Palladium-catalyzed methylation of terminal alkynes

Article abstract of DOI:10.1016/j.catcom.2019.105835

In this communication, a palladium-catalyzed procedure for the methylation of terminal alkynes has been developed. With N,N,N-trimethylbenzenaminium trifluoromethanesulfonate as the methyl source, various desired products were obtained in moderate to good yields. Both aromatic and aliphatic alkynes are applicable.

Full text of DOI:10.1016/j.catcom.2019.105835

FULL PAPER
Perovskite Solar Cells
www.afm-journal.de
Insights into Fullerene Passivation of SnO2 Electron
Transport Layers in Perovskite Solar Cells
Junke Wang, Kunal Datta, Christ H. L. Weijtens, Martijn M. Wienk,
and René A. J. Janssen*
compositions and device architectures,
resulting in power conversion efficiencies
(PCEs) well above 20%.[1–8] Compared to
other PV technologies, perovskites present
distinct advantages by combining high
PCEs, with cost-effectiveness and solution
processability, providing the incentive to
explore large-area implementation.[9–12] In
state-of-the-art n–i–p PSC architectures,
mesoporous TiO2 is commonly used as
electron transport layer (ETL). Mesoporous
TiO2, however, requires a high tempera-
ture (≥450 °C) process, which limits
its potential for upscalable processing
on flexible substrates.[9,12,13] Numerous
early studies[14–17] have pinpointed the
excellent characteristics of lead halide
perovskites, i.e., a high absorption coef-
ficient, long carrier diffusion length and
bipolar carrier transport, suggesting that
high PCEs can also be achieved in planar
PSCs, eliminating the mesoporous TiO2
layer and the need for high temperature
processing.
Interfaces between the photoactive and charge transport layers are crucial
for the performance of perovskite solar cells. Surface passivation of SnO2
as electron transport layer (ETL) by fullerene derivatives is known to
improve the performance of n–i–p devices, yet organic passivation layers
are susceptible to removal during perovskite deposition. Understanding
the nature of the passivation is important for further optimization of SnO2
ETLs. X-ray photoelectron spectroscopy depth profiling is a convenient
tool to monitor the fullerene concentration in passivation layers at a SnO2
interface. Through a comparative study using [6,6]-phenyl-C61-butyric acid
methyl ester (PCBM) and [6,6]-phenyl-C61-butyric acid (PCBA) passivation
layers, a direct correlation is established between the formation of interfacial
chemical bonds and the retention of passivating fullerene molecules at the
SnO2 interface that effectively reduces the number of defects and enhances
electron mobility. Devices with only a PCBA-monolayer-passivated SnO2 ETL
exhibit significantly improved performance and reproducibility, achieving
an efficiency of 18.8%. Investigating thick and solvent-resistant C60 and
PCBM-dimer layers demonstrates that the charge transport in the ETL is only
improved by chemisorption of the fullerene at the SnO2 surface.
In the planar n–i–p PSC architecture, n-type semiconducting
metal oxides with tailored compositions and structures have
been extensively explored as the ETLs.[18,19] Tin oxide (SnO2)
is considered as a superior candidate for high-performing
PSCs.[20–25] Compared to other metal oxides, it has been shown
to provide better energy level alignment with the perovskite
absorber, higher electron mobility, enhanced UV stability, and
visible light transparency.[21,23,25,26] Recently, several low-tem-
perature processing techniques such as the sol–gel layers,[21]
atomic layer deposition,[25–27] and chemical bath deposition[28]
have been utilized for high-quality planar SnO2 ETLs. More-
over, by using a commercialized SnO2 colloidal dispersion, the
best PCE of SnO2-based PSCs (23.32%) is now comparable to
that of mesoporous-TiO2 cells.[29] It is noteworthy, however, that
metal oxides are prone to contain imperfections. Depending
on the fabrication method, the number of defects (e.g., oxygen
vacancies) of SnO2 can vary dramatically, affecting device per-
formance and stability.[22,30] Specifically, defects at the SnO2/
perovskite interface can induce ionic charge accumulation[31,32]
and nonradiative interfacial recombination,[33] both con-
tributing to the hysteresis effect and performance loss in
PSCs.[31,34] Surface passivation of SnO2 can considerably sup-
press the formation of interfacial defects.[20,33] By introducing
an organic modification layer between the perovskite and SnO2
1. Introduction
Organic–inorganic hybrid perovskite solar cells (PSCs) have
attracted unprecedented interests from the photovoltaic (PV)
research community in recent years. Robust strategies have
been developed for small-area devices with various perovskite
J. Wang, K. Datta, Dr. C. H. L. Weijtens, Dr. M. M. Wienk,
Prof. R. A. J. Janssen
Molecular Materials and Nanosystems
Institute of Complex Molecular Systems
Eindhoven University of Technology
P.O. Box 513, 5600 MB Eindhoven, The Netherlands
E-mail: r.a.j.janssen@tue.nl
Prof. R. A. J. Janssen
Dutch Institute for Fundamental Energy Research
De Zaale 20, 5612 AJ Eindhoven, The Netherlands
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adfm.201905883.
© 2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co.
KGaA, Weinheim. This is an open access article under the terms of the
Creative Commons Attribution-NonCommercial License, which permits
use, distribution and reproduction in any medium, provided the original
work is properly cited and is not used for commercial purposes.
DOI: 10.1002/adfm.201905883
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (1 of 12)

www.advancedsciencenews.com
www.afm-journal.de
layers, PSCs exhibit reduced interfacial loss, less pronounced
hysteresis, and thereby better PV performance.[30,33,35–39]
layer, the charge transport in the ETL remains inefficient unless
thermal annealing is applied after depositing PCBM. Our results
highlight that chemisorption (PCBA) rather than physisorp-
tion (C60 and PCBM) to SnO2 is the most influential factor in
enhancing the interfacial charge transport process.
There are some prerequisites for the efficient surface passiva-
tion of SnO2.[19,40] First is the activation of the SnO2 surface.[40]
By UV–ozone treatment, organic binders and adventitious
contaminants of SnO2 can easily be removed while the number
of surface-bound hydroxyl groups is increased, providing more
anchoring sites for the passivation layer.[22,41,42] Second, a judi-
cious selection of organic surface modifier is required. Fullerene
derivatives, as well as other carboxylate self-assembled mono-
layers (SAMs), are among the most studied owing to the ease of
forming covalent bonds between the carboxylate moiety and SnO2
surface, and the broad availability of electron-accepting groups
which passivate the SnO2/perovskite interface.[19,33,35–39,43–46]
Third, the deposition condition of organic modifiers could also
affect their binding strength toward the substrate.[19] While
no special treatment is required for carboxylic acid-based
SAMs,[35,47–49] a thermal treatment is commonly used when
employing passivation by [6,6]-phenyl-C61-butyric acid methyl
ester (PCBM)[37,38,48] to enhance binding to the SnO2 surface.
However, it is remarkable that most organic modifiers are not
solvent-resistant and can easily be removed by spin coating of the
perovskite precursors.[50–56] Despite improving the PCE,[35,39] very
few studies were able to pinpoint the whereabouts of the pas-
sivation layer.[47,56–58] Therefore, the direct correlation between
the presence of interfacial modifiers and the PSC performance
remains obscure. On the other hand, tremendous research
efforts have been made to improve the robustness of the organic
2. Results and Discussion
Planar n–i–p PSCs were fabricated based on an indium tin
oxide (ITO)/SnO2/perovskite/Spiro-OMeTAD/MoO3/Au device
structure, in which the SnO2 ETL is spin coated from a col-
loidal dispersion and passivated by different fullerene layers.
A double-cation perovskite FA0.66MA0.34PbI2.85Br0.15 (FA is for-
mamidinium, and MA is methylammonium) was deposited
on the ETL via a facile two-step solution process reported by
Qiu et al.[55,65] To optimize processing conditions for the ETL,
we started using a SnO2 layer (≈20 nm), spin coated from an
≈2.8 wt% colloidal aqueous solution,[23] thermally annealed in
air at 150 °C, and treated by UV–ozone before depositing PCBM
from chlorobenzene (10 mg mL−1) and subsequent annealing at
100 °C for 20 min. The UV–ozone treatment increases the open-
circuit voltage (Voc) from 0.97 to 1.04 V (Figure S1 and Table S1,
Supporting Information). XPS confirmed that together with the
removal of carbon contaminant or organic binders of the SnO2
layer,[42] a more hydroxylated SnO2 surface is obtained after
UV–ozone treatment, which promotes the passivation of SnO2
by PCBM (Figure S2 and Table S2, Supporting Information).[66]
Different from a previous study,[23] we found that the Voc of
PSCs can be further improved when increasing the thickness of
the SnO2 layer from ≈20 to ≈110 nm (Figure S3 and Table S3,
Supporting Information) because of better coverage and
reduced shunting pathways.[67] Surprisingly, the fill factor (FF) is
not influenced when increasing the SnO2 layer thickness, owing
to its high electron mobility (Table S3, Supporting information).
As a result, optimum PSC performance was achieved by using
10 min. UV–ozone treated ≈110 nm SnO2 layers as ETL.
modification layer, for instance, by introducing thermal evapo-
[37,48,59]
rated C60
or by developing cross-linkable fullerene deriva-
tives,[53–55,60,61] after which a bilayered ETL in the device can be
formed. In this case, however, the impact of a chemical interac-
tion between the SnO2 and passivation layer is often neglected.
Herein, we use three representative fullerenes, C60, PCBM, and
[6,6]-phenyl-C61-butyric acid (PCBA) (Figure 1) to passivate a SnO2
ETL in planar n–i–p PSCs. Through X-ray photoelectron spectros-
copy (XPS) depth profiling we reveal the precise distribution of
fullerenes in the ETL. It is found that a PCBA monolayer easily
binds to the SnO2 interface, whereas a PCBM monolayer can only
be created and preserved when employing a thermal annealing
treatment prior to depositing the perovskite. We demonstrate that
fullerenes with the ability to react with surface hydroxyl groups
effectively passivate defects at the SnO2 interface and improve
the performance and reproducibility of the PSCs. After optimiza-
tion, the best-performing device with a PCE of 18.8% is achieved
by using PCBA-monolayer-passivated SnO2. Although thermally
evaporated C60 is mostly retained
To study the effect of fullerene passivation on the device per-
formance, C60, PCBM, and PCBA were introduced at the SnO2/
perovskite interface, either with or without a thermal annealing
treatment at 100 °C for 20 min. in N2 atmosphere. Figure 2 and
Figure S4 in the Supporting Information show the statistical
distribution of performance characteristics (short-circuit cur-
rent density Jsc, Voc, FF, and PCE) of the devices with different
passivation layers. The corresponding averaged PV parameters
are summarized in Table S4 (Supporting Information) for the
after the solvent treatment, the deteri-
orated PV performance indicates that
no efficient passivation is developed
at the SnO2 interface. By exposing an
as-deposited PCBM layer to UV light,
a less-soluble dimeric state of PCBM
is obtained,[62–64] which enabled us
to further study the role of fullerene
passivation in a bilayered SnO2/
fullerene ETL configuration. It is
found that even when the SnO2 sur-
face is fully covered by a PCBM-dimer Figure 1. Chemical structure of C60, PCBM, and PCBA.
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (2 of 12)

www.advancedsciencenews.com
www.afm-journal.de
23
1.16
1.12
1.08
1.04
1.00
0.96
0.92
a
b
22
21
20
19
w
w
w
w
w
0
w
0
wo
0
wo
18
0
6
6
6
6
/C
2
/PCBA wo
2
/C
2
/PCBM wo
2
/PCBM
2
/PCBA
2
/C
2
/PCBA wo
2
/C
/PCBM
2
/PCBA
2
/PCBM wo
2
2
2
2
O
O
O
O
O
O
O
O
O
O
n
SnO
SnO
SnO
Sn
S
Sn
SnO
Sn
Sn
Sn
Sn
Sn
Sn
Sn
17
0.8
0.7
0.6
0.5
0.4
0.3
0.2
20
18
16
14
12
10
8
d
c
w
w
w
w
A
w
0
w
0
wo
0
wo
6
BM
0
C
6
6
6
6
/C
/PCBA wo
2
/C
2
/PCBM wo
2
/P
/PCB
2
/C
2
/PCBA wo
2
/C
2
/PCBM wo
2
/PCBM
2
/PCBA
2
2
2
2
2
O
O
O
O
O
O
O
O
O
O
4
n
n
S
SnO
Sn
Sn
Sn
S
Sn
SnO
Sn
Sn
SnO
SnO
Sn
Sn
Figure 2. Statistical distribution of a) Jsc, b) Voc, c) FF, and d) PCE of the ITO/SnO2/fullerene/FA0.66MA0.34PbI2.85Br0.15/Spiro-OMeTAD/MoO3/Au PSCs
using different fullerene passivation layers, with (w) or without (wo) thermal annealing, and measured in a reverse scan.
reverse scan (from Voc to Jsc) and forward scan (from Jsc to Voc).
Figure 2 demonstrates that without a passivation layer the SnO2
control device exhibits a significant spread in performance,
showing an average PCE of 14.7 1.79% in the reverse scan
and 13.0 2.45% in the forward scan. The significant differ-
ences can be ascribed to an uncontrolled number of trap states,
which induces localized ionic charges at the SnO2/perovskite
interface and results in the formation of a potential barrier that
hampers the charge extraction process.[31]
To test the effect of interfacial passivation layers, we first
employed a thermally evaporated C60 layer. It has previously
been reported that ≥35 nm thick C60 layers are needed to
retain complete coverage of the substrate after processing the
perovskite layer from solution without changing its electron
transport properties.[59] However, we find that devices with a
thermally evaporated C60 layer (35 nm) performed inferior com-
pared to the SnO2 device. The Voc is reduced from 1.10 0.02 to
1.03 0.02 V, while the FF drops from 0.65 0.04 to 0.54 0.12,
resulting in a moderate PCE of 11.7 3.12% (in the reverse
scan). The low FF is ascribed to a high series resistance, resulting
from the thick C60 layer. Thermal annealing of the C60 layer pro-
vides only a marginal improvement, delivering an average PCE
of 12.6 1.38% (in the reverse scan). To reduce the effect of a
high series resistance, PSCs with thinner C60 interfacial layers
were tested but their performance remained inferior to the SnO2
device without passivation (Figure S5, Supporting Information).
These results demonstrate that inserting a C60 layer at the SnO2/
perovskite interface affords no effective passivation, but rather
restricts the electron extraction and induces nonradiative interfa-
cial recombination, lowering both the FF and Voc.
In a second step, we used an ≈15 nm PCBM interfacial
layer applied by solution processing from chlorobenzene
(10 mg mL−1). The PV characteristics are almost identical to
the SnO2 device without a passivation layer. A similar broad
distribution of PCEs is observed for both the reverse scan
(14.4
1.84%) and the forward scan (14.1
1.98%). How-
ever, thermal treatment of the PCBM layer (100 °C, 20 min.
under N2) prior to depositing the perovskite layer results in a
significant enhancement in performance and reproducibility.
Compared to the bare SnO2 device, the annealed PCBM device
displays a higher average PCE of 15.3% and a small standard
deviation of 0.84% in the reverse scan, due to an increased Jsc
of 21.7 0.56 mA cm−2, a comparable Voc of 1.09 0.01 V, and
an FF of 0.65 0.03. The PCE difference between both scan-
ning directions is negligible, by showing an average PCE of
15.5 0.86% in the forward scan. Apparently thermal annealing
leads to a passivated ETL interface that remains intact during
the perovskite solution processing. Possibly, thermal annealing
causes the ester side chains of PCBM to react with the hydroxyl
groups of the SnO2 surface, forming covalent bonds.
In accordance with this idea, a PCBA monolayer[47] spin coated
from a dilute (0.2 mg mL−1) solution in dichlorobenzene, exhibits
a distinctively improved passivation compared to PCBM or C60,
regardless of the thermal annealing process. Without thermal
annealing PCBA gives the best performance, showing a PCE of
16.5 0.84% in the reverse scan, with a reduced standard deviation
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (3 of 12)

www.advancedsciencenews.com
www.afm-journal.de
0
0
-4
a
b
SnO2, 15.3%
C60 wo, 14.7%
C60 w, 14.8%
PCBM wo, 15.8%
PCBM w, 16.9%
PCBA wo, 18.3%
PCBA w, 17.9%
Reverse
Forward
-4
-8
Slow sweep
-8
-12
-16
-20
-24
-12
-16
-20
-24
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Voltage (V)
Voltage (V)
100
25
20
15
10
5
25
20
15
10
5
d
c
21.92 mA cm-2
PCE = 18.8%
20
16
12
8
90
80
70
60
50
40
30
20
10
0
Jsc= 20.4 mA cm-2
4
0
0
0
400
500
600
700
800
0
200
400
Time (s)
600
800
Wavelength (nm)
Figure 3. a) Stabilized J–V curves of the best-performing ITO/SnO2/fullerene/FA0.66MA0.34PbI2.85Br0.15/Spiro-OMeTAD/MoO3/Au solar cells with
different passivation layers, with (w) or without (wo) thermal annealing. b) J–V curves and c) EQE spectra of the champion device with SnO2/PCBA
(not-annealed). d) Steady-state power and photocurrent output tracking over time of the champion device operated at the maximum power point under
the 100 mW cm−2 simulated AM1.5G illumination condition.
in a Jsc of 21.9 0.26 mA cm−2, a Voc of 1.09 0.01 V, and a sig-
nificantly improved FF of 0.69 0.03. Comparable high perfor-
mance is also obtained in the forward scan, with an average PCE
of 16.9 0.78%. This supports the idea that the higher reactivity of
the carboxylic acid side chain in PCBA toward the hydroxyl groups
of the SnO2 surface play a role in the improved passivation.
We note that the PCEs reported here are lower than the record
efficiencies,[29] but the clear trends in the statistical distributions of
device performance support the effectiveness of the fullerene pas-
sivation at the SnO2/perovskite interface. Figure 3a and Figure S6
in the Supporting Information display the stabilized current
density–voltage (J–V) curves and external quantum efficiency
(EQE) spectra of the champion (i.e., highest PCE) PSCs for the
different passivation conditions. The corresponding parameters
are summarized in Table 1. Similar to the distributions shown
in Figure 2, the devices based on not-annealed PCBA, annealed
PCBA, and annealed PCBM exhibit the best performance, with
PCEs of 18.3%, 17.9%, and 16.9%, respectively. The enhanced
performance of the PCBA modified devices is mainly a conse-
quence of the improved FF of 0.76, which suggests an improved
electron extraction, originating from an efficiently passivated
SnO2 ETL. The best-performing PSC device based on PCBA
Table 1. Photovoltaic parameters of the champion PSCs with different fullerene layers.
Passivation layer
SnO2 reference
C60 not-annealed
C60 annealed
Voca) [V]
1.10
1.03
1.04
1.09
1.08
1.10
1.06
FFa)
0.64
0.69
0.69
0.67
0.72
0.76
0.77
PCEa) [%]
15.4
PCEc) [%]
15.3
Jsca) [mA cm−2
22.0
]
Jscb) [mA cm−2
21.7
]
21.4
15.3
20.7
14.7
21.4
15.3
20.9
14.8
PCBM not-annealed
PCBM annealed
PCBA not-annealed
PCBA annealed
22.2
16.1
21.6
15.8
22.1
17.1
21.7
16.9
22.2
18.6
21.9
18.3
22.1
18.0
22.0
17.9
a)The data were extracted from stabilized J–V curves under simulated AM1.5G illumination (100 mW cm−2); b)Calculated by integrating the EQE spectrum with the AM1.5G
spectrum; c)Corrected PCE obtained by calculating the Jsc integrated from EQE spectrum and Voc and FF from the stabilized J–V measurement.
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (4 of 12)

www.advancedsciencenews.com
www.afm-journal.de
Figure 4. a) UV–vis–NIR spectra and b) XRD patterns of FA0.66MA0.34PbI2.85Br0.15 perovskite films deposited on different ETLs with (w) or without
(wo) thermal annealing. Top-view SEM images of perovskite films on c) SnO2, d) SnO2/C60 (annealed), e) SnO2/PCBM (not-annealed), f) SnO2/PCBM
(annealed), g) SnO2/PCBA (not-annealed), and h) SnO2/PCBA (annealed). Scale bars are 1 µm.
exhibits negligible hysteresis, by showing almost identical J–V
curves (Figure 3b) measured in reverse, forward, and stabilized
scans. The EQE measured with bias illumination provides an
estimated Jsc in close correspondence with the J–V data measured
with simulated AM1.5G (100 mW cm−2) illumination (Figure 3c).
Figure 3d shows the steady-state power and photocurrent output
diffraction (XRD) patterns of the as-prepared perovskite films
are virtually identical (Figure 4b). No significant shift in the dif-
fraction peak positions or intensities are observed, indicating
that the crystallinity of perovskites is largely independent on
the underlying fullerene layers. The primary diffraction peaks
at 14° and 28° are assigned to the (110) and (220) lattice planes
of the perovskite structure, respectively, and no significant trace
of PbI2 and nonperovskite phase can be identified. Figure 4c–h
displays the top-view scanning electron microscope (SEM)
images of the corresponding perovskite films deposited on dif-
ferent ETLs. All the layers exhibit a compact and pinhole-free
surface morphology, with an average grain size of ≈300 nm.
We attribute the independence of the perovskite morphology
and crystallinity on different substrates to the two-step solution
process utilized for depositing the perovskite layer, in which
the crystallization dynamics are affected by the inorganic PbI2
of the best device, measured at its maximum power point (VMPP
=
0.92 V) for 800 s. In this experiment the PCBA-modified device
exhibits a photocurrent of 20.4 mA cm−2 and a PCE of 18.8%,
which is very close to the values extracted from the J–V curves.
To validate our hypothesis that the device performance is
determined by the electron transport properties of the ETL, we
investigated the quality of the perovskite films deposited on dif-
ferent fullerene passivated SnO2 substrates. All perovskite films
display similar UV–vis–NIR absorption spectra, with a char-
acteristic onset at around 800 nm (Figure 4a). Also, the X-ray
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (5 of 12)

www.advancedsciencenews.com
www.afm-journal.de
framework rather than the ETL substrate.[33,68] In addition, com-
parable results are obtained from the steady-state photolumines-
cence (PL) spectra and light-intensity dependent measurements
(Figures S7 and S8, Supporting Information), showing that the
bulk properties of perovskite layers remain unchanged on dif-
ferent ETLs. Accordingly, we rule out the impact of perovskite
bulk properties on the variation trend of device performance.
To gain insight into the distribution of the fullerene in the
passivation layers, we performed XPS depth profiling. We
checked that the XPS depth profiles of pristine SnO2 layers
show no significant amounts of carbon (<1 at%) after UV–ozone
cleaning (Figure S9, Supporting Information). Also, after subse-
quent rinsing with N,N-dimethylformamide (DMF) the carbon
concentration remains negligible. Atomic force microscopy
revealed that the root-mean-square average surface roughness
of the nonpassivated and fullerene-passivated SnO2 layers was
less than 1 nm (Figure S10, Supporting Information). Hence,
by monitoring the carbon signal of fullerene-passivated SnO2
layers we can determine the fullerene concentration profile.
Figure 5a,b shows the XPS depth profiles of SnO2 substrates
with annealed and not-annealed PCBM layers after rinsing with
DMF. In both cases, the tin concentration increases abruptly
from the surface (t = 0 s), suggesting that DMF has removed
most of the capping PCBM layer. Meanwhile, the carbon con-
centration of the annealed PCBM substrate is remarkably
higher (29 at%) than that of the not-annealed PCBM substrate
(11 at%). From the attenuated Sn 3d signal (Figure 5e),[49] we
estimate the thickness of the annealed and not-annealed PCBM
layers at the SnO2 interface to be 1.3 and 0.2 nm, respectively.
This implies that without thermal annealing, the interfacial
PCBM layer is almost entirely removed by DMF, and no pas-
sivation can be developed at the SnO2 interface; however, with
annealing a PCBM-monolayer-passivated SnO2 interface is
formed, which remains intact during the solution processing.
Interestingly, both SnO2/PCBM substrates exhibit a plateau
of the carbon signal (>7 at%) during the sputtering process.
This suggests that PCBM infiltrates into the porous SnO2 layer
and that DMF does not remove such penetrated PCBM. These
SnO2 layers have been shown to have pore sizes on the order
of 10 nm.[42] On the other hand, for the annealed and not-
annealed PCBA-modified SnO2 substrates, after DMF rinsing,
the carbon concentrations at the surface (t = 0 s) are 43 and
40 at% (Figure 5c,d), much higher than their PCBM counter-
part. From Figure 5e, the thicknesses of the interfacial PCBA
layers are estimated to be 1.9 nm (annealed) and 1.6 nm (not-
annealed), respectively, indicating that a PCBA monolayer with
better coverage is obtained at the SnO2 interface. It is also
found that when compared to the nonpassivated SnO2, the shift
in the binding energies of Sn 3d core level for both the PCBA
modified SnO2 (>0.14 eV) is higher than that for the annealed-
and not-annealed-PCBM modified SnO2 (0.10 and 0.03 eV),
respectively. Here, the positive shift of the Sn 3d peaks is attrib-
uted to the decreased electron density near the Sn atoms, which
could be induced by the bonding between the SnO2 surface and
the fullerenes.[39,69] The results confirms the higher reactivity of
the carboxylic acid functional group of PCBA than the methyl
shown in Figure S11 in the Supporting Information. The depth
profiles (Figure 5c,d) also show that PCBA penetrates less deep
into the SnO2 layer than PCBM, which we attribute to the use
of a diluted PCBA solution (0.2 mg mL−1) compared to the
PCBM solution (10 mg mL−1). The fact that PCBA penetrates
less deep into the SnO2 layer as compared to PCBM, while at
the same time the PCBA concentration at the SnO2 surface is
higher, confirms that the carbon signals for the PCBM samples
are unlikely due to ion knock-on effects. Together with the
PV performance derived from the PCBA and PCBM modified
ETLs, we conclude that only the fullerenes located at the SnO2/
perovskite interface are essential for the interfacial passivation
process.
XPS depth profiles for the C60-modified SnO2 substrates after
DMF rinsing (Figure S12, Supporting Information) show that
the carbon concentration only starts to decrease after a long-
sputtering process when the as-deposited C60 layer is above
15 nm. This indicates that the thermally evaporated C60 layer is
resilient against DMF solvent treatment. However, for the 5 nm
thick C60 layer, the carbon concentration at the surface (t = 0 s)
is reduced to 18 at%, only slightly higher than that of the not-
annealed PCBM substrate (11 at%). Compared to the device
with not-annealed PCBM, the 5 nm thick C60 modified device
exhibits decreased PV performance (Figure S5, Supporting
Information).
To verify the observations from the XPS analysis, UV–vis–NIR
absorption spectra were measured for different fullerene modi-
fied SnO2 layers deposited on quartz substrates before and after
DMF rinsing (Figure 5f). As expected, the SnO2/C60 (15 nm)
substrate only displays a slight decrease in absorbance due
to the high resilience of the C60 film against DMF treatment.
In contrast, a significant reduction in absorbance is observed
for both SnO2/PCBM substrates with (w) or without (wo)
annealing, since DMF washes away most of the capping PCBM
layer. However, it is found that the absorbance of the annealed
PCBM is still higher than that of the not-annealed PCBM, which
confirms that more PCBM is retained on the substrate when
thermally annealed. For PCBA-modified SnO2 substrates, the
absorbance before DMF rinsing is already low due to the dilute
(0.2 mg mL−1) PCBA solution used. After DMF rinsing both
the annealed and not-annealed PCBA substrates show a neg-
ligible reduction in absorbance, confirming that DMF rinsing
removes very little PCBA. Among the DMF-washed PCBM and
PCBA passivation layers, the annealed PCBM layer exhibits the
highest absorbance. This is attributed to the PCBM infiltrated
into the SnO2 layer that cannot be removed by DMF, as is
evidenced by the XPS depth profiles (Figure 5a,b).
To explore the impact of PCBM (annealed) and PCBA mono-
layers on the passivation of the SnO2 ETL, their energy band
structures were characterized by ultraviolet photoelectron spec-
troscopy (UPS). Figure 6a depicts that the secondary electron
edge of the SnO2 layer shifts to lower binding energies after
PCBM and PCBA passivation. This corresponds to a significant
increase in the work function from 3.58 eV for the pristine SnO2,
to 3.80 and 4.00 eV for the PCBM- and PCBA-passivated SnO2,
respectively. From the onset of the valence band spectrum,
the ionization potential of the SnO2, SnO2/PCBM, and SnO2/
PCBA films are calculated to be 7.60, 6.11, and 6.28 eV, respec-
tively. Figure 6b illustrates a schematic energy level diagram

ester side of PCBM toward the hydroxyl ( OH) enriched SnO2
surface in forming stable monolayers. The corresponding
high-resolution XPS spectra of the C 1s and O 1s signals are
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (6 of 12)

www.advancedsciencenews.com
www.afm-journal.de
70
70
60
50
40
30
20
10
0
a
b
SnO2/PCBM wo + DMF
SnO2/PCBM w + DMF
60
50
40
30
20
10
0
C 1s
C 1s
O 1s
Si 2p
Sn 3d
O 1s
Si 2p
Sn 3d
0
50
100
150
200
250
0
50
100
150
200
250
Time (sec)
Time (sec)
70
60
50
40
30
20
10
0
70
60
50
40
30
20
10
0
d
c
SnO2/PCBA wo + DMF
SnO2/PCBA w + DMF
C 1s
C 1s
O 1s
Si 2p
Sn 3d
O 1s
Si 2p
Sn 3d
0
50
100
150
200
250
0
50
100
150
200
250
Time (s)
Time (s)
e
f
0.09
Sn 3d
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
PCBM w
PCBM w + DMF
PCBM wo
PCBM wo + DMF
PCBA w
PCBA w + DMF
PCBA wo
PCBA wo + DMF
C60
C60 + DMF
SnO2
PCBM w + DMF
PCBM wo + DMF
PCBA w + DMF
PCBA wo + DMF
0.06
0.03
SnO2 control + DMF
300
400
300
400
500
600
700
800
900
480
485
490
495
500
Binding energy (eV)
Wavelength (nm)
Figure 5. XPS depth profiles of a) SnO2/PCBM (annealed), b) SnO2/PCBM (not-annealed), c) SnO2/PCBA (annealed), and d) SnO2/PCBA (not-
annealed) ETLs deposited on glass substrates. Layers were rinsed by DMF solvent shortly before the XPS experiments. e) XPS high-resolution scan
spectra of the Sn 3d core level for different ETL substrate surfaces. f) UV–vis–NIR absorption spectra of different fullerene-passivated SnO2 ETLs
deposited on quartz substrates, measured before and after DMF rinsing. The initial SnO2 layer thickness was ≈20 nm for all depth profiles.
for different ETLs, assuming a fixed Fermi level for all the
measurements.[70] Consistent with the shift in Sn 3d binding
energies (Figure 5e), the increase in work function of PCBM-
and PCBA-passivated SnO2 ETLs can be attributed to the forma-
tion of surface dipoles, which induce an electric field that accel-
erates the charge collection at the SnO2 interface and reduces
the recombination losses.[33,71] Figure 6c illustrates that com-
pared to the pristine SnO2, the PCBM- and PCBA-passivated
SnO2 ETLs exhibit better conduction band alignment with the
perovskite, which could also be beneficial for the interfacial
charge transport.
Based on the XPS analysis, one might expect that introduc-
tion of a better-covered PCBM layer at the SnO2 interface would
improve the device performance. Alternatively, an interfacial
chemical reaction between the PCBM and SnO2 could be of
vital importance for improving the charge extraction. To gain
further understanding of the role of PCBM interlayer, we intro-
duced a solvent-resistant PCBM layer on the SnO2. Following
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (7 of 12)

www.advancedsciencenews.com
www.afm-journal.de
a
b
SnO2
SnO2/PCBM
SnO2/PCBA
Vacuum level
-4.00 eV
-3.80 eV
-2.31 eV
-3.58 eV
EF
-2.28 eV
SnO2/PCBA
-4.02 eV
SnO2/PCBM
SnO2
19
18
17 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0
Binding Energy (eV)
c
-3.8
-3.9
-4.1
-6.3
-4.1
-4.7
-5.6
-6.1
-7.6
Figure 6. a) UPS measurements of SnO2, SnO2/PCBM, and SnO2/PCBA, respectively. All samples were rinsed by DMF before the measurement.
b) Energy diagram of different ETL substrates, assuming a fixed Fermi level. c) Schematic conduction band diagram of perovskite film and different
ETL substrates.
the methods reported by Edman and co-workers,[62–64] we
exposed an as-prepared PCBM layer to UV light for some time,
before DMF or perovskite solution casting. UV light creates a
robust and less-soluble dimeric state of PCBM, yet with retained
electronic properties. According to the XPS depth profiles,
when using 10 mg mL−1 PCBM and UV light (Figure 7a,b),
both DMF rinsed SnO2/PCBM-dimer layers (w/wo thermal
annealing) exhibit much higher carbon concentrations at the
surface (>70 at%) than the SnO2/PCBM reference (Figure 5a,b)
due to a thicker PCBM layer on the SnO2. This observation is
well supported by the UV–vis–NIR measurement (Figure S13,
Supporting Information) and the thickness of the capping
PCBM layer (12 nm), measured by surface profilometry. Using
a higher (20 mg mL−1) PCBM concentration in combination
with UV illumination, the depth profiles of both SnO2/PCBM-
dimer layers (w/wo thermal annealing) are similar to that of the
SnO2/C60 (35 nm) layer (Figure S12, Supporting Information)
and show a constant carbon concentration of 100 at% during
entire the sputtering process. Independent of w/wo thermal
annealing, the PCBM thickness was 23 nm. This indicates the
successful fabrication of a DMF-resilient SnO2/PCBM bilayered
ETL in which the thickness and the chemical interaction of the
capping PCBM layer can be tuned, by modifying the PCBM
solution concentration and the thermal annealing treatment.
Figure 7a,b does not show a clear peak for the Sn and also the
onset of the Si signal is much less clear than in Figure 5 for
PCBM layers that were treated without UV light. We think that
this is due to the fact that after UV illumination the rate of
removal of the thin (12 nm) PCBM is drastically reduced and
the Sn and Si signals results from regions below the top PCBM
layer. This explanation is consistent with Figure 7c,d, where
the thicker (23 nm) UV-treated PCBM layer is not removed and
prevents observing Sn and Si below.
Figure 8 shows the J–V curves of PSC devices prepared
with different concentrations of PCBM (5–20 mg mL−1).
Without exposure to UV light, all devices deliver virtually
identical PCEs of 16.0% with negligible hysteresis when
the PCBM layer is thermally annealed (Figure 8a and Table
S6, Supporting Information). For low PCBM concentra-
tions (≤10 mg mL−1) only the PCBM chemically bonded to
the SnO2 is preserved at the interface. For the highest PCBM
concentration (20 mg mL−1) DMF rinsing does not wash
away the entire PCBM layer but leaves an ≈13 nm film as
determined by profilometry and confirmed by XPS depth
profiling (Figure S14, Supporting Information). The results
suggest that the passivation capacity of PCBM at the SnO2
interface is largely independent of the concentrations of the
PCBM solution and that, next to a surface bound monolayer,
a thin (≈13 nm) residual PCBM layer offers no additional
improvement.
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (8 of 12)

www.advancedsciencenews.com
www.afm-journal.de
100
100
90
80
70
60
50
40
30
20
10
0
a
b
PCBM 10 mg w + UV 30 min + DMF
PCBM 10 mg wo + UV 30 min + DMF
90
80
70
60
50
40
30
20
10
0
C 1s
C 1s
O 1s
Si 2p
Sn 3d
O 1s
Si 2p
Sn 3d
0
100
200
300
400
500
0
100
200
300
400
500
Time (s)
Time (s)
c
d
100
90
80
70
60
50
40
30
20
10
0
100
90
80
70
60
50
40
30
20
10
0
PCBM 20 mg w + UV 30 min+ DMF
PCBM 20 mg wo + UV 30 min + DMF
C 1s
C 1s
O 1s
Si 2p
Sn 3d
O 1s
Si 2p
Sn 3d
0
100
200
300
400
500
0
100
200
300
400
500
Time (s)
Time (s)
Figure 7. XPS depth profiles of glass/SnO2 substrates passivated by 10 mg mL−1 PCBM a,b) with (w) or without (wo) thermal annealing and 20 mg mL−1
PCBM c–d) w/wo thermal annealing, respectively. All the samples were illuminated under UV light for 30 min in the N2 atmosphere, before washing
by DMF solvent. The initial SnO2 layer thickness was ≈20 nm for all depth profiles.
In contrast, when using UV illumination to create PCBM
dimers at the SnO2 surface, a strong dependence of the
PCE on the PCBM concentration is found, both with and
without thermal annealing (Figure S15 and Tables S7 and
S8, Supporting Information). At low PCBM concentrations
(e.g., 10 mg mL−1), thermally annealed PCBM-dimer layers
provide high PCEs of ≈16.4%, comparable to devices without
UV exposure, indicating that a thin PCBM-dimer layer gives
good passivation (Figure 8b). Strikingly, the devices with
not-annealed PCBM-dimer layer exhibit much inferior perfor-
mance (PCE ≈ 14%; Figure 8b) despite the good coverage of the
fullerene on the SnO2 substrate (Figure 7b). We conclude that
0
0
a
b
PCBM x mg mL-1, annealed
PCBM 10 mg mL-1, UV 30 min
Annealed, reverse
5 mg, reverse
5 mg, forward
10 mg, reverse
10 mg, forward
15 mg, reverse
15 mg, forward
20 mg, reverse
20 mg, forward
-4
-8
-4
Annealed, forward
Not annealed, reverse
Not annealed, forward
-8
-12
-16
-20
-24
-12
-16
-20
-24
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0.0
0.2
0.4
0.6
0.8
1.0
Voltage (V)
Voltage (V)
Figure 8. J–V curves of ITO/SnO2/fullerene/FA0.66MA0.34PbI2.85Br0.15/Spiro-OMeTAD/MoO3/Au solar cells with SnO2/PCBM ETLs. a) For cells where
the PCBM concentration in the spin coating solution varied from 5 to 20 mg mL−1 and the layers was thermally annealed (100 °C, 20 min). b) For cells
made from a 10 mg mL−1 PCBM solution with or without thermal annealing and with UV-light illumination before the perovskite deposition.
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (9 of 12)

www.advancedsciencenews.com
www.afm-journal.de
the thermally induced chemical interaction between the PCBM
and SnO2 is the most influential factor in enhancing the inter-
facial charge transport process.
passivation layer between the photoactive and charge transport
layers. The new insights into the interface passivation of metal
oxide charge transport layer offers a strategy for the future
development toward high-efficiency PSCs.
As the concentration of PCBM solution increases to
20 mg mL−1, for both cells made with or without thermal
annealing, the corresponding J–V curves show strong s-shapes
with reduced Voc, Jsc, and FF, yielding PCEs lower than 10%
(Figure S15 and Tables S7 and S8, Supporting Information).
Again, we exclude an effect of perovskite bulk films on the
device performance, since they present virtually identical XRD
patterns and SEM images on different PCBM-dimer layers
(Figures S16 and S17, Supporting Information). In conductivity
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
measurements for different ETLs (Figure S18 and Table S9, Acknowledgements
Supporting Information), we found that compared to other thin
The authors thank Dr. Mengmeng Li for the help with the AFM
PCBM layers, the conductivity extracted from the J–V curve is
reduced when applying a slightly thicker (20 mg mL−1) PCBM-
dimer layer on the SnO2 substrate. This increased resistivity of
the PCBM-dimer layer limits the charge transport and induces
nonradiative recombination at the SnO2/perovskite interface,
resulting in the reduced device performance. Collectively,
our observations imply that the fabrication of a robust SnO2/
fullerene ETL does not necessarily improve the SnO2 passiva-
tion and that the introduction of a fullerene monolayer with
controlled chemical interaction with the SnO2 ETL is sufficient
to boost the PSC device performance.
measurement. The authors acknowledge funding from the European
Research Council under the European Union’s Seventh Framework
Programme (FP/2007-2013) (ERC Grant Agreement No. 339031), the
Netherlands Organization for Scientific Research (NWO) (project
680-91-011), the Ministry of Education, Culture and Science (Gravity
program 024.001.035), and the NWO Spinoza prize awarded to R.A.J.
Janssen by the Netherlands Organization for Scientific Research (NWO).
Conflict of Interest
The authors declare no conflict of interest.
3. Conclusion
Keywords
We performed a comprehensive study correlating the chemical
interactions at the SnO2/fullerene interface with the PSC device
performance. XPS depth profile measurements are a useful
tool to monitor the precise location of thin-film fullerene resi-
dues in the SnO2 ETL. In the case of depositing PCBM from
≤10 mg mL−1 solutions, thermal annealing is essential to create
a surface-bound PCBM monolayer that is resilient to depositing
a perovskite layer on top. The surface-bound PCBM monolayer
significantly improved the device performance and repro-
ducibility. Although PCBM partly infiltrates into the porous
SnO2 layer, only the fullerenes located at the SnO2/perovskite
interface are essential for electron extraction. The passivation
capacity of PCBM is not affected by the concentration of the
solution from which it is deposited because PCBM that is not
bound to the surface is removed by DMF in subsequent pro-
cessing. While thermally evaporated C60 showed high resilience
against DMF, the performance of the corresponding PSC was
inferior to the thermally annealed PCBM devices. This is attrib-
uted to the absence of a chemical bond between C60 and the
SnO2 surface. In accordance, the carboxylic acid-based PCBA
derivative exhibited much higher passivation efficiency. A
PCBA monolayer with excellent coverage and DMF-resilience
could be formed from dilute (0.2 mg mL−1) solutions without
any thermal treatment. After optimization, the best-performing
device using a PCBA-modified SnO2 ETL exhibited a PCE of
18.8%.
fullerene, metal halide perovskite, passivation, solar cell, tin oxide
Received: July 20, 2019
Revised: August 21, 2019
Published online:
[1] NREL Chart, https://www.nrel.gov/pv/assets/pdfs/best-research-
cell-efficiencies-190416.pdf (accessed: June 2019).
[2] M. Saliba, J.-P. Correa-Baena, C. M. Wolff, M. Stolterfoht, N. Phung,
S. Albrecht, D. Neher, A. Abate, Chem. Mater. 2018, 30, 4193.
[3] M. Saliba, T. Matsui, J. Y. Seo, K. Domanski, J. P. Correa-Baena,
M. K. Nazeeruddin, S. M. Zakeeruddin, W. Tress, A. Abate,
A. Hagfeldt, M. Gratzel, Energy Environ. Sci. 2016, 9, 1989.
[4] M. Saliba, T. Matsui, K. Domanski, J. Seo, A. Ummadisingu,
S. M. Zakeeruddin, J. P. Correa Baena, W. R. Tress, A. Abate,
A. Hagfeldt, M. Graetzel, Science 2016, 354, 206.
[5] W. S. Yang, B. W. Park, E. H. Jung, N. J. Joen, Y. C. Kim, D. U. Lee,
S. S. Shin, J. Seo, E. K. Kim, J. H. Noh, S. I. Seok, Science 2017, 356,
1376.
[6] D. Luo, W. Yang, Z. Wang, A. Sadhanala, Q. Hu, R. Su, R. Shivanna,
G. F. Trindade, J. F. Watts, Z. Xu, T. Liu, K. Chen, F. Ye, P. Wu,
L. Zhao, J. Wu, Y. Tu, Y. Zhang, X. Yang, W. Zhang, R. H. Friend,
Q. Gong, H. J. Snaith, R. Zhu, Science 2018, 360, 1442.
[7] Q. Jiang, Z. Chu, P. Wang, X. Yang, H. Liu, Y. Wang, Z. Yin, J. Wu,
X. Zhang, J. You, Adv. Mater. 2017, 29, 1703852.
[8] H. Tan, A. Jain, O. Voznyy, X. Lan, F. P. G. De Arquer, J. Z. Fan,
R. Quintero-Bermudez, M. Yuan, B. Zhang, Y. Zhao, F. Fan, P. Li,
L. N. Quan, S. Hoogland, E. H. Sargent, Science 2017, 355, 722.
[9] Y. Galagan, J. Phys. Chem. Lett. 2018, 9, 4326.
Interfaces are of significant importance for the PSCs based
on metal-oxide layers. Depending on the fabrication method,
the passivation of metal oxide is often found critical for high-
performing devices. Our study explored the essence of the
[10] J. Wang, F. Di Giacomo, J. Brüls, H. Gorter, I. Katsouras, P. Groen,
R. A. J. Janssen, R. Andriessen, Y. Galagan, Sol. RRL 2017, 1, 1700091.
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (10 of 12)

www.advancedsciencenews.com
www.afm-journal.de
[11] S. T. Williams, A. Rajagopal, C. C. Chueh, A. K. Jen, J. Phys. Chem.
Lett. 2016, 7, 811.
[12] Y. Rong, Y. Hu, A. Mei, H. Tan, M. I. Saidaminov, S. I. Seok,
M. D. McGehee, E. H. Sargent, H. Han, Science 2018, 361,
eaat8235.
[38] L. Kegelmann, C. M. Wolff, C. Awino, F. Lang, E. L. Unger, L. Korte,
T. Dittrich, D. Neher, B. Rech, S. Albrecht, ACS Appl. Mater.
Interfaces 2017, 9, 17245.
[39] D. Yang, R. Yang, K. Wang, C. Wu, X. Zhu, J. Feng, X. Ren, G. Fang,
S. Priya, S. F. Liu, Nat. Commun. 2018, 9, 3239.
[13] H. Kim, K.-G. Lim, T.-W. Lee, Energy Environ. Sci. 2016, 9, 12.
[14] M. Liu, M. B. Johnston, H. J. Snaith, Nature 2013, 501, 395.
[15] S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou,
M. J. P. Alcocer, T. Leijtens, L. M. Herz, A. Petrozza, H. J. Snaith,
Science 2013, 342, 341.
[16] M. A. Green, A. Ho-Baillie, H. J. Snaith, Nat. Photonics 2014,
8, 506.
[17] G. Xing, N. Mathews, S. Sun, S. S. Lim, Y. M. Lam, M. Graetzel,
S. Mhaisalkar, T. C. Sum, Science 2013, 342, 344.
[40] S. P. Pujari, L. Scheres, A. T. Marcelis, H. Zuilhof, Angew. Chem.,
Int. Ed. 2014, 53, 6322.
[41] W.-Q. Wu, D. Chen, R. A. Caruso, Y.-B. Cheng, J. Mater. Chem. A
2017, 5, 10092.
[42] Z. Tebby, T. Uddin, Y. Nicolas, C. Olivier, T. Toupance, C. Labrugere,
L. Hirsch, ACS Appl. Mater. Interfaces 2011, 3, 1485.
[43] K. Wojciechowski, S. D. Stranks, A. Abate, G. Sadoughi,
A. Sadhanala, N. Kopidakis, G. Rumbles, C.-Z. Li, R. H. Friend,
A. K.-Y. Jen, H. J. Snaith, ACS Nano 2014, 8, 12701.
[18] M. A. Haque, A. D. Sheikh, X. Guan, T. Wu, Adv. Energy Mater. 2017,
7, 1602803.
[19] A. Mingorance, H. Xie, H.-S. Kim, Z. Wang, M. Balsells,
A. Morales-Melgares, N. Domingo, N. Kazuteru, W. Tress,
J. Fraxedas, N. Vlachopoulos, A. Hagfeldt, M. Lira-Cantu, Adv.
Mater. Interfaces 2018, 5, 1800367.
[44] C. Wang, D. Zhao, C. R. Grice, W. Liao, Y. Yu, A. Cimaroli,
N. Shrestha, P. J. Roland, J. Chen, Z. Yu, P. Liu, N. Cheng,
R. J. Ellingson, X. Zhao, Y. Yan, J. Mater. Chem. A 2016, 4,
12080.
[45] C. Wang, D. Zhao, Y. Yu, N. Shrestha, C. R. Grice, W. Liao,
A. J. Cimaroli, J. Chen, R. J. Ellingson, X. Zhao, Y. Yan, Nano Energy
2017, 35, 223.
[46] C. Wang, C. Xiao, Y. Yu, D. Zhao, R. A. Awni, C. R. Grice,
K. Ghimire, I. Constantinou, W. Liao, A. J. Cimaroli, P. Liu, J. Chen,
N. J. Podraza, C.-S. Jiang, M. M. Al-Jassim, X. Zhao, Y. Yan, Adv.
Energy Mater. 2017, 7, 1700414.
[47] Y. Dong, W. Li, X. Zhang, Q. Xu, Q. Liu, C. Li, Z. Bo, Small 2016, 12,
1098.
[20] L. Xiong, Y. Guo, J. Wen, H. Liu, G. Yang, P. Qin, G. Fang, Adv.
Funct. Mater. 2018, 28, 1802757.
[21] W. Ke, G. Fang, Q. Liu, L. Xiong, P. Qin, H. Tao, J. Wang, H. Lei,
B. Li, J. Wan, G. Yang, Y. Yan, J. Am. Chem. Soc. 2015, 137, 6730.
[22] Q. Jiang, X. Zhang, J. You, Small 2018, 14, 1801154.
[23] Q. Jiang, L. Zhang, H. Wang, X. Yang, J. Meng, H. Liu, Z. Yin, J. Wu,
X. Zhang, J. You, Nat. Energy 2017, 2, 16177.
[24] Y. Lee, S. Lee, G. Seo, S. Paek, K. T. Cho, A. J. Huckaba,
M. Calizzi, D. W. Choi, J. S. Park, D. Lee, H. J. Lee, A. M. Asiri,
M. K. Nazeeruddin, Adv. Sci. 2018, 5, 1800130.
[48] K. K. Wong, A. Fakharuddin, P. Ehrenreich, T. Deckert,
M. Abdi-Jalebi, R. H. Friend, L. Schmidt-Mende, J. Phys. Chem. C
2018, 122, 10691.
[25] J. P. Correa Baena, L. Steier, W. Tress, M. Saliba, S. Neutzner,
T. Matsui, F. Giordano, T. J. Jacobsson, A. R. Srimath Kandada,
S. M. Zakeeruddin, A. Petrozza, A. Abate, M. K. Nazeeruddin,
M. Grätzel, A. Hagfeldt, Energy Environ. Sci. 2015, 8, 2928.
[26] M. F. Ayguler, A. G. Hufnagel, P. Rieder, M. Wussler, W. Jaegermann,
T. Bein, V. Dyakonov, M. L. Petrus, A. Baumann, P. Docampo, ACS
Appl. Mater. Interfaces 2018, 10, 11414.
[49] Q. An, P. Fassl, Y. J. Hofstetter, D. Becker-Koch, A. Bausch,
P. E. Hopkinson, Y. Vaynzof, Nano Energy 2017, 39, 400.
[50] Y. Zhang, P. Wang, X. Yu, J. Xie, X. Sun, H. Wang, J. Huang, L. Xu,
C. Cui, M. Lei, D. Yang, J. Mater. Chem. A 2016, 4, 18509.
[51] K. Wojciechowski, T. Leijtens, S. Siprova, C. Schlueter,
M. T. Horantner, J. T. Wang, C. Z. Li, A. K.-Y. Jen, T. L. Lee,
H. J. Snaith, J. Phys. Chem. Lett. 2015, 6, 2399.
[27] Y. Kuang, V. Zardetto, R. van Gils, S. Karwal, D. Koushik,
M. A. Verheijen, L. E. Black, C. Weijtens, S. Veenstra, R. Andriessen,
W. M. M. Kessels, M. Creatore, ACS Appl. Mater. Interfaces 2018, 10,
30367.
[28] E. H. Anaraki, A. Kermanpur, L. Steier, K. Domanski, T. Matsui,
W. Tress, M. Saliba, A. Abate, M. Grätzel, A. Hagfeldt,
J.-P. Correa-Baena, Energy Environ. Sci. 2016, 9, 3128.
[29] Q. Jiang, Y. Zhao, X. Zhang, X. Yang, Y. Chen, Z. Chu, Q. Ye, X. Li,
Z. Yin, J. You, Nat. Photonics 2019, 13, 460.
[30] K. Choi, J. Lee, H. I. Kim, C. W. Park, G.-W. Kim, H. Choi, S. Park,
S. A. Park, T. Park, Energy Environ. Sci. 2018, 11, 3238.
[31] S. A. L. Weber, I. M. Hermes, S.-H. Turren-Cruz, C. Gort,
V. W. Bergmann, L. Gilson, A. Hagfeldt, M. Graetzel, W. Tress,
R. Berger, Energy Environ. Sci. 2018, 11, 2404.
[52] C. Tao, S. Neutzner, L. Colella, S. Marras, A. R. Srimath Kandada,
M. Gandini, M. D. Bastiani, G. Pace, L. Manna, M. Caironi,
C. Bertarelli, A. Petrozza, Energy Environ. Sci. 2015, 8, 2365.
[53] C. Tao, J. Van Der Velden, L. Cabau, N. F. Montcada, S. Neutzner,
A. R. Srimath Kandada, S. Marras, L. Brambilla, M. Tommasini,
W. Xu, R. Sorrentino, A. Perinot, M. Caironi, C. Bertarelli,
E. Palomares, A. Petrozza, Adv. Mater. 2017, 29, 1604493.
[54] B. L. Watson, N. Rolston, K. A. Bush, T. Leijtens, M. D. McGehee,
R. H. Dauskardt, ACS Appl. Mater. Interfaces 2016, 8, 25896.
[55] W. Qiu, J. P. Bastos, S. Dasgupta, T. Merckx, I. Cardinaletti,
M. V. C. Jenart, C. B. Nielsen, R. Gehlhaar, J. Poortmans,
P. Heremans, I. McCulloch, D. Cheyns, J. Mater. Chem. A 2017, 5,
2466.
[56] X. Liu, Y. Zhang, L. Shi, Z. Liu, J. Huang, J. S. Yun, Y. Zeng, A. Pu,
K. Sun, Z. Hameiri, J. A. Stride, J. Seidel, M. A. Green, X. Hao, Adv.
Energy Mater. 2018, 8, 1800138.
[32] I. M. Hermes, Y. Hou, V. W. Bergmann, C. J. Brabec, S. A. L. Weber,
J. Phys. Chem. Lett. 2018, 9, 6249.
[33] L. Zuo, Q. Chen, N. De Marco, Y. T. Hsieh, H. Chen, P. Sun,
S. Y. Chang, H. Zhao, S. Dong, Y. Yang, Nano Lett. 2017, 17, 269.
[34] M. Stolterfoht, C. M. Wolff, J. A. Márquez, S. Zhang, C. J. Hages,
D. Rothhardt, S. Albrecht, P. L. Burn, P. Meredith, T. Unold,
D. Neher, Nat. Energy 2018, 3, 847.
[57] T. Bu, J. Li, F. Zheng, W. Chen, X. Wen, Z. Ku, Y. Peng, J. Zhong,
Y. B. Cheng, F. Huang, Nat. Commun. 2018, 9, 4609.
[58] B. Tu, Y. Shao, W. Chen, Y. Wu, X. Li, Y. He, J. Li, F. Liu, Z. Zhang,
Y. Lin, X. Lan, L. Xu, X. Shi, A. M. C. Ng, H. Li, L. W. Chung,
A. B. Djurisic, Z. He, Adv. Mater. 2019, 31, 1805944.
[35] K. Liu, S. Chen, J. Wu, H. Zhang, M. Qin, X. Lu, Y. Tu, Q. Meng,
X. Zhan, Energy Environ. Sci. 2018, 11, 3463.
[59] H. Yoon, S. M. Kang, J.-K. Lee, M. Choi, Energy Environ. Sci. 2016,
9, 2262.
[36] T. Gatti, E. Menna, M. Meneghetti, M. Maggini, A. Petrozza,
F. Lamberti, Nano Energy 2017, 41, 84.
[37] W. Ke, D. Zhao, C. Xiao, C. Wang, A. J. Cimaroli, C. R. Grice,
M. Yang, Z. Li, C.-S. Jiang, M. Al-Jassim, K. Zhu, M. G. Kanatzidis,
G. Fang, Y. Yan, J. Mater. Chem. A 2016, 4, 14276.
[60] K. Wojciechowski, I. Ramirez, T. Gorisse, O. Dautel, R. Dasari,
N. Sakai, J. M. Hardigree, S. Song, S. Marder, M. Riede, G. Wantz,
H. J. Snaith, ACS Energy Lett. 2016, 1, 648.
[61] S. Song, R. Hill, K. Choi, K. Wojciechowski, S. Barlow, J. Leisen,
H. J. Snaith, S. R. Marder, T. Park, Nano Energy 2018, 49, 324.
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (11 of 12)

www.advancedsciencenews.com
www.afm-journal.de
[62] J. Wang, C. Larsen, T. Wågberg, L. Edman, Adv. Funct. Mater. 2011,
21, 3723.
[63] J. Enevold, C. Larsen, J. Zakrisson, M. Andersson, L. Edman, Nano
Lett. 2018, 18, 540.
[67] K. H. Hendriks, J. J. van Franeker, B. J. Bruijnaers, J. A. Anta,
M. M. Wienk, R. A. J. Janssen, J. Mater. Chem. A 2017, 5, 2346.
[68] W. S. Yang, J. H. Noh, N. J. Joen, Y. C. Kim, S. Ryu, J. Seo, S. I. Seok,
Science 2015, 348, 1234.
[64] A. Dzwilewski, T. Wågberg, L. Edman, J. Am. Chem. Soc. 2009, 131,
4006.
[65] W. Qiu, A. Ray, M. Jaysankar, T. Merckx, J. P. Bastos, D. Cheyns,
R. Gehlhaar, J. Poortmans, P. Heremans, Adv. Funct. Mater. 2017,
27, 1700920.
[69] J. Tian, Q. Xue, X. Tang, Y. Chen, N. Li, Z. Hu, T. Shi, X. Wang,
F. Huang, C. J. Brabec, H. L. Yip, Y. Cao, Adv. Mater. 2019, 31,
1901152.
[70] S. Zhang, M. Stolterfoht, A. Armin, Q. Lin, F. Zu, J. Sobus, H. Jin,
N. Koch, P. Meredith, P. L. Burn, D. Neher, ACS Appl. Mater. Inter-
faces 2018, 10, 21681.
[66] O. Pachoumi, A. A. Bakulin, A. Sadhanala, H. Sirringhaus,
R. H. Friend, Y. Vaynzof, J. Phys. Chem.
18945.
C
2014, 118,
[71] Y. Vaynzof, D. Kabra, L. Zhao, P. K. H. Ho, A. T. S. Wee, R. H. Friend,
Appl. Phys. Lett. 2010, 97, 033309.
©
Adv. Funct. Mater. 2019, 1905883
2019 The Authors. Published by WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1905883 (12 of 12)