Review

Cite This: Chem. Rev. XXXX, XXX, XXX−XXX pubs.acs.org/CR

Halide Perovskite Photovoltaics: Background, Status, and Future

Prospects
Ajay Kumar Jena,† Ashish Kulkarni,† and Tsutomu Miyasaka*,†,‡
†
Graduate School of Engineering and ‡Faculty of Medical Engineering, Toin University of Yokohama, 1614 Kurogane-cho, Aoba,
Yokohama, Kanagawa 225-8503, Japan

ABSTRACT: The photovoltaics of organic−inorganic lead halide perovskite materials

have shown rapid improvements in solar cell performance, surpassing the top efficiency
of semiconductor compounds such as CdTe and CIGS (copper indium gallium
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

selenide) used in solar cells in just about a decade. Perovskite preparation via simple
Downloaded via UNIV OF TEXAS AT DALLAS on March 1, 2019 at 15:37:57 (UTC).

and inexpensive solution processes demonstrates the immense potential of this thin-film
solar cell technology to become a low-cost alternative to the presently commercially
available photovoltaic technologies. Significant developments in almost all aspects of
perovskite solar cells and discoveries of some fascinating properties of such hybrid
perovskites have been made recently. This Review describes the fundamentals, recent
research progress, present status, and our views on future prospects of perovskite-based
photovoltaics, with discussions focused on strategies to improve both intrinsic and
extrinsic (environmental) stabilities of high-efficiency devices. Strategies and challenges regarding compositional engineering of
the hybrid perovskite structure are discussed, including potentials for developing all-inorganic and lead-free perovskite materials.
Looking at the latest cutting-edge research, the prospects for perovskite-based photovoltaic and optoelectronic devices,
including non-photovoltaic applications such as X-ray detectors and image sensing devices in industrialization, are described. In
addition to the aforementioned major topics, we also review, as a background, our encounter with perovskite materials for the
first solar cell application, which should inspire young researchers in chemistry and physics to identify and work on challenging
interdisciplinary research problems through exchanges between academia and industry.

CONTENTS 5.2. Stability Issues with Hole Transport Materi-

als and Contacts AA
1. Discovery and Background of Perovskite Photo- 6. All-Inorganic Perovskites AD
voltaics B 6.1. CsPbI3: Stabilization of Black Phase AE
1.1. Photovoltaics of Halide Perovskites B 6.2. Cesium−Lead Mixed Halide Perovskites AH
1.2. Discovery and History of Perovskite Photo- 7. Lead-Free and Low-Lead Perovskites AJ
voltaics B 7.1. Lead-Free Perovskite Materials AJ
1.2.1. From Oxide to Halide Perovskites B 7.1.1. Tin (Sn)-Based Perovskites AJ
1.2.2. Discovery of Halide Perovskite Solar 7.1.2. Germanium (Ge)-Based Perovskites AL
Cells C 7.1.3. Lead-Free Binary Metal Halide Perov-
2. Fundamental Structure, Working Mechanism, skites AM
and Major Milestones of Progress E 7.1.4. Group 15 Metal-Based Perovskite/Non-
2.1. Semiconductor Properties of Organic Lead perovskite Materials AN
Halide Perovskites E 7.2. Low-Lead Perovskite Materials: Reducing
2.2. Working Principle G Toxicity and Enhancing Efficiency AQ
2.3. Major Milestones of Progress H 8. Toward Commercialization AR
3. Metal Oxide-Based Electron Transport Layers in 8.1. Scaling Up and Reproducibility Challenges AR
Perovskite Solar Cells I 8.2. Perovskite Tandem Cells AT
4. Compositional Engineering of Perovskites K 8.3. Low-Temperature Process and Flexible
4.1. Mixed Compositions K Device AV
4.1.1. A-Site Cations Mixture L 8.4. Potential Applications in Optoelectronic
4.1.2. X-Site Anions Mixture O Devices AW
4.1.3. Both Cations and Anions Mixture O 9. Conclusions and Future Prospects AZ
4.2. Mixed Dimensions S
5. Stability of Perovskite Solar Cells T
5.1. Stability Issues with Perovskites T Special Issue: Perovskites
5.1.1. Structural/Intrinsic Stability U
5.1.2. External/Environmental Stability Y Received: August 30, 2018

© XXXX American Chemical Society A DOI: 10.1021/acs.chemrev.8b00539

Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

9.1. Further Improvement in PCE AZ

9.2. How To Increase Intrinsic Stability Further? BA
9.3. How To Increase Environmental Stability
Further? BA
9.4. Potential of Lead-Free Perovskites BB
9.5. Steps toward Commercialization BB
Author Information BB
Corresponding Author BB
ORCID BB
Notes BB
Acknowledgments BB
References BC

1. DISCOVERY AND BACKGROUND OF PEROVSKITE

PHOTOVOLTAICS
1.1. Photovoltaics of Halide Perovskites
The evolution of organic−inorganic lead halide perovskite Figure 1. Year-wise citation history of the first paper1 on perovskite
solar cells (PSCs) has accomplished the most notable progress solar cells based on data obtained from Clarivate Analytics.
in the field of photovoltaics (PV). In the decade since the
publication of our peer-reviewed paper in 2009,1 the first on
stability are two crucial factors for device implementation, in
the topic, which was based on our experiments that had started
this Review, in addition to high efficiency aspects of PSCs, we
as early as 2005, the power conversion efficiency (PCE) of
focus on the stability issues and strategies to overcome them.
PSCs has rapidly increased to reach the latest record of 23.7%
Although there are challenges in industrialization of PSCs,
(reported by the U.S. National Renewable Energy Laboratory
results coming from both research laboratories and industries
(NREL), https://www.nrel.gov/pv/assets/pdfs/pv-efficiency-
show prominent potential.
chart.201812171.pdf), approaching the top values achieved
The tremendous success of perovskite PV was quite
with single-crystalline silicon solar cells. Not only the high
unexpected by our group.1 With a long silence (almost no
performance but also the low-cost solution-based processes
citations) for about 3 years after our first publication and a
used for device fabrication attest to the immense potential of
series of presentations at international conferences, we had not
PSCs to be a PV technology of the future. While solution-
anticipated PSCs becoming the subject of such a huge amount
based syntheses of the perovskite materials involve a lot of
of research activity. In fact, it was only this success that made
chemistry, crystallization engineering and the optical and
us realize the importance of the beginning. Hence, we take this
electrical characterization of solid-state crystals (semiconduc-
opportunity to share the background story of our discovery of
tors) have their backgrounds in physics. Because of the
perovskites as PV materials first, followed by reviews and
interdisciplinary nature of perovskite PV, necessitating
discussions on their performance, their stability, industrializa-
expertise in chemistry, physics, and optoelectronics, the
tion, present challenges, and future prospects.
research field has gained interest from a large community of
researchers around the world. As a result, research progress in 1.2. Discovery and History of Perovskite Photovoltaics
perovskite PV has been tremendous. Interestingly, with the 1.2.1. From Oxide to Halide Perovskites. The term
discoveries of its rare properties, perovskites have found many “perovskite solar cell” might sound familiar to most of us today,
applications beyond PV, expanding into the areas of light- but it was alien in 2005, when the journey to explore PV
emitting diodes, photodetectors, X-ray detectors, memory applications of halide perovskites began in the Miyasaka
devices, and so on. Such multiple functions of halide laboratory at Toin University in Yokohma, Japan. At that time,
perovskites have promoted considerable interdisciplinary “perovskite” generally meant metal oxides having perovskite
research. On the basis of information from Clarivate Analytics structures, most of which are classified as either ferroelectric or
in 2018, we estimate that researchers at more than 1000 piezoelectric materials. Perovskite generally represents a kind
institutes worldwide are presently working on halide perov- of crystal structure with chemical formula ABX3, in which A
skite-related photovoltaics and optoelectronics, and this has and B are cations and X is an anion. In an ideal cubic structure,
produced more than 8000 scientific papers in the field. Figure the B cation has a 6-fold coordination, surrounded by an
1 shows, by year (from 2009 to 2018), the history of citations octahedron of anions, and the A cation has a 12-fold
of our first paper,1 obtained from Clarivate Analytics, cuboctahedral coordination. The cubic unit cell of such
representing the explosive escalation of interest in perovskites. compounds is composed of A cations at cube corner positions,
Being a magic box of many mysterious properties, B sitting at the body-center position, and X anion occupying
perovskites have also triggered fundamental studies on ion the face-centered positions (see Figure 3a). In the history of
migration, defect tolerance, carrier dynamics in this soft minerals, perovskite was first discovered in a piece of chlorite-
semiconductor. However, while some of the mysterious rich skarn by the Prussian mineralogist Gustav Rose in 1839.2
properties of perovskite are very valuable, others are creating The mineral was composed of CaTiO3 and was named after
issues that impair practical applications of the new technology. the renowned Russian mineralogist Count Lev A. Perovskiy
For instance, while the defect tolerant nature of perovskite (1792−1856) upon the request of a notable Russian mineral
contributes to high efficiency the ion-migration phenomenon collector, August Alexander Kämmerer. Later, many inorganic
stands as a potential threat against stability. As efficiency and metal oxides, such as BaTiO3, PbTiO3, SrTiO3, BiFeO3, etc.,
B DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 2. Solid-state mixing routes for synthesis of (a) CH3NH3PbBr3 and (b) CH3NH3PbI3. Photographs of CH3NH3PbBr3 powder spread on
paper and exposed to (c) room light and (d) UV lamp (black light), showing strong emission under the UV light. Solutions of (e) CH3NH3PbBr3
and (f) CH3NH3PbI3. Photographs of (g) TiO2/CH3NH3PbBr3 and (h) TiO2/CH3NH3PbI3 photoanodes. Scanning electron micrographs of (i)
bare TiO2 and (j) TiO2 loaded with CH3NH3PbBr3. Photographs of (k) MAPbBr3 single crystals, (l) precursor solution (dated Aug 4, 2009), and
(m) carbon-based perovskite photovoltaic devices used as the first solid-state perovskite solar cell, all made by Kojima.

were found to have the perovskite structure, so therefore, researchers to use other cations in place of Cs. Weber6,7 found
perovskite compounds are more commonly known as metal that the organic cation methylammonium (CH3NH3+)
oxides, with formula ABO3. Although David Cahen et al. replaces Cs+ to form CH3NH3MX3 (M = Pb,6 Sn,7 X = I,
mentioned in their review3 that a perovskite might have been Br) and reported the first crystallographic study on organic
first synthesized in 1882 by the Danish chemist and lead halide perovskites. Toward the end of the 20th century, a
crystallographer Haldor Topsøe (1842−1935), it seems that large variety of halide perovskites were synthesized by David
the first synthesis had been attempted in 1851 by French Mitzi using small and large organic cations.10−12 Mitzi had
researcher Jacques-Joseph Ebelmen, who synthesized CaTiO3 focused his studies on the physical properties of two-
by a flux growth method.2 dimensional (2D) perovskite materials with a large organic
Oxide perovskites are in use in various ferroelectric, group.11 Based on Mitzi’s studies, in the late 1990s, Prof. Kohei
piezoelectric, dielectric, and pyroelectric applications, etc., Sanui was conducting a project through a Japanese national
But except for some limited compositions like LiNbO3, research program (JST-CREST). This project dealt with self-
PbTiO3, and BiFeO3, which show some PV effect due to organized quantum confinement structures using the above
ferroelectric polarization (known as ferroelectric photo- perovskites; optical properties of 2D13,14 and 3D crystals were
voltaics),4 these metal oxide perovskites do not exhibit good investigated.15,16 Although the research opened applications of
semiconducting properties that would make them suitable for these materials to nonlinear optics and electroluminescence
PV applications. However, a class of halide perovskites which (i.e., light-emitting diodes, LEDs) by utilizing sharp mono-
differ from oxide perovskites in having halide anions in place of chromatic optical absorption and luminescence,17,18 there was
oxide anions (ABX3; A = cation, B = divalent metal cation, X = no idea that these materials could utilize solar energy, because
halogen anion) shows the semiconducting properties that are 2D perovskites are not suitable for harvesting light over the
desired for PV applications. The discovery of such halide wide spectral range of sunlight.
perovskites dates back to the 1890s. In 1893, Wells et al. 1.2.2. Discovery of Halide Perovskite Solar Cells. First-
performed a comprehensive study on the synthesis of lead and second-generation solar cells comprising silicon wafer-
halide compounds from solutions including lead halide and based and thin-film solar cells, respectively, have done well in
cesium, CsPbX3 (X = Cl, Br, I),5 ammonium (NH4),6 or terms of efficiency and stability. However, ultra-high-pure
rubidium, RbPbX3.7 Much later, in 1957, the Danish researcher metallic silicon (>99.9999%), which can be obtained by
C. K. Møller found that CsPbCl3 and CsPbBr3 have the crystallization of melted Si in a furnace at more than 1400 °C,
perovskite structure,8,9 existing as a tetragonally distorted is required for the solar cells. Thus, the high cost of materials
structure which undergoes a transition to a pure cubic phase at and processing of the wafers has prevented the popular use of
high temperature.9 The simple solution process for synthesis of such solar cells as alternatives to fossil-fuel-based energy
these cesium lead halide ionic crystals might have inspired sources such as thermal power generation (present cost is
C DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 3. (a) Crystal structure of organo-lead halide perovskite compounds. (b) SEM image (scale bar = 10 nm) of particles (shown by an arrow)
of nanocrystalline CH3NH3PbBr3 deposited on the TiO2 surface. (c) Incident photon-to-current conversion efficiency (IPCE) action spectra for
photoelectrochemical cells using CH3NH3PbBr3/TiO2 (solid line) and CH3NH3PbI3/TiO2 (dashed line) photoanodes with a liquid electrolyte, 0.4
M LiBr and 0.04 M Br2 dissolved in acetonitrile for the former and 0.15 M LiI and 0.075 M I2 dissolved in methoxyacetonitrile for the latter
photoanode. (d) Photocurrent−voltage characteristics for cells using CH3NH3PbBr3/TiO2 (solid line, PCE = 3.13%) and CH3NH3PbI3/TiO2
(dashed line, PCE = 3.81%) under 100 mW cm−2 AM 1.5 irradiation. Reproduced with permission from ref 1. Copyright 2009 American Chemical
Society.

<0.05USD per kWh of electric power). A lot of research and studies from our team have been presented as a collaboration
development has therefore been devoted to the search for cost- between three universities (TPU, TUY, and UT) and Peccell.
effective alternatives, leading to the third-generation photo- Our perovskite-based PV cell first employed CH3NH3PbX3 (X
voltaics, including organic solar cells and dye-sensitized solar = I, Br) as the sensitizer on a TiO2 mesoporous electrode used
cells (DSSCs). Although these cells could be processed very in conjunction with a lithium halide-containing electrolyte
cheaply, their performance remained limited to PCE of around solution. Some of Kojima’s initial works, carried out before
10%, which limited their chances for commercialization. In 2009, are presented in Figure 2.
2005−2006, while we were putting our continuous efforts into On the assumption that the perovskite would function as a
the further development of DSSCs,19−21 we explored the use of quantum dot-like sensitizer, deposition of the perovskite was
an organic−inorganic lead halide perovskite as an absorber to done by spin-coating of the precursor solution, in which the
replace the organic dye in DSSCs. Although we had not loading amount of the perovskite was adjusted so as to obtain
imagined then that the material would open up a new world of the thinnest layer of nanocrystalline perovskite, to cover a large
PV with such immense potential, we were curious to exploit its surface area of a thick TiO2 layer (∼10 μm), similar to DSSCs.
semiconducting properties in photoelectrochemical cellsit This architecture was different from the present perovskite
was absolutely a blissful encounter with perovskites! Following solar cell that uses a thin TiO2 film (<0.1 μm) as an electron
is a brief background of how we started working on perovskites collector (or uses no mesoporous film). As dissolution of the
for solar cells. perovskite into the liquid electrolyte was a serious problem,
Our research on perovskite-based PV devices started in many trials were required to make the cell stable enough for
2005. This was the year after Miyasaka had established a measurements to be performed. Finally, we were able to
venture company, Peccell Technologies, on the campus of fabricate cells that yielded a PCE barely reaching 3.8%, and this
Toin University of Yokohama (TUY), specializing in was published as the first peer-reviewed paper (Figure 3) on
applications of photoelectrochemistry. In 2005, Akihiro perovskite-based PV cells in 2009.1 Having realized the
Kojima, then a graduate student at Tokyo Polytechnic necessity for a solid-state cell with a solid-state hole transport
University (TPU), came to Miyasaka’s laboratory to learn material (HTM), we attempted to prepare devices by
experiments on dye-sensitized solar cell through collaboration employing carbon/conductive polymer composites and
between TPU and TUY. This collaboration was initiated by obtained a low PCE value (<1%). This was the first example
Dr. Kenjiro Teshima, who was Kojima’s mentor at TPU and of a full solid-state perovskite PV cell, which we reported at an
later worked as a researcher in Peccell. The aim of their international conference (ECS) in 2008.23,24 Later, we realized
collaboration was to examine the possibility of using halide that the poor performance was apparently due to the
perovskites as a sensitizer on mesoporous TiO2 electrodes. significantly low loading amount of perovskite, which was
This was also an extension of the studies being done by also confirmed when Nam Gyu Park et al. showed an improved
Teshima and Kojima at TPU with regard to the quantum PCE even with the liquid electrolyte by simply increasing the
photochemistry of halide perovskites. Teshima had served as a perovskite loading (using a higher concentration precursor
cooperating member of the JST-CREST national project solution).25
(1997−2002) on self-organized quantum-confinement struc- In 2011, we started a collaboration with Henry Snaith of
tures using 2D perovskites.14,15 So it was Miyasaka’s encounter Oxford University for making solid-state perovskite cells. This
with Teshima, who came to join Peccell, that triggered of the opportunity was arranged by Dr. Takurou Murakami
start of perovskite photovoltaics. Experiments using methyl- (presently in AIST, Japan), who had joined the Swiss Federal
ammonium lead halide perovskites as nanocrystalline sensi- Institute of Technology in Lausanne (EPFL),26 where Snaith
tizers in DSSCs gave the first preliminary results which showed was also a researcher, to continue his research on DSSCs after
the possibility of visible light sensitization of TiO2 with having finished his Ph.D. in 2005 under the supervision
deposited lead halide perovskite. In October of 2006, we Miyasaka at TUY. In 2009, Takurou Murakami had started a
presented our first results at Electrochemical Society (ECS) JSPS international collaboration project between the U.K. and
Meeting in Mexico.22 We continued this study, and Kojima Japan on charge transport enhancement in DSSCs. Fabrication
joined our group as a doctoral student in the Graduate School of perovskite-based photovoltaic cells was planned as an
at the University of Tokyo (UT), where Miyasaka had served extension of this project. In 2011, Michael Lee, a Ph.D. student
as a guest professor (2005−2009). Therefore, early perovskite in the Snaith group at Oxford, spent several months in our
D DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 4. (a) Picture of perovskite devices bearing methylammonium lead bromo-iodide, CH3NH3Pb(BrxI1‑x)3‑yCly (0 ≤ x ≤ 1), perovskite films
with different Br/I molar ratios on mesoporous TiO2 substrates. (b) Absorption spectra of the perovskites films (absorption tail at long wavelengths
is due to the mesoporous layer light scattering). (c) Energy band gap extracted from the absorption measurements depending on the content of Br.
Reprinted with permission from ref 35. Copyright 2015 American Chemical Society. (d, e) Wavelength tunability for formamidinium lead bromo-
iodide perovskite, HC(NH2)2PbBr1−yIy. (d) Photographs of the HC(NH2)2PbBr1−yIy perovskite films with y increasing from 0 to 1 (left to right).
(e) UV−vis absorbance of the HC(NH2)2PbBr1−yIy perovskites with varying y. Reproduced with permission from ref 36. Copyright 2014 Royal
Society of Chemistry.

group to learn methods for making the perovskites. Because 2. FUNDAMENTAL STRUCTURE, WORKING
Snaith had worked on solid-state DSSCs using HTM,27,28 our MECHANISM, AND MAJOR MILESTONES OF
aim was to solidify the perovskite-sensitized cell with a spin- PROGRESS
coated layer of an organic hole transporter, spiro-OMeTAD
(2,2′,7,7′-tetrakis(N,N-dimethoxyphenylamine)-9,9′-spiro- 2.1. Semiconductor Properties of Organic Lead Halide
bifluorene). However, we immediately faced the difficulty of Perovskites
stabilizing perovskite crystals against dissolution in the solvent Despite the early discovery of halide perovskites and extensive
used in making the spiro-OMeTAD solution. As this work on the synthesis of 2D and 3D organic lead halide
collaboration progressed, in 2012, we found a way to suppress families, the solid-state physics of halide perovskite materials
dissolution of the perovskite by mixing chloride into the has been investigated in detail only recently. Both experimental
composition of the perovskite, CH3NH3PbI3‑xClx, which and theoretical investigations of their semiconducting proper-
eventually produced cells with PCE up to 10.9%.29 A surprising ties, carrier transport mechanisms, and PV applications have
feature of the device was the use of an insulating mesoporous been done intensively in recent years. These investigations
Al2O3, which worked well as a scaffold for CH3NH3PbI3 have been mainly focused on 3D perovskite materials, which
crystals. A cell using Al2O3 exhibited higher voltage and PCE have excellent PV functions. Unlike conventional semi-
than those obtained using TiO2. Indeed, it was a sign of the conductors like Si, CdTe, GaAs, etc. that are essentially
long diffusion length of carriers in perovskite, which was also covalent compounds, halide perovskites are ionic crystals that
indicated by our observation of intense green light photo- exhibit semiconducting properties. Similarly to the family of
luminescence (PL) from bromide perovskite (CH3NH3PbBr3) silver halide ionic materials, the optical absorption wavelength
on mesoporous Al2O3.30 The long carrier diffusion length of of halide perovskites shows a wide variation with the kind and
the lead halide perovskite, which was later justified by opto- molar ratio of halides (I, Br, Cl) present in the structure.
physical analysis31,32 and carrier mobility33 characterizations of Indeed, the simultaneously ionic and semiconducting nature of
this material, led to a revolution in device architecture and perovskite materials allows easy tuning of band gap and optical
brought about radical changes in researchers’ understanding of absorption by varying the halide ions (I, Br, Cl). Absorption
how PSCs work, distinguishing the new technology from any edge wavelength (band gap) can be freely tuned by mixing I
other organic and hybrid material-based solar cells. With that, a and Br or Cl and Br, forming mixed-halide solid solutions.
new star was born in PV technologies. This starPSCsgrew Figure 4 shows wavelength tuning for an I and Br mixture of
bigger and shone brighter in the shortest time ever in the methylammonium lead halide (CH3NH3Pb(BrxI1‑x)3‑yCly, 0 ≤
history of PV technology. Having demonstrated an incredibly x ≤ 1)35 and formamidinium lead halide (HC(NH2)2PbBr1‑yIy,
rapid ascent in PCE, reaching beyond 23%,34 which has been y = 0−1).36 The absorption edge wavelength that corresponds
accomplished through simple solution processes without much to the band gap energy of the mixed-halide material changes
sophistication involved in the fabrication methods, the new proportionally to the content of bromide. When iodide is
technology has shown its potential to compete with and added to the perovskite structure, HC(NH2)2PbBr1−yIy (Figure
conquer the existing PV technologies and, therefore, has 4 e), the absorption exhibits a constant red shift from the edge
received enormous attention in recent years. wavelength of ∼550 nm for pure bromide (HC(NH2)2PbI3) to
E DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 5. (a) Structure of organo metal halide perovskite crystal. (b) Band gap structure and energy levels of CH3NH3PbI3. (c) Electronic structure
of CH3NH3PbI3 based on the quasiparticle self-consistent GW approximation. Zero denotes the valence band maximum. Green solid line, red solid
line, and gray dotted line depict bands of I 5p, Pb 6p, and Pb 6s, respectively. Points denoted M and R are zone-boundary points. Reproduced with
permission from ref 40. Copyright 2014 American Physical Society.

∼830 nm for pure iodide (HC(NH2)2PbI3). A constant as shallow traps near the CB and VB. Carriers trapped in
emission peak shift similar to this absorption is also observed shallow defects can be detrapped easily and can contribute to
in the PL spectrum. These optical properties corroborate that current generation. Therefore, PV properties of perovskites are
mixed-halide perovskites can form good solid solutions as ionic not affected by defect formation. The defect-tolerant nature of
crystals. perovskites is reflected by the large carrier diffusion lengths,
As reflected by the strong absorption at the edge wavelength measured over the PL lifetime, which range from 1 μm
and the broad flat absorption at the shorter wavelengths, halide (polycrystalline film)31 to over 100 μm (single crystal).33
perovskites possess excellent optical absorption property useful Furthermore, as an intrinsic semiconductor, MAPbI3 has
for visible light optoelectronics. Methylammonium (MA) lead ambipolar carrier mobility, exhibiting similar effective mass
halide iodide (CH3NH3PbI3, abbreviated as MAPbI3), which is values for both electrons and holes (0.23−0.29),43 which is a
a standard perovskite absorber in PV cells, can be characterized rare property endowed to this ionic crystal. Similar to inorganic
as a rare intrinsic semiconductor,37 exhibiting excellent PV semiconductors such as Si and GaAs, photogenerated
mobility of both photogenerated electrons and holes. Density carriers in MAPbI3 behave as free carriers, and these carriers
functional theory (DFT) and first-principle theory-based can migrate in the perovskite absorber layer without
calculations have been applied by many researchers to recombination for a long time (PL measurements show carrier
corroborate the superior photophysical properties of organo lifetime of several hundreds of nanoseconds).31,44,45 It is also
metal halide perovskites which have been revealed by considered that high ionic density in halide perovskites helps
optoelectronic measurements.31,33 Figure 5 shows the lattice to suppress recombination between electrons and holes by a
structure of MA halide perovskite, MAPbX3 (X = I, Br, Cl), charge-screening effect against Coulombic interaction.
and the band structures of MAPbI338 based on studies by In summary, the important factors that support superior
Kondo et al.15,39 and Brivio et al.40 The valence band (VB) of performance and high efficiency of perovskite solar cells are the
MAPbI3 consists of approximately 70% of I 5p orbitals and following: (1) a high optical absorption coefficient (105 cm−1)
25% of Pb 6s2 orbitals (lone pair), while the conduction band that allows the use of a thin film, (2) a long carrier diffusion
(CB) consists of a mixture of Pb 6p and other orbitals. Here, length and suppressed recombination (defect tolerance), and
VB orbitals have strong coupling between Pb lone-pair 6s2 and (3) a well-balanced charge transfer. In PV applications, the
I 5p orbitals.41 This structure is opposite to that of GaAs, in defect-tolerant nature of halide perovskites is especially
which the VB and CB are formed by p and s orbitals, important, as it leads to generation of high voltage. The high
respectively. PCE of PSCs is a result of high photovoltage (1.1−1.2 V)
The high symmetry and therefore direct band gap of output rather than the amplitude of the photocurrent.38 The
MAPbI3 and p−p electronic transitions from VB to CB, open-circuit voltage (VOC) of all kinds of existing solar cells,
enabled by the Pb s orbital lone pair, contribute to except for GaAs single-junction solar cells, undergoes a large
exceptionally high optical absorption coefficients of this thermal loss from its band gap energy (eV). Such voltage loss
material (105 cm−1).42 The unique defect properties of often constitutes more than one-third of the band gap energy.
perovskite are attributed to strong Pb s−I p antibonding For example, Si p/n junction cells (band gap, 1.1 eV; VOC ≈
coupling, weak Pb p−I p coupling, and its ionic character- 0.7 V), which are highly sensitive to small concentrations of
istics.41 Weak antibonding coupling between Pb p and I p impurity defects, show significant reduction in VOC, especially
orbitals fixes the conduction band minimum (CBM) close to under operation at low incident light. Perovskite solar cells of
the Pb p orbital, and strong Pb s−I p antibonding coupling high efficiency usually generate VOC exceeding 1.1 V, or in the
lowers the valence band maximum (VBM) close to the I p best cells over 1.2 V (1.26 V obtained recently with
orbital. Therefore, for vacancies (defects) formed by removal MAPbI346), with respect to their band gap energy of 1.55−
of I−, the defect state lies between the Pb p atomic orbital level 1.6 eV, and this high voltage is maintained even under weak
and the CBM, and for the Pb2+ vacancy, the defect state is intensity light, which is comparable with the behavior of GaAs
formed between the I p and VBM levels. Hence, unlike other solar cells. There is still a substantial loss in VOC. One property
ionic semiconductors, where localized nonbonding orbitals related to the loss of efficiency that is limits the VOC and
surrounding the ion vacancies form trap states deep within the efficiency of PSCs is defect (trap) density in halide perovskite
band gap, such defects in perovskites (MAPbI3) generate trap crystals, which is estimated to be from 1010 cm−3 (single
states that either reside within the bands (VB or CB) or exist crystals)33,47 to 1016 cm−3 (polycrystals).45,48 This indicates
F DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 6. Variety of device architectures for perovskite solar cells.

that the purity of solution-processed polycrystalline films can TAD/metal contact). In a p-i-n type structure, which is
be an important factor to minimize defect and trap otherwise known as an inverted architecture, perovskite is
concentrations, and thus improve efficiency further. Equally sandwiched between a p-type material (for example, PEDOT-
important are the losses caused by interfacial recombination PSS) at the bottom and an n-type layer (for example, PCBM)
that depend on the properties of other layers in the device. at the top (e.g., FTO or TiO2/PEDOT-PSS/perovskite/
Hence, understanding the working of PSCs in standard PCBM/metal contact). Some typical architectures of PSCs
architectures and using optimized electron and hole transport are illustrated in Figure 6. The perovskite absorbs light to
layers are important for further improvements in performance. generate electrons and holes. The electrons are selectively
2.2. Working Principle collected by the n-type ETM layer, while the holes are
collected by the p-type HTM layer. Electrons flow through the
Based on our prior understanding of the working of DSSCs, we external circuit to reach the p-type layer and combine with the
presumed that perovskite in the very first studywhere it was holes. Figure 7 shows the band diagram of a typical metal oxide
coated on a TiO2 mesoporous photoanode in a liquid
electrolyte-based electrochemical cellworks the same way
as dye works in a DSSC.24 The perovskite absorber, adsorbed
on a TiO2 surface, undergoes photoexcitation by absorbing
light of wavelength matching with its band gap, and then
electrons from the excited state (LUMO) of the perovskite are
injected into the CB of TiO2 and transported through the layer
to reach the FTO substrate. From there it flows in an external
circuit to the counter electrode (cathode), where it combines
with oxidized species of electrolyte. The reduced species then
diffuses to the vicinity of the oxidized perovskite to reduce it.
However, studies on perovskite devices with an insulating
Al2O3 mesoporous layer,24,29 and even without any meso-
porous layer (planar heterojunction with compact TiO2 layer
for collecting electrons), implied that electrons and holes can
be transported through the perovskite film itself, without the
need for any semiconducting scaffold as is required in the case
of sensitized solar cells. This gave sufficient grounds to
differentiate the working of the perovskite solar cell from the
mechanism of sensitization, and further progress in the field Figure 7. Energy diagram of a typical perovskite solar cell using
MAPbI3 as the perovskite absorber, TiO2 as the ETM, and spiro-
and the diversity of device architectures made it clear that the OMeTAD as the HTM. FTO and Au are the front and back contacts.
PSCs function more like solid-state p-n junction solar cells, in
contrast to the sensitization concept. As widely accepted today,
the working principle of PSCs closely matches those of n-i-p ETM-based perovskite device (FTO/TiO2/perovskite/spiro-
and p-i-n solar cells, where perovskite works as an intrinsic OMTAD/Au). Although this simple working principle is
absorber sandwiched between two selective contacts (p and n). widely accepted at present, the processes of carrier generation,
In the n-i-p structure, TiO2 (compact or mesoporous layer) separation, and transport are not as straightforward as they are
works as an n-type electron transport material (ETM), in amorphous Si p-i-n solar cells.
perovskite functions as the intrinsic (i) absorber, and a In p-i-n or n-i-p solar cells, the difference between quasi
HTM like spiro-OMeTAD (an organic molecule) works as the Fermi levels at the selective contacts (i.e., between p-type and
p-type contact (FTO or ITO/TiO2/perovskite/spiro-OMe- n-type layers) limits the open-circuit voltage. In other words,
G DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 8. (Right) Major milestones and activities of progress in perovskite solar cells from 2005 to 2018. (Left) Power conversion efficiency
progress chart.

the difference between work functions of the selective contacts carrier and lattice dynamics are expected to help us understand
affects the VOC of p-i-n- or n-i-p-type solar cells. In PSCs, the the exceptional PV characteristics of PSCs more precisely.
lack of a clear trend and the dispersion of VOC dependences on 2.3. Major Milestones of Progress
different selective contacts are rather indicative of some
unusual behavior of carriers in the perovskite. In fact, organic− A number of developments at the fronts of both device
inorganic hybrid perovskites demonstrate a number of unusual performance and basic scientific understanding have happened
photophysical characteristics, which indeed culminate in during the journey from 3.8% to >23% PCE. Among the early
excellent PV properties in the device. It is generally accepted developments, fabrication of a high-quality perovskite film was
that the exceptionally high PCE and high VOC achieved with a major focus because it was the first and foremost requirement
perovskite-based solar cells are due to the perovskite’s for achieving high cell efficiency. A great amount of effort went
into development of processes to prepare perovskite films of
remarkably superior properties, such as high absorption
good quality. Initially, a wide variation in the processing
coefficient (α > 105 cm−1), ultralong carrier diffusion
methods and conditions followed by different groups, and the
length,31,42,49 long carrier lifetime,50,51 moderate carrier
high sensitivity of perovskite to even slight modifications in
mobility,32,52,53 unusually high defect tolerance, and slow
these methods, had unfortunately made it difficult to find a
carrier recombination.53 The origin of such exceptional PV
general correlation between the perovskite film characteristics
characteristics is still being investigated, and debates on the and performance. However, as more and more studies
nature of charge carriers in these materials are still going on. pertaining to the influences of perovskite film morphology
While some theoretical studies have suggested the potential on cell performance were undertaken, it became clear and
role of intrinsic ferroelectricity or bulk PV effects in the comprehensible that a flat and dense (pinhole-free) perovskite
effective separation of charges,54,55 some other studies have absorber film with large grains and high crystallinity is a
focused on the role of surface fields and diffusion.56 Carrier− prerequisite to high efficiency. A variety of methods, such as
lattice interactions57−59 and large polaron formation60,61 have anti-solvent dripping,63,64 solvent-assisted vapor anneal-
also been proposed to produce such unique characteristics of ing,65−68 inclusion of additives in precursor solution,69,70 and
the perovskites. More recently, Guzelturk et al. witnessed a even the use of special annealing conditions,71−73 have been
strong and coherent carrier−lattice coupling effect in perov- employed to achieve the required characteristics of perovskite
skite, which was found to be important for both resonant and films that culminate in high efficiency. The effects of
far-above-gap photoexcitation.62 Their study indicates that processing methods and film morphology on performance
ultrafast lattice distortions play a key role in the initial have been comprehensively reviewed.74,75 Almost all parame-
processes associated with charge transport. Hence, all of the ters involved in the solution processes(i) precursor solvent,
different observations and varying explanations reported in (ii) nature, concentration, and ratios of different precursor
different studies lead to ambiguity about the nature of defects ions, (iii) nature and concentration of additives (besides the
and carrier dynamics in perovskites. A clear and concrete active ions), (iv) drying temperature, time, and environment
understanding of the processes of carrier generation, (humidity level, controlled solvent or antisolvent vapor
separation, transport, and even accumulation is currently exposure), etc.influence the morphology of the films. In
lacking. It seems that the complexity of such studies lies in the fact, the interdependence of most of the above parameters
interdependence of a number of film characteristics: grain makes it difficult to differentiate the effects/contributions of
boundaries, compositional non-uniformity, phase impurity/ different parameters on the cell performance; for example, the
purity, interfaces, etc. Further deeper insights into in situ use of additives in the solution changes the solution
H DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 9. (a) Cross-sectional scanning electron micrograph of layered structure and (b) photocurrent−voltage (J−V) characteristics of a
mesoporous Al2O3-based perovskite solar cell. J−V curve shows hysteresis for forward and reverse scans of voltage. Reproduced with permission
from ref 24. Copyright 2015 Chemical Society of Japan.

concentration and film morphology as well as the composition, 3. METAL OXIDE-BASED ELECTRON TRANSPORT
the annealing temperature can change the morphology as well LAYERS IN PEROVSKITE SOLAR CELLS
as the phase purity in some mixed perovskites, and so on. Metal oxide semiconductors, typically TiO2, have been the
Regardless, a good-quality perovskite film (pinhole-free, large most popular electron transport materials (ETMs) and
grains) is the basic requirement for a device to perform well. scaffolds employed in PSCs. The use of TiO2 in PSCs is an
In those early years, a lot of attention was also given to issues extension of DSSC technology, as we had started developing
of anomalous J−V hysteresis and the important roles of the first perovskite-based device by using mesoporous TiO2 as
different interfaces in device performance. Critical effects of the the ETM.1 n-Type metal oxide mesoporous semiconductor
interfaces of perovskite with both electron and hole transport materials in PSCs play dual functions: one is selective electron
layers on device performance have been highlighted in a extraction as ETM and other is provision of a large surface
majority of studies, which are comprehensively reviewed in refs scaffold to accommodate perovskite crystals. The latter
76 and 77. Although there is no clear consensus on the origin function is assumed to be essential for collecting electrons
of J−V hysteresis as of now, a number of methods based on from the thick perovskite layer, especially in the case where the
interfacial engineering have been established to either reduce diffusion length of excited electrons is short due to
or eliminate hysteresis. The use of organic electron transport recombination inside the perovskite layer. However, because
layers like PCBM78 or C6079,80 and modification of the most of the long diffusion length of free carriers (>1 μm) in
widely used TiO2 ETM layer surface with organic mole- polycrystalline perovskite films,31,93 which generally work as
cules81−83 have become popular interface engineering absorbers with thicknesses of more or less than 0.4 μm,
strategies to reduce J−V hysteresis. While a majority of recent efficient electron extraction can also be achieved with use of a
studies suggest ion migration to be responsible for such nonporous flat ETM film, a dense TiO2 film, or even organic
anomalous hysteresis,84−87 some other studies claim that traps materials such as PCBM. In fact, the efficiency obtained with
at the interface and interfacial recombination cause the PCBM-based inverted structure devices94,95 has been close to
hysteresis.88−90 Possible mechanisms of origin of hysteresis those observed with TiO2-based devices. Nevertheless, many
and methods of reduction/elimination of hysteresis are high-performance devices with PCE exceeding 20% have been
thoroughly discussed in book chapters and also in some fabricated on mesoporous TiO2 layers with varying composi-
review articles.91,92 tions of perovskites. Although there is no comprehensive study
In recent years, key developments have been reported in on structural and compositional difference at the interface of
simultaneous improvement of efficiency and stability by perovskite with mesoporous TiO2 and compact TiO2,
compositional engineering of perovskites. Structural/intrinsic crystallization of perovskite on the two ETLs must be very
different, which can certainly alter the interface.
stability and long-term environmental stability of perovskites
In a regular device structure, in which ETM is coated on
have become the most important issues at the present time. In
transparent conductive substrates such as FTO and ITO, metal
the roadmap of progress, as presented in Figure 8, we can see
oxide ETMs are convenient to use in terms of their chemical
that the first few years of research saw a lot of effort focused on stability (insolubility) against perovskite precursor solutions
exploiting the “ability/efficiency” of perovskites and presently that are deposited on the ETM. Metal oxide ETMs are not
this focus has been shifted to “stability” of the material. Hence, necessarily employed in the form of mesoporous layers and can
in this Review, we first present a brief overview of role of metal be used as nonporous compact layers (CLs) in the planar
oxide electron transport layer in PSCs, touching on some of structure devices. The existence of a CL is essential to ensure
the key aspects, followed by more comprehensive discussions selective electron extraction and hole-blocking function at the
on different recent and important developments, such as (i) surface of the negative electrode. It is also a prerequisite even
compositional engineering of perovskites (mixtures of cations when a mesoporous metal oxide ETM layer is used as the
and anions that have resulted in concurrent enhancements in scaffold for the perovskite layer. Interestingly, when we used a
performance and stability), (ii) increasing potential of all- mesoporous Al2O3 layer on the TiO2 CL in a solid-state
inorganic perovskites, (iii) Pb-free perovskites, (iv) stability perovskite device,29 we found higher efficiency and VOC
issues with perovskite and with other layers, and (v) challenges compared to those observed with a meso-TiO2-based cell
and potentials of scaling up perovskite PV. Present challenges used as reference. In this architecture, because of its
are discussed, and our views on future prospects are presented. nonconductive (insulator) property, Al2O3 does not conduct
I DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 10. (a) Cross-sectional scanning electron micrograph of ALD-SnO2-based perovskite ((FAPbI3)0.85(MAPbBr3)0.15) solar cell. (b) J−V
characteristics of both ALD-TiO2- and ALD-SnO2-based planar heterojunction PSCs. (c) Schematic illustration of the existence (in the case of
TiO2) or absence (in the case of SnO2) of an energy barrier between the electron transport material (ALD-TiO2 or ALD-SnO2) and the mixed
perovskite. Reproduced with permission from ref 102. Copyright 2015 Royal Society of Chemistry, under Creative Commons Attribution
NonCommercial 3.0 Unported License.

Figure 11. SnO2- and ZnO-based planar structure perovskite solar cells made by a low-temperature process. (a) Illustration of device structure and
(b) energy level diagram showing the electron transport mechanism for a SnO2-based perovskite cell using MAPbI3 as absorber, in which PbI2 was
used as an intermediate layer to suppress recombination. (c) J−V characteristics of the SnO2 perovskite cell showing VOC = 1.08 V. Reproduced
with permission from ref 103. Copyright 2015 Royal Society of Chemistry. (d) J−V characteristics of the ZnO-based perovskite solar cell, using
Csx(MA0.17FA0.83)1−xPb(I0.83Br0.17)3 as absorber, showing VOC = 1.11 V. Reproduced with permission from ref 104. Copyright 2017 Royal Society
of Chemistry.

the electrons, but its role is to accommodate perovskite into its the case for PSCs. Perovskite solar cells using SnO2 and ZnO
porous network so that carriers are conveyed to the substrate exhibit VOC and JSC comparable with those of TiO2-based cells.
(electrode) through the infiltrated perovskite. This is, in fact, Indeed, one of the record efficiencies has been obtained with
enabled by the long distance diffusivity of carriers in SnO2.102 The differences between DSSCs and perovskite cells
perovskite. In this aspect, a meso-Al2O3-based perovskite cell must be due to differences in their device structure and
works basically the same as a planar heterojunction structure working mechanism. The thicknesses and volumes of metal
cell without a mesoporous ETM layer. As examples of further oxide semiconductors employed in both cells are very different.
developments in such Al2O3-based devices, we accomplished a It is reasonable that a thin film (compact layer, <100 nm) of
PCE of 17% (Figure 9),24 while Snaith et al.97 also obtained a such metal oxide semiconductors can behave differently than
PCE up to 16.7% in a meso-Al2O3-based MAPbI3 solar cell. the thick (5−10 μm) and mesoporous layers. Such thick metal
However, after those reports, much less work was done on the oxide layers in DSSCs predominantly exhibit bulk semi-
use of Al2O3 because devices with Al2O3 scaffolds often conductor properties, whereas thin metal oxide films in
showed low fill factors (FFs), attributed to high impedance of perovskite cells are assumed to function as current-rectifying
the Al2O3 network, which limited the cell’s performance. layers with high electron selectivity (hole-blocking ability).
However, the ability to use Al2O3 provided the first evidence Even then, their properties can vary with the methods of
that perovskite is capable of transporting carriers over long preparation. In PSCs this has been, in fact, strongly indicated
distances without significant recombination.24 by the results of atomic layer deposition (ALD)-based TiO2
A number of metal oxide semiconductors other than TiO2, and SnO2 employed in planar heterojunction structure
which had been already employed in DSSCs, have also been PSCs.102 Hagfeldt et al.102 used an ALD-based thin SnO2
used as ETMs in PSCs. SnO298,99 and ZnO100,101 have been layer (15 nm) as the ETM in planar structure cells with mixed-
frequently employed as alternatives to TiO2 in PSCs. In cation perovskite absorber ((FAPbI3)0.85(MAPbBr3)0.15), in
DSSCs, the energy level of the CB significantly influences the which formamidinium (FA) was mixed with MA. This cell
VOC and short-circuit photocurrent (JSC) of the cells. In achieved 18.4% PCE in the best conditions of film preparation
particular, as VOC in DSSCs is determined by the energy gap (Figure 10). Under the same conditions, an ALD-based TiO2
between the CB level of the semiconductor and the redox thin layer resulted in lower efficiency of the cell, while VOC
potential of the iodide-based redox agent in the electrolyte values in both cases were same. One thing that was surprising
(HTM), SnO2, having a deep CB level (large work function), in this ALD-TiO2-based cell was the photocurrent, which was
always results in low VOC. However, interestingly, this is not as low as 5 mA/cm2, in comparison to >20 mA/cm2 obtained
J DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 12. Energy levels of lead halide perovskite absorbers and various metal oxide electron transport materials and hole transport materials
(HTMs) employed in solar cell devices. In the perovskites, MA and FA stand for methylammonium and formamidinium, respectively. PCBM
denotes [6,6]-phenyl-C 61-butyric acid methyl ester. In HTMs, PTAA, P3HT, and PEDOT:PSS denote poly[bis(4-phenyl)(2,4,6-
trimethylphenyl)amine, poly(3-hexylthiophene-2,5-diyl), and poly(3,4-ethylenedioxythiophene)−polystyrene sulfonate, respectively.

with ALD-SnO2. As explained by the authors, the poor analyzed in high-vacuum conditions, such as by ultraviolet
photocurrent in the case of TiO2 was due to band photoelectron spectroscopy (UPS) and X-ray photoelectron
misalignment between perovskite and TiO2, forming an energy spectroscopy (XPS), do not necessarily indicate the actual
barrier against electron transfer. It must be noted that the low energy levels of thin films exposed to ambient air and or of
photocurrent in that case is not even comparable with those of those in chemical contact with other charge transport or
solution-based TiO2 films, which often produce current >20 absorber materials in the device structure. In this regard, the
mA/cm2 at even not-so-perfectly optimized conditions. This summary in Figure 12 is only a reference for approximate
result asserts the greater importance of thin-film properties comparison of energy levels to estimate the possibility of
over bulk properties of metal oxide ETMs in PSCs. charge-transfer processes.
As different metal oxides (TiO2, ZnO, SnO2, etc.) have All of these ETMs are capable of sufficiently high VOC. VOC
different work functions and conductivity, the rate of electron of the device tends to be more influenced by the quality of the
transfer from the perovskite varies with the different metal perovskite film and its heterojunction interfaces than by the
oxides used in the cell. However, TiO2, ZnO, and SnO2 ETMs bulk properties of the ETM. The kinds of defects and density
have all shown VOC > 1 V, indicating insignificance of the of defects that strongly affect the carrier collection and
energy level of these layers in determining the VOC of the cell. recombination at the ETM/perovskite interface are also
Figure 11 displays examples of planar structure PSCs using important to the overall performance of the cells. One of our
thin, compact films of SnO2103 and ZnO104 as ETMs, prepared studies, where we explored the use of a TiO2−MgO bilayer as
by a low-temperature-based (non-sintering) solution-coating the ETM layer (ETL), indicated that defects (trap states)
process.103 Although the device efficiency changes depending present in TiO2, which can be largely influenced by
on the metal oxide and perovskite is used, which affects the JSC preparation methods, can cause losses in the open-circuit
and FF, the VOC values of both devices are almost same (1.08 voltage of the cells.116 Defects/traps at the ETM/perovskite
and 1.12 V), comparable with those obtained with TiO2-based interface have been found to directly or indirectly influence the
cells. cell’s performance, especially J−V hysteresis. Different surface
In addition to TiO2, SnO2, and ZnO, a variety of other metal modification techniques have been adapted to change the
oxides, such as Nb2O5,105−107 WO3,108,109 Nb-doped TiO2,110 properties of the ETM so as to improve its interface with
Mg-doped ZnO,111 MgO/TiO2,112 etc., have been investigated perovskite. The roles of metal oxide ETMs and of the interface
for use in PSCs. Figure 12 shows a summary of the energy between perovskite and ETM in device performance are
levels (work function) of some typical metal oxide electron comprehensively reviewed by Mahmood et al.117 and
extraction materials in relation to those of perovskite absorber Fakharuddin et al.,77,118 respectively. In general, the physical
materials and HTMs. Energy level data listed here are based on and optoelectronic characteristics of ETMs influence the
our measurements103,105,113−115 and also published studies. performance of PSCs remarkably and need to be optimized
For metal oxide materials, the band gap energy and CB level specifically according to the device structure for high and stable
do not necessarily show good agreement with the open-circuit performance.
voltage observed in PSCs, probably due to its dependence on
the form of material subjected to measurement. Such 4. COMPOSITIONAL ENGINEERING OF PEROVSKITES
mismatches of values, if not large, can originate in the
differences between single bulk crystal and polycrystalline 4.1. Mixed Compositions
films, film samples in vacuum and in air, and pure and defect- In the past few years, while the efficiency of PSCs continued
rich samples. Furthermore, it is considered that energy levels increasing, the long-term stability of the cells also went up
K DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

remarkably. One major development that has stood out among methylammonium (MA), formamidinium (FA), Cs, and Rb
others and contributed substantially to enhanced stability is and anions like I, Br, and Cl, and their combinations, have
incorporation of different cations in the A-site and different been explored in the past few years. Table 1 provides a list of
halides in the B-site. This compositional play with the A-, B-, and X-site ions and their sizes that are or can be used in
perovskite has not only exerted a strong influence on efficiency certain combinations to form perovskite structures. Out of all
but also raised the stability substantially. Although such possible combinations,124 the mixed perovskite that has
compositional mixing does not always succeed in forming a become most popular in the recent years is (MA/FA/
homogeneous solid solution, certain combinations of cations in Cs)Pb(I/Br)3, which is commonly known as a triple-cation-
the A-site and halides in the B-site of the perovskites have based perovskite. The quadruple-cation-based perovskite,
demonstrated superiority over single-cation/halide perovskites including Rb (i.e., (MA/FA/Cs/Rb)Pb(I/Br)3) as the fourth
in terms of both efficiency and stability. Indeed, all the cells cation, has also gained interest due to its high cell efficiency
published in the NREL chart of certified efficiency, except for and stability. Although it was believed earlier that Rb occupies
the first one, comprise a mixture of cations or anions or both the A-site, a recent study has revealed that it likely does not sit
(Figure 13). in any lattice site and is instead expelled out to grain
boundaries.125,126
4.1.1. A-Site Cations Mixture. According to DFT
calculations, in lead halide perovskites, Pb 6s 6p−I 5p
interactions lead to generation of the two bands: the valence
band maximum (VBM) is formed by antibonding (σ*) Pb 6s−
I 5p interactions, while the conduction band minimum (CBM)
is formed by empty Pb 6p orbitals128 and/or by Pb 6p−I 5p
interactions.129,130 Therefore, the cations in the A-site are
considered not to contribute directly toward the band
structure, but they play a significant role in providing structural
stability by charge compensation within the PbI6 octahedra,
largely based on their electrostatic (van der Waals)
interactions131 with the inorganic cage. Nevertheless, any
change in the size of cations in the A-site can either contract or
expand the crystal lattice, thereby altering the optical
properties of the perovskite. Smaller cations like Cs and Rb
Figure 13. Perovskite solar cell performances certified by NREL are expected to contract the lattice and thus increase the band
(14.1%,119 16.2%,63 17.9%,120 20.1%,121 21.02%,122 and 22.1%123). gap, while larger cations like formamidinium (FA+) are
The compositions of perovskites used in three recent high-efficiency supposed to expand the lattice and decrease the Eg. FA+ was
cells (22.7%, 23.3%, and 23.7%) are not in the public domain yet. the first cation that was used instead of methylammonium
(MA+). FA+, having a larger ionic radius (r = 0.253 nm) than
Based on ionic size and geometrical tolerance factor (τ = MA+ (r = 0.217 nm), expands the crystal a bit, resulting in
rA + rX decreased Pb−I bond distance, which eventually lowers the
),
which is an empirical index widely used for
2 (rB + rX ) band gap. As measured, pure FAPbI3 shows a band gap (Eg) of
predicting perovskite crystal structure, different cations such as 1.47 eV,36 while that for pure MAPbI3 is 1.55 eV. Although,

Table 1. Ionic Radii of Some of Common A-Site Cations, B-Site Cations, and X-Site Anions Used in Hybrid Perovskitesa
effective ionic radius metal ion effective ionic radius anion effective ionic radius
cation (A-site) (pm) (B-site) (pm) (X-site) (pm)
NH4+ 146 Pb2+ 119 F− 129
methylammonium [CH3NH3]+, 217 Sn2+ 110 Cl− 181
(MA)
formamidinium, [CH(NH2)2]+, (FA) 253 Ge2+ 73 Br− 196
hydrazinium, [NH3NH2]+ 217 Mg2+ 72 I− 220
azetidinium, [(CH2)3NH2]+ 250 Ca2+ 100
hydroxylammonium, [NH3OH]+ 216 Sr2+ 118
imidazolium, [C3N2H5]+ 258 Ba2+ 135
ethylammonium, [(CH3CH2)NH3]+ 274 Cu2+ 73
dimethylammonium, [(CH3)2NH2]+ 272 Fe2+ 78
guanidinium, [(NH2)3C]+ 278 Pd2+ 86
tetramethylammonium, [(CH3)4N]+ 292 Eu2+ 117
thiazolium, [C3H4NS]+ 320 Bi3+ 103
3-pyrrolinium, [NC4H8]+ 272 Sb3+ 76
tropylium, [C7H7]+ 333
K+ 164
Rb+ 172
Cs+ 188

a
Adapted from ref 127. Copyright 2015 Springer Vienna.

L DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 14. (a) Schematic illustration of the importance of the size of the A-site cation in determining the tolerance factor and structural properties
of perovskite. Substitution of small amounts of FA with Cs on the A-site results in more structurally stable compounds. Changes in UV−vis spectra
of (b) FAPbI3 and (c) FA0.85Cs0.15PbI3 over 18 days. (d) XRD patterns of original FAPbI3 and FA0.85Cs0.15PbI3 films and those films after 30 days of
storage. (e) Change in XRD patterns of FAPbI3 thin film after its exposure to high humidity. (f) Photos of FAPbI3 and FA0.85Cs0.15PbI3 films under
high-humidity conditions. (g) J−V curves of FA0.85Cs0.15PbI3 solar cells at 0−15 days of storage under 15% RH. (h) Normalized PCE of FAPbI3
and FA0.85Cs0.15PbI3 solar cells at different storage times. Reproduced with permission from ref 141. Copyright 2016 American Chemical Society.

based on the Eg value (1.47 eV), FAPbI3 is supposed to degradation) as easily as MA.138 Therefore, partial replacement
perform better than MAPbI3 due to its extended absorption of MA with FA in MA-based perovskites also improves the
edge (800 nm), the best efficiency achieved so far with FAPbI3 thermal stability of the perovskites (FAxMA1‑xPbI3) remark-
is ∼18%,132 while that for pure MAPbI3 is >20%.133−135 With ably. Improved structural stability, associated with an increased
an extended absorption edge, FAPbI3 cells show higher tolerance factor and a stronger interaction of FA with iodide
photocurrent (JSC), but their efficiency is mainly limited by ions, results in greater thermal stability of the FAxMA1‑xPbI3
poor FF, which is possibly related to phase instability of perovskites. However, the FA inclusion strategy falls short of
FAPbI3. FAPbI3 readily crystallizes into a photo-inactive phase long-term durability because FA is, unfortunately, more
(δ-FAPbI3) at room temperature (RT), and this phase is hygroscopic than MA and, thus, shows greater vulnerability
transformed to a photo-active black phase (α-FAPbI3) at to humidity.138 Therefore, incorporation of the inorganic
temperatures between 125 and 165 °C.120 This α-FAPbI3, cation Cs became an alternate choice. Indeed, it has been
formed at high temperatures, slowly transforms into δ-FAPbI3 found that substitution of MA with Cs simultaneously
when kept at RT. However, partial substitution of FA with MA improves efficiency and thermal stability by a certain amount.
increases the stability of α-FAPbI3 at RT. Specifically, a As reported by Niu et al.,139 about 9 mol% of Cs in
trigonal α-FAPbI3 phase is stabilized at RT with a composition CsxMA1‑xPbI3 (x = 0.09) shows better performance (18.1%)
of 20 mol% of MA in MAxFA1‑xPbI3 (x = 0.2), although the and thermal stability (no color change when heated at 120 °C
best efficiency is reported (∼18.3%) for FA-based perovskites for 3 h) than pristine MAPbI3 (15.8%), while a higher
having a composition of MA0.4FA0.6PbI3. For 0.2 ≤ x ≤ 1, concentration of Cs shows surprisingly worse stability. The
MAxFA1‑xPbI3 exists in the tetragonal phase, indicating the unencapsulated Cs0.09MA0.91PbI3 cell retains above 80% of its
dominance of MAPbI3, which is stabilized in a tetragonal phase initial performance, whereas the performance of the pure
at RT. It has been proposed that, when a MA+, which has MAPbI3 cell deteriorates to less than 40% of its initial
almost 10 times greater dipole moment than FA+, is performance after heat treatment at 85 °C for 60 min. From
incorporated, it exhibits stronger interactions with the PbI64− temperature-dependent reflectance measurements, Gong et
octahedra and thus stabilizes the 3D arrangement of α-FAPbI3 al.140 also found that regular MAPbI3 degrades completely
with little lattice shrinkage or changes in the optical when heated at 200 °C for 10 min, changing its absorption
properties.136 Although the efficiency of FAPbI3 cells falls edge to that of PbI2, while the absorption spectrum of
behind that of the MAPbI3 cells, FAPbI3 has a significantly Cs0.05MA0.95PbI3 remains almost unchanged under the same
greater thermal stability compared to MAPbI3.137 FAPbI3, treatment conditions. To our knowledge, there is no study that
when heated at 150 °C for hours, does not change color, directly compares the moisture stability of MAPbI3 with that of
whereas MAPbI3 becomes yellow (formation of PbI2) after (Cs/MA)PbI3. However, there are several reports on
heating at 150 °C (or even lower) for just 30 min.36 It is comparisons of the moisture resistance of FAPbI3 with that
proposed that FA has a stronger interaction with iodide and, of (Cs/FA)PbI3, showing results that strongly endorse the fact
therefore, does not allow easy breaking of the network. It is that moisture stability is improved by Cs inclusion in the
also explained that FA, being less acidic than MA, does not perovskite. Small and large tolerance factors respectively for
undergo deprotonation to furnish HI (the first step of CsPbI3 (τ = 0.85) and FAPbI3 (τ = 0.98) stabilize the
M DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 15. Changes in UV−vis absorbance spectra of (a) FAPbI3, (b) Cs0.05FA0.95PbI3, and (c) Rb0.05FA0.95PbI3 perovskite films stored at 85% RH,
25 °C, and in the dark for different durations. As highlighted by this experiment, the stability of Rb0.05FA0.95PbI3 films was significantly superior
even to that of Cs0.05FA0.95PbI3. Reprinted with permission from ref 145. Copyright 2015 John Wiley and Sons.

Figure 16. UV−vis absorption spectra and photographs of MAPbI3‑xBrx. (a) UV−vis absorption spectra of FTO/c-TiO2/mp-TiO2/MAPbI3‑xBrx/
Au cells. (b) Photographs of TiO2/MAPbI3‑xBrx bilayer nanocomposites on FTO glass substrates. (c) Quadratic relationship of the band gaps of
MAPbI3‑xBrx as a function of Br concentration (x). (d) Power conversion efficiencies of the heterojunction MAPbI3‑xBrx solar cells as a function of
Br composition (x). (e) J−V characteristics of the MAPbI3‑xBrx cells (x = 0, 0.06, 0.13, 0.20, 0.29, 0.58, 1.0). Reprinted with permission from ref
158. Copyright 2013 American Chemical Society.

perovskites in orthorhombic (yellow) and hexagonal structures remain stable (Figure 14f), and correspondingly, the perform-
(Figure 14a), respectively. Due to the difficulty in determining ance of the FA0.85Cs0.15PbI3 devices are found to be more
the ionic radii of organic cations accurately, slight discrepancies stable than that of the pure FAPbI3 cells (Figure 14g,h).141 In a
may occur between the structures predicted from the tolerance similar study, Lee et al.143 observed that 10% of Cs in FAPbI3
factor values and the structures observed experimentally. For (i.e., Cs0.1FA0.9PbI3) improved the cell performance from
instance, while some reports141 agree on formation of the 16.3%, measured from pure FAPbI3 cells, to 17.1% in
hexagonal structure of FAPbI3 at RT, some other studies Cs0.1FA0.9PbI3. Interestingly, the absorbance of the Cs0.1FA0.9-
report a cubic structure142 of the same. Nonetheless, tuning of PbI3 film and performance of the cells made with Cs0.1FA0.9-
the tolerance factor by mixing ions of different sizes has been a PbI3 deteriorated much less in comparison to those of FAPbI3
successful strategy to stabilize the perovskites in cubic when the films and cells were aged under light/dark in a humid
structures. For example, the effective tolerance factor can be atmosphere. Similar results of higher efficiency and better
tuned to form a cubic structure (0.9 ≤ τ ≥ 1) by alloying stability were also reported for Cs0.2FA0.8PbI3-based devices by
CsPbI3 with FAPbI 3 (FA xCs 1‑x PbI3 ) over a range of Yi et al.144
proportions (Figure 14a). In comparison to pure FAPbI3, the In addition to Cs, Rb and K have also attracted attention, as
structural/phase stability and therefore the cell performance FAPbI3 perovskites including either Rb or K have demon-
are improved for a mixed FA/Cs composition with 15% of Cs strated improvements in the PV performance of the cells.
(i.e., FA0.85Cs0.15PbI3). It has been also observed that FAPbI3 Although its location in the crystals is still a matter of
perovskite thin films, when exposed to a humid environment controversy, it is believed that a small amount (x ≤ 0.05) of
for 18 days, degrade fast, while FA0.85Cs0.15PbI3 thin films Rb+ can be included in FAPbI3, and higher concentrations lead
N DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 17. (a) Dependence of powder XRD peaks on the value of x in (FAPbI3)1‑x(MAPbBr3)x single crystals. The hollow circles represent data
points, and the solid lines are Gaussian fits of the data. The shift of all the peaks to higher 2θ implies contraction of the lattice due to alloying of
FAPbI3 with MAPbBr3. Reproduced with permission from ref 163. Copyright 2017 American Chemical Society. (b) Schematic illustration of strain
relaxation after MAPBr3 alloying into FAPbI3 (side view). Reproduced with permission from ref 164. Copyright 2016 American Chemical Society.

to phase segregation.125 Devices based on Rb-mixed FAPbI3 observed clearly and directly. The influence of Br on the
(i.e., Rb0.05FA0.95PbI3) outperform those based on FAPbI3, and optical and electronic properties has been remarkable and
more importantly, the stability of this Rb0.05FA0.95PbI3 film, as recorded well. Incorporation of smaller Br − ions in
shown in Figure 15, against humid conditions is superior to MAPbI3‑xBrx increases the band gap of the mixed-halide
that of Cs0.05FA0.95PbI3.145 perovskite, and this increment in Eg follows a quadratic relation
K, being even smaller than Rb, has been also used in mixed- with the concentration of Br, as show in Figure 16c. Through
cation compositions, and its effect on the performance of the such band gap tuning, Noh et al.158 reported an initial best
cells has been found to be significant, which is discussed with efficiency above 12% with Br content <10% (Figure 16d). This
regard to mixed cation−mixed halide in the following section. was the first study on a MAPbI3‑xBrx mixed perovskite.
4.1.2. X-Site Anions Mixture. Like different cations in the Although the PCEs of cells with higher Br contents (>20%)
A-site, different halides, like Cl, Br, and even non-halide/ were lower, the cells displayed better resistance against high
pseudohalide ions like SCN−, have been incorporated into humidity (RH = 55%), which was correlated with a tetragonal-
pure MAPbI3, FAPbI3, or mixed-cation (FA/MA, FA/MA/Cs, to-pseudocubic structural transition (at x = 0.13). Following
FA/Cs, FA/MA/Cs/Rb, etc.) lead iodide perovskites. But, this first report, a number of reports came out showing
unlike cations, different halides mixed in the X-site impart a continuous improvement in cell efficiency, which was
dramatic effect on optical and electronic properties, absorption accomplished through optimization of the methods of
and emission spectra (band gap), and carrier lifetime and preparation. A certified PCE of 16.2% was reported by Jeon
diffusion length. Although the ambiguity of inclusion of Cl in et al.,63 which they achieved by using a solvent engineering
pure iodide perovskites like MAPbI3 remains unresolved, a method to prepare a uniform and dense MAPbI3‑xBrx film.
majority of reports claim that Cl easily sublimes out of the In a similar manner, FA-based mixed I/Br (FAPbI3‑xBrx)
perovskite film during preparation and, therefore, does not perovskites with a range of Br concentrations (x = 0−1)36 were
exist in the final film,146−148 despite the starting solution studied, and a similar effect of increasing band gap with Br
containing Cl-based precursors (PbCl2 or MACl). In contrary, incorporation was observed. But the interesting and surprising
a good number of studies also show evidence of the existence fact about FAPbI3‑xBrx is that it does not form any crystalline
of a trace amount of Cl in MAPbI3,149−152 even though the phase (amorphous) for 2.3 ≤ x ≥ 2.5.36 The origin of this
amount measured was always substantially lower than that amorphous regime is not yet understood. However, like in the
used in the starting materials. It is presumed that MAPbI3‑xClx case of MAPbI3‑xBrx,159 photoinduced phase segregation was
is either a metastable phase or has a high formation energy153 also observed in FAPbI3‑xBrx.160 A detailed discussion about
and, therefore, is not formed in the final perovskite film despite such photoinduced phase segregation is presented in section
the starting materials containing Cl. Nevertheless, distinct 5.1.2.
effects of Cl present in the precursor solution on the quality, A few studies have been undertaken on systems like
morphology, and crystallinity of the perovskite films have been MAPb(Br/Cl) and MAPb(I/Br/Cl). In contrast to MAPb(I/
observed in all the studies. PbCl2 as the source of Pb- or Cl- Cl), Cl has been found to coexist in MAPb(Br/Cl). The
based additives, such as HCl,154,155 NH4Cl,156 and MACl,157 smaller difference in ionic radii (RI− = 2.07 Å, RBr− = 1.84 Å,
improves the film quality by slowing down the crystallization, RCl− = 1.67 Å) and the higher degree of ionic character
thus resulting in more uniform and pinhole-free films, which between Br and Cl are responsible for the easier miscibility of
consequently improves the performance as well as the stability Br/Cl than I/Cl.161 To our knowledge, application of
of the devices. Besides, the longer diffusion length of electrons MAPbBr3‑xClx in PV devices has not been explored yet, but
in MAPbI3‑xClx than in MAPbI3 films157 has been credited devices based on triple-halide compositions (MAP-
with improving the performance of the cells. But, as the carrier bI3‑x‑yBrxCly) have been reported to show PCE > 16%.162
lifetime or diffusion length is dependent on the morphology 4.1.3. Both Cations and Anions Mixture. A number of
(grain size and grain boundaries) of polycrystalline films, it is “mixed cations and mixed halides” perovskites have been
not easy to separate the effect of Cl from effects of morphology synthesized and employed in PV devices. In fact, such
on the electronic properties. Hence, it seems that Cl− in the simultaneous mixing of cations and anions has led to further
precursor solution has an indirect but positive effect on the improvements in cell efficiency and stability. Based on results
overall performance of the cells. Unlike Cl, the presence and of enhanced VOC by Br inclusion and increased structural
effect of Br in the X-site in mixed Br/I perovskites have been stability in the FA-MA mix, (FA/MA)Pb(I/Br) perovskite
O DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 18. (a) XRD patterns and (b) corresponding UV−vis spectra (dashed lines) and photoluminescence (PL) spectra (solid lines) of perovskite
upon addition of Cs to the series Csx(MA0.17FA0.83)(1‑x)Pb(I0.83Br0.17)3, abbreviated as CsxM, where M stands for “mixed perovskite” and x = 0, 5,
10, and 15%. Cross-sectional scanning electron microscopy (SEM) images of (c) Cs0M- and (d) Cs5M-based perovskite solar cells. (e) Current−
voltage scans for the best-performing Cs5M device, showing PCE exceeding 21%. The inset shows the power output under maximum power point
tracking for 60 s, starting from forward bias and resulting in a stabilized power output of 21.1% (at 960 mV). (f) Aging for 250 h of high-
performance Cs5M and Cs0M devices in a nitrogen atmosphere held at RT under constant illumination and maximum power point tracking. The
maximum power point was updated every 60 s by measuring the current response to a small perturbation in potential. A J−V scan was taken
periodically to extract the device parameters. The device efficiency of Cs5M drops by about 20% (red curve, circles), and then it stays relatively
stable for at least 250 h. This is not the case for Cs0M (black curve, squares). Reproduced with permission from ref 167. Copyright 2016 Royal
Society of Chemistry, under Creative Commons Attribution-NonCommercial 3.0 Unported License.

Figure 19. (a) Cross-sectional scanning electron micrograph, (b) J−V curves, and (c) PCE histogram plot of triple-cation perovskite solar cells
fabricated in ambient air under controlled humidity (15−25%). Reproduced with permission from ref 38. Copyright 2018 Chemical Society of
Japan.

became a popular mixed perovskite for study. A good amount of FAPbI3, a slight amount of the photo-inactive yellow phase
of work has been done on this mixed perovskite to find the always remained in the resultant perovskite, which was
optimum composition for maximizing device performance and considered to be detrimental for long-term stability.
to understand more about the correlation between its In order to avoid this yellow phase completely, Saliba et al.
composition and optoelectronic properties. Incorporation of explored the inclusion of Cs+, which is significantly smaller
cations with smaller effective radius (MA+) into FAPbI3 can than MA+, as a third cation in mixed FA-MA and I-Br
adjust the Goldschmidt tolerance factor close to 1163 by perovskites.167 In their study, Cs prevented formation of the
contracting the lattice (Figure 17a) or relaxing the crystal yellow phase completely and the improved morphology of the
strain of FA-based perovskites (Figure 17b) to stabilize the
perovskite film through further grain growth. A triple-cation
cubic phase of the perovskite.164 As a result, simultaneous
mixing of FA-MA cations and Br-I anions improved the composition with 5% of Cs (i.e., Cs0.05(MA0.17FA0.83)0.95Pb-
performance and stability of the devices. Devices based on (I0.83Br0.17)3) produced excellent results, with a stabilized PCE
MA0.17FA0.83Pb(I0.83Br0.17)3 have been reported to work at PCE above 21%, which dropped to about 18% in a few hours and
> 20%, which has been accomplished either by combining it then remained stable for up to 250 h of continuous operation
with a newly designed HTM165 or by adding a slight excess of at maximum power point (Figure 18). Further, Cs (20%) in
PbI2 in the solution.166 Although MA, with a smaller effective the triple-cation mixed perovskites (Cs/FA/MA/Pb/I/Br)
size than FA, worked here as a crystallizer for the black phase demonstrated an incredibly long life (1000 h); the perform-
P DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 20. (a) Schematic illustration of the device architecture of perovskite solar cells, (b) maximum power point tracking of encapsulated PSCs
under constant 1 sun AM1.5G illumination measured in air, and (c−f) box chart presentation of photovoltaic parameters of PSCs. The current−
voltage (J−V) curves of 17 cells of each type were recorded at a scan rate of 0.1 V s−1. Reproduced with permission from ref 171. Copyright 2018
John Wiley and Sons.

Figure 21. Schematic illustration of (a) black single-cation α-FAPbI3, (b) black double- (CsFA, RbFA), triple- (CsMAFA), or quadruple-cation
(RbCsMAFA) compositions (X = I, Br), and (c) yellow non-perovskite δ-FAPbI3. The table shows the incorporation capacity of Rb+ and Cs+ into
the FAPbI3 lattice. Reproduced with permission from ref 125. Copyright 2017 American Chemical Society.

ance of cells made with Cs0.2FA0.8Pb2.84Br0.16 film almost did good-quality perovskite films with high reproducibility and
not change, even after being stored for 1000 h in the dark. lower sensitivity to the environment. Triple-cation perovskites,
The two identifiable attributes of these triple-cation-based either processed in a N2 environment or in ambient conditions
perovskite films are phase purity (no δ-phase formation) and (humidity −15−25%), result in high efficiency. Figure 19
uniform grains, which are apparently responsible for the shows one example of a high-efficiency (>21%) triple-cation
enhanced PV performance of the cells. Inclusion of Cs into the perovskite solar cell fabricated in ambient conditions in our
FA-MA-based perovskite completely prevents formation of the laboratory.169
δ-phase, improving the phase purity of the resultant perovskite The triple-cation recipe was then followed by a quadruple-
film. Recently, Zhou et al. tried to understand the underlying cation mixed perovskite that included Rb as the fourth
mechanism by which Cs helps in preventing formation of the cation.170 With this quadruple-cation mixed perovskite with
δ-phase.168 As found in their study, a triple-coordinated 5% of Rb, Saliba et al. achieved stabilized efficiencies up to
intermediate phase consisting of Pb2+, DMSO, and Cs+ retards 21.6% (average value: 20.2%) and open-circuit voltage of 1.24
crystallization of PbI2 in the precursor films, suppressing V for a band gap of 1.63 eV (0.39 V loss in potential), and the
formation of the yellow δ-phase. The greater structural stability polymer-coated cells maintained 95% of their initial perform-
of triple-cation mixed perovskites and the presence of the ance until 500 h of operation at the maximum power point at
intermediates/colloids (not well investigated yet) in the 85 °C.170 In comparison to Cs/FA/MA triple-cation devices,
precursor solution are probably the keys to formation of Rb-based quadruple-cation perovskite cells show slightly
Q DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 22. Effect of K inclusion on J−V hysteresis of different perovskites. J−V curves of perovskite solar cells employing different perovskite
materials FA0.85MA0.15PbI2.55Br0.45, FA0.85MA0.1Cs0.05PbI2.7Br0.3, MAPbI3, and FAPbI3 doped with (B, D, F, H) and without (A, C, E, G) 10 μmol of
KI, measured on reverse (filled circles) and forward (empty circles) scans at the scan rate of 130 mV/s (= voltage settling time of 200 ms) under
AM 1.5G 1 sun illumination (100 mW/cm2). Aperture mask area was 0.125 cm2. Reproduced with permission from ref 177. Copyright 2018
American Chemical Society.

Table 2. Summary of Key Effects of Cs, Rb, and K Inclusion into (FA/MA)Pb(I/Br)3 Mixed Perovskites
role and effects of Cs role and effects of Rb Role and effects of K
•gets incorporated into lattice at A-site •most likely not incorporated into lattice at •most likely not incorporated into lattice A-site
A-site
•helps in formation of cubic perovskite phase •segregation preferably near to the ETL •shift of XRD peaks/lattice expansion (contrary to what
(complete removal of yellow phase) and at grain boundaries expected if it was incorporated)
•enhanced stability •enhanced stability •reduction/elimination of hysteresis
•reduces trap states •enhancement in charge mobility •hinders formation of δ-FAPbI3
•resists phase segregation in mixed I/Br perovskites •no effect on trap landscape •grain boundary passivation
(improves photostability)

higher FF, reduced hysteresis, and greater photostability Like Rb+ (172 pm), K+ (164 pm), having a similar ionic size,
(Figure 20), which were essentially attributed to reduced is also considered not to occupy any crystal site in FAPbI3 or
recombination and less defects.171 mixed perovskites, but it enhances the cell performance by
Hu et al.171 found that Rb addition leads to increased charge passivating the defects sites in perovskite. Although K+ was
carrier mobility but has only a marginal effect on the trap assumed to occupy the A-site, as reported in earlier studies,
landscape of the perovskite layer. In contrast, Cs incorporation several independent and recent studies support its existence in
significantly reduces the number and the depth of trap states in interstitial sites172−174(evidenced from shift in XRD peaks and
the perovskite crystals, but it has barely any effect on the absorption edges), while a few other studies claim that it does
charge carrier mobility. The observed reduction in trap density not occupy any crystal site175 and is expelled out of the crystal,
is in excellent agreement with the enhancement in VOC and FF segregating at grain boundaries or on surface.176 Hence,
for Cs-containing devices compared to FA-MA (Figure 20). although its site of location in perovskite film and its active role
Upon combining Cs and Rb in quadruple-cation (Rb-Cs-FA- are not entirely clear, its positive impact on cell performance
MA) perovskite mixtures, the highest mobility and the lowest has been commonly observed in all the studies. K inclusion in a
trap density were observed, which subsequently resulted in variety of perovskite compositions has been found to eliminate
solar cells with the highest stabilized power output. This hysteresis in the J−V curves of the cells (Figure 22) and,
difference between the effects of Cs and Rb essentially arises therefore, has been proposed to be a universal method to
from incorporation or segregation of these cations in the eliminate hysteresis.177 Among all alkali metal ions (Li+, Na+,
crystal lattice. As evident from solid-state NMR studies,125 Cs K+, Rb+, Cs+), K+ works best for reducing/eliminating
can get incorporated into the crystal of (FA/MA)Pb(I/Br)3 up hysteresis. It is proposed that K+ prevents formation of Frenkel
to 15 mol%, while Rb does not occupy any crystal sites.125,126 defects, which are responsible for hysteresis, but direct
Instead, it segregates as rubidium-rich phases RbPbI3 mixed evidence supporting the active mechanism involved in
cesium−rubidium lead iodides, mixture of rubidium halides, elimination of hysteresis is yet to come.
various rubidium lead bromides, depending on the exact It has been also found that K+ incorporation hinders
composition (Figure 21). A ToF-SIMS depth profile also formation of photoinactive δ-FAPbI3 almost completely,
shows a preferable accumulation of Rb+ species at the TiO2 resulting in enhanced lifetime of charge carriers, reduced
interface, while Cs+ is very homogeneously distributed in the recombination, and shift of conduction band edge toward
film.171 Hence, it is believed that enrichment of Rb cations better energy alignment with SnO2 ETL.178 As a result,
near the electron transport layer interface is possibly related to K0.03(MA0.17FA0.83)0.97PbI2.5Br0.5 perovskite solar cell based on
a reduction of surface recombination in the vicinity of the tin oxide (SnO2) as ETL yields hysteresis-free PCE over 17%.
electron transporting layer, which in turn affects device Based on the reports so far, a summary of major roles and
hysteresis and power output stability. effects of Cs, Rb, and K in mixed FA-MA perovskite is given in
R DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 23. (a) UV−vis absorption and (b) steady-state PL emission spectra of FAMA mixed perovskite doped with Cs, Rb, K, Na, and
combinations of (Cs, Rb), (Cs, K), and (Cs, Rb, K). (c) Table listing corresponding band gap values. Reproduced with permission from ref 126.
Copyright 2018 American Chemical Society.

Figure 24. Schematic illustration summarizing the main difference between the perovskite materials containing two (MA/FA), three (Rb/MA/FA
and Cs/MA/FA), and four (Rb/Cs/MA/FA) cations, observed by HAXPES (hard X-ray photoelectron spectroscopy). Quantification of different
ions in bulk and on surface displays that unreacted FAI increases on surface with increasing addition of Cs or Rb in the precursor. Reproduced with
permission from ref 179. Copyright 2017 American Chemical Society.

Table 2. As the studies have conceded, Cs helps in formation all the studies which show improvement in performance by
of the black phase, preventing formation of the yellow phase slight modification of the surface of perovskite also support this
completely. Furthermore, it also imparts a significant effect of fact that compositional modification at the surface of
grain growth. As a result, thermal stability as well as moisture perovskite, which can come from different methods, can
stability of Cs-incorporated mixed perovskites has been have a strong effect on performance, especially on VOC.
witnessed to be remarkable. However, it is not known how It is expected that deeper understanding related to roles of
exactly a small amount of Cs prevents the reaction of the different cations or anions in the mixed perovskites will come
perovskites with water. It is unconvincing that this small through more studies in future, but one thing that has been
amount of Cs protects the perovskite from humidity just commonly noticed in most of these mixed perovskite studies is
because Cs is stable against moisture. Instead, it seems that the the enhanced structural/intrinsic stability, which is doubtlessly
enhanced structural stability improves the moisture resistance an important development.
of Cs/Rb-based mixed perovskites. Hence, structural stability 4.2. Mixed Dimensions
can be a major factor involved in moisture instability. In all Another recent strategy that has succeeded in improving
three cases, Cs, Rb, and K, the improvement in PCE is intrinsic stability or essentially the moisture stability of
basically due to higher VOC and better FF, which is a result of perovskites is by mixing the 2D structures with 3D structures.
traps-passivation. It has been found (Figure 23) that the It is well known that 2D perovskites are more stable to heat
optical band gap remains unaltered when (FA/MA)Pb(I/Br)3 and humidity181,182 but they lag behind the 3D perovskites in
is doped with either of Cs, Rb, K, or Na.126 This result terms of performance because of their narrow absorption band
complements the fact that improved performance in all three in addition to poor electron transport properties.183 However,
cases is related to defects passivation. However, the questions mixing a small amount of 2D perovskite to 3D perovskite
that remain to be answered are, “What kind of defects exist in structures have been found to work with higher efficiency and
the films, and how do these cations help in passivating them? improved long-term stability. Several 2D/3D mixed perovskite
Why do Na+ and Li+ not exhibit such traps-passivation effects? compositions and their cell efficiency and stability are listed in
Is it that the distributions of these ions are different and Table 3. For instance, incorporation of 0.8 mol% of
therefore they demonstrate different effects?” Not just ethylenediammonium iodide (EDAI) (which forms a 2D
distribution of these ions, it is rather very much possible that perovskite structure when mixed alone with PbI2) into 3D
these ions strongly impact distribution of other ions like Pb2+, MAPbI3 structure improves the PCE by reducing recombina-
I−, Br−, MA+, and FA+ in the films. As a matter fact, it has been tion184, and the MAPbI3-EDAI cell retains about 75% of its
found that inclusion of Cs and/or Rb in (FA/MA)Pb(I/Br)3 initial performance (18%) after 72 h of continuous operation
films increases the amount of unreacted FAI on the surface of under illumination while the regular MAPI3 loses 90% of its
the films (Figure 24), which is indeed responsible for an initial performance (17%) in just 15 h of operation (Figure
increase in VOC in the devices with Cs, or Rb doped FAMA 25). Similarly, a combination of properties of enhanced
perovskite.179 And, this understanding that slight excess of FA stability from 2D perovskite and excellent optoelectronic
on the surface improves the VOC is also consistent with the properties from 3D perovskite has been observed in a mixed-
results of increased VOC in the cases where slight excess of dimensionality and mixed-compositional (MDMC) lead iodide
organic cations were used in the precursor solution.180 In fact, perovskite based on [CF3CH2NH2]2(FA0.825MA0.15Cs0.025)n‑1-
S DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

186

187

184

185

187

188

189
ref

retains 80% of initial performance up to 1000 h (for non-encapsulated cells) or

stability measured at maximum power point up to 2 weeks; better stability than

retains 75% of initial performance after 72 h under AM 1.5G irradiation, 50%

better stability compared to 3D under 65% RH for up to 4 weeks

stability

better thermal stability than 3D perovskite

above 3000 h (for encapsulated cells)

RH, 50 °C device temperature

stability not measured

stability not checked

3D structure

Figure 25. Long-term evolution of normalized (a) PCE, (b) VOC, (c)
JSC, and (d) FF for the planar PSCs based on pristine MAPbI3 and
Table 3. Compositions of 2D/3D Mixed Perovskites, Corresponding Solar Cell Performance, and Stability

MA1−2xEDAxPbI3 (x = 0.008) tested under 1 sun irradiation, at 50 °C

and RH 50%. Absolute starting parameter values were PCE = 16.9%
and 17.6%, VOC = 1.03 and 1.04 V, JSC = 22.2 and 22.3 mA cm−2, FF
9.3

17.2

17.6

17.7

8.5
PCE
(%)

18

>18

= 0.74 and 0.76 for MAPbI3 and MA1−2xEDAxPbI3 (x = 0.008),

respectively. Reproduced with permission from ref 184. Copyright
2017 John Wiley and Sons.
ITO/NiOxperovskite/PCBM/C60/
FTO/SnO2/C60/perovskite/spiro-
FTO/c-TiO2/m-TiO2/perovskite/

FTO/c-TiO2/m-TiO2/perovskite/

ITO/PTAA/perovskite/PCBM/
FTO/c-TiO2/perovskite/spiro-

Pbn(I0.85Br0.15)3n+1 series (n = 1−∞, CF3CH2NH2-TFEA).185

FTO/TiO2/perovskite/spiro-
device structure

The hydrophobic nature of the trifluoroethylamine chain and

the multilayered structure of the mixed perovskite (Figure 26)
spiro-OMeTAD/Au

spiro-OMeTAD/Au

result in enhanced moisture resistance; MDMC PSCs (n = 30)

OMeTAD/Au

OMeTAD/Au
C60/BCP/Cu

without encapsulation maintain over 90% of the initial PCE

OMeTAD

under relative humidity of 65% at RT for up to approximately

28 days (Figure 26b). A more promising result has been
Ag

obtained with a 2D/3D mixed perovskite with about 3 mol%

of aminovaleric acid iodide (AVAI) in 3D MAPbI3 perovskite.
This mixed perovskite employed in a HTM-free carbon-based
[CF3CH2NH2]2(FA0.825MA0.15Cs0.025)n‑1Pbn(I0.85Br0.15)3n+1

device has exhibited stability more than 10 000 h under 1 Sun

illumination.

5. STABILITY OF PEROVSKITE SOLAR CELLS

5.1. Stability Issues with Perovskites
mixed perovskite

BA0.09(FA0.83Cs0.17)0.91Pb (I0.6Br0.4)3
(IC2H4NH3)2(CH3NH3)n‑1PbnI3n+1

At present, while PCE of above 20% in a lab-scale device is

being achieved by most of the leading laboratories, long-term
MA1‑2xEDAxPbI3 (x = 0.008)

stability and toxicity of Pb stand as two formidable obstacles

for commercialization of PSCs. For outdoor installation like Si
(BA)2(MA)n‑1PbnI3n+1

PV panels, PSCs must guarantee production of stable power at

PEA2MAn‑1PbnBr3n+1

operating conditions of real sun radiation, raised temperature

FAxPEA1‑xPbI3

due to heating, and under atmospheric moisture and oxygen

for a period of ∼25 years. Thus, these conditions are
considered as the requisite for commercialization. Long-term
stability/instability therefore include both intrinsic and
extrinsic stability issues with perovskite. Both structural/
intrinsic stability and stability under different external environ-
phenylethylamine
butylamine (BA)

butylamine (BA)
spacer cation

ethylenediamine

mental stresses such as heat, light, humidity, and oxygen are

CF3CH2NH2

critically important. Performance deterioration and/or material

IC2H4NH3

(EDA)

degradation issues under continuous operation of the cells

(PEA)

needs serious attention and should be solved in the coming

PEA

days.
T DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 26. (a) Schematic representation of the stacking structures of multidimension multicomposition (MDMC) perovskites based on
trifluoroethylamine ammonium cations. (b) Power conversion efficiency as a function of storage time (under a relative humidity level of 65% at
room conditions) for conventional 3D and MDMC (n = 30) perovskite solar cell, tested with an interval of 1 day. (c) XRD patterns of fresh and
aged perovskite film (4 weeks). Reproduced with permission from ref 185. Copyright 2017 John Wiley and Sons.

5.1.1. Structural/Intrinsic Stability. Structural stability of When τ is in the rage of 0.8 < τ < 1, ideal cubic perovskite
perovskite compounds can be primarily judged by the structures or distorted perovskite structures with tilted
Goldschmidt tolerance factor (τ), which is an empirical octahedra are favored. Specifically, 0.9 < τ < 1 favors a cubic
index widely used to predict formation of different crystal perovskite structure while for 0.8 < τ < 0. 9, a distorted
structures of ABX3. The value of τ (relation is given in section perovskite structure is formed. Values of τ < 0.8 and τ > 1
4.1) varies with the size of the ions in ABX3. Since ionic radii of diminish the possibility of formation of perovskite structures.
organic cations (A) cannot be determined accurately, a certain Therefore, it can be expected that τ close to the middle of the
amount of uncertainty lies in the calculated tolerance factors. range from 0.8 and 1, away from both the non-perovskite zones
Nevertheless, the values can be used to compare cations of (Figure 27), would form a stable perovskite.190 For FAPbI3, τ
reasonably different ionic radii. Figure 27 shows the calculated is close to 1, which is the near the upper boundary for
tolerance factors of APbI3 systems where A = Na, K, NH4, Rb, perovskite structure, and therefore, FAPbI3 is prone to
Cs, MA, FA, EA (ethylamine), and EDA (ethylenediamine). formation of a hexagonal δ-phase which is photoinactive.
CsPbI3 with τ ≈ 0.8, is at the edge of lower boundary for
perovskite structures, and it normally crystallizes into a δ-
phase. MAPbI3 with τ ≈ 0.9 is close to the middle of the
perovskite zone and forms a black photoactive perovskite
phase.
As discussed in section 4.1, compositional engineering of
perovskite; mixing of different cations and anions improves
structural stability of perovskites, essentially by adjusting the
value of τ close to middle of perovskite zone. Addition of Cs or
MA to FAPbI3 moves τ value down from 1 to stabilize the
cubic phase of FAPbI3. Although precise calculation of
resultant/effective tolerance factor (τ) for mixed perovskites
is not easy, a simple mixture rule can be applied to obtain an
approximate value. According to the mixture rule, for
(AxA′1‑x)B(XyX′1‑)3, rA(eff), and rX(eff) can be calculated by
using the following equations:
Figure 27. Calculated tolerance factors (τ) for different cations (A) in
APbI3 perovskite system. Commonly used cations like Cs, MA, and rA(eff) = xrA + (1 − x)rA′ , rX(eff) = yrX + (1 − y)rX′
FA give rise to a τ-value in the range of 0.8−1.0, indicating formation
of the cubic perovskite phase structure. Ionic size of A cations used in (1)
calculation are values referring to XII coordination, not VI The effective tolerance factor (τeff) is calculated by using the
coordination. Ethylammonium (EA) and ethylenediamine (EDA) values of rA(eff) and rx(eff) in eq 1. Applying this mixture rule,
cations are too big, giving rise to a tolerance factor >1.0 and thus fall
into the “upper forbidden zone” and cannot form perovskite alone.
τeff for a mixed perovskite, (MAPbBr3)0.15(FAPbI3)0.85 is
The group I alkali metal cations (Na, K, Rb) and NH4 have a τ-value calculated to be ∼0.98 while that for triple cation mixed
<0.8 and thus fall into the “lower forbidden zone”, not forming perovskite with 5% Cs (i.e., Cs0.05(MA0.17FA0.83)0.95Pb-
perovskite themselves. However, all cations of lower or upper (I0.83Br0.17)3) is ∼0.97. From such values of τeff, it can be
“forbidden zone” can be used as additives into those cations in “cubic predicted that more of Cs or MA in these mixed perovskite
perovskite formation zone” for phase stabilization. systems should result in even more stable perovskite structures.
U DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 28. (a) Proposed migration path of I− ions along the I−−I− edge of the PbI64− octahedron in the MAPbI3 crystal. (b) Illustration of the
migration paths for I− and Pb2+ ions in the Pb−I plane. (c) MA+ ions in the MA−I plane that was used for the calculation of activation energies.
Reproduced with permission from ref 192. Copyright 2015 Springer Nature. PTIR images of the distribution of MA+ before (d) and after electrical
poling for 100 (e) and 200 s (f). Reproduced with permission from ref 195. Copyright 2015 John Wiley and Sons. (g, h, i) Optical images of the
lateral MAPbI3 perovskite solar cell with a mobile PbI2 thread, which formed and migrated at 330 K along the applied electrical field direction.
Reproduced with permission from ref 193. Copyright 2015 John Wiley and Sons.

However, for such systems of mixed perovskites, formation of Br, I) decomposes to PbX2(s), MA(g), and HX(g). Energy of
secondary non-perovskite phases above certain concentration formation of these component products of degradation process
becomes another important factor that influences the stability. confirms that MAPbI3 and MAPbBr3 are thermodynamically
In other words, τeff close to 0.9 is not a sufficient condition for more stable than MAPbCl3.191 Similar study on mixed
predicting structural stability of mixed perovskites. Nonethe- perovskites with different cations and anions will be interesting
less, it provides guidance for selecting the ions or combination and can give a profound understanding about the structural
of ions.190 Cations with large differences in the ionic sizes stability of the perovskites.
between the additives (Na, K, Cs, Rb) and the matrix system Regardless of perovskite crystal structure (stable or
(FA/MA mix) are preferred for stabilizing the perovskite unstable), intrinsic ion migration persists to occur under the
structure as long as there is no non-perovskite secondary influence of electric field, either generated in the device during
structure is formed. In other words, phase purity also plays a its operation or biased externally in dark. In the recent years,
significant role in intrinsic long-term stability of the mixed with increasing direct or indirect evidence of ion migration and
perovskites. its association with anomalous J−V hysteresis, phase
Based on tolerance factor and phase purity, one would segregation, the possibilities that such ion migration can affect
believe MAPbI3 with τ ≈ 0.9 to be even more stable than long-term stability of PSCs have increased tremendously. In
mixed perovskites, which is in contrast to what is being any ionic solid, migration of ions must be mediated by defects
observed experimentally. Crystal structure difference between in the solid. The rate of migration of ions in such solids
MAPbI3 (tetragonal) and mixed perovskites (FA-MA-Cs) depends on the available interstitial space, Schottky defects
(usually cubic) might explain why MAPbI3 is less stable. As (cation or anion vacancies), ion jumping distance, and size and
cubic structures with higher symmetry than tetragonal charge of the ions; smaller ions and ions with smaller charge
structures are more stable, mixed perovskites with cubic tend to migrate faster than larger ions and ions with greater
structures show better stability than pure MAPbI3. Moreover, charge. It is believed that ions in perovskites essentially migrate
thermodynamic stability of different perovskite structures and through the cation and anion defects (i.e., MA+ and I−
its dependence on composition, which is lacking now, will vacancies). How easily an ion migrates in perovskite film can
provide better insights into structural/intrinsic stability of be assessed by its activation energy of migration. Theoretical
different perovskites. It has been found that MAPbX3 (X = Cl, calculation of activation energies of migration for different
V DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 29. (a) Schematic illustrating the stages of film formation and stress evolution in the film (the effect on molecular structure for antisolvent
processing of perovskite with solvent exchange during spin-coating with antisolvent drip). (b) Photographs of MAPbI3 on PET with externally
applied stresses ranging from −130 to 130 MPa after 24 h of damp air aging at 25 °C and 85% RH or dry heat aging at 85 °C and 25% RH, where
applied compressive stress enhanced film stability and applied tensile stress reduced film stability. Reproduced with permission from ref 199.
Copyright 2018 John Wiley and Sons.

Figure 30. (a, b) Photographs and (c, d) XRD pattern of (100) facet, (112) facet, and (001) facet of MAPbI3 single crystal before (a, c) and after
(b, d) exposure to water mist erosion for 60 s. Asterisks in (c) and (d) indicate peaks of PbI2. Reproduced with permission from ref 201. Copyright
2017 Royal Society of Chemistry.

constituting ions (i.e., I−, Pb2+, MA+, or FA+) in perovskites, as This conversely implies that channels other than grain
shown in Figure 28a−c, suggests iodide to be most easily boundaries also exist. (3) What are the effects of composition
migrating ion,192 which is consistent with experimental (mixed cation and anion) on ion migration? As mixed
studies.193,194 Electric-field-driven migration of PbI2 thread in perovskites with complex systems of cations and anions
MAPbI3 film under small electric field (Figure 28g−i) presents generally demonstrate more stable performance and longer
a direct evidence for macroscopic iodide migration.193 MA+ stability than pure MAPbI3 or FAPbI3, understanding the rate
ions also migrate in perovskite crystals, as evident from MA+ of ion migration in such mixed perovskites would certainly
ion distribution in MAPbI3 film, measured by photothermal help in finding good strategies to suppress the phenomenon. In
induced resonance (PTIR) microscopy.195 Increasing accumu- relation to the above, in a recent study, DFT calculation has
lation of MA+ near the cathode with time of biasing (the poling suggested increase in ion migration barrier energy by alkali
field is 1.6 V μm−1) confirms migration of MA+ ions. metals (Rb+, K+, Na+, and Li+) occupying interstitial sites of
Although a good amount of scientific understanding perovskite. This indicates that compositional engineering can
regarding ion migration in perovskites has come through a have a substantial effect on ion migration.
great number of studies in these years, there are still number of In addition to atomic/ionic level stability in structure,
open questions, which need further investigation. (1) As most characteristics of polycrystalline films can be intrinsic sources
of the studies have reported migration of ions under bias of instability too. For instance, thermal strain generated during
electric field, it is not known if ions can still migrate without the annealing process of perovskite films accelerates degrada-
field (stored in dark) and at RT. This may not represent the tion of the perovskite. Zhao et al. have found that such strain in
case of cell operation but it can be important for shelf life of the polycrystalline MAPbI3 film due to difference in thermal
the cells. (2) Although most of studies assumed ions in expansion of the substrate (ITO glass) and perovskite
perovskite migrate through the Schottky defects (vacancies), (MAPbI3) leads to faster degradation of the perovskite.198
there is no direct evidence of such defects. In other words, the Different amounts of strain created deliberately by bending the
actual channels of ion migration in polycrystalline films of MAPbI3 films on ITO glass substrate in different ways
perovskite are still not known. In polycrystalline films, grain displayed a remarkable difference in the speed of degradation.
boundaries are considered as preferred regions of ion The films with large strain (convex bending) degraded faster
migration than within the grains. Therefore, it is possible to while the film with least strain (concave bending) did not
minimize ion migration in perovskite films by reducing the degrade at all. Recently, Rolston et al. found that stability of
number of grain boundaries. However, reducing the grain MAPbI3 films coated on flexible substrate (PET) depended on
boundary volume might not eliminate the process of ion kind of applied stress on the film; films with applied tensile
migration because, as observed in several studies, halides also stress had significant visible degradation with PbI2 formation,
can migrate in single crystals of MAPbI3196 and MAPbBr3.197 whereas films with applied compressive stress resulted in intact
W DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 31. Diagram showing different strategies followed for improving moisture resistance of perovskite solar cells.

Figure 32. (a) VOC distributions of PSCs with and without PVP (i.e., FTO/c-TiO2/mp-TiO2/MAPbI3/with or without PVP/spiro-OMeTAD/Au),
showing increase in VOC from ∼900 to ∼980 mV by PVP coating on surface of perovskite. (b) Steady-state PL emission of MAPbI3 surface with
and without PVP. Higher emission intensity in case of PVP-modified MAPbI3 film implies passivation of surface traps. (c) PCE for devices with and
without PVP changing over days stored in 50% relative humidity at room temperature. (d) Photographs showing degradation (discoloration) of
standard PSC- and PVP-modified perovskite cells after dipping in water. PVP layer prevents moisture ingression into perovskite. Reproduced with
permission from ref 204. Copyright 2017 John Wiley and Sons.

perovskite after exposure to exposed to either damp air (25 °C, can be effective in reducing stress and improving the inherent
85% RH) or dry heat (85 °C, 25% RH) for 24 h (Figure moisture and thermal stability of perovskite films.
29).199 Further, Strelcov et al. discovered ferroelasticity (sponta-
Hence, after having improved structural stability by mixing neous phase transition from cubic to tetragonal under applied
cations and anions, strain-induced degradation should be also stress) in MAPbI3 perovskites. Such stress-induced phase
taken into account seriously for enhancing overall stability of transition, which generates intragrain strain, can also be an
important intrinsic source of instability.200 As grain boundaries
the perovskite devices. In fact, it will be interesting to study the
affect the long-term stability of organic−inorganic perovskite
relation between compositions of perovskite (mixed cations devices, and because the ferroelastic domain boundaries may
and anions) and thermal strain generated in the polycrystalline differ from regular grain boundaries, the latter becomes a new
films. As suggested by the study by Rolston et al.,199 selection variable to be considered for improving the stability. The other
of a low-temperature processing method or a polymeric significant source of instability can be the crystal orientation
substrate with higher coefficient of thermal expansion (as long because it has been found that moisture corrodes some crystal
as the device operation is below the annealing temperature) planes faster than the others. In case of MAPbI3 single crystals,
X DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

the (001) facet shows greater sensitivity to moisture and Hence, it is expected that combining surface passivation
degrades faster than the (100) and (112) facets (Figure 30).201 techniques to 2D/3D mixed perovskites can raise the moisture
5.1.2. External/Environmental Stability. Stability of the resistance further. However, protecting the perovskite film
cells to external conditions like moisture, oxygen (air), heat, from only moisture might not be enough because the material,
light, external bias, etc. is going to be important after having as being reported in literature, also shows quite high sensitivity
improved structural (intrinsic) stability of perovskites and the to vacuum, O2, N2, etc. Such influence of atmosphere (O2, N2,
HTMs. Sensitivity of perovskite to moisture has been a matter or vacuum) on optical and electronic properties of perovskites
of worry for its long-term stability. It is known that water might become rather a bigger concern than humidity. It has
molecules easily bind to perovskite by hydrogen bonding to been found that perovskite films show weak PL in a vacuum or
form hydrated compounds, which alter the properties of under N2 and a strong luminescence when aged in O2. For
perovskite locally. The losses caused by this hydrated instance, MAPbBr3 single crystal shows PL of intensity 2
compounds can be reversed, as shown for MAPbI3 in eq 2, orders of magnitude higher in air than in vacuum.211 As
but further ingression of water can cause irreversible evident from another study, O2 is reversibly adsorbed on the
degradation of perovskite to PbI2 and other components (eq surface of MAPbI3 (Cl) and thereby, passivates the deep
3).202 surface trap states, resulting in enhanced luminescence.212 It is
also suggested that a photochemical reaction between the
4CH3NH3PbI3 + 4H 2O ↔ 4[CH3NH3PbI3·H 2O] photogenerated carriers and oxygen-related species deactivates
the trap states in MAPbI3 perovskite, resulting in enhanced
↔ (CH3NH3)4 PbI6 ·2H 2O + 3PbI 2 + 2H 2O (2) PL.212 By investigating simultaneous effect of light, moisture,
and oxygen on perovskite (MAPbI3), Stranks et al. found
(CH3NH3)4 PbI6 ·2H 2O → 4CH3NH3I + PbI 2 + 2H 2O involvement of superoxide ion (O2•−) in defects passivation.
(3) According to them, O2̅ ions, formed in the presence of light,
Therefore, a proper encapsulation of the PSCs becomes a remove the shallow trap states at the surface of MAPbI3 and
prerequisite for prevention of degradation caused by moisture. thereby enhance the internal luminescence quantum efficiency
Although the cost and effort that will be required to have a (from 1% to 89%).93 Although several independent studies
rugged sealing are a little discouraging for industries at the consistently report about such improved luminescence of
present moment, it seems that, by choosing a suitable sealing perovskite in O2 environment, one recent report from Kong et
technology established in other industrial applications, perov- al. shows a contradictory result. According to their findings, as
skite can be protected well from the external humidity without detected by Raman spectroscopy, O2 can form Pb−O bonds
much addition to the overall cost. Nonetheless, a lot of efforts on the surface of perovskite film, and such oxidation of
have been made to improve the moisture resistance of the perovskite surface results in weak luminescence.213 Although
materials constituting the cell structure, either by introducing the direct impact of such changes on the surface of perovskite
hydrophobic layers like polymer/carbon composites in place of on the cell performance is not yet known, it is reasonable to
the widely used spiro-OMeTAD,203 by incorporating non- expect an effect of significance because the top surface of
hygroscopic interlayers between perovskite and HTM,204,205 perovskite forms the most important working interface with
by passivating (modifying) the perovskite surface by small the HTM in actual device. The surface trap states of
molecules,206,207 or by incorporating 2D perovskite having perovskite, if not healed by the HTM coated on its top, can
hydrophobic organic group into 3D perovskites.208 These be detrimental to carrier transfer at the interface. However, if
strategies are summarized in Figure 31. we consider the performance of cells either made on low
The results from the recent strategy of 2D/3D mixing are humidity ambient conditions or glovebox (N2 atmosphere), it
very promising (discussed in section 4.2). The method serves seems that such change on the surface may not have a
two purposes simultaneously: passivation of the traps and significant effect but the bigger concern is oxygen-induced
protection against moisture ingression. Large organic cation degradation of perovskite, which even takes place in dry
that has more hydrophobic nature than MA and FA helps in atmosphere (no moisture). Several studies have disclosed that
improving the moisture resistance. This moisture resistance perovskite films exposed simultaneously to oxygen and light
can also be highly enhanced by modifying perovskite surface by degrades faster than even the moisture-exposure case, and the
hydrophobic polymer. For instance, an ultra-thin layer (<10 suggested mechanism follows generation of superoxide (O2•−)
nm) of polyvinylpyridine (PVP) coated on perovskite surface by electron transfer from the photoexcited perovskite to
(MAPbI3) and in contact with spiro-OMeTAD improves water molecular oxygen on the surface. This superoxide ion then
resistance of MAPbI3 when dipped in water for up 100 s breaks down the perovskite by deprotonating the organic
(Figure 32d). The PVP-modified device also shows enhanced cation (MA), as shown in the equations below.214
open-circuit voltage and efficiency.204 Lewis base side chain of light, no moisture
PVP possibly coordinates the under-coordianted Pb2+ on the MAPbI3 ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ MAPbI3 *
surface of perovskite and thereby reduces the surface traps MAPbI3 *
density as observed by PL enhancement (Figure 32b). A O2 ⎯⎯⎯⎯⎯⎯⎯⎯→ O2•−
similar result has been also reported by Tress et al, who found
large effect of suppressing nonradiative recombination by using MAPbI3 + O2•− → MeNH 2 + PbI 2 + 1 2 I 2 + H 2O
PVP as a thin interlayer and succeeded in elevating VOC of the
perovskite cell to 1.20 V.209 Recently, continuous chemical Thus, incorporation of less acidic organic cation like FA or
functionalization of grain boundaries of MAPbI3 films by use of inorganic cation Cs shows longer durability of the films. In
triblock copolymers that exhibit both hydrophilicity and addition, oxygen tolerance of perovskites strongly depends on
hydrophobicity has been also found to boost the PCE as charge carrier density; cells aged at open-circuit conditions (no
well as stability of the MAPbI3 cells.210 charge collection) show faster degradation than the cells at
Y DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 33. (a) Schematic illustration of phase segregation (formation of I-rich and Br-rich regions) in mixed halide perovskite, MAPb(BrxI1−x)3
under light. Reproduced with permission from ref 222. Copyright 2018 American Chemical Society. (b) The (200) XRD peak of MAPb(BrxI1‑x)3
(x = 0.6) film before (black) and after (red) white-light soaking for 5 min at 50 mW cm2. XRD patterns of an x = 0.2 film (dashed green) and an x
= 0.7 film (dashed brown) are included for comparison. Reproduced with permission from ref 228. Copyright 2016 American Chemical Society.
(c) Photoluminescence (PL) spectra of an x = 0.4 thin film over 45 s in 5 s increments under 457 nm, 15 mW cm−2 light at 300 K. Inset:
temperature dependence of initial PL growth rate. Reproduced with permission from ref 228. Copyright 2016 American Chemical Society. (d) A
series of cathodoluminescence (CL) images of MAPb(I0.1Br0.9)3 taken at 30s intervals of light soaking (with a 405 nm LED), the scale bars are 2
μm. Reproduced with permission from ref 58. Copyright 2017 American Chemical Society. (e) Schematic of the proposed mechanism for photo-
induced trap formation through halide segregation. Reproduced with permission from ref 160. Copyright 2015 Royal Society of Chemistry, under
Creative Commons Attribution 3.0 Unported license.

short circuit.215 Hence, integration of efficient electron bathocuproine (BCP) as an interlayer between FTO and
collecting layer in the structure improves the stability by perovskite (i.e., FTO/BCP/triple cation perovskite/spiro-
reducing the yield of superoxide ions.216 As prevention of OMeTAD/Au) has been observed to demonstrate superior
oxygen insertion into the device may not be easy by stability under illumination (no UV filter) than TiO2-based
encapsulation, it is required to develop perovskite composi- devices.225 Besides, coating a UV-absorber material like 2-(2H-
tions that will be more resistant to oxygen-induced benzotriazol-2-yl)-4,6-bis(1-methyl-1-phenylethyl)phenol (UV
degradation. We believe that right composition and right 234)219 (device structure: UV-234/FTO/c-TiO2−KH370/
interfacial contacts can improve the stability of PSCs to CH3NH3PbI3/spiro-OMeTAD/Au) and YVO4:Eu3+ phos-
external conditions like moisture and O2. phor226 (device structure: YVO4:Eu3+ NPs/FTO/TiO2/
Instability against moisture and O2 can be partially/ CH3NH3PbI3/spiro-OMeTAD/Au) on the front side of
completely solved by external protection but the effect of substrate has also been effective in improving the UV
light (photo-instability) on perovskite, which need to be irradiation resistance of PSCs. Recently, a study on spectral-
approached intrinsically, is rather becoming a formidable issue. dependent UV-light degradation of perovskite (MAPbI3) solar
It is a matter of great worry that almost all of those chemical, cells with SnO2, compact−TiO2, electron-beam−TiO2, and
physical, or electronic changes/processes in the perovskites, nanoparticle−TiO2 ETLs has shown that the 311 nm UV
such as ion migration, defects generation, phase segregation, radiations heavily deteriorate the JSC of all the cells regardless
lattice disordering, etc., which are believed to be detrimental to of SnO2 or TiO2 ETLs whereas the 370 nm UV radiation
the or long-term stability of perovskites, are unfortunately causes lesser or no performance deterioration at all.227
found to be influenced or enhanced by light, and light has to Recently, a number of studies have shown direct evidence of
inevitably shine on the cells for their operation. Photocatalytic photo-induced phase segregation or ion redistribution in mixed
degradation by TiO2,217−220 photo-induced ion migration,197 halide (I and Br) perovskites. A reversible photo-induced
photo-induced trap state generation,221 photo-induced phase phase separation results in formation of iodide-rich and
segregation,58 photo-induced cation or halide redistribu- bromide-rich regions160 in mixed-halide perovskites (Figure
tion,222,223 large photo-induced dielectric constant, etc. have 33a). Changes observed in XRD peak (Figure 33b), PL spectra
been found to be the reasons for performance deterioration (Figure 33c), and cathodoluminescence images (Figure 33d)
under constant illumination. UV light-induced photocatalytic of MAPb(BrxI1‑x)3 films under continuous light soaking are
degradation by the most widely used ETL, TiO2 can be some of the evidence that support phase segregation in mixed
prevented either by complete replacement of TiO2 with an halide perovskites. Such I-rich and Br-rich phase separation
alternative less photoactive ETL, or by pacifying the surface of leads to generation of low band gap trap states (Figure 33e),160
TiO2, or by preventing UV light to reach TiO2 by coating UV- which causes the performance degradation of mixed halide
absorbing layers on top. For example, Al2O3217 or SnO2224 PSCs under illumination. While one of the proposed
mesoporous layer shows better stability than TiO2 against UV mechanisms for such phase segregation describes that
light, and surface modification of TiO2 with silane coupling electron−phonon coupling in the photoexcited state of
agents (KH570)219 has been found to enhance UV light mixed perovskites deforms the surrounding lattice,58 which
stability of the perovskite cells. Even, ETL-free devices using possibly promotes segregation of the ions, the other proposed
Z DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 34. Change in PCE (backward scan) of planar MAPbI3 solar cells when the devices were heated at 0, 60, 80, 100, and 120 °C for 1 h during
at different steps of fabrication: (a) complete cells with the Au film: FTO/c-TiO2/MAPbI3/spiro-OMeTAD/Au), (b) before spiro-OMeTAD and
Au deposition (FTO/c-TiO2/MAPbI3), (c) before Au film deposition (incomplete cells without the Au film: FTO/c-TiO2/MAPbI3/spiro-
OMeTAD). Cross-sectional scanning electron micrographs of devices (d) before and (e) after heating at 100 °C, showing severe deformation of
spiro-OMeTAD. Adapted from refs 235 and 239. Chemical depth profiles of ions (counts versus sputtering time for different ions) in the Au/spiro-
OMeTAD/perovskite/TiO2/FTO architecture, measured by time-of-flight secondary-ion mass spectrometry (ToF-SIMS) in (f) fresh cells and (g)
degraded devices with over 60% efficiency loss. (h) Schematic of the electric field distribution in perovskite solar cells and the corresponding
direction of ion migration. Reproduced with permission from refs 235 (Copyright 2018 Royal Society of Chemistry), 237 (Copyright 2017
American Chemical Society), and 239 (Copyright 2017 American Chemical Society).

origin is halogen defects.228 Based on the observation of less 5.2. Stability Issues with Hole Transport Materials and
severe photo-induced phase segregation in mixed halide Contacts
perovskites including Cs or FA as partial substitution to MA, As discussed in previous sections, efforts made in the direction
which eventually improves the crystallinity and morphology of improving structural (intrinsic) stability of perovskites have
through large grains, it is believed that defects like halogen brought excellent results. Not to mention, such structural
vacancies are essentially the main promoters of such phase stability of the perovskite is the first and foremost step toward
separation. As ion mobility in these materials is known to be solving the instability issues with the PSCs. But, not to be
more facile at the grain boundaries,229 reducing volume of ignored are the other layers, such as the hole transport and
grain boundaries (growing large grains) by compositional electron transport layers, even the metal (Au or Ag) contacts,
tuning with Cs or FA in mixed cation perovskites results in which also cause performance degradation in the long term.
improved photo-stability by slowing down the ion migration. Especially, the HTM, which forms the main working junction
Further, a more controlled study on effect of composition of with the perovskite, can be very critical for long-term stability
Cs and Br in the mixed perovskites reveals that compositions because perovskite/HTM junction is the beating heart of the
with low Br and high Cs demonstrate superior photostability whole device. Any chemical or physical change at this junction
than compositions with high Br and low Cs.230 over time or under different environmental stresses (light,
Hence, it seems that, for a given composition, reduction of temperature, humidity, oxygen, etc.) will definitely have a
halogen defects in the perovskite through fewer grain strong influence on long-term performance. Indeed, quite a
boundaries can have a significant effect on the photo-stability good amount of work has shown how different HTMs affect
of perovskites. As all these results were essentially obtained the long-term stability of the cells. A variety of HTMs have
with the perovskite films, where morphology was also changing been employed in PSCs. The most popular one that leads in
with composition, it will be more interesting to verify such terms of efficiency is an organic small molecule, spiro-
phase segregation behavior in single crystals of different OMeTAD, while conducting polymers like P3HT, PTAA,
composition. PEDOT:PSS, etc. and inorganic p-type materials like NiOx,
AA DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 35. (a) Schematic illustrating cell structure with SWCNT interlayer between perovskite and spiro-OMeTAD. A solution-processed SnO2
layer is the ETL, perovskite absorber is FA0.83MA0.17Pb(I0.87Br0.17)3, and p-type contact is formed by a dense multilayer of polymer-wrapped
SWNTs and a sequentially deposited layer of undoped spiro-OMeTAD. (b) Thermal stress test of devices with doped spiro-OMeTAD (triangles)
and the SWNT interlayer without dopant (squares). The devices were exposed to 85 °C in the dark at a RH of 45%. Each data point represents the
average SPO (tx) of 20 devices normalized by their initial SPO (t0) prior to heat exposure. Reproduced with permission from ref 238. Copyright
2017 American Chemical Society.

CuOx, CuPc, etc. have also shown promising results. perovskite solar cell in comparison to MA-lean (PbI2-rich),
Unfortunately, the spiro-OMeTAD, which shows higher MA-free (FA/Cs),235 and FA-treated MAPbI3 PSCs, as
efficiency, suffers most from degradation. As hole conductivity observed in our laboratory, also corroborates that diffusion
of pure spiro-OMeTAD is not adequate, LiTFSI (lithium of MA+ cation into spiro-OMeTAD is one of causes of
bis(trifluoromethanesulfonyl)imide) is added to it as a p- performance degradation. It is, therefore, believed that
dopant, which promotes oxidation of spiro-OMeTAD in the prevention of ion diffusion between the layers can be one of
presence of oxygen to enhance the hole mobility.231 It has the strategies to improve thermal stability of PSCs. Indeed, as
been found that LiTFSI, owing to its hygroscopic nature, shown in Figure 35, it has been found that introducing a single
allows easy ingress of moisture into the film, which eventually wall carbon nanotube (SWCNT) interlayer between perovskite
degrades the perovskite film (turning yellow). But, the (FA0.83MA0.17Pb(I0.83Br0.17)3) and spiro-OMeTAD (dopant
worrisome concern is that performance of the cells decreases free) improves the thermal stability of the cells.238
in many cases where the perovskite even remains unchanged. It Besides, it has been observed that spiro-OMeTAD with
has been found that performance of cells using spiro- LiTFSI and 4-tert-butylpyridine (TBP) added to it undergoes
OMeTAD and LiTFSI as additive decreases due to either severe morphological deformation, generating large voids in
crystallization232 or photo-oxidation233 of spiro-OMeTAD, or the layer235,239 at higher temperatures like 100 and 120 °C.
because of diffusion of Au into the spiro-OMeTAD film.234 Although these voids seem not to affect the cell efficiency
Moreover, one of our recent studies235 indicated that some much, they can be detrimental to long-term stability of the
physical or chemical exchange at the MAPbI3/spiro-OMeTAD cells. Nonetheless, it is certain that the additives (LiTFSI and
interface at elevated temperatures can be the main reason for TBP) used in the spiro-OMeTAD are responsible for
performance deterioration in MAPbI3 solar cells while slight degradation linked to spiro-OMeTAD, not just through their
degradation of perovskite may have almost no effect on such physical distribution or diffusion between the layers but also
performance degradation. As we found in our study, perform- via some chemical reactions. For instance, recently, Kaspar-
ance of the cells decreased only when the perovskite film was avicius et al. have found that the oxidized spiro-OMeTAD
heated along with the spiro-OMeTAD layer (Figure 34a−c) on (spiro-OMeTAD+) is reduced by TBP while a part of spiro-
its top (i.e., the complete cell) whereas the performance OMeTAD undergoes pyridination on its reaction with TBP.240
remained same for all the cases where perovskite films were In addition, based on our earlier findings239 and some
heated without the spiro-OMeTAD. In both the cases, preliminary results of ongoing work (cannot be disclosed
perovskite (MAPbI3) degraded to same extent but perform- now), we doubt that TBP also reacts with LiTFSI.
ance deteriorated only when spiro-OMeTAD was heated along To summarize, spiro-OMeTAD HTM has many stability
with perovskite film, suggesting that some chemical and/or issues such as (i) Li (in dopant) diffusion into perovskite, (ii)
physical change is happening at the interface of perovskite and Au diffusion into spiro-OMeTAD, (iii) reaction of iodide with
spiro-OMeTAD. In fact, some recent studies support this oxidized spiro-OMeTAD, (iii) migration of MA+ into spiro-
conjecture. For example, Carrilo et al. reported that spiro- OMeTAD, (iv) physical deformation at elevated temperature,
OMeTAD+ can react with migrating I− to reduce hole (v) crystallization of spiro-OMeTAD, (vi) chemical reaction
conductivity of the HTM, which results in progressive between the additives. Furthermore, spiro-OMeTAD is still
degradation in performance.236 Zhao et al. also found that commercially a highly expensive material. Therefore, although
MA+ ions from MAPbI3 can diffuse into spiro-OMeTAD the majority of high PCEs have been achieved with spiro-
(Figure 34f,g) and, thereby, cause performance degradation.237 OMeTAD until now, it needs to be replaced with an
Greater thermal degradation in case of MA-rich MAPbI3 alternative HTM with equivalent performance but with longer
AB DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 36. (a) Stability of perovskite solar cells with NiOx and PEDOT:PSS HTMs (inverted architecture; ITO/NiOx or PEDOT:PSS/
perovskite/PCBM/AZO/Ag and encapsulated), measured at 30 °C (∼50% RH) under ambient and dark condition and under MPPT condition
(the devices were kept under MPPT condition between the periodical J−V measurements). Reproduced with permission from ref 241. Copyright
2017 American Chemical Society. (b) Thermal stability of PSCs with spiro-OMeTAD (blue) and CuSCN (red) HTMs. The devices were heated
at 125 °C in the dark and in air with 40% relative humidity for different durations. Reproduced with permission from ref 249. Copyright 2016 John
Wiley and Sons.

Figure 37. (a) Schematic illustration, (b) HR-SEM cross-sectional image, and (c) J−V curve (best performing device) of HTM-free CH3NH3PbI3/
TiO2 heterojunction solar cell. Reproduced with permission from ref 253. Copyright 2013 Royal Society of Chemistry. (d) Schematic illustration,
(e) SEM cross-sectional micrograph, and (f) and J−V curves of HTM-free carbon electrode-based MAPbI3 solar cells. Reproduced with permission
from ref 259. Copyright 2016 John Wiley and Sons.

lifetime. Alternative organic HTMs like P3HT, PTAA, demonstrating better stability in comparison to PEDOT:PSS
PEDOT:PSS, etc. work with comparable high efficiency, but and spiro-OMeTAD respectively are given in Figure 36.
most of them also need diffusible p-dopant and/or additives Considering the importance of developing solar cells of all
similar to spiro-OMeTAD. Therefore, as alternative candi- inorganic structure, which can be more robust and durable
dates, a number of inorganic HTMs, such as NiOx,241,242 compared to organic−inorganic hybrid solar cells, use of
CuOx, 243,244 Cu 2 O, 245 CuI, 246,247 Cu(thiourea)I, 248 inorganic HTMs, especially those of metal oxides, looks
CuSCN,249,250 CuPc,251 etc., have been used as p-type promising for industrial applications of PSCs. Investment of
semiconductors. Promisingly, many of them have demon- more effort and energy in development of inorganic HTMs is
strated comparable efficiency and longer life than the spiro- expected to help in improving overall long-term stability of
OMeTAD. One example from each of NiOx and CuSCN PSCs further.
AC DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 38. (a) Schematic illustration of a doctor-bladed perovskite film and the chemical structure of F4TCNQ dopant. (b) Cross-sectional SEM
image of the MAPbI3 film deposited on ITO glass via bladed coating at 150 °C, showing the film thickness of around 500 nm. (c) Low- and (d)
high-magnification SEM images of bladed and doped MAPbI3 film prepared with methylammonium hypophosphite (MHP) (0.225 wt%) and
MACl (0.5 wt%) as additive, followed by co-solvents annealing treatment. (e) Schematic illustration of the HTL-free device configuration. (f) J−V
characteristics, (g) Steady-state current and stabilized PCE measured at a maximum power point (0.93 V) of the HTM-free MAPbI3:F4TCNQ
device. Reproduced with permission from ref 261. Copyright 2018 Springer Nature.

While, on one hand, inorganic HTMs are becoming popular enhancing the electron extraction efficiency of ETL258 implies
as potential alternatives to spiro-OMeTAD, HTM-free and/or that property optimization of the layers in combination with
carbon-based PSCs, on the other hand, are emerging as an each other is required to exploit the actual potential of the
attractive option, especially for industries. Even though the cells. Since almost all the research on HTM-free reported so far
HTM-free PSCs are lagging behind that of the cells with HTM have dealt with MAPbI3 perovskites, there is no information
in terms of PCE, which can be largely attributed to inadequate about effect of compositions of perovskite with varying carrier
understanding and imperfect device structures, they are type and density on performance of HTM-free PSCs. It is
interesting and important because they can be more cost- expected that work function matching through compositional
effective and processed by more simple and scalable fabrication variations of perovskite can also be significantly important for
methods, and more importantly, they can avoid the overall performance. In fact, this has been endorsed by findings
degradation issues related to HTM and perovskite/HTM of a recent study,261 which presented a remarkable achieve-
interface. ment of 20% PCE with a HTM-free doctor-bladed perovskite
With the ambipolar charge transport properties of halide solar cell (Figure 38). As found in the study, doping of
perovskite materials exploited, several hole extraction electro- MAPbI3 with a p-type molecular dopant, 2,3,5,6-tetrafluoro-
des like Au,252,253 Ni,254 and carbon255−259 have been explored 7,7,8,8-tetra-cyanoquinodimethane (F4TCNQ) modifies the
for use in HTM-free PSCs. Figure 37 shows examples of Au- ITO/MAPbI3 interface (ITO/perovskite-F4TCNQ/ETL/Cu)
and carbon-electrode-based HTM-free device structures with and actuates a favorable band bending, which facilitates the
their performances. A variety of carbon materials, carbon black, hole extraction. This result signifies that characteristics of
graphite, CNTs, etc. have been used, and a continuous perovskite that change both of its interfaces with ETL, and the
progress in performance has been reported in past few years. hole extracting electrode can play a dramatic role in overall
The commonly used carbon electrode materials (graphite and performance of the HTM free PSCs.
carbon black) suffer from recombination loss due to their
ineffective hole selectivity characteristics, which limit their hole 6. ALL-INORGANIC PEROVSKITES
extraction efficiency. However, graphene260 with a better Although organic−inorganic hybrid perovskites like MAPbI3,
matching with the Fermi level of MAPbI3 demonstrate FAPbI3, and the mixed-cation-based perovskites (FA/MA,
enhanced hole extraction and device performance. Therefore, MA/Cs, FA/Cs, FA/MA/Cs, etc.) are leading in terms of
it is anticipated that doped/modified carbon materials with efficiency, they have been facing a huge challenge of long-term
different physical and chemical properties, particularly with stability. It is the organic cation part (MA or FA), which is
better matched Fermi levels, would increase the device believed to be responsible for poor thermal and environmental
performance. Improved performance of HTM-free PSCs by stability of these materials. For instance, MAPbI3 degrades
AD DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Table 4. List of Studies Following Different Methods To Stabilize α-CsPbI3 and Corresponding Cell/Film Performance and
Stability
any special PCE
perovskite compositions device structure conditions (%) stability ref
CsPbI3:HI FTO/c-TiO2/CsPbI3(HI)/ inert 2.9 hours (films, not on device) 262
spiro-OMeTAD/Au atmosphere
CsPbI3 (HI and IPA treatment) FTO/c-TiO2/CsPbI3(HI, air 4.13 72 h (films, not on device) 280
IPA)/spiro-OMeTAD/Ag
CsPbI3 (QD) FTO/c-TiO2/CsPbI3(QD)/ air 10.77 months (films, not on the device) 267
spiro-OMeTAD/MoOx/Al
CsPbI3:HI ITO/PEDOT:PSS/ N2-filled 4.88 26% loss in PCE after 3 h in air (non-encapsulated cells) 281
CsPbI3(HI)/PCBM/BCP/ glovebox
LiF/Au
CsPbI3 (Bi) FTO/c-TiO2/CsPb0.96Bi0.04I3/ Ar-filled 13.21 maintained 68% of its initial PCE value for 168 h in air; non- 270
CuI/Au glovebox encapsulated cell
CsPbI3 (EDDI) FTO/c-TiO2/CsPbI3(0.025 air 11.86 retains >83% of initial PCE for 35 days in dry conditions; non- 266
EDDI)/spiro-OMeTAD/Ag encapsulated cells
CsPbI3 (QD) FTO/c-TiO2/CsPbI3(QD)/ air 13.4 not studied 268
spiro-OMeTAD/MoOx/Al
CsPbI0.98Cl0.02 (zwitterion) ITO/PTAA/ air 11.4 maintains ∼85% of its original efficiency after storage in air for over 282
CsPbI0.98Cl0.02:SB/PCBM/ 30 days
C60/BCP/Ag
Ca-doped CsPbI3 FTO/c-TiO2/mp-TiO2/ N2-filled 13.5 2 months; encapsulated cells stored under 85% humidity 272
CsPb1‑xCaxI3/ P3HT/Au glovebox
Sb-doped CsPbI3 FTO/c-TiO2/m-TiO2/m- air 5.3 retains 93% of initial PCE after 37 days of storage in air 271
Al2O3/CsPb0.96Sb0.04I3/
carbon/Au
CsPbI3(solvent engineering) ITO/SnO2/CsPbI3/spiro- N2-filled 15.7 500 h photostability (measured in glovebox) 263
OMeTAD/Au glovebox
Eu-doped CsPbI3 FTO/c-TiO2/CsPbI3:xEu/ air 6.8 films stable up to 30 days; additives in spiro-OMeTAD causes partial 273
spiro-OMeTAD/Au degradation
phenyltrimethylammonium FTO/c-TiO2/CsPbI3−PTABr/ N2-filled 17.06 photostability of PTABr-CsPbI3 PSC under continuous white light 278
bromide (PTABr) post- spiro-OMeTAD/Ag glovebox LED illumination (100 mW·cm−2) in a N2 glovebox for 500 h
treatment

easily at temperature as low as 120 °C, even at 80 °C if heated

for long time.137 Therefore, replacement of organic cation with
an inorganic cation is considered as a good strategy to improve
stability. One inorganic cation that has been found suitable and
successful in forming perovskite structures is cesium (Cs+),
although theoretical studies also predict some perovskite
compounds of Rb. Cesium-based perovskite structures like
CsPbX3 and CsSnX3 (X = Cl, Br, I) are indeed the first halide
perovskites that were studied by Wells et al. in 1983. With their
band gap controlled by the halide ions (CsPbI3, 1.77 eV;
CsPbBr3, 2.38 eV), CsPbX3 perovskites have been considered
as a model compound among all all-inorganic perovskites.
Despite its superior stability against heat, employment of Figure 39. Illustration of crystal structures of both yellow and black
CsPbI3 in PSCs has been difficult for several reasons, and, phases of CsPbI3. Reprinted with permission from ref 262. Copyright
2015 Royal Society of Chemistry.
therefore, the progress in PCE of these cells has been slow.
Nevertheless, recent reports have demonstrated some sig-
nificant leaps in PCE of CsPbI3 PSCs. PCE of CsPbI3 PSCs One of the ways to stabilize α-CsPbI3 is to increase the
has increased from 2.9% achieved in the very first attempt tolerance factor, which can be accomplished by partial
made by Snaith et al.262 in 2015 to 15.7% in a recent work substitution of either Cs+ (r = 167 pm) with bigger cations
undertaken by Pengyang et al.263 Table 4 provides a list of like MA+ (r = 217 pm) and FA+ (r = 253 pm) in the A-site or
some key studies reported until now on all inorganic CsPbI3 I− ions (r = 220 pm) with smaller anions like Br− (r = 196 pm)
solar cells. and Cl− (r = 181 pm) in the X-site. The other popular strategy
to stabilize a crystal structure with more symmetry is by
6.1. CsPbI3: Stabilization of Black Phase
reducing the crystal dimensions (size-controlled stabilization).
The imposing challenge with CsPbI3 is stabilization of its black Hence, suitable foreign dopants or even additives in the
photoactive phase (α-CsPbI3) at RT because CsPbI3 preferably precursors of CsPbI3 that can restrict the dimensions of
crystallizes in a yellow phase (δ-CsPbI3) at RT while the black crystals to few nanometers (nanocrystals) can effectively
phase is only stable at temperature above 310 °C262 (Figure stabilize the cubic CsPbI3. Successful implementation of such
39). The tolerance factor calculated for CsPbI3 is 0.8, which ideas in recent studies has produced encouraging results, and
favors formation of an orthorhombic yellow phase over cubic black phase (α-CsPbI3) has been stabilized at RT and in
black phase at RT. The size of Cs cation is too small to sustain ambient conditions. In addition to cell performance, stability of
the PbI6 polyhedra in cubic α-CsPbI3 at RT, and therefore, it the formed black phase (α-CsPbI3) has been enhanced but it is
readily degrades to the orthorhombic δ-CsPbI3. not so appreciable at present in comparison to organic lead
AE DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 40. (a) Photograph of α-CsPbI3, δ-CsPbI3, and Bi-incorporated CsPb1‑xBixI3 (0 ≤ x < 0.1) perovskite precursor solutions and
corresponding films. (b) Crystal structures and photographs of δ-CsPbI3 and α-CsPbI3. (c) Current density−voltage characteristics of the best cells
based on CsPb0.96Bi0.04I3 and inset shows the schematic illustration of the fully inorganic perovskite solar cells. Reproduced with permission from
ref 270. Copyright 2017 American Chemical Society.

Figure 41. Stabilization of black phase of CsPbI3 (α-CsPbI3) by inclusion of EuCl3 at ambient room temperature conditions. (a, b) Photographs
and (c, d) scanning electron micrographs (SEM) of CsPbI3 films with and without EuCl3. (d) I−V curve of cell based on CsPbI3:5Eu and spiro-
OMeTAD as hole transport layer. Reproduced with permission from ref 273. Copyright 2018 American Chemical Society.

halide solar cells. Although the crystal structure of this black and the cells made with an optimum concentration of the 2D
phase formed at RT has been claimed very recently to be perovskite display a PCE above 11.5%.
orthorhombic, not cubic as one formed at high temperature,264 Partial substitution of Pb2+ with Bi3+,270 Sb3+,271 or Ca2+272
incorporation of additives like HI,262 sulfobetaine zwitter- has also been found to serve the purpose of stabilizing the α-
ion,265 ethylenediamine (EDA)266 into CsPbI3 precursor has CsPbI3. At an optimal concentration of Bi3+ (4 mol% in
been found to stabilize a black photoactive phase at RT. As CsPbI3), δ-phase is completely converted to α-phase, and the
reported by Snaith et al., HI in the CsPbI3 solution induces device made with this perovskite and CuI as an HTM yields an
formation of small grains and creates strain in crystals, which efficiency of 13.2% (Figure 40).270 Two common observations
are likely responsible for stabilization of the black phase of in all the reports presenting formation/stabilization of black
CsPbI3 at RT. This HI-induced black phase CsPbI3 was used phase at RT are (i) reduction in crystal size (or grain size) and
in solar cells of different architectures, and among all different (ii) microstrain in crystals (splitting of the XRD peaks). We
also recently found that inclusion of Eu (both Eu2+ and Eu3+)
architectures, the planar-inverted cells showed the best PCE of
into CsPbI3 can stabilize the black α-CsPbI3 phase at RT
2.9% but they lasted only for several hours.262 Using the same
ambient conditions by reducing the grain size (Figure 41).273
HI additive in the CsPbI3 precursors, Lou et al. developed a
As we observed in this study, regardless of the size of the cation
sequential isopropanol treatment method to improve the PCE (Eu2+ and Eu3+) and stoichiometry of Eu in CsPbI3, black
to 4.1% and stability up to few days. Kim et al. also used HI- phase is formed at about 85 °C for both Eu2+ and Eu3+, and for
induced stabilized black CsPbI3 film in a PSC with an inverted both stoichiometry and non-stoichiometry ratio. This indicates
architecture, which performed with a PCE of 4.88%. However, that not the size of metal ion but its interaction with other ions
with respect to HI-induced stabilization, quantum dot-induced in the solution probably controls the crystallization and grain
stabilization (size-controlled phase stabilization) of α-CsPbI3 growth so as to confine the grains/crystals to nanometer order.
seems to work far more effectively by raising the PCE beyond It seems that stabilization of the black phase at low
10% and the stability near to a month. 267,268 CsPbI 3 temperature is largely sourced by smaller particles/grains and
nanocrystals-based devices have been reported to demonstrate inclusion of suitable metal ions limits crystal growth to even
PCE above 13%.269 It has been also found that inclusion of a nanometer size, and as result, the cubic phase is stabilized. As
small amount of ethylene diamine lead iodide (EDAPbI4) 2D of now, due to lack of understanding about the solution
perovskite in CsPbI3 stabilizes the α-CsPbI3 at RT for months, chemistry of such mixture of ions, there is no general rule for
AF DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 42. (a) Schematic illustration of CsPbI3 perovskite crystallization procedures via solvent-controlled growth (SCG). (b) Enlarged SEM
image of SCG CsPbI3 films (scale bar 5 mm). (c) J−V curves of the devices using CsPbI3 absorber layer prepared without and with SCG (SCG
time is 50 min). (d) Photostability measurement of the devices under continuous one-sun illumination (100 mW cm−2) with UV cut filter (420
nm) in nitrogen glovebox (temperature approximately 25 °C) for the unencapsulated devices. (e) J−V curves of the devices under different
continuous light-soaking time. Reproduced with permission from ref 263. Copyright 2018 Springer Nature.

Figure 43. (a) Schematic illustration of gradient Br doping and PTA organic cation surface passivation on CsPbI3 perovskite thin film. XRD
patterns evolution of (b) CsPbI3 and PTABr-CsPbI3 thin films heated 80 °C in a N2 glovebox for 72 h and (c) PTABr-CsPbI3 and CsPbI3 thin
films after exposed to 80 ± 5% RH at ∼35 °C for 0.5 h; inset is their photographs. Top-surface SEM images of (d) CsPbI3 and (d) PTABr-CsPbI3
thin films. (f) J−V characteristics of champion CsPbI3 and PTABr-CsPbI3-based PSCs under simulated AM 1.5G illumination of 100 mW·cm−2 in
reverse scan. (g) Efficiency histogram of CsPbI3 and PTABr-CsPbI3 PSCs. Reproduced with permission from ref 274. Copyright 2018 American
Chemical Society.

selection of the metal ions (dopants) that can serve the before heating at 350 °C, named as solvent controlled growth
purpose of stabilizing the black phase. Moreover, there is no (SCG), imparted a significant effect on black phase stability
investigation on the effect of charge neutrality in the doped and on the device performance. Besides, it was encouraging to
CsPbI3 (for black phase stabilization) on electronic and see that performance of the cells did not change much until
photovoltaic properties. In contrary to what has been found 500 h under continuous light soaking. Wang et al.274 have
with black phase stabilization by reduction of crystal size, some recently found that CsPbI3 film, even although the grains were
recent studies have reported formation of stable black CsPbI3 large, could retain the black phase by surface treatment with
phase even with large grains. For instance, black CsPbI3 films phenyltrimethylammonium bromide (PTABr) for long time,
prepared at high temperature (350 °C) in a glovebox filled and the solar cells made with the films demonstrated the best
with N2 was composed of huge grains of 5 μm size and the PCE of above 17%, which is the highest reported so far for
solar cells made employing this film demonstrated best PCE of CsPbI3 (Figure 43). These results, indeed, indicate that more
15.7% (Figure 42).263 A slight modification in the annealing understanding about crystallization and film formation of
steps by holding the precursor films at RT for several 10 min CsPbI3, which is lacking now, will certainly help in making
AG DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 44. (a) Schematic presentation of two-step annealing process followed for fabrication of CsPbI2Br thin film. (b) Statistical histogram of the
efficiency of PSCs based on three kinds of ETLs (FTO/NiOx/CsPbI2Br/ETL/Ag). (c) J−V curve and SPO as a function of time held at 0.89 V
forward bias (inset) for the best performing cell using ZnO-C60 bilayer as ETL. Reproduced with permission from ref 277. Copyright 2018
American Chemical Society.

Figure 45. (a) Schematic illustration of the vapor-based synthesis of Cs2TiBr6 thin film. (b) UV−vis spectra (inset: photograph of the final
Cs2TiBr6 thin film). (c) Schematic illustrations of energy levels of ETL (TiO2-C60), Cs2TiBr6, and HTL (P3HT). (d) J−V curves at both forward
(hollow circles) and reverse (solid circles) scans of the best PSCs without and with the presence of the C60 interfacial layer. (e) Stabilized PCE
output at the maximum power points of the PSCs without and with the presence of the C60 interfacial layer. Reproduced with permission from ref
278. Copyright 2018 Cell Press.

further developments in CsPbI3-based cells. Further optimiza- reported works, it can be also judged that the device structures
tion of annealing conditions, finding more suitable dopants, are not optimized; how the ETL and HTL match with CsPbI3
and surface treatments are expected to raise the performance is also not known well. Another factor which even can be more
and long-term stability of CsPbI3-based all-inorganic PSCs important in stabilizing the black phase at RT for longer time is
remarkably high. environment (air, humidity, and light). Similar to the study of
Hence, based on what have been accomplished so far with organic−inorganic perovskites, degradation mechanism of
CsPbI3-based solar cells, it can be foreseen that a lot more CsPbI3 needs in-depth study to know potential of long-term
efforts are needed to be put into different aspects of CsPbI3 to stability of CsPbI3-based solar cells by sincerely addressing/
exploit the actual potential of CsPbI3. As there are not many finding answers to the above-mentioned questions. With more
studies to date, insightful scientific knowledge about several understanding, more alternative methods can be established to
key areas of CsPbI3 photovoltaics are lacking in literature. stabilize the black phase of CsPbI3 at RT and in ambient
Although it is believed that doping or use of additives can environment for sufficiently long time.
stabilize the black phase at RT by limiting to crystal size to
6.2. Cesium−Lead Mixed Halide Perovskites
possibly nanometer size, the actual/active role of these
additives or dopants are still not clear. Can the same Unlike CsPbI3, CsPbBr3 having a band gap of ∼2.3 eV, is not
characteristics of small crystal be obtained by simple solvent an ideal candidate for PV applications. Nevertheless, as the
engineering or by controlling the crystal growth by modifying photoactive phase of CsPbBr3 is stable at RT and in ambient
the annealing/crystallization method? Lack of studies on conditions, several studies have employed CsPbBr3 as an active
charge carriers and their dynamics inside CsPbI3 is also one of absorber in PSCs in different architectures to find that it works
the reasons why the progress has been so much. From the with efficiency around 6%.275 Although this PCE is slightly
AH DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 46. (a) MASnI3‑xBrx crystal structure and (b) its energy level diagram along with TiO2 (ETL) and spiro-OMeTAD (HTL). Reproduced
with permission from ref 290. Copyright 2016 Springer Nature. Top-surface SEM image of FASnI3 (c) without pyrazine and (d) with pyrazine.
Reproduced with permission from ref 291. Copyright 2016 American Chemical Society. (e) Long-term stability of pristine FASnI3 and with 20%
PEA. Reproduced with permission from ref 292. Copyright 2017 American Chemical Society. (f) J−V characteristic plot of best-performing FASnI3
(with 0.08 M PEA) device. Reproduced with permission from ref 293. Copyright 2017 John Wiley and Sons.

lower than that obtained for MAPbBr3, CsPbBr3 shows thermal Br substitution in CsPbI3 lowers the black phase formation
stability up to temperature as high as 580 °C while the former temperature from 350 °C to about 250 °C, and in comparison
(i.e., MAPbBr3) degrades at 220 °C.275 Therefore, having to CsPbI3, CsPbI2Br perovskite structure shows better stability
taken the advantage of thermal stability of CsPbBr3 and lower against phase transition. Hence, usage of CsPbI2Br is
band gap of CsPbI3 (black phase), different combinations of promising for further development but bandgap of 1.9 eV
bromide and iodide (i.e., CsPbI3‑xBrx) have been explored and will be a limiting factor for light harvesting ability. Lowering of
used in PSCs. The band gap of CsPbI3‑xBrx can be precisely this bandgap to ∼1.5 eV while maintaining the phase stability
controlled by I/Br ratio; Eg increasing with more Br and can be challenging because tuning the band gap needs
decreasing with more I in CsPbI3‑xBrx.276 CsPbI2Br with a significant change in the composition. However, mixed cation
band gap of 1.9 V, when used in PSCs with configuration and anion approach is expected to help in improving the
glass/FTO/c-TiO2 /CsPbI2 Br/spiro-OMeTAD/Ag, shows performance and stability of CsPbX3 PSCs. For instance, first-
PCE of 9.8%. Recently, by slightly modifying the method of principles calculations predict that mixing Rb with Cs in
annealing of CsPbI2Br film (two-step annealing process) and CsPbI3 (Rb1‑xCsxPbI3) can stabilize perovskite black phase and
then employing the formed perovskite (CsPI2Br) in cells of provide other beneficial effects from bandgap engineering.
inverted configuration using NiOx as HTM layer (HTL) and Simultaneous mixing of Cs and Rb in A-site, Pb, Sn, or Ge in
ZnO-C60 bilayer as ETM layer (ETL), Liu et al.277 have B-site, and I and Br in X-site should be explored. Indeed, based
obtained PCE of 13.3% in best cells (Figure 44). In on the recent results of titanium-containing inorganic perov-
comparison to inorganic−organic PSCs skite (Cs2TiBr6),278 it can be anticipated that more interesting
(Cs0.04FA0.8MA0.16PbI0.85Br0.15), the CsPbI2Br cells demonstra- perovskite structures with different metals in the B-site of
ted superior long-term stability when the non-encapsulated CsPbX3 are yet to be discovered. Reaction of CsBr with TiBr4
devices were heated continuously at 85 °C under N 2 vapor forming Ti-based vacancy-ordered double perovskite,
atmosphere for 360 h. CsPbI2Br cells suffered only 20% loss Cs2TiBr6 (Figure 45), and a PCE of above 3% achieved with
in PCE while Cs0.04FA0.8MA0.16PbI0.85Br0.15 cells lost 50% of this perovskite in a simple device structure (FTO/TiO2/
their initial PCE after ∼190 h. Cs2TiBr6/P3HT/Au)278 opens up possibilities of new families
AI DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

of perovskite with different metals in the B-site of CsPbX3. effective masses.287 In 2012, Chen et al. employed CsSnI3‑xBrx-
Theoretical and experimental efforts have shown that based Sn perovskite material because of low bandgap (∼1.3
Cs2TiI2Br4 possesses bandgaps of ∼1.38, which is also suitable eV) and exciton binding energies (∼18 meV) and reported
for single-junction PSCs.279 Theoretical calculations also show initial efficiency of 0.9%.288 Subsequently, in 2014, Noel et
that vacancy ordered perovskites like Cs2TiI6, Rb2TiI6, K2TiI6, al.289 and Hao et al.290 independently employed MASnI3 and
and In2TiI6 possess desired electronic and optical properties as MASnI3‑xBrx (Figure 46a), respectively, in TiO2 mesoporous
visible-light absorber materials for PV applications.279 Hence, architecture reporting efficiency of 6.4% in former case and
we are optimistic to see even some more new families of 5.2% in later one. With incorporation of Br−, the VOC was
perovskites to be discovered in near future. significantly increased, attributing to raised conduction band
edge and reduced series resistance in MASnI3‑xBrx, as shown in
7. LEAD-FREE AND LOW-LEAD PEROVSKITES Figure 46b. Although the device performance was not
In addition to stability issues associated with PSCs, lead reproducible and degraded after storing for 12 h, these two
toxicity has been a profound concern for commercialization. reports triggered the field of Sn perovskites. A tremendous
Although “lead” leads in terms of performance of PSCs, its amount of work including different cations and halides in Sn-
toxicity might be an obstacle for industrialization. Lead toxicity based perovskites has been done since then. Tables 5 and 6
is an insidious hazard with the potential of causing irreversible summarize the employed architecture and device parameters of
health effects; interfering with a number of body functions, majority of Sn-based perovskites reported so far.
primarily affecting the central nervous, hematopoietic, hepatic,
and renal system producing serious disorders.283 World Health Table 5. Various Organic/Inorganic Cation-Based Sn
Organization (WHO) has recently considered Pb as one of the Perovskites and Their Corresponding Device Architecture
ten most toxic materials for human health and environment and Parameters
and is making strict policies to avoid usage of Pb. The device JSC VOC PCE
maximum permissible amount of Pb, according to United Sn perovskite structurea (mA/cm2) (V) FF (%) ref
States Environmental Protection Agency (U.S. EPA), is 15 and MASnI3 inverse 4.5 0.15 38 0.3 294
0.15 μg/L in water and air, respectively,284 which are much MASnI3 inverse 12.1 0.38 37 1.7 295
lower than the amount of Pb (0.4 g) estimated in a 1 m2 solar MASnI3 regular 17.4 0.27 39 1.9 296
panel with 300 nm thick perovskite layer. However, based on MASnI3 inverse 17.8 0.6 30 3.2 297
poor solubility of PbI2 in water, it is estimated that the amount MASnI3 regular 16.8 0.88 42 6.4 289
of Pb that can potentially get into environment by complete MASnI2.6Br0.4 inverse 1.0 0.15 51 0.1 298
dissolution of PbI2 from perovskite solar panels will be MASnI1.5Br1.5 regular 5.0 0.45 48 1.1 299
significantly smaller than the content of Pb existing naturally in MASnI2Br regular 11.7 0.82 57 5.7 290
soil.285 Moreover, strong adsorption of Pb on topsoil can MASnBr3 regular 2.2 0.49 46 0.5 299
prevent easy spreading of the pollutant and help in minimizing MASnBr3 regular 4.3 0.5 4.9 1.1 300
the effect. Additionally, mitigation approaches such as putting MASnBr3 regular 7.9 0.88 59 4.3 301
membranes on ground, fence of solar farms and effective FASnI3 inverse 11.7 0.04 23 0.1 302
recycling235 can also be employed to protect the environment FASnI2Br inverse 6.8 0.47 54 1.7 303
from Pb toxicity. Nevertheless, this undesired lead toxicity CsSnI3 inverse 9.5 0.52 61 3 304
issues has provided a compelling motivation to the research a
Regular and inverse device structures represent n-i-p and p-i-n
communities to pave a path toward eliminating (lead-less or architectures.
lead-free) and reducing (less-lead) lead from the perovskites.
7.1. Lead-Free Perovskite Materials Even though Sn perovskites have shown promising
Based on ionic size and Goldschimdt tolerance factor that optoelectronic properties such as narrow optical band gap
predicts formability and stability of perovskite structures, a with potential to absorb light up to 1000 nm, bulk n-type
wide range of cations are predicted to be replacements for Pb electrical conductivity (5 × 10−2 S cm−1), long diffusion
in Pb perovskites. Group 14 elements like Sn2+ and Ge2+, length, and superior electron mobility (∼2000 cm2 V−1 s−1)
alkaline earth metals like Be2+, Mg2+, Ca2+, Sr2+, and Ba2+, than traditional semiconductors such as CdTe and Si,
transition metals such as V2+, Mn2+, Fe2+, Co2+, Ni2+, Pd2+, multivalent nature of Sn and easy oxidation of Sn2+ to Sn4+
Cu2+, Zn2+, Cd2+, and Hg2+, lanthanides like Eu2+, Tm2+, and in ambient conditions hav been constant challenges for its
Yb2+, and p-block elements like Ga2+ and In2+ can be employment in PSCs.305 The oxidized Sn4+ ion acts as a p-type
considered for alternative Pb-free perovskites. However, dopant, causing self-doping of the perovskite layer and, thus,
considering stability of formed perovskite structure and PV limitng PCE. Another factor that limits the PCE is the poor
properties, the candidates which look promising are cations coverage and inhomogeneity of Sn perovskite films due to
Sn2+, Ge2+, Mg2+, Mn2+, Ni2+, and Co2+.127 rapid crystallization,290 resulting in large number of pin-
7.1.1. Tin (Sn)-Based Perovskites. Sn-based perovskites holes.301 Addressing this, Kanatzidis et al. investigated
have been explored as the first lead-free perovskite because of influence of different solvents on crystallization and found
its similar ionic radii (1.35 Å) as of Pb2+ (1.49 Å). that pinhole free high quality films could be obtained in case of
Encouragingly, Sn-based perovskites possess lower bandgap DMSO and NMP, induced by intermediate phase, which
and higher charge carrier mobility of 102−103 cm2/V·s resulted in unprecedented rise in photocurrent, up to 21 mA/
compared to their Pb analogues.286 Moreover, the exciton cm2.306 Additionally, to control the crystallization and improve
ionization is not limited in these materials as majority of 3D the film quality, various deposition techniques such as vapor
Sn-based perovskites exhibit binding energies (2−50 meV) deposition and hybrid vapor deposition−solution methods
similar to that of Pb perovskites, owing to the exceptionally low have been also tried. Morphology and crystallite size of
AJ DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Table 6. Photovoltaic Parameters of Sn-Based Perovskites with Different Additives Used in the Processing
Sn perovskite (with additives) device structure JSC (mA/cm2) VOC (V) FF PCE (%) ref
MASnI3 (with 20% SnF2) regular 3.3 0.23 41 0.3 315
MASnI3 (with 20% SnF2) regular 26.2 0.23 3 2.3 316
MASnI3 (with 20% SnF2) regular 21.4 0.32 46 3.2 306
MASnI3 (with 20% SnF2 and N2H4) regular 19.9 0.38 52 3.9 309
MASnI3 (with 20% SnF2) inverse 11.8 0.45 40 2.1 317
FASnI3 (with 20% SnF2) regular 24.5 0.24 36 2.1 311
FASnI3 (with 10% SnF2 and pyrazine) regular 23.7 0.42 63 4.8 291
FASnI3 (with 20% SnF2) regular 23.1 0.38 60 5.3 318
FASnI3 (with 10% SnF2) inverse 22.1 0.47 60 6.2 319
CsSnI3 (with 10% SnF2) regular 22.7 0.24 37 2.0 314
CsSnI3 (with 20% SnF2 and N2H4) regular 30.8 0.17 35 1.8 309
CsSnI2Br (with 60% SnF2 and H3PO2) regular 16.7 0.33 53 3.2 320
CsSnBr3 (2.5% SnF2) inverse 2.4 0.4 55 0.6 321
CsSnBr3 (20% SnF2) regular 6.6 0.41 48 1.3 322
CsSnBr3 (with 20% SnF2) regular 9.1 0.4 56 2.1 321
CsSnBr3 (with 20% SnF2 and N2H4) regular 14.0 0.37 59 3.0 317
MA0.9Cs0.1SnI3 inverse 4.5 0.2 36 0.3 302
FA0.8Cs0.2SnI3 inverse 16.1 0.24 36 1.4 302
FA0.5MA0.5SnI3 (with SnF2) inverse 21.3 0.53 52.4 1.3 323
FA0.75MA0.25SnI3 (with SnF2) inverse 21.2 0.61 62 8.12 323
FASnI3 (with SnF2) regular 22.5 0.48 66 7.14 324
(BA)2(MA)3Sn4I13 (with SnF2) regular 24.1 0.229 45.7 2.53 325
(PEA)2(FA)8Sn9I28 inverse 14.4 0.59 69 5.97 292

MASnI3 made by two-step deposition method, in which SnI2 in dense, smooth, pinhole free perovskite layer contrasting the
obtained by physical vapor deposition was converted to case of without pyrazine, as shown in Figure 46c,d and also
MASnI3 by spin coating the MAI solution, were found to be prevents oxidation of Sn2+. The resulting device demonstrated
highly dependent on the concentration of MAI and very PCE of 4.8% with high reproducibility, and the encapsulated
interestingly, the MASnI3 films did not show Sn4+ peaks after cells showed stability over 100 days against ambient
storing for 20 days in glovebox and for 90 min in ambient atmosphere. To suppress the adverse effect of excess SnF2,
atmosphere. This superior stability was attributed to high additives (in combination with SnF2) such as hydrazine
quality and dense evaporated SnI2 films.307 Qi et al. found that, (N2H4) and hypophosphorous acid (H3PO2), which act like
while MASnBr3 perovskite deposited by co-evaporation forms reducing agents, have been incorporated into Sn perovskites in
Sn−Br oxide on the surface (capping layer), which restrains combination with SnF2 and have been found to show
generation of excitons and charge transport across the interface improved stability. In comparison to SnF2, SnCl2 as an
and leads to low efficiency (0.35%), MASnBr3 made by additive has shown superior stability. Marshall et al. observed
sequential evaporation method yields a PCE of 1.12% due to that SnCl2 was concentrated on the surface of CsSnI3 and the
much reduced oxidation by the top MABr layer.300 From these devices showed higher stability in ambient atmosphere at 50
results, it seems that two-step deposition technique can be °C, which is believed to be due to formation of SnCl2 layer on
advantageous for stabilizing Sn perovskites. top of CsSnI3 further preventing ingress of water and oxygen.
As mentioned earlier that Sn2+ is prone to oxidation, Additionally, SnCl2 gets converted to SnO2 when exposed to
addition of extra Sn2+ precursors like SnI2,308 SnF2309 and ambient atmosphere, and SnO2 acts as an electron transport
SnCl2310 to precursor solution have been tried to compensate layer.310
missing Sn2+ (see Table 6). Among the mentioned additives, To stabilize Sn perovskite systems, a mixed-cation approach,
SnF2 has been used widely. Incorporation of SnF2 in FASnI3 for instance, addition of Cs+ in MASnI3 or FASnI3, has been
reduces background carrier density, prevents the oxidation, and also employed. However, the device efficiencies were below
improves film morphology to finally result in PCE of 2.1%.311 ∼2% (see Table 6).302 Mixing approach was further extended
Incorporation of SnF2 into CsSnI3 perovskites (1.3 eV)312 also toward larger cations. Cao et al. decreased the dimensionality
resulted in reduction of intrinsic defects associated with Sn by mixing CH3(CH2)3NH3+ (BA+) and CH3NH3+ (MA+) to
vacancies313 and enhanced photocurrent.314 Additionally, it obtain 2D Ruddlesden−Popper perovskite
eliminates the formation of non-perovskite phase of CsSnI3. To ((CH3(CH2)3NH3)2-(CH3NH3)n‑1SnnI3n+1) possessing an op-
note, CsSnI3 is more stable in its yellow phase but addition of timal bandgap of 1.42 and 1.5 eV for n = 4 and 3, respectively.
SnF2, which does not get incorporated into the crystal lattice, In terms of moisture stability and device performance, the
distorts the structure due to much smaller ionic radii of F− lower dimensional perovskite (n = 4) outperformed its 3D
compared to I−291 and thereby prevents formation of the analogue.325 In another work, Liao et al. incorporated 20% of
yellow phase. The best PCE of 2.02% has been obtained with phenylethylammonium (PEA) into FASnI 3 perovskite
20% SnF2 in CsSnI3.314 However, addition of excess SnF2 leads ((PEA)2(FA)8Sn9I28) which also exhibited markedly enhanced
to severe phase separation, which is remarkably reduced by stability against pure FASnI3. The inverted structure device
binding SnF2 to pyrazine.291 Addition of SnF2-pyrazine exhibited best PCE of 5.94%, and the non-encapsulated device
complex combined with solvent engineering process results showed superior stability up to 1000 h.292 Importantly,
AK DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 47. (a) Optical absorption spectra of AGeI3 (A = Cs, MA, FA) in comparison with CsSnI3 and (b) corresponding schematic energy level
diagram. Reproduced with permission from ref 331. Copyright 2015 American Chemical Society.

reducing the amount of PEA (0.08 M) in FASnI3 (0.92 M) SnCl2 furnish Sn2+ to compensate the Sn2+ vacancies and thus
gave highly homogeneous growth with high crystallinity and improve the stability, but the role of halides (F and I) in this is
oriented FASnI3 grains at relatively low temperature. The not known. Besides, strong reducing nature of SnF2 might also
number of grain boundaries was reduced along with reduction work in the same way as other reducing agents (H3PO2) work
of Sn2+ vacancies, background carrier density was reduced in stabilizing Sn2+. Hence, better understanding of mechanism
more than 1 order of magnitude compared to pristine FASnI3 of Sn2+ stabilization (prevention of oxidation of Sn2+) by
films, lifetime of charge carriers increased. This led to additives will certainly help find us suitable agents that can
enhancement in PCE up to 9% which is highest efficiency improve the performance and stability of Sn perovskites
reported for pure Sn perovskite-based solar cells (Figure 1e) to further. Compositional changes occurring at the interface,
date. Furthermore, they reported that the perovskite film depending on the nature of charge transport layers, suggest
showed a high degree of structural disorder and randomly that suitable charge transport layers can help in further
oriented 3D grains when SnF2 was added as an additive. As a enhancement of performance of Sn perovskites. In addition,
result, high density of Sn vacancies was generated causing high cation mixing and dimensional mixing need to be explored
p-doping level. Consequently, the device with only SnF2 more to find ways to stabilize the Sn perovskites longer.
additive demonstrated PCE of 6%.293 7.1.2. Germanium (Ge)-Based Perovskites. Germanium
As stability of Sn-based perovskite is also affected by (Ge), being in the same group as Sn, also finds its place in
additives like LiTFSI in the HTM, spiro-OMeTAD, efforts perovskite family. Ge perovskites (AGeX3) possess similar
have been made to find suitable charge carrier transport transport and optical properties as lead and tin analogues due
materials for Sn perovskite systems. For instance, Hatton et al. to its divalent nature.328 As metal ions containing outer ns2
employed CuI as a HTL on which CsSnI3 perovskite was electron with low ionization energy is considered to impart
deposited (with SnI2 additive) to fabricate a device of inverse good optical and transport properties in AMX3 perovskite
structure. In the device, excess SnI2 was found at the interface structures,43 Ge2+ with 4s2 electron and smaller ionic radii than
of CsSnI3/CuI, which acted further as HTL across the Pb2+ and Sn2+ exhibits good optical and transport properties in
interface to improve the hole extraction efficiency.326 However, AGeX3 (A = MA, Cs) like Sn and Pb perovskites. Both
this kind of phenomenon was not observed in devices with theoretical329 and experimental330 studies have shown depend-
PEDOT:PSS-based CsSnI3 (with excess SnI2 additive), ence of band gap of AGeX3 (A = Cs and MA) on the halide
implying specificity of the phenomenon to nature of the ions; band gap increasing with decreasing size of halide ions.
under layer HTM (CuI or PEDOT:PSS). Such dependence of For instance, in case of CsGeX3, Eg changes from 1.6 to 2.3 to
composition or phase of Sn perovskite on the under layer was 3.2 eV when X changes from I to Br to Cl, respectively. On the
also evident from the work done by Sun et al., who employed other hand, contrary to Pb perovskites, Ge-based perovskites
CsSnI3 in three different architectures, namely, mesoscopic show increasing band gap with an increase in the radii of A-
architecture, meso-super-structured, and planar heterojunction cations, as shown in Figure 47a,b. Moreover, due to
structure, and witnessed formation of yellow phase of CsSnI3 stereochemical activation of the 4s2 lone pair, incorporation
and Cs2SnI6 only in the former two architectures. This of bulkier cations such as guanidinium, trimethylammonium,
indicates non-ideal solid−solid interfaces between the perov- and isopropylammonium not only increases the band gap but
skite and scaffold material.327 also reduces the dimensionality (one-dimensional (1D)) and
In summary, despite the excellent optoelectronic properties nature of band gap from direct to indirect one. This results in
of Sn perovskites, they are far behind Pb perovskites in terms poor absorption by such Ge perovskites.331
of performance. Such poor performance, as can be Mhaisalkar et al. demonstrated a device incorporating
comprehended from literature, is a result of combined effects CsGeI3 and MAGeI3 with PCE of only 0.2% and 0.11%,
of poor film quality, oxidation of Sn2+ to Sn4+, degradation respectively. It is believed that low binding energy of Ge 4s2
caused by additives in the HTM, and sensitivity of composition electrons makes Ge perovskite unstable when exposed to air
to the interfacial layers. Although the use of additives like SnF2 and, thus, limits the PCE.330 However, due to large band gap,
and SnCl2 as Sn2+ compensator and strong reducing agents like Ge perovskites are being considered more relevant as a top
H3PO2 has been found effective in preventing oxidation of absorber in tandem devices. Although there is no study on
Sn2+, a complete understanding of the mechanism/role of such tandem cells using Ge perovskites until now, and there are only
additives in doing so is still lacking. It is assumed that SnF2 and few studies showing device performance based on Ge
AL DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 48. (a) J−V characteristic curve of Sn−Ge binary perovskite best performing device and (b) its corresponding stability test. Reproduced
with permission from ref 336. Copyright 2018 American Chemical Society.

Figure 49. (a) Crystal structure of MA3Bi2I9 showing local structure of the Bi2I93− anion (left) and cation and anion positions in the unit cell
(right). Reproduced with permission from ref 343. Copyright 2016 Royal Society of Chemistry, under Creative Commons Attribution 3.0
Unported License. (b) Photograph of MA3Bi2I9 and MAPbI3 deposited on quartz substrates showing stability. Reproduced with permission from
ref 345. Copyright 2015 John Wiley and Sons. Top-surface SEM images of MA3Bi2I9 (c) without and (d) with NMP. Reproduced with permission
from ref 348. Copyright 2017 Royal Society of Chemistry, under Creative Commons Attribution-NonCommercial 3.0 Unported License. (e)
Device characteristic J−V curve of MA3Bi2I9 and (f) the corresponding stability. Reproduced with permission from ref 350. Copyright 2018
Elsevier.

perovskites, a recent theoretical investigation showing 7.1.3. Lead-Free Binary Metal Halide Perovskites. In
remarkable electron and hole conductive behavior of addition to Sn- and Ge-based single metallic perovskites,
bimetallic perovskite compositions, containing two metals
MAGeI3 with ample stability compared to MAPbI3332
simultaneously (e.g., Tl-Bi, Sn-Ge, etc.), have been studied. In
promises potential applications of Ge perovskites as an fact, such bimetallic compositions have shown better perform-
alternative to Pb perovskites. ance and stability. A theoretical investigation on a Tl/Bi
AM DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Table 7. Band Gap and Photovoltaic Parameters of Various Perovskite, Perovskite-like, and Non-perovskite Materials Based
on Bi and Sb
active layer Eg (eV) JSC (mA/cm2) VOC (V) FF PCE (%) ref
MA3Bi2I9 2.1 1.00 0.67 62.5 0.42 358
MA3Bi2I9 2.22 1.39 0.83 34 0.39 359
MA3Bi2I9 2.1 1.10 0.65 0.50 0.36 360
MA3Bi2I9 − 0.94 0.51 0.61 0.31 348
MA3Bi2I9 − 0.83 0.56 49 0.26 347
C5H6NBiI4 1.98 2.71 0.62 0.54 0.9 361
(H3NC6H12NH3)BiI5 2.1 0.12 0.40 43 0.03 362
Cs3Bi2I9 2.03 2.15 0.85 60 1.09 346
CsBi3I10 1.77 3.4 0.31 38 0.4 362
MA3Sb2I9 2.14 1.0 0.9 55 0.49 363
MA3Sb2I9 1.95 5.41 0.62 60 2.04 356
Rb3Sb2I9 2.1 2.11 0.55 57 0.66 355
Cs3Sb2I9 2.05 <0.1 0.31 − <0.1 353
Cs3Sb2I9 2.0 2.91 0.6 48.1 0.84 356

aliovalent ionic pair (i.e., MATl0.5Bi0.5I3) shows a calculated (Sb)- and bismuth (Bi)-based perovskites have been
band gap of 1.03 eV without considering spin−orbit coupling theoretically and experimentally investigated. Sitting in the
(SOC) and 0.35 eV with SOC.333 However, no information on neighborhood of Pb, Sb and Bi possess similar electronic
the device performance has been reported yet. Another binary configuration and comparable ionic radii with that of Pb,
metal perovskite explored is Sn-Ge. First, Ju et al. performed allowing them to incorporate effectively into the perovskite
theoretical studies and predicted the possibility of a series of lattice. In this regard, ternary non-toxic metal halides-based
mixed Sn-Ge perovskites, and among all the explored materials, perovskite materials (A3M2X9; A = Cs+, MA+, Rb+; M = Bi, Sb;
RbSn0.5Ge0.5I3 was found to possess a direct band gap within X = I−, Cl−) have emerged as potential candidate due to their
an optimal range of 0.9−1.6 eV and an optical absorption non-toxicity and high moisture stability. Optical properties,
similar to that of MAPbI3.334 Later, Sadhanala et al. crystal structure, dielectric, and quantum physical properties of
synthesized Sn−Ge perovskites by varying the metal ion various ternary bismuth halide perovskite materials have been
composition (i.e., Sn-to-Ge ratio) and studied their optoelec- formerly investigated,338−342 but attempts to use them in PV
tronic properties.335 With the incorporation of Ge2+ into devices were only recently addressed. Eckhardt et al.
MASnI3, the authors observed change in the crystal structure systematically investigated the single-crystal structure of
from tetragonal (0% Ge2+) to trigonal (75% Ge2+) owing to MA3Bi2I9 and found that three equivalent symmetrical I−
smaller ionic radius of Ge2+ (73 pm) which caused distortion helps to combine every two Bi3+, and these I− are situated
in the structure. The tolerance factor increased from 0.84 for on the other sides of mirror planes as shown in Figure 49a.343
pure MASnI3 to 0.87, 0.90, and 0.93 for 25, 50, and 75% Ge2+, The absorption coefficients for different Bi perovskites were
respectively, and enthalpy of decomposition decreased from estimated to be 1 × 105 cm−1 at 450 nm, lower than 2 × 105
0% Ge to 50% Ge, indicating greater intrinsic stability of cm−1 for MAPbI3 at same wavelength. Additionally, MA3Bi2I9
MASn0.5Ge0.5I3. While the band gap of this Sn−Ge system showed Wannier−Mott exciton binding energy of 70 meV
ranged from ∼1.3 eV (0% Ge) to 1.9 eV (75% Ge), 50% Ge which is slightly higher than lead halide perovskites (12 ± 4
case showed a band gap of ∼1.5 eV with lowest structural meV).344 Crystal structure, optoelectronic properties, and
disorder, thus suitable for single junction solar cells. Although stability of MA3Bi2I9 deposited by conventional two-step
no device employing MASn0.5Ge0.5I3 has been reported so far, solution and vapor processed deposition were systematically
Ito et al. have fabricated devices with dual cation (FA-MA), studied by Hoye et al.345 Phase pure MA3Bi2I9 showed indirect
dual metal (Sn-Ge)-based perovskites (FA0.75MA0.25Sn1‑xGexI3; band gap of 2.04 eV which is close to the value reported by
x = 0−0.2).336 The cell with 5% Ge showed PCE of 4.48% Park et al. They also compared long-term stability of MA3Bi2I9
(Figure 48a) versus 3.31% for pure Sn-based perovskite (i.e., with MAPbI3 by exposing both films to relative humidity of
FA0.75MA0.25SnI3). Interestingly, the authors observed en- 61% (Figure 47b). The color of MAPbI3 changed from brown
hancement in PCE up to 6.9% after storing the cell for 72 h to yellow in 5 days while MA3Bi2I9 maintained its color even
in N2 atmosphere. With Ge addition, the PCE retained 80% of after 13 days and showed slightly bright color after 26 days,
its original performance after exposing to air, affirming which was accounted for the formation of Bi2O3 or BiOI layer
significant stability (Figure 48b). Enhancement in PCE due on the surface. Moreover, vapor processed film exhibited
to Ge doping was attributed to low disorder defects and characteristic PL decay times over 0.76 ns, with bulk lifetime
passivation effects of Ge. Another class of lead-free binary possibly close to 0.56 ns, promising its application in PV
metal perovskites was reported by Vargas et al., who devices.345 Park et al.346 integrated three different ternary
incorporated Cu 2+ into antimony perovskite, yielding bismuth halide absorbers, Cs 3 Bi 2 I 9 , MA 3 Bi 2 I 9 , and
Cs4CuSb2Cl12 having a direct band gap of 1 eV. Cs4CuSb2Cl12 MA3Bi2I9‑xClx, with optical band gaps of 2.2, 2.1, and 2.4 eV,
is stable against heat-stress, light, and humidity and possesses respectively, into PV devices. Best performing devices showed
better conductivity than MAPbI3, highlighting its promising 1%, 0.1%, and 0.003% PCE with Cs3Bi2I9, MA3Bi2I9, and
PV applications.337 MA3Bi2I9‑xClx perovskites respectively.346 We investigated the
7.1.4. Group 15 Metal-Based Perovskite/Nonperov- effect of various TiO2 layers on performance of MA3Bi2I9 cells
skite Materials. Lately, from the Group 15 metals, antimony to find that MA3Bi2I9 forms a uniform capping layer with
AN DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Table 8. Optical Band Gap and Photovoltaic Parameters of Devices with Silver Bismuth Halide and Cu-Based Non-perovskite
and Perovskite-like Materials
active layer Eg JSC (mA/cm2) VOC (V) FF PCE (%) ref
AgBi2I7 1.87 3.3 0.56 67.4 1.22 374
AgBi2I7 1.78 1.6 0.46 56 0.4 375
AgBi2I5 1.85 6.8 0.49 63 2.1 375
AgBi2I5 − 4.898 0.592 60 1.74 378
AgBi2I5 (with excess BiI3) − 6.16 0.612 59.21 2.22 378
Ag3BiI6 1.83 10.7 0.63 0.64 4.3 284
(CH3(CH2)3NH3)2CuBr4 1.76 1.78 0.88 40 0.63 379
(p−F-C6H5C2H4−NH3)2-CuBr4 1.74 1.46 0.87 40 0.51 379
MA2CuCl2Br2 2.12 0.22 0.26 32 0.02 380
MA2CuCl0.5Br0.35 1.8 0.021 0.29 28 0.002 380
C6H4NH2CuBr2I 1.64 6.20 0.2 46 0.46 381
SbSI 2.15 6.42 0.55 56.6 2.00 382
SbSI (P3HT HTL) 2.15 7.06 0.53 56.7 2.12 382
SbSI (PCBDTBT HTL) 2.15 9.06 0.56 56 2.84 382
BiI3 1.83 3.4 0.31 0.4 0.49 383
BiI3 1.8 7.0 0.364 39.2 1.02 384

improved morphology on anatase TiO2 under layer which of 3.17% with VOC = 1.01 V (Figure 49e) which is the highest
results in PCE of 0.2%.347 Since fast crystallization of MA3Bi2I9 value reported for ternary bismuth-based PSCs until date.
is responsible for poor morphology and carrier transport across Moreover, the devices (in case of 25 min MAI vapor) showed
the interface, we incorporated N-methyl-2-pyrrolidone (NMP) enhanced stability up to 60 days, as shown in Figure 49f.350
as an additive to slow down the crystallization of MA3Bi2I9, and Similar to bismuth, antimony (Sb)-based perovskites exhibit
as a result, the morphology improved with fewer pinholes,348 good optoelectronic properties and, due to similar electronic
as shown in Figure 49c,d. However, the obtained performance configuration, comparable electronegativity, smaller ionic
was not enhanced much (∼0.31%), indicating that tuning the radius (76 pm), and 3+ state, A3Sb2X9-based perovskites also
morphology does not help to improve the performance and form lower dimensional structures, limited up to 2D.352
attention should be rather directed toward tuning intrinsic Cs3Sb2I9 is reported to exist in 0D dimer (space group P63/
optoelectronic properties. Later, Zhang et al. developed a novel mmc) and 2D layered forms (space group P3m1).353 In the
two-step approach to obtain MA3Bi2I9 films in which BiI3 was crystal structure of Cs3Sb2I9, one-third of Sb layers along the
initially deposited under high vacuum and then BiI3 was ⟨111⟩ direction are vacant resulting in layered 2D structure
transformed to MA3Bi2I9 under low vacuum. As a result, highly with a direct band gap of 2.05 eV, which is similar to that of
compact, pinhole free layers with large grains were obtained. MA3Sb2I9. Cs3Sb2I9 also shows high charge carrier mobilities
Moreover, the films exhibited larger absorption coefficient in and good tolerance to defects.354 The layered 2D form of
low energy regime, and charge diffusion length and density of Cs3Sb2I9 has high absorption coefficient, comparable to Pb
trap states were comparable with lead perovskite films. Even perovskite, which derives from their electronic structure.
after obtaining uniform morphology (without pinholes), the However, when deposited via solution process, the dimer
best devices showed PCE of ∼1.6%, which is far behind that of form of Cs3Sb2I9 is preferentially formed over 2D layered
MAPbI3.349 This was further in the line of our observations.348 form.353 In contrast, when Cs is replaced by Rb, the Rb3Sb2I9 is
Regardless, the wide band gap (2.1 eV) of MA3Bi2I9 was one effectively stabilized in 2D layered form with a slight blue shift
of the limiting factor for performance. To address the bandgap of the optical gap (∼2.24 eV).355 Device employing this
issue, Vigneshwaran et al. attempted doping of sulfur into Rb3Sb2I9 perovskite shows PCE of 0.66%. As summarized in
MA3Bi2I9 perovskites at relatively lower temperature (120 °C) Table 7, even after several attempts the efficiency of Sb
and observed band gap reduction to 1.45 eV, which is even perovskites was limited to <1% until Boopathi et al. tried an
lower than MAPbI3.351 Moreover, Hall effect measurements additive assisted approach to deposit A3Sb2I9 (A = Cs, MA),
suggested that the resultant perovskite behave as a p-type which improved the morphology significantly and thereby
semiconductor with higher carrier concentration and mobility resulted in PCE of 2.04%.356 In MA3Sb2I9, incorporation of
compared to undoped MA3Bi2I9. Several other attempts were Cl− has been found to transform the dimer form to 2D layered
made to reduce the band gap and tune the morphology and structure which reaching a record PCE of 2%.357
dimensionality, which are summarized in Table 7. In addition to A3Sb2I9 class of materials, Sb2S3 having 1D
One important finding was reported by Jain et al., who chained structure and A2Sb8S13 (A = Cs+, MA+) with 2D
developed MA3Bi2I9 perovskite by vapor-assisted solution layered structures show semiconducting properties with band
process (VASP) in which CH3NH3I (MAI) vapor were gaps 1.72, 1.85, and 2.08 eV for Sb2S3, Cs2Sb8S13, and
exposed on to the solution processed BiI3 film for different MA2Sb8S13, respectively.364 From the electronic structures, the
times.350 Interestingly, concentration of Bi0 in BiI3 was work function and ionization potential were also calculated,
substantially reduced by exposing MAI vapor for longer time shedding light on possible contact materials for PV
(25 and 45 min) and at particular exposure time of 25 min, the applications.
MA3Bi2I9 showed improved morphology compared to other It is well known that 3D layered structures work better as
processed conditions owing to mitigation of metal defects light harvesters than lower dimensional structures because of
sights. The device at this condition resulted in record efficiency their lower band gap and exciton binding energy. Thus,
AO DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Table 9. Band Gap, Device Architecture, VOC, and Power Conversion Efficiency (PCE) of Sn-Pb Binary Perovskites of
Different Compositions
perovskite device architecture band gap VOC (V) PCE (%) ref
MASn0.1Pb0.9I3 inverse 1.24 0.78 10 385
MASn0.15Pb0.85I3 inverse 1.4 0.76 9.8 387
MASn0.2Pb0.8I3 inverse − 0.77 5.6 297
MASn0.25Pb0.75I3 regular 1.24 0.73 7.4 301
MASn0.25Pb0.75I3 inverse 1.31 0.7 7 385
MASn0.25Pb0.75I3 inverse 1.26 0.8 11.2 388
MASn0.25Pb0.75I3 inverse 1.3 0.77 15.2 386
FASn0.25Pb0.75I3 inverse 1.31 0.75 7.3 294
FASn0.25Pb0.75I3 inverse − 0.75 10.3 302
MA0.9Cs0.1Sn0.25Pb0.75I3 inverse 1.39 0.84 14.6 302
MASn0.25Pb0.75(I0.9Br0.1)3 inverse 1.43 0.85 14.5 298
MASn0.25Pb0.75Br3 inverse 2.03 0.76 1.5 298
MASn0.4Pb0.6I3 inverse − 0.75 7.3 297
FA0.75Cs0.25Sn0.5Pb0.5I3 inverse 1.2 0.74 14.8 389
MA0.9Cs0.1Sn0.5Pb0.5I3 inverse 1.28 0.7 12.7 302
FA0.6MA0.4Sn0.6Pb0.4I3 (with SnF2) inverse 1.2 0.8 15.1 390
MASn0.05Cu0.05Pb0.9I0.95Br0.1 inverse 1.58 1.086 21.08 391

attempts were made to form 3D structures consisting of such Beyond AMX3 or A2MM′X6 perovskite structures, design
Sb and Bi trivalent metal ions. This led to explore the and electronic structure of lead-free chalcogenide-halide mixed
possibilities of heterovalent substitution of Pb2+ by incorporat- perovskites with formula MABiXY2 (X = chalcogen; Y =
ing trivalent metals (Bi, Sb) in combination with monovalent halogen) have been theoretically and experimentally inves-
metals such as silver (Ag), gold (Au), copper (Cu), and tigated. In these compounds, the atomic structure is similar to
potassium (K) into the perovskite structure, forming double tetragonal structure of MAPbI3, and compounds like MABiSI2
perovskites possessing molecular structure of A2MM′X6 (A = and MABiSeI2 possess similar optical gap (∼1.3−1.4 eV) as of
Cs, MA; M = Bi, Sb; M′ = Ag, Au, Cu, K; X = I, Cl, Br).365 lead perovskites, and hence, are quite suitable for single
Among many combinations of double-perovskite materials, junction solar cells.372 Seok et al. made the first experimental
Cs2BiAgBr6 and Cs2BiAgCl6 have been explored earnestly. attempt to synthesize such chalcogenide-halide mixed perov-
Both Cs2BiAgBr6365,366 and Cs2BiAgCl6366 exhibit wide and skite (MASbSI2).373 When employed in a solar cell, MASbSI2
indirect band gaps. Only a few studies have reported demonstrated the best PCE of 3.08%. Moreover, non-
employment of Cs2BiAgBr6 in PV devices. Efficiencies of the encapsulated cells retained 90% of its initial efficiency when
cells of different architectures including Cs2BiAgBr6 as the stored in relative humidity of ∼60%.373 Silver bismuth halides,
absorber vary in the range 1−3%, depending on architecture for example AgBi2I7, have also emerged as one of the potential
and preparation conditions.367−369 Even though these candidates for lead-free PSCs. In the first study on such
attempts367,368 have shown that double perovskites based on compounds, Sargent et al. prepared AgBi2I7 (non-perovskite
Ag−Bi combination can be a promising replacement for lead cubic phase) thin film by dissolving bismuth iodide (BiI3) and
silver iodide (AgI) in n-butylamine, and the cell made with this
perovskites (summarized in Table 8), Savory et al. reported its
film exhibited 1.2% efficiency and improved long-term stability
limitations to achieve high PCE (maximum limit is 10%),
against humidity and light-soaking.374 However, this efficiency
owing to its wide indirect band gap and large carrier effective
was not reproduced in the follow-up studies by Johansson et
masses.370 Theoretical investigation suggests that this limi-
al.375 and Shao et al.376 Such poor reproducibility of the results
tation can be overcome by replacing Ag with indium (In) or can be due to sensitivity of crystal structure of AgBi2I7 to
thallium (Tl). Bi-In- and Bi-Tl-based double-perovskite annealing conditions, which, although has not been inves-
materials possess direct band gap of ∼2 eV.370 It has been tigated thoroughly, is indicative from the works reported by
found experimentally that replacement of Ag with Tl in Sargent et al. and Mitzi et al., while Sargent et al. reported that
(MA)2AgBiBr6 also results in direct band gap of 2 eV. crystallization of AgBi2I7 occurs at relatively high temperature
Unfortunately, in addition to such wide band gap (∼2 eV) (∼150 °C) due to strong complex formation of BiI3 and AgI
which is not suitable for PV applications, higher toxicity of Tl with n-butylamine.374 Mitzi et al. presented that annealing
(than Pb) limits use of Bi−Tl perovskites in PSCs.371 AgBi2I7 films at high temperature (150 °C) results in loss of
As it is evident from these studies, wide and indirect band BiI3 from the crystal structure, forming Ag1.16Bi1.04I4.00 instead
gap of Ag−Bi perovskites is largely limiting the performance of of AgBi2I7.377 Addressing this phase impurity issue, we
cells based on such double perovskites. Hence, development of followed a solvent engineering process to obtain phase pure
direct band gap double-perovskite material (through doping) AgBi2I7. As a result, PCE of 2% was obtained, which is, to date,
becomes the present challenge for further development in the highest efficiency reported for AgBi2I7-based cells (the
performance of such double perovskites. Moreover, VOC ≈ 1 details cannot be disclosed here as the work is not published
V,367 demonstrated by Cs2AgBiBr6 double-perovskite and yet). Nevertheless, a clear understanding about formation of
MA3Bi2I9-based devices, indicates their suitability for tandem. different phases (i.e., AgBi2I7, AgBiI4, AgBi2I5 or Ag3BiI6) is
Applications of these perovskites in tandem structures should lacking now.284,378 Some of such phases along with their device
be also explored. performances are listed in the Table 8. It is to be noted that
AP DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 50. Common scalable methods for perovskite deposition for large are perovskite solar modules fabrication; (a) blade coating, (b) slot-die
coating, (c) spray coating, (d) inkjet printing, and (e) screen printing. Reprinted with permission from ref 409. Copyright 2018 Springer Nature.

excess of BiI3 in the Ag−Bi iodide compounds plays a role of indicates that more studies are required to understand this
passivating the defects/traps.378 This suggests that optimiza- anomalous behavior of band gap of Pb−Sn perovskites and its
tion of the ratio of BiI3 to AgI and understanding of the sensitivity to processing conditions.
mechanism of defects passivation in such Ag−Bi iodide However, like in case of Pb perovskites, tuning of
compounds can help us improve the performance further. morphology and crystallinity,386 and multi-cation approaches36
7.2. Low-Lead Perovskite Materials: Reducing Toxicity and have been adopted for Pb/Sn perovskites to improve
Enhancing Efficiency performance and stability. Performances of PSCs based on
In earlier sections, we have discussed lead-free perovskite, non- Pb/Sn perovskites of different compositions are summarized in
perovskite, and perovskite-like materials. We have also Table 9.
discussed binary metal halide perovskites. In all these studies It has been found that, in FA-based Pb/Sn perovskites,
reviewed in the previous section, the objective was to eliminate FASnI3 prevents the formation of gamma-phase of FAPbI3 and
the lead completely from the structure. However, as the simultaneously, Sn2+ is stabilized by Pb2+.36 Incorporation of
performance of PSCs based on these Pb-free materials are still Cs+ into FAPb1−xSnxI3, stabilizes the black phase, shifts the
low, efforts aiming at reducing the content/toxicity of lead Fermi level toward the valence band and makes the valence
instead of eliminating it completely have been made. band more shallower (compared to Cs-free perovskites),
Partial substitution of Pb with Sn in Pb perovskites, resulting making an ideal contact with PEDOT:PSS (HTL) and, thus,
in Pb−Sn alloy perovskites, has been studied extensively. Pb− results in improved performance.143,392 Cs+ was also
Sn alloy perovskites with concentration of Sn higher or equal incorporated into pure MA-based Pb−Sn perovskite system.
to 50% often crystallizes in the space group of pure tin Cs in MA(Pb/Sn)I3 does not affect the band gap and
perovskite (tetragonal, P4mm), while Sn lower than 50% crystallinity but improves the film morphology and thus, the
adopts space group as that of the lead perovskite (tetragonal, PCE and stability against ambient atmosphere for 20 days.302
I4cm).301 A great discrepancy exists in the band gap values Interestingly, XPS results revealed that MA2SnI6 is formed only
(Table 9) of Pb−Sn alloy perovskites reported in different in Cs-free perovskite layer upon exposure to ambient
studies,297,301,385 possibly linked to the methods of perovskite atmosphere, caused by oxidation of Sn2+ to Sn4+. This confirms
formation. While Hao et al.301 observed an anomalous change that Cs+ incorporation is effectively reducing the oxidation of
in band gap values of MA(Pb1‑xSnx)I3 for x varying between 0 Sn2+ when exposed to ambient atmosphere. Efforts were also
and 1 (prepared by RT drying of solution), which instead of made to incorporate Br− into the perovskite structure to
falling between the two extremes (i.e., 1.55 eV for MAPbI3 and increase the band gap. The band gap of mixed halide Pb/Sn
1.35 eV for MASnI3) became even smaller (<1.3 eV), Zhao et perovskites were found to be tunable from 1.35 to 2.03 eV by
al.297 the obtained higher bad gaps for MA(Pb1‑xSnx)I3 changing the halide ratio (I to Br), and importantly, these
(prepared by solution hot casting) with 20% (1.5 eV) and perovskites are stable under continuous illumination, showing
40% (1.39 eV) Sn. Importantly, this strategy to lower the band no halide segregation.392 The devices with
gap (by combining Pb and Sn) down to ∼1.1−1.3 eV can help MAPb0.75Sn0.25(I0.4Br0.6)3 (Eg = 1.73 eV) showed PCE of
to amalgamate particular ratio of Sn−Pb perovskite materials 12.6% with VOC = 1.04 V with improved stability against
as a bottom cell configuration in tandem architecture. This continuous illumination under N2 atmosphere and high
AQ DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 51. Examples of some large-area perovskite solar modules. (a) Module of 6 in. × 6 in. dimensions prepared by slot-die coating, showing a
module PCE of 10%. Reproduced with permission from ref 410. Copyright 2018 Elsevier. (b) Module with dimensions of 8 cm × 8 cm prepared by
pressure-spreading of perovskite precursor solution, performing at a PCE of 13.9%. Reproduced with permission from ref 411. Copyright 2017
Springer Nature. (c) A 10 cm × 10 cm module fabricated by a two-step method (formed PbI2 film was dipped in MAI solution) and with interfacial
modification with graphene, showing PCE of 12.6%. Reproduced with permission from ref 412. Copyright 2017 American Chemical Society. (d) A
carbon-based HTM-free module of 10 cm × 10 cm fabricated by screen printing, exhibiting a PCE of 10.4%. Reproduced with permission from ref
413. Copyright 2017 John Wiley and Sons.

temperature in inert atmosphere, making them promising and (iv) determination of impacts of such change in
candidates for tandem applications. Interestingly, incorpora- architecture on performance. Despite these challenges, close
tion of CuBr2 as a source for Br− in Sn−Pb perovskite leads to matching of performances of cells made by scalable methods
formation of a ternary perovskite (MAPb0.9Sn0.05Cu0.05I2.9Br0.1) like blade coating393 with that of spin-coating methods
with improved crystallization, enlarged grains.391 Addition of promises great potential of scaling up of PSCs.
Cu2+ forms a pinhole-free denser film and minimizes the Methods like blade coating,393,394 slot-die coating,395−397 ink
extrinsic trap sites at the crystal boundaries. Moreover, electron jet printing,398−400 screen printing,401,402 spray coating,403,404
and hole mobility of MAPb0.9Sn0.05Cu0.05I2.9Br0.1 was extrapo- vapor phase deposition,405,406 electrodeposition407,408 are a
lated to be 0.83 and 1.20 cm2 V−1 S1−, which are even higher good number of scale-up methods that can be used for
than that of MAPbI3. Among all the studied compositions, fabrication of largearea PSCs. While some of these methods
MAPb0.9Sn0.05Cu0.05I2.9Br0.1 device showed PCE of 21.08% can be integrated into roll-to-roll (R2R) printing methods
which is, to date, the highest efficiency reported for low-lead- (Figure 50), others can be done only through screen-to-screen
based perovskite material.391 (S2S) manufacturing. Figure 51 provides some examples of
recent achievements with the largearea perovskite solar
8. TOWARD COMMERCIALIZATION modules.
A PCE of 10−14% in large-area modules, as shown in Figure
Power conversion efficiency >20% achieved in a smallarea lab
51, is quite encouraging but a substantial gap still remains
cell is certainly encouraging but easy translation of this
between performance of small-area cells and large-area
performance in largearea cells/modules is a prerequisite to
modules. Unoptimized design of large-area cells/modules can
industrialization. Factors such as increase in sheet/series
limit the output power significantly and thus may be the main
resistance, homogeneity of the film over large area, loss in
reason for the performance gap between small- and large-area
active area due to carrier collectors, which contribute to loss of
cells. A study based on modeling of structure of large-area
performance on scaling up the size, are inevitable. Therefore,
modules reveals that optimized design of the front electrode
development of manufacturing methods for large-area modules
with reliable and reproducible performance with minimum loss with metal grids (fingers and bus bars) of suitable dimensions
from the small cell performance is a first step toward can yield a PCE of 20%, which is same as small-area cells, in
industrialization. Progress, potentials, and challenges of the modules of dimensions 156 mm × 156 mm.414 The
different available scaling-up methods for PSCs are discussed performance gap between small- and large-area cells also
below. largely comes from the difference in morphology/quality of
film made by spin-coating and scale-up methods like balding or
8.1. Scaling Up and Reproducibility Challenges slot-die coating. Recently, formulation of ink/solution and
Considering the state-of-the-art device architectures (n-i-p or modification of annealing/drying conditions specific to the
p-i-n) that generally produce high efficiency (for example, scale up methods has improved the performance of largearea
FTO/c-TiO2/with or without mp-TiO2/perovskite/HTM modules, narrowing the difference between smallarea cells and
(spiro-OMETAD)/Au), up-scaling of PSCs such as (i) largearea modules. Constraints of adopting some important
development of scalable deposition strategies for coating all steps that are used in small-scale fabrication in laboratories for
the layers including perovskite absorber layer, ETL, HTL, and large-area fabrication mandate alterations in the process to
electrodes over large area with same uniformity as obtained control nucleation and crystal formation. For example, an
with spin-coating method for smallarea cells, (ii) modification antisolvent dripping step, usually employed in spin-coating
of the precursor solution chemistry required to match with processing of lab scale cells for preparing films of high quality,
scale-up processes to achieve same control over film formation is not suitable for methods like blading or slot die coating.
over large area, (iii) establishment of procedures for fabricating Nevertheless, antisolvent spraying415 and antisolvent bath-
modules that have to differ in architecture from smallarea cells, ing416 methods can be followed for largearea fabrication. As
AR DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 52. (a) Chemical structure of ITIC-Th (3,9-bis(2-methylene-(3-(1,1-dicyanomethylene)-indanone))-5,5,11,11-tetrakis(5-hexylthienyl)-

dithieno[2,3-d:2′,3′-d′]-s-indaceno[1,2-b:5,6-b′] dithiophene). (b) Schematics showing films fabricated from the fresh and aged precursor
solutions without and with ITIC-Th. (c) A series of photographs of films fabricated by using perovskite precursor solutions without and with ITIC-
Th, aged for different durations. XRD patterns of the films (d) without and (e) with ITIC-Th. The symbols *, #, + , and • denote the diffraction
peaks corresponding to the α-phase, δ-phase, PbI2, and TiO2/FTO substrate, respectively. J−V curves of perovskite (MA0.17FA0.83Pb(I0.83Br0.17)3)
solar cells (FTO/TiO2/perovskite/spiro-OMeTAD/Au) made from solutions (f) without and (g) with ITIC-Th aged for different times.
Reproduced with permission from ref 421. Copyright 2018 John Wiley and Sons.

annealing the films on hot plate for durations as long as 1 h, as Solution processing is a simple and cost-effective method
is done in laboratories for small cells, is not desirable for large- and, therefore, is certainly an advantage for PSCs. Indeed, that
area manufacturing, rapid thermal annealing techniques like is the reason why the amount of work that has been done in
IR-sintering, intense pulsed-light sintering,417 photonic flash just less than 10 years is tremendous. But, the dependence of
annealing, 418 etc. need to be employed in large-area optoelectronic properties and PV performance of the cells
fabrication. In order to control the crystal growth (film predominantly on the film characteristics, such as grain size,
formation) over large areas during the rapid thermal annealing crystallinity, phase purity, defects density, etc., which are
processes, some special solvent mixture and/or some chemical directly translated from the solution properties, makes one
additives techniques394 are necessary. Therefore, adjusting the point clear: a thorough understanding of the chemistry of the
precursor chemistry according to the scale-up method becomes solutions is important for achieving high performance with
critical for large-area manufacturing of PSCs. Control of film high reproducibility. Reproducibility has been a constant
formation over large area through special techniques included challenge since the beginning years of development. Earlier,
in the processes is expected to close the performance gap the challenge lay essentially in obtaining perovskite films with
full coverage (pinhole-free) and with large grains, which has
between small- and largearea cells.
been addressed well in these years. Processing in controlled
Another important aspect of largearea manufacturing
humid conditions, use of additives in precursors, antisolvent
methods that needs serious attention is reproducibility. As
dripping or special annealing conditions have provided the
almost all developments on smallarea lab cells have happened
solutions for preparing a good quality film. However, as
through processes made under controlled environment complexity of the precursor solution has grown up in recent
(controlled humidity and temperature and even in glovebox), years with inclusion of multiple cations and anions in the
and as controlling the environment in industrial scale may not system, the challenge of reproducibility remains in preparation
be cost-effective, it is desired to develop methods which are of polycrystalline films with same morphology, phase purity,
more tolerant/insensitive to ambient conditions. Two-step and chemical distribution by controlling the fabrication
method is one such option but it may need modifications to method. It has been found that aging of solution has a drastic
obtain films with morphology same as one-step deposition. In effect on film morphology and eventually on cell performance.
addition to environment, several other parameters also need to For instance, Tsai et al. witnessed a progressive increase in
be taken into account for establishing reliable and reproducible grain size and crystallinity of MAPbI3‑xClx perovskite thin films
methods. One important factor is solution recipe. Since with solution aging time.419 Their results suggested that once
solution properties (aging) affect the final cell performance the precursor solutions were aged for more than 24 h, the
dramatically, formulation of more stable solution/ink for crystallinity and grain-like features of perovskite dramatically
industrial methods is essential. improved. The device performance increased continuously
AS DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 53. Structure of a perovskite-Si tandem cell (a) exhibiting EQE spectral sensitivities of top and bottom cells (b) and current−voltage
characteristics (c). Reproduced with permission from ref 427. Copyright 2017 Springer Nature.

with solution aging time. A recent study on effect of solution view, solutions to most of the present problems might come
aging on performance of triple cation perovskite from solution itself. Therefore, understanding the solution
(Cs0.05(FA0.83MA0.17)0.95Pb(I0.83Br0.17)3) devices has rather chemistry of perovskite is going to be a key to improve
shown a dramatic change in performance with aging time.420 reproducibility. Development of a proper solution recipe based
Precursor solution that was aged longer than the optimum on this understanding is important for industrial manufacturing
time (6 h) degraded the phase purity of the resultant of PSCs.
perovskite films and led to poorer device performance, which In terms of device structure, HTM-free carbon-based PSCs
was attributed to the chemical inhomogeneity of the micron- are most attractive for industrial-scale manufacturing because
sized colloidal intermediates (solvated species) seen from the of their cost effectiveness and potential to avoid instability
the dynamic light scattering (DLS) measurements. This issues related to HTM and metal contacts. Although the
conclusion was further supported by the fact that this effect performance of carbon-based PSCs in large area is behind that
of aged solution was not observed for MAPbI3, in which, the made with HTM, potential of such structures are evident. PCE
aggregates were presumed to be chemically homogeneous. above 20% shown by an HTM-free carbon-based perovskite
Interestingly, Qin et al. recently found that a mixed cation cell made with doped-MAPbI3261 affirms that a matching
solution can be stabilized for long time, without affecting the combination of carbon and perovskite composition can
device performance, by incorporation of a small organic produce more efficient and stable PSCs even in large area.
molecule, ITIC-Th (Figure 52). 421 Notably, ITIC-Th Besides that, tandem and flexible PSCs are two promising
molecule suppressed formation of the undesired photo-inactive strategies for industrialization of PSCs. Recent progress and
δ-FAPbI 3 from the aged mixed cation perovskite present status of tandem and flexile PSCs are reviewed in
(MA0.17FA0.83Pb(I0.83Br0.17)3) solution and thus, resulted in sections below.
almost same performance even after aging up to 39 days 8.2. Perovskite Tandem Cells
(Figure 52).
In addition to such solution aging effects, solvents and The top efficiency of single-junction perovskite cell (>23%)34
additives used in the solution and the pH of the solution also has already surpassed top efficiencies of CIGS (copper indium
influence the morphology and crystallinity of the films. In the gallium selenide) and CdTe and will come in competition with
past few years, a large number and varieties of additives have the record efficiency of single crystalline silicon (25.8%) soon.
been employed in precursor solution and their effects on The characteristics like high PCE with VOC ≈ 1.15 V, steep
coverage, grain growth, crystallinity, traps passivation have absorption edge, easily tunable band gap (1.4∼2.3 eV), and
been well reported. Although understanding about the low-temperature (100−150 °C) processability of perovskites
mechanism of involvement of these additives and solvated make PSCs an attractive option for tandem cells. Moreover,
intermediates thereof in the crystallization process is gaining advantages from already established technologies like
inadequate at present, it is conspicuous that perovskite film Si and CIGS, perovskite is considered to find an easy way to be
formation can be influenced strongly by composition of the commercialized through tandem cells using a perovskite top
solution. Moreover, it is reasonable that these additives can cell and a Si or CIGS bottom cell. However, additional process
also affect the chemical distribution and electronic properties cost in making tandem structures should be compared with the
like carrier type and their density in the films. As there is not gain of efficiency over the top efficiency of a single cell.
much understanding on these facts, it is not easy to find a Furthermore, for stable operation of a tandem cell, some key
general correlation between the properties of additives and characteristics of top cell are ought to match with that of the
their effects on perovskite film. Other equally important but bottom cell. Rate of performance degradation (lifetime) and
not-well-investigated observations are influence of solution on dependence of voltage output on variation of light intensity
crystal orientation of the perovskite films and thermal strain in and temperature are two examples of such characteristics.
the films developed during the process of fabrication, which Crystalline Si solar cells exhibit large intensity422 and
subsequently results in altered performance and stability. temperature423 dependence of VOC (drop of VOC under weak
Therefore, it is becoming loud and clear that the influence of light) while perovskite cell generates relatively stable voltage
solution on the end results of PSCs can be significant. In our against light variation.424−426 Although VOC change under light
AT DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

and temperature variation will not seriously influence on the a certified PCE of 27.3% with a perovskite and crystalline Si
degradation of the cell and merely leads to loss in integrated tandem cell (active area, 1 cm2). The PCE value achieved
power generation, tandem cells with perovskite and crystalline exceeds the top PCE of single junction Si cell (25.8%) by 1.5%
Si may not be a stable power source for practical use due to and is more than the PCE (26.7%) achieved by a
unbalanced performance of top and bottom cells. Mechan- heterojunction Si cell (fabricated by Kaneka Co.); 27.3% is
ically, influence of such unbalanced performance can be considered to be close to the theoretical limit of PCE.
avoided by making four-terminal tandem cells in which powers For bottom cell, CIGS cells having tunable band gap (by
from top and bottom cells are independently collected. Taking changing composition) are also useful.429,430 CIGS has better
the life of perovskite cell, which is regarded not as high as Si characteristics than Si in terms of weak light responsivity that
and CIGS, into consideration, four-terminal structures are also can match the perovskite cell. Ideally, best combination for
safer approach to developing tandem solar cell, although its tandem cell must be the perovskite−perovskite junction in
cost is more than that of two-terminal structures. Apart from which top and bottom cells have balanced performance and
such practical issues, design of perovskite-based tandem cells the whole structure can be made by low cost printing process.
has scientific significance as a challenge to measure future Band gap tunability of perovskite materials enables flexible
potential of reachable efficiency. Because photon harvesting design for the kind of top and bottom cells. In reality, however,
efficiencies (or EQEs) of perovskite cell and candidate bottom efficiency of dual perovskite tandems is behind those of Si
cells (Si and CIGS) are already on the level close to the perovskite and CIGS perovskite tandems due to large voltage
theoretical limits, photocurrent density of tandem cell has little loss occurring in top and bottom cells. Snaith et al. made a 2-
room for improvement with the given top and bottom cells terminal tandem cell comprising narrow band gap FA0.75Cs0.25-
except for an optical loss due to transparent carrier transport Sn0.5Pb0.5I3 (1.2 eV, VOC = 0.74 V) and wide band gap
layer at the heterojunction. Therefore, reachable efficiency of a FA0.83Cs0.17Pb(I0.5Br0.5)3 (1.8 eV, VOC = 1.12 V). This cell
tandem cell depends on VOC values and fill factors (FFs) of top yielded PCE of 17% with VOC = 1.66 V389 while their four-
and bottom cells. In short, how much gain in voltage at terminal version using 1.6 eV top cell yielded a higher 20.3%
maximum power point is obtained at the cost of part of JSC PCE (VOC ≈ 1.7 V). Here, relatively large voltage loss in
from the bottom cell determines the ultimate efficiency of a narrow (<1.5 eV) and wide (>1.7 eV) gap perovskite cells is
tandem cell. To balance photocurrent density between top and limiting PCE of the tandem. Superior merit of small voltage
bottom cells, band gap (absorption edge wavelength) of loss in perovskite solar cell is special to the 1.6 eV band gap
perovskite top cell is adjusted by tuning halide composition so perovskite which achieves PCE > 21%.34,123,167,424,431 Jen et al.
that the numbers of photons absorbed by the top and bottom improved VOC of a tandem of similar perovskite combination
cells are equivalent. Theoretically, by sharing the solar (MAPb0.5Sn0.5I3 of 1.22 eV and MA0.9Cs0.1Pb(I0.6Br0.4)3 of 1.82
spectrum (photon flux) with perovskite top cell (1.6 eV eV) up to 1.98 V by suppressing recombination (voltage loss)
band gap) capable of VOC = 1.2 V and crystalline Si bottom cell due to interfacial contact, achieving 18.5% PCE.432 Other two-
(1. 1 eV band gap) capable of 0.8 V, PCE of a perovskite/ terminal tandems using MAPbI3 (1.6 eV) as bottom cell and
silicon tandem cell can reach close to 30%, depending on the Cs0.15FA0.85Pb(I0.3Br0.7)3 top cell achieved higher VOC = 2.3 V,
gain in VOC and FF.392,427 Figure 53 shows spectral sensitivity although PCE remains at 18%.433 As shown by these examples,
and current−voltage characteristics of a perovskite−Si tandem no dual-perovskite tandem cells have been able to exceed the
cell developed by McGehee’s group.427 Composition of best efficiency of single junction perovskite cell that works with
perovskite in the top cell was Cs0.17FA0.83Pb(Br0.17I 0.83)3 low voltage loss.34,123,167,424,431
(band gap of 1.63 eV), which was a good choice for being To estimate potentially reachable efficiency of a perovskite-
thermally more stable than MA-containing perovskites. This based tandem cell, Shockley−Queisser (SQ) limit of VOC is
tandem cell achieved a PCE of 23.6%, which is slightly higher considered most important. Taking an example of combination
than the best PCE (23.2%) of perovskite single junction cell. of MAPbI3 and CIGS (band gap 1.15 eV), SQ limits of VOC are
The top perovskite cell was of inverse structure, coated with assumed to be 1.27−1.32 V for MAPbI396,434 and 0.9 V for
PCBM as electron collector and transparent conductive top CIGS. If VOC of perovskite-CIGS tandem cell is realized to be
layer (ITO), yielding PCE close to 15%, which is lower than close to the sum of SQ limits of top and bottom subcells,
efficiency obtained by the perovskite single cell due to namely 2.1 V, we can estimate that PCE will reach a theoretical
influence of optical loss at the light incident side. Photocurrent limit around 33% (VOC = 2.1 V, JSC = 20 mA cm−2, FF = 0.8).
density was adjusted as equivalent (18.5−18.9 mA cm−2) And, in reality, PCE of 30% can be managed.
between top and bottom cells. VOC of the tandem was 1.65 V. The highest PCE of perovskite tandem cell is presently
As extension of this example and method, value of reachable obtained around 27% as achieved by Oxford PV, which is
PCE can be estimated on the basis of ideal values of JSC and several % more than the champion PCE of single perovskite
VOC. Minimizing optical loss (surface reflection and parasitic cell (∼23%). On the other hand, fabrication of tandem type
loss), JSC can reach 23 mA cm−2, a 23% increase. VOC can be cells and modules needs extra cost in materials and processes.
improved up to 1.95 V (18% increase) compared to 1.2 V for In practical use, different lifetimes between perovskite and the
perovskite cell and 0.75 V for Si cell. Such changes will result in bottom Si or CIGS cells can turn out to be an issue to be
PEC close to 27%. Duong et al. reported 26.4% PCE by using a solved for ensuring the cell stability. In this respect, question
4-terminal tandem cell with a mechanical stack of Rb- will arise if design of tandem cells is truly going to be cost-
containing quadruple cation perovskite (PCE of 17.4% with efficient as long as high voltage perovskite cell is employed.
band gap of 1.73 eV) and silicon cells (PCE of 23.9%).428 Merit of tandem cell design is to earn high VOC by combining
What rendered PCE higher than the above 2-terminal cell is cells of modest VOC (bearing large loss in VOC). However, this
the higher PCE of the Si cell and gain of VOC, which was 1.81 merit is less applicable in the case of PSCs that work at high
V (1.12 V of top, + 0.69 V of bottom). Besides challenges of VOC by itself. On the contrary, there is a more realistic example
these studies, a venture company Oxford PV recently reported that achieves PCE close to 29%. This is a single cell of GaAs
AU DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 54. Amorphous TiOx CL-based low-temperature processed perovskite (MAPbI3) solar cell (a) Surface morphology of ITO substrate
without TiOx layer and (b) with TiOx layer. (c) Cross-sectional SEM image of MAPbI3 perovskite solar cells with TiOx (thickness ∼8 nm)/
mesoporous brookite TiO2 electron collector. (d) J−V performance and (e) EQE spectrum of the device. (f) Dependence of PCE on the thickness
of amorphous TiOx CL. Reproduced with permission from ref 439. Copyright 2018 American Chemical Society.

(band gap 1.42 V, highest VOC = 1.12 V). Absorbing light up to by high temperature sintering process, or can be deposited by
900 nm, GaAs shows PCE up to 28.8%. SQ limit PCE of GaAs vacuum technologies. As an alternative method, thin metal
is around 33%.435 However, the GaAs device is highly oxide layers can be deposited at low temperature (without the
expensive compared to Si and manufacture of largearea GaAs requirement of high temperature sintering) via chemical
modules is not practical. In contrast, perovskite solar cell, now reactions involving dehydration−condensation reaction of
achieving PCE over 23%, does not have such drawback. It also precursors or metal oxide nanoparticles. This method has
has an exceptional potential in realizing high VOC, similar to been applied in our group to fabricate dye-sensitized solar cells
GaAs but with a cost much less than GaAs. Hence, a smart (DSSCs) on plastic substrates.19 For ETL for DSSCs, TiO2
strategy can be development of a single high-performance cell mesoporous layers are prepared either by dehydration−
that can compete the best performance of GaAs. In short, high- condensation reaction of TiO2 nanocrystals or by in situ
performance single cell of perovskite (toward PCE > 28%) can oxidation of titanium precursors such as TiCl4. The former
be more advantageous than tandem cell in terms of cost to yields a polycrystalline TiO2 layer while the latter forms a
performance. To catch up to GaAs, narrow band gap versions partially oxidized and amorphous TiOx layer. These low-
of perovskites have been synthesized by mixing Sn2+ to Pb2+, temperature-based ETLs with poorer conductivity compared
but unfortunately, they tend to suffer from increased VOC loss, to sintered films give relatively low efficiency in DSSCs because
leading to lower PCE values.323,390,436 Therefore, it is a highly a significantly thick TiO2 layer (several micrometers or more)
difficult goal for perovskite PV to create a high efficiency cell of required to load sufficient dye can have high electric resistance,
narrow band gap (absorption up to 900 nm) without losing the leading to lowering of the sensitized photocurrent. In contrast,
merit of high VOC output. Despite such difficulty, efforts in such drawback of this low-temperature method does not apply
compositional development for perovskite absorber and in PSCs because metal oxide ETL in perovskite cells is
optimization of interfacial connects are expected to find a significantly thin (<100 nm). Such thin films of metal oxide
breakthrough solution to the issue. provide the surface for perovskite crystallization as scaffold
and, as ETL, they do not necessarily convey all electrons from
8.3. Low-Temperature Process and Flexible Device
perovskite to the substrate because perovskite infiltrated in
Low-temperature (<150 °C) processing of all the active layers mesoporous TiO2 structure can transfer electrons. This is not
including perovskite in PSCs offers a noteworthy advantage for the case for non-porous TiO2 CL (without infiltrated
industrialization of the PSCs. In short, similar to organic thin- perovskite), which has to transport electron perfectly. In fact,
film PV devices, simple printing methods can be applied to a perovskite cell which uses mesoporous Al2O3 (insulator) as a
PSCs too. However, application of full printing process scaffold can exhibit good PV performance with PCE = 15−
requires all layers surrounding perovskite absorber, such as 17% and VOC = 1.07 V.24 Therefore, low-temperature TiO2-
ETM layer (ETL) and HTM layer (HTL), to be printed at low based PSCs are capable of high efficiency close to those of
temperature. Organic ETLs like fullerene derivatives and sintered TiO2-based solar cells. For instance, high efficiency
conductive polymers and HTLs like PEDOT−PSS immedi- and high VOC have been obtained with low-temperature-
ately meet this requirement. However, use of metal oxide prepared ZnO104 and SnO2437,438 as ETLs and scaffolds.
materials (semiconductors such as TiO2, SnO2, NiOx), which Special quality is required for a CL made from TiO2 that
are thermally more stable, is always preferred for long-term covers the surface of the conductive substrate (FTO, ITO,
stability of the cells. Generally, these metal oxide layers are etc.) to block back-electron transfer (recombination). The
prepared from metal precursors or nanoparticles of metal oxide quality and compactness of CL that blocks recombination
AV DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

determines VOC output of the cell. To prepare a pinhole-free possesses high photoconductive gain in photon to current
dense CL, we have applied an ultra-thin amorphous film of conversion. Several studies have clarified the rare ability of
TiOx (∼8 nm), prepared on the conductive ITO substrate by CH3NH3PbI3 as photodetector to achieve high gain of photo-
low-temperature synthesis. A mesoporous layer of brookite induced current in which quantum yield of photon to electron
type TiO2 (40 nm) was also coated as a scaffold on the CL at (current) far exceeds 100%.446−448 The perovskite-based
low temperature. As displayed in Figure 54d, the device using photodiode device is operated by applying a reverse bias in
MAPbI3 and spiro-OMeTAD HTL showed a best PCE of external circuit to enhance short-circuit photocurrent. The
21.6% (back scan) with high VOC up to 1.18 V.439 The device device is characterized as a photon-mode light sensor and an
performance was highly sensitive to the thickness of energy consuming device in which efforts to minimize the
amorphous layer of TiOx, which has high resistance, unlike operating voltage becomes important. Performance of photo-
conductive TiO2. Figure 54 shows morphologies and device detector is evaluated by measuring ratio of photocurrent and
performance of the low-temperature-processed MAPbI3 device dark current, namely, the S/N ratio in photon detection. Here,
and PCE value as a function of amorphous TiOx thickness. amplified output of photocurrent leads to increase of S/N
Tuning the quality of CL in terms of surface coverage and ratio. We found that the reverse-biased CH3NH3PbI3 cell (bias
electrical resistance is the key to enhance the device voltage, −0.3 to −0.9 V), which has a heterojunction structure
performance. Properties and qualities of CL and the contact basically similar to solar cells, is capable of generating a large
of CL and meso-TiO2 also influence recombination mecha- density photoconductive current (200−400 mA cm−2) under
nism116,424 and generation/suppression of hysteretic J−V exposure to light (100 mW cm−2).449 EQE values of maximal
performance.440,441 photocurrent exceeded 1700 under incidence of 100 mW cm−2
The low-temperature metal oxide coating method enables and striking increase in EQE occurs when the incident
fabrication of plastic substrate-based lightweight flexible intensity is reduced to 0.01 mW cm−2. EQE reached a value
perovskite devices. Using thickness-controlled amorphous more than 2400 times of 100%, indicating significant
TiO2 CLs, we fabricated a thin plastic film solar cell (thickness, amplification of photoresponse (gain value of 2400). Photo-
126 μm) using a mixed cation perovskite sensitivity (photoresponsivity) of such perovskite-based photo-
Csx(FA0.83MA0.17)(1‑x)Pb(I0.83Br0.17)3 as absorber, which yields detector in terms of generated current per incident power can
PCE up to 18% with VOC = 1.11 V.442 Because of the thin reach as high as 620 A W− under weak monochromatic light
layered structure of active layers (thickness around 1 μm), input power (550 nm) at 10 μW cm−2. This amplified
mechanical bending of the plastic device basically causes only a photocurrent increased S/N ratio of the device to more than
small strain between layers under curvature. This enables the 102. Figure 55a shows the responsivity characteristics of a
device to be durable against mechanical bending with high CH3NH3PbI3-based photodetector measured against intensity
curvature (1 cm diameter) up to 1,000 times, without of 550 nm light. We attributed the current amplification effect
substantial reduction in PV performance.442 Lightweight to photoconductivity of CH3NH3PbI3 polycrystalline which
flexible PV devices meet requirements of various small power can switch on by absorption of weak light and allow for flow of
devices in demand in an “Internet of Things” (IoT) society, large dark current (in addition to photocurrent) under reduced
which is showing quick growth in consumer electronics. The electric resistance. Namely, diode current is photo-switched in
PV device for use in IoT is required to have good light
harvesting ability under weak light including LED illumination.
Perovskite solar cells were found to have ability to maintain
high VOC against low-intensity light. For example, triple cation
perovskite-based device exhibited an ideality factor as low as
1.3, which indicates drop of VOC by decreasing light intensity is
sufficiently small (∼70 mV per one order of intensity).424 For
light harvesting applications, it is highly promising that energy-
saving type wireless communication devices are powered by
high efficiency (high voltage) perovskite PV cells in an indoor
weak-light environment.
8.4. Potential Applications in Optoelectronic Devices
With its excellent optoelectronic properties, perovskite finds a
number of interesting applications beyond photovoltaics,
namely, light-emitting diodes (LEDs), photodetectors, mem-
resistors, X-ray detectors, etc. Among them, LEDs have been
considered a major application, and intense studies have been
done on fabrication of perovskite-based LEDs.443 Perovskite
LEDs can be designed so that the high luminance device can
be operated by low driving voltage which has superior
advantage over the existing LED.444,445 Here, we are not
reviewing the progress of perovskites LEDs but focusing our
Figure 55. Photoresponsivity characteristics of CH3NH3PbI3-based
discussion on light-to-electric conversion devices such as photodetector under incidence of 550 nm light (a) and mechanism of
photodetector and X-ray detectors. High responsivity photo- amplified current generation due to photoconductivity switching
detecting devices or light sensors based on perovskites have behavior of devices with (b) and without (c) blocking CL.
been successfully fabricated. Similar to GaAs, CH3NH3PbI3 Reproduced with permission from ref 449. Copyright 2015 American
exhibits excellent performance as a photodiode which Chemical Society.

AW DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 56. A full-color perovskite detector. (a) Vertically stacked color sensor design, where the sensors function both as active detecting layers and
as long-pass filters for the underlying layers. (b) Measured light absorption of each perovskite used in a stack. Each layer in the detector absorbs and
filters incoming light. MAPbCl3 absorbs and filters out blue wavelengths, MAPbBr3 absorbs and filters out the remaining green light, and the
MAPb(Br/I)3 absorbs the remaining light from the visible spectrum. The IR fraction of light remains unabsorbed. (c) Schematics of stacked
photodetector and its electrical connection scheme. (d) Photograph of the prototype detector assembled from three perovskite single crystals
stacked on a chip carrier: MAPbCl3 on the top, MAPbBr3 in the middle and MAPb(Br/I)3 on the bottom. (e) Normalized photoconductivity
spectra of the individual single crystals in the stacked detector presented in (c). Reproduced with permission from ref 450. Copyright 2017 Springer
Nature.

Figure 57. An X-ray detector based on hybrid perovskite absorbers. (a) Illustration of an all-solution-processed digital X-ray detector. (b) Left, an
optical image of spin-cast polyimide (PI)-MAPbI3 on an a-Si:H TFT backplane. Inset, a single-pixel structure of TFT (scale bar 30 μ m) in which
the collection electrode (white outline) is connected to the drain contact of the TFT through a via (circular pad). Right, photograph of printed
MAPbI3 photoconductors on the PI-MAPbI3. (c) Charge collection and sensitivity characteristics of the detector measured at 100 kVp. The inset
shows W± in the pixelated (blue symbols) and diode (red symbols) detectors. (d) A hand phantom X-ray image obtained from a detector (using
100 kVp and 5 mGyair s−1 for 5 ms exposure, resulting in a dose of 25 μGyair and a bias voltage of 50 V).Reproduced with permission from ref 453.
Copyright 2017 Springer Nature.

amplified mode as a result of photoconductive response of the (Figure 55b,c). Advantage of such perovskite-based photo-
device. Furthermore, we verified that amplified current under detectors is its low voltage (<1 V) in operation. In commercial
reverse bias is only enabled by incorporating a pinhole-rich applications, the device can be powered by a single small
(bypass-rich) TiO2 compact layer (CL) between FTO and battery.
perovskite, which permits flow of large dark current, while Applications of perovskite photodetectors are extended to
highly dense CL exhibits no amplification of diode current color image sensors for use in digital cameras. For color image
AX DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 58. Proton irradiation tolerance of MAPbI3 (left) and mixed cation/halide perovskite (right) cells. Changes in photovoltaic characteristics
(JSC, VOC, Pmax, and PCE) are plotted against proton fluence. Values are normalized for unity at initial magnitudes. Reproduced with permission
from ref 456. Copyright 2018 Cell Press.

sensing, spectral sensitivity (absorption wavelength) of the perovskites still need large thickness as absorber. However,
perovskite device can be tuned by changing the halide anion solution processability of perovskite is expected to make
(Br, I) and cation group (organic or inorganic), which is absorber preparation much easier than the inorganic materials
similar to the spectral sensitivity tuning of silver halide grains which need high-temperature sintering, vacuum deposition,
(AgX, X = Cl, Br, I) in photographic science. Figure 56 shows etc. Needless to say, perovskites are much cheaper than
an example of such experiment.450 The study proposes an previous semiconductor materials. Utilizing the elemental
image sensor with a multilayered structure of blue-, green-, and nature of lead in absorber and detector of radiations (X-ray,
red-sensitive absorbers comprising chloride, bromide, and γ-ray, etc.) is highly promising for future commercialization.
iodide perovskites, respectively, which mimics the method of Although high sensitivity (high current response) has been
color photography using silver halides. Here, low-temperature corroborated, next subject of development is considered to
coating usability of perovskite materials may have a big merit ensure high S/N ratio in responsivity by suppressing dark
because underlying 2D pattern pixel electrode of current background response.
collection which is not robust against heat is not necessarily In connection to radiation sensing, space industry will be
exposed to high temperature in fabricating the image sensor. also a playground of perovskite optoelectronic devices.
These studies on the perovskite-based photodetector and Monitoring of radiation intensities is an important mission in
color-sensing device, taking advantage of low-cost, printable space exploration. However, device should have robust stability
light-sensing semiconductors, are moving toward R&D for against continuous exposure to high energy space irradiations.
commercial applications. This requirement becomes most strict for solar cells mounted
Other important application of perovskite-based detectors is on satellites. As power source, solar cells are also continuously
X-ray detection for use in medical diagnosis. This application is exposed to sunlight which can heat the device up to 100 °C.
highly promising because heavy lead-containing perovskite Therefore, we have conducted examination on space
materials have relatively high absorption coefficients for X-ray irradiation tolerance of PSCs as laboratory simulation by
radiation and use of lead in X-ray imaging can be accepted in using irradiation of high-energy electron and proton particle
diagnostic equipment as lead is already used there in X-ray beams. It is expected that defect tolerance nature of organo
diagnosis system. In X-ray imaging, generally employed lead halide perovskite semiconductors enables PSCs to survive
method is combination of scintillator (such as Tl-doped CsI) under exposure to high energy particles that are absorbed by
as light emitter and semiconductor photodetector (such as perovskite films. Several papers have reported relatively high
silicon) that detects light emitted by scintillator, which is an stability of perovskite materials (MAPbI3) against proton
indirect two-step detection system. Other method is a direct irradiation of particle energy tuned at 68 MeV.454,455 We have
method in which photodetector such as CdTe or amorphous investigated stability and particle radiation tolerance of PSCs
Se convert X-ray signal to current output. The direct system using MAPbI3 and mixed-cation/halide perovskites absorbers.
basically has merit for higher resolution of 2D image because Our study included thermal stability tests, before conducting
of none of optical crosstalk involved in indirect system. particle irradiation experiments, by exposing solar cells at high
However, previous direct methods with inorganic photo- and low temperatures of 100 and −80 °C, which correspond to
detectors have problems in poor absorptivity of X-rays. satellite orbit in hemispheres of globe with and without
Challenge to develop direct X-ray detector by using lead sunlight, respectively. This test revealed fragility of organic
halide perovskite absorbers have been made by several groups HTMs rather than the perovskite absorber. Spiro-OMeTAD
to show promising performance of the perovskite detec- deteriorated very quickly at 100 °C, while a polymer
tors.451,452 The most practical data of actual X-ray image conductor, P3HT, as HTM survived by keeping VOC of the
detection using 2D patterned image sensor was reported by cell. In proton and electron irradiations, both MAPbI3 and
research group of Samsung and Prof. Park.453 The detector was mixed-cation PSCs using P3HT HTM showed fairly high
a solution-processed perovskite device with a 0.83 mm thick X- tolerance to high fluence and large dose of proton (50 keV)
ray absorber layer as displayed in Figure 57. This device and electron (1 MeV), which can destroy crystalline Si- and
showed good sensitivity for X-ray detection, which was 1 order GaAs-based solar cells.456 Figure 58 shows how PSCs are
of magnitude higher than the sensitivities achieved with Ti- tolerant to fluence of proton irradiation up to 1014 particles
doped CsI-based indirect detectors and amorphous Se-based cm−2. Such robust durability of perovskite cells is enabled by
direct detectors. Because of absorptivity of X-rays, lead halide use of very thin absorber films (for visible light conversion)
AY DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Figure 59. (a) SQ PCE limit for a solar cell operated under AM 1.5G illumination at 298.15 K as a function of the band gap energy (Eg) and record
laboratory PCEs of different kinds of solar cells. (b) Current ratio (JSC/JSQ) versus the product of VOC and FF fraction (FF×VOC/FF×SQVSQ) for
the record efficiency solar cells. Reproduced with permission from ref 458. Copyright 2018 American Chemical Society.

which permit penetration of high energy particles without has made a significant leap in structural stability of perovskite.
causing collision damages. Needless to say, the defect tolerance However, a number of issues that have persisted throughout
property of perovskite semiconductors greatly contributes to these years of progress are long-term stability and thermal and
such rare performance in space environment. moisture stability. Instability of perovskite under humid
conditions, thermal stress, and illumination has been great
9. CONCLUSIONS AND FUTURE PROSPECTS concern. Hence, we anticipate more developments on the
It has been so good so far, but the question still remains, “Is it issues of long-term stability to happen in near future. In our
so far?”. The progress on all fronts of perovskite solar cells has views, and as the literature up to now suggests, the following
been tremendous so far, questions are still knocking on the challenges and strategies should be considered seriously for
doors of research laboratories: “When and how can it be taken further developments.
out of lab, and how far is it from commercial success?” 9.1. Further Improvement in PCE
Involvement of different companies (Saule Technologies, Shockley−Queisser (SQ) limit (theoretical PCE limit) of a
Toshiba, Oxford PV, Slliance, Solaronix, etc.) worldwide in
perovskite solar cell (Figure 59) using absorber with band gap,
scaling-up PSCs and their successful efforts already made in
Eg = 1.6 eV can be calculated to 30.14% with JSC = 25.47 mA/
fabricating largearea modules of good efficiency (12−14%) are
cm2, VOC = 1.309 V, FF = 90.5%.458 However, considering the
encouraging, but more efforts are needed to make perovskite
achievable FF (i.e., 85% not 90%), which is in reference to the
PV with a guaranteed commercial success. In fact, in
best case of single crystalline Si (85%), an efficiency close to
agreement to Henry’s views,457 we also affirm that, for further
developments toward commercialization of perovskite PV, 27% is more realistically achievable.459 Comparison of JSC,
interests and financial support in the forms of large-scale VOC, and FF of best efficiency cases (PCE = 22−23%) of PSCs
projects (not through small grants in aid) from both with that of SQ limits shows that, while JSC (>24 mA/cm2) is
government agencies and industries are much needed at the approaching close to the SQ limit (25.47 mA/cm2), VOC
present stage of the new technology. There are a number of (∼1.15 V) and FF (∼81%) are lower than the SQ limits
challenges, but it must be noted that, there has been no (i.e., 1.3 V and 90%) . Cells that show JSC lower than the SQ
technology which did not have issues and became a limit suffer from optical loss (inefficient light harvest) or
commercial success in just a decade. If the progress and incomplete carrier collection, whereas a reduced VOC or FF
major milestones are tracked from the beginning to the present implies undesired bulk or interfacial carrier recombination,
time, it can be rationalized that the first few years were invested parasitic resistance, or other electrical losses. It suggests that
largely in development of methods or process optimization to recombination losses in perovskite (Eg = 1.6 V) solar cells are
achieve high efficiency, which was followed by a lot of work possibly limiting the performance by limiting FF and VOC.
related to interface engineering addressing simultaneously high Therefore, identification of the recombination processes, both
efficiency and issues of anomalous hysteresis and reliability of in bulk and at interfaces, and passivation of the traps is believed
measured performance. Although maximum efficiency of the to push the PCE of PSCs further. Indeed, most of recent
cells had reached about 20% by the end of 2015, studies reporting high efficiency (20−23%) attribute enhance-
reproducibility, I−V hysteresis, reliability of measured perform- ment in PCE to suppressed recombination in some or other
ance, and stability remained worrisome issues then. Later, way. Inclusion of Cs, Rb, K into FA-MA mixed perovskites do
interface engineering in combination with high quality (i.e., not change the optical properties (absorption edge or band
pinhole free, large grains) perovskite films produced excellent gap) of mixed perovskites much, but it improves performance
results of minimizing or even eliminating hysteresis in cases. (FF and VOC) through passivation of defects and suppressed
However, structural or inherent intrinsic stability of perovskite recombination. Although mechanism of working of these ions
was still a bothering problem. In the recent years, among all toward passivation of defects is not entirely understood, Cs
developments, compositional engineering of perovskite (mixed and Rb inclusion in the FA-MA-based mixed perovskites
perovskites) that has concurrently improved efficiency and actually increases the unreacted FAI on the surface of the film
stability has been remarkable. Compositional tuning of and the latter becomes the actual reason for enhancement of
perovskites by mixing of cations (MA/FA/Cs) and anions VOC in the cells. It indicates that the effect of inclusion of these
(I/Br), which complemented with morphology of the films, cations in the precursor on distribution of the all component
AZ DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

ions (chemical homogeneity) in the film can be a more degradation even up to 1.5 year in 20−30% RH while the
important factor for the cell performance. We believe that films made by the widely used solution method degrades in few
further understanding of the roles of these ions in structure months. We find a direct relation between the differences in
evolution and chemical distribution can help in improving the crystal structures to the difference in such stability of films
PCE (∼ 27%) by minimizing recombination and increasing made by the two methods. We refrain from disclosing the
VOC and FF up to ∼1.3 V (against Eg = 1.6 V) and ∼85%, details here, as the investigation is still going on and is
respectively. intended to be reported in complete form elsewhere.
However, to achieve PCE beyond 27%, exceeding the level Nevertheless, it strongly suggests that the structural/intrinsic
of 28%, perovskite absorber with a narrow band gap (Eg) is instability can come from the solution method itself.
needed to gain on photocurrent. This is essentially aiming at 9.3. How To Increase Environmental Stability Further?
performance similar to GaAs (Eg = 1.41 eV), which is capable
of producing JSC of 28 mA/cm2. As a more realistic example, if Thermal stability of perovskites has doubtlessly improved by
the perovskite absorber with a similar Eg (1.4 eV) produces JSC using mixed cation perovskites (FA-MA-Cs). It is again related
of 28 mA/cm2 while maintaining the VOC and FF at 1.2 V and to the improved structural stability of such compositions.
85% respectively, PCE of 28.6% can be achieved. Although no Thermal decomposition of organic−inorganic lead halide
work has been so far succeeded in this direction, it is a highly perovskites depends on interaction/bonding of the cations
important challenge. In other words, as we have mentioned in (MA+, FA+) with the inorganic lead halide sub-lattice. It is
the section of tandem cells, efforts toward developing a known that weak interaction of such organic molecular ions
perovskite single junction solar cell as a low cost alternative to with the inorganic sub-lattice is responsible for easy thermal
GaAs should be encouraged and emphasized. As Sn-based degradation of these compounds. Therefore, strategies to
perovskites, although not as stable and efficient as Pb-based strengthen chemical interaction between such organic cations
perovskites, show lower band gaps (1.2∼1.3 eV), alloying of Sn and the inorganic part must be developed to increase the
with Pb in perovskites can be effective in tuning the band gap thermal stability further. Recently, Zhu et al. reported
to the SQ limit maximum (∼1.4 eV).460 A major challenge amazingly high chemical stability of MAPbI3 by divalent
arises during fabrication of such mixed Pb/Sn perovskites due anion (Se2−) doping, which is believed to enhance hydrogen-
to unstable Sn2+. Ways to stabilize Sn2+ in the Sn containing bond-like interactions between the organic cations and the
perovskites are being explored. Recent finding of the effect of inorganic framework.461 This study in fact becomes a starting
Ge2+ mixed in Sn2+ in stabilizing Sn-based perovskite under point for exploration of ions that can function in a way to
ambient air336 is a sign that there will be a method of making strengthen the chemical interaction between organic cations
stable Sn-alloy-based perovskites. If suitable methods to and the inorganic framework. Small cations and/or anions that
stabilize Sn2+ can be found then Pb−Sn perovskites will can occupy cuboctahedral site (A-site) along with the organic
become a material of choice. cation are expected to influence the ionic interaction between
the two sub-lattices.
9.2. How To Increase Intrinsic Stability Further? Superior thermal stability of all inorganic perovskites
For the purpose of commercialization, ensuring stability of (CsPbX3) up to above 400 °C makes them a strong alternative
PSCs becomes more important than pushing the PCE close to choice for future. Rapidly increasing interests in all-inorganic
the SQ limit. Although overall device stability is determined by perovskites in the recent years have made good progress on
the stability of ETMs and HTMs besides perovskites, intrinsic/ PCE. However, a more difficult challenge is to enhance
structural stability of perovskites is a prerequisite to improve stability against moisture and light, separately and in
overall stability. An ideal cubic perovskite structure is desired coexistence. 2D/3D mixed perovskites demonstrate greater
for better structural stability. Tolerance factor τ between 0.97 resistance against humidity. It is believed that the water
and 0.99, like in the case of triple cation (Cs/FA/MA)Pb(I/ repelling nature of larger organic cations used in 2D/3D mixed
Br)3, supports formation of such cubic perovskite structures perovskites is responsible for enhanced moisture resistance in
geometrically. Mixing of cations and anions has effectively the mixed dimensional perovskites. However, besides the
formed cubic perovskites and thereby raised the intrinsic water-repelling nature, we believe, such results must be
stability. As mixed perovskites have been quite extensively implicitly linked to structural/thermodynamic stability of
explored, it seems that further compositional changes without such mixed dimensional perovskites. Deeper investigation
compromising the optical and electronic properties of the and further understanding of nature of chemical interactions
resultant perovskite may not produce results of significance in between larger cations and the inorganic sub-lattice (PbI64−)
terms of ideal cubic structure. Instead, other sources of are needed. This can help in improving the moisture resistance
intrinsic instability, such as thermal strain generated between further. Also, investigation and development of materials for
substrate and the perovskite film during process of annealing, sealing/encapsulation of the cells should constitute an
possible strain in the crystals due to interphase boundaries in important step toward overall long-term stability. However,
phase impure cases, anisoptropic sensitivity of crystal faces to the issue which requires urgent attention is photo-induced
environment, etc., which have not been investigated well until degradation effects: photo-induced ion migration, phase
now, deserve serious attention. Is it possible to eliminate such segregation, lattice distortion, etc. Variation in extents of
bulk and interfacial strain? Can perovskite films with preferred such effects with composition of perovskite (cations and
crystal face orientation be grown? Fabrication of polycrystalline anions combination), like Cs inclusion in FA-MA mixed
films without strain and with crystals oriented along more air- perovskites slowing down photo-induced ion migration,
stable crystal facets is expected to raise the intrinsic stability of indicates that certain combination of ions might be robust to
films and cells substantially. In an ongoing work in our lab, we light. Studies on lattice-photon interaction can help in finding
are observing that MAPbI3 films made from solid-state mixing the right combination of such cations and anions. Such
of the precursor powders can remain stable without developments are doubtlessly important but performance
BA DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

degradation caused by other layers in the device has become an 9.5. Steps toward Commercialization
equally important issue and should be addressed with same Scaling up the processes of fabrication, reproducibility of
sincerity. Performance deterioration caused not by degradation results, and cost to performance assessment must be done to
of perovskite but due to undesired changes at the interfaces of bring the new technology close to commercialization. Although
perovskite/spiro-OMeTAD and deformation of spiro-OMe- tandem cells with a perovskite top cell and Si or CIGS bottom
TAD strongly indicates that replacement of organic HTMs like cells have been considered to be a closer possibility toward
spiro-OMeTAD is essential. Inorganic HTMs and HTM-free commercialization, material specificity and process complexity
carbon-electrode-based PSCs are promising. More develop- required for accomplishing the desired effective PCE in such
ments in such alternatives are expected to happen in near tandem cells needs to be critically reviewed. It seems that
future. Regarding all-inorganic cells, despite high thermal single-junction PSCs with efficiency above 24% and extended
stability, stabilization of the photoactive black phase of CsPbI3 long-term stability can be more cost-effective than tandem cells
at RT and in ambient conditions for long time is a big which may work at a PCE of 27−28%. Hence, more efforts
challenge.The effect of environment (air, humidity, and light) should be made in fabrication and scaling up of single-junction
on the formation and stability of the photoactive black phase PSCs with high efficiency, high yield, and long-term stability.
needs to be investigated well to help make further improve- Development of low-cost large-scale fabrication methods with
ments in all-inorganic PSCs. highly reproducible results is required for commercialization.
9.4. Potential of Lead-Free Perovskites In commercialization, influence of solution aging on the cell
performance should be minimized. Influence of solution aging
As toxicity of Pb stands as a potential threat for on the performance of cells indicates that compositional
commercialization of Pb-based perovskite PVs, quest for uniformity in the crystallized perovskite film varies with aging
suitable perovskite structures free of Pb or with a lower time, leading to even structural difference. Chemical structure
amount of Pb as an alternative to Pb perovskites gained of the colloids/clusters in the solution, which eventually grows
tremendous force. For all the Sn perovskite systems studied so into crystals in the film, can be the core origin of such
far, rapid oxidation of Sn2+ to Sn4+ and uncontrolled performance variation in final cells. Investigation of inter-ionic
crystallization of Sn perovskites from precursor solution interactions and chemical structure of the clusters/colloids in
resulting in poor film morphology appear to be two daunting the solution will provide critical information about crystal-
challenges restricting utilization of actual potential of Sn lization of perovskite and chemical distribution in the film. In
perovskites. In fact, instability of Sn perovskites in ambient fact, understanding nature of these clusters in the solution will
conditions becomes the reason for its slow progress in be important for addressing the challenges of reproducibility.
comparison to Pb-based perovskites. Although the mechanism All these investigations require in-depth study of solution
is not clear yet, some additives like SnF2, SnCl2, H3PO2 have chemistry of perovskites. We believe that most of the present
been found to improve the stability of Sn perovskites. challenges will be addressed in the coming years to help PSCs
However, the stability is still far behind that of Pb perovskites. to be commercialized.
It is considered that such additives compensate the missing/
oxidized Sn2+ by furnishing additional Sn2+ into the film. AUTHOR INFORMATION
However, in our view, the bonding interactions between the Corresponding Author
anionic counterparts (F−, Cl−, PO23−, etc.) from such additives
and Sn2+ in Sn perovskites need to be investigated for a better *E-mail: miyasaka@toin.ac.jp.
understanding. Electronegativity difference between F−, Cl−, ORCID
and I− can be believed to alter the bonding (overlapping) of Ajay Kumar Jena: 0000-0002-9279-5079
orbitals between Sn2+ and halides to eventually affect oxidation Ashish Kulkarni: 0000-0002-7945-208X
of Sn2+. Further understanding of role of such additives will Tsutomu Miyasaka: 0000-0001-8535-7911
certainly help in finding more suitable additives to stabilize the
Sn2+ and, thus, raise performance and stability further. Notes
Germanium perovskites also suffer from same problem of The authors declare no competing financial interest.
easy oxidation of Ge(II) state to Ge(IV) state. However,
bimetallic Sn-Ge and Pb-Sn perovskites have shown promising ACKNOWLEDGMENTS
results on performance and stability. In general, compositional Our studies discussed in this Review were financially supported
variations in Sn halide perovskites is expected to help in by Japan Science and Technology Agency (JST), Grant-in-Aid
improving the stability. Bismuth- and antimony-based lower for Scientific Research B from Japanese Society for Promotion
(A3M2I9; M = Bi3+, Sb3+) and higher (double perovskite) of Science (JSPS), and New Energy and Industrial Technology
dimensional perovskite materials show promising stability Development Organization (NEDO) of Japan. The authors
against humidity, heat, and continuous light exposure. thank Japan Aerospace Exploration Agency (JAXA) for
However, their efficiency has been limited (in most of the collaboration on the space tolerance study of perovskites.
reports) to ∼2% by indirect and wide band gap (∼2 eV). The authors acknowledge Prof. Xiao-Feng Wang of Jilin
Efforts are needed to reduce band gap of these perovskites. University, China, for his study on ZnO- and SnO2-based
Moreover, it seems that the widely used TiO2 layer and spiro- perovskite cells, Dr. Hsin-Wei Chen of Showa Shell Sekiyu
OMeTAD as n-type and p-type contacts in the devices with K.K. for his study on photodetectors, Prof. Tzu Chien Wei of
these Pb-free perovskites do not work well. Better under- National Tsing-Hua University, Taiwan, for his study on
standing of carrier dynamics at these interfaces and use of perovskite film preparation, Dr. Alessandra Alberti and her
suitable electron and hole extracting layers can be effective in group in IMM-CNR, Italy, for their investigation on the
improving the performance and stability of these Pb-free PSCs stability of perovskite materials, and Prof. Alexei Emeline for
further. his study on lead-free perovskite materials. The authors also
BB DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

thank Prof. Masashi Ikegami, Prof. Youhei Numata, Prof. (20) Ikeda, N.; Teshima, K.; Miyasaka, T. Conductive polymer−
Ayumi Ishii, Dr. Trilok Singh, and Dr. Yoshitaka Sanehira, Dr. carbon−imidazolium composite: a simple means for constructing
Peerathat Pinpithak, and Bhumika Chaudhary for their solid-state dye-sensitized solar cells. Chem. Commun. 2006, 16, 1733−
contributions to the study of perovskite device fabrication in 1735.
our group. (21) Miyasaka, T. Toward Printable Sensitized Mesoscopic Solar
Cells: Light-Harvesting Management with Thin TiO2 Films. J. Phys.
Chem. Lett. 2011, 2, 262−269.
REFERENCES (22) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Novel
(1) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Organometal Photoelectrochemical Cell with Mesoscopic Electrodes Sensitized by
Halide Perovskites as Visible-Light Sensitizers for Photovoltaic Cells. Lead-halide Compounds (5). Electrochem. Soc., 210th Meeting Abstr.
J. Am. Chem. Soc. 2009, 131, 6050−6051. 2007, MA2007-02, 352.
(2) Chakhmouradian, A. R.; Woodward, P. M. Celebrating 175 (23) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Novel
Years of Perovskite Research: A Tribute to Roger H. Mitchell. Phys. Photoelectrochemical Cell with Mesoscopic Electrodes Sensitized by
Chem. Miner. 2014, 41, 387−391. Lead-halide Compounds (11). Electrochem. Soc., 214th Meeting Abstr.
(3) Brenner, T. M.; Egger, D. A.; Kronik, L.; Hodes, G.; Cahen, D. 2008, MA2008-02, 27.
Hybrid Organic−inorganic Perovskites: Low-cost Semiconductors (24) Miyasaka, T. Perovskite Photovoltaics: Rare Functions of
with Intriguing Charge-transport Properties. Nat. Rev. Mater. 2016, 1 Organo Lead Halide in Solar Cells and Optoelectronic Devices. Chem.
(15007), 1−16. Lett. 2015, 44, 720−729.
(4) Yuan, Y.; Xiao, Z.; Yang, B.; Huang, J. Arising Applications of (25) Im, J.-H.; Lee, C.-R.; Lee, J.-W.; Park, S.-W.; Park, N.-G. 6.5%
Ferroelectric Materials in Photovoltaic Devices. J. Mater. Chem. A Efficient Perovskite Quantum-dot-sensitized Solar Cell. Nanoscale
2014, 2, 6027−6041. 2011, 3, 4088−4093.
(5) Wells, H. L. Ü ber die Cäsium- und Kalium-Bleihalogenide. (26) Murakami, T. N.; Ito, S.; Wang, Q.; Nazeeruddin, M. K.;
Zeitschrift für anorganische Chemie 1893, 3, 195−210. Bessho, T.; Cesar, I.; Liska, P.; Humphry-Baker, R.; Comte, P.; Péchy,
(6) Weber, D. CH3NH3PbX3, ein Pb(II)-System mit kubischer P.; et al. Highly Efficient Dye-Sensitized Solar Cells Based on Carbon
Perowskitstruktur/CH3NH3PbX3, a Pb(II)-System with Cubic Per-
Black Counter Electrodes. J. Electrochem. Soc. 2006, 153, A2255−
ovskite Structure. Z. Naturforsch., B: J. Chem. Sci. 1978, 33, 1443−
A2261.
1445.
(27) Snaith, H. J.; Petrozza, A.; Ito, S.; Miura, H.; Grätzel, M.
(7) Weber, D. CH3NH3SnBrxI3‑x (x = 0−3), ein Sn(II)-System mit
kubischer Perowskitstruktur/CH3NH3SnBrxI3‑x(x = 0−3), a Sn(II)- Charge Generation and Photovoltaic Operation of Solid-State Dye-
System with Cubic Perovskite Structure. Z. Naturforsch., B: J. Chem. Sensitized Solar Cells Incorporating a High Extinction Coefficient
Sci. 1978, 33, 862−865. Indolene-Based Sensitizer. Adv. Funct. Mater. 2009, 19, 1810−1818.
(8) Møller, C. K. A Phase Transition in Cæsium Plumbochloride. (28) Snaith, H. J.; Humphry-Baker, R.; Chen, P.; Cesar, I.;
Nature 1957, 180, 981−982. Zakeeruddin, S. M.; Grätzel, M. Charge Collection and Pore Filling
(9) Møller, C. K. Crystal Structure and Photoconductivity of in Solid-state Dye-sensitized Solar Cells. Nanotechnology 2008, 19,
Cæsium Plumbohalides. Nature 1958, 182, 1436−1436. 424003.
(10) Mitzi, D. B. Synthesis, Structure, and Properties of Organic- (29) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.;
Inorganic Perovskites and Related Materials. Progress in Inorganic Snaith, H. J. Efficient Hybrid Solar Cells Based on Meso-Super-
Chemistry; John Wiley and Sons, 1999; Vol. 48, pp 1−122. structured Organometal Halide Perovskites. Science 2012, 338, 643−
(11) Mitzi, D. B. Organic−Inorganic Perovskites Containing 647.
Trivalent Metal Halide Layers: The Templating Influence of the (30) Kojima, A.; Ikegami, M.; Teshima, K.; Miyasaka, T. Highly
Organic Cation Layer. Inorg. Chem. 2000, 39, 6107−6113. Luminescent Lead Bromide Perovskite Nanoparticles Synthesized
(12) Mitzi, D. B.; Wang, S.; Feild, C. A.; Chess, C. A.; Guloy, A. M. with Porous Alumina Media. Chem. Lett. 2012, 41, 397−399.
Conducting Layered Organic-inorganic Halides Containing < 110>- (31) Stranks, S. D.; Eperon, G. E.; Grancini, G.; Menelaou, C.;
Oriented Perovskite Sheets. Science 1995, 267, 1473. Alcocer, M. J. P.; Leijtens, T.; Herz, L. M.; Petrozza, A.; Snaith, H. J.
(13) Era, M.; Oka, S. PbBr-based layered perovskite film using the Electron-Hole Diffusion Lengths Exceeding 1 Micrometer in an
Langmuir−Blodgett technique. Thin Solid Films 2000, 376, 232−235. Organometal Trihalide Perovskite Absorber. Science 2013, 342, 341−
(14) Takeoka, Y.; Asai, K.; Rikukawa, M.; Sanui, K. Incorporation of 342.
Conjugated Polydiacetylene Systems into Organic−inorganic Quan- (32) Wehrenfennig, C.; Eperon, G. E.; Johnston, M. B.; Snaith, H. J.;
tum-well Structures. Chem. Commun. 2001, 2592−2593. Herz, L. M. High Charge Carrier Mobilities and Lifetimes in
(15) Kondo, T.; Iwamoto, S.; Hayase, S.; Tanaka, K.; Ishi, J.; Organolead Trihalide Perovskites. Adv. Mater. 2014, 26, 1584−1589.
Mizuno, M.; Ema, K.; Ito, R. Resonant Third-order Optical (33) Dong, Q.; Fang, Y.; Shao, Y.; Mulligan, P.; Qiu, J.; Cao, L.;
Nonlinearity in The Layered Perovskite-type Material Huang, J. Electron-hole Diffusion Lengths > 175 μm in Solution-
(C6H13NH3)2PbI4. Solid State Commun. 1998, 105, 503−506.
grown CH3NH3PbI3 Single Crystals. Science 2015, 347, 967−969.
(16) Tanaka, K.; Kondo, T. Bandgap and Exciton Binding Energies
(34) Jeon, N. J.; Na, H.; Jung, E. H.; Yang, T.-Y.; Lee, Y. G.; Kim,
in Lead-iodide-based Natural quantum-well Crystals. Sci. Technol. Adv.
G.; Shin, H.-W.; Il Seok, S.; Lee, J.; Seo, J. A Fluorene-terminated
Mater. 2003, 4, 599−604.
(17) Ema, K.; Umeda, K.; Toda, M.; Yajima, C.; Arai, Y.; Kunugita, Hole-transporting Material for Highly Efficient and Stable Perovskite
H.; Wolverson, D.; Davies, J. J. Huge Exchange Energy and Fine Solar Cells. Nat. Energy 2018, 3, 682−689.
Structure of Excitons in an Organic-inorganic Quantum Well (35) Suarez, B.; Gonzalez-Pedro, V.; Ripolles, T. S.; Sanchez, R. S.;
Material. Phys. Rev. B: Condens. Matter Mater. Phys. 2006, 73, 241310. Otero, L.; Mora-Sero, I. Recombination Study of Combined Halides
(18) Ishi, J.; Mizuno, M.; Kunugita, H.; Ema, K.; Iwamoto, S.; (Cl, Br, I) Perovskite Solar Cells. J. Phys. Chem. Lett. 2014, 5, 1628−
Hayase, S.; Kondo, T.; Ito, R. Third-Order Optical Nonlinearity Due 1635.
to Excitons and Biexcitons in a Self-Organized Quantum-Well (36) Eperon, G. E.; Stranks, S. D.; Menelaou, C.; Johnston, M. B.;
Material (C6H13NH3)2PbI4. J. Nonlinear Opt. Phys. Mater. 1998, 07, Herz, L. M.; Snaith, H. J. Formamidinium Lead Trihalide: A Broadly
153−159. Tunable Perovskite for Efficient Planar Heterojunction Solar Cells.
(19) Miyasaka, T.; Ikegami, M.; Kijitori, Y. Photovoltaic Perform- Energy Environ. Sci. 2014, 7, 982−988.
ance of Plastic Dye-Sensitized Electrodes Prepared by Low-Temper- (37) Miyano, K.; Tripathi, N.; Yanagida, M.; Shirai, Y. Lead Halide
ature Binder-Free Coating of Mesoscopic Titania. J. Electrochem. Soc. Perovskite Photovoltaic as a Model p−i−n Diode. Acc. Chem. Res.
2007, 154, A455−A461. 2016, 49, 303−310.

BC DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

(38) Miyasaka, T. Lead Halide Perovskites in Thin Film Photo- (56) Jiang, C.-S.; Yang, M.; Zhou, Y.; To, B.; Nanayakkara, S. U.;
voltaics: Background and Perspectives. Bull. Chem. Soc. Jpn. 2018, 91, Luther, J. M.; Zhou, W.; Berry, J. J.; van de Lagemaat, J.; Padture, N.
1058−1068. P.; et al. Carrier Separation and Transport in Perovskite Solar Cells
(39) Tanaka, K.; Takahashi, T.; Ban, T.; Kondo, T.; Uchida, K.; Studied by Nanometre-scale Profiling of Electrical Potential. Nat.
Miura, N. Comparative Study on The Excitons in Lead-halide-based Commun. 2015, 6, 8397.
Perovskite-type Crystals CH3NH3PbBr3 CH3NH3PbI3. Solid State (57) Wu, X.; Tan, L. Z.; Shen, X.; Hu, T.; Miyata, K.; Trinh, M. T.;
Commun. 2003, 127, 619−623. Li, R.; Coffee, R.; Liu, S.; Egger, D. A.; et al. Light-induced
(40) Brivio, F.; Butler, K. T.; Walsh, A.; van Schilfgaarde, M. Picosecond Rotational Disordering of The Inorganic Sublattice in
Relativistic Quasiparticle Self-consistent Electronic Structure of Hybrid Perovskites. Sci. Adv. 2017, 3, No. e1602388.
Hybrid Halide Perovskite Photovoltaic Absorbers. Phys. Rev. B: (58) Bischak, C. G.; Hetherington, C. L.; Wu, H.; Aloni, S.;
Condens. Matter Mater. Phys. 2014, 89, 155204. Ogletree, D. F.; Limmer, D. T.; Ginsberg, N. S. Origin of Reversible
(41) Yin, W.-J.; Shi, T.; Yan, Y. Superior Photovoltaic Properties of Photoinduced Phase Separation in Hybrid Perovskites. Nano Lett.
Lead Halide Perovskites: Insights from First-Principles Theory. J. 2017, 17, 1028−1033.
Phys. Chem. C 2015, 119, 5253−5264. (59) Neukirch, A. J.; Nie, W.; Blancon, J.-C.; Appavoo, K.; Tsai, H.;
(42) Green, M. A.; Ho-Baillie, A.; Snaith, H. J. The Emergence of Sfeir, M. Y.; Katan, C.; Pedesseau, L.; Even, J.; Crochet, J. J.; et al.
Perovskite Solar Cells. Nat. Photonics 2014, 8, 506−514. Polaron Stabilization by Cooperative Lattice Distortion and Cation
(43) Giorgi, G.; Fujisawa, J.-I.; Segawa, H.; Yamashita, K. Small Rotations in Hybrid Perovskite Materials. Nano Lett. 2016, 16, 3809−
Photocarrier Effective Masses Featuring Ambipolar Transport in 3816.
Methylammonium Lead Iodide Perovskite: A Density Functional (60) Miyata, K.; Meggiolaro, D.; Trinh, M. T.; Joshi, P. P.; Mosconi,
Analysis. J. Phys. Chem. Lett. 2013, 4, 4213−4216. E.; Jones, S. C.; De Angelis, F.; Zhu, X. Y. Large Polarons in Lead
(44) de Quilettes, D. W.; Vorpahl, S. M.; Stranks, S. D.; Nagaoka, Halide Perovskites. Sci. Adv. 2017, 3, No. e1701217.
H.; Eperon, G. E.; Ziffer, M. E.; Snaith, H. J.; Ginger, D. S. Impact of (61) Zhu, H.; Miyata, K.; Fu, Y.; Wang, J.; Joshi, P. P.; Niesner, D.;
Microstructure on Local Carrier Lifetime in Perovskite Solar Cells. Williams, K. W.; Jin, S.; Zhu, X. Y. Screening in Crystalline Liquids
Science 2015, 348, 683−686. Protects Energetic Carriers in Hybrid Perovskites. Science 2016, 353,
(45) Stranks, S. D.; Snaith, H. J. Metal-halide Perovskites for 1409−1413.
Photovoltaic and Light-emitting Devices. Nat. Nanotechnol. 2015, 10, (62) Guzelturk, B.; Belisle, R. A.; Smith, M. D.; Bruening, K.;
391−402. Prasanna, R.; Yuan, Y.; Gopalan, V.; Tassone, C. J.; Karunadasa, H. I.;
(46) Liu, Z.; Krückemeier, L.; Krogmeier, B.; Klingebiel, B.; McGehee, M. D.; Lindenberg, A. M.; et al. Terahertz Emission from
Márquez, J. A.; Levcenko, S.; Ö z, S.; Mathur, S.; Rau, U.; Unold, Hybrid Perovskites Driven by Ultrafast Charge Separation and Strong
T.; et al. Open-Circuit Voltages Exceeding 1.26 V in Planar Electron−Phonon Coupling. Adv. Mater. 2018, 30, 1704737.
Methylammonium Lead Iodide Perovskite Solar Cells. ACS Energy (63) Jeon, N. J.; Noh, J. H.; Kim, Y. C.; Yang, W. S.; Ryu, S.; Seok, S.
I. Solvent Engineering for High-performance Inorganic−organic
Lett. 2019, 4, 110−117.
(47) Shi, D.; Adinolfi, V.; Comin, R.; Yuan, M.; Alarousu, E.; Buin, Hybrid Perovskite Solar Cells. Nat. Mater. 2014, 13, 897−903.
(64) Xiao, M.; Huang, F.; Huang, W.; Dkhissi, Y.; Zhu, Y.;
A.; Chen, Y.; Hoogland, S.; Rothenberger, A.; Katsiev, K.; et al. Low
Etheridge, J.; Gray-Weale, A.; Bach, U.; Cheng, Y.-B.; Spiccia, L. A
trap-state Density and Long Carrier Diffusion in Organolead
Fast Deposition-Crystallization Procedure for Highly Efficient Lead
Trihalide Perovskite Single Crystals. Science 2015, 347, 519−522.
Iodide Perovskite Thin-Film Solar Cells. Angew. Chem. 2014, 126,
(48) Stranks, S. D.; Burlakov, V. M.; Leijtens, T.; Ball, J. M.; Goriely,
10056−10061.
A.; Snaith, H. J. Recombination Kinetics in Organic-Inorganic
(65) Liu, D.; Wu, L.; Li, C.; Ren, S.; Zhang, J.; Li, W.; Feng, L.
Perovskites: Excitons, Free Charge, and Subgap States. Phys. Rev.
Controlling CH3NH3PbI3−xClx Film Morphology with Two-Step
Appl. 2014, 2, No. 034007. Annealing Method for Efficient Hybrid Perovskite Solar Cells. ACS
(49) Xing, G.; Mathews, N.; Sun, S.; Lim, S. S.; Lam, Y. M.; Grätzel, Appl. Mater. Interfaces 2015, 7, 16330−16337.
M.; Mhaisalkar, S.; Sum, T. C. Long-Range Balanced Electron- and (66) Liu, J.; Gao, C.; He, X.; Ye, Q.; Ouyang, L.; Zhuang, D.; Liao,
Hole-Transport Lengths in Organic-Inorganic CH3NH3PbI3. Science C.; Mei, J.; Lau, W. Improved Crystallization of Perovskite Films by
2013, 342, 344−347. Optimized Solvent Annealing for High Efficiency Solar Cell. ACS
(50) Bi, Y.; Hutter, E. M.; Fang, Y.; Dong, Q.; Huang, J.; Savenije, T. Appl. Mater. Interfaces 2015, 7, 24008−24015.
J. Charge Carrier Lifetimes Exceeding 15 μs in Methylammonium (67) Yu, H.; Liu, X.; Xia, Y.; Dong, Q.; Zhang, K.; Wang, Z.; Zhou,
Lead Iodide Single Crystals. J. Phys. Chem. Lett. 2016, 7, 923−928. Y.; Song, B.; Li, Y. Room-temperature Mixed-solvent-vapor Annealing
(51) Chouhan, A. S.; Jasti, N. P.; Hadke, S.; Raghavan, S.; Avasthi, S. for High Performance Perovskite Solar Cells. J. Mater. Chem. A 2016,
Large Grained and High Charge Carrier Lifetime CH3NH3PbI3 Thin- 4, 321−326.
films: Implications for Perovskite Solar Cells. Curr. Appl. Phys. 2017, (68) Xiao, S.; Bai, Y.; Meng, X.; Zhang, T.; Chen, H.; Zheng, X.; Hu,
17, 1335−1340. C.; Qu, Y.; Yang, S. Unveiling a Key Intermediate in Solvent Vapor
(52) Johnston, M. B.; Herz, L. M. Hybrid Perovskites for Postannealing to Enlarge Crystalline Domains of Organometal Halide
Photovoltaics: Charge-Carrier Recombination, Diffusion, and Radia- Perovskite Films. Adv. Funct. Mater. 2017, 27, 1604944.
tive Efficiencies. Acc. Chem. Res. 2016, 49, 146−154. (69) Mabrouk, S.; Bahrami, B.; Gurung, A.; Reza, K. M.; Adhikari,
(53) Ponseca, C. S.; Savenije, T. J.; Abdellah, M.; Zheng, K.; Yartsev, N.; Dubey, A.; Pathak, R.; Yang, S.; Qiao, Q. Higher Efficiency
A.; Pascher, T.; Harlang, T.; Chabera, P.; Pullerits, T.; Stepanov, A.; Perovskite Solar Cells Using Additives of LiI, LiTFSI and BMImI in
et al. Organometal Halide Perovskite Solar Cell Materials the PbI2 Precursor. Sustainable Energy Fuels 2017, 1, 2162−2171.
Rationalized: Ultrafast Charge Generation, High and Microsecond- (70) You, S.; Bi, S.; Huang, J.; Jia, Q.; Yuan, Y.; Xia, Y.; Xiao, Z.;
Long Balanced Mobilities, and Slow Recombination. J. Am. Chem. Soc. Sun, Z.; Liu, J.; Sun, S.; et al. Additive-Enhanced Crystallization of
2014, 136, 5189−5192. Solution Process for Planar Perovskite Solar Cells with Efficiency
(54) Frost, J. M.; Butler, K. T.; Brivio, F.; Hendon, C. H.; van Exceeding 19%. Chem. - Eur. J. 2017, 23, 18140−18145.
Schilfgaarde, M.; Walsh, A. Atomistic Origins of High-Performance in (71) Kim, M.; Kim, G.-H.; Oh, K. S.; Jo, Y.; Yoon, H.; Kim, K.-H.;
Hybrid Halide Perovskite Solar Cells. Nano Lett. 2014, 14, 2584− Lee, H.; Kim, J. Y.; Kim, D. S. High-Temperature−Short-Time
2590. Annealing Process for High-Performance Large-Area Perovskite Solar
(55) Tan, L. Z.; Zheng, F.; Young, S. M.; Wang, F.; Liu, S.; Rappe, Cells. ACS Nano 2017, 11, 6057−6064.
A. M. Shift Current Bulk Photovoltaic Effect in Polar Materials− (72) Cao, Q.; Yang, S.; Gao, Q.; Lei, L.; Yu, Y.; Shao, J.; Liu, Y. Fast
hybrid and Oxide Perovskites and Beyond. Npj Comput. Mater. 2016, and Controllable Crystallization of Perovskite Films by Microwave
2, 16026. Irradiation Process. ACS Appl. Mater. Interfaces 2016, 8, 7854−7861.

BD DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

(73) Lavery, B. W.; Druffel, T.; Draper, G.; Sunkara, M. K. Intense (90) Almosni, S.; Cojocaru, L.; Li, D.; Uchida, S.; Kubo, T.; Segawa,
Pulsed Light Annealing of Perovskite Solar Cells. Electrochem. Soc., H. Tunneling-Assisted Trapping as One of the Possible Mechanisms
Meeting Abstr. 2015, MA2015-02, 801. for the Origin of Hysteresis in Perovskite Solar Cells. Energy Technol.
(74) Li, Y.; Ji, L.; Liu, R.; Zhang, C.; Mak, C. H.; Zou, X.; Shen, H.- 2017, 5, 1767−1774.
H.; Leu, S.-Y.; Hsu, H.-Y. A Review on Morphology Engineering for (91) Li, C.; Guerrero, A.; Zhong, Y.; Huettner, S. Origins and
Highly Efficient and Stable Hybrid Perovskite Solar Cells. J. Mater. Mechanisms of Hysteresis in Organometal Halide Perovskites. J.
Chem. A 2018, 6, 12842−12875. Phys.: Condens. Matter 2017, 29, 193001.
(75) Salim, T.; Sun, S.; Abe, Y.; Krishna, A.; Grimsdale, A. C.; Lam, (92) Song, D. H.; Jang, M. H.; Lee, M. H.; Heo, J. H.; Park, J. K.;
Y. M. Perovskite-based Solar Cells: Impact of Morphology and Device Sung, S.-J.; Kim, D.-H.; Hong, K.-H.; Im, S. H. A Discussion on the
Architecture on Device Performance. J. Mater. Chem. A 2015, 3, Origin and Solutions of Hysteresis in Perovskite Hybrid Solar Cells. J.
8943−8969. Phys. D: Appl. Phys. 2016, 49 (47), 473001.
(76) Zhou, Z.; Pang, S.; Liu, Z.; Xu, H.; Cui, G. Interface (93) Brenes, R.; Guo, D.; Osherov, A.; Noel, N. K.; Eames, C.;
Engineering for High-performance Perovskite Hybrid Solar Cells. J. Hutter, E. M.; Pathak, S. K.; Niroui, F.; Friend, R. H.; Islam, M. S.;
Mater. Chem. A 2015, 3, 19205−19217. et al. Metal Halide Perovskite Polycrystalline Films Exhibiting
(77) Cho, A.-N.; Park, N.-G. Impact of Interfacial Layers in Properties of Single Crystals. Joule 2017, 1, 155−167.
Perovskite Solar Cells. ChemSusChem 2017, 10, 3687−3704. (94) Zhou, L.; Chang, J.; Liu, Z.; Sun, X.; Lin, Z.; Chen, D.; Zhang,
(78) Heo, J. H.; Han, H. J.; Kim, D.; Ahn, T. K.; Im, S. H. C.; Zhang, J.; Hao, Y. Enhanced Planar Perovskite Solar Cell
Hysteresis-less Inverted CH3NH3PbI3 Planar Perovskite Hybrid Solar Efficiency and Stability using a Perovskite/PCBM Heterojunction
Cells with 18.1% Power Conversion Efficiency. Energy Environ. Sci. formed in One Step. Nanoscale 2018, 10, 3053−3059.
2015, 8, 1602−1608. (95) Yu, J. C.; Hong, J. A.; Jung, E. D.; Kim, D. B.; Baek, S.-M.; Lee,
(79) Yoon, H.; Kang, S. M.; Lee, J.-K.; Choi, M. Hysteresis-free S.; Cho, S.; Park, S. S.; Choi, K. J.; Song, M. H. Highly efficient and
Low-temperature-processed Planar Perovskite Solar Cells with 19.1% stable inverted perovskite solar cell employing PEDOT:GO
Efficiency. Energy Environ. Sci. 2016, 9, 2262−2266. composite layer as a hole transport layer. Sci. Rep. 2018, 8 (1), 1070.
(80) Wang, Y.-C.; Li, X.; Zhu, L.; Liu, X.; Zhang, W.; Fang, J. (96) Park, N.-G.; Grätzel, M.; Miyasaka, T.; Zhu, K.; Emery, K.
Efficient and Hysteresis-Free Perovskite Solar Cells Based on a Towards Stable and Commercially Available Perovskite Solar Cells.
Solution Processable Polar Fullerene Electron Transport Layer. Adv. Nat. Energy 2016, 1, 16152.
Energy Mater. 2017, 7, 1701144. (97) Pathak, S.; Sepe, A.; Sadhanala, A.; Deschler, F.; Haghighirad,
(81) Cai, F.; Yang, L.; Yan, Y.; Zhang, J.; Qin, F.; Liu, D.; Cheng, Y.- A.; Sakai, N.; Goedel, K. C.; Stranks, S. D.; Noel, N.; Price, M.; et al.
B.; Zhou, Y.; Wang, T. Eliminated Hysteresis and Stabilized Power Atmospheric Influence upon Crystallization and Electronic Disorder
Output Over 20% in Planar Heterojunction Perovskite Solar Cells by and Its Impact on the Photophysical Properties of Organic−Inorganic
Compositional and Surface Modifications to The Low-temperature- Perovskite Solar Cells. ACS Nano 2015, 9, 2311−2320.
processed TiO2 Layer. J. Mater. Chem. A 2017, 5, 9402−9411. (98) Jung, K.-H.; Seo, J.-Y.; Lee, S.; Shin, H.; Park, N.-G. Solution-
(82) Zhou, Y.-Q.; Wu, B.-S.; Lin, G.-H.; Xing, Z.; Li, S.-H.; Deng, L.- processed SnO2 Thin Film for a Hysteresis-free Planar Perovskite
L.; Chen, D.-C.; Yun, D.-Q.; Xie, S.-Y. Interfacing Pristine C60 onto Solar Cell with a Power Conversion Efficiency of 19.2%. J. Mater.
TiO2 for Viable Flexibility in Perovskite Solar Cells by a Low- Chem. A 2017, 5, 24790−24803.
Temperature All-Solution Process. Adv. Energy Mater. 2018, 8, (99) Jiang, Q.; Zhang, X.; You, J. SnO2: A Wonderful Electron
1800399. Transport Layer for Perovskite Solar Cells. Small 2018, 14, 1801154.
(83) Wojciechowski, K.; Stranks, S. D.; Abate, A.; Sadoughi, G.; (100) Zhang, P.; Wu, J.; Zhang, T.; Wang, Y.; Liu, D.; Chen, H.; Ji,
Sadhanala, A.; Kopidakis, N.; Rumbles, G.; Li, C.-Z.; Friend, R. H.; L.; Liu, C.; Ahmad, W.; Chen, Z. D.; et al. Perovskite Solar Cells with
Jen, A. K. Y.; et al. Heterojunction Modification for Highly Efficient ZnO Electron-Transporting Materials. Adv. Mater. 2018, 30, 1703737.
Organic−Inorganic Perovskite Solar Cells. ACS Nano 2014, 8, (101) An, Q.; Fassl, P.; Hofstetter, Y. J.; Becker-Koch, D.; Bausch,
12701−12709. A.; Hopkinson, P. E.; Vaynzof, Y. High Performance Planar Perovskite
(84) Richardson, G.; O’Kane, S. E. J.; Niemann, R. G.; Peltola, T. A.; Solar Cells by ZnO Electron Transport Layer Engineering. Nano
Foster, J. M.; Cameron, P. J.; Walker, A. B. Can Slow-moving Ions Energy 2017, 39, 400−408.
Explain Hysteresis in the Current−voltage Curves of Perovskite Solar (102) Correa Baena, J. P.; Steier, L.; Tress, W.; Saliba, M.; Neutzner,
Cells? Energy Environ. Sci. 2016, 9, 1476−1485. S.; Matsui, T.; Giordano, F.; Jacobsson, T. J.; Srimath Kandada, A. R.;
(85) Zhang, Y.; Liu, M.; Eperon, G. E.; Leijtens, T. C.; McMeekin, Zakeeruddin, S. M.; et al. Highly Efficient Planar Perovskite Solar
D.; Saliba, M.; Zhang, W.; de Bastiani, M.; Petrozza, A.; Herz, L. M.; Cells through Band Alignment Engineering. Energy Environ. Sci. 2015,
et al. Charge Selective Contacts, Mobile Ions and Anomalous 8, 2928−2934.
Hysteresis in Organic−inorganic Perovskite Solar Cells. Mater. Horiz. (103) Song, J.; Zheng, E.; Bian, J.; Wang, X.-F.; Tian, W.; Sanehira,
2015, 2, 315−322. Y.; Miyasaka, T. Low-temperature SnO2-based Electron Selective
(86) Zhang, T.; Chen, H.; Bai, Y.; Xiao, S.; Zhu, L.; Hu, C.; Xue, Q.; Contact for Efficient and Stable Perovskite Solar Cells. J. Mater. Chem.
Yang, S. Understanding the Relationship Between Ion Migration and A 2015, 3, 10837−10844.
the Anomalous Hysteresis in High-efficiency Perovskite Solar Cells: A (104) Song, J.; Liu, L.; Wang, X.-F.; Chen, G.; Tian, W.; Miyasaka,
Fresh Perspective from Halide Substitution. Nano Energy 2016, 26, T. Highly Efficient and Stable Low-temperature Processed ZnO Solar
620−630. Cells with Triple Cation Perovskite Absorber. J. Mater. Chem. A 2017,
(87) Ravishankar, S.; Almora, O.; Echeverría-Arrondo, C.; 5, 13439−13447.
Ghahremanirad, E.; Aranda, C.; Guerrero, A.; Fabregat-Santiago, F.; (105) Kogo, A.; Numata, Y.; Ikegami, M.; Miyasaka, T. Nb2O5
Zaban, A.; Garcia-Belmonte, G.; Bisquert, J. Surface Polarization Blocking Layer for High Open-circuit Voltage Perovskite Solar Cells.
Model for the Dynamic Hysteresis of Perovskite Solar Cells. J. Phys. Chem. Lett. 2015, 44, 829−830.
Chem. Lett. 2017, 8, 915−921. (106) Ling, X.; Yuan, J.; Liu, D.; Wang, Y.; Zhang, Y.; Chen, S.; Wu,
(88) Calado, P.; Telford, A. M.; Bryant, D.; Li, X.; Nelson, J.; H.; Jin, F.; Wu, F.; Shi, G.; et al. Room-Temperature Processed
O’Regan, B. C.; Barnes, P. R. F. Evidence for Ion Migration in Hybrid Nb2O5 as the Electron-Transporting Layer for Efficient Planar
Perovskite Solar Cells with Minimal Hysteresis. Nat. Commun. 2016, Perovskite Solar Cells. ACS Appl. Mater. Interfaces 2017, 9, 23181−
7, 13831. 23188.
(89) Weber, S. A. L.; Hermes, I. M.; Turren-Cruz, S.-H.; Gort, C.; (107) Shen, D.; Zhang, W.; Li, Y.; Abate, A.; Wei, M. Facile
Bergmann, V. W.; Gilson, L.; Hagfeldt, A.; Graetzel, M.; Tress, W.; Deposition of Nb2O5 Thin Film as an Electron-Transporting Layer
Berger, R. How the Formation of Interfacial Charge Causes Hysteresis for Highly Efficient Perovskite Solar Cells. ACS Appl. Nano Mater.
in Perovskite Solar Cells. Energy Environ. Sci. 2018, 11, 2404−2413. 2018, 1, 4101−4109.

BE DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

(108) Wang, K.; Shi, Y.; Li, B.; Zhao, L.; Wang, W.; Wang, X.; Bai, (126) Jacobsson, T. J.; Svanström, S.; Andrei, V.; Rivett, J. P. H.;
X.; Wang, S.; Hao, C.; Ma, T. Amorphous Inorganic Electron- Kornienko, N.; Philippe, B.; Cappel, U. B.; Rensmo, H.; Deschler, F.;
Selective Layers for Efficient Perovskite Solar Cells: Feasible Strategy Boschloo, G. Extending the Compositional Space of Mixed Lead
Towards Room-Temperature Fabrication. Adv. Mater. 2016, 28, Halide Perovskites by Cs, Rb, K, and Na Doping. J. Phys. Chem. C
1891−1897. 2018, 122, 13548−13557.
(109) Gheno, A.; Thu Pham, T. T.; Di Bin, C.; Bouclé, J.; Ratier, B.; (127) Hoefler, S. F.; Trimmel, G.; Rath, T. Progress on Lead-free
Vedraine, S. Printable WO3 Electron Transporting Layer for Metal Halide Perovskites for Photovoltaic Applications: A review.
Perovskite Solar Cells: Influence on Device Performance and Stability. Monatsh. Chem. 2017, 148, 795−826.
Sol. Energy Mater. Sol. Cells 2017, 161, 347−354. (128) Brivio, F.; Walker, A. B.; Walsh, A. Structural and Electronic
(110) Numata, Y.; Ishikawa, R.; Sanehira, Y.; Kogo, A.; Shirai, H.; Properties of Hybrid Perovskites for High-efficiency Thin-film
Miyasaka, T. Nb-doped Amorphous Titanium Oxide Compact Layer Photovoltaics from First-principles. APL Mater. 2013, 1, No. 042111.
for Formamidinium-based High Efficiency Perovskite Solar Cells by (129) Wang, B.; Xiao, X.; Chen, T. Perovskite Photovoltaics: A
Low-temperature Fabrication. J. Mater. Chem. A 2018, 6, 9583−9591. High-efficiency Newcomer to the Solar Cell Family. Nanoscale 2014,
(111) Song, J.; Zheng, E.; Liu, L.; Wang, X.-F.; Chen, G.; Tian, W.; 6, 12287−12297.
Miyasaka, T. Magnesium-doped Zinc Oxide as Electron Selective (130) Umebayashi, T.; Asai, K.; Kondo, T.; Nakao, A. Electronic
Contact Layers for Efficient Perovskite Solar Cells. ChemSusChem Structures of Lead Iodide based Low-dimensional Crystals. Phys. Rev.
2016, 9, 2640−2647. B: Condens. Matter Mater. Phys. 2003, 67, 155405.
(112) Guo, X.; Dong, H.; Li, W.; Li, N.; Wang, L. Multifunctional (131) Chen, Q.; De Marco, N.; Yang, Y.; Song, T.-B.; Chen, C.-C.;
MgO Layer in Perovskite Solar Cells. ChemPhysChem 2015, 16, Zhao, H.; Hong, Z.; Zhou, H.; Yang, Y. Under the spotlight: The
1727−1732. organic−inorganic Hybrid Halide Perovskite for Optoelectronic
(113) Song, J.; Hu, W.; Wang, X.-F.; Chen, G.; Tian, W.; Miyasaka, Applications. Nano Today 2015, 10, 355−396.
T. HC(NH2)2PbI3 as a Thermally Stable Absorber for Efficient ZnO- (132) Zhou, Y.; Yang, M.; Pang, S.; Zhu, K.; Padture, N. P.
based Perovskite Solar Cells. J. Mater. Chem. A 2016, 4, 8435−8443. Exceptional Morphology-Preserving Evolution of Formamidinium
(114) Kunugita, H.; Hashimoto, T.; Kiyota, Y.; Udagawa, Y.; Lead Triiodide Perovskite Thin Films via Organic-Cation Displace-
Takeoka, Y.; Nakamura, Y.; Sano, J.; Matsushita, T.; Kondo, T.; ment. J. Am. Chem. Soc. 2016, 138, 5535−5538.
Miyasaka, T.; et al. Excitonic Feature in Hybrid Perovskite (133) Son, D.-Y.; Lee, J.-W.; Choi, Y. J.; Jang, I.-H.; Lee, S.; Yoo, P.
CH3NH3PbBr3 Single Crystals. Chem. Lett. 2015, 44, 852−854. J.; Shin, H.; Ahn, N.; Choi, M.; Kim, D.; Park, N.-G. Self-formed
(115) Zhang, Q.-H.; Hu, W.-D.; Wang, X.-F.; Chen, G.; Zhang, J.-P.; Grain Boundary Healing Layer for Highly Efficient CH3NH3PbI3
Xiao, L.-X.; Miyasaka, T. Fullerene Multiadducts as Electron Perovskite Solar Cells. Nat. Energy 2016, 1, 16081.
(134) Wu, Y.; Xie, F.; Chen, H.; Yang, X.; Su, H.; Cai, M.; Zhou, Z.;
Collection Layers for Perovskite Solar Cells. Chem. Lett. 2017, 46,
Noda, T.; Han, L. Thermally Stable MAPbI3 Perovskite Solar Cells
101−103.
with Efficiency of 19.19% and Area over 1 cm2 Achieved by Additive
(116) Kulkarni, A.; Jena, A. K.; Chen, H.-W.; Sanehira, Y.; Ikegami,
Engineering. Adv. Mater. 2017, 29, 1701073.
M.; Miyasaka, T. Revealing and Reducing the Possible Recombination
(135) Momblona, C.; Gil-Escrig, L.; Bandiello, E.; Hutter, E. M.;
Loss within TiO2 Compact Layer by Incorporating MgO Layer in
Sessolo, M.; Lederer, K.; Blochwitz-Nimoth, J.; Bolink, H. J. Efficient
Perovskite Solar Cells. Sol. Energy 2016, 136, 379−384.
Vacuum Deposited p-i-n and n-i-p Perovskite Solar Cells Employing
(117) Mahmood, K.; Sarwar, S.; Mehran, M. T. Current Status of
Doped Charge Transport Layers. Energy Environ. Sci. 2016, 9, 3456−
Electron Transport Layers in Perovskite Solar Cells: Materials and
3463.
Properties. RSC Adv. 2017, 7, 17044−17062. (136) Binek, A.; Hanusch, F. C.; Docampo, P.; Bein, T. Stabilization
(118) Fakharuddin, A.; Schmidt-Mende, L.; Garcia-Belmonte, G.;
of the Trigonal High-Temperature Phase of Formamidinium Lead
Jose, R.; Mora-Sero, I. Interfaces in Perovskite Solar Cells. Adv. Energy Iodide. J. Phys. Chem. Lett. 2015, 6, 1249−1253.
Mater. 2017, 7, 1700623. (137) Smecca, E.; Numata, Y.; Deretzis, I.; Pellegrino, G.; Boninelli,
(119) Burschka, J.; Pellet, N.; Moon, S.-J.; Humphry-Baker, R.; Gao, S.; Miyasaka, T.; La Magna, A.; Alberti, A. Stability of Solution-
P.; Nazeeruddin, M. K.; Gratzel, M. Sequential Deposition as a Route processed MAPbI3 and FAPbI3 layers. Phys. Chem. Chem. Phys. 2016,
to High-performance Perovskite-sensitized Solar Cells. Nature 2013, 18, 13413−13422.
499, 316−319. (138) Leijtens, T.; Bush, K.; Cheacharoen, R.; Beal, R.; Bowring, A.;
(120) Jeon, N. J.; Noh, J. H.; Yang, W. S.; Kim, Y. C.; Ryu, S.; Seo, McGehee, M. D. Towards Enabling Stable Lead Halide Perovskite
J.; Seok, S. I. Compositional Engineering of Perovskite Materials for Solar Cells; Interplay between Structural, Environmental, and
High-performance Solar Cells. Nature 2015, 517, 476−480. Thermal Stability. J. Mater. Chem. A 2017, 5, 11483−11500.
(121) Yang, W. S.; Noh, J. H.; Jeon, N. J.; Kim, Y. C.; Ryu, S.; Seo, (139) Niu, G.; Li, W.; Li, J.; Liang, X.; Wang, L. Enhancement of
J.; Seok, S. I. High-performance Photovoltaic Perovskite Layers Thermal Stability for Perovskite Solar Cells through Cesium Doping.
Fabricated through Intramolecular Exchange. Science 2015, 348, RSC Adv. 2017, 7, 17473−17479.
1234−1237. (140) Gong, J.; Guo, P.; Benjamin, S. E.; Van Patten, P. G.; Schaller,
(122) Bi, D.; Yi, C.; Luo, J.; Décoppet, J.-D.; Zhang, F.; R. D.; Xu, T. Cation Engineering on Lead Iodide Perovskites for
Zakeeruddin, S. M.; Li, X.; Hagfeldt, A.; Grätzel, M. Polymer- Stable and High-performance Photovoltaic Applications. J. Energy
templated Nucleation and Crystal Growth of Perovskite Films for Chem. 2018, 27, 1017−1039.
Solar Cells with Efficiency Greater than 21%. Nat. Energy 2016, 1, (141) Li, Z.; Yang, M.; Park, J.-S.; Wei, S.-H.; Berry, J. J.; Zhu, K.
16142. Stabilizing Perovskite Structures by Tuning Tolerance Factor:
(123) Yang, W. S.; Park, B.-W.; Jung, E. H.; Jeon, N. J.; Kim, Y. C.; Formation of Formamidinium and Cesium Lead Iodide Solid-State
Lee, D. U.; Shin, S. S.; Seo, J.; Kim, E. K.; Noh, J. H.; et al. Iodide Alloys. Chem. Mater. 2016, 28, 284−292.
Management in Formamidinium-lead-halide−based Perovskite Layers (142) Weller, M. T.; Weber, O. J.; Frost, J. M.; Walsh, A. Cubic
for Efficient Solar Cells. Science 2017, 356, 1376−1379. Perovskite Structure of Black Formamidinium Lead Iodide, α-
(124) Ono, L. K.; Juarez-Perez, E. J.; Qi, Y. Progress on Perovskite [HC(NH2)2]PbI3, at 298 K. J. Phys. Chem. Lett. 2015, 6, 3209−3212.
Materials and Solar Cells with Mixed Cations and Halide Anions. ACS (143) Lee, J.-W.; Kim, D.-H.; Kim, H.-S.; Seo, S.-W.; Cho, S. M.;
Appl. Mater. Interfaces 2017, 9, 30197−30246. Park, N.-G. Formamidinium and Cesium Hybridization for Photo-
(125) Kubicki, D. J.; Prochowicz, D.; Hofstetter, A.; Zakeeruddin, S. and Moisture-Stable Perovskite Solar Cell. Adv. Energy Mater. 2015, 5,
M.; Grätzel, M.; Emsley, L. Phase Segregation in Cs-, Rb- and K- 1501310.
Doped Mixed-Cation (MA)x(FA)1−xPbI3 Hybrid Perovskites from (144) Yi, C.; Luo, J.; Meloni, S.; Boziki, A.; Ashari-Astani, N.;
Solid-State NMR. J. Am. Chem. Soc. 2017, 139, 14173−14180. Gratzel, C.; Zakeeruddin, S. M.; Rothlisberger, U.; Gratzel, M.

BF DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Entropic Stabilization of Mixed A-cation ABX3 Metal Halide (163) Xie, L.-Q.; Chen, L.; Nan, Z.-A.; Lin, H.-X.; Wang, T.; Zhan,
Perovskites for High Performance Perovskite Solar Cells. Energy D.-P.; Yan, J.-W.; Mao, B.-W.; Tian, Z.-Q. Understanding the Cubic
Environ. Sci. 2016, 9, 656−662. Phase Stabilization and Crystallization Kinetics in Mixed Cations and
(145) Park, Y. H.; Jeong, I.; Bae, S.; Son, H. J.; Lee, P.; Lee, J.; Lee, Halides Perovskite Single Crystals. J. Am. Chem. Soc. 2017, 139,
C.-H.; Ko, M. J. Inorganic Rubidium Cation as an Enhancer for 3320−3323.
Photovoltaic Performance and Moisture Stability of HC(NH2)2PbI3 (164) Zheng, X.; Wu, C.; Jha, S. K.; Li, Z.; Zhu, K.; Priya, S.
Perovskite Solar Cells. Adv. Funct. Mater. 2017, 27 (16), 1605988. Improved Phase Stability of Formamidinium Lead Triiodide Perov-
(146) Yantara, N.; Yanan, F.; Shi, C.; Dewi, H. A.; Boix, P. P.; skite by Strain Relaxation. ACS Energy Lett. 2016, 1, 1014−1020.
Mhaisalkar, S. G.; Mathews, N. Unravelling the Effects of Cl Addition (165) Saliba, M.; Orlandi, S.; Matsui, T.; Aghazada, S.; Cavazzini,
in Single Step CH3NH3PbI3 Perovskite Solar Cells. Chem. Mater. M.; Correa-Baena, J.-P.; Gao, P.; Scopelliti, R.; Mosconi, E.; Dahmen,
2015, 27, 2309−2314. K.-H.; et al. A Molecularly Engineered Hole-transporting Material for
(147) Yu, H.; Wang, F.; Xie, F.; Li, W.; Chen, J.; Zhao, N. The Role Efficient Perovskite Solar Cells. Nat. Energy 2016, 1, 15017.
of Chlorine in the Formation Process of “CH3NH3PbI3‑xClx” (166) Bi, D.; Tress, W.; Dar, M. I.; Gao, P.; Luo, J.; Renevier, C.;
Perovskite. Adv. Funct. Mater. 2014, 24, 7102−7108. Schenk, K.; Abate, A.; Giordano, F.; et al. Correa Baena, J.-P.et al.
(148) Xie, F. X.; Zhang, D.; Su, H.; Ren, X.; Wong, K. S.; Grätzel, Efficient Luminescent Solar Cells Based on Tailored Mixed-cation
M.; Choy, W. C. H. Vacuum-Assisted Thermal Annealing of Perovskites. Sci. Adv. 2016, 2, No. e1501170.
CH3NH3PbI3 for Highly Stable and Efficient Perovskite Solar Cells. (167) Saliba, M.; Matsui, T.; Seo, J.-Y.; Domanski, K.; Correa-Baena,
ACS Nano 2015, 9, 639−646. J.-P.; Nazeeruddin, M. K.; Zakeeruddin, S. M.; Tress, W.; Abate, A.;
(149) Heo, J. H.; Lee, M. H.; Jang, M. H.; Im, S. H. Highly efficient Hagfeldt, A.; et al. Cesium-containing Triple Cation Perovskite Solar
CH3NH3PbI3‑xClx Mixed Halide Perovskite Solar Cells Prepared by Cells: Improved Stability, Reproducibility and High Efficiency. Energy
Re-dissolution and Crystal Grain Growth via Spray Coating. J. Mater. Environ. Sci. 2016, 9, 1989−1997.
Chem. A 2016, 4, 17636−17642. (168) Zhou, G.; Wu, J.; Zhao, Y.; Li, Y.; Shi, J.; Li, Y.; Wu, H.; Li,
(150) Starr, D. E.; Sadoughi, G.; Handick, E.; Wilks, R. G.; Alsmeier, D.; Luo, Y.; Meng, Q. Application of Cesium on the Restriction of
J. H.; Köhler, L.; Gorgoi, M.; Snaith, H. J.; Bär, M. Direct Observation Precursor Crystallization for Highly Reproducible Perovskite Solar
of an Inhomogeneous Chlorine Distribution in CH3NH3PbI3‑xClx Cells Exceeding 20% Efficiency. ACS Appl. Mater. Interfaces 2018, 10,
Layers: Surface Depletion and Interface Enrichment. Energy Environ. 9503−9513.
Sci. 2015, 8, 1609−1615. (169) Singh, T.; Miyasaka, T. Stabilizing the Efficiency Beyond 20%
(151) Colella, S.; Mosconi, E.; Pellegrino, G.; Alberti, A.; Guerra, V. with a Mixed Cation Perovskite Solar Cell Fabricated in Ambient Air
L. P.; Masi, S.; Listorti, A.; Rizzo, A.; Condorelli, G. G.; De Angelis, under Controlled Humidity. Adv. Energy Mater. 2018, 8, 1700677.
F.; et al. Elusive Presence of Chloride in Mixed Halide Perovskite (170) Saliba, M.; Matsui, T.; Domanski, K.; Seo, J.-Y.;
Solar Cells. J. Phys. Chem. Lett. 2014, 5, 3532−3538. Ummadisingu, A.; Zakeeruddin, S. M.; Correa-Baena, J.-P.; Tress,
(152) Unger, E. L.; Bowring, A. R.; Tassone, C. J.; Pool, V. L.; Gold- W. R.; Abate, A.; Hagfeldt, A.; et al. Incorporation of Rubidium
Parker, A.; Cheacharoen, R.; Stone, K. H.; Hoke, E. T.; Toney, M. F.; Cations into Perovskite Solar Cells Improves Photovoltaic Perform-
McGehee, M. D. Chloride in Lead Chloride-Derived Organo-Metal ance. Science 2016, 354, 206−209.
Halides for Perovskite-Absorber Solar Cells. Chem. Mater. 2014, 26, (171) Hu, Y.; Hutter, E. M.; Rieder, P.; Grill, I.; Hanisch, J.; Aygüle,
7158−7165. M. F.; Hufnagel, A. G.; Handloser, M.; Bein, T.; et al. Understanding
(153) Fan, L.; Ding, Y.; Luo, J.; Shi, B.; Yao, X.; Wei, C.; Zhang, D.; the Role of Cesium and Rubidium Additives in Perovskite Solar Cells:
Wang, G.; Sheng, Y.; Chen, Y.; et al. Elucidating the Role of Chlorine Trap States, Charge Transport, and Recombination. Adv. Energy
in Perovskite Solar Cells. J. Mater. Chem. A 2017, 5, 7423−7432. Mater. 2018, 8 (16), 1703057.
(154) Li, G.; Zhang, T.; Zhao, Y. Hydrochloric Acid Accelerated (172) Zhao, P.; Yin, W.; Kim, M.; Han, M.; Song, Y. J.; Ahn, T. K.;
Formation of Planar CH3NH3PbI3 Perovskite with High Humidity Jung, H. S. Improved Carriers Injection Capacity in Perovskite Solar
Tolerance. J. Mater. Chem. A 2015, 3, 19674−19678. Cells by Introducing A-site Interstitial Defects. J. Mater. Chem. A
(155) Pan, J.; Mu, C.; Li, Q.; Li, W.; Ma, D.; Xu, D. Room- 2017, 5, 7905−7911.
Temperature, Hydrochloride-Assisted, One-Step Deposition for (173) Bu, T.; Liu, X.; Zhou, Y.; Yi, J.; Huang, X.; Luo, L.; Xiao, J.;
Highly Efficient and Air-Stable Perovskite Solar Cells. Adv. Mater. Ku, Z.; Peng, Y.; Huang, F.; et al. A Novel Quadruple-cation Absorber
2016, 28, 8309−8314. for Universal Hysteresis Elimination for High Efficiency and Stable
(156) Chen, Y.; Zhao, Y.; Liang, Z. Non-Thermal Annealing Perovskite Solar Cells. Energy Environ. Sci. 2017, 10, 2509−2515.
Fabrication of Efficient Planar Perovskite Solar Cells with Inclusion of (174) Muzammal uz Zaman, M.; Imran, M.; Saleem, A.; Kamboh, A.
NH4Cl. Chem. Mater. 2015, 27, 1448−1451. H.; Arshad, M.; Khan, N. A.; Akhter, P. Potassium Doped
(157) Xie, F. X.; Su, H.; Mao, J.; Wong, K. S.; Choy, W. C. H. Methylammonium Lead Iodide (MAPbI3) Thin Films as a Potential
Evolution of Diffusion Length and Trap State Induced by Chloride in Absorber for Perovskite Solar Cells; Structural, Morphological,
Perovskite Solar Cell. J. Phys. Chem. C 2016, 120, 21248−21253. Electronic and Optoelectric Properties. Phys. B 2017, 522, 57−65.
(158) Noh, J. H.; Im, S. H.; Heo, J. H.; Mandal, T. N.; Seok, S. I. (175) Kubicki, D. J.; Prochowicz, D.; Hofstetter, A.; Zakeeruddin, S.
Chemical Management for Colorful, Efficient, and Stable Inorganic− M.; Grätzel, M.; Emsley, L. Phase Segregation in Potassium-Doped
Organic Hybrid Nanostructured Solar Cells. Nano Lett. 2013, 13, Lead Halide Perovskites from 39K Solid-State NMR at 21.1 T. J. Am.
1764−1769. Chem. Soc. 2018, 140, 7232−7238.
(159) Yang, X.; Yan, X.; Wang, W.; Zhu, X.; Li, H.; Ma, W.; Sheng, (176) Abdi-Jalebi, M.; Andaji-Garmaroudi, Z.; Cacovich, S.;
C. Light Induced Metastable Modification of Optical Properties in Stavrakas, C.; Philippe, B.; Richter, J. M.; Alsari, M.; Booker, E. P.;
CH3NH3PbI3‑xBrx Perovskite Films: Two-step Mechanism. Org. Hutter, E. M.; Pearson, A. J.; et al. Maximizing and Stabilizing
Electron. 2016, 34, 79−83. luminescence From Halide Perovskites with Potassium Passivation.
(160) Hoke, E. T.; Slotcavage, D. J.; Dohner, E. R.; Bowring, A. R.; Nature 2018, 555, 497−501.
Karunadasa, H. I.; McGehee, M. D. Reversible Photo-induced Trap (177) Son, D.-Y.; Kim, S.-G.; Seo, J.-Y.; Lee, S.-H.; Shin, H.; Lee, D.;
Formation in Mixed-halide Hybrid Perovskites for Photovoltaics. Park, N.-G. Universal Approach Toward Hysteresis-Free Perovskite
Chem. Sci. 2015, 6, 613−617. Solar Cell via Defect Engineering. J. Am. Chem. Soc. 2018, 140, 1358−
(161) Yin, W.-J.; Yan, Y.; Wei, S.-H. Anomalous Alloy Properties in 1364.
Mixed Halide Perovskites. J. Phys. Chem. Lett. 2014, 5, 3625−3631. (178) Yao, D.; Zhang, C.; Pham, N. D.; Zhang, Y.; Tiong, V. T.; Du,
(162) Chiang, C.-H.; Lin, J.-W.; Wu, C.-G. One-step Fabrication of a A.; Shen, Q.; Wilson, G. J.; Wang, H. Hindered Formation of
Mixed-halide Perovskite Film for a High-efficiency Inverted Solar Cell Photoinactive δ-FAPbI3 Phase and Hysteresis-Free Mixed-Cation
and Module. J. Mater. Chem. A 2016, 4, 13525−13533. Planar Heterojunction Perovskite Solar Cells with Enhanced

BG DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Efficiency via Potassium Incorporation. J. Phys. Chem. Lett. 2018, 9, (196) Wu, T.; Ahmadi, M.; Hu, B. Giant Current Amplification
2113−2120. Induced by Ion Migration in Perovskite Single Crystal Photo-
(179) Philippe, B.; Saliba, M.; Correa-Baena, J.-P.; Cappel, U. B.; detectors. J. Mater. Chem. C 2018, 6, 8042−8050.
Turren-Cruz, S.-H.; Grätzel, M.; Hagfeldt, A.; Rensmo, H. Chemical (197) Luo, Y.; Khoram, P.; Brittman, S.; Zhu, Z.; Lai, B.; Ong, S. P.;
Distribution of Multiple Cation (Rb+, Cs+, MA+, and FA+) Perovskite Garnett, E. C.; Fenning, D. P. Direct Observation of Halide Migration
Materials by Photoelectron Spectroscopy. Chem. Mater. 2017, 29, and its Effect on the Photoluminescence of Methylammonium Lead
3589−3596. Bromide Perovskite Single Crystals. Adv. Mater. 2017, 29 (43),
(180) Jacobsson, T. J.; Correa-Baena, J.-P.; Halvani Anaraki, E.; 1703451.
Philippe, B.; Stranks, S. D.; Bouduban, M. E. F.; Tress, W.; Schenk, (198) Zhao, J.; Deng, Y.; Wei, H.; Zheng, X.; Yu, Z.; Shao, Y.;
K.; Teuscher, J.; Moser, J.-E.; et al. Unreacted PbI2 as a Double-Edged Shield, J. E.; Huang, J. Strained Hybrid Perovskite Thin Films and
Sword for Enhancing the Performance of Perovskite Solar Cells. J. Their Impact on the Intrinsic Stability of Perovskite Solar Cells. Sci.
Am. Chem. Soc. 2016, 138, 10331−10343. Adv. 2017, 3, eaao5616.
(181) Smith, I. C.; Hoke, E. T.; Solis-Ibarra, D.; McGehee, M. D.; (199) Rolston, N.; Bush, K. A.; Printz, A. D.; Gold-Parker, A.; Ding,
Karunadasa, H. I. A Layered Hybrid Perovskite Solar-Cell Absorber Y.; Toney, M. F.; McGehee, M. D.; Dauskardt, R. H. Engineering
with Enhanced Moisture Stability. Angew. Chem. 2014, 126, 11414− Stress in Perovskite Solar Cells to Improve Stability. Adv. Energy
11417. Mater. 2018, 8, 1802139.
(182) Cao, D. H.; Stoumpos, C. C.; Farha, O. K.; Hupp, J. T.; (200) Strelcov, E.; Dong, Q.; Li, T.; Chae, J.; Shao, Y.; Deng, Y.;
Kanatzidis, M. G. 2D Homologous Perovskites as Light-Absorbing Gruverman, A.; Huang, J.; Centrone, A. CH3NH3PbI3 Perovskites:
Materials for Solar Cell Applications. J. Am. Chem. Soc. 2015, 137, Ferroelasticity Revealed. Sci. Adv. 2017, 3, No. e1602165.
7843−7850. (201) Lv, Q.; He, W.; Lian, Z.; Ding, J.; Li, Q.; Yan, Q. Anisotropic
(183) Etgar, L. The merit of perovskite’s dimensionality; can this Moisture Erosion of CH3NH3PbI3 Single Crystals. CrystEngComm
replace the 3D halide perovskite? Energy Environ. Sci. 2018, 11, 234− 2017, 19, 901−904.
242. (202) Leguy, A. M. A.; Hu, Y.; Campoy-Quiles, M.; Alonso, M. I.;
(184) Lu, J.; Jiang, L.; Li, W.; Li, F.; Pai, N. K.; Scully, A. D.; Tsai, C. Weber, O. J.; Azarhoosh, P.; van Schilfgaarde, M.; Weller, M. T.; Bein,
M.; Bach, U.; Simonov, A. N.; Cheng, Y. B.; Spiccia, L.; et al. T.; Nelson, J.; et al. Reversible Hydration of CH3NH3PbI3 in Films,
Diammonium and Monoammonium Mixed-Organic-Cation Perov- Single Crystals, and Solar Cells. Chem. Mater. 2015, 27, 3397−3407.
skites for High Performance Solar Cells with Improved Stability. Adv. (203) Habisreutinger, S. N.; Leijtens, T.; Eperon, G. E.; Stranks, S.
Energy Mater. 2017, 7, 1700444. D.; Nicholas, R. J.; Snaith, H. J. Carbon Nanotube/Polymer
(185) Ye, J.; Zheng, H.; Zhu, L.; Liu, G.; Zhang, X.; Hayat, T.; Pan, Composites as a Highly Stable Hole Collection Layer in Perovskite
X.; Dai, S. Enhanced Moisture Stability of Perovskite Solar Cells With Solar Cells. Nano Lett. 2014, 14, 5561−5568.
Mixed-Dimensional and Mixed-Compositional Light-Absorbing Ma- (204) Chaudhary, B.; Kulkarni, A.; Jena, A. K.; Ikegami, M.;
terials. Solar RRL 2017, 1, 1700125. Udagawa, Y.; Kunugita, H.; Ema, K.; Miyasaka, T. Poly(4-Vinyl-
(186) Koh, M.; Shanmugam, V.; Schlipf, J.; Oesinghaus, L.; Müller- pyridine)-Based Interfacial Passivation to Enhance Voltage and
Buschbaum, P.; Ramakrishnan, N.; Swamy, V.; Mathews, N.; Boix, P. Moisture Stability of Lead Halide Perovskite Solar Cells. Chem-
P.; Mhaisalkar, S. G. Nanostructuring Mixed-Dimensional Perov- SusChem 2017, 10, 2473−2479.
skites: A Route Toward Tunable, Efficient Photovoltaics. Adv. Mater. (205) Hou, Y.; Zhou, Z. R.; Wen, T.; Qiao, H.; Lin, Z. Q.; Ge, B.;
2016, 28, 3653−3661. Yang, H. Enhanced Moisture Stability of Metal Halide Perovskite
(187) Wang, Z.; Lin, Q.; Chmiel, F. P.; Sakai, N.; Herz, L. M.; Solar Cells Based on Sulfur-oleylamine Surface Modification.
Snaith, H. J. Efficient Ambient-air-stable Solar Cells with 2D−3D Nanoscale Horiz. 2019, 4 (1), 208.
Heterostructured Butylammonium-caesium-formamidinium Lead (206) Abdelmageed, G.; Sully, H. R.; Bonabi Naghadeh, S.; El-Hag
Halide Perovskites. Na. Energy 2017, 2, 17135. Ali, A.; Carter, S. A.; Zhang, J. Z. Improved Stability of Organometal
(188) Li, N.; Zhu, Z.; Chueh, C. C.; Liu, H.; Peng, B.; Petrone, A.; Halide Perovskite Films and Solar Cells Toward Humidity via Surface
Li, X.; Wang, L.; Jen, A. K.-Y. Mixed Cation FAxPEA1−xPbI3 with Passivation with Oleic Acid. ACS Appl. Energy Mater. 2018, 1, 387−
Enhanced Phase and Ambient Stability toward High-Performance 392.
Perovskite Solar Cells. Adv. Energy Mater. 2017, 7, 1601307. (207) Cao, J.; Yin, J.; Yuan, S.; Zhao, Y.; Li, J.; Zheng, N. Thiols as
(189) Cohen, B. E.; Wierzbowska, M.; Etgar, L. High Efficiency and Interfacial Modifiers to Enhance the Performance and Stability of
High Open Circuit Voltage in Quasi 2D Perovskite Based Solar Cells. Perovskite Solar Cells. Nanoscale 2015, 7, 9443−9447.
Adv. Funct. Mater. 2017, 27, 1604733. (208) Grancini, G.; Roldán-Carmona, C.; Zimmermann, I.;
(190) Han, G.; Hadi, H. D.; Bruno, A.; Kulkarni, S. A.; Koh, T. M.; Mosconi, E.; Lee, X.; Martineau, D.; Narbey, S.; Oswald, F.; De
Wong, L. H.; Soci, C.; Mathews, N.; Zhang, S.; Mhaisalkar, S. G. Angelis, F.; Graetzel, M.; et al. One-Year Stable Perovskite Solar Cells
Additive Selection Strategy for High Performance Perovskite by 2D/3D Interface Engineering. Nat. Commun. 2017, 8, 15684.
Photovoltaics. J. Phys. Chem. C 2018, 122, 13884−13893. (209) Yavari, M.; Mazloum-Ardakani, M.; Gholipour, S.; Tavakoli,
(191) Brunetti, B.; Cavallo, C.; Ciccioli, A.; Gigli, G.; Latini, A. On M. M.; Taghavinia, N.; Hagfeldt, A.; Tress, W. Reducing Surface
the Thermal and Thermodynamic (In)Stability of Methylammonium Recombination by a Poly(4-vinylpyridine) Interlayer in Perovskite
Lead Halide Perovskites. Sci. Rep. 2016, 6 (1), 31896. Solar Cells with High Open-Circuit Voltage and Efficiency. ACS
(192) Eames, C.; Frost, J. M.; Barnes, P. R. F.; O’Regan, B. C.; Omega 2018, 3, 5038−5043.
Walsh, A.; Islam, M. S. Ionic Transport in Hybrid Lead Iodide (210) Zong, Y.; Zhou, Y.; Zhang, Y.; Li, Z.; Zhang, L.; Ju, M.-G.;
Perovskite Solar Cells. Nat. Commun. 2015, 6, 7497. Chen, M.; Pang, S.; Zeng, X. C.; Padture, N. P. Continuous Grain-
(193) Yuan, Y.; Wang, Q.; Shao, Y.; Lu, H.; Li, T.; Gruverman, A.; Boundary Functionalization for High-Efficiency Perovskite Solar Cells
Huang, J. Electric-Field-Driven Reversible Conversion Between with Exceptional Stability. Chem. 2018, 4, 1404−1415.
Methylammonium Lead Triiodide Perovskites and Lead Iodide at (211) Fang, H.-H.; Adjokatse, S.; Wei, H.; Yang, J.; Blake, G. R.;
Elevated Temperatures. Adv. Energy Mater. 2016, 6, 1501803. Huang, J.; Even, J.; Loi, M. A. Ultrahigh Sensitivity of
(194) Minns, J. L.; Zajdel, P.; Chernyshov, D.; van Beek, W.; Green, Methylammonium Lead Tribromide Perovskite Single Crystals to
M. A. Structure and Interstitial Iodide Migration in Hybrid Perovskite Environmental Gases. Sci. Adv. 2016, 2, No. e1600534.
Methylammonium Lead Iodide. Nat. Commun. 2017, 8, 15152. (212) Zohar, A.; Kedem, N.; Levine, I.; Zohar, D.; Vilan, A.; Ehre,
(195) Yuan, Y.; Chae, J.; Shao, Y.; Wang, Q.; Xiao, Z.; Centrone, A.; D.; Hodes, G.; Cahen, D. Impedance Spectroscopic Indication for
Huang, J. Photovoltaic Switching Mechanism in Lateral Structure Solid State Electrochemical Reaction in (CH3NH3)PbI3 Films. J. Phys.
Hybrid Perovskite Solar Cells. Adv. Energy Mater. 2015, 5, 1500615. Chem. Lett. 2016, 7, 191−197.

BH DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

(213) Kong, W.; Rahimi-Iman, A.; Bi, G.; Dai, X.; Wu, H. Oxygen Methylammonium Lead Halide Perovskite Solar Cells. Adv. Energy
Intercalation Induced by Photocatalysis on the Surface of Hybrid Mater. 2016, 6 (13), 1600330.
Lead Halide Perovskites. J. Phys. Chem. C 2016, 120, 7606−7611. (230) Bush, K. A.; Frohna, K.; Prasanna, R.; Beal, R. E.; Leijtens, T.;
(214) Aristidou, N.; Sanchez-Molina, I.; Chotchuangchutchaval, T.; Swifter, S. A.; McGehee, M. D. Compositional Engineering for
Brown, M.; Martinez, L.; Rath, T.; Haque, S. A. The Role of Oxygen Efficient Wide Band Gap Perovskites with Improved Stability to
in the Degradation of Methylammonium Lead Trihalide Perovskite Photoinduced Phase Segregation. ACS Energy Lett. 2018, 3, 428−435.
Photoactive Layers. Angew. Chem., Int. Ed. 2015, 54, 8208−8212. (231) Cappel, U. B.; Daeneke, T.; Bach, U. Oxygen-Induced Doping
(215) Pearson, A. J.; Eperon, G. E.; Hopkinson, P. E.; of Spiro-MeOTAD in Solid-State Dye-Sensitized Solar Cells and Its
Habisreutinger, S. N.; Wang, J. T.-W.; Snaith, H. J.; Greenham, N. Impact on Device Performance. Nano Lett. 2012, 12, 4925−4931.
C. Oxygen Degradation in Mesoporous Al2O3/CH3NH3PbI3‑xClx (232) Zhao, X.; Kim, H.-S.; Seo, J.-Y.; Park, N.-G. Effect of Selective
Perovskite Solar Cells: Kinetics and Mechanisms. Adv. Energy Contacts on the Thermal Stability of Perovskite Solar Cells. ACS
Mater. 2016, 6, 1600014. Appl. Mater. Interfaces 2017, 9, 7148−7153.
(216) O’Mahony, F. T. F.; Lee, Y. H.; Jellett, C.; Dmitrov, S.; (233) Sanchez, R. S.; Mas-Marza, E. Light-induced Effects on Spiro-
Bryant, D. T. J.; Durrant, J. R.; O’Regan, B. C.; Graetzel, M.; OMeTAD Films and Hybrid Lead Halide Perovskite Solar Cells. Sol.
Nazeeruddin, M. K.; Haque, S. A. Improved Environmental Stability Energy Mater. Sol. Cells 2016, 158, 189−194.
of Organic Lead Trihalide Perovskite-based Photoactive-layers in the (234) Domanski, K.; Correa-Baena, J.-P.; Mine, N.; Nazeeruddin, M.
Presence of Mesoporous TiO2. J. Mater. Chem. A 2015, 3, 7219− K.; Abate, A.; Saliba, M.; Tress, W.; Hagfeldt, A.; Grätzel, M. Not All
7223. That Glitters Is Gold: Metal-Migration-Induced Degradation in
(217) Leijtens, T.; Eperon, G. E.; Pathak, S.; Abate, A.; Lee, M. M.; Perovskite Solar Cells. ACS Nano 2016, 10, 6306−6314.
Snaith, H. J. Overcoming Ultraviolet Light Instability of Sensitized (235) Jena, A. K.; Numata, Y.; Ikegami, M.; Miyasaka, T. Role of
TiO2 with Meso-superstructured Organometal Tri-halide Perovskite spiro-OMeTAD in Performance Deterioration of Perovskite Solar
Solar Cells. Nat. Commun. 2013, 4, 2885. Cells at High Temperature and Reuse of the Perovskite Films to
(218) Li, W.; Zhang, W.; Van Reenen, S.; Sutton, R. J.; Fan, J.; Avoid Pb-waste. J. Mater. Chem. A 2018, 6, 2219−2230.
Haghighirad, A. A.; Johnston, M. B.; Wang, L.; Snaith, H. J. Enhanced (236) Carrillo, J.; Guerrero, A.; Rahimnejad, S.; Almora, O.; Zarazua,
UV-light Stability of Planar Heterojunction Perovskite Solar Cells I.; Mas-Marza, E.; Bisquert, J.; Garcia-Belmonte, G. Ionic Reactivity at
with Caesium Bromide Interface Modification. Energy Environ. Sci. Contacts and Aging of Methylammonium Lead Triiodide Perovskite
2016, 9, 490−498. Solar Cells. Adv. Energy Mater. 2016, 6, 1502246.
(219) Sun, Y.; Fang, X.; Ma, Z.; Xu, L.; Lu, Y.; Yu, Q.; Yuan, N.; (237) Zhao, Y.; Zhou, W.; Tan, H.; Fu, R.; Li, Q.; Lin, F.; Yu, D.;
Ding, J. Enhanced UV-light Stability of Organometal Halide Walters, G.; Sargent, E. H.; Zhao, Q. Mobile-Ion-Induced
Perovskite Solar Cells with Interface Modification and a UV Degradation of Organic Hole-Selective Layers in Perovskite Solar
Absorption Layer. J. Mater. Chem. C 2017, 5, 8682−8687. Cells. J. Phys. Chem. C 2017, 121, 14517−14523.
(220) Lee, S.-W.; Kim, S.; Bae, S.; Cho, K.; Chung, T.; Mundt, L. E.; (238) Habisreutinger, S. N.; Wenger, B.; Snaith, H. J.; Nicholas, R. J.
Lee, S.; Park, S.; Park, H.; Schubert, M. C.; et al. UV Degradation and Dopant-Free Planar n−i−p Perovskite Solar Cells with Steady-State
Recovery of Perovskite Solar Cells. Sci. Rep. 2016, 6 (1), 38150. Efficiencies Exceeding 18%. ACS Energy Lett. 2017, 2, 622−628.
(221) Nie, W.; Blancon, J.-C.; Neukirch, A. J.; Appavoo, K.; Tsai, H.; (239) Jena, A. K.; Ikegami, M.; Miyasaka, T. Severe Morphological
Chhowalla, M.; Alam, M. A.; Sfeir, M. Y.; Katan, C.; Even, J.; et al. Deformation of Spiro-OMeTAD in (CH3NH3)PbI3 Solar Cells at
Light-activated Photocurrent Degradation and Self-healing in Perov- High Temperature. ACS Energy Lett. 2017, 2, 1760−1761.
skite Solar Cells. Nat. Commun. 2016, 7, 11574. (240) Kasparavicius, E.; Magomedov, A.; Malinauskas, T.; Getautis,
(222) Brennan, M. C.; Draguta, S.; Kamat, P. V.; Kuno, M. Light- V. Long-Term Stability of the Oxidized Hole-Transporting Materials
Induced Anion Phase Segregation in Mixed Halide Perovskites. ACS used in Perovskite Solar Cells. Chem. - Eur. J. 2018, 24, 9910−9918.
Energy Lett. 2018, 3, 204−213. (241) Islam, M. B.; Yanagida, M.; Shirai, Y.; Nabetani, Y.; Miyano,
(223) deQuilettes, D. W.; Zhang, W.; Burlakov, V. M.; Graham, D. K. NiOx Hole Transport Layer for Perovskite Solar Cells with
J.; Leijtens, T.; Osherov, A.; Bulović, V.; Snaith, H. J.; Ginger, D. S.; Improved Stability and Reproducibility. ACS Omega 2017, 2, 2291−
Stranks, S. D. Photo-induced halide redistribution in organic− 2299.
inorganic perovskite films. Nat. Commun. 2016, 7 (1), 11683. (242) You, J.; Meng, L.; Song, T.-B.; Guo, T.-F.; Yang, Y.; Chang,
(224) Roose, B.; Baena, J.-P. C.; Gödel, K. C.; Graetzel, M.; W.-H.; Hong, Z.; Chen, H.; Zhou, H.; Chen, Q.; et al. Improved Air
Hagfeldt, A.; Steiner, U.; Abate, A. Mesoporous SnO2 Electron Stability of Perovskite Solar Cells via Solution-Processed Metal Oxide
Selective Contact Enables UV-stable Perovskite Solar Cells. Nano Transport Layers. Nat. Nanotechnol. 2016, 11, 75−81.
Energy 2016, 30, 517−522. (243) Rao, H.; Ye, S.; Sun, W.; Yan, W.; Li, Y.; Peng, H.; Liu, Z.;
(225) Zhao, P.; Han, M.; Yin, W.; Zhao, X.; Kim, S. G.; Yan, Y.; Bian, Z.; Li, Y.; Huang, C. A 19.0% Efficiency Achieved in CuOx-
Kim, M.; Song, Y. J.; Park, N. G.; Jung, H. S. Insulated Interlayer for based Inverted CH3NH3PbI3‑xClx Solar Cells by an Effective Cl
Efficient and Photostable Electron-Transport-Layer-Free Perovskite Doping Method. Nano Energy 2016, 27, 51−57.
Solar Cells. ACS Appl. Mater. Interfaces 2018, 10, 10132−10140. (244) Sun, W.; Li, Y.; Ye, S.; Rao, H.; Yan, W.; Peng, H.; Li, Y.; Liu,
(226) Chander, N.; Khan, A. F.; Chandrasekhar, P. S.; Thouti, E.; Z.; Wang, S.; Chen, Z.; et al. High-performance Inverted Planar
Swami, S. K.; Dutta, V.; Komarala, V. K. Reduced Ultraviolet Light Heterojunction Perovskite Solar Cells Based on a Solution-processed
Induced Degradation and Enhanced Light Harvesting using CuOx Hole Transport Layer. Nanoscale 2016, 8, 10806−10813.
YVO4:Eu3+ Down-shifting Nano-phosphor Layer in Organometal (245) Hossain, M. I.; Alharbi, F. H.; Tabet, N. Copper Oxide as
Halide Perovskite Solar Cells. Appl. Phys. Lett. 2014, 105, No. 033904. Inorganic Hole Transport Material for Lead Halide Perovskite Based
(227) Farooq, A.; Hossain, I. M.; Moghadamzadeh, S.; Schwenzer, J. Solar Cells. Sol. Energy 2015, 120, 370−380.
A.; Abzieher, T.; Richards, B. S.; Klampaftis, E.; Paetzold, U. W. (246) Christians, J. A.; Fung, R. C. M.; Kamat, P. V. An Inorganic
Spectral Dependence of Degradation under Ultraviolet Light in Hole Conductor for Organo-Lead Halide Perovskite Solar Cells.
Perovskite Solar Cells. ACS Appl. Mater. Interfaces 2018, 10, 21985− Improved Hole Conductivity with Copper Iodide. J. Am. Chem. Soc.
21990. 2014, 136, 758−764.
(228) Slotcavage, D. J.; Karunadasa, H. I.; McGehee, M. D. Light- (247) Wang, H.; Yu, Z.; Jiang, X.; Li, J.; Cai, B.; Yang, X.; Sun, L.
Induced Phase Segregation in Halide-Perovskite Absorbers. ACS Efficient and Stable Inverted Planar Perovskite Solar Cells Employing
Energy Lett. 2016, 1, 1199−1205. CuI as Hole-Transporting Layer Prepared by Solid−Gas Trans-
(229) Yun, J. S.; Seidel, J.; Kim, J.; Soufiani, A. M.; Huang, S.; Lau, formation. Energy Technology 2017, 5, 1836−1843.
J.; Jeon, N. J.; Seok, S. I.; Green, M. A.; Ho-Baillie, A. Critical Role of (248) Ye, S.; Rao, H.; Zhao, Z.; Zhang, L.; Bao, H.; Sun, W.; Li, Y.;
Grain Boundaries for Ion Migration in Formamidinium and Gu, F.; Wang, J.; Liu, Z.; et al. A Breakthrough Efficiency of 19.9%

BI DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Obtained in Inverted Perovskite Solar Cells by Using an Efficient (266) Zhang, T.; Dar, M. I.; Li, G.; Xu, F.; Guo, N.; Grätzel, M.;
Trap State Passivator Cu(thiourea)I. J. Am. Chem. Soc. 2017, 139, Zhao, Y. Bication Lead Iodide 2D Perovskite Component to Stabilize
7504−7512. Inorganic α-CsPbI3 Perovskite Phase for High-efficiency Solar Cells.
(249) Jung, M.; Kim, Y. C.; Jeon, N. J.; Yang, W. S.; Seo, J.; Noh, J. Sci. Adv. 2017, 3, No. e1700841.
H.; Il Seok, S. Thermal Stability of CuSCN Hole Conductor-Based (267) Swarnkar, A.; Marshall, A. R.; Sanehira, E. M.; Chernomordik,
Perovskite Solar Cells. ChemSusChem 2016, 9, 2592−2596. B. D.; Moore, D. T.; Christians, J. A.; Chakrabarti, T.; Luther, J. M.
(250) Arora, N.; Dar, M. I.; Hinderhofer, A.; Pellet, N.; Schreiber, Quantum dot−induced Phase Stabilization of α-CsPbI3 Perovskite for
F.; Zakeeruddin, S. M.; Grätzel, M. Perovskite Solar Cells with High-efficiency Photovoltaics. Science 2016, 354, 92−95.
CuSCN Hole Extraction Layers Yield Stabilized Efficiencies Greater (268) Sanehira, E. M.; Marshall, A. R.; Christians, J. A.; Harvey, S.
than 20%. Science 2017, 358, 768−771. P.; Ciesielski, P. N.; Wheeler, L. M.; Schulz, P.; Lin, L. Y.; Beard, M.
(251) Jiang, X.; Yu, Z.; Li, H.-B.; Zhao, Y.; Qu, J.; Lai, J.; Ma, W.; C.; Luther, J. M. Enhanced Mobility CsPbI3 Quantum Dot Arrays for
Wang, D.; Yang, X.; Sun, L. A Solution-processable Copper(ii) Record-efficiency, High-voltage Photovoltaic Cells. Sci. Adv. 2017, 3,
Phthalocyanine Derivative as a Dopant-free Hole-transporting Ma- eaao4204.
terial for Efficient and Stable Carbon Counter Electrode-based (269) Hazarika, A.; Zhao, Q.; Gaulding, E. A.; Christians, J. A.; Dou,
Perovskite Solar Cells. J. Mater. Chem. A 2017, 5, 17862−17866. B.; Marshall, A. R.; Moot, T.; Berry, J. J.; Johnson, J. C.; Luther, J. M.
(252) Etgar, L.; Gao, P.; Xue, Z.; Peng, Q.; Chandiran, A. K.; Liu, B.; Perovskite Quantum Dot Photovoltaic Materials beyond the Reach of
Nazeeruddin, M. K.; Grätzel, M. Mesoscopic CH3NH3PbI3/TiO2 Thin Films: Full-Range Tuning of A-Site Cation Composition. ACS
Heterojunction Solar Cells. J. Am. Chem. Soc. 2012, 134, 17396− Nano 2018, 12, 10327−10337.
17399. (270) Hu, Y.; Bai, F.; Liu, X.; Ji, Q.; Miao, X.; Qiu, T.; Zhang, S.
(253) Laban, W. A.; Etgar, L. Depleted Hole Conductor-free Lead Bismuth Incorporation Stabilized α-CsPbI3 for Fully Inorganic
Halide Iodide Heterojunction Solar Cells. Energy Environ. Sci. 2013, 6, Perovskite Solar Cells. ACS Energy Lett. 2017, 2, 2219−2227.
3249−3253. (271) Xiang, S.; Li, W.; Wei, Y.; Liu, J.; Liu, H.; Zhu, L.; Chen, H.
(254) Ku, Z.; Xia, X.; Shen, H.; Tiep, N. H.; Fan, H. J. A Synergistic Effect of Non-stoichiometry and Sb-doping on Air-stable
Mesoporous Nickel Counter Electrode for Printable and Reusable α-CsPbI3 for Efficient Carbon-based Perovskite Solar Cells. Nanoscale
Perovskite Solar Cells. Nanoscale 2015, 7, 13363−13368. 2018, 10, 9996−10004.
(255) Ku, Z.; Rong, Y.; Xu, M.; Liu, T.; Han, H. Full Printable (272) Lau, C. F. J.; Deng, X.; Zheng, J.; Kim, J.; Zhang, Z.; Zhang,
Processed Mesoscopic CH3NH3PbI3/TiO2 Heterojunction Solar M.; Bing, J.; Wilkinson, B.; Hu, L.; Patterson, R.; et al. Enhanced
Cells with Carbon Counter Electrode. Sci. Rep. 2013, 3, 3132.
Performance via Partial Lead Replacement with Calcium for a CsPbI3
(256) Mei, A.; Li, X.; Liu, L.; Ku, Z.; Liu, T.; Rong, Y.; Xu, M.; Hu,
Perovskite Solar Cell Exceeding 13% Power Conversion Efficiency. J.
M.; Chen, J.; Yang, Y.; et al. A Hole-conductor−free, Fully Printable
Mater. Chem. A 2018, 6, 5580−5586.
Mesoscopic Perovskite Solar Cell with High Stability. Science 2014,
(273) Jena, A. K.; Kulkarni, A.; Sanehira, Y.; Ikegami, M.; Miyasaka,
345, 295−298.
T. Stabilization of α-CsPbI3 in Ambient Room Temperature
(257) Wei, Z.; Yan, K.; Chen, H.; Yi, Y.; Zhang, T.; Long, X.; Li, J.;
Conditions by Incorporating Eu into CsPbI3. Chem. Mater. 2018,
Zhang, L.; Wang, J.; Yang, S. Cost-efficient Clamping Solar Cells
using Candle Soot for Hole Extraction from Ambipolar Perovskites. 30, 6668−6674.
(274) Wang, Y.; Zhang, T.; Kan, M.; Zhao, Y. Bifunctional
Energy Environ. Sci. 2014, 7, 3326−3333.
(258) Chen, H.; Wei, Z.; Yan, K.; Yi, Y.; Wang, J.; Yang, S. Liquid Stabilization of All-Inorganic α-CsPbI3 Perovskite for 17% Efficiency
Phase Deposition of TiO2 Nanolayer Affords CH3NH3PbI3/Nano- Photovoltaics. J. Am. Chem. Soc. 2018, 140, 12345−12348.
carbon Solar Cells with High Open-circuit Voltage. Faraday Discuss. (275) Kulbak, M.; Gupta, S.; Kedem, N.; Levine, I.; Bendikov, T.;
2014, 176, 271−286. Hodes, G.; Cahen, D. Cesium Enhances Long-Term Stability of Lead
(259) Chen, H.; Wei, Z.; He, H.; Zheng, X.; Wong, K. S.; Yang, S. Bromide Perovskite-Based Solar Cells. J. Phys. Chem. Lett. 2016, 7,
Solvent Engineering Boosts the Efficiency of Paintable Carbon-Based 167−172.
Perovskite Solar Cells to Beyond 14%. Adv. Energy Mater. 2016, 6 (8), (276) Sutton, R. J.; Eperon, G. E.; Miranda, L.; Parrott, E. S.;
1502087. Kamino, B. A.; Patel, J. B.; Hörantner, M. T.; Johnston, M. B.;
(260) Yan, K.; Wei, Z.; Li, J.; Chen, H.; Yi, Y.; Zheng, X.; Long, X.; Haghighirad, A. A.; Moore, D. T.; et al. Bandgap-Tunable Cesium
Wang, Z.; Wang, J.; Xu, J.; et al. High-Performance Graphene-Based Lead Halide Perovskites with High Thermal Stability for Efficient
Hole Conductor-Free Perovskite Solar Cells: Schottky Junction Solar Cells. adv. Energy Mater. 2016, 6, 1502458.
Enhanced Hole Extraction and Electron Blocking. Small 2015, 11, (277) Liu, C.; Li, W.; Zhang, C.; Ma, Y.; Fan, J.; Mai, Y. All-
2269−2274. Inorganic CsPbI2Br Perovskite Solar Cells with High Efficiency
(261) Wu, W.-Q.; Wang, Q.; Fang, Y.; Shao, Y.; Tang, S.; Deng, Y.; Exceeding 13%. J. Am. Chem. Soc. 2018, 140, 3825−3828.
Lu, H.; Liu, Y.; Li, T.; Yang, Z.; et al. Molecular Doping Enabled (278) Chen, M.; Ju, M.-G.; Carl, A. D.; Zong, Y.; Grimm, R. L.; Gu,
Scalable Blading of Efficient Hole-transport-layer-free Perovskite Solar J.; Zeng, X. C.; Zhou, Y.; Padture, N. P. Cesium Titanium(IV)
Cells. Nat. Commun. 2018, 9 (1), 1625. Bromide Thin Films Based Stable Lead-free Perovskite Solar Cells.
(262) Eperon, G. E.; Paterno, G. M.; Sutton, R. J.; Zampetti, A.; Joule 2018, 2, 558−570.
Haghighirad, A. A.; Cacialli, F.; Snaith, H. J. Inorganic Caesium Lead (279) Ju, M.-G.; Chen, M.; Zhou, Y.; Garces, H. F.; Dai, J.; Ma, L.;
Iodide Perovskite Solar Cells. J. Mater. Chem. A 2015, 3, 19688− Padture, N. P.; Zeng, X. C. Earth-Abundant Nontoxic Titanium(IV)-
19695. based Vacancy-Ordered Double Perovskite Halides with Tunable 1.0
(263) Wang, P.; Zhang, X.; Zhou, Y.; Jiang, Q.; Ye, Q.; Chu, Z.; Li, to 1.8 eV Bandgaps for Photovoltaic Applications. ACS Energy Lett.
X.; Yang, X.; Yin, Z.; You, J. Solvent-controlled Growth of Inorganic 2018, 3, 297−304.
Perovskite Films in Dry Environment for Efficient and Stable Solar (280) Luo, P.; Xia, W.; Zhou, S.; Sun, L.; Cheng, J.; Xu, C.; Lu, Y.
Cells. Nat. Commun. 2018, 9 (1), 2225. Solvent Engineering for Ambient-Air-Processed, Phase-Stable CsPbI3
(264) Sutton, R. J.; Filip, M. R.; Haghighirad, A. A.; Sakai, N.; in Perovskite Solar Cells. J. Phys. Chem. Lett. 2016, 7, 3603−3608.
Wenger, B.; Giustino, F.; Snaith, H. J. Cubic or Orthorhombic? (281) Kim, Y. G.; Kim, T.-Y.; Oh, J. H.; Choi, K. S.; Kim, Y.-J.; Kim,
Revealing the Crystal Structure of Metastable Black-phase CsPbI3 by S. Y. Cesium Lead Iodide Solar Cells Controlled by Annealing
Theory and Experiment. ACS Energy Lett. 2018, 3, 1787−1794. Temperature. Phys. Chem. Chem. Phys. 2017, 19, 6257−6263.
(265) Wang, Q.; Zheng, X.; Deng, Y.; Zhao, J.; Chen, Z.; Huang, J. (282) Wang, Q.; Zheng, X.; Deng, Y.; Zhao, J.; Chen, Z.; Huang, J.
Stabilizing the α-Phase of CsPbI3 Perovskite by Sulfobetaine Stabilizing the α-Phase of CsPbI3 Perovskite by Sulfobetaine
Zwitterions in One-Step Spin-Coating Films. Joule 2017, 1, 371−382. Zwitterions in One-Step Spin-Coating Films. Joule 2017, 1, 371−382.

BJ DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

(283) Babayigit, A.; Ethirajan, A.; Muller, M.; Conings, B. Toxicity (301) Hao, F.; Stoumpos, C. C.; Chang, R. P. H.; Kanatzidis, M. G.
of Organometal Halide Perovskite Solar Cells. Nat. Mater. 2016, 15, Anomalous Band Gap Behavior in Mixed Sn and Pb Perovskites
247−251. Enables Broadening of Absorption Spectrum in Solar Cells. J. Am.
(284) Turkevych, I.; Kazaoui, S.; Ito, E.; Urano, T.; Yamada, K.; Chem. Soc. 2014, 136, 8094−8099.
Tomiyasu, H.; Yamagishi, H.; Kondo, M.; Aramaki, S. Photovoltaic (302) Liu, X.; Yang, Z.; Chueh, C.-C.; Rajagopal, A.; Williams, S. T.;
Rudorffites: Lead-Free Silver Bismuth Halides Alternative to Hybrid Sun, Y.; Jen, A. K. Y. Improved Efficiency and Stability of Pb−Sn
Lead Halide Perovskites. ChemSusChem 2017, 10, 3754−3759. Binary Perovskite Solar Cells by Cs Substitution. J. Mater. Chem. A
(285) Hailegnaw, B.; Kirmayer, S.; Edri, E.; Hodes, G.; Cahen, D. 2016, 4, 17939−17945.
Rain on Methylammonium Lead Iodide Based Perovskites: Possible (303) Zhang, M.; Lyu, M.; Yun, J.-H.; Noori, M.; Zhou, X.; Cooling,
Environmental Effects of Perovskite Solar Cells. J. Phys. Chem. Lett. N. A.; Wang, Q.; Yu, H.; Dastoor, P. C.; Wang, L. Low-temperature
2015, 6, 1543−1547. Processed Solar Cells with Formamidinium Tin Halide Perovskite/
(286) Stoumpos, C. C.; Malliakas, C. D.; Kanatzidis, M. G. Fullerene Heterojunctions. Nano Res. 2016, 9, 1570−1577.
Semiconducting Tin and Lead Iodide Perovskites with Organic (304) Todorov, T. K.; Reuter, K. B.; Mitzi, D. B. High-Efficiency
Cations: Phase Transitions, High Mobilities, and Near-Infrared Solar Cell with Earth-Abundant Liquid-Processed Absorber. Adv.
Photoluminescent Properties. Inorg. Chem. 2013, 52, 9019−9038. Mater. 2010, 22, E156−E159.
(287) Manser, J. S.; Christians, J. A.; Kamat, P. V. Intriguing (305) Konstantakou, M.; Stergiopoulos, T. A Critical Review on Tin
Optoelectronic Properties of Metal Halide Perovskites. Chem. Rev. Halide Perovskite Solar Cells. J. Mater. Chem. A 2017, 5, 11518−
2016, 116, 12956−13008. 11549.
(288) Chen, Z.; Wang, J. J.; Ren, Y.; Yu, C.; Shum, K. Schottky solar (306) Hao, F.; Stoumpos, C. C.; Guo, P.; Zhou, N.; Marks, T. J.;
cells based on CsSnI3 Thin-films. Appl. Phys. Lett. 2012, 101, Chang, R. P. H.; Kanatzidis, M. G. Solvent-Mediated Crystallization
No. 093901. of CH3NH3SnI3 Films for Heterojunction Depleted Perovskite Solar
(289) Noel, N. K.; Stranks, S. D.; Abate, A.; Wehrenfennig, C.; Cells. J. Am. Chem. Soc. 2015, 137, 11445−11452.
Guarnera, S.; Haghighirad, A.-A.; Sadhanala, A.; Eperon, G. E.; (307) Weiss, M.; Horn, J.; Richter, C.; Schlettwein, D. Preparation
Pathak, S. K.; Johnston, M. B.; et al. Lead-free Organic-inorganic Tin and Characterization of Methylammonium Tin Iodide Layers as
Halide Perovskites for Photovoltaic Applications. Energy Environ. Sci. Photovoltaic Absorbers. Phys. Status Solidi A 2016, 213, 975−981.
2014, 7, 3061−3068. (308) Song, T.-B.; Yokoyama, T.; Aramaki, S.; Kanatzidis, M. G.
(290) Hao, F.; Stoumpos, C. C.; Cao, D. H.; Chang, R. P. H.; Performance Enhancement of Lead-Free Tin-Based Perovskite Solar
Kanatzidis, M. G. Lead-free Solid-state Organic-inorganic Halide Cells with Reducing Atmosphere-Assisted Dispersible Additive. ACS
Perovskite Solar Cells. Nat. Photonics 2014, 8, 489−494. Energy Lett. 2017, 2, 897−903.
(291) Lee, S. J.; Shin, S. S.; Kim, Y. C.; Kim, D.; Ahn, T. K.; Noh, J. (309) Song, T.-B.; Yokoyama, T.; Stoumpos, C. C.; Logsdon, J.;
H.; Seo, J.; Seok, S. I. Fabrication of Efficient Formamidinium Tin Cao, D. H.; Wasielewski, M. R.; Aramaki, S.; Kanatzidis, M. G.
Iodide Perovskite Solar Cells through SnF2−Pyrazine Complex. J.
Importance of Reducing Vapor Atmosphere in the Fabrication of Tin-
Am. Chem. Soc. 2016, 138, 3974−3977.
Based Perovskite Solar Cells. J. Am. Chem. Soc. 2017, 139, 836−842.
(292) Liao, Y.; Liu, H.; Zhou, W.; Yang, D.; Shang, Y.; Shi, Z.; Li, B.;
(310) Marshall, K. P.; Walker, M.; Walton, R. I.; Hatton, R. A.
Jiang, X.; Zhang, L.; Quan, L. N.; et al. Highly Oriented Low-
Enhanced Stability and Efficiency in Hole-transport-layer-free CsSnI3
Dimensional Tin Halide Perovskites with Enhanced Stability and
Perovskite Photovoltaics. Nat. Energy 2016, 1, 16178.
Photovoltaic Performance. J. Am. Chem. Soc. 2017, 139, 6693−6699.
(311) Koh, T. M.; Krishnamoorthy, T.; Yantara, N.; Shi, C.; Leong,
(293) Shao, S.; Liu, J.; Portale, G.; Fang, H.-H.; Blake, G. R.; ten
W. L.; Boix, P. P.; Grimsdale, A. C.; Mhaisalkar, S. G.; Mathews, N.
Brink, G. H.; Koster, L. J. A.; Loi, M. A. Highly Reproducible Sn-
Based Hybrid Perovskite Solar Cells with 9% Efficiency. Adv. Energy Formamidinium Tin-based Perovskite with Low Eg for Photovoltaic
Mater. 2018, 8, 1702019. Applications. J. Mater. Chem. A 2015, 3, 14996−15000.
(294) Yang, Z.; Rajagopal, A.; Chueh, C.-C.; Jo, S. B.; Liu, B.; Zhao, (312) Chung, I.; Song, J.-H.; Im, J.; Androulakis, J.; Malliakas, C. D.;
T.; Jen, A. K. Y. Stable Low-Bandgap Pb−Sn Binary Perovskites for Li, H.; Freeman, A. J.; Kenney, J. T.; Kanatzidis, M. G. CsSnI3:
Tandem Solar Cells. Adv. Mater. 2016, 28, 8990−8997. Semiconductor or Metal? High Electrical Conductivity and Strong
(295) Yu, Y.; Zhao, D.; Grice, C. R.; Meng, W.; Wang, C.; Liao, W.; Near-Infrared Photoluminescence from a Single Material. High Hole
Cimaroli, A. J.; Zhang, H.; Zhu, K.; Yan, Y. Thermally Evaporated Mobility and Phase-Transitions. J. Am. Chem. Soc. 2012, 134, 8579−
Methylammonium Tin Triiodide Thin Films for Lead-free Perovskite 8587.
Solar Cell Fabrication. RSC Adv. 2016, 6, 90248−90254. (313) Shum, K.; Chen, Z.; Qureshi, J.; Yu, C.; Wang, J. J.;
(296) Yokoyama, T.; Cao, D. H.; Stoumpos, C. C.; Song, T.-B.; Pfenninger, W.; Vockic, N.; Midgley, J.; Kenney, J. T. Synthesis and
Sato, Y.; Aramaki, S.; Kanatzidis, M. G. Overcoming Short-Circuit in Characterization of CsSnI3 Thin Films. Appl. Phys. Lett. 2010, 96,
Lead-Free CH3NH3SnI3 Perovskite Solar Cells via Kinetically 221903.
Controlled Gas−Solid Reaction Film Fabrication Process. J. Phys. (314) Kumar, M. H.; Dharani, S.; Leong, W. L.; Boix, P. P.;
Chem. Lett. 2016, 7, 776−782. Prabhakar, R. R.; Baikie, T.; Shi, C.; Ding, H.; Ramesh, R.; Asta, M.;
(297) Zhao, B.; Abdi-Jalebi, M.; Tabachnyk, M.; Glass, H.; Kamboj, et al. Lead-Free Halide Perovskite Solar Cells with High Photo-
V. S.; Nie, W.; Pearson, A. J.; Puttisong, Y.; Gödel, K. C.; Beere, H. E.; currents Realized Through Vacancy Modulation. Adv. Mater. 2014,
et al. High Open-Circuit Voltages in Tin-Rich Low-Bandgap 26, 7122−7127.
Perovskite-Based Planar Heterojunction Photovoltaics. Adv. Mater. (315) Ma, L.; Hao, F.; Stoumpos, C. C.; Phelan, B. T.; Wasielewski,
2017, 29, 1604744. M. R.; Kanatzidis, M. G. Carrier Diffusion Lengths of over 500 nm in
(298) Yang, Z.; Rajagopal, A.; Jo, S. B.; Chueh, C.-C.; Williams, S.; Lead-Free Perovskite CH3NH3SnI3 Films. J. Am. Chem. Soc. 2016,
Huang, C.-C.; Katahara, J. K.; Hillhouse, H. W.; Jen, A. K. Y. 138, 14750−14755.
Stabilized Wide Bandgap Perovskite Solar Cells by Tin Substitution. (316) Do Kim, H.; Miyamoto, Y.; Kubota, H.; Yamanari, T.; Ohkita,
Nano Lett. 2016, 16, 7739−7747. H. Open-circuit Voltage Loss in CH3NH3SnI3 Perovskite Solar Cells.
(299) Yokoyama, T.; Song, T.-B.; Cao, D. H.; Stoumpos, C. C.; Chem. Lett. 2017, 46, 253−256.
Aramaki, S.; Kanatzidis, M. G. The Origin of Lower Hole Carrier (317) Fujihara, T.; Terakawa, S.; Matsushima, T.; Qin, C.; Yahiro,
Concentration in Methylammonium Tin Halide Films Grown by a M.; Adachi, C. Fabrication of High Coverage MASnI3 Perovskite
Vapor-Assisted Solution Process. ACS Energy Lett. 2017, 2, 22−28. Films for Stable, Planar Heterojunction Solar Cells. J. Mater. Chem. C
(300) Jung, M.-C.; Raga, S. R.; Qi, Y. Properties and Solar Cell 2017, 5, 1121−1127.
Applications of Pb-free Perovskite Films Formed by Vapor (318) Ke, W.; Stoumpos, C. C.; Logsdon, J. L.; Wasielewski, M. R.;
Deposition. RSC Adv. 2016, 6, 2819−2825. Yan, Y.; Fang, G.; Kanatzidis, M. G. TiO2−ZnS Cascade Electron

BK DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Transport Layer for Efficient Formamidinium Tin Iodide Perovskite (335) Nagane, S.; Ghosh, D.; Hoye, R. L. Z.; Zhao, B.; Ahmad, S.;
Solar Cells. J. Am. Chem. Soc. 2016, 138, 14998−15003. Walker, A. B.; Islam, M. S.; Ogale, S.; Sadhanala, A. Lead-Free
(319) Liao, W.; Zhao, D.; Yu, Y.; Grice, C. R.; Wang, C.; Cimaroli, Perovskite Semiconductors Based on Germanium−Tin Solid
A. J.; Schulz, P.; Meng, W.; Zhu, K.; Xiong, R.-G.; et al. Lead-Free Solutions: Structural and Optoelectronic Properties. J. Phys. Chem.
Inverted Planar Formamidinium Tin Triiodide Perovskite Solar Cells C 2018, 122, 5940−5947.
Achieving Power Conversion Efficiencies up to 6.22%. Adv. Mater. (336) Ito, N.; Kamarudin, M. A.; Hirotani, D.; Zhang, Y.; Shen, Q.;
2016, 28, 9333−9340. Ogomi, Y.; Iikubo, S.; Minemoto, T.; Yoshino, K.; Hayase, S. Mixed
(320) Li, W.; Li, J.; Li, J.; Fan, J.; Mai, Y.; Wang, L. Addictive- Sn−Ge Perovskite for Enhanced Perovskite Solar Cell Performance in
assisted Construction of All-inorganic CsSnIBr2 Mesoscopic Perov- Air. J. Phys. Chem. Lett. 2018, 9, 1682−1688.
skite Solar Cells with Superior Thermal Stability up to 473 K. J. Mater. (337) Vargas, B.; Ramos, E.; Pérez-Gutiérrez, E.; Alonso, J. C.; Solis-
Chem. A 2016, 4, 17104−17110. Ibarra, D. A Direct Bandgap Copper−Antimony Halide Perovskite. J.
(321) Gupta, S.; Bendikov, T.; Hodes, G.; Cahen, D. CsSnBr3, A Am. Chem. Soc. 2017, 139, 9116−9119.
Lead-Free Halide Perovskite for Long-Term Solar Cell Application: (338) Kawai, T.; Ishii, A.; Kitamura, T.; Shimanuki, S.; Iwata, M.;
Insights on SnF2 Addition. ACS Energy Lett. 2016, 1, 1028−1033. Ishibashi, Y. Optical Absorption in Band-Edge Region of
(322) Sabba, D.; Mulmudi, H. K.; Prabhakar, R. R.; Krishnamoorthy,
(CH3NH3)3Bi2I9 Single Crystals. J. Phys. Soc. Jpn. 1996, 65, 1464−
T.; Baikie, T.; Boix, P. P.; Mhaisalkar, S.; Mathews, N. Impact of
1468.
Anionic Br− Substitution on Open Circuit Voltage in Lead Free
(339) Kawai, T.; Shimanuki, S. Optical Studies of (CH3NH3)3Bi2I9
Perovskite (CsSnI3‑xBrx) Solar Cells. J. Phys. Chem. C 2015, 119,
Single Crystals. Phys. Status Solidi B 1993, 177, K43−K45.
1763−1767.
(340) Arakcheeva, A.; Bonin, M.; Chapuis, G.; Zaitsev, A. I. The
(323) Zhao, Z.; Gu, F.; Li, Y.; Sun, W.; Ye, S.; Rao, H.; Liu, Z.; Bian,
Z.; Huang, C. Mixed-Organic-Cation Tin Iodide for Lead-Free phases of Cs3Bi2I9 between RT and 190 K. Z. Kristallogr. - Cryst.
Perovskite Solar Cells with an Efficiency of 8.12%. Adv. Sci. 2017, 4, Mater. 1999, 214, 279−283.
1700204. (341) Arakcheeva, A.; Chapuis, G.; Meyer, M. The LT phase of
(324) Ke, W.; Stoumpos, C. C.; Zhu, M.; Mao, L.; Spanopoulos, I.; Cs3Bi2I9. Z. Kristallogr. - Cryst. Mater. 2001, 216, 199−205.
Liu, J.; Kontsevoi, O. Y.; Chen, M.; Sarma, D.; Zhang, Y.; et al. (342) Szklarz, P.; Pietraszko, A.; Jakubas, R.; Bator, G.; Zieliński, P.;
Enhanced Photovoltaic Performance and Stability with a New Type of Gałazka, M. Structure, Phase Transitions and Molecular Dynamics of
Hollow 3D Perovskite {en}FASnI3. Sci. Adv. 2017, 3, No. e1701293. [C(NH2)3]3[M2I9], M = Sb, Bi. J. Phys.: Condens. Matter 2008, 20,
(325) Cao, D. H.; Stoumpos, C. C.; Yokoyama, T.; Logsdon, J. L.; 255221.
Song, T.-B.; Farha, O. K.; Wasielewski, M. R.; Hupp, J. T.; Kanatzidis, (343) Eckhardt, K.; Bon, V.; Getzschmann, J.; Grothe, J.; Wisser, F.
M. G. Thin Films and Solar Cells Based on SemiconductingTwo- M.; Kaskel, S. Crystallographic Insights into (CH3NH3)3(Bi2I9): A
DimensionalRuddlesden−Popper New Lead-free Hybrid Organic−inorganic Material as a Potential
(CH3(CH2)3NH3)2(CH3NH3)n−1SnnI3n+1 Perovskites. ACS Energy Absorber for Photovoltaics. Chem. Commun. 2016, 52, 3058−3060.
Lett. 2017, 2, 982−990. (344) Yang, Z.; Surrente, A.; Galkowski, K.; Bruyant, N.; Maude, D.
(326) Marshall, K. P.; Walton, R. I.; Hatton, R. A. Tin Perovskite/ K.; Haghighirad, A. A.; Snaith, H. J.; Plochocka, P.; Nicholas, R. J.
Fullerene Planar Layer Photovoltaics: Improving the Efficiency and Unraveling the Exciton Binding Energy and the Dielectric Constant in
Stability of Lead-free Devices. J. Mater. Chem. A 2015, 3, 11631− Single-Crystal Methylammonium Lead Triiodide Perovskite. J. Phys.
11640. Chem. Lett. 2017, 8, 1851−1855.
(327) Wang, N.; Zhou, Y.; Ju, M.-G.; Garces, H. F.; Ding, T.; Pang, (345) Hoye, R. L. Z.; Brandt, R. E.; Osherov, A.; Stevanović, V.;
S.; Zeng, X. C.; Padture, N. P.; Sun, X. W. Solar Cells: Stranks, S. D.; Wilson, M. W. B.; Kim, H.; Akey, A. J.; Perkins, J. D.;
Heterojunction-Depleted Lead-Free Perovskite Solar Cells with Kurchin, R. C.; et al. Methylammonium Bismuth Iodide as a Lead-
Coarse-Grained B-γ-CsSnI3 Thin Films. Adv. Energy Mater. 2016, 6, Free, Stable Hybrid Organic−Inorganic Solar Absorber. Chem. - Eur. J.
1601130. 2016, 22, 2605−2610.
(328) Sun, P.-P.; Li, Q.-S.; Yang, L.-N.; Li, Z.-S. Theoretical Insights (346) Park, B.-W.; Philippe, B.; Zhang, X.; Rensmo, H.; Boschloo,
into a Potential Lead-free Hybrid Perovskite: Substituting Pb2+ with G.; Johansson, E. M. J. Bismuth Based Hybrid Perovskites A3Bi2I9 (A:
Ge2+. Nanoscale 2016, 8, 1503−1512. Methylammonium or Cesium) for Solar Cell Application. Adv. Mater.
(329) Tang, L.-C.; Chang, Y.-C.; Huang, J.-Y.; Lee, M.-H.; Chang, 2015, 27, 6806−6813.
C.-S. First Principles Calculations of Linear and Second-Order (347) Singh, T.; Kulkarni, A.; Ikegami, M.; Miyasaka, T. Effect of
Optical Responses in Rhombohedrally Distorted Perovskite Ternary Electron Transporting Layer on Bismuth-Based Lead-Free Perovskite
Halides, CsGeX3 (X = Cl, Br, and I). Jpn. J. Appl. Phys. 2009, 48, (CH3NH3)3Bi2I9 for Photovoltaic Applications. ACS Appl. Mater.
112402. Interfaces 2016, 8, 14542−14547.
(330) Krishnamoorthy, T.; Ding, H.; Yan, C.; Leong, W. L.; Baikie, (348) Kulkarni, A.; Singh, T.; Ikegami, M.; Miyasaka, T. Photo-
T.; Zhang, Z.; Sherburne, M.; Li, S.; Asta, M.; Mathews, N.; et al.
voltaic Enhancement of Bismuth Halide Hybrid Perovskite by N-
Lead-free Germanium Iodide Perovskite Materials for Photovoltaic
methyl Pyrrolidone-assisted Morphology Conversion. RSC Adv. 2017,
Applications. J. Mater. Chem. A 2015, 3, 23829−23832.
(331) Stoumpos, C. C.; Frazer, L.; Clark, D. J.; Kim, Y. S.; Rhim, S. 7, 9456−9460.
(349) Zhang, Z.; Li, X.; Xia, X.; Wang, Z.; Huang, Z.; Lei, B.; Gao, Y.
H.; Freeman, A. J.; Ketterson, J. B.; Jang, J. I.; Kanatzidis, M. G.
Hybrid Germanium Iodide Perovskite Semiconductors: Active Lone High-Quality (CH3NH3)3Bi2I9 Film-Based Solar Cells: Pushing
Pairs, Structural Distortions, Direct and Indirect Energy Gaps, and Efficiency up to 1.64%. J. Phys. Chem. Lett. 2017, 8 (17), 4300−4307.
Strong Nonlinear Optical Properties. J. Am. Chem. Soc. 2015, 137, (350) Jain, S. M.; Phuyal, D.; Davies, M. L.; Li, M.; Philippe, B.; De
6804−6819. Castro, C.; Qiu, Z.; Kim, J.; Watson, T.; Tsoi, W. C.; et al. An
(332) Ming, W.; Shi, H.; Du, M.-H. Large Dielectric Constant, High Effective Approach of Vapour Assisted Morphological Tailoring for
Acceptor Density, and Deep Electron Traps in Perovskite Solar Cell Reducing Metal Defect Sites in Lead-free, (CH3NH3)3Bi2I9 Bismuth-
Material CsGeI3. J. Mater. Chem. A 2016, 4, 13852−13858. based Perovskite Solar Cells for Improved Performance and Long-
(333) Giorgi, G.; Yamashita, K. Alternative, Lead-free, Hybrid term Stability. Nano Energy 2018, 49, 614−624.
Organic−Inorganic Perovskites for Solar Applications: A DFT (351) Vigneshwaran, M.; Ohta, T.; Iikubo, S.; Kapil, G.; Ripolles, T.
Analysis. Chem. Lett. 2015, 44, 826−828. S.; Ogomi, Y.; Ma, T.; Pandey, S. S.; Shen, Q.; Toyoda, T.; et al.
(334) Ju, M.-G.; Dai, J.; Ma, L.; Zeng, X. C. Lead-Free Mixed Tin Facile Synthesis and Characterization of Sulfur Doped Low Bandgap
and Germanium Perovskites for Photovoltaic Application. J. Am. Bismuth Based Perovskites by Soluble Precursor Route. Chem. Mater.
Chem. Soc. 2017, 139, 8038−8043. 2016, 28, 6436−6440.

BL DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

(352) Hu, H.; Dong, B.; Zhang, W. Low-toxic Metal Halide Diffusion Length in High-Quality Lead-Free Double Perovskite Films.
Perovskites: Opportunities and Future Challenges. J. Mater. Chem. A Adv. Mater. 2018, 30, 1706246.
2017, 5, 11436−11449. (370) Savory, C. N.; Walsh, A.; Scanlon, D. O. Can Pb-Free Halide
(353) Saparov, B.; Hong, F.; Sun, J.-P.; Duan, H.-S.; Meng, W.; Double Perovskites Support High-Efficiency Solar Cells? ACS Energy
Cameron, S.; Hill, I. G.; Yan, Y.; Mitzi, D. B. Thin-Film Preparation Lett. 2016, 1, 949−955.
and Characterization of Cs3Sb2I9: A Lead-Free Layered Perovskite (371) Zhou, C.; Lin, H.; Lee, S.; Chaaban, M.; Ma, B. Organic−
Semiconductor. Chem. Mater. 2015, 27, 5622−5632. inorganic Metal Halide Hybrids Beyond Perovskites. Mater. Res. Lett.
(354) Brandt, R. E.; Stevanović, V.; Ginley, D. S.; Buonassisi, T. 2018, 6, 552−569.
Identifying Defect-tolerant Semiconductors with High Minority- (372) Sun, Y.-Y.; Shi, J.; Lian, J.; Gao, W.; Agiorgousis, M. L.;
carrier Lifetimes: Beyond Hybrid Lead Halide Perovskites. MRS Zhang, P.; Zhang, S. Discovering Lead-free Perovskite Solar Materials
Commun. 2015, 5, 265−275. with a Split-anion Approach. Nanoscale 2016, 8, 6284−6289.
(355) Harikesh, P. C.; Mulmudi, H. K.; Ghosh, B.; Goh, T. W.; (373) Nie, R.; Mehta, A.; Park, B.-w.; Kwon, H.-W.; Im, J.; Seok, S. I.
Teng, Y. T.; Thirumal, K.; Lockrey, M.; Weber, K.; Koh, T. M.; Li, S.; Mixed Sulfur and Iodide-Based Lead-Free Perovskite Solar Cells. J.
et al. Rb as an Alternative Cation for Templating Inorganic Lead-Free Am. Chem. Soc. 2018, 140, 872−875.
Perovskites for Solution Processed Photovoltaics. Chem. Mater. 2016, (374) Kim, Y.; Yang, Z.; Jain, A.; Voznyy, O.; Kim, G.-H.; Liu, M.;
28, 7496−7504. Quan, L. N.; García de Arquer, F. P.; Comin, R.; Fan, J. Z.; et al. Pure
(356) Boopathi, K. M.; Karuppuswamy, P.; Singh, A.; Hanmandlu, Cubic-Phase Hybrid Iodobismuthates AgBi2I7 for Thin-Film Photo-
C.; Lin, L.; Abbas, S. A.; Chang, C. C.; Wang, P. C.; Li, G.; Chu, C. voltaics. Angew. Chem. 2016, 128, 9738−9742.
W. Solution-processable Antimony-based Light-absorbing Materials (375) Zhu, H.; Pan, M.; Johansson, M. B.; Johansson, E. M. J. High
Beyond Lead Halide Perovskites. J. Mater. Chem. A 2017, 5, 20843− Photon-to-Current Conversion in Solar Cells Based on Light-
20850. Absorbing Silver Bismuth Iodide. ChemSusChem 2017, 10, 2592−
(357) Jiang, F.; Yang, D.; Jiang, Y.; Liu, T.; Zhao, X.; Ming, Y.; Luo, 2596.
B.; Qin, F.; Fan, J.; Han, H.; et al. Chlorine-Incorporation-Induced (376) Shao, Z.; Le Mercier, T.; Madec, M. B.; Pauporté, T.
Formation of the Layered Phase for Antimony-Based Lead-Free Exploring AgBixI3x+1 Semiconductor Thin Films for Lead-free
Perovskite Solar Cells. J. Am. Chem. Soc. 2018, 140, 1019−1027. Perovskite Solar Cells. Mater. Des. 2018, 141, 81−87.
(358) Zhang, X.; Wu, G.; Gu, Z.; Guo, B.; Liu, W.; Yang, S.; Ye, T.; (377) Khazaee, M.; Sardashti, K.; Sun, J.-P.; Zhou, H.; Clegg, C.;
Chen, C.; Tu, W.; Chen, H. Active-layer Evolution and Efficiency Hill, I. G.; Jones, J. L.; Lupascu, D. C.; Mitzi, D. B. A Versatile Thin-
Improvement of (CH3NH3)3Bi2I9-based Solar Cell on TiO2-deposited Film Deposition Method for Multidimensional Semiconducting
ITO substrate. Nano Res. 2016, 9, 2921−2930. Bismuth Halides. Chem. Mater. 2018, 30, 3538−3544.
(359) Ran, C.; Wu, Z.; Xi, J.; Yuan, F.; Dong, H.; Lei, T.; He, X.; (378) Jung, K. W.; Sohn, M. R.; Lee, H. M.; Yang, I. S.; Sung, S. D.;
Hou, X. Construction of Compact Methylammonium Bismuth Iodide Kim, J.; Wei-Guang Diau, E.; Lee, W. I. Silver Bismuth Iodides in
Film Promoting Lead-Free Inverted Planar Heterojunction Organo- Various Compositions as Potential Pb-free Light Absorbers for
halide Solar Cells with Open-Circuit Voltage over 0.8 V. J. Phys. Hybrid Solar Cells. Sustainable Energy Fuels 2018, 2, 294−302.
Chem. Lett. 2017, 8, 394−400. (379) Cui, X.-P.; Jiang, K.-J.; Huang, J.-H.; Zhang, Q.-Q.; Su, M.-J.;
(360) Mali, S. S.; Kim, H.; Kim, D.-H.; Kook Hong, C. Anti-Solvent Yang, L.-M.; Song, Y.-L.; Zhou, X.-Q. Cupric Bromide Hybrid
Assisted Crystallization Processed Methylammonium Bismuth Iodide Perovskite Heterojunction Solar Cells. Synth. Met. 2015, 209, 247−
Cuboids towards Highly Stable Lead-Free Perovskite Solar Cells. 250.
ChemistrySelect 2017, 2, 1578−1585. (380) Cortecchia, D.; Dewi, H. A.; Yin, J.; Bruno, A.; Chen, S.;
(361) Li, T.; Hu, Y.; Morrison, C. A.; Wu, W.; Han, H.; Robertson, Baikie, T.; Boix, P. P.; Grätzel, M.; Mhaisalkar, S.; Soci, C.; et al. Lead-
N. Lead-free Pseudo-three-dimensional Organic−inorganic Iodobis- Free MA2CuClxBr4−x Hybrid Perovskites. Inorg. Chem. 2016, 55,
muthates for Photovoltaic Applications. Sustainable Energy Fuels 2017, 1044−1052.
1, 308−316. (381) Li, X.; Zhong, X.; Hu, Y.; Li, B.; Sheng, Y.; Zhang, Y.; Weng,
(362) Johansson, M. B.; Zhu, H.; Johansson, E. M. J. Extended C.; Feng, M.; Han, H.; Wang, J. Organic−Inorganic Copper(II)-
Photo-Conversion Spectrum in Low-Toxic Bismuth Halide Perovskite Based Material: A Low-Toxic, Highly Stable Light Absorber for
Solar Cells. J. Phys. Chem. Lett. 2016, 7, 3467−3471. Photovoltaic Application. J. Phys. Chem. Lett. 2017, 8, 1804−1809.
(363) Hebig, J.-C.; Kühn, I.; Flohre, J.; Kirchartz, T. Optoelectronic (382) Nie, R.; Yun, H.-s.; Paik, M.-J.; Mehta, A.; Park, B.-w.; Choi,
Properties of (CH3 NH3)3Sb2I 9 Thin Films for Photovoltaic Y. C.; Seok, S. I. Efficient Solar Cells Based on Light-Harvesting
Applications. ACS Energy Lett. 2016, 1, 309−314. Antimony Sulfoiodide. Adv. Energy Mater. 2018, 8, 1701901.
(364) Yang, R. X.; Butler, K. T.; Walsh, A. Assessment of Hybrid (383) Kulkarni, A.; Singh, T.; Jena, A. K.; Pinpithak, P.; Ikegami, M.;
Organic−Inorganic Antimony Sulfides for Earth-Abundant Photo- Miyasaka, T. Vapor Annealing Controlled Crystal Growth and
voltaic Applications. J. Phys. Chem. Lett. 2015, 6, 5009−5014. Photovoltaic Performance of Bismuth Triiodide Embedded in
(365) Slavney, A. H.; Hu, T.; Lindenberg, A. M.; Karunadasa, H. I. A Mesostructured Configurations. ACS Appl. Mater. Interfaces 2018,
Bismuth-Halide Double Perovskite with Long Carrier Recombination 10, 9547−9554.
Lifetime for Photovoltaic Applications. J. Am. Chem. Soc. 2016, 138, (384) Hamdeh, U. H.; Nelson, R. D.; Ryan, B. J.; Bhattacharjee, U.;
2138−2141. Petrich, J. W.; Panthani, M. G. Solution-Processed BiI3 Thin Films for
(366) McClure, E. T.; Ball, M. R.; Windl, W.; Woodward, P. M. Photovoltaic Applications: Improved Carrier Collection via Solvent
Cs2AgBiX6 (X = Br, Cl): New Visible Light Absorbing, Lead-Free Annealing. Chem. Mater. 2016, 28, 6567−6574.
Halide Perovskite Semiconductors. Chem. Mater. 2016, 28, 1348− (385) Zhu, L.; Yuh, B.; Schoen, S.; Li, X.; Aldighaithir, M.;
1354. Richardson, B. J.; Alamer, A.; Yu, Q. Solvent-molecule-mediated
(367) Greul, E.; Petrus, M. L.; Binek, A.; Docampo, P.; Bein, T. manipulation of crystalline grains for efficient planar binary lead and
Highly Stable, Phase Pure Cs2AgBiBr6 Double Perovskite Thin Films tin triiodide perovskite solar cells. Nanoscale 2016, 8, 7621−7630.
for Optoelectronic Applications. J. Mater. Chem. A 2017, 5, 19972− (386) Zhu, H. L.; Xiao, J.; Mao, J.; Zhang, H.; Zhao, Y.; Choy, W. C.
19981. H. Controllable Crystallization of CH3NH3Sn0.25Pb0.75I3 Perovskites
(368) Wu, C.; Zhang, Q.; Liu, Y.; Luo, W.; Guo, X.; Huang, Z.; for Hysteresis-Free Solar Cells with Efficiency Reaching 15.2%. Adv.
Ting, H.; Sun, W.; Zhong, X.; Wei, S.; et al. The Dawn of Lead-Free Funct. Mater. 2017, 27, 1605469.
Perovskite Solar Cell: Highly Stable Double Perovskite Cs2AgBiBr6 (387) Zuo, F.; Williams, S. T.; Liang, P.-W.; Chueh, C.-C.; Liao, C.-
Film. Adv. Sci. 2018, 5, 1700759. Y.; Jen, A. K. Y. Binary-Metal Perovskites Toward High-Performance
(369) Ning, W.; Wang, F.; Wu, B.; Lu, J.; Yan, Z.; Liu, X.; Tao, Y.; Planar-Heterojunction Hybrid Solar Cells. Adv. Mater. 2014, 26,
Liu, J.-M.; Huang, W.; Fahlman, M.; et al. Long Electron−Hole 6454−6460.

BM DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

(388) Li, L.; Zhang, F.; Hao, Y.; Sun, Q.; Li, Z.; Wang, H.; Cui, Y.; (405) Ono, L. K.; Leyden, M. R.; Wang, S.; Qi, Y. Organometal
Zhu, F. High Efficiency Planar Sn−Pb Binary Perovskite Solar Cells: Halide Perovskite Thin Films and Solar Cells by Vapor Deposition. J.
Controlled Growth of Large Grains via a One-step Solution Mater. Chem. A 2016, 4, 6693−6713.
Fabrication Process. J. Mater. Chem. C 2017, 5, 2360−2367. (406) Leyden, M. R.; Jiang, Y.; Qi, Y. Chemical Vapor Deposition
(389) Eperon, G. E.; Leijtens, T.; Bush, K. A.; Prasanna, R.; Green, Grown Formamidinium Perovskite Solar Modules with High Steady
T.; Wang, J. T.-W.; McMeekin, D. P.; Volonakis, G.; Milot, R. L.; State Power and Thermal Stability. J. Mater. Chem. A 2016, 4, 13125−
May, R.; et al. Perovskite-perovskite Tandem Photovoltaics with 13132.
Optimized Bandgaps. Science 2016, 354, 861−865. (407) Chen, H.; Wei, Z.; Zheng, X.; Yang, S. A Scalable
(390) Liao, W.; Zhao, D.; Yu, Y.; Shrestha, N.; Ghimire, K.; Grice, Electrodeposition Route to the Low-cost, Versatile and Controllable
C. R.; Wang, C.; Xiao, Y.; Cimaroli, A. J.; Ellingson, R. J.; et al. Fabrication of Perovskite Solar Cells. Nano Energy 2015, 15, 216−
Fabrication of Efficient Low-Bandgap Perovskite Solar Cells by 226.
Combining Formamidinium Tin Iodide with Methylammonium Lead (408) Charles, U. A.; Ibrahim, M. A.; Teridi, M. A. M.
Iodide. J. Am. Chem. Soc. 2016, 138, 12360−12363. Electrodeposition of Organic−inorganic Tri-halide Perovskites Solar
(391) Li, M.; Wang, Z.-K.; Zhuo, M.-P.; Hu, Y.; Hu, K.-H.; Ye, Q.- Cell. J. Power Sources 2018, 378, 717−731.
Q.; Jain, S. M.; Yang, Y.-G.; Gao, X.-Y.; Liao, L.-S. Pb−Sn−Cu (409) Li, Z.; Klein, T. R.; Kim, D. H.; Yang, M.; Berry, J. J.; van
Ternary Organometallic Halide Perovskite Solar Cells. Adv. Mater. Hest, M. F. A. M.; Zhu, K. Scalable Fabrication of Perovskite Solar
2018, 30, 1800258. Cells. Nat. Rev. Mater. 2018, 3, 18017.
(392) McMeekin, D. P.; Sadoughi, G.; Rehman, W.; Eperon, G. E.; (410) Di Giacomo, F.; Shanmugam, S.; Fledderus, H.; Bruijnaers, B.
Saliba, M.; Hörantner, M. T.; Haghighirad, A.; Sakai, N.; Korte, L.; J.; Verhees, W. J. H.; Dorenkamper, M. S.; Veenstra, S. C.; Qiu, W.;
Rech, B.; et al. A mixed-cation Lead Mixed-halide Perovskite Gehlhaar, R.; Merckx, T.; et al. Up-scalable Sheet-to-Sheet Production
Absorber for Tandem Solar Cells. Science 2016, 351, 151−155. of High Efficiency Perovskite Module and Solar Cells on 6-in.
(393) Tang, S.; Deng, Y.; Zheng, X.; Bai, Y.; Fang, Y.; Dong, Q.; Substrate using Slot die Coating. Sol. Energy Mater. Sol. Cells 2018,
Wei, H.; Huang, J. Composition Engineering in Doctor-Blading of 181, 53−59.
Perovskite Solar Cells. Adv. Energy Mater. 2017, 7, 1700302. (411) Chen, H.; Ye, F.; Tang, W.; He, J.; Yin, M.; Wang, Y.; Xie, F.;
(394) Yang, M.; Li, Z.; Reese, M. O.; Reid, O. G.; Kim, D. H.; Siol, Bi, E.; Yang, X.; Grätzel, M.; Han, L.; et al. A solvent- and Vacuum-
S.; Klein, T. R.; Yan, Y.; Berry, J. J.; van Hest, M. F. A. M.; et al. free Route to Large-area Perovskite Films for Efficient Solar Modules.
Perovskite Ink with Wide Processing Window for Scalable High- Nature 2017, 550, 92−95.
efficiency Solar Cells. Nat. Energy 2017, 2, 17038. (412) Agresti, A.; Pescetelli, S.; Palma, A. L.; Del Rio Castillo, A. E.;
(395) Kim, J.-E.; Jung, Y.-S.; Heo, Y.-J.; Hwang, K.; Qin, T.; Kim,
Konios, D.; Kakavelakis, G.; Razza, S.; Cinà, L.; Kymakis, E.;
D.-Y.; Vak, D. Slot Die Coated Planar Perovskite Solar Cells via
Bonaccorso, F.; et al. Graphene Interface Engineering for Perovskite
Blowing and Heating Assisted One Step Deposition. Sol. Energy
Solar Modules: 12.6% Power Conversion Efficiency over 50 cm2
Mater. Sol. Cells 2018, 179, 80−86.
Active Area. ACS Energy Lett. 2017, 2, 279−287.
(396) Whitaker, J. B.; Kim, D. H.; Larson, B. W.; Zhang, F.; Berry, J.
(413) Hu, Y.; Si, S.; Mei, A.; Rong, Y.; Liu, H.; Li, X.; Han, H. Stable
J.; van Hest, M. F. A. M.; Zhu, K. Scalable Slot-die Coating of High
Large-Area (10 × 10 cm2) Printable Mesoscopic Perovskite Module
Performance Perovskite Solar Cells. Sustainable Energy Fuels 2018, 2,
Exceeding 10% Efficiency. Solar RRL 2017, 1, 1600019.
2442−2449.
(414) Wilkinson, B.; Chang, N. L.; Green, M. A.; Ho-Baillie, A. W.
(397) Lee, D.; Jung, Y.-S.; Heo, Y.-J.; Lee, S.; Hwang, K.; Jeon, Y.-J.;
Kim, J.-E.; Park, J.; Jung, G. Y.; Kim, D.-Y. Slot-Die Coated Perovskite Y. Scaling Limits to Large Area Perovskite Solar Cell Efficiency. Prog.
Films Using Mixed Lead Precursors for Highly Reproducible and Photovoltaics 2018, 26, 659−674.
(415) Kim, J.; Yun, J. S.; Cho, Y.; Lee, D. S.; Wilkinson, B.; Soufiani,
Large-Area Solar Cells. ACS Appl. Mater. Interfaces 2018, 10, 16133−
16139. A. M.; Deng, X.; Zheng, J.; Shi, A.; Lim, S.; et al. Overcoming the
(398) Mathies, F.; Eggers, H.; Richards, B. S.; Hernandez-Sosa, G.; Challenges of Large-Area High-Efficiency Perovskite Solar Cells. ACS
Lemmer, U.; Paetzold, U. W. Inkjet-Printed Triple Cation Perovskite Energy Lett. 2017, 2, 1978−1984.
Solar Cells. ACS Appl. Energy Mater. 2018, 1, 1834−1839. (416) Zhou, Y.; Yang, M.; Wu, W.; Vasiliev, A. L.; Zhu, K.; Padture,
(399) Mathies, F.; Abzieher, T.; Hochstuhl, A.; Glaser, K.; N. P. Room-temperature Crystallization of Hybrid-perovskite Thin
Colsmann, A.; Paetzold, U. W.; Hernandez-Sosa, G.; Lemmer, U.; Films via Solvent−solvent Extraction for High-Performance Solar
Quintilla, A. Multipass Inkjet Printed Planar Methylammonium Lead Cells. J. Mater. Chem. A 2015, 3, 8178−8184.
Iodide Perovskite Solar Cells. J. Mater. Chem. A 2016, 4, 19207− (417) Lavery, B. W.; Kumari, S.; Konermann, H.; Draper, G. L.;
19213. Spurgeon, J.; Druffel, T. Intense Pulsed Light Sintering of CH3NH3-
(400) Li, S.-G.; Jiang, K.-J.; Su, M.-J.; Cui, X.-P.; Huang, J.-H.; PbI3 Solar Cells. ACS Appl. Mater. Interfaces 2016, 8, 8419−8426.
Zhang, Q.-Q.; Zhou, X.-Q.; Yang, L.-M.; Song, Y.-L. Inkjet Printing of (418) Troughton, J.; Carnie, M. J.; Davies, M. L.; Charbonneau, C.;
CH3NH3PbI3 on a Mesoscopic TiO2 Film for Highly Efficient Jewell, E. H.; Worsley, D. A.; Watson, T. M. Photonic Flash-annealing
Perovskite Solar Cells. J. Mater. Chem. A 2015, 3, 9092−9097. of Lead Halide Perovskite Solar Cells in 1 ms. J. Mater. Chem. A 2016,
(401) Cao, K.; Zuo, Z.; Cui, J.; Shen, Y.; Moehl, T.; Zakeeruddin, S. 4, 3471−3476.
M.; Grätzel, M.; Wang, M. Efficient Screen Printed Perovskite Solar (419) Tsai, H.; Nie, W.; Lin, Y.-H.; Blancon, J. C.; Tretiak, S.; Even,
Cells Based on Mesoscopic TiO2/Al2O3/NiO/carbon Architecture. J.; Gupta, G.; Ajayan, P. M.; Mohite, A. D. Effect of Precursor
Nano Energy 2015, 17, 171−179. Solution Aging on the Crystallinity and Photovoltaic Performance of
(402) Priyadarshi, A.; Bashir, A.; Gunawan, J. T.; Haur, L. J.; Bruno, Perovskite Solar Cells. Adv. Energy Mater. 2017, 7, 1602159.
A.; Akhter, Z.; Mathews, N.; Mhaisalkar, S. G. Simplified Architecture (420) Boonmongkolras, P.; Kim, D.; Alhabshi, E. M.; Gereige, I.;
of a Fully Printable Perovskite Solar Cell Using a Thick Zirconia Shin, B. Understanding Effects of Precursor Solution Aging in Triple
Layer. Energy Technol. 2017, 5, 1866−1872. Cation Lead Perovskite. RSC Adv. 2018, 8, 21551−21557.
(403) Mohamad, D. K.; Griffin, J.; Bracher, C.; Barrows, A. T.; (421) Qin, M.; Cao, J.; Zhang, T.; Mai, J.; Lau, T.-K.; Zhou, S.;
Lidzey, D. G. Spray-Cast Multilayer Organometal Perovskite Solar Zhou, Y.; Wang, J.; Hsu, Y.-J.; Zhao, N.; et al. Fused-Ring Electron
Cells Fabricated in Air. Adv. Energy Mater. 2016, 6, 1600994. Acceptor ITIC-Th: A Novel Stabilizer for Halide Perovskite Precursor
(404) Lan, D.-H.; Hong, S.-H.; Chou, L.-H.; Wang, X.-F.; Liu, C.-L. Solution. Adv. Energy Mater. 2018, 8, 1703399.
High Throughput Two-step Ultrasonic Spray Deposited CH3NH3- (422) Chegaar, M.; Hamzaoui, A.; Namoda, A.; Petit, P.; Aillerie,
PbI3 Thin Film Layer for Solar Cell Application. J. Power Sources M.; Herguth, A. Effect of Illumination Intensity on Solar Cells
2018, 390, 270−277. Parameters. Energy Procedia 2013, 36, 722−729.

BN DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

(423) Singh, P.; Singh, S. N.; Lal, M.; Husain, M. Temperature (440) Jena, A. K.; Chen, H.-W.; Kogo, A.; Sanehira, Y.; Ikegami, M.;
Dependence of I−V Characteristics and Performance Parameters of Miyasaka, T. The Interface between FTO and the TiO2 Compact
Silicon Solar Cell. Sol. Energy Mater. Sol. Cells 2008, 92, 1611−1616. Layer Can Be One of the Origins to Hysteresis in Planar
(424) Singh, T.; Ö z, S.; Sasinska, A.; Frohnhoven, R.; Mathur, S.; Heterojunction Perovskite Solar Cells. ACS Appl. Mater. Interfaces
Miyasaka, T. Sulfate-Assisted Interfacial Engineering for High Yield 2015, 7, 9817−9823.
and Efficiency of Triple Cation Perovskite Solar Cells with Alkali- (441) Cojocaru, L.; Uchida, S.; Jayaweera, P. V. V.; Kaneko, S.;
Doped TiO2 Electron-Transporting Layers. Adv. Funct. Mater. 2018, Nakazaki, J.; Kubo, T.; Segawa, H. Origin of the Hysteresis in I−V
28, 1706287. Curves for Planar Structure Perovskite Solar Cells Rationalized with a
(425) Liu, M.; Endo, M.; Shimazaki, A.; Wakamiya, A.; Tachibana, Surface Boundary-induced Capacitance Model. Chem. Lett. 2015, 44,
Y. Light Intensity Dependence of Performance of Lead Halide 1750−1752.
Perovskite Solar Cells. J. Photopolym. Sci. Technol. 2017, 30, 577−582. (442) Singh, T.; Ikegami, M.; Miyasaka, T. ACS Appl. Energy Mater.
(426) Tress, W.; Yavari, M.; Domanski, K.; Yadav, P.; Niesen, B.; 2018, 1 (12), 6741−6747.
Correa Baena, J. P.; Hagfeldt, A.; Graetzel, M. Interpretation and (443) Tan, Z.-K.; Moghaddam, R. S.; Lai, M. L.; Docampo, P.;
Evolution of Open-circuit Voltage, Recombination, Ideality Factor Higler, R.; Deschler, F.; Price, M.; Sadhanala, A.; Pazos, L. M.;
and Subgap Defect States during Reversible Light-soaking and Credgington, D.; et al. Bright Light-emitting Diodes Based on
Irreversible Degradation of Perovskite Solar Cells. Energy Environ. Organometal Halide Perovskite. Nat. Nanotechnol. 2014, 9, 687−692.
Sci. 2018, 11, 151−165. (444) Yang, X.; Zhang, X.; Deng, J.; Chu, Z.; Jiang, Q.; Meng, J.;
(427) Bush, K. A.; Palmstrom, A. F.; Yu, Z. J.; Boccard, M.; Wang, P.; Zhang, L.; Yin, Z.; You, J. Efficient Green Light-emitting
Cheacharoen, R.; Mailoa, J. P.; McMeekin, D. P.; Hoye, R. L. Z.; Diodes Based on Quasi-two-dimensional Composition and Phase
Bailie, C. D.; Leijtens, T.; et al. 23.6%-efficient Monolithic Perovskite/ Engineered Perovskite with Surface Passivation. Nat. Commun. 2018,
Silicon Tandem Solar Cells with Improved Stability. Nat. Energy 9, 570.
2017, 2, 17009. (445) Wang, J.; Wang, N.; Jin, Y.; Si, J.; Tan, Z.-K.; Du, H.; Cheng,
(428) Duong, T.; Wu, Y.; Shen, H.; Peng, J.; Fu, X.; Jacobs, D.; L.; Dai, X.; Bai, S.; He, H.; et al. Interfacial Control Toward Efficient
Wang, E.-C.; Kho, T. C.; Fong, K. C.; Stocks, M.; et al. Rubidium and Low-Voltage Perovskite Light-Emitting Diodes. Adv. Mater. 2015,
Multication Perovskite with Optimized Bandgap for Perovskite- 27, 2311−2316.
Silicon Tandem with over 26% Efficiency. Adv. Energy Mater. 2017, 7, (446) Dong, R.; Fang, Y.; Chae, J.; Dai, J.; Xiao, Z.; Dong, Q.; Yuan,
1700228. Y.; Centrone, A.; Zeng, X. C.; Huang, J. High-Gain and Low-Driving-
(429) Kranz, L.; Abate, A.; Feurer, T.; Fu, F.; Avancini, E.; Voltage Photodetectors Based on Organolead Triiodide Perovskites.
Lö ckinger, J.; Reinhard, P.; Zakeeruddin, S. M.; Grätzel, M.; Adv. Mater. 2015, 27, 1912−1918.
Buecheler, S.; et al. High-Efficiency Polycrystalline Thin Film Tandem (447) Moehl, T.; Im, J. H.; Lee, Y. H.; Domanski, K.; Giordano, F.;
Solar Cells. J. Phys. Chem. Lett. 2015, 6, 2676−2681. Zakeeruddin, S. M.; Dar, M. I.; Heiniger, L.-P.; Nazeeruddin, M. K.;
(430) Guchhait, A.; Dewi, H. A.; Leow, S. W.; Wang, H.; Han, G.;
Park, N.-G.; et al. Strong Photocurrent Amplification in Perovskite
Suhaimi, F. B.; Mhaisalkar, S.; Wong, L. H.; Mathews, N. Over 20%
Solar Cells with a Porous TiO2 Blocking Layer under Reverse Bias. J.
Efficient CIGS−Perovskite Tandem Solar Cells. ACS Energy Lett.
Phys. Chem. Lett. 2014, 5, 3931−3936.
2017, 2, 807−812.
(448) Li, D.; Dong, G.; Li, W.; Wang, L. High Performance Organic-
(431) Tan, H.; Jain, A.; Voznyy, O.; Lan, X.; García de Arquer, F. P.;
inorganic Perovskite-optocoupler Based on Low-voltage and Fast
Fan, J. Z.; Quintero-Bermudez, R.; Yuan, M.; Zhang, B.; Zhao, Y.;
Response Perovskite Compound Photodetector. Sci. Rep. 2015, 5,
et al. Efficient and Stable Solution-processed Planar Perovskite Solar
Cells via Contact Passivation. Science 2017, 355, 722−726. 7902.
(432) Rajagopal, A.; Yang, Z.; Jo, S. B.; Braly, I. L.; Liang, P.-W.; (449) Chen, H.-W.; Sakai, N.; Jena, A. K.; Sanehira, Y.; Ikegami, M.;
Hillhouse, H. W.; Jen, A. K. Y. Highly Efficient Perovskite−Perovskite Ho, K.-C.; Miyasaka, T. A Switchable High-Sensitivity Photodetecting
Tandem Solar Cells Reaching 80% of the Theoretical Limit in and Photovoltaic Device with Perovskite Absorber. J. Phys. Chem. Lett.
Photovoltage. Adv. Mater. 2017, 29, 1702140. 2015, 6, 1773−1779.
(433) Forgács, D.; Gil-Escrig, L.; Pérez-Del-Rey, D.; Momblona, C.; (450) Yakunin, S.; Shynkarenko, Y.; Dirin, D. N.; Cherniukh, I.;
Werner, J.; Niesen, B.; Ballif, C.; Sessolo, M.; Bolink, H. J. Efficient Kovalenko, M. V. Non-dissipative Internal Optical Filtering with
Monolithic Perovskite/Perovskite Tandem Solar Cells. Adv. Energy Solution-grown Perovskite Single Crystals for Full-colour Imaging.
Mater. 2017, 7, 1602121. NPG Asia Mater. 2017, 9, No. e431.
(434) Sha, W. E. I.; Ren, X.; Chen, L.; Choy, W. C. H. The (451) Yakunin, S.; Sytnyk, M.; Kriegner, D.; Shrestha, S.; Richter,
Efficiency Limit of CH3NH3PbI3 Perovskite Solar Cells. Appl. Phys. M.; Matt, G. J.; Azimi, H.; Brabec, C. J.; Stangl, J.; Kovalenko, M. V.;
Lett. 2015, 106, 221104. et al. Detection of X-ray Photons by Solution-processed Lead Halide
(435) Miller, O. D.; Yablonovitch, E.; Kurtz, S. R. Strong Internal Perovskites. Nat. Photonics 2015, 9, 444−449.
and External Luminescence as Solar Cells Approach the Shockley− (452) Wei, W.; Zhang, Y.; Xu, Q.; Wei, H.; Fang, Y.; Wang, Q.;
Queisser Limit. IEEE J. Photovolt. 2012, 2, 303−311. Deng, Y.; Li, T.; Gruverman, A.; Cao, L.; et al. Monolithic Integration
(436) Ogomi, Y.; Morita, A.; Tsukamoto, S.; Saitho, T.; Fujikawa, of Hybrid Perovskite Single Crystals with Heterogenous Substrate for
N.; Shen, Q.; Toyoda, T.; Yoshino, K.; Pandey, S. S.; Ma, T.; et al. Highly Sensitive X-ray Imaging. Nat. Photonics 2017, 11, 315−321.
CH3NH3SnxPb(1−x)I3 Perovskite Solar Cells Covering up to 1060 nm. (453) Kim, Y. C.; Kim, K. H.; Son, D.-Y.; Jeong, D.-N.; Seo, J.-Y.;
J. Phys. Chem. Lett. 2014, 5, 1004−1011. Choi, Y. S.; Han, I. T.; Lee, S. Y.; Park, N.-G. Printable
(437) Song, J.; Zheng, E.; Wang, X.-F.; Tian, W.; Miyasaka, T. Low- Organometallic Perovskite enables Large-area, Low-dose X-ray
temperature-processed ZnO−SnO2 Nanocomposite for Efficient Imaging. Nature 2017, 550, 87−91.
Planar Perovskite Solar Cells. Sol. Energy Mater. Sol. Cells 2016, (454) Lang, F.; Nickel, N. H.; Bundesmann, J.; Seidel, S.; Denker,
144, 623−630. A.; Albrecht, S.; Brus, V. V.; Rappich, J.; Rech, B.; Landi, G.; et al.
(438) Pinpithak, P.; Chen, H.-W.; Kulkarni, A.; Sanehira, Y.; Radiation Hardness and Self-Healing of Perovskite Solar Cells. Adv.
Ikegami, M.; Miyasaka, T. Low-temperature and Ambient Air Mater. 2016, 28, 8726−8731.
Processes of Amorphous SnOx-based Mixed Halide Perovskite Planar (455) Brus, V. V.; Lang, F.; Bundesmann, J.; Seidel, S.; Denker, A.;
Solar Cell. Chem. Lett. 2017, 46, 382−384. Rech, B.; Landi, G.; Neitzert, H. C.; Rappich, J.; Nickel, N. H. Defect
(439) Kogo, A.; Sanehira, Y.; Numata, Y.; Ikegami, M.; Miyasaka, T. Dynamics in Proton Irradiated CH3NH3PbI3 Perovskite Solar Cells.
Amorphous Metal Oxide Blocking Layers for Highly Efficient Low- Adv. Electron. Mater. 2017, 3, 1600438.
Temperature Brookite TiO2-based Perovskite Solar Cells. ACS Appl. (456) Miyazawa, Y.; Ikegami, M.; Chen, H.-W.; Ohshima, T.;
Mater. Interfaces 2018, 10, 2224−2229. Imaizumi, M.; Hirose, K.; Miyasaka, T. Tolerance of Perovskite Solar

BO DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX
Chemical Reviews Review

Cell to High-Energy Particle Irradiations in Space Environment.

iScience 2018, 2, 148−155.
(457) Snaith, H. J. Present Status and Future Prospects of Perovskite
Photovoltaics. Nat. Mater. 2018, 17, 372−376.
(458) Park, N.-G.; Segawa, H. Research Direction toward
Theoretical Efficiency in Perovskite Solar Cells. ACS Photonics
2018, 5, 2970−2977.
(459) Green, M. A.; Hishikawa, Y.; Warta, W.; Dunlop, E. D.; Levi,
D. H.; Hohl-Ebinger, J.; Ho-Baillie, A. W. H. Solar Cell Efficiency
Tables (version 50). Prog. Photovoltaics 2017, 25, 668−676.
(460) Tang, Z.-K.; Xu, Z.-F.; Zhang, D.-Y.; Hu, S.-X.; Lau, W.-M.;
Liu, L.-M. Enhanced Optical Absorption via Cation Doping Hybrid
Lead Iodine Perovskites. Sci. Rep. 2017, 7, 7843.
(461) Gong, J.; Yang, M.; Rebollar, D.; Rucinski, J.; Liveris, Z.; Zhu,
K.; Xu, T. Divalent Anionic Doping in Perovskite Solar Cells for
Enhanced Chemical Stability. Adv. Mater. 2018, 30, 1800973.

BP DOI: 10.1021/acs.chemrev.8b00539
Chem. Rev. XXXX, XXX, XXX−XXX