Adjusting the Introduction of Cations for Highly Efficient and Stable Perovskite Solar Cells Based on (FAPbI3)0.9(FAPbBr3)0.1

Article abstract of DOI:10.1002/cssc.201800658

Although the power conversion efficiency (PCE) of perovskite solar cells (PSCs) has increased to 22.7 %, the instability when exposed to moisture and heat has hindered their further practical development. In this study, to gain highly efficient and stable perovskite components, methylammonium (MA), Cs, and Rb cations are introduced into a (FAPbI₃)_0.9(FAPbBr₃)_0.1 (FA=formamidine) film, which is rarely used because of its poor photovoltaic performance. The effects of different contents of MA, Cs, or Rb cations on the performance of (FAPbI₃)_0.9(FAPbBr₃)_0.1 films and devices are systematically studied. The results show that the devices with Cs cations exhibit markedly improved photovoltaic performance and stability, attributed to the clearly enhanced quality of films and their intrinsic stability. The (FAPbI₃)_0.9(FAPbBr₃)_0.1 devices with 10 % Cs show a PCE as high as 19.94 %. More importantly, the unsealed devices retain about 80 % and 90 % of the initial PCE at 85 °C after 260 h and under 45±5 % relative humidity (RH) after 1440 h, respectively, which are better than that with 15 % MA and 5 % Rb under the same conditions. This indicates that a highly efficient and stable perovskite component has been achieved, and PSCs based on this component are expected to promote their further development.

Full text of DOI:10.1002/cssc.201800658

Accepted Article
Title: Adjusting the introduction of cations (MA, Cs or Rb) to obtain
highly efficient and stable perovskite solar cells based on
(FAPbI3)0.9(FAPbBr3)0.1
Authors: Guozhen Liu, Haiying Zheng, Liangzheng Zhu, Ahmed
Alsaedi, Tasawar Hayat, Xu Pan, Li’e Mo, and Songyuan Dai
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.201800658
Link to VoR: http://dx.doi.org/10.1002/cssc.201800658
A Journal of
www.chemsuschem.org

ChemSusChem
10.1002/cssc.201800658
Adjusting the introduction of cations (MA, Cs or Rb) to
obtain highly efficient and stable perovskite solar cells based
on (FAPbI
3
)0.9(FAPbBr
)
3 0.1
Guozhen Liu,[a,b] Haiying Zheng,[a,b] Liangzheng Zhu,[a,b] Ahmed Alsaedi , Tasawar
[c]
Hayat[c,e], Xu Pan,* Li’e Mo* and Songyuan Dai*
[a]
[a]
[a,c,d]
aKey Laboratory of Photovoltaic and Energy Conservation Materials, Institute of Applied
Technology, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031,
China.
bUniversity of Science and Technology of China, Hefei 230026, China.
cNAAM Research Group, Department of Mathematics, Faculty of Science, King Abdulaziz
University, Jeddah 21589, Saudi Arabia.
dState Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources, North
China Electric Power University, Beijing 102206, China.
eDepartment of Mathematics, Quaid-I-Azam University, Islamabad 44000, Pakistan.
E-mail: xpan@rntek.cas.cn, lemo@rntek.cas.cn, sydai@ncepu.edu.cn
Abstract: Although power conversion efficiency (PCE) of perovskite solar cells
(PSCs) has increased to 22.7%, the instability when exposed to moisture and heat
hindered their further practical development. In this study, to gain highly efficient and
stable perovskite component, MA, Cs and Rb cations are respectively introduced into
the (FAPbI
performance. The effects of different contents of MA, Cs or Rb cations on the
performance of (FAPbI 0.1 films and devices are systematically studied.
3
)
0.9(FAPbBr )0.1 film which is rarely used due to the poor photovoltaic
3
3
)
0.9(FAPbBr )
3
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
The results show that the devices with Cs cation exhibit markedly improved
photovoltaic performance and stability, attributing to the obviously enhanced quality of
films and their intrinsic stability. The (FAPbI 0.1 devices with 10% Cs
obtain a PCE as high as 19.94%. More importantly, the unsealed devices retain about
0% and 90% of the initial PCE at 85 °C after 260 h and under 45±5% relative humidity
RH) after 1440 h, respectively, which are more brilliant than that with 15% MA and
% Rb under the same condition. It indicates that a highly efficient and stable
3
)
0.9(FAPbBr )
3
8
(
5
perovskite component has been achieved and the PSCs based on this component will
expect to promote the further development.
Keywords: perovskite solar cells, cations introduction, high efficiency, humidity
stability, thermal stability
Introduction
In the last few years, organic-inorganic metal halide perovskite solar cells (PSCs) have
attracted unprecedented attention and been supposed to be one of the most promising
superstar for next-generation photovoltaic industry due to their outstanding properties,
such as wide absorption band, low defect density, high charge carrier mobility and long
diffusion length.[1-9] Recently, the certified power conversion efficiency (PCE) in
laboratory has increased to 22.7%,[10] which has surpassed the records of the second-
generation solar cell kept by CIGS/CdTe and been still continuously updated.[11]
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
Unfortunately, the poor stability of PSCs under high humidity and temperature
have been reported to impede their commercialized development.[12-15] More and more
attempts have been proposed to enhance the humidity and temperature stability of PSCs,
such as changing the electron transport materials (ETM),[16,17] introducing the additives
into light absorber layer,[18-21] employing the inorganic hole-transporting materials
(
HTM) without dopant.[22,23]
Among these efforts, to obtain a stability perovskite material by compositional
engineering is one of the simplest and most essential method.[24-29] Pure FAPbI
suffers
3
from poor humidity stability due to the easy phase transition from the α-phase
perovskite (black) to the δ-phase nonperovskite (yellow), which hinders its intensive
study and widespread use.[24,27] Seok and co-workers[30] introduced MAPbBr
component into FAPbI to gain (FAPbI 0.15 PSCs which present high
PCE and phase stability. Park and co-workers[31] adopted inorganic Rb as an enhancer
to improve the photovoltaic performance and moisture stability of FAPbI PSCs. Zhou
and co-workers[32] acquired highly efficient and stable FA1−xCs
PbI (Cl) PSCs by the
intercalation of CsI. Moreover, (FAPbI 1-y(FAPbBr have been supposed to be more
3
3
3
)
0.85(FAPbBr )
3
+
3
x
3
3
)
)
3 y
stable, whereas they are rarely used in solar cells due to the poor photovoltaic
performance.[33,34] Therefore, it is an urgent need to improve the PCE and stability of
devices based on this component.
In this study, we report embedding different contents of MA, Cs or Rb cations into
(
FAPbI
3
)
0.9(FAPbBr )0.1 perovskite film to fabricate highly efficient and stable PSCs.
3
Finally, a perovskite composition exhibits superior photovoltaic performance and
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
excellent stability by introducing 10% Cs into (FAPbI
3
)
0.9(FAPbBr
3
)0.1 film. The
devices with a PCE of 19.94% under AM 1.5G solar illumination can be achieved due
to the excellent photoelectric property and film quality. In addition, the unsealed
devices display remarkable humidity and heat stability after aging at 85 °C and under
4
5±5% relative humidity (RH), respectively, which is more outstanding than the
composition by introducing 15% MA and 5% Rb.
Results and discussion
Figure 1. (a) Crystal structure of APbX perovskite. (b) Schematic architecture of
3
mesoscopic PSC. (c) Ionic structures and radius of different A-site cations (FA, MA,
Cs and Rb).
Herein, based on the (FAPbI
3
)
0.9(FAPbBr )0.1 component, different perovskite films
3
were fabricated by one-step method. The schematic architecture of mesoscopic PSC is
depicted in Figure 1b. The ionic structure and radius of four different A-site cations
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
(FA, MA, Cs and Rb) are shown in Figure 1c. After introducing different contents of
cations (MA, Cs or Rb), we prepared (A
x
FA1-xPbI
3
)
0.9(FAPbBr
3
)0.1 films with x ranging
from 0 to 0.05, 0.1 and 0.15 (Labeled as FA, A
recognition, A= MA, Cs or Rb), respectively.
5
-FA, A10-FA and A15-FA for easy
Figure 2. XRD patterns of (a) MA
perovskite films. (d) The magnified XRD patterns of the (110) peak of MA
FA and Rb -FA (x=5, 10 and 15) perovskite films. (e) XRD patterns of FA, MA15-FA,
Cs10-FA and Rb -FA perovskite films.
x
-FA, (b) Cs
x
-FA and (c) Rb
x
-FA (x=5, 10 and 15)
x
-FA, Cs
x
-
x
5
To uncover the crystal structure of perovskite films after introducing different
cations (MA, Cs or Rb), XRD patterns are shown in Figure 2. Figure 2a, b and c
demonstrate the XRD patterns of perovskite films with different contents of MA, Cs
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
and Rb, respectively. After introducing MA or Cs cations, the films display good
crystallinity of α-phase perovskites. As shown in Figure 2a, when the content of MA
cation is 5%, the diffraction peaks at 12.2° and 12.7° are attributed to the“orange-red”
phase and residual PbI
cation increases to 15%, there is a weak diffraction peak at 13.12°, which can be
speculated as the formation of δ-CsPbI -FA films in Figure 2c
.[31] However, the Rb
have poor crystalline structure. As the content of Rb only increases beyond 5%, the
diffraction peaks at 10.1° and 26.9° indicate the formation of δ-RbPbI . It is attributable
to the big difference in ionic radius between Rb (152 pm) and FA (253 pm).[31] In Figure
d, with the increase of MA cation content from 5% to 10% and 15%, the diffraction
peaks attributed to (110) plane of α-FAPbI display a little shift to higher angle, in
agreement with the smaller size of MA cation relative to FA, which shrinks the crystal
lattice.[34] The same shifts can be found in Cs
-FA and Rb -FA perovskite films. On the
2
presenting in the films.[35] In Figure 2b, when the content of Cs
3
x
3
2
3
x
x
one hand, it may caused by the lattice distortion after inserting of a cation (Cs or Rb)
smaller than FA cation.[31,40] On the other hand, the Cs and Rb cations may be acted as
scavengers to eliminat iodide selectively (resulting in the blue shift in UV-vis
absorption spectra and PL spectra) or by regulate the vacancies though the
incorp/oration of excess cations/anions from the addition of excess AX (CsI or RbI).
The real reason is very valuable to explore in the further work. As a result, Cs and Rb
cations play a role in preventing bromide phase-segregation of the perovskite
effectively.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
Figure 3. (a) UV-vis absorption spectra and (b) normalized PL spectra of FA, MA
Cs -FA and Rb -FA (x=5, 10, and 15) perovskite films.
To make clear the different influence of mixed cations (MA, Cs or Rb) on the
crystallization, Figure 2e shows the XRD patterns of (FAPbI 0.9(FAPbBr )0.1 film and
x
-FA,
x
x
3
)
3
perovskite films with MA, Cs or Rb cations in the case of the optimal proportion.
Without any mixed cations, the film owns an obvious diffraction peak at about 12.23°,
suggesting the formation of a new material or phase which displays as orange-red.[32]
The diffraction peaks at about 11.6° and 12.7° are attributed to the δ-phase of FAPbI
3
and residual PbI , respectively, which are common phenomena in FA-based PSCs and
2
lead to reduced properties.[25,31] Whereas, for the MA15-FA and Cs10-FA films, no
diffraction peaks are observed around 11.61°, 12.23° and 12.69°, indicating that there
are no δ-phase of FAPbI
3
, “orange-red” phase and residual PbI in the films. With
2
regards to Rb
5
-FA film, obvious diffraction peaks can be found at about 12.2° and 12.7°.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
It means that, compared with MA and Cs, the introducing of Rb cation takes the weakest
majorization of promoting great crystallinity of α-phase perovskites. The results are
consistent with the performance of the corresponding devices.
The UV-vis absorption spectra and normalized PL spectra are utilized to examine
the optical properties of (FAPbI
3
)
0.9(FAPbBr )0.1 perovskite films with different
3
contents of mixed cations (MA, Cs or Rb). As shown in Figure 3a, after introducing
MA, Cs or Rb cations, all the films exhibit visibly enhanced light absorption from 550
to 800 nm due to the improved crystallinity. For Rb
owe to the easily separated of δ-RbPbI from the films. A slight blue-shift can be
observed as the increase of cations (MA, Cs or Rb) contents, which is in good
agreement with the corresponding PL peaks in Figure 3b. Compared to MA -FA, the
Cs -FA and Rb -FA films present a blue shift in absorption band edge because of the
x
-FA, the poor light absorption may
3
x
x
x
smaller ionic radius of Cs and Rb. The same trend can be found on the normalized PL
spectra. Interestingly, a second peak can be found in the PL spectra of FA film (Figure
3
b), which is caused by the phase-segregated consisting of Br-rich and Br-free
regions.[38,39] The second phase (orange-red phase) displays a XRD diffraction peak at
θ= 12.23° (Figure 2e). It is a simply intrinsic to all (FAPbI perovskite
compositions (Figure S7) and the potential reason for (FAPbI devices
2
3
)
1-x(FAPbBr
3
)
x
3
)1-x(FAPbBr
)
3 x
behave a poorer photovoltaic property. When introducing mixed cations (MA, Cs or
Rb), no second PL peak is observed. The results demonstrate that the existence of mixed
cations (MA, Cs or Rb) in the (FAPbI
3
)
0.9(FAPbBr )0.1 films can effectively restrain the
3
phase separation. This conclusion is consistent with the result of the XRD patterns.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
Figure 4. Top surface SEM images of MA
perovskite films.
Scanning electron microscopy (SEM) measurements are used to perform the
surface morphology of (FAPbI 0.1 perovskite films with mixed cations
MA, Cs or Rb). Figure 4 shows the top surface SEM images of mixed cations
x
-FA, Cs
x
-FA and Rb
x
-FA (x=5, 10 and 15)
3
)
0.9(FAPbBr )
3
(
perovskite films with different contents. In the experiment, all the films display
compact surface without pinhole. Among the films with increased contents of MA
cation, the morphologies turn to be smooth. In addition, when the contents of Cs and
Rb cations increases to 15%, the morphologies (Cs15-FA and Rb15-FA) are rough and
there are many irregular crystals growing on the surfaces. Those could be caused by the
formation of δ-CsPbI
3
and δ-RbPbI , respectively. Especially, the films with 15% MA
3
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
and 10% Cs present relatively smooth and uniform top surface, which will effectively
reduce the defect and then improve the photoelectricity properties.
Figure 5. (a) Normalized transient absorption (TA) responses of glass/perovskite films
(
MA15-FA, Cs10-FA and Rb
conversion efficiency (IPCE) spectra of PSCs based on FA, MA15-FA, Cs10-FAand Rb
FA perovskite films. (d) J–V curves of Cs10-FA PSC under reverse and forward scan
directions. (f) The PCE histogram fitted with a Gaussian distribution of the FA, MA15
5
-FA). (b) J–V curves and (c) incident photon to current
5
-
-
FA, Cs10-FA and Rb
5
-FA devices among 30 measured devices.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
To investigate the recombination behavior between electron and hole in the
perovskite layer with different components, we measured transient absorption (TA)
spectra (Figure 5a). The internal recombination occurred in perovskite layers has a great
influence on the photovoltaic performance of devices. The longer recombination time
means the slower recombination of electron and hole, leading to better device
performance.[36,37] According to fitting results in single exponential decays, the Cs10
FA has the longest lifetime of 132 ns, which is 106 ns and 82 ns for MA15-FA and Rb
FA, respectively. The Rb
-
-
5
5
-FA displays the shortest lifetime due to its worse
crystallization and morphology. In other words, a high-quality film with perfect
morphology and fewer defect states could be achieved by introducing 10% Cs cation.
To further characterize the photovoltaic performance of perovskite films, the
devices with structure of FTO/c-TiO
fabricated by introducing different contents of mixed cations (MA, Cs or Rb) into
FAPbI 0.1 perovskite photoactive layer. The current–voltage (J–V)
2
/m-TiO /perovskite/spiro-MeOTAD/Au were
2
(
3
)
0.9(FAPbBr )
3
curves and corresponding parameters measured under standard AM1.5G illumination
are shown in Figure 5b and Table 1, respectively. The device based on
(
FAPbI
3
)
0.9(FAPbBr )0.1 film displays an inferior PCE of 10.81% with a much lower Jsc
3
-2
of 14.95 mA cm , which is caused by the presence of orange-red phase. Fortunately,
by introducing mixed cations into the perovskite films, the photoelectric property can
be increased effectively, due to the adjustment of bandgap and the improvement of
absorption. Among the PSCs with different components, the MA15-FA and Cs10-FA
devices own a better performance, which is in accord with the crystallization,
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
morphology and TA characterization. Particularly, the device based on the Cs10-FA
-2
gains the optimal photovoltaic performance with a Jsc of 22.82 mA cm , a Voc of 1.09
V, and a FF of 79.92%, and ultimately an excellent PCE of 19.94%. When refer to Rb
5
-
FA, the Voc is as low as 0.99 V, which can be put down to the worse film quality. The
incident photon to current conversion efficiency (IPCE) spectra of the optimal devices
are shown in Figure 5c. The results are consistent with the Jsc presenting in
corresponding J–V curves. A slight blue shift also can be find which has mentioned in
the UV-vis absorption spectra (Figure 3a). The hysteresis phenomenon, an important
index to evaluate the actual photovoltaic performances, is verified by collecting J–V
curves under reverse and forward scan directions. In Figure 5d, the results exhibit that
the device with a component of Cs10-FA has a 0.95% PCE hysteresis, which is lower
than most reported FA-based PSCs.[25,32] Moreover, Figure 5e presents PCE histograms
of different devices with a Gaussian distribution. An apparent variation of PCE can be
observed among the four kinds of PSCs. When refer to Cs10-FA, an average PCE of
1
8.81% can be achieved among 30 measured devices, indicating a high reproducibility.
Table 1. Photovoltaic parameters of FA, MA15-FA, Cs10-FA and Rb
5
-FA PSCs.
PCE (%)
10.81
-2
Device
FA
J
sc (mA cm )
Voc (V)
FF (%)
61.55
74.89
79.92
68.33
16.86
1.04
1.09
1.09
0.99
MA15-FA
Cs10-FA
22.77
18.58
22.82
19.94
Rb
5
-FA
19.12
12.93
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
Figure 6. Normalized PCE variation curves of unsealed FA, MA15-FA, Cs10-FA and
Rb -FA PSCs when exposure to (a) 85 °C and (c) 45±5% RH. (b) Thermal stability
5
measurements for Cs10-FA PSCs at different temperatures: 20, 60 and 85 °C. (d)
Humidity stability measurements for Cs10-FA PSCs under different humility: 25±5%,
4
5±5%, and 65±5% RH.
The instability of PSCs which are prone to degradation upon exposure to moisture
and heat has hindered their realistic development. Therefore, it is a serious matter to
improve the stability of PSCs. We firstly measured the thermal stability of high
performance four kinds of devices (FA, MA15-FA, Cs10-FA and Rb -FA) stored in the
5
condition of 85 °C where the RH is about 20% without any encapsulation (in the dark).
Figure 6a shows the normalized PCE variation curves as a function of the testing time.
Compared with Rb -FA and FA devices, which only retain about 18% and 10% of the
5
initial PCE, respectively, the Cs10-FA PSCs display the most outstanding thermal
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
stability. After aging for 260 hours (h), the best device of Cs10-FA only decreases by
0%. Although the MA15-FA PSCs display a good stability at the first 50 h, only 45%
2
of the initial values is kept after 260 h. According to GTF calculation, it is easy for MA+
to desorb from the crystal, which will lead to the thermal instability.[26] Meanwhile, the
stability of Cs10-FA PSCs under 20 °C and 60 °C was also measured (Figure 6b). After
2
60 h, the PCE only decrease by about 3% and 9%, respectively. In our case, the
enhanced thermal stability should be ascribed to the more stable structural of Cs/FA
double cations component.
We further conducted the humidity stability of the FA, MA15-FA, Cs10-FA and
Rb
in the dark) and all the devices were not encapsulated. It can be found from Figure 6c,
after 1440-h aging test, the Cs10-FA devices retain about 88% of the initial PCE.
Whereas, at the same condition, the PCE of FA, MA15-FA and Rb -FA devices decrease
5
-FA devices. Here, the test was proceeding under 45±5% RH at room temperature
(
5
by about 87%, 79% and 49%, respectively, while only 12% for the Cs10-FA.
Furthermore, the Cs10-FA PSCs still maintain about 95% and 76% of the initial PCE
after aging under 25±5% and 65±5% RH for 1440 h, respectively (Figure 6d). The
results exhibit that a device with superior humidity stability can be realized by
introducing Cs cation into the (FAPbI
3
)
0.9(FAPbBr )0.1 perovskite films.
3
Conclusions
In conclusion, we systematically studied the impact of MA, Cs or Rb cations
introduction on the performance of (FAPbI 0.1 film and device. By
3
)
0.9(FAPbBr )
3
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
regulating the contents of MA, Cs or Rb cations, perovskite films with different
components were fabricated. As a result, the (FAPbI 0.1 devices with 10%
Cs obtain a highest PCE up to 19.94%, compared to the devices without doping (PCE=
0.81%), with 15% MA (PCE= 18.58%) and 5% Rb (PCE= 12.93%). Meanwhile, the
3
)
0.9(FAPbBr )
3
1
PSCs with 10% Cs display more outstanding stability against moisture and heat than
other devices under the same fabricating and aging conditions. The unsealed
(
Cs0.1FA0.9PbI
3
)
0.9(FAPbBr
3
)0.1 devices maintain about 80% of their initial PCE under
8
5 °C after 260 h and the PCE only decrease by about 10% after 1440 h under 45±5%
RH. This work provides a valuable method to higher efficient and stable PSCs as a
next-generation photovoltaic industry.
Experimental Section
Materials and preparation
Formamidine iodide HC(NH
2
)
2
I (FAI) and formamidine bromine HC(NH
2
2
) Br (FABr):
FAI and FABr were prepared by reacting formamidine acetate powder with 57 wt%
hydroiodic acid (for FAI) or 47 wt% hydrobromic acid (for FABr) with
stoichiometrically. After adding HI or HBr, the solution was stirred for 2 h in the ice
bath. Then, the solution was evaporated at about 55 °C using rotary evaporation to
remove the solvent under reduced pressure. The powders were then purified by
dissolving in ethanol and recrystallizing in anhydrous ether. Finally, the products were
dried in vacuum at 60 °C overnight.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
Preparation of perovskite precursors: The (FAPbI
3
)
0.9(FAPbBr
3
)0.1 precursor
2
+
solution (1.35 M Pb ) was prepared from dissolving the lead iodide (PbI
2
), lead
bromine (PbBr ), formamidine iodide (FAI) and formamidine bromine (FABr) powders
2
in mixed solvent of DMSO (20% by volume) and DMF (80% by volume). MAI/CsI/RbI
were incorporated into the precursor solution by replacing the corresponding content of
FAI. All the solutions were then stirred for 1 h at 70 °C and filtered using 0.45 µm
PVDF filters before spin-coating.
Device fabrication
The FTO substrate was washed by sonication for 30 min and sequentially cleaned by
ultrapure water and ethanol for several times. The compact layer of TiO was coated on
2
the substrate by spray pyrolysis at 450 °C with the precursor solution of 7-mL
isopropanol, 0.6-mL titanium diisopropoxide and 0.4-mL bis(acetylacetonate). The
TiO
s, followed by 10 min thermal drying at 100 °C. Then, the substrate was annealed at
10 °C for 30 min. After cooling to room temperature, perovskite precursor solution
was poured onto the substrate and spin-coated for 50 s (1100 r.p.m for 15 s and then
600 r.p.m for 35 s) in an air flowing glovebox (About 20%RH). 120-μL chlorobenzene
2
mesoporous layer was then spin-coated on the compact layer at 4000 r.p.m for 20
5
4
was drop-casted on the rotating film 15 s before the end of the spin coating program.
The deposited film was then heated at 150 °C for 30 min when the component is
(
FAPbI
3
)
0.9(FAPbBr )0.1. When MAI/CsI/RbI were incorporated into the precursor
3
solutions, the deposited film was heated at 100 °C for 30 min. The HTL, with a
composed of spiro-OMeTAD (73 mg), 4-tert-butylpyridine (29 μL), lithium
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
bis(trifluoromethylsulphonyl) imide (LiTFSI, 17 μL) and Co(III)-complex (8 μL) in 1-
mL chlorobenzene solvent, was spin-coated on the perovskite film with a speed of 3000
r.p.m for 20 s. Finally, about 60-nm Au electrode were deposited on the HTL by thermal
evaporating.
Characterization
X-ray diffraction (XRD) patterns of perovskite films were recorded on an X’pert PRO
MPD (PANalytical) with Cu Kα radiation and the 2θ range was 10-50°. Ultraviolet-
visible (UV-vis) absorption spectra were carried out by an UV-vis spectrophotometer
(UV−vis, U-3900H, HITACHI, Japan) with the wavelength range from 500 nm to 900
nm. Steady-state fluorescence (PL) spectra were measured and recorded by a
spectrofluorometer (photon technology international), analyzed by a software
Fluorescence. The excitation light wavelength was 473 nm from a standard 450 W
xenon CW lamp. Scanning electron microscopy (SEM) measurements were performed
on a Schottky Field Emission SEM (FEI Sirion200). Transient absorption spectra (TAS)
were tested by LKS.80 and LP.920. Incident monochromatic photon-to-current
efficiency (IPCE) of perovskite solar cells were recorded by QE/IPCE measurement
system (Newport, USA). Photocurrent and voltage curves (J–V) were obtained by a
2
solar simulator (Newport, Oriel Class A, 91195A) at 100 mW/cm illumination AM
2
1
.5G. The active area of devices was defined as 0.09 cm by a black mask. The setup
was calibrated with a certified silicon solar cell (Fraunhofer ISE) prior to measurements.
The humidity aging tests were measured in the containers remained at room
temperature, under dark and the humidity of about 25%, 45% and 65% was controlled
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
by humidifier and desiccant. The temperature aging tests were measured in the drying
ovens with the humidity of about 20% under dark, the temperatures were retained at
0 °C, 60 °C and 85 °C.
2
Acknowledgements
This work was financially supported by the National High Technology Research and
Development Program of China under Grant No.2015AA050602, STS project of
Chinese Academy of Sciences (KFJ-SW-STS-152).
References
[
1] A. Kojima, K. Teshima, Y. Shirai, T. Miyasaka, J. Am. Chem. Soc. 2009, 131, 6050-
051.
2] J. M. Ball, M. M. Lee, A. Hey, H. J. Snaith, Energy Environ. Sci. 2013, 6, 1739-
743.
6
[
1
[
[
3] H. S. Jung, N.-G. Park, Small 2015, 11, 10-25.
4] M. Saliba, T. Matsui, K. Domanski, J.-Y. Seo, A. Ummadisingu, S. M. Zakeeruddin,
J.-P. Correa-Baena, W. R. Tress, A. Abate, A. Hagfeldt, M. Grätzel, Science 2016, 354,
06-209.
5] W. S. Yang, B.-W. Park, E. H. Jung, N. J. Jeon, Y. C. Kim, D. U. Lee, S. S. Shin, J.
Seo, E. K. Kim, J. H. Noh, S. I. Seok, Science 2017, 356, 1376-1379.
2
[
[
[
6] M. Grätzel, Nat. Mater. 2014, 13, 838-842.
7] Q. Chen, H. Zhou, Z. Hong, S. Luo, H.-S. Duan, H.-H. Wang, Y. Liu, G. Li, Y. Yang,
J. Am. Chem. Soc. 2014, 136, 622-625.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
[
8] Q. Dong, Y. Fang, Y. Shao, P. Mulligan, J. Qiu, L. Cao, J. Huang, Science 2015, 347,
9
67-970.
[
9] Y. Wu, X. Yang, W. Chen, Y. Yue, M. Cai, F. Xie, E. Bi, A. Islam, L. Han, Nat.
Energy 2016, 1, 16148.
10] https://www.nrel.gov/pv/assets/images/efficiency-chart.png, The National
Renewable Energy Laboratory (NREL), 2017.
11] M. A. Green, K. Emery, Y. Hishikawa, W. Warta, E. D. Dunlop, Prog.
Photovoltaics: Res. Appl. 2015, 23, 1-9.
12] X. Meng, Y. Bai, S. Xiao, T. Zhang, C. Hu, Y. Yang, X. Zheng, S. Yang, Nano
Energy 2016, 30, 341-346.
13] M. Saliba, T. Matsui, K. Domanski, J.-Y. Seo, A. Ummadisingu, S. M. Zakeeruddin,
J.-P. Correa-Baena, W. R. Tress, A. Abate, A. Hagfeldt, M. Grätzel, Science 2016, 354,
06-209.
14] E. J. Juarez-Perez, Z. Hawash, S. R. Raga, L. K. Ono, Y. Qi, Energy Environ. Sci.
016, 9, 3406-3410.
15] K. A. Bush, C. D. Bailie, Y. Chen, A. R. Bowring, W. Wang, W. Ma, T. Leijtens, F.
Moghadam, M.D. McGehee, Adv. Mater. 2016, 28, 3937-3943.
16] J. Huang, X. Yu, J. Xie, C.-Z. Li, Y. Zhang, D. Xu, Z. Tang, C. Cui, D. Yang, ACS
Appl. Mater. Interfaces 2016, 8, 34612-34619.
17] A. Agresti, S. Pescetelli, L. Cinà, D. Konios, G. Kakavelakis, E. Kymakis, A D.
Carlo, Adv. Funct. Mater. 2016, 26, 2686-2694.
[
[
[
[
2
[
2
[
[
[
[
18] J. Pan, C. Mu, Q. Li, W. Li, D. Ma, D. Xu, Adv. Mater. 2016, 28, 8309-8314.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
[
[
19] X. Song, W. Wang, P. Sun, W. Ma, Z.-K. Chen, Appl. Phys. Lett. 2015, 106, 033901.
20] P.-W. Liang, C.-Y. Liao, C.-C. Chueh, F. Zuo, S. T. Williams, X.-K. Xin, J. Lin, A.
K. Y. Jen, Adv. Mater. 2014, 26, 3748-3754.
21] Y. Sheng, Y. Hu, A. Mei, P. Jiang, X. Hou, M. Duan, L. Hong, Y. Guan, Y. Rong,
Y. Xiong, H. Han, J. Mater. Chem. A 2016, 4, 16731-16736.
[
[
[
22] Y. Xue, Y. Wu, Y. Li, J. Power Sources 2017, 344, 160-169.
23] X. Zhao, F. Zhang, C. Yi, D. Bi, X. Bi, P. Wei, J. Luo, X. Liu, S. Wang, X. Li, S.
M. Zakeeruddin, M. Gratzel, J. Mater. Chem. A 2016, 4, 16330-16334.
24] J.-W. Lee, D.-H. Kim, H.-S. Kim, S.-W. Seo, S.-M. Cho, N.-G. Park, Adv. Energy
Mater. 2015, 5, 1501310.
25] 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.
26] X. Zhang, X. Ren, B. Liu, R. Munir, X. Zhu, D. Yang, J. Li, Y. Liu, D.-M. Smilgies,
R. Li, Z. Yang, T. Niu, X. Wang, A. Amassian, K. Zhao, S. Liu, Energy Environ. Sci.
017, 10, 2095-2102.
27] G. E. Eperon, S. D. Stranks, C. Menelaou, M. B. Johnston, L. M. Herz, H. J. Snaith,
Energy Environ. Sci. 2014, 7, 982-988.
[
[
[
2
[
[
[
28] H.-S. Kim, J.-Y. Seo, N.-G. Park, ChemSusChem 2016, 9, 2528-2540.
29] R. G. Niemann, L. Gouda, J. Hu, S. Tirosh, R. Gottesman, P. J. Camerona, A.
Zaban, J. Mater. Chem. A 2016, 4, 17819-17827.
30] N. J. Jeon, J. H. Noh, W. S. Yang, Y. C. Kim, S. Ryu, J. Seo, S. I. Seok, Nature
015, 517, 476.
[
2
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
[
31] Y. H. Park, I. Jeong, S. Bae, H. J. Son, P. Lee, J. Lee, C.-H. Lee, M. J. Ko, Adv.
Funct. Mater. 2017, 27, 1605988.
[
32] N. Zhou, Y. Shen, Y. Zhang, Z. Xu, G. Zheng, L. Li, Q. Chen, H. Zhou, Small 2017,
3, 1700484.
33] G. Divitini, S. Cacovich, F. Matteocci, L. Cina, A. Di Carlo, C. Ducati, Nat. Energy
016, 1, 15012.
34] D. Bi, P. Gao, R. Scopelliti, E. Oveisi, J. Luo, M. Grätzel, A. Hagfeldt, M. K.
Nazeeruddin, Adv. Mater. 2016, 28, 2910-2915.
35] X. Zhang, J. Ye, L. Zhu, H. Zheng, G. Liu, X. Liu, B. Duan, X. Pan, S. Dai,
Nanoscale 2017, 9, 4691-4699.
36] Q. Shen, Y. Ogomi, J. Chang, T. Toyoda, K. Fujiwara, K. Yoshino, K. Sato, K.
Yamazaki, M. Akimoto, Y. Kuga, K. Katayama, S. Hayase, J. Mater. Chem. A 2015, 3,
308-9316.
37] Y. Ogomi, K. Kukihara, S. Qing, T. Toyoda, K. Yoshino, S. Pandey, H. Momose,
S. Hayase, ChemPhysChem 2014, 15, 1062-1069.
38] B. Slimi, M. Mollar, I. B. Assaker, A. Kriaa, R. Chtourou, B. Marıí, Monatshefte
für Chemie-Chemical Monthly 2017, 148, 835.
39] A. S. Subbiah, S. Agarwal, N. Mahuli, P. Nair, M. v. Hest, S, K. Sarkar, Adv. Mater.
Interfaces 2017, 4, 1601143.
40] C. Yi, J. Luo, S. Meloni, A. Boziki, N. Ashari-Astani, C. Grätzel, S. M.
1
[
2
[
[
[
9
[
[
[
[
Zakeeruddin, U. Rothlisberger, M. Grätzel, Energy Environ. Sci. 2016, 9, 656-662.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800658
Entry for the Table of Contents
Films and solar cells based on (FAPbI
contents of MA, Cs or Rb cations are respectively prepared and systematically studied.
The results demonstrate that the (FAPbI 0.1 devices with 10% Cs display
a highest PCE up to 19.94% and excellent humidity and heat stability.
3
)
0.9(FAPbBr
3
)0.1 by introducing different
3
)
0.9(FAPbBr )
3
2
3
0
-
2
2
J
sc=22.82 mA cm
Voc=1.09 V
FF=79.92 %
PCE=19.94 %
1
6
2
8
4
1
0
.0
0.2
0.4
0.6
0.8
1.0 1.1
Voltage (V)
This article is protected by copyright. All rights reserved.