calculated according to the following equa-
tion:
The PL quenching spectra of the P3HT:acceptor blend
films with the same composite ratios (1:0.7) and thick-
nesses (100 nm) were also recorded to observe the elec-
tron-transfer properties in the solid state. The PL emis-
sion spectra of pure P3HT and P3HT:M/DM_OXCMA,
P3HT:M/DM_OXCBA, P3HT:M/DM_OXCTA, and
P3HT:PC61BM blend films are shown in Figure 2b. The
peak emission of the pure P3HT film is at 665 nm with
excitation at 450 nm, and the PL intensity of the pure
P3HT film dramatically decreases upon the addition of
the acceptors. The PL emission of the P3HT:PC61BM
blend was significantly quenched (95.3% relative to
P3HT); in contrast, the PL emissions of the P3HT:M/
DM_OXCMA, P3HT:M/DM_OXCBA, and P3HT:M/
DM_OXCTA blends were less efficiently quenched with
measured values of 84.3%/82.4%, 75.8%/78.9%, and
68.6%/70.8%, respectively. The PL quenching efficiencies
of the fullerene monoadduct blend films are similar and
only slightly lower than that of the P3HT:PC61BM blend,
whereas the bis- and trisadduct blend films exhibit consid-
erably lower quenching efficiencies. The obtained PL
quenching efficiencies are consistent with the Ksv values
obtained for the P3HT:acceptor solutions. It is known
that the electron affinities of multi-adducts, such as the
bisadduct (56 sp2 carbons) and trisadduct (54 sp2 carbons),
are lower than those of the monoadduct (58 sp2 carbons)
and basic fullerene (60 sp2 carbons) because they contain
fewer sp2 carbons.[10] The measured Ksv values and PL
quenching efficiencies of the P3HT:acceptor solutions
and blend films are summarized in Table 2.
YLUMO ¼ À½ðEonsetredÀEfcÞ þ 4:8Š ðeVÞ
ð2Þ
where Efc is 0.26 eV versus Ag/Ag+. The calculated
LUMO energy levels of M_OXCMA, M_OXCBA, M_
OXCTA, DM_OXCMA, DM_OXCBA, DM_OXCTA,
and PC61BM are thus À3.65, À3.53, À3.48, À3.66, À3.54,
À3.51, and À3.70 eV, respectively. The LUMO energy
levels of all the synthesized fullerene derivatives are
higher than that of PC61BM. The higher LUMO energy
levels of the bis- and trisadduct fullerene derivatives are
attributed to interruption of the p-conjugated system of
the fullerenes. The number of p electrons decreases grad-
ually from 58 in M/DM_OXCMA to 56 in M/DM_
OXCBA and 54 in M/DM_OXCTA; these values corre-
late with the decreasing number of sp2 carbon atoms in
the fullerenes. The high LUMO energy levels of the full-
erene derivatives are expected to result in high open-cir-
cuit voltages in photovoltaic cells.[8]
2.4. Charge-Carrier Mobilities
The charge-carrier mobilities of the synthesized fullerene
derivatives blended with P3HT were studied using the
space-charge-limited current (SCLC) method. To measure
the SCLC mobility, we fabricated hole-only and electron-
only devices, i.e., ITO/PEDOT:PSS/P3HT:acceptor/Au
and ITO/ZnO/P3HT:acceptor/LiF/Al, respectively. The
current densityÀvoltage (JÀV) characteristics were mea-
sured in the absence of light. The JÀV measurements
were recorded in the range of 0–8 V, and the results were
estimated using the MottÀGurney square law:[15]
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.ijc.wiley-vch.de
1020

Full Paper
and cathode. The series and contact resistance of the
device (~15 W) was measured using a blank device (ITO/
ZnO/Al), and the voltage drop due to this resistance (Vr)
was subtracted from the applied voltage. Figure S5 in the
Supporting Information shows the JÀV curves of the
P3HT:M_OXCMA,
P3HT:M_OXCBA,
P3HT:M_
OXCTA, P3HT:DM_OXCMA, P3HT:DM_OXCBA,
P3HT:DM_OXCTA, and P3HT:PC61BM blend films for
hole-only devices; very similar hole mobility values of
3.39”10À4, 3.94”10À4, 3.69”10À4, 3.59”10À4, 3.88”10À4,
3.71”10À4, and 3.48”10À4 cm2 VÀ1 sÀ1, respectively, were
obtained.
Table 4. Hole and electron mobility values of P3HT:fullerene deriva-
tives and P3HT:PC61BM blend films calculated using the SCLC
method.
P3HT:Acceptor
mh (cm2 VÀ1 sÀ1
)
me (cm2 VÀ1 sÀ1
)
mh/me
1.61
1.76
242.36
1.64
1.68
226.22
1.53
M_OXCMA
M_OXCBA
M_OXCTA
DM_OXCMA
DM_OXCBA
DM_OXCTA
PC61BM
3.39”10À4
3.94”10À4
3.69”10À4
3.59”10À4
3.88”10À4
3.71”10À4
3.48”10À4
2.11”10À4
2.24”10À4
1.52”10À6
2.19”10À4
2.31”10À4
1.64”10À6
2.28”10À4
Figure 4 shows the plots from the electron-only devi-
ces; from these plots, the electron mobilities of
P3HT:M_OXCMA,
P3HT:M_OXCBA,
P3HT:M_
Figure 3. Cyclic voltammograms of (a) M_OXCMA, M_OXCBA, and
M_OXCTA and (b) DM_OXCMA, DM_OXCBA, and DM_OXCTA in
1,2-dichlorobenzene solution.
OXCTA, P3HT:DM_OXCMA, P3HT:DM_OXCBA,
P3HT:DM_OXCTA, and P3HT:PC61BM blends were
determined to be 2.11”10À4, 2.24”10À4, 1.52”10À6, 2.19”
10À4, 2.31”10À4, 1.64”10À6, and 2.28”10À4 cm2 VÀ1 sÀ1, re-
spectively. The measured SCLC hole and electron mobili-
ties of the P3HT:mono- and bisadduct fullerene deriva-
tives (i.e., M/DM_OXCMA and M/DM_OXCBA) are
similar to that of the P3HT:PC61BM device. On the other
hand, the electron mobilities of the P3HT:trisadduct full-
erene (i.e., M/DM_OXCTA) devices are two orders of
magnitude lower. This result is consistent with previously
reported values for devices containing fullerene trisad-
ducts and probably results from the increased disorder of
the fullerene phase caused by the addition of the solubi-
lizing group and presence of a number of isomers of the
trisadduct fullerenes; both of these factors prevent effi-
cient packing of the fullerenes and decrease electron
transport through the fullerene phase.[10a] Well-balanced
charge-carrier transport is important for increasing the fill
factor (FF) of BHJ solar cells.[16] We therefore compared
the hole mobility to electron mobility ratios (mh/me). The
mh/me values for the P3HT:M_OXCMA, P3HT:M_
OXCBA, P3HT:M_OXCTA, P3HT:DM_OXCMA,
Table 3. Electrochemical properties of M_OXCMA, M_OXCBA, M_
OXCTA, DM_OXCMA, DM_OXCBA, DM_OXCTA, and PC61BM.
Acceptor
E1 (eV) E2 (eV) Eonset (eV) LUMO (eV)
red red red
M_OXCMA
M_OXCBA
M_OXCTA
À1.09
À1.23
À1.46
À1.51
À1.65
À1.87
À1.49
À1.65
À1.85
À1.38
À0.89
À3.65
À3.53
À3.48
À3.66
À3.54
À3.51
À3.70
À1.01
À1.06
À0.88
À1.00
À1.03
À0.84
DM_OXCMA À1.07
DM_OXCBA
DM_OXCTA
PC61BM
À1.25
À1.42
À0.99
8
9
V2
L3
ð3Þ
J ¼ ereom
where J is the current density, er is the dielectric constant
of the polymer or fullerene derivative, e0 is the permittivi-
ty of vacuum, m is the hole or electron mobility, L is the
film thickness, and V=VapplÀVbiÀVr, where Vappl is the ap-
plied potential and Vbi is the built-in voltage that results
from the difference in the work functions of the anode
P3HT:DM_OXCBA,
P3HT:PC61BM devices were 1.61, 1.76, 242.36, 1.64, 1.68,
226.22, and 1.53, respectively. The P3HT blend films with
P3HT:DM_OXCTA,
and
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.ijc.wiley-vch.de
1021

Full Paper
2.5. Photovoltaic Performances
Photovoltaic devices were fabricated using P3HT as the
electron donor and the fullerene derivatives as the elec-
tron acceptors with device configurations of ITO/PE-
DOT:PSS/P3HT:acceptor/LiF/Al. The most efficient de-
vices fabricated using the fullerene derivatives were ob-
tained using a P3HT:acceptor blend ratio of 1:0.7 (w/w)
with annealing at 1508 for 10 min. Figure 5 shows the JÀV
curves of these photovoltaic cells and the photovoltaic
characteristics of the fabricated devices are summarized
in Table 5 along with the value for a P3HT:PC61BM ref-
erence device. The highest power conversion efficiency
(PCE) values of the P3HT:M_OXCMA, P3HT:M_
OXCBA, P3HT:M_OXCTA, P3HT:DM_OXCMA,
P3HT:DM_OXCBA,
P3HT:DM_OXCTA,
and
P3HT:PC61BM devices were 2.52%, 4.15%, 1.39%,
3.19%, 4.77%, 1.50%, and 3.25%, respectively. The solar
cells with the bisadducts M_OXCBA and DM_OXCBA
gave the best performance. In particular, the device with
DM_OXCBA as an electron acceptor had a PCE of
4.77%, a Voc of 0.88 V, a short-circuit current density (Jsc)
of 10.52 mAcmÀ2, and an FF of 0.51; this PCE value is
among the highest reported for devices fabricated with
P3HT as a donor.
The Voc values for the P3HT:M_OXCMA, P3HT:M_
OXCBA, P3HT:M_OXCTA, P3HT:DM_OXCMA,
P3HT:DM_OXCBA,
P3HT:DM_OXCTA,
and
P3HT:PC61BM devices were 0.65, 0.88, 0.93, 0.66, 0.88,
0.92, and 0.64 V, respectively. The Voc values of the devi-
ces fabricated using monoadducts (i.e., M/DM_OXCMA)
as electron acceptors are similar to that of the PC61BM
device, while the values for the devices with bisadducts
(i.e., M/DM_OXCBA) and trisadducts (i.e., M/DM_
OXCTA) are significantly higher. These increased Voc
values are consistent with the LUMO energy levels of the
acceptors because the Voc value is directly proportional to
the energy gap between the LUMO energy level of the
electron acceptor and the HOMO energy level of the
electron donor.[8]
Figure 4. Measured space-charge-limited current (SCLC) JÀV char-
acteristics of (a) P3HT:M_OXCMA, P3HT:M_OXCBA, P3HT:M_OXC-
TA, and P3HT:PC61BM and (b) P3HT:DM_OXCMA, P3HT:DM_OXC-
BA, P3HT:DM_OXCTA, and P3HT:PC61BM blend devices under dark
conditions for electron-only devices.
mono- or bisadducts exhibited relatively well balanced mh/
me values, unlike the blended films with trisadducts; the
P3HT:M_OXCTA and P3HT:DM_OXCTA films have
very unbalanced mh/me ratios because of the very low elec-
tron mobilities of the acceptors. Unbalanced mobility
ratios cause the carriers with lower mobility to accumu-
late in a solar cell. The additional electric field in the
device due to the accumulated carriers can hinder extrac-
tion of the charge carriers, resulting in decreased mobility
and FF for the solar cell. A serious imbalance of mobility
can thus produce S-shaped JÀV curves and a very low FF,
both of which lead to poor overall device performance.[17]
The parameters calculated using the SCLC method are
summarized in Table 4.
The measured Jsc values of the P3HT:M_OXCMA,
P3HT:M_OXCBA, P3HT:M_OXCTA, P3HT:DM_
OXCMA, P3HT:DM_OXCBA, P3HT:DM_OXCTA,
and P3HT:PC61BM devices were 6.92, 9.85, 3.96, 8.75,
10.52, 4.47, and 8.53 mAcmÀ2, respectively, with the bi-
sadduct (i.e., M/DM_OXCBA) devices showing the high-
est values; in contrast, the monoadduct (i.e., M/DM_
OXCMA) devices exhibited similar values to that of the
PC61BM device, while the Jsc values of the trisadduct (i.e.,
M/DM_OXCTA) devices were notably lower than those
of all the other devices. The relatively high Jsc of the
P3HT:DM_OXCBA device is likely related to its rela-
tively high electron mobility.
The P3HT:M_OXCMA and P3HT:DM_OXCMA de-
vices show higher FF values (0.56) than the other devices
(0.36–0.51). The FF of the device fabricated using
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.ijc.wiley-vch.de
1022

Full Paper
Figure 6. EQE curves of P3HT:M_OXCBA and P3HT:DM_OXCBA
devices annealed at 1508 for 10 min.
because of their unbalanced mh/me ratios, as determined
using the SCLC method (Table 4).
The external quantum efficiency (EQE) values were
measured under optimized device conditions for the
P3HT:M_OXCBA and P3HT:DM_OXCBA devices, as
shown in Figure 6. The maximum EQE values of the
P3HT:M_OXCBA and P3HT:DM_OXCBA devices at
500 nm were 58.4% and 61.5%, respectively. The EQE
value of the P3HT:DM_OXCBA device is slightly higher
than that of the P3HT:M_OXCBA device; this observa-
tion is consistent with the Jsc values obtained for the fab-
ricated devices.
Figure 5. JÀV curves of the PSCs fabricated using (a) P3HT:M_
OXCMA/M_OXCBA/M_OXCTA and (b) P3HT:DM_OXCMA/DM_
OXCBA/DM_OXCTA active layers under AM 1.5G illumination of
2.6. Morphologies
100 mWcmÀ2
.
The surface morphologies of the blend films of P3HT and
the fullerene derivatives (1:0.7, w/w) spin-coated from
1,2-dichlorobenzene solutions were studied using tapping-
mode atomic force microscopy (AFM). Figure 7 shows
AFM images of the blend films of the P3HT and fullerene
Table 5. Photovoltaic performances of PSCs based on P3HT:accep-
tor (1:0.7, w/w) and P3HT:PC61BM blends, with annealing at 1508
for 10 min, under AM 1.5G illumination of 100 mWcmÀ2
.
P3HT:Acceptor
Voc (V)
Jsc (mAcmÀ2
)
FF
PCE (%)
M_OXCMA
M_OXCBA
M_OXCTA
DM_OXCMA
DM_OXCBA
DM_OXCTA
PC61BM
0.65
0.88
0.93
0.66
0.88
0.92
0.64
0.61
6.92
9.85
3.96
8.75
10.52
4.47
8.53
9.02
0.56
0.48
0.38
0.56
0.51
0.36
0.60
0.64
2.52
4.15
1.39
3.19
4.77
1.50
3.25
3.55
PC71BM
DM_OXCBA is slightly lower than those of the
P3HT:M_OXCMA and P3HT:DM_OXCMA devices.
The P3HT:M_OXCTA and P3HT:DM_OXCTA devices
had the lowest FF values of all the devices; this could be
Figure 7. AFM images of the blend films of P3HT and the fullerene
derivatives; (a) P3HT:M_OXCMA, (b) P3HT:M_OXCBA, (c)
P3HT:M_OXCTA, (d) P3HT:DM_OXCMA, (e) P3HT:DM_OXCBA, and
(f) P3HT:DM_OXCTA.
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.ijc.wiley-vch.de
1023

Full Paper
derivatives annealed at 1508 for 10 min; these were found
to be the optimum device fabrication conditions. In the
AFM images, the brighter islands and darker valleys cor-
respond to the P3HT and fullerene domains, respectively.
The surface root-mean-square (rms) roughness values of
4. Experimental Section
4.1. Materials
Lithium aluminum hydride (LiAlH4), phosphorus tribro-
mide, 5-methylisobenzofuran-1,3-dione, 5,6-dimethyliso-
benzofuran-1,3-dione, potassium iodide, 18-crown-6, and
1,2-dichlorobenzene were purchased from Aldrich. Fuller-
ene (C60) was purchased from Nano-C. All chemicals
were used without further purification.
P3HT:M_OXCMA,
P3HT:M_OXCBA,
P3HT:M_
OXCTA, P3HT:DM_OXCMA, P3HT:DM_OXCBA,
and P3HT:DM_OXCTA were 5.32, 5.24, 4.96, 5.47, 5.16,
and 4.87 nm, respectively. In general, a rougher surface of
P3HT-based blend films indicates P3HT self-organization,
which enhances the formation of ordered structures.[18]
Therefore, the P3HT in P3HT:mono- and bisadduct blend
films might have a more ordered structure than that in
the P3HT:trisadduct blend films. Also, the blend films of
P3HT with mono- and bisadducts (i.e., M/DM_OXCMA
and M/DM_OXCBA) demonstrate that the donor and ac-
ceptor interpenetrating networks of these blend films are
better than those of the P3HT blend films with trisad-
ducts. Notably, the P3HT:trisadduct (i.e., M/DM_
OXCTA) blend films had the lowest roughness values;
their overly smooth morphologies may have been caused
by homogeneous separation of P3HT and the fullerene
derivatives. Such homogeneous blend films might lead to
insufficient phase separation and fragmented electron
channels.[19]
4.2. Synthesis of Fullerene Derivatives
Synthesis of (4-Methyl-1,2-phenylene)dimethanol (1)
A solution of 5-methylisobenzofuran-1,3-dione (5.0 g,
30.8 mmol) in THF (150 mL) was cooled to 08, and
LiAlH4 (2.9 g, 77.1 mmol) was slowly added. The reaction
mixture was stirred at 08 for 1 h and then warmed to
room temperature. After stirring overnight, the reaction
was quenched by the dropwise addition of 20 wt % am-
monium chloride solution. The resulting mixture was par-
titioned between diethyl ether and brine, and the organic
layer was dried with anhydrous MgSO4. The crude prod-
uct was purified by flash column chromatography using
hexane/ethyl acetate as the eluent to yield 1 (4.28 g,
91.3%). 1H NMR (300 MHz, CDCl3, d): 7.19 (d, 1H),
7.13 (s, 1H), 7.06 (d, 1H), 4.54 (s, 2H), 4.52 (s, 2H), 4.31
(broad OÀH, 2H), 2.36 (s, 3H); 13C NMR (75 MHz,
CDCl3, d): 137.25, 136.68, 136.24, 129.27, 127.26, 128.09,
62.51, 62.42, 21.62.
3. Conclusions
We designed and synthesized methyl- and dimethyl-sub-
stituted o-xylenyl-based fullerene derivatives, i.e., M_
OXCMA, M_OXCBA, M_OXCTA, DM_OXCMA,
DM_OXCBA, and DM_OXCTA, and investigated their
use as electron acceptors in P3HT-based BHJ solar cells.
The new fullerene derivatives effectively quenched the
PL of P3HT as a fluorophore. M_OXCMA, M_OXCBA,
M_OXCTA, DM_OXCMA, DM_OXCBA, and DM_
OXCTA were used as electron acceptors in P3HT-based
BHJ solar cells with device structures of ITO/PE-
DOT:PSS/P3HT:acceptor/LiF/Al. The highest PCE
values observed for M_OXCMA, M_OXCBA, M_
OXCTA, DM_OXCMA, DM_OXCBA, and DM_
OXCTA were 2.52%, 4.15%, 1.39%, 3.19%, 4.77%, and
1.50%, respectively, compared to a value of 3.25% for
PC61BM. The higher PCE values of the devices contain-
ing the bisadducts M_OXCBA and DM_OXCBA likely
resulted from their higher Voc and Jsc values because of
the higher LUMO energy levels and slightly higher elec-
tron mobilities, respectively, than those of PC61BM. In
particular, bisadducts such as M_OXCBA and DM_
OXCBA are useful electron acceptors for PSCs and could
be promising replacements for PC61BM to achieve PSCs
with higher Voc values.
Synthesis of (4,5-Dimethyl-1,2-phenylene)dimethanol (2)
Compound 2 (4.42 g, 93.7%) was prepared from 5,6-di-
methylisobenzofuran-1,3-dione via the same procedure as
1
that used to make 1. H NMR (300 MHz, acetone-d6, d):
7.12 (s, 2H), 4.61 (s, 4H), 2.22 (s, 6H); 13C NMR
(75 MHz, CDCl3, d): 135.89, 133.68, 129.17, 62.72, 19.15.
Synthesis of 1,2-Bis(bromomethyl)-4-methylbenzene (3)
Phosphorus tribromide (5.5 g, 60.9 mmol) was added to
a solution of compound 1 (3.0 g, 19.7 mmol) in diethyl
ether (150 mL), and the resulting mixture was stirred at
08 for 1 h and then refluxed overnight. The resulting mix-
ture was partitioned between diethyl ether and brine, and
the organic layer was dried with anhydrous MgSO4. The
crude product was purified by flash column chromatogra-
phy using hexane/ethyl acetate as the eluent to yield 3
1
(4.58 g, 83.5%). H NMR (300 MHz, acetone-d6, d): 7.37
(d, 1H), 7.30 (s, 1H), 7.17 (d, 1H), 4.78 (s, 2H), 4.76 (s,
2H), 3.03 (broad OÀH, 2H) 2.31 (s, 3H); 13C NMR
(75 MHz, CDCl3, d): 138.95, 137.26, 134.88, 131.52,
129.57, 129.26, 30.47, 30.16, 21.66.
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.ijc.wiley-vch.de
1024

Full Paper
Synthesis of 1,2-Bis(bromomethyl)-4,5-dimethylbenzene (4)
as that used to make the methyl o-xylene-substituted full-
erene derivatives from 3. DM_OXCMA: 1H NMR
(300 MHz, CDCl3, d): 7.42 (s, 2H), 4.76 (d, 2H), 4.37 (d,
2H), 2.44 (s, 6H); 13C NMR (75 MHz, CDCl3/CS2, d):
161.08, 158.56, 157.87, 157.65, 148.88, 148.51, 147.87,
147.54, 146.46, 146.13, 145.77, 144.51, 144.20, 143.63,
143.09, 142.91, 142.44, 142.03, 141.63, 141.33, 141.03,
139.83, 138.36, 128.36, 64.65, 43.11, 43.06, 20.63; MALDI-
TOF MS (M+1, C70H12): calcd, 852.09; found, 852.17;
purity: 99.3% by HPLC. DM_OXCBA: 1H NMR
(300 MHz, CDCl3, d): 7.67–7.03 (m, 4H), 4.92–3.82 (m,
8H), 2.66–1.93 (m, 12H); 13C NMR (75 MHz, CDCl3, d):
162.54, 157.26, 156.56, 153.51, 152.69, 148.42, 146.56,
146.10, 145.73, 145.16, 144.87, 144.57, 144.32, 144.12,
144.06, 143.87, 143.23, 142.68, 142.54, 141.12, 138.75,
138.12, 137.87, 128.32, 127.43, 64.21, 63.77, 45.53, 44.87,
20.15, 19.87; MALDI-TOF MS (M+1, C80H26): calcd,
986.20; found, 986.73; purity: 98.8% by HPLC. DM_
Compound 4 (4.17 g, 79.1%) was prepared from 2 via the
same procedure as that used to make 3. 1H NMR
(300 MHz, acetone-d6, d): 7.24 (s, 2H), 4.76 (s, 4H), 2.24
(s, 6H); 13C NMR (75 MHz, CDCl3, d): 137.33, 134.87,
132.42, 30.47, 19.23.
Synthesis of M_OXCMA, M_OXCBA, and M_OXCTA
Compound 3 (1.52 g, 5.52 mmol), KI (1.38 g, 8.34 mmol),
18-crown-6 (3.64 g, 13.8 mmol), and C60 (1.0 g, 1.39 mmol)
were dissolved in anhydrous 1,2-dichlorobenzene at
reflux for 48 h under an argon atmosphere and dark con-
ditions. Upon cooling to room temperature, the crude
product was precipitated by adding methanol. The result-
ing solid was purified using column chromatography on
silica gel with toluene/hexane as the eluent. This separa-
tion yielded M_OXCMA (130 mg, 11.2%), M_OXCBA
(269 mg, 20.3%), and M_OXCTA (261 mg, 17.5%).
M_OXCMA: 1H NMR (300 MHz, CDCl3, d): 7.57 (d,
1H), 7.49 (s, 1H), 7.37 (d, 1H), 4.76 (d, 2H), 4.40 (d, 2H),
2.49 (s, 3H); 13C NMR (75 MHz, CDCl3/CS2, d): 157.65,
157.37, 149.36, 149.17, 148.65, 148.26, 147.89, 147.55,
147.13, 146.56, 146.22, 145.86, 144.75, 144.52, 143.67,
143.29, 143.07, 142.92, 142.56, 142.31, 142.16, 141.72,
139.92, 137.53, 128.87, 127.56, 65.93, 46.53, 45.22, 21.32;
MALDI-TOF MS (M+1, C69H10): calcd, 838.08; found,
1
OXCTA: H NMR (300 MHz, CDCl3, d): 7.62–6.88 (m,
6H), 4.71–3.65 (m, 12H), 2.56–1.88 (m, 18H); 13C NMR
(75 MHz, CDCl3, d): 162.31, 155.46, 152.23, 150.93,
150.32, 149.40, 146.87, 146.23, 145.64, 145.33, 144.59,
144.23, 143.67, 143.20, 142.66, 142.36, 142.07, 141.71,
141.26, 140.92, 140.67, 140.15, 139.77, 139.16, 138.73,
138.55, 138.32, 136.12, 135.78, 134.96, 129.57, 128.35,
128.23, 127.51, 65.17, 67.57, 66.82, 65.77, 64.52, 64.22,
63.67, 63.18, 46.17, 45.82, 44.35, 43.87, 42.21, 42.07, 20.15,
20.01, 19.96; MALDI-TOF MS (M+1, C90H38): calcd,
1118.30; found, 1118.57; purity: 98.4% by HPLC.
1
838.21; purity: 99.2% by HPLC. M_OXCBA: H NMR
(300 MHz, CDCl3, d): 7.87–7.07 (m, 6H), 4.92–3.21 (m,
8H), 2.62–2.03 (m, 6H); 13C NMR (75 MHz, CDCl3, d):
157.52, 156.16, 151.59, 150.07, 147.55, 147.17, 146.89,
146.13, 146.06, 145.92, 145.71, 145.58, 145.42, 144.83,
144.35, 143.76, 143.58, 143.35, 143.14, 139.35, 138.66,
138.38, 137.22, 128.92, 128.43, 128.07, 66.84, 65.78, 65.06,
64.76, 64.12, 47.41, 46.27, 45.53, 22.54, 22.32, 21.87, 21.56;
MALDI-TOF MS (M+1, C78H22): calcd, 958.17; found,
4.3. Measurements
1H and 13C NMR spectra were recorded using a Varian
300 spectrometer. MALDI-TOF mass spectra were mea-
sured using a ZMS-DX303 mass spectrometer (JEOL
Ltd.), and TGA was performed on a TGA Q500 V6.7
Build 203 (TA Instruments) using a 2910 MDSC V4.4E
(TA Instruments) analyzer under a N2 atmosphere with
heating and cooling rates of 108 minÀ1. UVÀvisible ab-
sorption spectra were measured using a Model S-300
Scinco UVÀvis spectrophotometer, and photolumines-
cence (PL) spectra were recorded using a Shimadzu RF
5301 PC fluorometer. Cyclic voltammetry (CH Instru-
ments 600D) was performed using a 0.10 M solution of
tetrabutylammonium tetrafluoroborate in 1,2-dichloro-
benzene with the analyte present at a concentration of
10À3 M and employing a scan rate of 100 mVsÀ1 at room
temperature under an argon atmosphere. A Pt electrode
was used as the working electrode, and a Pt wire and Ag/
Ag+ electrode were used as the counter and reference
electrodes, respectively. Atomic force microscopy was
performed using a Digital Instruments NanoScope IV op-
erating in the tapping mode in air after annealing at 1508
for 10 min.
1
957.53; purity: 98.7% by HPLC. M_OXCTA: H NMR
(300 MHz, CDCl3, d): 7.62–6.93 (m, 9H), 4.75–3.16 (m,
12H), 2.43–1.98 (m, 9H); 13C NMR (75 MHz, CDCl3, d):
156.14, 155.57, 153.82, 152.65, 151.55, 149.13, 148.63,
147.53, 147.21, 146.64, 145.21, 144.46, 144.35, 143.62,
143.31, 143.02, 142.99, 142.67, 142.12, 142.03, 141.93,
141.35, 140.56, 140.17, 139.75, 139.33, 138.64, 138.57,
130.87, 129.81, 129.26, 128.50, 128.11, 67.46, 62.77, 62.06,
67.78, 66.76, 65.27, 64.21, 46.85, 46.55, 46.13, 45.56, 45.23,
43.83, 43.22, 42.61, 42.39, 21.85, 21.66, 21.35, 21.06;
MALDI-TOF MS (M+1, C87H32): calcd, 1076.25; found,
1075.73; purity: 98.1% by HPLC.
Synthesis of DM_OXCMA, DM_OXCBA, and DM_OXCTA
4,5-Dimethyl o-xylene-substituted fullerene derivatives,
i.e., DM_OXCMA (161 mg, 13.6%), DM_OXCBA
(302 mg, 22.1%), and DM_OXCTA (262 mg, 16.9%),
were prepared from compound 4 via the same procedure
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.ijc.wiley-vch.de
1025

Full Paper
4.4. Fabrication of Photovoltaic Devices
References
[1] a) B. C. Thompson, J. M. J. FrØchet, Angew. Chem. Int. Ed.
2008, 47, 58–77; b) C. J. Brabec, N. S. Sariciftci, J. C. Hum-
melen, Adv. Funct. Mater. 2011, 11, 15–26.
[2] a) Y.-J. Cheng, S.-H. Yang, C.-S. Hsu, Chem. Rev. 2009, 109,
5868–5923; b) T. D. Nielsen, C. Cruickshank, S. Foged, J.
Thorsen, F. C. Krebs, Sol. Energy Mater. Sol. Cells 2010, 94,
1553–1571.
[3] a) R. A. J. Janssen, J. C. Hummelen, K. Lee, K. Pakbaz,
N. S. Sariciftci, A. J. Heeger, F. Wudl, J. Chem. Phys. 1995,
103, 788–793; b) J. C. Hummelen, B. W. Knight, F. LePeq,
F. Wudl, J. Org. Chem. 1995, 60, 532–538.
[4] a) M. Granstrçm, K. Petritsch, A. C. Arias, A. Lux, M. R.
Andersson, R. H. Friend, Nature 1998, 395, 257–260; b) G.
Yu, J. Gao, J. C. Hummelen, F. Wudl, A. J. Heeger, Science
1995, 270, 1789–1791.
Mixed solutions of P3HT and fullerene derivatives were
prepared in 1,2-dichlorobenzene. The concentration was
kept within the range of 1.0–2.0 wt%. Polymer photovol-
taic devices were then fabricated employing the following
sandwich structure: ITO/PEDOT:PSS/P3HT:acceptor/
LiF/Al. The ITO-coated glass substrates were cleaned by
sonication in a detergent solution followed by successive
sonication in distilled water, acetone, and 2-propanol. A
40 nm thick layer of PEDOT:PSS (Clevious P VP
AI4083) was spin-coated onto the cleaned ITO substrate
after exposing the ITO surface to ozone for 10 min. This
PEDOT:PSS layer was heated on a hot plate at 1508 for
15 min, after which the composite solution was filtered
using a 0.45 mm syringe filter and spin-coated onto the
PEDOT:PSS layer at 900 rpm up to a thickness of
100 nm. Finally, a top electrode consisting of a 0.5 nm
thick layer of LiF and a 120 nm thick layer of Al was de-
posited onto the polymer active layer in a thermal evapo-
rator under a vacuum of 3”10À6 Torr. The current densi-
tyÀvoltage (JÀV) characteristics of all photovoltaic cells
were analyzed under simulated solar light (intensity:
100 mWcmÀ2; AM 1.5G) provided by an Oriel 300 W
solar simulator. The currentÀvoltage behavior was mea-
sured using a Keithley 2400, and all characterizations
were carried out in the ambient environment. The intensi-
ty of the simulated sunlight used was calibrated using
a standard Si photodiode detector (BS520, Bunkoh-keiki)
that had been calibrated at the National Renewable
Energy Laboratory (NREL). EQE values were measured
as a function of wavelength from 300 to 800 nm using
a device (PV Measurements, Inc.) equipped with a halo-
gen lamp light source, while calibration was performed
using the Si reference photodiode. The measurements
were carried out after masking all but the active cell
areas of the fabricated devices. All characterization steps
were carried out under ambient laboratory atmosphere.
The active areas of the solar cells were 9 mm2. The thick-
nesses of the thin films were measured using a KLA
Tencor Alpha-step IQ surface profilometer with an accu-
racy of Æ1 nm.
[5] a) Y. Zhang, H.-L. Yip, O. Acton, S. K. Hau, F. Huang,
A. K.-Y. Jen, Chem. Mater. 2009, 21, 2598–2600; b) Y. Li, Y.
Zou, Adv. Mater. 2008, 20, 2952–2958; c) J. Chen, Y. Cao,
Acc. Chem. Res. 2009, 42, 1709–1718.
[6] 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, 1480–1487; b) C. Yang, J. Y. Kim, S. Cho, J. K.
Lee, A. J. Heeger, F. Wudl, J. Am. Chem. Soc. 2008, 130,
6444–6450; c) D. Mi, J.-H. Kim, H. U. Kim, F. Xu, D.-H.
Hwang, J. Nanosci. Nanotechnol. 2014, 14, 1064–1084.
[7] a) H. U. Kim, J.-H. Kim, H. Kang, A. C. Grimsdale, B. J.
Kim, S. C. Yoon, D.-H. Hwang, ACS Appl. Mater. Interfaces
2014, 6, 20776–20785; b) P. A. Troshin, H. Hoppe, J. Renz,
M. Eggineger, J. Y. Mayorova, A. E. Goryachev, A. S. Pere-
gudov, R. N. Lyubovskaya, G. Gobsch, N. S. Sariciftci, V. F.
Razumov, Adv. Funct. Mater. 2009, 19, 779–788; c) J. A.
Mikroyannidis, A. N. Kabanakis, S. S. Sharma, G. D.
Sharma, Adv. Funct. Mater. 2011, 21, 746–755.
[8] a) M. C. Scharber, D. Mühlbacher, M. Koppe, P. Denk, C.
Waldauf, A. J. Heeger, C. J. Brabec, Adv. Mater. 2006, 18,
789–794.
[9] a) Y. He, H.-Y. Chen, J. Hou, Y. Li, J. Am. Chem. Soc. 2010,
132, 1377–1382; b) G. Zhao, Y. He, Y. Li, Adv. Mater. 2010,
22, 4355–4358; c) H. Kang, C.-H. Cho, H.-H. Cho, T. E.
Kang, H. J. Kim, K.-H. Kim, S. C. Yoon, B. J. Kim, ACS
Appl. Mater. Interfaces 2012, 4, 110–116.
[10] a) K.-H. Kim, H. Kang, S. Y. Nam, J. Jung, P. S. Kim, C.-H.
Cho, C. Lee, S. C. Yoon, B. J. Kim, Chem. Mater. 2011, 23,
5090–5095; b) K.-H. Kim, H. Kang, H. J. Kim, P. S. Kim,
S. C. Yoon, B. J. Kim, Chem. Mater. 2012, 24, 2373–2381;
c) X. Meng, Q. Xu, W. Zhang, Z. Tan, Y. Li, Z. Zhang, L.
Jiang, C. Shu, C. Wang, ACS Appl. Mater. Interfaces 2012, 4,
5966–5973.
[11] a) G. Ye, S. Chen, Z. Xiao, Q. Zuo, Q. Wei, L. Ding, J.
Mater. Chem. 2012, 22, 22374–22377; b) D. He, X. Du, Z.
Xiao, L. Ding, Org. Lett. 2014, 16, 612–615.
[12] a) H. U. Kim, D. Mi, J.-H. Kim, J. B. Park, S. C. Yoon, U. C.
Yoon, D.-H. Hwang, Sol. Energy Mater. Sol. Cells 2012, 105,
6–14.
[13] a) M. A. Ruderer, S. Guo, R. Meier, H.-Y. Chiang, V. Kçrst-
gens, J. Wiedersich, J. Perlich, S. V. Roth, P. Müller-Busch-
baum, Adv. Funct. Mater. 2011, 21, 3382–3391.
[14] a) C. Fan, S. Wang, J. W. Hong, G. C. Bazan, K. W. Plaxco,
A. J. Heeger, Proc. Natl. Acad. Sci. U. S.A. 2003, 100, 6297–
6301; b) H. U. Kim, J.-H. Kim, H. Suh, J. Kwak, D. Kim,
A. C. Grimsdale, S. C. Yoon, D.-H. Hwang, Chem.
Commun. 2013, 49, 10950–10952.
Acknowledgements
This work was supported by the New and Renewable
Energy Program of the Korea Institute of Energy Tech-
nology Evaluation and Planning (KETEP) funded by the
Korean Government Ministry of Trade, Industry &
Energy (No. 20113010010030), and National Research
Foundation (NRF) grants funded by the Korean Govern-
ment (NRF-2014R1A2A2A01007318 and No. 2011-
0030013 through GCRC SOP).
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.ijc.wiley-vch.de
1026

Full Paper
[15] a) A. M. Goodman, A. Rose, J. Appl. Phys. 1971, 42, 2823–
2830.
[16] a) V. D. Mihailetchi, H. Xie, B. de Boer, J. A. Koster,
P. W. M. Blom, Adv. Funct. Mater. 2006, 16, 699–708;
b) P. W. M. Blom, V. D. Mihailetchi, L. J. A. Koster, D. E.
Markov, Adv. Mater. 2007, 19, 1551–1566.
[19] a) C. J. Schaffer, C. M. Palumbiny, M. A. Niedermeier, C.
Jendrzejewski, G. Santoro, S. V. Roth, P. Müller-Buschbaum,
Adv. Mater. 2013, 25, 6760–6764; b) M. Pfaff, P. Müller, P.
Bockstaller, E. Müller, J. Subbiah, W. W. H. Wong, M. F. G.
Klein, A. Kiersnowski, S. R. Puniredd, W. Pisula, A. Cols-
mann, D. Gerthsen, D. J. Jones, ACS Appl. Mater. Interfaces
2013, 5, 11554–11562.
[17] a) C. Melzer, E. J. Koop, V. D. Mihailetchi, P. W. M. Blom,
Adv. Funct. Mater. 2004, 14, 865–870.
[18] a) G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K.
Emery, Y. Yang, Nat. Mater. 2005, 4, 864–868.
Received: December 26, 2014
Accepted: March 23, 2015
Published online: May 27, 2015
Isr. J. Chem. 2015, 55, 1017 – 1027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.ijc.wiley-vch.de
1027