Spectral Engineering of Semitransparent Polymer Solar Cells for Greenhouse Applications

Abstract

In this study, a wavelength selective semitransparent polymer solar cell (ST‐PSC) with a proper transmission spectrum for plant growth is proposed for greenhouse applications. A ternary strategy combining a wide bandgap polymer donor with a near‐infrared absorbing nonfullerene acceptor and a high electron mobility fullerene acceptor is introduced to achieve PSCs with power conversion efficiency (PCE) over 10%. The addition of PC71BM into J52:IEICO‐4F binary blend contributes to the suppressed trap‐assisted recombination, enhanced charge extraction, and improved open‐circuit voltage simultaneously. ST‐PSC based on the J52:IEICO‐4F:PC71BM ternary blend shows an optimized performance with PCE of 7.75% and a defined crop growth factor of 24.8%. Such high‐performance ST‐PSC is achieved by carefully engineering the absorption spectrum of the light harvesting materials. As a result, the transmission spectra of the semitransparent devices are well‐matched with the absorption spectra of the photoreceptors, such as chlorophylls, in green plants, which provides adequate lighting conditions for photosynthesis and plant growth, and therefore making it a competitive candidate for photovoltaic greenhouse applications.

Full text

www.advenergymat.de
Full paper
1803438 (1 of 8) © 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Spectral Engineering of Semitransparent Polymer Solar
Cells for Greenhouse Applications
Hui Shi, Ruoxi Xia, Guichuan Zhang, Hin-Lap Yip,* and Yong Cao
DOI: 10.1002/aenm.201803438
power generating windows, for which the
color property and transparency will have
to be tailored depending on the applica-
tion requirement.[5–8] Recent demonstra-
tion suggested that it is even possible to
combine the functions of power genera-
tion and power saving in semitransparent
polymer solar cells (ST-PSCs) by balancing
the key performance parameters including
efficiency, transmittance, and heat insu-
lating properties of the cells, providing a
promising application for building inte-
grated photovoltaics (BIPV).[9,10]
Another potential application of ST-
PSCs is for self-powered greenhouse, but
so far only very few attempts had been
made to explore their suitability for such
new application.[11,12] A successful inte-
gration of photovoltaic power generation
with greenhouse requires the facility to
maintain good agricultural activities. Pre-
liminary evaluation suggests that ST-PSCs
could be ideal candidate for greenhouse
applications as their transmission properties can be readily
tuned to provide incoming light with suitable wavelengths
for plant growth.[12] Green plants utilize specialized pigments
(chlorophyll a, chlorophyll b, carotenoid, etc.) to capture photon
energy for growth, and the crop growth depends closely on the
level of photosynthesis and photomorphogenesis, which are
carried out by the absorption and conversion of light energy
within the wavelength range of ≈400–700 nm.[13–15] It was also
demonstrated that ST-PSCs for greenhouse applications should
provide at least 10–50% transparency in the photosynthetically
active radiation (PAR) range to ensure sufficient agricultural
activities.[6] In order to achieve the balance between electricity
generation and optical transparency within the semitransparent
devices, the introduction of strong near-infrared light absorbing
materials with weak visible absorption becomes particularly
useful. Encouragingly, low-bandgap nonfullerene acceptors
(NFAs) fulfill such requirement and they were also shown to
be very efficient materials for high-performance PSCs,[16–19]
showing champion opaque cell efficiencies exceeding 14%.[20,21]
In addition, high-performance NFA-based ST-PSCs have also
been obtained by maximizing NIR light harvesting and bal-
ancing transmission and absorption for visible photons.[7,22–26]
For greenhouse application, the optical property of ST-PSCs
needs to be specially designed as the action spectrum for plants
is highly wavelength dependent, with the rate of photosynthesis
peaked at the red and blue parts of the spectrum. In previous
report, optical engineering strategy based on the introduction
In this study, a wavelength selective semitransparent polymer solar cell
(ST-PSC) with a proper transmission spectrum for plant growth is proposed
for greenhouse applications. A ternary strategy combining a wide bandgap
polymer donor with a near-infrared absorbing nonfullerene acceptor and
a high electron mobility fullerene acceptor is introduced to achieve PSCs
with power conversion efficiency (PCE) over 10%. The addition of PC71BM
into J52:IEICO-4F binary blend contributes to the suppressed trap-assisted
recombination, enhanced charge extraction, and improved open-circuit
voltage simultaneously. ST-PSC based on the J52:IEICO-4F:PC71BM ternary
blend shows an optimized performance with PCE of 7.75% and a defined
crop growth factor of 24.8%. Such high-performance ST-PSC is achieved by
carefully engineering the absorption spectrum of the light harvesting mate-
rials. As a result, the transmission spectra of the semitransparent devices
are well-matched with the absorption spectra of the photoreceptors, such as
chlorophylls, in green plants, which provides adequate lighting conditions
for photosynthesis and plant growth, and therefore making it a competitive
candidate for photovoltaic greenhouse applications.
Dr. H. Shi, R. Xia, Dr. G. Zhang, Prof. H.-L. Yip, Prof. Y. Cao
Institute of Polymer Optoelectronic Materials and Devices
State Key Laboratory of Luminescent Materials and Devices
South China University of Technology
Guangzhou 510640, P. R. China
E-mail: msangusyip@scut.edu.cn
Dr. G. Zhang, Prof. H.-L. Yip
Innovation Center for Printed Photovoltaics
South China Institute of Collaborative Innovation
Dongguan 523808, P. R. China
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/aenm.201803438.
Polymer Solar Cells
1. Introduction
Organic photovoltaics with benefits of being light weight,
mechanical flexible, and solution processible are considered as
one of the most promising photovoltaic technologies for future
energy generation.[1–4] Another unique property of organic
materials is their absorption can be easily tuned by molecular
design and their absorption spectrum can be tailored to have
a relatively localized absorbance at particular wavelengths. A
classic example is chlorophyll, which has a localized absorp-
tion for blue and red lights. Such versatility in spectral engi-
neering of organic materials offers new niche applications of
organic photovoltaics, such as semitransparent solar cells as
Adv. Energy Mater. 2018, 1803438

www.advenergymat.de
www.advancedsciencenews.com
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1803438 (2 of 8)
of photonic crystals to selectively enhance the transmission of
targeted wavelengths had been applied to ST-PSCs to improve
the utilization of incoming sunlight for plants.[11] In fact,
spectral engineering of light harvesting materials is a more
effective strategy to enhance the performance of ST-PSCs for
greenhouse applications due to the facile absorption spectrum
control through molecular design.[27,28] With careful design
or selection of photovoltaic materials for the ST-PSCs, we can
ensure a proper transmission of the portion of sunlight that is
relevant to photosynthesis and maximize the harvesting of light
that is unnecessary for crop growth to fulfill the requirements
of improving electricity generation and crop productivity at the
same time.
Toward these goals, herein, we fabricated ST-PSCs based
on an optimized ternary blend composed of a large bandgap
polymer donor (poly{4,8-bis[5-(2-ethylhexyl)thiophen-2-yl]
benzo[1,2-b:4,5-b′]dithiophene-5,5′-diyl-alt-4,7-bis(thien-2-yl)-
5,6-difluoro-2-(2-hexyldecyl)-2H-benzo[d][1,2,3]triazole}, J52)[29]
and two different acceptors, including a narrow-bandgap NFA
(2,2′-((2Z,2′Z)-(((4,4,9,9-tetrakis(4-hexylphenyl)-4,9-dihydro-
sindaceno[1,2-b:5,6-b′]-dithiophene-2,7-diyl)bis(4-((2-ethylhexyl)
oxy)thiophene-5,2-diyl))bis(methanylylide-ne))bis(5,6-difluoro-
3-oxo-2,3-dihydro-1H-indene-2,1-diylidene))dimalononitrile,
IEICO-4F)[30] and a fullerene acceptor (PC71BM). The polymer
donor and NFA were selected as their blend film shows a weak
absorption for blue and red light but an intense absorption
for green and NIR light, which is favorable for greenhouse
application. The introduction of PC71BM contributed simul-
taneously to the suppressed recombination, enhanced charge
extraction and transport, and improved open-circuit voltage.
Opaque devices with optimized blend achieved high PCE of
10.68%. ST-PSCs were then fabricated by replacing the thick
mirror Ag electrode with an ultrathin (10–20 nm) transparent
Ag electrode to fine-tune the overall property of the semitrans-
parent cells. As a result, an optimized performance with a
PCE of 7.75% and a defined crop growth factor of 24.8% was
achieved. The achievement of the decent value of crop growth
factor is attributed to the well matching of the transmittance
spectrum of the ST-PSCs to the absorption spectra of the main
photoreceptors, such as chlorophylls, in green plants.[31] To the
best of our knowledge, these performance parameters are the
best reported ones for greenhouse photovoltaic application, fur-
ther suggesting that spectral engineering is an effective strategy
to improve the performance of ST-PSCs that are tailored for
greenhouse applications.
2. Results and Discussion
The chemical structures and thin-film absorption spectra of
the light harvesting materials are shown in Figure 1. It can
be seen that J52 and PC71BM show main absorption in visible
range while IEICO-4F shows strong absorption in near-infrared
range. The lowest unoccupied molecular orbital (LUMO)
levels for J52, PC71BM, and IEICO-4F are −2.99, −3.90, and
−4.19 eV (Figure 1c), respectively. As the LUMO of PC71BM
located between J52 and IEICO-4F, cascade-type electron
transfer from J52 to IEICO-4F through PC71BM is energetically
favorable, which may facilitate more efficient charge separa-
tion.[32] We fabricated the devices with a conventional config-
uration of ITO/PEDOT:PSS/J52:IEICO-4F:PC71BM/PFN-Br/
Ag, and systematically optimized the weight ratio of the two
acceptors to achieve high-performance PSCs. Typical current
density–voltage (J–V) curves for the binary and ternary solar
cells are shown in Figure S1a (Supporting Information), and
the corresponding photovoltaic performance is summarized
Adv. Energy Mater. 2018, 1803438
Figure 1. a) Chemical structures of J52, IEICO-4F, and PC71BM. b) Optical absorption spectra of the solid thin films of J52, IEICO-4F, and PC71BM.
c) Energy level diagram for J52, IEICO-4F, and PC71BM extracted from the literature.

www.advenergymat.dewww.advancedsciencenews.com
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1803438 (3 of 8)
in Table 1. PSCs based on the reference J52:IEICO-4F binary
system showed a PCE of 9.21%, with an open-circuit voltage (Voc)
of 0.68 V, a short-circuit current density (Jsc) of 22.27 mA cm−2,
and a fill factor (FF) of 61.3%. While the J52:PC71BM binary
system exhibited a relatively large Voc and FF, the Jsc is much
lower due to the poor absorption of PC71BM. In the ternary
system, when 20 and 40 wt% of PC71BM with respect to the
total acceptor concentration were used, the Voc, Jsc, and FF
were all enhanced when compared to the J52:IEICO-4F binary
system, which led to an overall enhancement in PCE. The
optimal PC71BM content was identified to be 40 wt%, achieving
a prominent efficiency of 10.68%, with Voc, Jsc, and FF values of
0.698 V, 22.70 mA cm−2, and 67.4%, respectively. The increased
photocurrent density was also revealed by the external quantum
efficiency (EQE) data presented in Figure S1b (Supporting
Information). It is worth noting that the ternary blend shows
a broad composition tolerance for achieving high-performance
devices, with an even higher PC71BM content (60 wt%), the
PSC still produces a PCE of 9.73%, which is also superior to
that of the binary counterparts.
The charge carrier mobility of the binary and ternary
photo active films was evaluated by space-charge limited cur-
rent (SCLC) method,[33] which were carried out by fabricating
hole-only and electron-only devices with the architectures of
ITO/PEDOT:PSS/J52:IEICO-4F:PC71BM/MoO3/Ag and ITO/
ZnO/J52:IEICO-4F:PC71BM/PFN Br/Ag, respectively. As shown
in Table 2, the hole mobilities for all the blends maintained at
the same order of magnitude of about 10−3 cm2 V−1 s−1, while the
electron mobilities changed with the weight ratio of the accep-
tors. The hole and electron mobilities for J52:IEICO-4F binary
blend are calculated to be 1.42 × 10−3 and 9.36 × 10−5 cm2 V−1 s−1,
respectively, indicating a relatively imbalanced charge carrier
mobility and leading to a low FF of 61.3%. The J52:PC71BM
blend exhibited a hole/electron mobility ratio of 0.85, showing
the most balanced charge carrier mobility with the highest FF
of 69.7%. In the ternary blends, the electron mobilities were
increased with the PC71BM content, and the difference between
the hole mobility and electron mobility therefore decreased
gradually, suggesting more balanced charge carrier transport,
which is beneficial for enhancing the FF values.[34]
The analysis of the dependency of Jsc and Voc on light inten-
sity (Plight) was also performed to study the predominant recom-
bination processes in different cases. In principle, a power-law
dependence of Jsc upon light intensity can be expressed as
Jsc ∝ (Plight)S in organic solar cells, where S is the exponential
factor, and weak bimolecular recombination in the device would
result in a S value close to unity.[35,36] Figure 2a depicts Jsc as
a function of light intensity varying from 10 to 100 mW cm−2
for the binary and optimized ternary devices. The J52:PC71BM
device exhibits a S value of 0.995, which is very close to unity,
indicating very weak bimolecular recombination in the device.
The J52:IEICO-4F device shows the S value of 0.962, while
the value is increased to 0.986 for the J52:IEICO-4F:PC71BM
(1:0.9:0.6) ternary device, suggesting less bimolecular recombi-
nation in the optimized ternary PSCs, which agrees with the
enhanced FF and Jsc.
Figure 2b shows the relationship between Voc and Plight. The
slope of Voc versus the natural logarithm of Plight determines
the degree of trap-assisted recombination in the devices.[36,37]
A slope close to kBT/q indicates the dominating mechanism of
bimolecular recombination, while the slope of 2kBT/q stands
for trap-assisted recombination, where kB, T, and q are the
Boltzmann constant, temperature, and elementary charge,
respectively. The binary devices based on J52:IEICO-4F and
J52:PC71BM showed slopes of 1.48 and 1.80 kBT/q, respec-
tively, while the ternary J52:IEICO-4F:PC71BM device attained
a much smaller slope of 1.22 kBT/q. The results suggest that
the trap density is reduced and therefore led to an effective
suppression in trap-assisted recombination in the ternary
devices, which could also contribute to the enhanced current
density.
To investigate the exciton dissociation and charge collec-
tion dynamics for the binary and optimized ternary PSCs, the
plots of the photocurrent density versus the effective voltage
(Jph–Veff) were measured, as depicted in Figure 2c. The Jph can
be defined as Jph = JL − JD, where JL and JD represent the cur-
rent density under illumination and dark conditions, respec-
tively. The Veff is determined by V0 − Va, where V0 is the voltage
at Jph = 0, and Va is the applied bias voltage.[38] It is generally
assumed that all photogenerated excitons are dissociated into
free charge carriers and all photogenerated charge carriers can
be adequately collected by the electrodes at high Veff (2 V in this
case). Thus, Jsat is only limited by the maximum exciton genera-
tion rate (Gmax), which is defined as Jsat = qLGmax, where q is
elementary charge and L is the active layer thickness. The cal-
culated Jph, Jsat, and Gmax values of the PSCs are summarized in
Table S1 (Supporting Information). The Gmax for the J52:IEICO-
4F:PC71BM (1:0.9:0.6) ternary device was 1.47 × 1028 m−3 s−1
(Jsat = 23.53 mA cm−2), while those of the J52:IEICO-4F and
J52:PC71BM binary devices were 1.43 × 1028 m−3 s−1 (Jsat =
22.92 mA cm−2) and 6.88 × 1027 m−3 s−1 (Jsat = 11.01 mA cm−2),
respectively, which suggested that light absorption was
enhanced in the ternary blend.
Adv. Energy Mater. 2018, 1803438
Table 1. Photovoltaic parameters of the conventional opaque devices
with different weight ratios of PC71BM.
Ratio PC71BM
content
Voc [V] Jsc [mA cm−2]FF [%] PCE [%]
1:1.5:0 0% 0.675 22.27 61.3 9.21 (9.14)a)
1:1.2:0.3 20% 0.690 22.60 65.7 10.26 (10.16)
1:0.9:0.6 40% 0.698 22.70 67.4 10.68 (10.54)
1:0.6:0.9 60% 0.713 19.67 69.3 9.73 (9.50)
1:0:1.5 100% 0.723 10.15 69.7 5.12 (5.03)
a)Numbers in parentheses indicate the average values from ten devices.
Table 2 . Hole and electron mobilities of J52:IEICO-4F:PC71BM films with
different weight ratios of PC71BM.
Ratio PC71BM content [%]
μ
h [cm2 V−1 s−1]
μ
e [cm2 V−1 s−1]
μ
h/
μ
e
1:1.5:0 0 1.42 × 10−29.36 × 10−515
1:1.2:0.3 20 1.56 × 10−38.22 × 10−41.9
1:0.9:0.6 40 1.41 × 10−39.35 × 10−41.5
1:0.6:0.9 60 1.51 × 10−31.12 × 10−31.3
1:0:1.5 100 1.66 × 10−31.95 × 10−30.85

www.advenergymat.de
www.advancedsciencenews.com
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1803438 (4 of 8)
To take one step further, we also compared the exciton dis-
sociation probabilities [P(E,T)] in different blends (see Table S1,
Supporting Information), and the value of P(E,T) is determined
by normalizing Jph with Jsat (Jph/Jsat). Under short-circuit con-
ditions, a high P(E,T) of 98.1% was achieved for the optimal
J52:IEICO-4F:PC71BM (1:0.9:0.6) ternary device, which is
higher than those of the J52:IEICO-4F-based device (95.2%)
and the J52:PC71BM-based device (93.4%), indicating the addi-
tion of PC71BM could also facilitate more efficient dissociation
of excitons into free carriers. These results indicate that the
optimized ternary device offers more efficient exciton dissocia-
tion and charge extraction, therefore led to the much enhanced
device efficiency.
In addition to the demand of high efficiency, stability is
also a critical requirement for practical application of PSCs.[39]
Therefore, we also fabricated PSCs based on J52:IEICO-4F
binary system and J52:IEICO-4F:PC71BM (1:0.9:0.6) ternary
system to investigate the influence of PC71BM incorporation on
the photostability of the devices. The stability of encapsulated
devices was tested under continuously light-soaked using white
light-emitting diodes (LEDs) irradiation under ambient con-
dition. The J–V characteristics of both the binary and ternary
devices were probed periodically, and the normalized PCE data
are shown in Figure 2d. The efficiency of J52:IEICO-4F-based
device decreased by about 50% after 100 h of light exposure,
while the J52:IEICO-4F:PC71BM-based ternary device main-
tained more than 80% of its initial PCE value, exhibiting much
better photostability. These combined merits of high efficiency
and good stability of the ternary system further triggered us to
study the performance of the corresponding ST-PSCs.
The J52:IEICO-4F:PC71BM (1:0.9:0.6) ternary ST-PSCs were
constructed by replacing the opaque Ag electrode with an
ultrathin Ag electrode as the top transparent electrode. The
photovoltaic parameters and transmittance properties of the
semitransparent devices with different electrode thicknesses
are summarized in Table 3. In order to quantitatively assess
the impact of semitransparent devices on plant growth, we
employed a crop growth factor (G) to define the light transmit-
tance properties of the ST-PSCs, which can be calculated by the
following equation
λλ
λλ
λλλ
() ()()
()()
=∫
∫
d
d
s
s
GTb a
ba
(1)
where T(
λ
) is the transmission of the semitransparent solar cell,
bs(
λ
) is AM 1.5 solar spectral irradiance, and a(
λ
) represents the
plant action spectrum, which is obtained from the averaged
action spectrum of 27 herbaceous plants, as exhibited in Figure S2
(Supporting Information).[40] The rate of photo synthesis in
a crop is governed by the integral of the solar spectrum and
the plant action spectrum, as seen in the denominator of
Equation (1). Thus, the crop growth factor represents the ratio
of the rate of photosynthesis through the semitransparent solar
cells with spectrally dependent transparency under a clear sky.
To further evaluate the relationship between the crop growth
factor and the efficiency of the J52:IEICO-4F:PC71BM (1:0.9:0.6)
ST-PSCs, we achieved a gradient change of the transmittance
of the cells by linearly increasing the thickness of the top Ag
electrode from 10 to 20 nm, while keeping the light harvesting
layer thickness of 100 nm. As summarized in Table 3, the Jsc
Adv. Energy Mater. 2018, 1803438
Figure 2. a) JSC as a function of light intensity for the binary and ternary PSCs. b) VOC as a function of light intensity for the binary and ternary PSCs.
c) Photocurrent density (Jph) versus effective voltage (Veffect) curves for the binary and ternary PSCs. d) Normalized PCEs of encapsulated J52:IEICO-4F
and J52:IEICO-4F:PC71BM devices with ≈100 h of light exposure (100 mW cm−2) at ambient condition.

www.advenergymat.dewww.advancedsciencenews.com
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1803438 (5 of 8)
decreases and the transmittance increases with decreasing the
thickness of the ultrathin Ag cathode because of the decrease
in reflectivity of thinner Ag film, which reflects less light back
to the light harvesting layer but allowing more light to pass
through the device.[41] We also observed that the FFs are slightly
decreased with decreasing Ag thickness due to the increase
in series resistance (Rs) of the devices, originated from the
reduced conductivity of thinner Ag cathode.[42] As a result, a
higher PCE of 8.83% with a lower G of 20.1% and a lower PCE
of 5.93% with a higher G of 28.8% were achieved for the ST-
PSCs with a 20 and 10 nm Ag top electrode, respectively. The
ST-PSC with Ag thickness of 15 nm is particularly of interest
because of the balanced performance with a crop growth factor
of 24.8% and a corresponding PCE of 7.75%. Therefore, we
also fabricated J52:IEICO-4F binary and J52:IEICO-4F:PC71BM
(1:1.2:0.3) ternary system based ST-PSCs with Ag thickness of
15 nm for comparison and both of these reference cells showed
inferior efficiency and transmittance compared to that of the
J52:IEICO-4F:PC71BM (1:0.9:0.6) ST-PSC having the same elec-
trode thickness. The EQE spectra and transmittance curves of
the three different ST-PSCs are shown in Figure 3. In all the
cases, due to the weak absorption of the blended films for blue
and red lights, the EQE curves show two valleys located at ≈400
and ≈650 nm, while the transmittance spectra show the corre-
sponding peaks at those wavelengths.
In Figure 4, the transmittance curve of the J52:IEICO-
4F:PC71BM (1:0.9:0.6) ST-PSC is plotted together with the
absorption spectra of the main photoreceptors such as chloro-
phylls and carotenoid in green plants for comparison, showing
a very well-matched transmittance with the characteristic
absorption peaks of plants. For a better evaluation of the suit-
ability of our spectrally engineered ST-PSCs for greenhouse
applications, we also compares the crop growth factor G to
the transmittance parameter, visible light transmittance (VLT),
which is traditionally used to evaluate optical property of ST-
PSCs for BIPV application. VLT corresponds to the integrated
transmission spectrum of a semitransparent device over the
whole visible spectrum, weighted by the spectral response
of a typical human eye and the AM 1.5 solar spectral irradi-
ance.[43] Therefore, as opposed to the crop growth factor, visible
light transmittance is weighted significantly at green light and
weakly weighted at both blue and red lights. As a result, the
visible light transmittance values are ≈20% lower than the cor-
responding crop growth factors, indicating that the ST-PSCs
demonstrated here are much better satisfied for the conditions
needed for plant growth instead of for human vision, therefore
suitable for greenhouse applications. The color coordinates of
the representative ST-PSCs are also displayed in the CIE 1931
chromaticity diagram in Figure 5a. The color coordinates of
the J52:IEICO-4F:PC71BM (1:0.9:0.6) ST-PSCs with ultrathin
Ag electrodes of 15 and 10 nm are located at (0.317, 0.283) and
(0.312, 0.281), respectively, which were calculated from the
transmission spectra of ST-PSCs. In addition, Figure 5b shows
a photo that was taken through the ST-PSC based on the 15 nm
Ag electrode. The photo appears as purplish due to the combi-
nation of blue and red transmitted lights, which are useful for
photosynthesis in plants.
Finally, we also provide a further analysis of the transmit-
tance of the ST-PSCs weighted by the absorption spectra of the
three most commonly found photoreceptors in green plants,
Adv. Energy Mater. 2018, 1803438
Table 3. Photovoltaic parameters, crop growth factor (G), and visible light transmittance (VLT) of the ternary ST-PSCs with various Ag top electrode
thicknesses and different weight ratios of PC71BM.
Ag Ratio Voc [V] Jsc [mA cm−2]FF [%] PCE [%] G [%] VLT [%]
20 nm 1:0.9:0.6 0.690 19.04 67.2 8.83 (8.58)a) 20.1 15.8
15 nm 1:0.9:0.6 0.685 16.90 66.9 7.75 (7.68) 24.8 19.9
10 nm 1:0.9:0.6 0.689 13.23 65.0 5.93 (5.81) 28.8 23.7
15 nm 1:1.2:0.3 0.680 17.25 65.4 7.68 (7.09) 23.7 19.0
15 nm 1:1.5:0 0.671 16.92 62.7 7.12 (7.55) 23.8 18.8
a)Numbers in parentheses indicate the average values from ten devices.
Figure 3. a) EQE spectra and b) transmission spectra of different ST-PSCs composed of a 15 nm Ag film as transparent cathode.

www.advenergymat.de
www.advancedsciencenews.com
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1803438 (6 of 8)
including chlorophyll a, chlorophyll b, and carotenoid. When
substituting the plant action spectrum in Equation (1) by the
absorption spectra of chlorophyll a, chlorophyll b, and carot-
enoid, the corresponding transmittances of the J52:IEICO-
4F:PC71BM (1:0.9:0.6) based ST-PSC were calculated to be
27.7, 28.3, and 26.9%, respectively, with an average value of
27.6%. This value is slightly higher than that defined by the
crop growth factor (24.8%), which was weighted by an averaged
action spectrum of 27 different types of plants.[40] As the use of
semitransparent solar cells for greenhouse photovoltaics is still
just an emerging application, it still lacked of a consensus on
choosing the best evaluation factor for the technology. There-
fore, we believed that both the analysis using crop growth factor
and averaging absorption spectra of key photoreceptors in
plants are valuable references to the research community.
3. Conclusion
In summary, we have developed a high-performance, wave-
length selective semitransparent polymer solar cell with a
proper transmission spectrum for plant growth. J52:IEICO-
4F:PC71BM ternary PSCs were fabricated by combining the
merits of broad absorption of the narrow-bandgap nonfullerene
acceptor and high electron mobility of the fullerene acceptor.
The suppressed recombination, enhanced charge extraction,
and improved open-circuit voltage derived from the incorpora-
tion of the third component PC71BM have improved the effi-
ciency of the ternary PSCs reaching up to 10.68%. As a result,
the corresponding semitransparent ternary solar cells achieved
an optimized PCE of 7.75% together with a crop growth factor
of 24.8%. In addition, the semitransparent devices exhibited
well-matched transmission spectra with the absorption spectra
of the main photoreceptors such as chlorophylls in green
plants, providing efficient light energy for photosynthesis. The
spectral engineering concept presented in this work is expected
to provide important guidelines for designing high-perfor-
mance ST-PSCs, which paves the way for future applications
of semitransparent photovoltaic technology for self-powered
greenhouses.
4. Experimental Section
Materials: J52 and IEICO-4F were purchased from Solarmer Materials
Inc. PC71BM was purchased from Solenne B. V. Chlorobenzene (CB)
Adv. Energy Mater. 2018, 1803438
Figure 4. Optical absorption spectra of chlorophyll a, chlorophyll b, and
carotenoid, and the transmission spectra of the J52:IEICO-4F:PC71BM
(1:0.9:0.6) ST-PSC with 15 nm Ag film as transparent cathode.
Figure 5. a) The CIE 1931xyY chromaticity diagram showing the color coordinates of the representative ST-PSCs. b) Outdoor photo taken through the
ST-PSC with crop growth factor of 24.8%.

www.advenergymat.dewww.advancedsciencenews.com
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1803438 (7 of 8)
Adv. Energy Mater. 2018, 1803438
and 1,8-diiodooctane (DIO) were obtained from Sigma-Aldrich Inc. All
reagents and solvents were used directly as received.
Devices Fabrication: A PEDOT:PSS layer (≈40 nm) was first spin-
coated onto the precleaned ITO substrate and dried at 140 °C for
20 min in air. The substrates were then transferred into a nitrogen-filled
glove box. J52:IEICO-4F and J52:PC71BM binary blend, and J52:IEICO-
4F:PC71BM ternary blend with various weight ratios in CB:DIO (99:1
volume ratio) solution were spin-coated to form active layers with
thickness of ≈100 nm, followed by thermal annealing at 100 °C for
10 min. Subsequently, a 5 nm thick film of PFN-Br was spin-coated as
the electron transport layer. Finally, a silver electrode (100 nm for opaque
devices and 10–20 nm for semitransparent devices) was deposited by
thermal evaporation through a shadow mask under a base pressure
of 10−7 mbar.
Device Characterization: The current density–voltage (J–V) curves were
measured by using a Keithley 2400 source meter, and the photocurrent
was obtained using an AM 1.5 G solar simulator (Taiwan, Enlitech)
with the light intensity of 100 mW cm−2 calibrated by a standard silicon
solar cell. A nonrefractive mask was put in close contact and aligned
with the solar cells to define the active area to be 0.04 cm2. The EQE
spectra were performed on a commercial EQE measurement system
(Taiwan, Enlitech, QE-R3011). The optical absorption spectra of the
thin film and the transmission spectra of semitransparent devices were
recorded with the TMS-I instrument (Guangdong Jinuosh Technology
Co., Ltd.). The experimental transmission of each device is folded with
the AM1.5 spectrum to obtain the perceived transmission under solar
illumination, and the resulting data are coupled with the CIE 1931 2°
standard observer color matching functions to obtain the corresponding
xyY points.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
This work was financially supported by the Ministry of Science and
Technology (No. 2017YF0206600), the Science and Technology Program
of Guangzhou, China (Nos. 201607020010 and 2017A050503002), the
Natural Science Foundation of China (Nos. 21761132001 and 91633301),
and the China Postdoctoral Science Foundation (No. 2017M622681).
Conflict of Interest
The authors declare no conflict of interest.
Keywords
crop growth factor, photovoltaic greenhouse, semitransparent polymer
solar cells, spectral engineering
Received: November 7, 2018
Published online:
[1] L. Lu, T. Zheng, Q. Wu, A. M. Schneider, D. Zhao, L. Yu, Chem. Rev.
2015, 115, 12666.
[2] G. Li, R. Zhu, Y. Yang, Nat. Photonics 2012, 6, 153.
[3] G. Yu, J. Gao, J. C. Hummelen, F. Wudl, A. J. Heeger, Science 1995,
270, 1789.
[4] L. Meng, Y. Zhang, X. Wan, C. Li, X. Zhang, Y. Wang, X. Ke, Z. Xiao,
L. Ding, R. Xia, H.-L. Yip, Y. Cao, Y. Chen, Science 2018, 361, 1094.
[5] C. J. Traverse, R. Pandey, M. C. Barr, R. R. Lunt, Nat. Energy 2017,
2, 849.
[6] S.-Y. Chang, P. Cheng, G. Li, Y. Yang, Joule 2018, 2, 1039.
[7] S. Dai, X. Zhan, Adv. Energy Mater. 2018, 8, 1800002.
[8] Q. Xue, R. Xia, C. J. Brabec, H.-L. Yip, Energy Environ. Sci. 2018, 11,
1688.
[9] K. Forberich, F. Guo, C. Bronnbauer, C. J. Brabec, Energy Technol.
2015, 3, 1051.
[10] C. Sun, R. Xia, H. Shi, H. Yao, X. Liu, J. Hou, F. Huang, H.-L. Yip,
Y. Cao, Joule 2018, 2, 1.
[11] F. Yang, Y. Zhang, Y. Hao, Y. Cui, W. Wang, T. Ji, F. Shi, B. Wei, Appl.
Opt. 2015, 54, 10232.
[12] C. J. M. Emmott, J. A. Röhr, M. Campoy-Quiles, T. Kirchartz,
A. Urbina, N. J. Ekins-Daukes, J. Nelson, Energy Environ. Sci. 2015,
8, 1317.
[13] D. M. Gates, H. J. Keegan, J. C. Schleter, V. R. Weidner, Appl. Opt.
1965, 4, 11.
[14] M. Bernardi, J. C. Grossman, J. Phys. Chem. C 2013, 117,
26896.
[15] T. Wang, G. Wu, J. Chen, P. Cui, Z. Chen, Y. Yan, Y. Zhang, M. Li,
D. Niu, B. Li, H. Chen, Renewable Sustainable Energy Rev. 2017, 70,
1178.
[16] P. Cheng, G. Li, X. Zhan, Y. Yang, Nat. Photonics 2018, 12, 131.
[17] C. Yan, S. Barlow, Z. Wang, H. Yan, A. K. Y. Jen, S. R. Marder,
X. Zhan, Nat. Rev. Mater. 2018, 3, 18003.
[18] J. Hou, O. Inganas, R. H. Friend, F. Gao, Nat. Mater. 2018,
17, 119.
[19] X. Song, N. Gasparini, L. Ye, H. Yao, J. Hou, H. Ade, D. Baran, ACS
Energy Lett. 2018, 3, 669.
[20] H. Zhang, H. Yao, J. Hou, J. Zhu, J. Zhang, W. Li, R. Yu, B. Gao,
S. Zhang, J. Hou, Adv. Mater. 2018, 30, 1800613.
[21] Z. Xiao, X. Jia, L. Ding, Sci. Bull. 2017, 62, 1562.
[22] Y. Cui, C. Yang, H. Yao, J. Zhu, Y. Wang, G. Jia, F. Gao, J. Hou, Adv.
Mater. 2017, 29, 1703080.
[23] F. Liu, Z. Zhou, C. Zhang, J. Zhang, Q. Hu, T. Vergote, F. Liu,
T. P. Russell, X. Zhu, Adv. Mater. 2017, 29, 1606574.
[24] Y. Li, J. D. Lin, X. Che, Y. Qu, F. Liu, L. S. Liao, S. R. Forrest, J. Am.
Chem. Soc. 2017, 139, 17114.
[25] X. Shi, J. Chen, K. Gao, L. Zuo, Z. Yao, F. Liu, J. Tang, A. K. Y. Jen,
Adv. Energy Mater. 2018, 8, 1702831.
[26] X. Ma, Z. Xiao, Q. An, M. Zhang, Z. Hu, J. Wang, L. Ding, F. Zhang,
J. Mater. Chem. A 2018, 6, 21485.
[27] M. E. Loik, S. A. Carter, G. Alers, C. E. Wade, D. Shugar, C. Corrado,
D. Jokerst, C. Kitayama, Earth’s Future 2017, 5, 1044.
[28] C. S. Allardyce, C. Fankhauser, S. M. Zakeeruddin, M. Grätzel,
P. J. Dyson, Sol. Energy 2017, 155, 517.
[29] H. Bin, Z. G. Zhang, L. Gao, S. Chen, L. Zhong, L. Xue, C. Yang,
Y. Li, J. Am. Chem. Soc. 2016, 138, 4657.
[30] H. Yao, Y. Cui, R. Yu, B. Gao, H. Zhang, J. Hou, Angew. Chem., Int.
Ed. 2017, 56, 3045.
[31] D. Singh, C. Basu, M. Meinhardt-Wollweber, B. Roth, Renewable
Sustainable Energy Rev. 2015, 49, 139.
[32] L. Lu, M. A. Kelly, W. You, L. Yu, Nat. Photonics 2015, 9, 491.
[33] C. Melzer, E. J. Koop, V. D. Mihailetchi, P. W. M. Blom, Adv. Funct.
Mater. 2004, 14, 865.
[34] R. Yu, H. Yao, J. Hou, Adv. Energy Mater. 2018, 8, 1702814.
[35] I. Riedel, J. Parisi, V. Dyakonov, L. Lutsen, D. Vanderzande,
J. C. Hummelen, Adv. Funct. Mater. 2004, 14, 38.
[36] S. R. Cowan, A. Roy, A. J. Heeger, Phys. Rev. B 2010, 82, 245207.
[37] L. J. A. Koster, V. D. Mihailetchi, R. Ramaker, P. W. M. Blom, Appl.
Phys. Lett. 2005, 86, 123509.
[38] L. Lu, T. Xu, W. Chen, E. S. Landry, L. Yu, Nat. Photonics 2014, 8,
716.

www.advenergymat.de
www.advancedsciencenews.com
© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1803438 (8 of 8)
Adv. Energy Mater. 2018, 1803438
[39] 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.
[40] K. Inada, Plant Cell Physiol. 1976, 17, 355.
[41] K.-S. Chen, J.-F. Salinas, H.-L. Yip, L. Huo, J. Hou, A. K. Y. Jen,
Energy Environ. Sci. 2012, 5, 9551.
[42] H. Shi, R. Xia, C. Sun, J. Xiao, Z. Wu, F. Huang, H.-L. Yip, Y. Cao,
Adv. Energy Mater. 2017, 7, 1701121.
[43] R. Betancur, P. Romero-Gomez, A. Martinez-Otero, X. Elias,
M. Maymó, J. Martorell, Nat. Photonics 2013, 7, 995.

Copyright WILEY-VCH Verlag GmbH & Co. KGaA, 69469 Weinheim, Germany, 2018.
Supporting Information
for Adv. Energy Mater., DOI: 10.1002/aenm.201803438
Spectral Engineering of Semitransparent Polymer Solar Cells
for Greenhouse Applications
Hui Shi, Ruoxi Xia, Guichuan Zhang, Hin-Lap Yip,* and Yong
Cao

1
Supporting Information
Spectral Engineering of Semitransparent Polymer Solar Cells for
Greenhouse Applications
Hui Shi, Ruoxi Xia, Guichuan Zhang, Hin-Lap Yip,* Yong Cao
Dr. H. Shi, R. Xia, Dr. G. Zhang, Prof. H.-L. Yip, Prof. Y. Cao
Institute of Polymer Optoelectronic Materials and Devices, State Key Laboratory of
Luminescent Materials and Devices, South China University of Technology,
Guangzhou 510640, P. R. China
Dr. G. Zhang, Prof. H.-L. Yip
Innovation Center for Printed Photovoltaics, South China Institute of Collaborative
Innovation, Dongguan 523808, P. R. China
E-mail: msangusyip@scut.edu.cn
Figure S1. a) J–V curves and b) EQE spectra of PSCs with different weight ratios of
PC71BM.

2
Figure S2. Relative action spectrum for plants and AM1.5 solar spectrum.
Table S1. Exciton dissociation probabilities P(E,T) and maximum exciton generation rate
(Gmax) of the binary and optimized ternary solar cells.
Ratio
Jsat (mA cm-2)
Jph (mA cm-2)
P(E,T)
Gmax (m-3 s-1)
1:1.5:0
22.92
21.84
95.2%
1.43 × 1028
1:0.9:0.6
23.53
23.10
98.1%
1.47 × 1028
1:0:1.5
11.01
10.28
93.4%
6.88 × 1027