Article A Layer-by-Layer Architecture for Printable Organic Solar Cells Overcoming the Scaling Lag of Module Efficiency The layer-by-layer (LbL) strategy exhibits unique advantages of combining the merits of high photo-absorption rate, suitable vertical phase separation, and good practicability, endowing the LbL-bladed devices with a higher power conversion efficiency (PCE) of 16.35% compared to the bulk heterojunction (BHJ)-bladed device (15.37%). Importantly, this LbL approach can effectively reduce the scaling lag of module efficiency. An LbL-bladed solar module with a geometrical fill factor of over 90% exhibited an outstanding PCE of 11.86%. Rui Sun, Qiang Wu, Jie Guo, ..., Fei Huang, Yongfang Li, Jie Min [email protected]HIGHLIGHTS A layer-by-layer (LbL)-bladed OSC shows a PCE of 16.35% LbL approach can also be employed for optimizing other photovoltaic systems An LbL-bladed solar module exhibits a PCE of 11.86% LbL strategy can effectively reduce the scaling gap of module efficiency Sun et al., Joule 4, 407–419 February 19, 2020 ª 2019 Elsevier Inc. https://doi.org/10.1016/j.joule.2019.12.004
38
Embed
A Layer-by-Layer Architecture for Printable Organic Solar ... · Article A Layer-by-Layer Architecture for Printable Organic Solar Cells Overcoming the Scaling Lag of Module Efficiency
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Article
A Layer-by-Layer Architecture for PrintableOrganic Solar Cells Overcoming the ScalingLag of Module Efficiency
A Layer-by-Layer Architecture forPrintable Organic Solar Cells Overcomingthe Scaling Lag of Module EfficiencyRui Sun,1 Qiang Wu,1 Jie Guo,1 Tao Wang,1 Yao Wu,1 Beibei Qiu,2 Zhenghui Luo,3 Wenyan Yang,1
Zhicheng Hu,4 Jing Guo,1 Mumin Shi,1 Chuluo Yang,3 Fei Huang,4 Yongfang Li,2 and Jie Min1,2,5,6,*
Context & Scale
With the improvement of
photovoltaic efficiencies,
solution-processed organic solar
cells (OSCs) have shown a bright
prospect for inexpensive and
sustainable light-to-energy
conversion. However, when we
adopt the donor-acceptor bulk
heterojunction (BHJ) strategy to
fabricated large-scale OSC
modules, there is a huge gap in
efficiency. In this work, we
introduced an alternative method
layer-by-layer (LbL) approach into
SUMMARY
To date, organic solar cells (OSCs) with the development of photovoltaic ma-
terials have realized high power conversion efficiencies (PCEs) through the
solution processing strategy with bulk heterojunction (BHJ) structure, but
the BHJ morphology is difficult to control in large-scale fabrication of OSCs.
Herein, we report an alternative film-forming technology known as layer-by-
layer (LbL). As compared to its BHJ counterpart, LbL presents many unique
advantages including controllable ‘‘p-i-n’’ morphology, good charge transport
and extraction properties, and great universality. By using the LbL-bladed
coating strategy, a high PCE of 16.35% was achieved in the PM6:Y6 OSCs.
Notably, a large-area solar module of 11.52 cm2 with a geometrical fill factor
of over 90% exhibited an outstanding PCE of 11.86%, which represents the
highest efficiency of large-area solar modules. The results may pave the way
for the fabrication of the photoactive layer in the future industrial production
of OSCs.
the fabrication of OSCs through
blading-coating to obtain higher
photovoltaic performance as
compared to its BHJ counterpart.
In addition, we found the LbL
coating is a successful and general
processing technology that can
quickly bridge the huge gap
between the ‘‘hero’’ lab-scale
produced solar cells and large-
area solar modules.
INTRODUCTION
Solution-processed organic solar cells (OSCs) have been regarded as one of next-
generation photovoltaics owing to their key advantages such as light-weight,
low cost solution processing, and easy fabrication of flexible and semitransparent
devices.1–6 During the past decade, continuous efforts have been devoted to the
development of OSCs, including synthesis of numerous donor and acceptor
photovoltaic materials,4,7–10 morphology control strategies,11,12 and development
of device engineering techniques.13–16 With these tremendous exertions, power
conversion efficiency (PCE) has consequently increased up to 16% for polymer-
non-fullerene single OSCs.4,9 Although the efficiency of OSCs is now high
enough for commercial applications,5,17 there are still certain obstacles that
should be overcome to enter the industrial application in the near future. On
one hand, almost all high efficient OSCs and small-area OSC modules have
been fabricated by spin-coating method so far, which cannot be transferred to
high-throughput roll-to-roll (R2R) manufacturing for competing with other photo-
voltaic technologies.5 On the other hand, it is well known and understood, that
many up-scaling coating technologies cannot easily bridge the huge efficiency
gap between the ‘‘hero’’ lab-scale cells and large-area modules, while reduce
the geometric fill factor (GFF) losses.17 Thus, it is urgent to demonstrate that
high-performance large-area OSCs and OSC modules can be fabricated by scal-
able printing strategies under optimized conditions for standing out their great
potential applications.
Joule 4, 407–419, February 19, 2020 ª 2019 Elsevier Inc. 407
1The Institute for Advanced Studies, WuhanUniversity, Wuhan 430072, China
2Beijing National Laboratory for MolecularSciences, Beijing 100190, China
3Hubei Key Lab on Organic and PolymericOptoelectronic Materials, Department ofChemistry, Wuhan University, Wuhan 430072,P. R. China
4Institute of Polymer Optoelectronic Materialsand Devices, State Key Laboratory ofLuminescent Materials and Devices, South ChinaUniversity of Technology, Guangzhou 510640,People’s Republic of China
5Key Laboratory of Materials Processing andMold (Zhengzhou University), Ministry ofEducation, Zhengzhou 450002, China
2,7-diyl]-2,5-thiophenediyl]))/Ag. Of note is that the active layer and the interface
layers, including PEDOT:PSS and PNDIT-F3N-Br, are all fabricated by blade-
coating techniques in air (Figure 1A).
Joule 4, 407–419, February 19, 2020 409
Figure 1C exhibits the optical properties of the bladed BHJ and LbL active layers. As
compared to the BHJ active layer, the LbL blend shows a slightly red-shifted absorp-
tion spectra and a remarkably higher absorption coefficient of approximately 8.33
104 cm�1, indicating the enhanced molecular ordering of Y6 acceptors. The high ab-
sorption coefficient of the LbL blend also suggests that a higher number of photons
can be absorbed and converted into energy.41 Furthermore, the difference of optical
absorption drove us to employ spectroscopic ellipsometry to determine accurate
optical constants (n and k) of the BHJ and LbL blends (Figure S1). The results were
calculated from the used films with different thicknesses (Figure S2). Figures 1D
and 1E show the field distribution for the wavelength range of 300–900 nm for visu-
alizing the modulation of the electric field inside BHJ and LbL devices with the
bladed active layer thickness of 120 nm, respectively. As compared to the BHJ
blend, the photon absorption rate profile of LbL is slightly stronger. Notably, those
images can be viewed as the charge generation profiles within the BHJ and LbL de-
vices because most of excitions can be effectively separated into free charges,
demonstrated by the photoluminescence (PL) measurements (Figure S3). It should
be noted that the simulated photo-absorption rates of BHJ and LbL devices are
different with the absorption spectra of BHJ and LbL blends, probably resulting
from the complexities in the absorption, transmission, refraction, and reflection of
solar photons in devices. In addition, the morphology complexity of the LbL active
layer as well as the same layer thickness set in the simulation also lead to the
above-mentioned difference. Undoubtedly, the above-mentioned optical absorp-
tion profiles indicate the distinctly different three-dimensional microstructures of
BHJ and LbL blends as described in the following section.
To clarify the morphological characteristics of the BHJ- and LbL-bladed PM6:Y6
active layers, we applied grazing-incidence wide-angle X-ray scattering (GIWAXS),
photo-induced force microscopy (PiFM) and time-of-flight secondary ion mass spec-
trometry (ToF-SIMS) measurements. From the two-dimensional (2D) GIWAXS in Fig-
ure S4, the neat PM6 film shows a disordered arrangement implying a less ordered
nature of the film, in contrast the neat Y6 film is revealed to adopt a face-on prefer-
ential orientation with respect to the substrate. Notably, the 2D GIWAXS measure-
ments (Figures S4C and S4D) did not reveal distinctly different scattering patterns of
the BHJ and LbL blends acquired at the critical incident angle of 0.13�. Despite this,
the shallow incidence angle of 0.02� was further chosen to investigate the crystalli-
zation close to the top surface of the BHJ and LbL films, as exhibited in Figures 2A
and 2B, respectively. It was found that Y6 acceptor is much more ordered at the
top of LbL film than that of the BHJ blend, as evidenced by such diffraction (Fig-
ure 2C). The mean size of the crystallites was acquired by calculating the crystal
coherence length (CCL) using the Scherrer equation.45 The CCL values of BHJ and
LbL films are 24.5 and 25.1 nm, respectively, indicating the existence of ordered
Y6 top layers in the LbL blend. TheGIWAXS results are consistent with the PiFMmea-
surements. The PiFM images at the characteristic Fourier transform infrared (FTIR)
wavelengths corresponding to absorption peaks of Y6 acceptor (1,536 cm�1) with
BHJ and LbL blading approaches are shown in Figures 2D and 2E, respectively.
The LbL-bladed blend shows the enhanced molecular aggregation and increased
domain size in comparison with the BHJ film. The observed crystallization phenom-
ena explained in terms of the dynamical patterns for the corresponding BHJ and LbL
morphologies were derived from the competing processes between molecular
interdiffusion and molecular aggregation during the film formation process.43
We further conducted TOF-SIMS measurement, which quantitatively monitors the
vertical profiles of each component across the whole thickness of the active layers.46
410 Joule 4, 407–419, February 19, 2020
A B C
D E F
G H
Figure 2. Film-Forming Properties of BHJ and LbL Blends as well as Their Schematic Representation of 3DMorphological Characteristics and Possible
Physical Dynamics
(A and B) 2D GIWAXS profiles for BHJ (A) and (B) LbL films bladed on the PEDOT:PSS layer. All images are corrected for monitor and film thickness and
displayed on the same logarithmic color scale.
(C) The 1D GIWAXS line curves with respect to the in-plane (IP) direction and out-of-plane (OOP) direction. The IP and OOP profiles of BHJ and LbL films
acquired at the critical incident angle of 0.02�.(D and E) PiFM topography images of relevant (D) BHJ and (E) LbL films based on FTIR absorption at 1,536 cm�1 (Y6).
(F) TOF-SIMS ion yield as a function of sputtering time for BHJ and LbL samples. The depth profile of Y6 by trancing N element is shown here.
(G and H) Visual illustrations of the morphological characteristics and possible physical dynamics of BHJ (G) and LbL (H) blend-based devices.
Of note is that nitrogen (N) was used to track the Y6 acceptor. Combining with the N
signal as depicted in Figure 2F, we can easily conclude that Y6 acceptors were
assembled at the LbL/air surface, and PM6 polymer donors were enriched in the bot-
tom of LbL blend. In contrast, Y6 acceptors were evenly distributed throughout the
BHJ active layer. Therefore, by combining all findings from these morphological
characterizations, we can depict the detailed description of the surface aggregation
patterns and vertical phase separation in optimal BHJ and LbL blends as presented
in Figures 2G and 2H, respectively. The obviously different microstructures of BHJ
and LbL blends are directly reflected in their physical mechanisms (Figures 2G and
2H). Of note is that for a full dynamical analysis and explanation of corresponding
microstructures the reader is referred to Sun et al. (2019).
Joule 4, 407–419, February 19, 2020 411
A B C
D E F
Figure 3. Photovoltaic Parameters of BHJ and LbL Devices and Their Physical Dynamics
(A and B) J-V (A) and EQE (B) curves of the best-performing BHJ and LbL-based devices.
(C) Histograms of the PCE counts for 30 individual BHJ- and 30 individual LbL-based devices.
(D) Hole-only mobilities measured in single carrier diodes obtained from BHJ and LbL films; carrier mobilities of BHJ and LbL-based devices calculated
from photo-CELIV.
(E) Numbers of the extracted carrier in the BHJ and LbL devices as a function of delay time, obtained from photo-CELIV, and the fit.
(F) Charge carrier lifetime t, obtained from TPV, as a function of charge density n, calculated from CE under Voc conditions (from 0.15 to 2.50 suns). The
dashed lines represent linear fits of the data.
The key fabrication conditions of the devices, including the speed of doctor-blade
and the temperature of the bottom, plate to adjust the thicknesses and microstruc-
tures of the BHJ and LbL blends. Here, the device performance based on BHJ and
LbL layers were optimized by the adjustment of blade speeds as provided in Fig-
ure S5, and the related photovoltaic parameters are summarized in Table S1. The
optimal blade speeds were found to be 35 mm/s for BHJ blend and 12 mm/s
(donor)/12 mm/s (acceptor) for LbL blend, respectively. The current-density-voltage
(J-V) characteristics of the optimized devices and relevant parameters are shown in
Figure 3A and Table 1. The LbL device delivered the best PCE of 16.35%, along with
a short-circuit current density (Jsc) of 25.90 mAcm�2, an open-circuit voltage (Voc) of
0.834 V, and a fill factor (FF) of 75.68%, which is higher than that of the BHJ-based
device with an optimized PCE of 15.37% and a Jsc of 25.22 mAcm�2. The slightly
higher Jsc is probably attributed to the increased absorption spectra and higher
photo-absorption rate as well as the better charge dissociation probability in the
LbL devices (Figure S6), which is directly demonstrated by the external quantum ef-
ficiency (EQE) curves provided in Figure 3B. Figure 3C further presents the statistical
photovoltaic metrics and PCE histogram obtained from BHJ and LbL devices, which
indicate the good reproducibility of photovoltaic performance of the PM6:Y6 based
OSCs. Besides, our results provide the impetus for measurements on other high-
performance systems to probe the generality of these LbL morphology guidelines
to maximize performance. Thus, LbL blade-coating approach was employed
for optimizing other efficient active layers, including PM6:Y6-2Cl,9 PM6:Y6-C2,47
412 Joule 4, 407–419, February 19, 2020
Table 1. Photovoltaic Parameters of the BHJ and LbL Devices Based on Various Photovoltaic
Systems under the Illumination of AM 1.5 G at 100 mW cm�2
Active Layer Area (cm2) BHJ/LbL Voc (V) Jsc (mA cm�2) FF (%) PCE (%)
PM6:Y6 0.04 BHJ 0.840 25.22 72.49 15.37 (15.17 G 0.2)
1 BHJ 0.835 25.52 65.74 14.01 (13.71 G 0.3)
PM6/Y6 0.04 LbL 0.834 25.90 75.68 16.35 (16.15 G 0.2)
1 LbL 0.831 25.64 71.42 15.23 (15.03 G 0.2)
PM6:Y6-2Cl 0.04 BHJ 0.847 25.67 70.74 15.38 (15.18 G 0.2)
PM6/Y6-2Cl 0.04 LbL 0.849 25.88 72.30 15.89 (15.69 G 0.2)
PTQ10:Y6 0.04 BHJ 0.855 22.62 70.32 13.62 (13.42 G 0.2)
PTQ10/Y6 0.04 LbL 0.849 24.49 72.63 15.10 (14.90 G 0.2)
PM6:Y6-C2 0.04 BHJ 0.844 25.76 71.69 15.59 (15.39 G 0.2)
PM6/Y6-C2 0.04 LbL 0.834 25.82 73.99 15.93 (15.73 G 0.2)
and PTQ10:Y6.48 The chemical structures of relevant photovoltaic systems as well
as the J-V curves of their devices are provided in Figures S7–S9, and the correspond-
ing photovoltaic parameters are summarized in Table 1. All the LbL devices
exhibited higher PCE values than those of relevant BHJ devices, indicating that
LbL is a universal and effective strategy for a wide range of highly efficient photovol-
taic systems.
The better device properties of the photovoltaic systems fabricated by the blading
LbL approach are mainly attributed to their vertical phase morphology as depicted
in Figure 2H. As is well known, the relevant physical dynamics determined by the
blendmicrostructures caused the difference of photovoltaic performance in relevant
devices. Taking the PM6:Y6 system as an example, we calculated the hole mobilities
from space charge limited current (SCLC) measurements (Figure S10), as presented
in Figure 3D and summarized in Table S2. It was found that the hole mobility values
of LbL blends were slightly lower than that of the BHJ blends, even though their de-
vices exhibited better photovoltaic performance. It is because a more aggregation
of Y6 acceptors at the top layer probably suppressed the hole transport in the hole-
only devise. Here the photo-induced charge carrier extraction by linearly increasing
the voltage (photo-CELIV) over the nanosecond-microsecond (ns-ms, Figure S11)
was employed to determine the ambipolar charge extraction from an actual photo-
voltaic device. As shown in Figure 3D, the average mobility of LbL device (1.07 3
10�4 cm2V�1s�1) is higher than BHJ device (5.94 3 10�5 cm2V�1s�1), indicating
that charge carriers can be transmitted more effectively in a real LbL device.
The time dependence of charge carrier density resulting from the photo-CELIV mea-
surements was further employed to investigate the charge carrier recombination
mechanisms in the BHJ and LbL blends. As shown in Figure 3E, the number of ex-
tracted carriers reduce with increasing delay time between photogeneration and
extraction due to various recombination processes in the devices. Using the
following equation nðtÞ= nð0Þ=1+�
ttB
�g
, where nð0Þ is the initial density of photo-
generated carriers at t = 0 and g is the time-independent parameter), the effective
2nd order recombination coefficient (tB) were calculated.43 Relevant parameters
fitted and calculated by the above-mentioned equation are summarized in Table
S2. The LbL device showed the shorter 2nd order recombination coefficient than
that of BHJ device, which partially fit the description in Figures 2G and 2H. In addi-
tion, the transient time ttr values calculated from Figure 3E are 3.373 10�7 s for BHJ
Joule 4, 407–419, February 19, 2020 413
device and 1.87 3 10�7 s for LbL device. This result is identical to the transient pho-
tovoltage (TPV) measurements (Figure S12 and Table S2). The BHJ device showed a
carrier lifetime of 7.22 ms when the light intensity is around 1 sun, whereas the life-
time of PM6/Y6 system using an LbL structure decreased to 5.84 ms at the same in-
tensity. Additionally, combining the TPV and charge extraction (CE) techniques, a
non-geminate recombination order R (R = l + 1) can be calculated via the equation
t = t0ðn0=nÞl, where t0 (calculated from TPV curves) and n0 (adopted from CE curves,
Figure S13) are constants and l is the so-called recombination exponent.43 As shown
in Figure 3F, a slightly higher recombination order value (R = 2.11) for the BHJ device
as compared to the LbL device (R = 2.02) can be found. Overall, the results of the
carrier recombination dynamic analysis coupled with the charge carrier mobility
finally underpin the complex morphology outline above and give detailed insight
into subtle mechanisms being responsible for device parameters.
The above-mentioned photovoltaic systems demonstrated that the photovoltaic
performance of LbL-based devices is superior to corresponding BHJ devices with
an active area of 0.04 cm2. Due to the special advantages as mentioned above,
we further manufactured large-area LbL-bladed devices with an active area of
1 cm2 (1 3 1 cm). The best performances of the LbL-bladed devices reach a higher
performance of 15.23% compared to BHJ-based devices (14.01% PCE, Figure S14
and Table 1), mainly attributed to different FF values (65.74% for the BHJ device
and 71.42% for the LbL device, respectively). Importantly, direct, trap-free charge
percolation routes through mixed phases and excellent charge extraction and
collection properties of LbL devices with the low non-geminate recombination los-
ses are the likely reasons for the high FF values in the LbL devices. More importantly,
in the LbL-based conventional devices, the acceptors located on the surface and do-
nors deposited on the bottom can effectively suppress the surface recombination
demonstrated in Figure S15,49 and depicted in Figures 2G and 2H, respectively.
Thus, the excellent device area insensitivity of LbL strategy for fabricating single-
junction OSCs investigated in this work is crucial for processing the large-scale solar
modules by printing methods.
As we known, modules have become an important step from the laboratory to the
market toward large-scale production. However, the number of reports of large solar
module with the active area more than 10 cm2 is rather low until now. To demon-
strate the compatibility of the LbL approach with large-area printing techniques
and find an effective processing strategy to overcome the scaling lag of large-area
devices, we fabricated the large-area PM6:Y6-based solar modules with a
3.74 cm2 total area consisting of three series-connected cells with a dimension of
34 3 11 mm and a 12.60 cm2 total area consisting of four series-connected cells
with a dimension of 3.5 3 36 mm, respectively. The deal area of each cell in the
3.74 cm2 based module was determined to be 2 3 11 mm, resulting in an active
area of 3.3 cm2 and a GFF17,50,51 (defined as the ratio between active area and total
area of a monolithically interconnected module) of 88.2%. While the real area of the
12.6 cm2 based module was determined to be 13 36 mm, resulting in an active area
of 11.52 cm2 and aGFF of 91.4%. The LbL-based process of relevant solar modules is
depicted in Figure 4A, and the detailed fabrication processes investigated in this
study are explained in the Experimental Procedures. Figure 4B exhibits a digital
photo of the solar modules based on LbL PM6/Y6 film with an active area of
11.52 cm2 and a GFF of 91.4%. A schematic image of the large-area module fabri-
cation design is provided in Figure S16. In addition, Figure 4C shows the corre-
sponding J-V curves of the resulting BHJ and LbL modules with active areas of 3.3
414 Joule 4, 407–419, February 19, 2020
A B
C D E
F G
Figure 4. Fabrication and Device Performance of Large-Area Solar Modules
(A) Process flow diagram of the LbL-based process for large-area solar modules.
(B) Image of solar modules based on LbL PM6/Y6 film with an active area of 11.52 cm2 and an optimal GFF of 91.4%.
(C) Illuminated J-V curves of BHJ and LbL devices using a doctor-blade coating.
(D) Histograms of the PCE counts for 15 individual BHJ- and 15 individual LbL-based solar modules.
(E) Series resistance values of BHJ and LbL devices and solar modules.
(F) Chronological evolution of PCEs determined by J-V measurements of optimized devices with different device areas and processing technologies
and distribution of the PCEs of solar modules in terms of the device area fabricated with different printing or coating techniques. Here, SC is a spin-
coating method, DB is a doctor-blade-coating approach, and SD is a slot-die processing method.
(G) Distribution of the PCEs of organic solar modules in terms of the GFF values.
and 11.52 cm2, respectively, and the relevant photovoltaic parameters are summa-
rized in Table 2. As presented in Figure 4C and Table 2, the 11.52 cm2 solar module
fabricated by the LbL-based process exhibits a PCE of 11.86% with a Voc of 3.20 V, a
Jsc of 6.41 mAcm�2, and a FF of 57.85%, which is much better than that of the BHJ
module with a PCE of 10.15% and a FF of 50.12%. Moreover, the statistical photo-
voltaic metrics and PCE histogram obtained from BHJ and LbL modules with an
active area of 11.52 cm2, presented in Figure 4D, further indicating the good repro-
ducibility of the solar modules. Besides, the LbL module with an active area of
3.3 cm2 showed better photovoltaic performance (13.88%) than those of BHJ mod-
ule (11.86%), as exhibited in Figure 4C. In short, the LbL processing strategy strongly
Joule 4, 407–419, February 19, 2020 415
Table 2. Photovoltaic Parameters of the BHJ and LbL-Based Devices and Module Devices with
Various Areas under the Illumination of AM 1.5 G at 100 mW cm�2
Active Layer Area (cm2) Voc (V) Jsc (mA cm�2) Fill Factor (%) PCE (%)
PM6:Y6 3.3 2.51 8.61 55.78 12.06 (11.76 G 0.3)
11.52 3.21 6.25 50.12 10.15 (9.75 G 0.4)
PM6/Y6 3.3 2.49 8.72 63.92 13.88 (13.68 G 0.2)
11.52 3.20 6.41 57.85 11.86 (11.56 G 0.3)
exhibits an outstanding tolerance to device area variation in fabricating large-scale
OSCs and solar modules.
Obviously, the photovoltaic performance difference of BHJ and LbL-based modules
are mainly attributed to the difference of FF values (3.3 cm2: 55.78% for BHJ and
63.92% for LbL, respectively; 11.52 cm2: 50.12% for BHJ and 57.85% for LbL, respec-
tively). These can be partially explained by the above-mentioned advantages of LbL
architecture, including suitable blend morphology, high photo-absorption rate, and
good charge transport and extraction properties on the one hand. On the other
hand, the increase of FF values of the LbL modules as compared to that of the
BHJ modules can also be caused by the decrease of the series resistance (Rs), as
shown in Figure 4E. The lower series resistance of the larger area LbL OSCs could
result from the weakening or elimination of the above-mentioned ‘‘islands’’ and
‘‘bad peninsulas’’ of donor and/or acceptor aggregations in devices as depicted in
Figure 2G. Besides, we also measured the atomic force microscopy (AFM) images
of the large-scale BHJ and LbL films collected at different spots, as presented in Fig-
ure S17. All the collected AFM images (bottom, Figure S17) exhibited the smoother
surfaces in the whole LbL film with a size of 5 3 5 cm, suggesting a perfectly homo-
geneous LbL-coated active layers.
It should be noted that the dependence of photovoltaic performance (especially for
the FF values) on the device area by enlarging effective area from 0.04 to 11.52 cm2
are obvious and unavoidable.17 This is understandable since electrical loss from the
bottom ITO electrode, geometric loss induced by increasing cell width, and addi-
tional losses caused by film inhomogeneity, defects, or particles can significantly
limit the photovoltaic parameters of BHJ- and LbL-based OSCs with the enlarged
active area,17,52 as presented in Tables 1 and 2. Anyway, the LbL-bladed large-
area OSC modules of 3.3 and 11.52 cm2 exhibits outstanding PCEs of 13.88% and
11.86%, respectively. To the best of our knowledge, this is the highest efficiencies
reported thus far for large-area OSC modules, as depicted in Figure 4F. In addition,
as compared to the BHJ-bladed devices, our results demonstrated that the LbL
coating process is a successful technology which can quickly bridge the huge gap
between the ‘‘hero’’ lab-scale produced cells and large-area solar modules. Of
note is that the GFF values of all manufactured BHJ and LbL modules with an active
area of 11.52 cm2 are over 90% (Figure 4G), further underscoring the good repro-
ducibility for this highly attractive module layout.
Conclusions
In summary, we reported an LbL processing approach as a printable strategy for
high-performance large-area solar cells and modules. The LbL method exhibits
unique advantages of combining the merits of high photo-absorption rate, suitable
vertical phase separation, and good practicability, endowing the LbL devices with
excellent charge transport and extraction properties. As a result, the LbL-based
PM6:Y6 OSCs showed a higher PCE of 16.35% compared to the BHJ-bladed device
416 Joule 4, 407–419, February 19, 2020
(15.37%). Furthermore, the other three high-performance non-fullerene systems
investigated in this study, including PM6:Y6-2Cl, PTQ10:Y6, and PM6:Y6-C2, further
demonstrated the excellent universality of the LbL coating approach. Benefiting
from the excellent physical dynamics and good surface homogeneity of the LbL
blends, we applied this LbL processing strategy to fabricate solar modules with
larger active areas. As compared to its BHJ counterpart (10.15%), the LbL-based
modules with an active area of 11.52 cm2 and a GFF of over 90% deliver 11.86% po-
wer conversion efficiency, indicating that the LbL processing strategy can signifi-
cantly reduce the scaling gap, which taken as a sign of technological maturity.
Besides, to the best of our knowledge, this is the highest efficiency thus far for
large-area organic solar modules. Overall, this work not only further sheds light on
the unique advantages of LbL coating strategy but also demonstrates a successful
printing technique of processing the photoactive layer via the LbL strategy for up-
scaling organic solar cells toward high-performance large-scale production and in-
dustrial applications.
EXPERIMENTAL PROCEDURES
Full details of experimental procedures can be found in the Supplemental
Information.
SUPPLEMENTAL INFORMATION
Supplemental Information can be found online at https://doi.org/10.1016/j.joule.
2019.12.004.
ACKNOWLEDGMENTS
This work was financially supported by the National Natural Science Foundation of
China (NSFC) (grant nos. 21702154 and 51773157). We also thank the support of
the opening project of Key Laboratory of Materials Processing and Mold and Beijing
National Laboratory for Molecular Sciences (BNLMS201905).
AUTHOR CONTRIBUTIONS
R.S. and J.M. conceived and developed the ideas. R.S. designed the experiments
and performed device fabrications. R.S. and Q.W. fabricated the module devices.
J.G. performed optical simulation and analysis. T.W. synthesized the PM6 material.
Y.W. synthesized the Y6-2Cl material. Z.H.L. and C.L.Y. provided the Y6-C2, and
Z.C.H. and F.H. provided the PNDIT-F3N-Br material. B.B.Q. conducted the P-iFM
measurement. R.S. and J.M. wrote the manuscript.
DECLARATION OF INTERESTS
The authors declare no competing interests.
Received: September 29, 2019
Revised: November 14, 2019
Accepted: December 4, 2019
Published: December 31, 2019
REFERENCES
1. Huang, W., Cheng, P., Yang, Y.M., Li, G., andYang, Y. (2018). High-performanceorganic bulk-heterojunction solar cellsbased on multiple-donor or multiple-acceptor components. Adv. Mater. 30,1705706.
2. Zhang, G., Zhao, J., Chow, P.C.Y., Jiang, K.,Zhang, J., Zhu, Z., Zhang, J., Huang, F., andYan, H. (2018). Nonfullerene acceptormolecules for bulk heterojunctionorganic solar cells. Chem. Rev. 118,3447–3507.
3. Zhao, F., Wang, C., and Zhan, X. (2018).Morphology control in organic solar cells. Adv.Energy Mater. 8, 1703147.
4. Xu, X., Feng, K., Bi, Z., Ma, W., Zhang, G., andPeng, Q. (2019). Single-junction polymer solar
cells with 16.35% efficiency enabled by aplatinum(II) complexation strategy. Adv.Mater.31, e1901872.
5. Guo, J., and Min, J. (2019). A cost analysis offully solution-processe d ITO- free organicsolar modules. Adv. Energy Mater. 9, 1802521.
6. Min, J., Bronnbauer, C., Zhang, Z.-G., Cui, C.,Luponosov, Y.N., Ata, I., Schweizer, P.,Przybilla, T., Guo, F., Ameri, T., et al. (2016).Fully solution-processed small moleculesemitransparent solar cells: optimization oftransparent cathode architecture and fourabsorbing layers. Adv. Funct. Mater. 26, 4543–4550.
7. Wu, Y., Yang, H., Zou, Y., Dong, Y., Yuan, J.,Cui, C., and Li, Y. (2019). A new dialkylthio-substituted naphtho[2,3-c]thiophene-4,9-dione based polymer donor for high-performance polymer solar cells. EnergyEnviron. Sci. 12, 675–683.
8. Yuan, J., Zhang, Y., Zhou, L., Zhang, G., Yip,H.-L., Lau, T.-K., Lu, X., Zhu, C., Peng, H.,Johnson, P.A., et al. (2019). Single-junctionorganic solar cell with over 15% efficiency usingfused-ring acceptor with electron-deficientcore. Joule 3, 1140–1151.
9. Cui, Y., Yao, H., Zhang, J., Zhang, T., Wang, Y.,Hong, L., Xian, K., Xu, B., Zhang, S., Peng, J.,et al. (2019). Over 16% efficiency organicphotovoltaic cells enabled by a chlorinatedacceptor with increased open-circuit voltages.Nat. Commun. 10, 2515.
10. Wang, W., Sun, R., Guo, J., Guo, J., and Min, J.(2019). An oligothiophene-fullerene moleculewith a balanced donor-acceptor backbonetoward high-performance single componentorganic solar cells. Angew. Chem. Int. Ed. Engl.58, 14556–14561.
11. Yu, R., Yao, H., Chen, Z., Xin, J., Hong, L., Xu, Y.,Zu, Y., Ma, W., and Hou, J. (2019). Enhancedp–p Interactions of nonfullerene acceptors byvolatilizable solid additives in efficient polymersolar cells. Adv. Mater. 31, e1900477.
12. Chen, J., Bi, Z., Xu, X., Zhang, Q., Yang, S., Guo,S., Yan, H., You, W., and Ma, W. (2019). Fineoptimization of morphology evolution kineticswith binary additives for efficient non-fullereneorganic solar cells. Adv Sci. 6, 1801560.
13. Yip, H.-L., and Jen, A.K.-Y. (2012). Recentadvances in solution-processed interfacialmaterials for efficient and stable polymer solarcells. Energy Environ. Sci. 5, 5994–6011.
14. Li,W., Chen,M., Cai, J., Spooner, E.L.K., Zhang,H., Gurney, R.S., Liu, D., Xiao, Z., Lidzey, D.G.,Ding, L., and Wang, T. (2019). Molecular ordercontrol of non-fullerene acceptors for high-efficiency polymer solar cells. Joule 3, 819–833.
15. Yao, H., Cui, Y., Qian, D., Ponseca, C.S.,Honarfar, A., Xu, Y., Xin, J., Chen, Z., Hong, L.,Gao, B., et al. (2019). 14.7% efficiency organicphotovoltaic cells enabled by active materialswith a large electrostatic potential difference.J. Am. Chem. Soc. 141, 7743–7750.
16. Yao, H., Qian, D., Zhang, H., Qin, Y., Xu, B., Cui,Y., Yu, R., Gao, F., and Hou, J. (2018). Criticalrole of molecular electrostatic potential oncharge generation in organic solar cells. Chin.J. Chem. 36, 491–494.
418 Joule 4, 407–419, February 19, 2020
17. Lucera, L., Kubis, P., Fecher, F.W., Bronnbauer,C., Turbiez, M., Forberich, K., Ameri, T.,Egelhaaf, H.-J., and Brabec, C.J. (2015).Guidelines for closing the efficiency gapbetween hero solar cells and roll-to-roll printedmodules. Energy Technol. 3, 373–384.
18. Badgujar, S., Lee, G.-Y., Park, T., Song, C.E.,Park, S., Oh, S., Shin, W.S., Moon, S.-J., Lee,J.-C., and Lee, S.K. (2016). High-performancesmall molecule via tailoring intermolecularinteractions and its application in large-areaorganic photovoltaic modules. Adv. EnergyMater. 6, 1600228.
19. Jin, H., Tao, C., Velusamy, M., Aljada, M.,Zhang, Y., Hambsch, M., Burn, P.L., andMeredith, P. (2012). Efficient, large area ITO-and-PEDOT-free organic solar cell sub-modules. Adv. Mater. 24, 2572–2577.
20. Gasparini, N., Lucera, L., Salvador, M., Prosa,M., Spyropoulos, G.D., Kubis, P., Egelhaaf,H.-J., Brabec, C.J., and Ameri, T. (2017). High-performance ternary organic solar cells withthick active layer exceeding 11% efficiency.Energy Environ. Sci. 10, 885–892.
21. Li, N., Kubis, P., Forberich, K., Ameri, T., Krebs,F.C., and Brabec, C.J. (2014). Towards large-scale production of solution-processed organictandemmodules based on ternary composites:design of the intermediate layer, deviceoptimization and laser based moduleprocessing. Sol. Energy Mater. Sol. Cells 120,701–708.
22. Spyropoulos, G.D., Kubis, P., Li, N., Baran, D.,Lucera, L., Salvador, M., Ameri, T., Voigt, M.M.,Krebs, F.C., and Brabec, C.J. (2014). Flexibleorganic tandem solar modules with 6%efficiency: combining roll-to-roll compatibleprocessing with high geometric fill factors.Energy Environ. Sci. 7, 3284–3290.
24. Krebs, F.C., Hosel, M., Corazza, M., Roth, B.,Madsen, M.V., Gevorgyan, S.A., Søndergaard,R.R., Karg, D., and Jørgensen, M. (2013). Freelyavailable OPV-the fast way to progress. EnergyTechnol. 1, 378–381.
25. Krebs, F.C., Espinosa, N., Hosel, M.,Søndergaard, R.R., and Jørgensen, M. (2014).25th anniversary article: rise to power - OPV-based solar parks. Adv. Mater. 26, 29–38.
26. Agrawal, N., Zubair Ansari, M., Majumdar, A.,Gahlot, R., and Khare, N. (2016). Efficient up-scaling of organic solar cells. Sol. EnergyMater.Sol. Cells 157, 960–965.
27. Li, Y., Xu, G., Cui, C., and Li, Y. (2018). Flexibleand semitransparent organic solar cells. Adv.Energy Mater. 8, 1701791.
28. Li, N., and Brabec, C.J. (2015). Air-processedpolymer tandem solar cells with powerconversion efficiency exceeding 10%. EnergyEnviron. Sci. 8, 2902–2909.
29. Zhao, K., Hu, H., Spada, E., Jagadamma, L.K.,Yan, B., Abdelsamie, M., Yang, Y., Yu, L., Munir,R., Li, R., et al. (2016). Highly efficient polymersolar cells with printed photoactive layer:rational process transfer from spin-coating.J. Mater. Chem. A 4, 16036–16046.
30. Na, S., Seo, Y., Nah, Y., Kim, S., Heo, H., Kim, J.,et al. (2019). High performance roll-to-rollproduced fullerene-free organic photovoltaicdevices via temperature-controlled slot diecoating. Adv. Funct. Mater. 29, 1805825.
31. Green, M.A., Hishikawa, Y., Dunlop, E.D., Levi,D.H., Hohl-Ebinger, J., Yoshita, M., and Ho-Baillie, A.W.Y. (2019). Solar cell efficiencytables, version 53. Pro. Pho. Res. Appl. 27, 3–12.
32. Zhang, J., Zhao, Y., Fang, J., Yuan, L., Xia, B.,Wang, G., Wang, Z., Ma, W., Yan, W., Su, W.,et al. (2017). Enhancing performance of large-area organic solar cells with thick film viaternary Stratery. Small 12, 1700388.
33. Krebs, F.C. (2009). Polymer solar cell modulesprepared using roll-to-roll methods: knife-over-edge coating, slot-die coating and screenprinting. Sol. Energy Mater. Sol. Cells 93,465–475.
34. Hong, S., Kang, H., Kim, G., Lee, S., Kim, S.,Lee, J.H., Lee, J., Yi, M., Kim, J., Back, H., et al.(2016). A series connection architecture forlarge-area organic photovoltaic modules with a7.5% module efficiency. Nat. Commun. 7,10279.
35. Zhang, K., Chen, Z., Armin, A., Dong, S., Xia, R.,Yip, H.-L., Shoaee, S., Huang, F., and Cao, Y.(2018). Efficient large area organic solarcells processed by blade-coating withsingle-component green solvent. Sol. RRL 2,1700169.
36. Wu, Q., Guo, J., Sun, R., Guo, J., Jia, S., Li, Y.,Wang, J., and Min, J. (2019). Slot-die printednon-fullerene organic solar cells with thehighest efficiency of 12.9% for low-cost PV-driven water splitting. Nano Energy 61,559–566.
37. Sun, R., Guo, J., Sun, C., Wang, T., Luo, Z.,Zhang, Z., Jiao, X., Tang, W., Yang, C., Li, Y.,and Min, J. (2019). A universal layer-by-layersolution-processing approach for efficient non-fullerene organic solar cells. Energy Environ.Sci. 12, 384–395.
38. Ayzner, A.L., Tassone, C.J., Tolbert, S.H., andSchwartz, B.J. (2009). Reappraising the need forbulk heterojunctions in polymer-fullerenephotovoltaics: the role of carrier transport in all-solution-processed P3HT/PCBM bilayer solarcells. J. Phys. Chem. C 113, 20050–20060.
39. Cui, Y., Zhang, S., Liang, N., Kong, J., Yang, C.,Yao, H., Ma, L., and Hou, J. (2018). Towardefficient polymer solar cells processed by asolution-processed layer-by-layer approach.Adv. Mater. 30, e1802499.
40. Li, H., Qi, Z., andWang, J. (2013). Layer-by-layerprocessed polymer solar cells with self-assembled electron buffer layer. Appl. Phys.Lett. 102, 213901.
41. Lin, Y., Ma, L., Li, Y., Liu, Y., Zhu, D., and Zhan, X.(2013). A solution-processable small moleculebased on benzodithiophene anddiketopyrrolopyrrole for high-performanceorganic solar cells. Adv. Energy Mater. 3, 1166–1170.
42. Dong, S., Zhang, K., Xie, B., Xiao, J., Yip, H.-L.,Yan, H., Huang, F., and Cao, Y. (2019). High-performance large-area organic solar cellsenabled by sequential bilayer processing vianonhalogenated Solvents. Adv. Energy Mater.9, 1802832.
43. Sun, R., Guo, J., Wu, Q., Zhang, Z., Yang, W.,Guo, J., Shi, M., Zhang, Y., Kahmann, S., Ye, L.,et al. (2019). A multi-objective optimization-based layer-by-layer blade-coating approachfor organic solar cells: rational control ofvertical stratification for high performance.Energy Environ. Sci. 12, 3118–3132.
44. Zhang, M., Guo, X., Ma, W., Ade, H., and Hou,J. (2015). A large-bandgap conjugatedpolymer for versatile photovoltaic applicationswith high performance. Adv. Mater. 27, 4655–4660.
46. Min, J., Guldal, N.S., Guo, J., Fang, C., Jiao,X., Hu, H., Heumuller, T., Ade, H., andBrabec, C.J. (2017). Gaining further insightinto the effects of thermal annealing and
solvent vapor annealing on timemorphological development anddegradation in small molecule solar cells.J. Mater. Chem. A 5, 18101–18110.
47. Luo, Z., Sun, R., Zhong, C., Liu, T., Ma, R.,Zhang, G., Zou, H., Jiao, X., Min, J., and Yang,C. (2019). Altering Alkyl-Chains BranchingPositions for Boosting the Performance ofSmall-MoleculeA cceptors for Highly EfficientNonfullerene Organic Solar Cells. Sci. China.Chem. https://doi.org/10.1007/s11426-019-9670-2.
48. Wu, Y., Zheng, Y., Yang, H., Sun, C., Dong, Y.,Cui, C., Yan, H., and Li, Y. (2019). Rationallypairing photoactive materials for high-performance polymer solar cells with efficiencyof 16.53%. Sci. China Chem. 62, 1–2.
49. Min, J., Luponosov, Y.N., Zhang, Z.-G.,Ponomarenko, S.A., Ameri, T., Li, Y., andBrabec, C.J. (2014). Interface design toimprove the performance and stability of
solution-processed small-moleculeconventional solar cells. Adv. Energy Mater.4, 1400816.
50. Strohm, S., Machui, F., Langner, S., Kubis, P.,Gasparini, N., Salvador, M., McCulloch, I.,Egelhaaf, H.-J., and Brabec, C.J. (2018). P3HT:non-fullerene acceptor based large area, semi-transparent PVmodules with power conversionefficiencies of 5%, processed by industriallyscalable methods. Energy Environ. Sci. 11,2225–2234.
51. Zhao, W., Zhang, Y., Zhang, S., Li, S., He, C.,and Hou, J. (2019). Vacuum assisted annealingmethod for high efficiency printable large-areapolymer solar cell modules. J. Mater. Chem. C7, 14070–14078.
52. Fan, B., Zhong, W., Ying, L., Zhang, D., Li, M.,Lin, Y., Xia, R., Liu, F., Yip, H.L., Li, N., et al.(2019). Surpassing the 10% efficiency milestonefor 1-cm2 all-polymer solar cells. Nat. Commun.10, 4100.
Rui Sun, Qiang Wu, Jie Guo, Tao Wang, Yao Wu, Beibei Qiu, Zhenghui Luo, WenyanYang, Zhicheng Hu, Jing Guo, Mumin Shi, Chuluo Yang, Fei Huang, YongfangLi, and Jie Min
SUPPLEMENTAL INFORMATION
Supplemental Data Items
Figure S1. Optical constants (n and k) of the involved active layers.
Figure S2. Transmission and reflection spectra of (A) BHJ and (B) LbL blends with
different thicknesses.
Figure S3. Photoluminescence (PL) spectra of neat donor PM6 and acceptor Y6 films,
and the BHJ and LbL blends. Comparative studies of PL quenching (PLQ) efficiency
shows that LbL blend (93.4%) is slightly more efficient PL quencher than the BHJ blend