Content uploaded by Jingde Chen
Author content
All content in this area was uploaded by Jingde Chen on Aug 28, 2018
Content may be subject to copyright.
Content uploaded by Fuwen Zhao
Author content
All content in this area was uploaded by Fuwen Zhao on Jun 19, 2018
Content may be subject to copyright.
CommuniCation
1706363 (1 of 8) © 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.advmat.de
Balanced Partnership between Donor and Acceptor
Components in Nonfullerene Organic Solar Cells
with >12% Efficiency
Yuze Lin, Fuwen Zhao, Shyamal K. K. Prasad, Jing-De Chen, Wanzhu Cai,
Qianqian Zhang, Kai Chen, Yang Wu, Wei Ma, Feng Gao, Jian-Xin Tang,
Chunru Wang, Wei You, Justin M. Hodgkiss,* and Xiaowei Zhan*
DOI: 10.1002/adma.201706363
Solar cells, a renewable, clean energy tech-
nology that efficiently converts sunlight
into electricity, are a promising long-term
solution for energy and environmental
problems caused by the use of fossil
fuels. Solution-processed organic solar
cells (OSCs) based on an electron donor
(D)/electron acceptor (A) bulk hetero-
junction (BHJ)[1,2] are a promising cost-
effective alternative for conversion of solar
energy, and have attracted much attention
because of their advantages over silicon
solar cells, such as low cost, light weight,
and the capability to fabricate flexible
large-area modules.[3–6] Recently, power
conversion efficiencies (PCEs) of OSCs
have rapidly increased due to sophisti-
cated control over electronic structures
and morphology of photoactive materials
as well as interface modification for effi-
cient charge collection by the electrodes.
The certified PCEs of single-junction
Relative to electron donors for bulk heterojunction organic solar cells (OSCs),
electron acceptors that absorb strongly in the visible and even near-infrared
region are less well developed, which hinders the further development of
OSCs. Fullerenes as traditional electron acceptors have relatively weak visible
absorption and limited electronic tunability, which constrains the optical and
electronic properties required of the donor. Here, high-performance fullerene-
free OSCs based on a combination of a medium-bandgap polymer donor
(FTAZ) and a narrow-bandgap nonfullerene acceptor (IDIC), which exhibit
complementary absorption, matched energy levels, and blend with pure
phases on the exciton diffusion length scale, are reported. The single-junction
OSCs based on the FTAZ:IDIC blend exhibit power conversion efficiencies up
to 12.5% with a certified value of 12.14%. Transient absorption spectroscopy
reveals that exciting either the donor or the acceptor component efficiently
generates mobile charges, which do not suffer from recombination to
triplet states. Balancing photocurrent generation between the donor and
nonfullerene acceptor removes undesirable constraints on the donor imposed
by fullerene derivatives, opening a new avenue toward even higher efficiency
for OSCs.
Organic Solar Cells
J.-D. Chen, Prof. J.-X. Tang
Institute of Functional Nano and Soft Materials (FUNSOM)
Soochow University
Suzhou 215123, China
W. Cai, Prof. F. Gao
Biomolecular and Organic Electronics
IFM
Linköping University
Linköping 58183, Sweden
Q. Zhang, Prof. W. You
Department of Chemistry
University of North Carolina at Chapel Hill
Chapel Hill, NC 27599, USA
Y. Wu, Prof. W. Ma
State Key Laboratory for Mechanical Behavior of Materials
Xi’an Jiaotong University
Xi’an 710049, China
Dr. Y. Lin, Prof. X. Zhan
Department of Materials Science and Engineering
College of Engineering
Key Laboratory of Polymer Chemistry and Physics of Ministry
of Education
Peking University
Beijing 100871, China
E-mail: xwzhan@pku.edu.cn
F. Zhao, Prof. C. Wang
Institute of Chemistry
Chinese Academy of Sciences
Beijing 100190, China
S. K. K. Prasad, K. Chen, Prof. J. M. Hodgkiss
MacDiarmid Institute for Advanced Materials and Nanotechnology
School of Chemical and Physical Sciences
Victoria University of Wellington
Wellington 6010, New Zealand
E-mail: Justin.Hodgkiss@vuw.ac.nz
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.201706363.
Adv. Mater. 2018, 1706363
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1706363 (2 of 8)
www.advmat.dewww.advancedsciencenews.com
OSCs based on fullerene acceptors reported in the literature
were as high as 11.5%.[7]
A broad absorption spectrum and efficient exciton dissocia-
tion are prerequisites for achieving a high PCE in solar cells.
A single organic semiconductor is unlikely to absorb the full
visible and near-infrared (NIR) solar spectrum (300–1100 nm).
Thus, it is critical for D/A BHJ in OSCs to have complementary
absorption in the visible-NIR region. The exciton dissociation
process in OSCs, during which the photogenerated excitons are
converted into free charges, plays a key role in the photon-to-
electron conversion. Finely tuned energy levels of D and A are
needed to achieve sufficient driving force of exciton dissocia-
tion, small energy loss, and high open-circuit voltage (VOC).[8]
Most of the efficient OSCs developed to date have employed
fullerene-based acceptors; however, their weak visible absorp-
tion and limited energy-level variability impose constraints on
the donor. These outstanding issues have imposed fundamental
limitations to the further improvement of efficiencies of OSCs.
For example, low-bandgap donors whose energy levels are tuned
to absorb across the visible spectrum and preserve maximal VOC
when paired with fullerene would inevitably have energetically
accessible triplet states that accelerate recombination.[9]
As a promising alternative, nonfullerene acceptors
have become a new research focus in the field of
OSCs.[10] Compared with the typical imide-functionalized
rylenes,[8,11–20] fused ring electron acceptors (FREAs) based on
electron-donating extended fused rings, e.g., indaceno-
dithiophene and indacenodithieno[3,2-b]thiophene, with
acceptor–donor–acceptor (A–D–A) structure[21–33] exhibit
stronger absorption in the NIR region, tunable energy levels,
and consequently higher PCEs amongst nonfullerene OSCs.
In fact, FREA-based OSCs have shown both higher PCEs and
better stability than their fullerene counterparts when paired
with the same donors.[26] Here, the low-bandgap FREA is ide-
ally matched with a medium-bandgap donor.
Here, we report high-performance fullerene-free OSCs based
on a combination of a medium-bandgap polymer donor[34]
(FTAZ, Figure 1a) and a narrow-bandgap FREA acceptor[22]
(IDIC, Figure 1a), which together exhibit complementary
absorption, matched energy levels, and relatively high hole
and electron mobilities. The PCE of FTAZ:IDIC-based single-
junction OSCs can reach up to 12.5% with a certified PCE of
12.14%, by combining an advanced light manipulation scheme
for broadband self-enhanced light absorption. More impor-
tantly, transient absorption (TA) spectroscopy reveals that
exciting either the donor or acceptor component efficiently gen-
erates mobile charges, which do not suffer from recombination
to triplet states. Balancing photocurrent generation between
the donor and acceptor removes undesirable constraints on the
donor, and unlocks the next step of efficiency gains through
tuning energy levels.
Unlike fullerene acceptors like 6,6-phenyl C61 butyric acid
methyl ester (PCBM) that show weak absorption in the vis-
ible region, IDIC shows stronger absorption at ≈600–800 nm,
where the solar spectrum has the maximum photon flux den-
sity. IDIC possesses a narrow optical bandgap of 1.6 eV, while
FTAZ has a medium bandgap of 2.0 eV; IDIC and FTAZ have
complementary absorption (400–800 nm, Figure 1b).[35] IDIC
Adv. Mater. 2018, 1706363
Figure 1. a) Chemical structures of FTAZ and IDIC. b) UV–vis absorption spectra of the thin film of FTAZ, IDIC, and FTAZ:IDIC (1:1.5, w/w).
c) J–V curves of devices with the structure ITO/ZnO/FTAZ:IDIC/MoOx/Ag with 0.25% DIO on flat and patterned substrates.
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1706363 (3 of 8)
www.advmat.dewww.advancedsciencenews.com
possesses a lowest unoccupied molecular orbital energy level
of −3.9 eV, similar to that of PCBM; IDIC and FTAZ have
matched energy levels (Figure S1, Supporting Information),
which are important for a properly matched D/A pair and high
performance of solar cells.
We fabricated inverted BHJ OSCs with a device structure
of indium tin oxide (ITO)/ZnO/FTAZ:IDIC or PCBM/MoOx/
Ag (Figure 1a). The optimization details on device performance
can be seen in Tables S1 and S2 in the Supporting Informa-
tion. As-cast devices based on the optimized FTAZ:IDIC weight
ratio of 1:1.5 and optimized thickness of ≈85 nm, yield a PCE
of up to 9.70% (Figure S2a, Supporting Information). Further-
more, the OSCs using 0.25% additive 1,8-diiodooctane (DIO) in
chloroform show higher PCE of up to 11.6% (average PCE for
20 devices: 11.4%) when the thickness increases up to ≈110 nm
(Table 1 and Figure 1c). The increased optimized thickness
of active layer with DIO can be attributed to enhanced charge
mobilities. The hole and electron mobilities of FTAZ:IDIC
with 0.25% DIO calculated by space charge limited current
method are 1–2 orders of magnitude higher than those of as-
cast FTAZ:IDIC film (Figure S3 and Table S3, Supporting
Information). In comparison, as-cast control devices based
on FTAZ:PCBM with an optimized thickness of ≈115 nm,
yield a lower PCE of 5.99% (Figure S2a, Supporting Infor-
mation). FTAZ:PCBM-based OSCs with 0.25% DIO additive
show slightly lower PCEs of 5.14–5.34%, when the active layer
thickness is 105 to 135 nm (Table S4, Supporting Informa-
tion). FTAZ:PCBM blended film shows higher charge mobili-
ties (Figure S3 and Table S3, Supporting Information) than
FTAZ:IDIC blended film, leading to the thicker optimized
FTAZ:PCBM layer with the thickness of up to 135 nm.
Compared with FTAZ:PCBM-based OSCs, the significantly
enhanced short-circuit current density (JSC) of FTAZ:IDIC-
based OSCs is mainly attributed to the strong absorption
of IDIC between 600 and 800 nm, which is confirmed by
the broad external quantum efficiency (EQE) plateau at
350–780 nm (Figure S4a, Supporting Information). The opti-
mized devices based on FTAZ:IDIC with 0.25% DIO show
high EQE up to 79% in the NIR region. In contrast, the
FTAZ:PCBM-based devices show negligible EQE from 650 to
780 nm. More comparison on morphology, and photophysics
between FTAZ:IDIC and FTAZ:PCBM systems are discussed
in sections below.
To further improve the device performance of FTAZ:IDIC-
based OSCs, we patterned a deterministic aperiodic nano-
structure (DAN) into ZnO films on the substrates to optimize
the light-harvesting property of the OSCs.[36] As a result, the
device JSC is enhanced to 20.8 mA cm−2, leading to a best
PCE of 12.5% (Figure 1c), with an average value of 12.1% for
20 devices. Statistics of the PCE distributions of 20 devices
fabricated are shown in Figure S2b in the Supporting Infor-
mation, over 50% of the devices show PCEs larger than 12%.
Patterned devices with encapsulation were sent to the National
Center of Supervision and Inspection on Solar Photovoltaic
Products Quality of China (CPVT) for certification, and a certi-
fied PCE of 12.14% was obtained (Figure S5, Supporting Infor-
mation). Compared with the control FTAZ:IDIC device with a
flat structure, the devices patterned with DANs have an obvious
EQE improvement between 520 and 750 nm (Figure S4b,
Supporting Information), which can mostly be attributed to
the enhanced absorption arising from a combination of sev-
eral factors, including the pattern-induced antireflection, light
scattering, and surface plasmonic resonance.[36] From cross-
sectional scanning electron microscope images (Figure S6,
Supporting Information), we can see that the patterned ZnO
film is coated with slightly more material than the flat ZnO
substrate (although the measured difference is within experi-
mental error); the active layer thicknesses of optimized devices
on flat and patterned ZnO substrates are 110 ± 3 nm and 113 ±
6 nm, respectively. The additional photoactive materials on pat-
terned ZnO may make a small contribution to the increased
JSC observed from patterning. The effects of patterned ZnO on
device performance are further clarified through the following
investigations of morphology and photophysics.
From atomic force microscopy images (Figure S7, Sup-
porting Information), both FTAZ:IDIC and FTAZ:PCBM
blended films with 0.25% DIO additive on flat ZnO substrate
show smooth surfaces with root-mean-square (RMS) rough-
ness of ≈1.2–1.4 nm. However, RMS roughness (3.67 nm) of
FTAZ:IDIC layer with 0.25% DIO deposited on patterned ZnO
is higher than that (1.22 nm) on flat ZnO, which can be mainly
attributed to the much rougher surface (Figure S8, Supporting
Information) of patterned ZnO substrate than flat ZnO (RMS
roughness: 13.7 nm vs 1.67 nm[37]). Relative to original pat-
terned ZnO surface, the surface of FTAZ:IDIC layer deposited
on it becomes much smoother. This result indicates that the
deep ridges in patterned ZnO are filled with more material and
the hills are coated with thinner active layer.
Grazing incidence wide angle X-ray scattering (GIWAXS)
was used to investigate the molecular orientation and packing
of FTAZ:IDIC and FTAZ:PCBM blended films on flat ZnO
substrate, as well as FTAZ:IDIC BHJ on patterned ZnO sub-
strates. According to the scattering of the FTAZ and IDIC neat
films (Figure S9, Supporting Information), the peaks located at
q ≈ 0.3 and 0.6 Å−1 originate from the lamellar packing of
FTAZ, while the peaks at q ≈ 0.47 and 0.94 Å−1 correspond to
Adv. Mater. 2018, 1706363
Table 1. Device data of OSCs based on FTAZ:IDIC under the illumination of AM 1.5 G, 100 mW cm−2. The average and standard values of 20 devices
are in the brackets.
D:A [w/w] DIO ratio [%] VOC [V] JSC [mA cm−2]FF PCE [%]
1:1.5 0.25 0.850 (0.845 ± 0.005) 18.5 (18.4 ± 0.18) 0.733 (0.731 ± 0.01) 11.6 (11.4 ± 0.18)
1:1.5a) 0.25 0.840 (0.838 ± 0.002) 20.8 (20.7 ± 0.42) 0.718 (0.697 ± 0.02) 12.5 (12.1 ± 0.4)
1:1.5a,b) 0.25 0.840 21.4 0.675 12.14
a)With patterned structure; b)Tested by CPVT.
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1706363 (4 of 8)
www.advmat.dewww.advancedsciencenews.com
the lamellar packing of IDIC (Figure 2a,b). The
π
–
π
stacking
peak of FTAZ and IDIC is located at q ≈ 1.68 and 1.80 Å−1,
respectively. The coherence length (CL) of IDIC
π
–
π
stacking
in the out-of-plane direction is calculated (via the Scherrer
equation: CL = 2
π
K/FWHM, where K is a shape factor (here
uses 0.9) and FWHM is the full-width at half-maximum of the
peak) to be 3.5 and 3.3 nm for the FTAZ:IDIC blended films
with 0.25% DIO deposited on flat and patterned ZnO substrate,
respectively.[38] The slightly deceased coherence length for the
FTAZ:IDIC blended films on patterned substrate is unfavorable
for charge transport. In FTAZ:PCBM system, the aggregation
peak of PCBM locates at q ≈ 1.36 Å−1, and both donor and
acceptor exhibit weak scattering peaks in blended films, which
indicates the weaker molecular packing of FTAZ and PCBM.
It is noted that the FTAZ and IDIC
π
–
π
stacking in the out-of-
plane direction of all the blended films are visibly stronger than
that in the in-plane direction, indicating both FTAZ and IDIC
take predominantly face-on orientation relative to the electrode/
substrate. The vertical
π
–
π
stacking is well known to benefit
charge transport between anode and cathode of solar cells.
Resonant soft X-ray scattering (R-SoXS) was used to probe
the phase separation in FTAZ:IDIC blended films on flat
or patterned ZnO substrates and the control FTAZ:PCBM
blended film on flat ZnO substrate. The resonant energy of
284.2 eV was selected to provide highly enhanced materials
contrast. The mode of the distribution smode of the scattering
corresponds to the characteristic mode length scale,
ξ
, of the
corresponding log-normal distribution in real space with
ξ
= 1/smode, a model independent statistical quantity. It is noted
that the mode domain size is the half of
ξ
. The mode domain
size of 52 nm is obtained for the FTAZ:IDIC blended films on
flat substrate. On the patterned ZnO substrate, the domain size
of FTAZ:IDIC blend is reduced with domain size of 22 nm, and
these smaller domain sizes are beneficial to exciton dissociation
in BHJ. Meanwhile, the domain size of 22 nm is obtained for
FTAZ:PCBM blend with 0.25% DIO. Furthermore, R-SoXS can
also reveal the average composition variation (relative domain
purity) via integrating scattering profile, which is indicative of
the average purity of donor and acceptor domains. Considering
that the optical contrasts between FTAZ and IDIC or PCBM are
far different (Figure S10, Supporting Information), the relative
purity derived from R-SoXS is normalized by their respective
material contrasts measured using near edge X-ray absorp-
tion fine structure spectroscopy. The relative domain purity of
FTAZ:IDIC with 0.25% DIO on flat and patterned ZnO sub-
strates is 1 and 0.89, respectively. FTAZ:PCBM blended film on
flat ZnO substrate exhibits much lower total scattering inten-
sity with the relative domain purity of 0.30.
Adv. Mater. 2018, 1706363
Figure 2. a) The 2D GIWAXS patterns, b) scattering profiles of in-plane and out-of-plane, and c) R-SoXS profiles for FTAZ:IDIC blended films with
0.25% DIO on flat and patterned ZnO substrates and FTAZ:PCBM blended film with 0.25% DIO on flat ZnO substrate.
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1706363 (5 of 8)
www.advmat.dewww.advancedsciencenews.com
We summarize GIWAXS and R-SoXS analysis above to
understand the effect of the patterned substrate on mor-
phology of active layer, as well as to compare the morphology
between FTAZ:IDIC and FTAZ:PCBM systems, as follows.
First, the patterned ZnO substrate results in FTAZ:IDIC
having both smaller and less pure domains, along with slightly
weaker IDIC
π
–
π
stacking, and commensurately lower electron
mobility (Figure S11 and Table S3, Supporting Information).
These morphological differences induced by the substrate
patterning could affect photovoltaic performance in different
ways; however, the observed negligible impact on charge gen-
eration and recombination dynamics described in the photo-
physics section below suggests that enhanced light absorption
is the most important factor. Second, relative to the FTAZ:IDIC
blend, FTAZ:PCBM blended films show weaker molecular
packing and much lower relative domain purity. In order to
obtain complementary information on the morphology of
FTAZ:IDIC and FTAZ:PCBM, we performed optical spectros-
copy measurements that can effectively resolve how morpho-
logical differences are manifested in charge generation and
recombination behavior.
Figure 3a shows the relative photoluminescence (PL) spectra
of the pristine IDIC film and the FTAZ:IDIC films processed
with 0.25% DIO. We find that the emission intensity of the
IDIC band is only ≈5% for the blend with DIO of the pristine
IDIC film emission. This indicates efficient exciton quenching
via photoinduced hole transfer, with IDIC domains on the order
of the exciton diffusion length. For FTAZ:PCBM blends with
0.25% DIO (Figure 3b), the remaining emission is ≈10%—unu-
sually high for PCBM blends. While not a significant efficiency
loss, this residual PL suggests that, in spite of the compara-
tively lower relative domain purity indicated by R-SoXS, FTAZ
domains still have a smaller interface area (larger or purer
domains) than typical PCBM blends.
We carried out ultrafast TA spectroscopy in order to resolve
the exciton-to-charge conversion process in nonfullerene and
fullerene blends via donor- and acceptor excitation channels.
When exciting the acceptor component of the FTAZ:IDIC blend
on a flat fused silica substrate (Figure 4a), the initial TA spec-
trum resembles the exciton spectrum of neat IDIC;[27] it is dom-
inated by ground-state bleach peaks at ≈1.9 and 1.7 eV (mixed
with stimulated emission on the red edge), and a photoinduced
absorption peak at 1.4 eV. Over time, the IDIC exciton spec-
trum decays to reveal a TA spectrum of charge pairs resulting
from hole transfer; bleaching of the IDIC is complemented
with bleaching of the FTAZ polymer above 2 eV, the contribu-
tion of IDIC stimulated emission disappears, and the exciton-
based photoinduced absorption peak at 1.4 eV is replaced by
new peak at 1.3 eV. This photoinduced absorption is attributed
to electrons in IDIC, confirmed by its correspondence in IDIC
blends with a different donor polymer,[27] and it overlaps with
a broader NIR photoinduced absorption from holes in FTAZ
(confirmed below). Exciting the donor component of the FTAZ:
IDIC blend at 532 nm also results in efficient charge genera-
tion. The main difference in that case (Figure 4b) is that the
initial TA spectrum contains signatures of excitons in both
FTAZ and IDIC, which undergo charge generation via electron-
and hole transfer. The TA series for the FTAZ:PCBM blend in
Figure 4c exhibits a progression from polymer-based excitons
to charges after electron transfer. Holes in FTAZ exhibit broad
NIR absorption spanning between peaks at 1.9 and 1.3 eV
(the former likely arising from electroabsorption), and these
polymer-based features are presumed to be mixed with IDIC
features to account for the NIR shape of the nonfullerene blend
in Figure 4a.
In order to compare charge dynamics for each of the blends,
and for electron- versus hole-transfer channels, the TA surfaces
were mathematically decomposed into spectra and popula-
tions of excitons and charges (Figure S12, Supporting Infor-
mation), whose dynamics are shown in Figure 4d. In typical
polymer:fullerene blends, a high fraction (typically 70–90%)
of charges is generated on an ultrafast (≈100 fs) timescale,[39]
which has even been proposed as essential for free charge photo-
generation.[40] Contrary to this expectation, only ≈40% of exci-
tons undergo ultrafast charge generation in the FTAZ:PCBM
blend, with the remainder diffusion in the following pico-
seconds. Consistent with the residual PL (Figure 3b), these
dynamics suggest that FTAZ polymer domains are larger and
purer in its fullerene-based blend than in typical fullerene-
based blends. A similar charge generation profile is seen for
the FTAZ:IDIC blend when exciting the donor, which is con-
sistent with the EQE spectra in the donor absorption region for
blends with PCBM and IDIC (Figure S4, Supporting Informa-
tion), and suggests that the FTAZ phases maintain a suitable
Adv. Mater. 2018, 1706363
Figure 3. Relative PL spectra following 450 nm excitation for a) a pristine IDIC film and FTAZ:IDIC (1:1.5, w/w) blend film with 0.25% DIO, and
b) a pristine FTAZ film and FTAZ:PCBM (1:1.5, w/w) blend film with 0.25% DIO.
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1706363 (6 of 8)
www.advmat.dewww.advancedsciencenews.com
size and purity for charge generation in both IDIC and PCBM
blends.
A different charge photogeneration profile, however, is seen
when exciting the IDIC acceptor component of the FTAZ:IDIC
blend. In this hole-transfer channel, which accounts for sig-
nificant photocurrent, only about 10% of charges are generated
promptly for the blend processed with DIO, with the remainder
forming comparatively slowly via diffusion on the early pico-
second timescale. This slow charge generation profile confirms
the very high IDIC phase purity that was found in the morpho-
logical studies. Importantly, the pure phases indicated by slow
charge generation kinetics are favorable for suppressing charge
recombination. We also found that ZnO (flat or patterned) does
not affect these charge generation dynamics (Figures S13–S15,
Supporting Information), suggesting that the local environment
experienced by excitons is not strongly affected by the morpho-
logical differences as a result of patterning resolved via R-SoXS.
This supports the conclusion that enhanced light absorption is
the main benefit of substrate patterning, rather than enhancing
internal quantum efficiency.
Having confirmed the efficient, yet surprisingly slow, charge
photogeneration via both electron- and hole-transfer channels
Adv. Mater. 2018, 1706363
Figure 4. a) Series of TA spectra evolution for a FTAZ:IDIC blended film (1:1.5, w/w, 0.25% DIO) following 100 fs excitation at 712 nm (0.26 µJ cm−2),
and b) following 100 fs excitation at 532 nm (2.1 µJ cm−2). c) Series of TA spectra evolution for a FTAZ:PCBM (1:1.5, w/w, 0.25% DIO) following
100 fs excitation at 532 nm (0.60 µJ cm−2). d) Charge dynamics extracted from a decomposition of the TA surfaces for the range of blends and excita-
tion wavelengths indicated (≈0.6–2 µJ cm−2), with charge populations normalized assuming unit efficiency of exciton-to-charge conversion in the low
fluence limit. e) Fluence-dependent charge dynamics for the FTAZ:IDIC blend (1:1.5, w/w, 0.25% DIO) following 532 nm excitation, along with the
bimolecular recombination constant,
β
, extracted from a moving window global fit (Figure S15, Supporting Information). f) Normalized series of TA
spectra for the FTAZ:IDIC blend (1:1.5, w/w, 0.25% DIO) following 532 nm with either 100 fs (2.1 µJ cm−2) or ≈700 ps (0.6 µJ cm−2) pulses, for delays
less than or greater than 5 ns, respectively.
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1706363 (7 of 8)
www.advmat.dewww.advancedsciencenews.com
in FTAZ:IDIC blends, we now examine charge recombination.
Figure 4d reveals recombination from ≈100 ps into the micro-
second regime, with similar recombination profiles observed
for all blends. Rapid recombination is typically viewed as unfa-
vorable because it usually implies a geminate recombination
loss channel. However, the pronounced acceleration of charge
recombination with increasing fluence (Figure 4e)—even in a
relatively low fluence range—confirms nongeminate recombi-
nation as the major recombination channel and suggests that
charges are highly mobile.
In order to evaluate the charge mobility for different blends
and excitation channels, we globally fit the fluence-dependent
data using a moving window fit with a time-dependent bimo-
lecular recombination constant,
β
. Figure 4e shows that
β
exceeds 10−14 cm3 s−1 on the sub-nanosecond timescale and
then decays as a result of mobility dispersion.[41] The magni-
tude and dynamics of bimolecular recombination constants
are similar when exciting the IDIC, and for the PCBM blend
(Figures S16–S18, Supporting Information). The bimolecular
recombination constants are not expected to hamper device
performance under solar illumination; steady-state excita-
tion densities of 1016 cm−3 would correspond to charge life-
times exceeding microseconds, sufficient for charge extraction
to prevail (indeed, electron extraction into ZnO is observed
by the loss of IDIC electronic signatures by the microsecond
timescale, Figure S13b,c, Supporting Information). Instead,
the high bimolecular recombination constants show that both
donor and acceptor excitation channels produce highly mobile
charges that can rapidly diffuse away from interfaces.
Recent TA studies of polymer:fullerene blends have observed
rapid bimolecular recombination to triplet states that lie below
charge-transfer states when maximal voltage is sought using a
low-bandgap donor,[9,42,43] and Menke et al. concluded that such
behavior may impose a fundamental efficiency limit for OSCs.[9]
We investigated whether this was the case for FTAZ:IDIC. Dis-
tinguishing between charge pairs and triplets is simplified for
FTAZ:IDIC, because both components contribute to the spec-
trum of charge pairs and only one component would contribute
to the spectrum of a triplet exciton; in contrast, for fullerene
blends, each of the spectra is dominated by the donor. Figure 4f
shows a series of normalized TA spectra for FTAZ:IDIC
throughout the recombination timescale. Triplet formation is
ruled out on the basis that the spectral signatures from FTAZ
and IDIC maintain a constant ratio throughout the recombina-
tion timescale. Specifically, holes in FTAZ contribute the bleach
above 2 eV and a broad offset to NIR absorption, and electrons
in IDIC contribute bleaching around 1.7 eV and photoinduced
absorption at 1.3 eV, which is clearly seen against the reference
spectrum of holes in FTAZ:PCBM. The only spectral evolu-
tion occurs in the polymer subgap region, which likely reflects
changes in electroabsorption signals associated with interfacial
electric fields.
The absence of triplets here likely reflects an energy-level
alignment whereby the driving energy lost in charge transfer
makes subsequent triplet formation thermodynamically inac-
cessible. This band alignment—similar to P3HT:PCBM—
(Figure S1, Supporting Information) remains viable for
this >12% PCE blend because complementary absorption by
both components makes the optimal donor a medium- rather
than a low-bandgap material. This larger bandgap raises the
poly mer triplet energy level away from the charge-transfer state,
and the wasted voltage from excess driving energy is compen-
sated by reduced energy loss from exciton relaxation to a higher
bandgap. In spite of achieving >12% PCE, it is still likely that
too much voltage is wasted in charge transfer. The pathway to
capturing this extra voltage and further increasing PCEs will be
to reduce band-edge offsets, while retaining similar bandgaps,
until the triplet-imposed limit is reached. On the other hand,
low-bandgap:fullerene blends have apparently already reached
this limit, at lower PCE, when donor energy levels are opti-
mized against PCBM for both light absorption and voltage.[9]
In summary, we fabricated OSCs using a combination of a
medium-bandgap polymer donor FTAZ and a narrow-bandgap
acceptor IDIC, which exhibit complementary absorption,
matched energy levels, and relatively high hole and electron
mobilities. The single-junction OSCs based on FTAZ:IDIC
with a light manipulation scheme yield PCE up to 12.5% with
a certified value of 12.14%, which is much higher than the
control devices based on FTAZ:PCBM (<6%). The FTAZ:IDIC
blend shows phase separation on the exciton diffusion range,
but high domain purity, which is beneficial to charge separa-
tion and transport. The high domain purity results in surpris-
ingly slow charge generation dynamics compared with typical
fullerene blends, whether exciting the donor or acceptor
component, while domains remain small enough for charge
generation to still be efficient. Fluence-dependent measure-
ments reveal photogenerated charges to be highly mobile, and
unlike in fullerene blends that require low-bandgap polymers
to absorb across the visible spectrum, the FTAZ:IDIC blend
does not suffer from recombination to triplet excitons. Thus,
we find that nonfullerene acceptors that can form suitable mor-
phologies have unlocked the next efficiency gains; their strong
contribution to photocurrent relieves undesirable constraints
on donor energy levels, and should permit further tuning of
energy levels to boost cell voltage and efficiency.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
Y.L., F.Z., S.K.K.P., and J.-D.C. contributed equally to this work. X.Z.
thanks NSFC (Grant Nos. 21734001 and 51761165023). Y.L. thanks
NSFC (Grant No. 21504058). J.-X.T. thanks NSFC (Grant No. 11474214)
and the Ministry of Science and Technology of China (Grant No.
2014CB932600). Q.Z. and W.Y. thank the Office of Naval Research
(Grant No. N000141410221) and NSF (Grant No. DMR-1507249).
W.M. thanks the Ministry of Science and Technology of China (Grant
No. 2016YFA0200700) and NSFC (Grant No. 21504066). X-ray data
were acquired at beamlines 7.3.3[44] and 11.0.1.2[45] at the Advanced
Light Source, which is 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. S.K.K.P., K.C., and J.M.H.
gratefully acknowledge Yingzhuang Ma for assistance with sample
preparation, along with support from a Rutherford Discovery fellowship.
W.C. and F.G. gratefully acknowledge Prof. Olle Inganäs for his support
and insightful discussions. The research at Linköping was supported
Adv. Mater. 2018, 1706363
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1706363 (8 of 8)
www.advmat.dewww.advancedsciencenews.com
by the Knut and Alice Wallenberg Foundation through a Wallenberg
Scholar grant to O. Inganäs and the Swedish Government Strategic
Research Area in Materials Science on Functional Materials at Linköping
University (Faculty Grant SFO-Mat-LiU No. 2009-00971).
Conflict of Interest
The authors declare no conflict of interest.
Keywords
fused ring electron acceptors, nonfullerene solar cells, organic solar
cells, photophysics
Received: November 2, 2017
Revised: December 31, 2017
Published online:
[1] G. Yu, J. Gao, J. C. Hummelen, F. Wudl, A. J. Heeger, Science 1995,
270, 1789.
[2] J. J. M. Halls, C. A. Walsh, N. C. Greenham, E. A. Marseglia,
R. H. Friend, S. C. Moratti, A. B. Holmes, Nature 1995, 376, 498.
[3] L. Lu, T. Zheng, Q. Wu, A. M. Schneider, D. Zhao, L. Yu, Chem. Rev.
2015, 115, 12666.
[4] F. C. Krebs, N. Espinosa, M. Hösel, R. R. Søndergaard,
M. Jørgensen, Adv. Mater. 2014, 26, 29.
[5] G. Li, R. Zhu, Y. Yang, Nat. Photonics 2012, 6, 153.
[6] Y. Lin, X. Zhan, Acc. Chem. Res. 2016, 49, 175.
[7] J. Zhao, Y. Li, G. Yang, K. Jiang, H. Lin, H. Ade, W. Ma, H. Yan, Nat.
Energy 2016, 1, 15027.
[8] J. Liu, S. Chen, D. Qian, B. Gautam, G. Yang, J. Zhao, J. Bergqvist,
F. Zhang, W. Ma, H. Ade, O. Inganäs, K. Gundogdu, F. Gao,
H. Yan, Nat. Energy 2016, 1, 16089.
[9] S. M. Menke, A. Sadhanala, M. Nikolka, N. A. Ran, M. K. Ravva,
S. Abdel-Azeim, H. L. Stern, M. Wang, H. Sirringhaus, T. Q. Nguyen,
J. L. Bredas, G. C. Bazan, R. H. Friend, ACS Nano 2016, 10, 10736.
[10] Y. Lin, X. Zhan, Mater. Horiz. 2014, 1, 470.
[11] Y. Zhong, M. T. Trinh, R. Chen, G. E. Purdum, P. P. Khlyabich,
M. Sezen, S. Oh, H. Zhu, B. Fowler, B. Zhang, W. Wang, C.-Y. Nam,
M. Y. Sfeir, C. T. Black, M. L. Steigerwald, Y.-L. Loo, F. Ng, X. Y. Zhu,
C. Nuckolls, Nat. Commun. 2015, 6, 8242.
[12] D. Meng, D. Sun, C. Zhong, T. Liu, B. Fan, L. Huo, Y. Li, W. Jiang,
H. Choi, T. Kim, J. Y. Kim, Y. Sun, Z. Wang, A. J. Heeger, J. Am.
Chem. Soc. 2015, 138, 375.
[13] P. E. Hartnett, A. Timalsina, H. S. S. R. Matte, N. Zhou,
X. Guo, W. Zhao, A. Facchetti, R. P. H. Chang, M. C. Hersam,
M. R. Wasielewski, T. J. Marks, J. Am. Chem. Soc. 2014, 136, 16345.
[14] Q. Wu, D. Zhao, A. M. Schneider, W. Chen, L. Yu, J. Am. Chem. Soc.
2016, 138, 7248.
[15] T. Kim, J. H. Kim, T. E. Kang, C. Lee, H. Kang, M. Shin, C. Wang,
B. W. Ma, U. Jeong, T. S. Kim, B. J. Kim, Nat. Commun. 2015, 6, 8547.
[16] Y.-J. Hwang, H. Li, B. A. E. Courtright, S. Subramaniyan,
S. A. Jenekhe, Adv. Mater. 2016, 28, 124.
[17] Y. Diao, Y. Zhou, T. Kurosawa, L. Shaw, C. Wang, S. Park,
Y. K. Guo, J. A. Reinspach, K. Gu, X. D. Gu, B. C. K. Tee, C. H. Pang,
H. P. Yan, D. H. Zhao, M. F. Toney, S. C. B. Mannsfeld, Z. A. Bao,
Nat. Commun. 2015, 6, 7955.
[18] Y. Z. Lin, Y. F. Wang, J. Y. Wang, J. H. Hou, Y. F. Li, D. B. Zhu,
X. W. Zhan, Adv. Mater. 2014, 26, 5137.
[19] H. Yan, Z. H. Chen, Y. Zheng, C. Newman, J. R. Quinn, F. Dotz,
M. Kastler, A. Facchetti, Nature 2009, 457, 679.
[20] X. Zhan, Z. a. Tan, B. Domercq, Z. An, X. Zhang, S. Barlow, Y. Li,
D. Zhu, B. Kippelen, S. R. Marder, J. Am. Chem. Soc. 2007, 129,
7246.
[21] Y. Z. Lin, J. Y. Wang, Z. G. Zhang, H. T. Bai, Y. F. Li, D. B. Zhu,
X. W. Zhan, Adv. Mater. 2015, 27, 1170.
[22] Y. Lin, Q. He, F. Zhao, L. Huo, J. Mai, X. Lu, C.-J. Su, T. Li, J. Wang,
J. Zhu, Y. Sun, C. Wang, X. Zhan, J. Am. Chem. Soc. 2016, 138,
2973.
[23] Y. Z. Lin, F. W. Zhao, Q. He, L. J. Huo, Y. Wu, T. C. Parker, W. Ma,
Y. M. Sun, C. R. Wang, D. B. Zhu, A. J. Heeger, S. R. Marder,
X. W. Zhan, J. Am. Chem. Soc. 2016, 138, 4955.
[24] H. J. Bin, Z. G. Zhang, L. Gao, S. S. Chen, L. Zhong, L. W. Xue,
C. Yang, Y. F. Li, J. Am. Chem. Soc. 2016, 138, 4657.
[25] S. Dai, F. Zhao, Q. Zhang, T. K. Lau, T. Li, K. Liu, Q. Ling, C. Wang,
X. Lu, W. You, X. Zhan, J. Am. Chem. Soc. 2017, 139, 1336.
[26] W. Zhao, D. Qian, S. Zhang, S. Li, O. Inganäs, F. Gao, J. Hou, Adv.
Mater. 2016, 28, 4734.
[27] Y. Lin, F. Zhao, Y. Wu, K. Chen, Y. Xia, G. Li, S. K. Prasad, J. Zhu,
L. Huo, H. Bin, Z. G. Zhang, X. Guo, M. Zhang, Y. Sun, F. Gao,
Z. Wei, W. Ma, C. Wang, J. Hodgkiss, Z. Bo, O. Inganas, Y. Li,
X. Zhan, Adv. Mater. 2017, 29, 1604155.
[28] S. Holliday, R. S. Ashraf, A. Wadsworth, D. Baran, S. A. Yousaf,
C. B. Nielsen, C. Tan, S. D. Dimitrov, Z. Shang, N. Gasparini,
M. Alamoudi, F. Laquai, C. J. Brabec, A. Salleo, J. R. Durrant,
I. McCulloch, Nat. Commun. 2016, 7, 11585.
[29] F. Zhao, S. Dai, Y. Wu, Q. Zhang, J. Wang, L. Jiang, Q. Ling, Z. Wei,
W. Ma, W. You, C. Wang, X. Zhan, Adv. Mater. 2017, 29, 1700144.
[30] J. Zhu, Z. Ke, Q. Zhang, J. Wang, S. Dai, Y. Wu, Y. Xu, Y. Lin, W. Ma,
W. You, X. Zhan, Adv. Mater. 2018, 30, 1704713.
[31] D. Baran, R. S. Ashraf, D. A. Hanifi, M. Abdelsamie, N. Gasparini,
J. A. Rohr, S. Holliday, A. Wadsworth, S. Lockett, M. Neophytou,
C. J. Emmott, J. Nelson, C. J. Brabec, A. Amassian, A. Salleo,
T. Kirchartz, J. R. Durrant, I. McCulloch, Nat. Mater. 2017, 16, 363.
[32] H. Bin, L. Gao, Z. G. Zhang, Y. Yang, Y. Zhang, C. Zhang, S. Chen,
L. Xue, C. Yang, M. Xiao, Y. Li, Nat. Commun. 2016, 7, 13651.
[33] Z. Li, K. Jiang, G. F. Yang, J. Y. L. Lai, T. X. Ma, J. B. Zhao, W. Ma,
H. Yan, Nat. Commun. 2016, 7, 13094.
[34] S. C. Price, A. C. Stuart, L. Yang, H. Zhou, W. You, J. Am. Chem. Soc.
2011, 133, 4625.
[35] W. Li, S. Albrecht, L. Yang, S. Roland, J. R. Tumbleston, T. McAfee,
L. Yan, M. A. Kelly, H. Ade, D. Neher, W. You, J. Am. Chem. Soc.
2014, 136, 15566.
[36] J. D. Chen, C. H. Cui, Y. Q. Li, L. Zhou, Q. D. Ou, C. Li, Y. F. Li,
J. X. Tang, Adv. Mater. 2015, 27, 1035.
[37] F. Zhao, Z. Wang, J. Q. Zhang, X. W. Zhu, Y. J. Zhang, J. Fang,
D. Deng, Z. X. Wei, Y. F. Li, L. Jiang, C. R. Wang, Adv. Energy Mater.
2016, 6, 1502120.
[38] D. M. Smilgies, J. Appl. Crystallogr. 2009, 42, 1030.
[39] A. J. Barker, K. Chen, J. M. Hodgkiss, J. Am. Chem. Soc. 2014, 136,
12018.
[40] L. G. Kaake, D. Moses, A. J. Heeger, J. Phys. Chem. Lett. 2013, 4,
2264.
[41] A. Melianas, F. Etzold, T. J. Savenije, F. Laquai, O. Inganas,
M. Kemerink, Nat. Commun. 2015, 6, 8778.
[42] A. Rao, P. C. Y. Chow, S. Gelinas, C. W. Schlenker, C. Z. Li, H. L. Yip,
A. K. Y. Jen, D. S. Ginger, R. H. Friend, Nature 2013, 500, 435.
[43] F. Etzold, I. A. Howard, N. Forler, A. Melnyk, D. Andrienko,
M. R. Hansen, F. Laquai, Energy Environ. Sci. 2015, 8, 1511.
[44] 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.
[45] 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.
Adv. Mater. 2018, 1706363
