Enhancing Performance of Large-Area Organic Solar Cells with Thick Film via Ternary Strategy

Abstract

Large-scale fabrication of organic solar cells requires an active layer with high thickness tolerability and the use of environment-friendly solvents. Thick films with high-performance can be achieved via a ternary strategy studied herein. The ternary system consists of one polymer donor, one small molecule donor, and one fullerene acceptor. The small molecule enhances the crystallinity and face-on orientation of the active layer, leading to improved thickness tolerability compared with that of a polymer-fullerene binary system. An active layer with 270 nm thickness exhibits an average power conversion efficiency (PCE) of 10.78%, while the PCE is less than 8% with such thick film for binary system. Furthermore, large-area devices are successfully fabricated using polyethylene terephthalate (PET)/Silver gride or indium tin oxide (ITO)-based transparent flexible substrates. The product shows a high PCE of 8.28% with an area of 1.25 cm(2) for a single cell and 5.18% for a 20 cm(2) module. This study demonstrates that ternary organic solar cells exhibit great potential for large-scale fabrication and future applications.

Full text

(1 of 8) 1700388
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.small-journal.com
Enhancing Performance of Large-Area Organic Solar
Cells with Thick Film via Ternary Strategy
Jianqi Zhang, Yifan Zhao, Jin Fang, Liu Yuan, Benzheng Xia, Guodong Wang,
Zaiyu Wang, Yajie Zhang, Wei Ma,* Wei Yan,* Wenming Su, and Zhixiang Wei*
Scholars ultimately aim to commercialize solution-process-
able organic solar cells (OSCs) through large-scale indus-
trial fabrication. The PCE of OSCs has rapidly increased
and surpassed the highest record of 12%.[1] Current pro-
cessing solvents and conditions must be considered for prac-
tical applications. Generally, industrial fabrication of OSCs
requires the following: fabrication of an active layer with
high thickness tolerability and use of environment-friendly
solvents.
DOI: 10.1002/smll.201700388
Organic Solar Cells
Large-scale fabrication of organic solar cells requires an active layer with high
thickness tolerability and the use of environment-friendly solvents. Thick films with
high-performance can be achieved via a ternary strategy studied herein. The ternary
system consists of one polymer donor, one small molecule donor, and one fullerene
acceptor. The small molecule enhances the crystallinity and face-on orientation of
the active layer, leading to improved thickness tolerability compared with that of a
polymer-fullerene binary system. An active layer with 270 nm thickness exhibits an
average power conversion efficiency (PCE) of 10.78%, while the PCE is less than
8% with such thick film for binary system. Furthermore, large-area devices are
successfully fabricated using polyethylene terephthalate (PET)/Silver gride or indium
tin oxide (ITO)-based transparent flexible substrates. The product shows a high PCE
of 8.28% with an area of 1.25 cm2 for a single cell and 5.18% for a 20 cm2 module.
This study demonstrates that ternary organic solar cells exhibit great potential for
large-scale fabrication and future applications.
Dr. J. Zhang, Y. Zhao, Dr. J. Fang, Dr. L. Yuan, B. Xia, G. Wang,
Dr. Y. Zhang, Prof. Z. Wei
CAS Key Laboratory of Nanosystem and Hierarchical Fabrication
CAS Center for Excellence in Nanoscience
National Center for Nanoscience and Technology
Beijing 100190, China
E-mail: weizx@nanoctr.cn
Z. Wang, Prof. W. Ma
State Key Laboratory for Mechanical Behavior of Materials
Xi’an Jiaotong University
Xi’an 710049, China
E-mail: msewma@mail.xjtu.edu.cn
Prof. W. Yan
Department of Environmental
Science and Engineering
Xi’an Jiaotong University
Xi’an 710049, China
E-mail: yanwei@mail.xjtu.edu.cn
Prof. W. Su
Division of Printed Electronics
Suzhou Institute of Nano-Tech and Nano-Bionics (SINANO)
Chinese Academy of Sciences
Suzhou 215123, China
Fabrication of uniform films with 100 nm thickness is dif-
ficult using large-area printing techniques. An active layer
with high thickness tolerability must be achieved for large-
area fabrication. Researchers have developed donor mate-
rials with high hole carrier mobility or high crystallinity to
obtain high-performance OSCs with thick active layers.[2,3]
Materials with face-on orientation could achieve high PCE
with thick films because the face-on orientation facilitates
the out-of-plane charge transport.[4,5] In addition to synthesis
small 2017, 1700388
www.advancedsciencenews.com

communications
1700388 (2 of 8) www.small-journal.com © 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
new materials, ternary blend strategy can realize the required
morphology. Ternary systems consisting of one donor pol-
ymer, one donor small molecules with high crystallinity, and
one acceptor exhibit enhanced crystallinity and face-on ori-
entation.[6] The use of these systems may provide a promising
route for fabricating high-performance OSCs with thick films
and large-area OSCs through printing techniques. Further-
more, various studies have demonstrated that the perfor-
mance of OSCs fabricated via ternary strategy using a donor
or an acceptor material as the third component is better
than their binary counterparts.[7] Compared with adding an
acceptor, incorporation of a donor would achieve high-per-
formance system due to complementary absorption, improve-
ment of morphology, and increase charge mobility.[7–10]
On the other hand, the use of environment-friendly sol-
vents is crucial for fabrication of large-area OSCs. So far
chlorinated solvents, such as chlorobenzene (CB), o-dichlo-
robenzene (DCB), or chloroform (CF), are widely used to
dissolve photoactive materials, including solution-process-
able polymers and small-molecule OSCs.[11] Compared with
chlorine-free solvents, chlorinated solvents exhibit bicon-
tinuous morphology because of their improved solubility
and viscosity in solutions. However, chlorinated solvents are
unsuitable for large-scale manufacturing of OSCs because of
their toxicity, which causes environmental issues and health
problems for industrial workers. Therefore, scholars have
focused on developing solvents to replace chlorinated sol-
vents. Few works show that the PCE of a device processed
with chlorine-free solvent is similar to that of its counter-
part processed with chlorinated ones.[12–27] Vogt and Li and
co-workers used binary chlorine-free solvent mixtures to
mimic the Hansen solubility parameters of chlorine ana-
logs.[15,28] They found that P3HT:[6,6]-phenyl C71 butyric
acid methyl ester (PC71BM) blends exhibited a PCE of 3.8%
after the addition of 20 vol% benzaldehyde to p-xylene.[28]
Hou and co-workers reported that polymer solar cells (PSC)
can be successfully fabricated using o-xylene as solvent and
nine types of polymers, of which, PBDT-TS1:PC71BM-based
PSC achieved a PCE of 9.47%.[29] Yan and co-workers used
1,2,4-trimethylbenzene as processing solvent to prepare
PffBT4T-C9C13:PC71BM blends, which achieved a PCE of
11.7%.[30]
In this study, a ternary inverted OSC is designed using
o-xylene as processing solvent. Poly[4,8-bis(5-(2-ethylhexyl)
thiophen-2-yl)benzo[1,2-b:4,5-b′]dithiophene-co-3-fluoro-
thieno [3,4-b]thiophene-2-carboxylate] (PTB7-Th) and
7,7-(4,4-bis(2-ethylhexyl)-4H-silolo[3,2-b:4,5-b′]dithiophene-
2,6-diyl)bis(6-fluoro-4-(5′-hexyl-[2,2′-bithiophen]-5-yl)-
benzo[c][1,2,5] thiadiazole) (p-DTS(FBTTH2)2) are used as
donors. PC71BM is used as acceptor. The chemical structures
of PTB7-Th and p-DTS(FBTTH2)2 are shown in Figure 1a.
The fabricated ternary system with CB as processing solvent
can achieve a PCE of 10.5%.[6] o-Xylene is selected because
of its similar boiling point and viscosity to those of CB
(Table S1, Supporting Information). The ternary system with
15% weight ratio of p-DTS(FBTTH2)2 achieves an average
PCE of 10.78% under the optimized conditions. The ternary
system shows high performance with thickness of the active
layer ranging from 200 to 270 nm; hence, the system exhibits
potential for large-area printing techniques. The main reason
for obtaining high efficiency with thick films is that adding
small molecules into the ternary system enhances domain
small 2017, 1700388
www.advancedsciencenews.com
Figure 1. a) The chemical structures of PTB7-Th, p-DTS(FBTTH2)2, and PC71BM. b) Schematics of the binary system and ternary system.

(3 of 8) 1700388
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.small-journal.com
purity, face-on orientation, and increases crystalline domain
size as illustrated schematically in Figure 1b. The optimized
morphology facilitates charge transport and collection. Slot-
die coating is used to fabricate large-area flexible OSCs. A
PCE of 5.18% is obtained with an area of 20 cm2 on flexible
PET/ITO substrate, and a PCE of 8.28% is obtained with an
area of 1.25 cm2 on flexible PET/Ag-grid substrate.
o-Xylene is selected as chlorine-free processing solvent
for the present ternary system because of its comparable
properties to the halogenated analog (CB); such properties
include high boiling point, adequate viscosity, and solubility
for conjugated polymer/small molecules and fullerene deriva-
tives (Table S1, Supporting Information). The overall donor
(PTB7-Th and p-DTS(FBTTH2)2)-to-PC71BM ratio is main-
tained at 1:2. The inverted architecture type of OSCs with of
ITO/ZnO (40 nm)/active layer/MoOx (5 nm)/Ag (100 nm)
were fabricated due to their better stability than that of con-
ventional device architecture. The active area of the OSCs is
4 mm2, and their photovoltaic parameters are summarized in
Table 1.
Figure 2a illustrates the current density–voltage curve
of the optimized binary and ternary OSCs under AM 1.5G
illumination at 100 mW cm−2. The cells with 15% weight
ratio of p-DTS(FBTTH2)2 show the highest PCE, which is
mainly attributed to the increased fill factor (FF) compared
with the reference device PTB7-Th:PC71BM. The optimized
devices show an average PCE of 10.78%, with Voc = 0.737 V,
Jsc = 21.67 mA cm−2, and FF = 67.50%. The reference devices
of the binary system exhibit an average PCE of 9.16%, with
Voc = 0.797 V, Jsc = 20.7 mA cm−2, and FF = 55.5%, similar
to those of devices fabricated with CB.[6] The EQE spectra
(Figure 2b) show a “square-like” behavior with a plateau of
around 80% for all ternary systems. This finding indicates that
sunlight is used very efficiently at different wavelengths from
380 to 780 nm, resulting in very high Jsc. The calculated Jsc
values are given in Table 1, and the error between the values
obtained by sun simulator and EQE calculation is within 5%.
Variations in PCE with active layer thickness are also
investigated. Uniform films with 100 nm thickness are
difficult to fabricate in large scale by using industrial solution
small 2017, 1700388
www.advancedsciencenews.com
Table 1. Summary of photovoltaic parameters of solar cells with different weight ratios of p-DTS(FBTTH2)2 under AM 1.5G illumination at
100 mW cm−2.
PTB7-Th:p-DTS(FBTTH2)2:PC71BM Voc (ave) [V] Jsc (ave) [mA cm−2]Jcal [mA cm−2] FF (ave) [%] PCE (ave) [%]
1:0:2 0.797 ± 0.006 20.70 ± 0.35 19.88 55.50 ± 1.44 9.16 ± 0.13
0.95:0.05:2 0.771 ± 0.007 21.45 ± 0.34 20.42 58.22 ± 0.73 9.63 ± 0.22
0.90:0.10:2 0.753 ± 0.007 21.47 ± 0.49 20.43 62.92 ± 1.20 10.17 ± 0.20
0.85:0.15:2 0.737 ± 0.009 21.67 ± 0.70 20.75 67.50 ± 2.06 10.78 ± 0.10
0:1:2 0.677 ± 0.001 11.96 ± 0.38 12.01 47.22 ± 0.61 3.82 ± 0.15
Figure 2. a) J–V curves of binary and ternary systems under AM 1.5G illumination from a calibrated solar simulator with an irradiation intensity of
100 mW cm−2. b) EQE curves of the OSCs corresponding to the devices in (a). c) PCE of the reference device and ternary device and d) variations
in Jsc and FF as a function of the thickness of active layer.

communications
1700388 (4 of 8) www.small-journal.com © 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
casting techniques. Devices with high PCE and thick films
are important for large-scale industrial fabrication of OSCs.
Interestingly, ternary devices with 15% p-DTS(FBTTH2)2
exhibit a PCE of ≈10.7% with the thickness of 200–270 nm
(Figure 2c). By contrast, the PCE of the reference cell signifi-
cantly decreases when the thickness of the active layer
is larger than 200 nm. For the ternary blend, the absorp-
tion of the incident photos can be significantly improved in
such a thick film as demonstrated by the increased Jsc value
(Figure 2d). The Jsc increases by 20% as the film thickness
increases from 135 to 270 nm and maintains an extremely
high value of ≈21 mA cm−2 until 400 nm. Typically, the per-
formance of OSCs degrades with decreasing FF as the
thickness of the active layer increases. This rule is followed
by the binary reference cell (Figure 2d). However, in the
present study, FF decreases by 6% only as the film thickness
increases from 135 to 270 nm (Figure 2d). The FF signifi-
cantly decreases when the film thickness exceeds 300 nm.
To study the effect of p-DTS(FBTTH2)2 on charge dis-
sociation in ternary systems, exciton dissociation probabili-
ties are determined based on the relation of photocurrent
density (Jph) versus effective voltage (Veff). Figure 3a pre-
sents the dependence of Jph on Veff for the reference cell and
the ternary devices. Veff is defined as Veff = V0 − Va, where
V0 is the voltage at Jph = 0 and Va is the applied voltage.[31]
Under short-circuit condition, the ratios of Jph/Jph,sat for ter-
nary systems are 93.5% (0% p-DTS(FBTTH2)2), 94.5%
(5% p-DTS(FBTTH2)2), 95.8% (10% p-DTS(FBTTH2)2),
and 97.1% (15% p-DTS(FBTTH2)2), respectively. In the
maximal power output circumstances (Veff = 0.2 V), the
ratios of Jph/Jph,sat for the abovementioned cells are 68.1%,
75.6%, 76.4%, and 80.8%, respectively. This trend could be
attributed to the low bimolecular recombination after adding
p-DTS(FBTTH2)2 to PTB7-Th.
Variations in Jph with illumination intensity are investi-
gated to determine the influence of p-DTS(FBTTH2)2 on
recombination and charge transport processes. The Jph shows
a power-law dependence on illumination intensity for organic
solar cells, that is, Jph = I
α
, where I is the illumination inten-
sity and
α
is the exponential factor.[32] The deviation of
α
from the unity implies space charge buildup and subsequent
bimolecular recombination. Compared with Jph obtained
at high Veff, the Jph obtained at low Veff is more suitable for
studying the recombination behavior.[33] Figure 3b displays
the dependence of photocurrent on the illumination intensity
on a log-log scale at Veff = 0.1 V. The fitted
α
for the reference
cell is 0.894, whereas the
α
values are 0.907, 0.913, and 0.928
for adding 5%, 10%, and 15% p-DTS(FBTTH2)2, respec-
tively. These results indicate that bimolecular recombination
can be effectively suppressed by adding p-DTS(FBTTH2)2.
Thus, the ternary systems show high FF.
Figure 3c,d shows the ratio of Jph/Jph,sat and
α
as a func-
tion of film thickness. For the reference cell, Jph/Jph,sat and
α
significantly decrease, indicating less efficient exciton disso-
ciation and high bimolecular recombination. For the ternary
solar cell, Jph/Jph,sat is almost constant at ≈97% when the film
thickness is less than 300 nm; this finding indicates very effi-
cient exciton dissociation. The
α
value gradually decreases
from 0.95 to 0.93 as the film thickness increases from 130 to
300 nm. Therefore, bimolecular recombination can be effec-
tively suppressed at different film thicknesses and the ternary
solar cells exhibit high Jsc and FF.
The superior properties of OSCs are related to the
morphology of the active layer. Resonant soft X-ray
small 2017, 1700388
www.advancedsciencenews.com
Figure 3. a) Photocurrent density versus effective voltage curves. b) Photocurrent density at Veff = 0.1 V versus light intensity for the ternary systems.
c) Variation in Jph/Jsat and d)
α
as functions of the thickness of active layer.

(5 of 8) 1700388
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.small-journal.com
scattering (R-SoXS), transmission electron microscopy
(TEM), and grazing incidence wide-angle X-ray scattering
(GIWAXS) analyses are performed to determine the effect
of p-DTS(FBTTH2)2 on the nanoscale morphology and
molecular structures in the ternary system. R-SoXS is used to
probe the characteristic length scales of the morphology and
domain purity. Figure 4a shows the scattering profiles for the
reference cell and the ternary systems at photon energy of
284.2 eV. The selected photon energy ensures that the scat-
tering originates from the contrast between the materials
(polymer/small molecules and fullerene), not the mass–thick-
ness contrast. The reference cell (PTB7-Th:PC71BM) shows
a scattering peak at q = 0.16 nm−1, which corresponds to a
characteristic length scale of ≈39 nm. The active layer made
by p-DTS(FBTTH2)2:PC71BM has larger domain spacing
because of its low q value (0.06 nm−1). Notably, the scat-
tering peaks for the ternary systems are also located at
q = 0.16 nm−1, indicating that p-DTS(FBTTH2)2 has almost
no influence on the domain size of phase separation in ter-
nary systems. This result is consistent with the TEM results
shown in Figure S1 (Supporting Information); the images
reveal that blending p-DTS(FBTTH2)2 with up to 15%
PTB7-Th:PC71BM does not significantly change the phase
separation. The relative domain purity can be determined
by integrating the R-SoXS profiles. Scattering originates
from the contrast between polymer/small molecules and
fullerene; thus, the relative purity represents the mixture
degree of fullerene in the polymer/small molecule phases,
not the mixture of polymer and small molecule because of
the low contrast. Our previous study indicated that PTB7-Th
and p-DTS(FBTTH2)2 can form alloys;[6] the same results are
obtained in the present study (GIWAXS results). Moreover,
0%, 5%, 10%, and 15% p-DTS(FBTTH2)2 possess average
relative domain purities of 0.64, 0.75, 0.88, and 1, respec-
tively. The addition of p-DTS(FBTTH2)2 facilitates the phase
separation to produce pure domains. More importantly, the
trend of the relative domain purity on p-DTS(FBTTH2)2 is
similar to that of FF on p-DTS(FBTTH2)2 (Figure 4b). High
FF would result from more efficient charge transport and
reduced bimolecular charge recombination caused by the
pure domains.
GIWAXS analysis is employed to probe crystallo-
graphic changes in the donor molecules and PC71BM
agglomeration in the blended films. Figure S2 (Supporting
Information) shows the 2D GIWAXS patterns and the cor-
responding out-of-plane and in-plane cuts. Compared with
pure p-DTS(FBTTH2)2 and the ternary system with 20%
p-DTS(FBTTH2)2 showing macrophase separation behavior
(Figure S3, Supporting Information), the (100), (200), and
(300) diffraction peaks disappear in the out-of-plane direction
for the ternary system with less than 15% p-DTS(FBTTH2)2.
Therefore, p-DTS(FBTTH2)2 does not form an individual
phase in the active layer when its weight ratios are less than
15%. Similarly, our previous data showed that an alloy was
formed in the ternary system.[6] The broad peaks located
near 1.35 and 1.75 Å−1 correspond to PC71BM aggregates
and
π
–
π
stacking, respectively. The correlation length reflects
nanocrystallite size and can be determined using the full-
width at half-maximum (FWHMs) of the scattering peak
by the Scherrer equation.[34] The FWHMs of the PC71BM
aggregates and
π
–
π
stacking are obtained by fitting the out-
of-plane profiles with the Gaussian equation (Figure S4,
Supporting Information). Figure 4d shows the variations in
PC71BM and
π
–
π
correlation length with p-DTS(FBTTH2)2
weight ratio. The variations in both parameters exhibit a
monotonic trend, indicating increased crystallite sizes caused
small 2017, 1700388
www.advancedsciencenews.com
Figure 4. a) R-SoXS profiles in log scale for binary and ternary systems. b) FF and domain purity, c) charge mobility, and d) correlation length of
PCBM and
π
–
π
stacking as a function of weight ratios of p-DTS(FBTTH2)2.

communications
1700388 (6 of 8) www.small-journal.com © 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
by the formation of pure domains, as demonstrated by the
R-SoXS results. The pure domains and increased correlation
length facilitate the charge carrier transport, leading to high
charge carrier mobility (Figure 4c); this phenomenon may
assist in reducing monomolecular and bimolecular recombi-
nation losses to produce high FF for the ternary system.
The ternary system with 15% p-DTS(FBTTH2)2 is
selected to fabricate large-area flexible devices. A device
was fabricated by spin-coating with a flexible PET substrate
with the embedded silver grid. The sketch, the photo of the
PET/silver-grid substrate, the scanning electron microscopy
(SEM) image of the substrate, and the fabricated OSC
device are presented in Figure S5 (Supporting Informa-
tion). The device with an active layer of 1.25 cm2 achieves
a PCE of 8.28%, with Voc of 0.728 V, Jsc of 18.38 mA cm−2,
and FF of 61.88%. To the best of our knowledge, this PCE is
the highest value in large-area solar cells with flexible sub-
strates. The devices are fabricated by spin-coating because
of substrate limitations. In our future work, a new substrate
will be designed for slot-die processing. Moreover, the PET/
silver-grid substrate exhibits potential for fabrication of
large-area OSCs. Furthermore, large-area printed mod-
ules can be developed using commercially available PET/
ITO substrate by carefully designing the slot-die procedure.
The photovoltaic parameters are shown in Table 2, and the
module cell is shown in Figure 5. The one, two, and four
modules, with active areas of 5, 10, and 20 cm2, show PCE of
5.75%, 5.82%, and 5.18%, respectively. Voc is nearly n (n is
the number of modules) times larger than that of the single
cell, indicating that the subcells are perfectly interconnected.
Given this high Voc, the module of OSC is used to power a
series of LED lamps (Figure 5d) and power a fan (Video S1,
Supporting Information). The results confirm the potential
application of flexible solar cells as power source of elec-
tronic devices.
To check the recombination and ideal factor of the
module, the variations of Jsc and Voc with light intensity and
dark current were measured as shown in Figure S6 (Sup-
porting Information). It is seen that the Jsc versus light inten-
sity dependence is linear with a slope of 0.938 suggesting only
a small contribution of the bimolecular recombination. Theo-
retically, the slope of the Voc dependence should be nkT/q,
where n is 1 for bimolecular recombination and 2 in the case
of monomolecular Shockley–Read-Hall (SRH) recombina-
tion.[35] However, the slope should be double if we assume
the subcells work perfectly the same for the module series
connected by two subcells. The slope of the Voc versus light
intensity is 2.13 kT/q (Figure S6b, Supporting Information),
which means the bimolecular recombination dominates at the
open-circuit condition. From the dark current characteristics,
the ideal factor is 1.95 which is larger than unity, it would be a
sign of presence of electrically active traps.[36]
In summary, high-performance ternary OSCs with high
thickness tolerability were fabricated using o-xylene as pro-
cessing solvent. The ternary system with 15% weight ratio
p-DTS(FBTTH2)2 achieves an average PCE of 10.78% under
the optimized conditions. The ternary system shows high-per-
formance and possesses active layer with thickness ranging
from 200 to 270 nm; hence, the system exhibits potential for
large-area fabrication. The crystallinity and face-on orienta-
tion of the ternary system are enhanced by incorporating
highly crystalline small molecules, resulting in high hole
mobility and less charge recombination. Thus, Jsc, FF, and
PCE increase. Moreover, PET/Ag-grid as substrate exhibits
a PCE of 8.28% with an active layer of 1.25 cm2. To the best
of our knowledge, this PCE is the highest among large-area
solar cells with flexible substrates. Slot-die is used to fabricate
large-area flexible OSCs, with PCE of 5.18% and an active
area of 20 cm2. In the series connection of four cells, the
module demonstrates the ability to power fan and a series
of LED lamps and thus can be potential materials for future
applications.
Experimental Section
Photovoltaic Device Fabrication: Small-Area Device Processing:
PTB7-Th and p-DTS(FBTTH2)2 were purchased from 1-material
Chemscitech Inc. (Canada) and used as received. Patterned ITO
glass with sheet resistance of 15 Ω sq−1 was purchased from CSG
HOLDING Co., Ltd. (China). The ITO glass was cleaned by sequen-
tial sonication in soap with deionized (DI) water, acetone, and
isopropanol for 15 min at each step. After ultraviolet–ozone (Ultra-
violet Ozone Cleaner, Jelight Company, USA) treatment for 4 min,
a ZnO electron transport layer was prepared by spin-coating at
3000 rpm. The ZnO nanoparticles were synthesized following the
reported protocol.[37] Active layer solutions (D/A ratio 1:2) with
polymer concentrations of 8.5 mg mL−1 were prepared in o-xylene
with 1% (volume fraction) of 1,8-diiodooctane (DIO). Active layers
were spin-coated from the polymer solution on the substrate in an
N2 glove box. At a vacuum level of ≈1.0 × 10−6 mbar, a thin layer
(5 nm) of MoOx was deposited as the anode interlayer and Ag was
deposited as the top electrode at 100 nm.
Large-Area and Module Device Processing: The inverted solar
cell devices were processed under ambient conditions. The bottom
electrode was ITO coated on PET substrates with a striped pattern
(10 Ω sq−1, T ≥ 80%), which were purchased from South China
Xiang Science & Technology Co., Ltd. The R2R instrument (Mini
Roll Coater, FOM Technologies) was used to fabricate the large-
area flexible solar cells. Modules with two and four serial stripes
small 2017, 1700388
www.advancedsciencenews.com
Table 2. Summary of photovoltaic parameters of solar cells fabricated with flexible substrate under AM 1.5G illumination at 100 mW cm−2.
Module Substrate Area [cm2]Voc [V] Jsc [mA cm−2] FF [%] PCE [%]
Single PET/Ag-grid 1.25 0.728 18.38 61.88 8.28
Single PET/ITO 5 0.704 16.01 51.06 5.75
2 PET/ITO 10 1.43 7.65 53.25 5.82
4 PET/ITO 20 2.71 3.89 49.19 5.18

(7 of 8) 1700388
© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.small-journal.com
small 2017, 1700388
www.advancedsciencenews.com
were fabricated. After multiple optimization procedures, the ZnO
NPs dispersed in isopropanol were selected to make the elec-
tronic transport layer (ETL) and the concentration was maintained
at 15 mg mL−1. The donors and acceptor blends (85:15:200, w/w)
were dissolved in o-xylene as 15 mg mL−1; 3% (v/v) DIO was added
as additive solvent. The slot-die was carried out at room tempera-
ture under the ambient condition. During the fabrication, ZnO solu-
tion was injected by an injection pump with a speed of 6 mL h−1,
and the rolling speed of the mini-roller setup was 60 m h−1. The
active layer was dispensed at 3 mL h−1 and coated at 40 m h−1.
After the fabrication, the sample was placed into a vacuum evapo-
ration system to deposit 10 nm MoOx and 200 nm Ag. The active
areas of all flexible polymer solar cells modules were 10 and
20 cm2, respectively.
PET/silver-grid substrates were provided by a collaborator.
PET/Ag-grid/PH1000 hybrid electrodes were fabricated to improve
the surface flatness and wettability of the PET/Ag-grid substrates.
The high conductivity PEDOT:PSS (Clevios PH1000, Heraeus, Ger-
many) was spun-coated on the PET/Ag-grid substrates at 800 rpm
and baked at 120 °C for 15 min in ambient air. The OSC fabrication
process was the same for the reference cells made on the glass/
ITO substrate.
Performance Measurement of OSCs: The J–V characteristics of
the device were assessed under AM 1.5G (100 mW cm−1) with a
Newport Thermal Oriel 91159A solar simulator. Light intensity was
calibrated with a Newport oriel PN 91150V Si-based solar cell. J–V
characteristics were recorded with a Keithley 2400 source meter
unit. The masked and unmasked tests had consistent results
with relative errors within 2%. EQEs were performed in air with an
Oriel Newport system (Model 66902) equipped with a standard Si
diode. Monochromatic light was generated from a Newport 300 W
lamp source.
Methods: The thicknesses of the active layer were measured
with an Alpha-atep D-120 stylus profilometer (Kla-Tencor). GIWAXS
analysis was conducted at the beamline 7.3.3 of the Advanced
Light Source (ALS),[38] Berkeley, USA. Samples were prepared on
Si substrates with a thin ZnO layer. The incident angle was set as
0.14°, which maximized the scattering intensity from the samples.
The scattered X-rays were detected by a Dectris Pilatus 2M photon
counting detector. R-SoXS transmission measurements were per-
formed at the beamline 11.0.1.2 of the ALS.[39] Samples for R-SoXS
measurements were prepared on the PEDOT:PSS modified glass/
ITO substrate and transferred by floating in water to a 100 nm thick
Si3N4 membrane (1.5 mm × 1.5 mm) supported by a 200 mm thick
Figure 5. a) Slot-die setup, b) fabricated OSC module, c) schematic of the OSC module, d) powered LED lamps, and e) J–V curve of the large-area
OSCs.

communications
1700388 (8 of 8) www.small-journal.com © 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Si frame (5 mm × 5 mm) (Norcada Inc.). 2D scattering patterns
were collected by an in vacuum CCD camera (Princeton Instrument
PI-MTE). The beam size at the sample was 100 mm × 200 mm.
Supporting Information
Supporting Information is available from the Wiley Online Library
or from the author.
Acknowledgements
J.Z. and Y.Z. contributed equally to this work. The authors acknowl-
edge the financial support from the Ministry of Science and Tech-
nology (Grant No. 2016YFA0200700), the National Natural Science
Foundation of China (Grant Nos. 21125420, 21604017, 21504066,
and 21534003), and the “Strategic Priority Research Program” of
the Chinese Academy of Sciences (Grant No. XDA0909040200).
X-ray data were acquired at beamlines 7.3.3 at the Advanced Light
Source, which was supported by the Director, Office of Science,
Office of Basic Energy Sciences, of the U.S. Department of Energy
under Contract No. DE-AC02-05CH11231. The authors thank
Chenhui Zhu at beamline 7.3.3, and Cheng Wang at beamline
11.0.1.2 for assistance with data acquisition.
Conflict of Interest
The authors declare no conflict of interest.
[1] S. Li, L. Ye, W. Zhao, S. Zhang, S. Mukherjee, H. Ade, J. Hou, Adv.
Mater. 2016, 28, 9423.
[2] T. L. Nguyen, H. Choi, S. J. Ko, M. A. Uddin, B. Walker, S. Yum,
J. E. Jeong, M. H. Yun, T. J. Shin, S. Hwang, J. Y. Kim, H. Y. Woo,
Energy Environ. Sci. 2014, 7, 3040.
[3] W. Li, K. H. Hendriks, W. S. Roelofs, Y. Kim, M. M. Wienk,
R. A. Janssen, Adv. Mater. 2013, 25, 3182.
[4] I. Osaka, T. Kakara, N. Takemura, T. Koganezawa, K. Takimiya,
J. Am. Chem. Soc. 2013, 135, 8834.
[5] X. W. Zhu, J. Fang, K. Lu, J. Q. Zhang, L. Y. Zhu, Y. F. Zhao,
Z. G. Shuai, Z. X. Wei, Chem. Mater. 2014, 26, 6947.
[6] J. Zhang, Y. Zhang, J. Fang, K. Lu, Z. Wang, W. Ma, Z. Wei, J. Am.
Chem. Soc. 2015, 137, 8176.
[7] L. Y. Lu, M. A. Kelly, W. You, L. P. Yu, Nat. Photonics 2015, 9, 491.
[8] Q. S. An, F. J. Zhang, J. Zhang, W. H. Tang, Z. B. Deng, B. Hu,
Energy Environ. Sci. 2016, 9, 281.
[9] P. Cheng, Y. F. Li, X. W. Zhan, Energy Environ. Sci. 2014, 7, 2005.
[10] F. Bonaccorso, N. Balis, M. M. Stylianakis, M. Savarese, C. Adamo,
M. Gemmi, V. Pellegrini, E. Stratakis, E. Kymakis, Adv. Funct.
Mater. 2015, 25, 3870.
[11] L. Lu, T. Zheng, Q. Wu, A. M. Schneider, D. Zhao, L. Yu, Chem. Rev.
2015, 115, 12666.
[12] J. H. Chang, H. F. Wang, W. C. Lin, K. M. Chiang, K. C. Chen,
W. C. Huang, Z. Y. Huang, H. F. Meng, R. M. Ho, H. W. Lin, J. Mater.
Chem. A 2014, 2, 13398.
[13] X. Guo, M. Zhang, C. Cui, J. Hou, Y. Li, ACS Appl. Mater. Interfaces
2014, 6, 8190.
[14] Z. Zhang, X. J. Zhang, J. C. Zhang, X. Gong, Y. H. Liu, H. Lu, C. H. Li,
Z. S. Bo, RSC Adv. 2016, 6, 39074.
[15] C. D. Park, T. A. Fleetham, J. Li, B. D. Vogt, Org. Electron. 2011, 12,
1465.
[16] C. Sprau, F. Buss, M. Wagner, D. Landerer, M. Koppitz, A. Schulz,
D. Bahro, W. Schabel, P. Scharfer, A. Colsmann, Energy Environ.
Sci. 2015, 8, 2744.
[17] L. G. Xiao, C. Liu, K. Gao, Y. J. Yan, J. B. Peng, Y. Cao, X. B. Peng,
RSC Adv. 2015, 5, 92312.
[18] Y. Chen, S. Zhang, Y. Wu, J. Hou, Adv. Mater. 2014, 26,
2744.
[19] Y. Chen, Y. Cui, S. Q. Zhang, J. H. Hou, Polym. Chem. 2015, 6,
4089.
[20] C. C. Chueh, K. Yao, H. L. Yip, C. Y. Chang, Y. X. Xu, K. S. Chen,
C. Z. Li, P. Liu, F. Huang, Y. W. Chen, W. C. Chenb, A. K. Y. Jen,
Energy Environ. Sci. 2013, 6, 3241.
[21] J. Griffin, A. J. Pearson, N. W. Scarratt, T. Wang, A. D. F. Dunbar,
H. Yi, A. Iraqi, A. R. Buckley, D. G. Lidzey, Org. Electron. 2015, 21,
216.
[22] M. V. Srinivasan, N. Tsuda, P. K. Shin, S. Ochiai, RSC Adv. 2015, 5,
56262.
[23] M. Bin, Y. Y. Fu, Z. Y. Xie, J. Liu, L. X. Wang, Polym. Chem. 2015, 6,
805.
[24] J. J. van Franeker, M. Turbiez, W. Li, M. M. Wienk, R. A. Janssen,
Nat. Commun. 2015, 6, 6229.
[25] Y. Zhang, J. Y. Zou, C. C. Cheuh, H. L. Yip, A. K. Y. Jen, Macromol-
ecules 2012, 45, 5427.
[26] M. E. Farahat, C. S. Tsao, Y. C. Huang, S. H. Chang,
W. Budiawan, C. G. Wu, C. W. Chu, J. Mater. Chem. A 2016, 4,
7341.
[27] C. H. Duan, W. Z. Cai, B. B. Y. Hsu, C. M. Zhong, K. Zhang,
C. C. Liu, Z. C. Hu, F. Huang, G. C. Bazan, A. J. Heeger, Y. Cao,
Energy Environ. Sci. 2013, 6, 3022.
[28] C. D. Park, T. Fleetham, J. Li, Org. Electron. 2015, 16, 95.
[29] W. C. Zhao, L. Ye, S. Q. Zhang, M. L. Sun, J. H. Hou, J. Mater. Chem.
A 2015, 3, 12723.
[30] J. Zhao, Y. Li, G. Yang, K. Jiang, H. Lin, H. Ade, W. Ma, H. Yan, Nat.
Energy 2016, 1, 15027.
[31] V. D. Mihailetchi, J. Wildeman, P. W. Blom, Phys. Rev. Lett. 2005,
94, 126602.
[32] A. K. Kyaw, D. H. Wang, V. Gupta, W. L. Leong, L. Ke, G. C. Bazan,
A. J. Heeger, ACS Nano 2013, 7, 4569.
[33] Z. He, C. Zhong, X. Huang, W. Y. Wong, H. Wu, L. Chen, S. Su,
Y. Cao, Adv. Mater. 2011, 23, 4636.
[34] D. M. Smilgies, J. Appl. Crystallogr. 2009, 42, 1030.
[35] V. Gupta, A. K. K. Kyaw, D. H. Wang, S. Chand, G. C. Bazan,
A. J. Heeger, Sci. Rep. 2013, 3, 1965.
[36] V. V. Brus, C. M. Proctor, N. A. Ran, T.-Q. Nguyen, Adv. Energy
Mater. 2016, 6, 1502250.
[37] S. B. Dkhil, D. Duché, M. Gaceur, A. K. Thakur, F. B. Aboura,
L. Escoubas, J.-J. Simon, A. Guerrero, J. Bisquert, G. Garcia-
Belmonte, Q. Bao, M. Fahlman, C. Videlot-Ackermann, O. Margeat,
J. Ackermann, Adv. Energy Mater. 2014, 4, 1400805.
[38] A. Hexemer, W. Bras, J. Glossinger, E. Schaible, E. Gann, R. Kirian,
A. MacDowell, M. Church, B. Rude, H. Padmore, J. Phys. Conf. Ser.
2010, 247, 012007.
[39] E. Gann, A. T. Young, B. A. Collins, H. Yan, J. Nasiatka,
H. A. Padmore, H. Ade, A. Hexemer, C. Wang, Rev. Sci. Instrum.
2012, 83, 045110.
Received: February 3, 2017
Revised: February 26, 2017
Published online:
small 2017, 1700388
www.advancedsciencenews.com