Advancements in all-solid-state hybrid solar cells based on organometal halide perovskites

Abstract

Over the past several years, organic–inorganic hybrid perovskites have gained considerable research attention due to their direct band gap, large absorption coefficient, ambipolar diffusion and long carrier diffusion length, and have revolutionized the prospects of emerging photovoltaic technologies, with the highest power conversion efficiency of over 19% achieved under laboratory conditions. In this perspective, we summarize the recent developments in perovskite solar cells (from April 2009 to December 2014), describe the unique properties of organometal halide perovskites leading to their rapid emergence, and discuss challenges such as stability and environmental issues to be faced in the future.

---

Shaowei Shi

Shaowei Shi received a BS degree at Beijing University of Chemical Technology in 2010. Now he is a Ph.D. student in Prof. Haiqiao Wang's group, at the State Key Laboratory of Organic–Inorganic Composites, Beijing University of Chemical Technology. His research interests include the synthesis of organic semiconducting materials, and their application in electronic devices.

Yongfang Li

Dr Yongfang Li has been a professor in the Institute of Chemistry, Chinese Academy of Sciences (ICCAS) since 1993, a professor in the College of Chemistry, Chemical Engineering and Materials Science, Soochow University since 2012, and a Member of the Chinese Academy of Sciences since 2013. He obtained his M.Sc. in Chemistry from Eastern China University of Science and Technology in 1982, and his Ph.D. in physical chemistry from Fudan University in 1986. His major research areas include electrochemistry of conducting polymers, and photovoltaic materials and devices for polymer solar cells. He has published more than 490 papers, and the published papers have been cited by others more than 15 000 times with an H-index of 64.

Xiaoyu Li

Prof. Xiaoyu Li received his B.S. in the Chemistry Department, Shandong University (1981), his M.Sc. in the Department of Polymer Science, Beijing University of Chemical Technology (1985), and his Ph.D. in the College of Materials Science and Engineering, Beijing University of Chemical Technology (1998). He is currently a Professor of the State Key Laboratory of Organic–Inorganic Composites, Head of Department of Organic Functional Materials, Beijing University of Chemical Technology. His research areas involve emulsion polymerization theory, synthesis and application of water-based paint, adhesive and ink materials, and new types of photoelectric functional materials.

Haiqiao Wang

Prof. Haiqiao Wang received his B.S. (1983), and M.Sc. (1989) in the Department of Chemistry, Huazhong University of Science and Technology, and his Ph.D. (1998) in the Department of Photoelectric Engineering, Huazhong University of Science and Technology. After postdoctoral work at Tsinghua University (1999–2001), he joined Beijing University of Chemical Technology. He is currently a Professor of the State Key Laboratory of Organic–Inorganic Composites, Beijing University of Chemical Technology. His research areas involve materials and devices of organic light emitting diodes, polymer solar cells and organic field-effect transistors, and synthesis and application of water-based polymers.

---

1. Introduction

Increasing energy demands and concerns about global warming drive the exploration/development of clean, inexpensive and renewable energy sources. Several new energy technologies for converting solar energy into electricity, including organic photovoltaic cells (OPVs), dye sensitized solar cells (DSSCs) and quantum dot solar cells (QDSCs), have attracted significant attention as low-cost alternatives to conventional silicon-based solar cells. Recently, hybrid organometal halide perovskites, which were initially employed in DSSCs as light absorbers, have gradually become one of the most important active materials for all-solid-state solar cells (which can be named as perovskite solar cells) due to their direct band gap, large absorption coefficient, ambipolar diffusion and long carrier diffusion length, with the highest efficiency of up to ∼20%. Two important journals, Science¹ and Nature,² both highlighted perovskite photovoltaics as one of the biggest breakthroughs of the year 2013.³

“Perovskite”, which is named after the Russian mineralogist, L. A. Perovski (1792–1856), is defined as one class of compounds which crystallise in the ABX₃ structure (e.g. CaTiO₃).⁴ An ideal perovskite structure has a cubic Pm3̄m crystal structure, which consists of a three-dimensional (3-D) framework of corner-sharing BX₆ octahedron with the A ion placed in the cuboctahedral interstices, as shown in Fig. 1. In the case of organometal trihalide perovskites, A is an organic cation (typically CH₃NH₃⁺ or HN=CHNH₃⁺), B is a metal cation (typically Sn²⁺ or Pb²⁺), and X is a halide anion (typically Cl⁻, Br⁻ or I⁻). Goldschmidt's tolerance factor (t), t = (R_A + R_X)/[2^1/2(R_B + R_X)], (where R_A, R_B and R_X are the ionic radii for the ions in the A, B and X sites, respectively; t = 1 corresponds to a perfectly packed perovskite structure) and octahedral factor (μ), μ = R_B/R_X, can be used to estimate the stability and distortion of the perovskite structure.^5–7

Fig. 1 The basic perovskite structure (ABX₃). Reprinted with permission.⁷

In the present paper, we aim to offer a brief review on the application of hybrid organic–inorganic perovskites (typically methylammonium lead halide perovskite) in different photovoltaic devices, including the mesoporous metal oxide solar cell, meso-superstructured solar cell and planar heterojunction solar cell. Although great progress has been achieved in the past three years, continued development of perovskite solar cells will require a better understanding of the relationships between the perovskite structure, device architecture, hole–electron conductor, and device performance than is currently available.

2. Progress in hybrid organic–inorganic perovskite solar cells

Since O'Regan and Grätzel first introduced the concept of DSSCs in 1991,⁸ in the following two decades, DSSCs have drawn great attention both in scientific and technological aspects and are considered to be a potential alternative to conventional inorganic silicon-based solar cells due to the low processing costs and inexpensive constituent materials. Generally, a DSSC contains a transparent conducting oxide electrode (typically fluorine-doped tin oxide (FTO) or indium-doped tin oxide (ITO)), a dye-sensitized mesoporous semiconductor metal oxide (typically nanocrystalline TiO₂) film, a platinum (Pt) counter electrode, and an electrolyte containing redox couples (typically I⁻/I³⁻) dissolved in a solvent. However, although power conversion efficiencies (PCEs) of close to 13% have been achieved with this device architecture,⁹ such cells suffer from potential leakage problems associated with the corrosive and volatile nature of the liquid electrolyte and, thus, may be impractical for large-scale applications. In 1998, the first example of solid-state dye-sensitized solar cells (ss-DSSCs) emerged by using a solid hole-transporting material (HTM) 2,2′,7,7′-tetrakis(N,N-di-p-methoxyphenylamine)-9,9′-spirobifluorene (spiro-OMeTAD or spiro) instead of the conventional liquid redox electrolyte.¹⁰ With this change, ss-DSSCs appeared to be quickly becoming both efficient and stable. Nevertheless, in the following long period of time, this type of cell did not reach the predetermined level of high efficiencies, yielding a maximum PCE of 7.2% based on tris(2-(1H-pyrazol-1-yl)pyridine)cobalt(III) doped spiro-OMeTAD and an organic photosensitizer with a high molar extinction coefficient.¹¹

The real breakthrough in ss-DSSCs came in 2012, when Grätzel et al.¹² and Snaith et al.¹³ independently employed CH₃NH₃PbI₃ (MAPbI₃) and CH₃NH₃PbI₂Cl (MAPbI₂Cl) perovskite nanocrystals as light harvesters using submicron thick mesoporous TiO₂ film and spiro-MeOTAD as an electron- and a hole-transporting layer. PCEs of 9.7% and 7.6% were achieved under AM1.5G illumination along with excellent long term stability. Moreover, by replacing the n-type mesoporous TiO₂ with insulating mesoporous Al₂O₃, evolution of the ss-DSSC, which is termed as the meso-superstructured solar cell (MSSC) was developed by Snaith et al.¹³ The FTO/blocking layer (bl)-TiO₂/mesoporous (mp)-Al₂O₃/MAPbI₂Cl/spiro-OMeTAD/Ag-based device showed a PCE of 10.9%.¹³ Etgar and coworkers also fabricated a hole conductor-free MAPbI₃/mp-TiO₂ heterojunction solar cell (FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/Au) and achieved a PCE of 5.5%.¹⁴ All the above mentioned studies proved that this organometal halide perovskite can not only act as an absorber (dye) but also as an ambipolar charge transporter, indicating the possibility for applying in various device architectures.

Actually, early in 2009, organometal halide perovskites MAPbBr₃ and MAPbI₃ were initially attempted in conventional liquid electrolyte-based DSSCs, yielding PCEs of 3.13% and 3.81%, respectively (see Fig. 2).¹⁵ Later, in 2011, via optimization of the titania surface and perovskite (MAPbI₃) processing, Park and coworkers further improved the PCE to 6.54% in a perovskite quantum dot-sensitized 3.6 μm thick TiO₂ film under AM1.5G 1 sun illumination.¹⁶ The authors also reported a (CH₃CH₂NH₃)PbI₃-sensitized solar cell with an iodide-based redox electrolyte and achieved a PCE of 2.4%.¹⁷ This pioneering work opened the prelude to organometal halide perovskites to be applied as active layers in ss-DSSCs. However, electrolyte-based devices were usually unstable and the performance degraded rapidly due to the dissolution or decomposition of the perovskite in the liquid electrolyte.

Fig. 2 (a) The schematic of perovskite-sensitized TiO₂ undergoing photoexcitation and electron transfer. (b) The incident photon-to-electron conversion efficiency (IPCE) spectra for MAPbBr₃/MAPbI₃-sensitized solar cells. Reprinted with permission.¹⁵

Fig. 3 Solution-based deposition methods (e.g. MAPbI₃). (a) One step precursor deposition (OSPD). Reprinted with permission.⁵³ (b) Sequential deposition process (SDP). Reprinted with permission.⁵⁴ (c) Two step spin-coating deposition (TSSD). Reprinted with permission.²¹

After the breakthrough work of ss-DSSCs based on organometal halide perovskites,^12,13 in 2013, two papers published in the journal Nature lifted the study of perovskite solar cells to a new level.^18,19 First, by using a sequential deposition process to fabricate MAPbI₃ films, a remarkable PCE of 15% was measured in a FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/spiro-OMeTAD/Au-based device.¹⁸ Subsequently, Snaith and coworkers deposited a high-quality MAPbI_3−xCl_x film via dual source vacuum deposition in a planar heterojunction (PHJ) perovskite solar cell (FTO/bl-TiO₂/MAPbI_3−xCl_x/spiro-OMeTAD/Ag) and achieved a PCE of 15.4%.¹⁹ Seok and coworkers also obtained a PCE of 12% in a FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/PTAA/Au-based device by replacing the conventional spiro-OMeTAD HTM with poly(triarylamine) (PTAA).²⁰ Since then, follow-up work such as the morphology and crystal optimization, HTM/ETM (electron-transporting material) adjustment and interface/band-gap engineering further improved the device performance to nearly 20% efficiency, opening up a new direction for developing highly efficient solar cells with a low cost and good stability.

At present, two types of perovskite solar cells, based on a porous or planar architecture, are the focus of research. It is worth pointing out that in a porous device, mesoporous “scaffolds” can not only act as conductors (e.g. TiO₂, ZnO, NiO) but also as insulators (e.g. Al₂O₃, ZrO₂, SiO₂). For the purpose of distinguishing these two types of devices, herein, we classify this porous device into two groups, including mesoporous metal oxide perovskite solar cell (MMOPSC) (with mesoporous conductors) and the above-mentioned MSSC (with mesoporous insulators). So far the highest PCEs achieved from MMOPSC, MSSC and planar heterojunction perovskite solar cell (PHJPSC) are 17.01%,²¹ 15.9% (ref. 22) and 19.3%,²³ respectively.

3. Deposition processes for perovskite films

As the core part of perovskite solar cells, the morphology and crystal structure of perovskite absorbers are important for achieving high-performance devices. So far seven main deposition methods have been reported including one step precursor deposition (OSPD), sequential deposition process (SDP), two step spin-coating deposition (TSSD), dual source vacuum deposition (DSVD), sequential vapour deposition (SVD), vapour-assisted solution process (VASP) and spray-coating deposition (SCD). Among the above mentioned methods, devices fabricated using OSPD, SDP, TSSD, DSVD or SVD have yielded PCEs of over 15%, while PCEs of over 10% have been achieved using VASP or SCD (Table 1).

3.1 Solution-based deposition methods (OSPD, SDP, TSSD, SCD) (Fig. 3)

The initial report of one step precursor deposition for organo-lead halide perovskite solar cells was from Miyasaka and coworkers in 2009, which involved spin-coating a precursor solution containing MAX and PbX₂ (X = Br, I) on the mp-TiO₂ layer to form nanocrystalline MAPbBr₃ or MAPbI₃.¹⁵ For the widely studied MAPbI₃ or MAPbI_3−xCl_x-based perovskite solar cells, γ-butyrolactone (GBL) and N,N-dimethylformamide (DMF) are commonly used solvents for dissolving PbX₂ (X = I, Cl) and MAI, in which the mole ratio of MAI and PbI₂ is 1 : 1 while the MAI and PbCl₂ are present in a molar ratio of 3 : 1. Generally, the perovskite precursor solution (∼40 wt%) is firstly stirred at a certain temperature (e.g. 80 °C) overnight before spin-coating on the substrate. Then thermal annealing (∼100 °C) is performed to remove the solvent and complete the transformation of the precursor to the resulting crystalline perovskite. The thickness of the perovskite film is tuned by changing the precursor concentration and/or spin speed of the perovskite precursor solution. The problem in this deposition method is the uncontrolled crystallization process of the perovskite, which always results in a wide spread of photovoltaic performance and low reproducibility in the resulting devices (especially for porous structure-based devices). To avoid the large perovskite grains and uncovered pin-hole areas, follow-up morphology optimizations such as solvent-engineering technology (e.g. using a mixed solvent or fast crystallization–deposition),^24–30 thermal annealing,^31–34 and additive treatment (e.g. 1,8-diiodooctane (DIO), NH₄Cl and MACl)^35–38 have been reported by several groups and excellent efficiencies have been achieved. To date, OSPD is still considered as one of the simplest and most important deposition methods, and the record efficiency (19.3%) of perovskite solar cells was achieved via this approach.²³

Another important solution-based deposition is the sequential deposition process, which was originally introduced by Mitzi et al. in 1998,³⁹ then employed in perovskite solar cells by Grätzel et al. in 2013.¹⁸ In this process, a PbX₂ (X = I, Cl⁴⁰) solution in DMF (or dimethylsulphoxide (DMSO)⁴¹) is first spin-coated (one or more times⁴²) on the top of porous or planar substrates, followed by dipping it into a solution of MAI in 2-propanol (seconds to tens of minutes) to transform into the perovskite. The perovskite crystal size and orientation on the substrate can be controlled by modifying the prewetting time of 2-propanol or reaction temperature in the second step.^43–45 So far, using this deposition, perovskite solar cells based on mesoporous metal oxide, meso-superstructured or planar structures have achieved the highest PCE of 15%,¹⁸ 10.8% (ref. 46) and 15.7%,⁴⁷ respectively. Compared with OSPD, SDP is more beneficial to improve the coverage, pore-filling, and morphology of the deposited perovskite in porous substrates, thereby further improving the device performance. However, for most PHJ perovskite solar cells, SDP cannot perform as successfully as in nanostructured devices. Without the porous “scaffold”, it is hard for MAI to penetrate into the PbX₂ films due to the limited reaction interface area, leading to a long reaction time and/or low conversion rate of PbX₂.⁴⁸ Also, large randomly distributed grains always appear during the quick crystallization of perovskite, which results in a very rough perovskite film and low efficiency.⁴⁹

In 2014, Huang and coworkers reported an efficient planar perovskite solar cell (PCE = 15.4%) using a two step spin-coating deposition method,⁴⁹ in which the solution of PbI₂ in DMF and the solution of MAI in 2-propanol were spin-coated step by step on the ITO/PEDOT:PSS-based substrate at first, followed by thermal annealing at 100 °C to drive the interdiffusion of precursors (this step can be skipped via controlling well the thickness of PbI₂ and the process of adding MAI⁵⁰). Later, Park et al. found that the size of the MAPbI₃ cuboids was very strongly dependent on MAI concentration and the exposure time of PbI₂ to the MAI solution before spin coating, and high PCE of 17.01% was achieved using this modified TSSD.^21,51 Lidzey et al. explored the use of ultra-sonic spray-coating as a deposition technique to create MAPbI_3−xCl_x perovskite thin-films in ambient conditions.⁵² Precursor solution of MAI and PbCl₂ (3 : 1 molar ratio) in DMF or DMSO was deposited on the substrates from a single pass of the spray-head, then transformed into MAPbI_3−xCl_x through thermal annealing. A planar device with an ITO/PEDOT:PSS/MAPbI_3−xCl_x/[6,6]-phenyl C61-butyric acid methyl ester (PC60BM)/Ca/Al structure yielded a PCE of 11.1% with an average efficiency of 7.8%, which was comparable to the devices fabricated via spin-coating, showing the commercial potential of SCD used in the large-area, low-cost, efficient manufacture of perovskite solar cells.

3.2 Vapor-based deposition methods (DSVD, SVD, VASP) (Fig. 4)

Vapor-based deposition processes are not used as commonly as solution-based deposition processes (especially OSPD and SDP), mainly due to the increasing manufacturing cost. In 2013, Snaith and coworkers initially fabricated a MAPbI_3−xCl_x-based PHJ perovskite solar cell (FTO/bl-TiO₂/MAPbI_3−xCl_x/spiro-OMeTAD/Ag) via dual source vacuum deposition and achieved a high PCE of 15.4%.¹⁹ In this process, MAI and PbCl₂ were evaporated simultaneously from separate sources and a superior uniformity of the perovskite films was observed over a range of length scales. Subsequently, using DSVD, Bolink and Sarkar groups reported three types of PHJ devices based on ITO/PEDOT:PSS/poly-TPD/MAPbI₃/PC60BM/Au, FTO/NiO/MAPbI_3−xCl_x/PC60BM/Ag and FTO/CuSCN/MAPbI_3−xCl_x/PC60BM/Ag, yielding PCEs of 14.8%, 7.26%, and 3.8%, respectively.

In 2014, Cui⁵⁵ and Lin⁵⁶et al. developed a sequential vapor deposition process, in which PbX₂ (X = I, Cl) and MAI were vapour-deposited on the substrates layer by layer. PbX₂ reacts with MAI in situ, followed by thermal annealing to complete the perovskite crystal transformation. This deposition process is very similar to the above mentioned precursor interdiffusion deposition method (TSSD), except that SVD uses vapor deposition. Compared with the perovskite films fabricated via DSVD, SVD fabricated films demonstrate a larger crystal domain size, which is beneficial in improving transport properties of carriers. Also, SVD enhances the control on the perovskite morphology by independent manipulation of PbX₂ and MAI. Using SVD, Cui and coworkers prepared a uniform, smooth MAPbI₃ film and fabricated an easy planar perovskite solar cell consisting of only a MAPbI₃/fullerene (C60) layer sandwiched between two electrical contacts.⁵⁵ No hole conductor (e.g. PEDOT:PSS or NiO_x) was needed in this system and a PCE of 5.4% was achieved under AM1.5G one sun illumination. Lin et al. also achieved a high PCE of 15.4% in an ITO/PEDOT:PSS/MAPbI_3−xCl_x/C60/bathophenanthroline (Bphen)/Ca/Ag-based device.⁵⁶

Yang and coworkers reported a blended deposition method called vapour-assisted solution process to fabricate MAPbI₃ films.⁴⁸ First, the solution of PbI₂ in DMF was spin-coated on the FTO/bl-TiO₂ substrates, and dried at 110 °C for 15 min. Then the PbI₂-coated substrates were annealed in MAI vapor at 150 °C in a N₂ atmosphere for the desired time to yield the perovskite films. After cooling down, the as-prepared substrates were washed with isopropanol, dried and annealed to complete the deposition. The perovskite film derived from this approach exhibited full surface coverage, uniform grain structure with grain size up to micrometers, and ∼100% precursor transformation completeness. The PHJ device based on FTO/bl-TiO₂/MAPbI_3−xCl_x/spiro-OMeTAD/Ag yielded a PCE of 12.1%. Later, via VASP, the authors introduced a controllable self-induced passivation technique by tuning the amount of PbI₂ species in perovskite grain boundaries and at the relevant interfaces, which was helpful for understanding the carrier behavior along the heterojunctions and the polycrystalline nature of hybrid perovskite thin films.^57–59

Table 1 Optimal device performance reported so far via different perovskite deposition methods and corresponding device configurations

Deposition process (DP)	Device structure	J sc (mA cm^−2)	V oc (V)	FF (%)	PCE (%)	Ref.
OSPD	ITO/PEIE/bl-TiO2(Y)/MAPbI3−xClx/spiro/Au	22.75	1.13	75.01	19.3	23
SDP	ITO/bl-ZnO/MAPbI3/spiro/Ag	20.4	1.03	74.9	15.7	47
TSSD	ITO/bl-TiO2/mp-TiO2/MAPbI3/spiro/Au	21.64	1.056	74.1	17.01	21
DSVD	FTO/bl-TiO2/MAPbI3−xClx/spiro/Ag	21.5	1.07	67	15.4	19
SVD	ITO/PEDOT:PSS/MAPbI3−xClx/C60/Bphen/Ca/Ag	20.9	1.02	72.2	15.4	56
VASP	FTO/bl-TiO2/MAPbI3−xClx/spiro/Ag	19.8	0.924	66.3	12.1	48
SCD	ITO/PEDOT:PSS/MAPbI3−xClx/PC60BM/Ca/Al	16.8	0.92	72	11.1	52

---

4. Device architectures of perovskite solar cells

4.1 Mesoporous metal oxide-based perovskite solar cells

4.1.1 Device based on mesoporous n-type TiO₂. As the most common device architecture for perovskite solar cells, devices based on a perovskite sensitized mesoporous TiO₂ scaffold have yielded PCEs from 9.7% (ref. 12) to over 16%.^21,24,60 This porous n-type TiO₂ film not only extracts photoexcited electrons generated in the absorber layer⁶¹ but also increases the perovskite crystal transformation when the perovskite layer is fabricated using SDP.⁶² For a typical mesoporous TiO₂ based perovskite solar cell, the fabrication process always starts with the deposition of a compact TiO₂ layer on the top of pre-cleaned FTO conductive substrates. Three strategies are widely used at present for depositing the compact TiO₂ layer, which acts as a blocking layer to prevent direct contact between FTO and the infiltrated HTM layer, including (1) spin-coating the colloidal dispersion of TiO₂ nanoparticles followed by a thermal treatment (titanium source: TiCl₄,²² titanium isopropoxide,^63,64 tetra-n-butyl-titanate⁶⁵); (2) spin-coating titanium precursor solutions followed by a thermal treatment (titanium source: TiCl₄,⁶⁶ titanium isopropoxide,⁶⁷ titanium diisopropoxide bis(acetylacetonate)¹²); (3) spray pyrolysis deposition (titanium source: titanium diisopropoxide bis(acetylacetonate)¹⁸). Also, other methods such as atomic layer deposition (ALD)^68–70 and thermal oxidation of Ti film^71,72 were also reported. Then a mesoporous n-type TiO₂ layer is constructed by spin-coating, screen-printing or doctor-blading the TiO₂ nanoparticle paste on the top of the compact TiO₂ layers followed by sintering. After that, OSPD or SDP is employed to deposit the hybrid organic–inorganic perovskite on the mesoporous TiO₂ films. Then the organic or inorganic HTM layer is deposited on the top of the perovskite layer (the existence of HTMs not only favours the hole transport but also blocks the electron transfer from perovskites to the electrode). This step can be skipped in a HTM-free mp-TiO₂/perovskite heterojunction solar cell.¹⁴ Finally, a metal electrode is deposited via a thermal evaporator to complete the solar cell.

At present, tremendous attention has been paid on the optimization of this mesoporous TiO₂-based device, including the modification of the TiO₂ nanostructure, perovskite layer and HTM, along with the in-depth understanding on the detailed mechanism behind this high efficiency device architecture.^73–77 So far, anatase TiO₂ nanoparticles are the most frequently and successfully used nanostructures to construct mesoscopic perovskite solar cells, with the highest PCE of over 17%. In 2014, Park and coworkers used rutile TiO₂ nanoparticles to fabricate a FTO/bl-TiO₂/mp-TiO₂(rutile)/MAPbI₃/spiro-MeOTAD/Au device.⁷⁸ Compared to the anatase TiO₂-based device, higher electron diffusion coefficients and lower electron recombination were observed in the rutile TiO₂-based device, suggesting that more electrons were injected from the perovskite to the rutile TiO₂ layer. As a result, although a relatively lower open-circuit voltage (V_oc) was observed due to the lower Fermi energy level at equilibrium between TiO₂ and perovskite in the rutile-based device, the optimal device demonstrated a PCE of 14.46%, with a short-circuit current density (J_sc) of 20.02 mA cm⁻², a V_oc of 1.022 V, and a fill factor (FF) of 0.71 at AM1.5G one sun illumination.

Besides the mostly used TiO₂ nanoparticles, many other nanostructured TiO₂ materials, such as nanosheets,^14,79,80 nanorods,^81–85 nanotubes,⁸⁶ nanofibers/nanowires^87–89 and single crystals,⁹⁰ were also tried to fabricate the mesoscopic TiO₂ structure, due to the reason that the physical and chemical properties of TiO₂ nanocrystals are affected not only by the intrinsic electronic structure, but also by their size, shape, organization, and surface properties. Jung and coworkers also fabricated a TiO₂ nanoparticle/ITO nanowire nanocomposite for use as a photoelectrode material.⁹¹ However, compared with the state-of-the-art TiO₂ nanoparticle-based device, lower efficiencies were observed in these nanostructured TiO₂-based devices. Furthermore, researchers^70,82,85 also introduced a way of surface modification to improve the perovskite-based device performance by doping metal (e.g. Y, Mg, Nb) ions into the nanostructured TiO₂.

Fig. 4 Vapor-based deposition methods. (a) Dual source vacuum deposition (DSVD). Reprinted with permission.¹⁹ (b) Sequential vapour deposition (SVD) (e.g. MAPbI₃). Reprinted with permission.⁵⁵ (c) Vapor assisted solution process (VASP). Reprinted with permission.⁴⁸

4.1.2 Device based on mesoporous n-type ZnO (Table 2). Nanostructured ZnO is a viable n-type alternative scaffold to mesoporous TiO₂ for perovskite solar cells due to its comparable energy levels (band gap: ∼3.37 eV at 25 °C) as well as relatively higher electron mobility.^92,93 The first example of such a device was reported by Hagfeldt et al. in 2013, in which the MAPbI₃ perovskite layer was one-step deposited on the top of vertically ordered ZnO nanorod arrays (NRAs) with spiro-OMeTAD as the HTM.⁹⁴ In this configuration, no dense TiO₂ layer was needed but replaced with a compact ZnO layer (bl-ZnO), which acted as not only a hole blocking layer but also a seed layer for the growth of the ZnO nanorods. ZnO NRAs were prepared by a hydrothermal process, using equimolar zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O) and hexamethylenetetramine (HMTA) as precursors. The diameter and length of ZnO nanorods could be controlled by changing the precursor concentration and growth time. A noteworthy feature is that although an increased nanorod length can result in a higher light harvesting efficiency due to the increased perovskite loading, an increased charge recombination could also come up, which will deteriorate the photovoltaic performance. Thus, it is necessary to seek a balance to achieve the optimal performance. Based on this ITO/bl-ZnO/NRA-ZnO (diameter/length = 50 nm/1000 nm)/MAPbI₃/spiro-OMeTAD/Ag device, the initial study yielded a PCE of 5%, with a J_sc of 12.7 mA cm⁻², V_oc of 0.68 V, and FF of 0.58 under 100 mW cm⁻² AM1.5G illumination.⁹⁴ Later, by systematically controlling the diameter/length (82 nm/1000 nm) of ZnO NRAs and depositing MAPbI₃ using SDP (see Fig. 5), Park et al. effectively increased the pore-filling of perovskite in ZnO NRA film and further improved the PCE to 11.13%, with both increased J_sc (20.08 mA cm⁻²) and V_oc (0.991 V).⁹⁵

Fig. 5 SEM images of (a) bare ZnO nanorod grown on an FTO substrate, (b) MAPbI₃-deposited ZnO nanorods, and (c) full cell. The inset in (b) is the surface SEM image of the MAPbI₃ capping layer. (d) The fabrication procedure of the perovskite solar cell (via SDP) based on the ZnO nanorod electrode. Reprinted with permission.⁹⁵

Mathews and coworkers also reported a ZnO NRA-based all-low-temperature processed device, in which the ZnO nanorods were fabricated using chemical bath deposition (CBD) while the ZnO compact layer was formed by electrodeposition.⁹⁶ The typical ZnO nanorod diameters were in the range of 100–150 nm and the lengths were between 400 and 500 nm. Four types of devices, including FTO/bl-ZnO/MAPbI₃/HTM/Au (T₁), FTO/bl-ZnO/NRA-ZnO/MAPbI₃/HTM/Au (T₂), polyethylene terephthalate (PET)/FTO/bl-ZnO/MAPbI₃/HTM/Au (T₃), and PET/FTO/bl-ZnO/NRA-ZnO/MAPbI₃/HTM/Au (T₄), were fabricated and characterized. Relatively low PCEs were obtained under the condition without ZnO NRAs and at last, PCEs of 8.90% and 2.62% were achieved from the T₂ and T₄ devices, respectively.

In 2014, using plasma-enhanced chemical vapor deposition (PECVD), Ahmad et al. prepared a nanocolumnar ZnO thin film on the compact TiO₂-coated FTO substrate at low temperature.⁹⁷ Under AM1.5G solar illumination (100 mW cm⁻²), the MAPbI₃-sensitized solar cell combined with spiro-OMeTAD or PTAA as the HTM yielded PCEs of 4.8% and 1.3%. Compared with the early result from Mathews et al.,⁹⁶ the FTO/bl-TiO₂/nanocolumns ZnO/MAPbI₃/spiro-OMeTAD/Au-based device demonstrated a similar J_sc but significantly decreased FF and V_oc, which can be attributed to the higher charge recombination.

Recently, Mahmood and coworkers fabricated a mesoporous ZnO or Al-doped mesoporous ZnO (AZO) film using the electrospraying method and studied the performance of the resulting perovskite solar cells.⁹⁸ For the optimal device based on a pure ZnO film (∼440 nm thick), a PCE of 10.8% was achieved, with a J_sc of 16 mA cm⁻², V_oc of 1.01 V, and FF of 0.67. When the ZnO film was doped with Al, the device yielded a higher PCE of 12%, with obviously increased V_oc (1.045 V) and FF (0.76). According to the reports,⁹⁹ doping of ZnO with metals can not only enhance its n-type characteristics (e.g. increasing the conductivity) but also help to shift the Fermi level in the direction of the conduction band. As the V_oc is mainly determined by the difference between the quasi-Fermi levels of the electrons in the n-type semiconductor and the holes in the HTM, a lower electron recombination rate and higher electron concentration of the conduction band can be obtained, which is beneficial in enhancing both the V_oc and FF. Similar performance increase was also observed by Meng et al., by modifying the ZnO nanorod surface with a thin Al-doped ZnO layer, in a FTO/bl-ZnO/NRA-ZnO/AZO/MAPbI₃/HTM/Au-based device (J_sc = 19.77 mA cm⁻², V_oc = 0.90 V, FF = 0.60 and PCE = 10.7%).¹⁰⁰

Table 2 A summary of published representative results of mesoporous n-type ZnO-based MMOPSC performance parameters with different device fabrication methods and configurations

Perovskite layer (PL)	DP	Device structure	J sc (mA cm^−2)	V oc (V)	FF (%)	PCE (%)	Ref.
MAPbI3	OSPD	ITO/bl-ZnO/NRA-ZnO/PL/spiro/Ag	12.7	0.68	58	5.0	94
MAPbI3	SDP	FTO/bl-ZnO/NRA-ZnO/PL/spiro/Au	20.08	0.991	56	11.13	95
MAPbI3	SDP	FTO/bl-ZnO/PL/spiro/Au (T1)	11.27	1.08	45.44	5.54	96
MAPbI3	SDP	FTO/bl-ZnO/NRA-ZnO/PL/spiro/Au (T2)	16.98	1.02	51.11	8.90
MAPbI3	SDP	PET/ITO/bl-ZnO/PL/spiro/Au (T3)	5.57	0.99	39.58	2.18
MAPbI3	SDP	PET/ITO/bl-ZnO/NRA-ZnO/PL/spiro/Au (T4)	7.52	0.80	43.14	2.62
MAPbI3	SDP	FTO/bl-TiO2/nanocolumn ZnO/PL/spiro/Au	16	0.718	41.2	4.8	97
MAPbI3	SDP	FTO/bl-TiO2/nanocolumn ZnO/PL/PTAA/Au	8.3	0.481	32.7	1.3
MAPbI3	SDP	FTO/bl-ZnO/mp-ZnO/PL/spiro/Ag	16	1.01	67	10.8	98
MAPbI3	SDP	FTO/bl-ZnO(Al)/mp-AZO/PL/spiro/Ag	15.1	1.045	76	12
MAPbI3	SDP	FTO/bl-ZnO/NRA-ZnO/AZO/PL/spiro/Au	19.77	0.90	60	10.7	100

---

4.1.3 Device based on mesoporous p-type NiO. Apart from devices based on mesoporous n-type metal oxides such as TiO₂ and ZnO, in 2014, Guo's group demonstrated a mesoscopic NiO/MAPbI₃/fullerene derivative-based architecture using nanocrystalline p-type NiO as the selective contact (see Fig. 6).¹⁰¹ This inverted mesoporous configuration was initially developed from a planar perovskite solar cell with a flat electrode of oxide ITO/NiO_x (without the NiO nanocrystalline layer).^102,103 Compared with planar devices with thin NiO films, fabrication of the mesoscopic NiO layer provided an increased film thickness to host the light absorbing perovskite material and prevented the risk of morphological defects that decreased the photovoltaic performance. The PCE of devices with the ITO/bl-NiO_x/mp-NiO/MAPbI₃/[6,6]-phenyl C71-butyric acid methyl ester (PC70BM)/BCP/Al and ITO/bl-NiO_x/mp-NiO/MAPbI₃/PC60BM/BCP/Al structure reached 9.44% and 9.51%, which are higher than that (7.4%) of the device based on the ITO/bl-NiO_x/MAPbI₃/PC60BM/BCP/Al system. Later, using a low-temperature sputtered bl-NiO_x thin film instead of the solution-processed bl-NiO_x thin film, a higher PCE of 11.6% was achieved in this mp-NiO/MAPbI₃ heterojunction solar cell.¹⁰⁴

Fig. 6 (a) The photos and illustrations of patterned ITO glass, ITO glass with bl-NiO_x and mp-NiO with a perovskite coated electrode. (b) The schematic of the whole device (ITO/bl-NiO_x/mp-NiO/MAPbI₃/PC60BM (or PC70BM)/BCP/Al). (c) The energy level diagram of the mp-NiO/perovskite/PC60BM heterojunction. Reprinted with permission.¹⁰¹

4.2 Meso-superstructured perovskite solar cells (Table 3)

In 2012, Snaith and coworkers put forward the concept of “meso-superstructured solar cell” for the first time.¹³ In this case, a Cl mixed perovskite MAPbI_3−xCl_x was infiltrated within an insulating mp-Al₂O₃ film rather than the conventional mp-TiO₂ film. No photoexcited electrons were injected into Al₂O₃ but directly transported throughout the perovskite layer and were collected at the compact TiO₂-coated FTO electrode (see Fig. 7a). Unlike the mesoporous n-type TiO₂-based perovskite solar cells, Al₂O₃ acts only as a “scaffold” and the perovskite layer functions both as an intrinsic absorber and electron transporter. Faster electronic charge transportation in the perovskite layer was observed than in the mesoporous TiO₂, and the device based on mp-Al₂O₃ yielded a PCE of 10.9%, with a J_sc of 17.8 mA cm⁻², V_oc of 0.98 V, and FF of 0.63. It was found that a lower V_oc (0.80 V) was measured for mp-TiO₂-based devices than for the mp-Al₂O₃-based device (0.98 V). Initial theories attributed this to the declined electron quasi-Fermi levels in the TiO₂-based device, which resulted in a narrowed splitting of hole and electron quasi-Fermi levels. Since the V_oc is directly related to the difference between the hole- and electron quasi-Fermi levels, this decreased electron quasi-Fermi levels will lead to a lower V_oc in the TiO₂-based device. A similar change of V_oc was also reported by Hodes et al. in MAPbBr₃-based MSSCs using PDI as the HTM,¹⁰⁵ where the device based on the FTO/bl-TiO₂/mp-Al₂O₃/MAPbBr₃/PDI/Au structure yielded a V_oc as high as 1.3 V compared to the mp-TiO₂ based device with a V_oc of 1 V.

Fig. 7 (a) The schematic of perovskite-coated TiO₂ and Al₂O₃, illustrating electron and hole transfer. Reprinted with permission.¹³ (b) The schematic of MSSC with a thin Al₂O₃ layer. Reprinted with permission.⁶⁷

In 2013, using a thin Al₂O₃ film (∼80 nm) processed at low temperature (<150 °C), Snaith et al. fabricated a “flat-junction” thin film solar cell with a thick perovskite “capping layer” on the top of the scaffold (see Fig. 7b), and improved the FTO/bl-TiO₂/mp-Al₂O₃/MAPbI_3−xCl_x/spiro-OMeTAD/Ag-based device efficiency to 12.3% (J_sc = 18 mA cm⁻², V_oc = 1.02 V, and FF = 0.67).⁶⁷ Later, by incorporating core–shell Au@SiO₂ nanoparticles (np-Au@SiO₂) into the porous Al₂O₃ scaffold, the authors achieved significantly enhanced J_sc and PCE in the “Au@SiO₂” device (J_sc = 16.91 mA cm⁻², PCE = 11.4%) compared to the “Al₂O₃-only” device (J_sc = 14.76 mA cm⁻², PCE = 10.7%), and introduced a new enhancement mechanism of reduced exciton binding energy with the incorporation of the metal nanoparticles, rather than enhanced light absorption.¹⁰⁶ Worsley and coworkers reported a simplified one-step deposition process for the Al₂O₃–perovskite layer: a mixed DMF solution containing Al₂O₃ nanoparticles (np-Al₂O₃) and the MAPbI_3−xCl_x perovskite precursor was directly spin-coated on the compact TiO₂ layer, followed by a thermal treatment at low temperature (<110 °C).¹⁰⁷ The device based on this co-deposited Al₂O₃–perovskite film, in which 5 wt% alumina was added to the precursor solution prior to spin-coating, yielded the highest PCE of 7.16%, with a J_sc of 12.78 mA cm⁻², V_oc of 0.925 V, and FF of 0.61. In 2014, Snaith and coworkers introduced a method to passivate the hole trapping states at the perovskite (MAPbI_3−xCl_x) surface by assembling iodopentafluorobenzene (IPFB) via supramolecular halogen bonding.¹⁰⁸ The passivation of the undercoordinated halides effectively reduced the density of accumulated charge at the perovskite/HTM heterojunction, leading to an increased FF (0.67) compared to that (0.57) in the unpassivated device and a high PCE of 15.7%. Later, using a low Al-doped TiO₂ blocking layer (0.3 mol%), the authors also observed an increased J_sc (20 mA cm⁻²) and PCE (13.8%) compared with the undoped device (J_sc = 16.04 mA cm⁻², PCE = 11.13%) due to the improved conductivity of Al-doped TiO₂.¹⁰⁹ Recently, by replacing the organic HTM with P3HT-functionalized single-walled carbon nanotubes (SWNTs) embedded in an insulating poly(methyl methacrylate) (PMMA) matrix, an MSSC device with unprecedented resilience against thermal stressing and moisture ingress was fabricated, and yielded a PCE of 15.3% with an average efficiency of 10 ± 2%.¹¹⁰

In spite of the elimination of high temperature sintering steps for the mesoporous Al₂O₃ scaffold, a high sintering temperature is still needed for the compact TiO₂ layer (∼500 °C) in MSSCs, which lacks the compatibility with flexible substrates and thus presents a drawback in view of large scale industrial manufacture. In the end of 2013, by spin-coating the nanocomposites of graphene nanoflakes/TiO₂ nanoparticles on the top of FTO-coated substrates, Snaith and coworkers fabricated low temperature processed electron collection layers at 150 °C.¹¹¹ The superior charge mobility and proper work function of graphene effectively improved the electrical conductivity and reduced the formation of energy barriers at the material interfaces, leading to an excellent device efficiency of 15.6%. Subsequently, the authors reported a new route for fabricating a low temperature processed TiO₂ compact layer (<150 °C). Compared with the previous MSSCs based on a high temperature sintered compact TiO₂ layer, this all-low-temperature processed device yielded a PCE of up to 15.9%, with a J_sc of 21.5 mA cm⁻², V_oc of 1.02 V, and FF of 0.71 under simulated AM1.5 100 mW cm⁻² sunlight. Also, the PCE of 15.9% is the highest reported value for Al₂O₃-based MSSCs.²² Furthermore, Yuan and coworkers also reported the use of compact ZnO layers which were fabricated by ALD at 70 °C, and achieved PCEs of 13.1% under standard AM1.5 illumination.¹¹²

Besides the Al₂O₃-based MSSCs, in 2013, Johansson⁴⁶ and Park¹¹³et al. reported another type of MSSC based on a nanostructured ZrO₂ scaffold. Like Al₂O₃, the much higher conduction band of ZrO₂ than that of perovskite blocks the electron injection with no charge separation at the ZrO₂/perovskite interface. Finally, using the SDP method, a PCE of 10.8% was achieved from this mp-ZrO₂/MAPbI₃-based device, with a high V_oc up to 1.07 V. In 2014, Jiang and coworkers also fabricated a scaffold layer composed of SiO₂ nanoparticles for perovskite solar cells.¹¹⁴ By controlling the size of the SiO₂ nanoparticles, optimal FTO/bl-TiO₂/mp-SiO₂/MAPbI_3−xCl_x/spiro-OMeTAD/Au-based devices yielded a PCE of 11.45%, with a J_sc of 16.4 mA cm⁻², V_oc of 1.05 V, and FF of 0.66.

Table 3 A summary of published representative results of MSSC performance parameters with different device fabrication methods and configurations

PL	DP	Device structure	J sc (mA cm^−2)	V oc (V)	FF (%)	PCE (%)	Ref.
MAPbI3−xClx	OSPD	FTO/bl-TiO2/mp-Al2O3/PL/spiro/Ag	17.8	0.98	63	10.9	13
MAPbI3−xClx	OSPD	FTO/bl-TiO2/mp-Al2O3/PL/spiro/Ag	18	1.02	67	12.3	67
MAPbI3−xClx	OSPD	FTO/bl-TiO2/mp-Al2O3(np-Au@SiO2)/PL/spiro/Ag	16.91	1.02	64	11.4	106
MAPbI3−xClx	OSPD	FTO/bl-TiO2/PL(np-Al2O3)/spiro/Au	12.78	0.925	61	7.16	107
MAPbI3−xClx	OSPD	FTO/bl-TiO2/mp-Al2O3/PL/IPFB/spiro/Ag	23.38	1.06	67	15.7	108
MAPbI3−xClx	OSPD	FTO/bl-TiO2(Al)/mp-Al2O3/PL/spiro/Ag	20	1.07	65	13.8	109
MAPbI3−xClx	OSPD	FTO/bl-TiO2/mp-Al2O3/PL/P3HT-SWNTs-PMMA/Ag	22.71	1.02	66	15.3	110
MAPbI3−xClx	OSPD	FTO/bl-TiO2(graphene)/mp-Al2O3/PL/spiro/Au	21.9	1.04	73	15.6	111
MAPbI3−xClx	OSPD	FTO/bl-TiO2/mp-Al2O3/PL/spiro/Ag	21.5	1.02	71	15.9	22
MAPbI3	OSPD	FTO/bl-ZnO/mp-Al2O3/PL/spiro/Ag	20.4	0.976	66	13.1	112
MAPbI3	SDP	FTO/bl-TiO2/mp-ZrO2/PL/spiro/Ag	17.3	1.07	59	10.8	46
MAPbI3−xClx	OSPD	FTO/bl-TiO2/mp-SiO2/PL/spiro/Au	16.4	1.05	66	11.45	114

---

4.3 Planar heterojunction perovskite solar cells

During the initial studies on MSSCs with an Al₂O₃ scaffold, Snaith and coworkers also fabricated a planar heterojunction perovskite solar cell based on a simple FTO/bl-TiO₂/MAPbI_3−xCl_x/spiro-OMeTAD/Ag structure. No porous metal-oxide structure was needed in this configuration and perovskite actually functioned as the ambipolar layer in a p–i–n junction, where the intrinsic (i) layer is the perovskite absorber. Preliminary research findings yielded PCEs from 1.8% (ref. 13) to 4.9%,⁶⁷ and nearly 100% internal quantum efficiency (IQE) was measured.⁶⁷ Also, it has been discovered that the typical 3-D organolead trihalide perovskites exhibit large charge carrier diffusion lengths (∼100 nm for the triiodide perovskite and >1 μm for the mixed halide perovskite with Cl).^115,116 All the above mentioned studies indicate the possibility of this planar configuration to be a highly efficient architecture. In fact, the PHJ structure can not only simplify the device fabrication process but also avoid the pore filling problem in the mesosuperstucture device, which always leads to a large standard deviation. At present, two configurations for this planar device including a positive and inverted structure have been reported, in which a n-type (e.g. TiO₂, ZnO) or p-type conductor (e.g. PEDOT:PSS, NiO, CuSCN, graphene oxide, polythiophene) was coated on a conductive glass, and both yielded excellent PCEs.

4.3.1 Device based on a positive configuration (Table 4). For this type of PHJ perovskite solar cell (bl-TiO₂/perovskite/spiro-OMeTAD), the morphology of the absorber layer is the critical factor in determining the resulting device performance. In 2013, through carefully varying the processing conditions, such as the annealing temperature and perovskite film thickness, compact TiO₂ layers with a high MAPbI_3−xCl_x perovskite (via OSPD) coverage were obtained by Snaith et al., and a breakthrough PCE of 11.4% was achieved under standard AM1.5G illumination, with a J_sc of 20.3 mA cm⁻², V_oc of 0.89 V, and FF of 0.64.³¹ Later, using DSVD or SDP, Snaith¹⁹ and Bein^43,117et al. further improved the bl-TiO₂/MAPbI_3−xCl_x/spiro-OMeTAD-based device efficiency to ∼15%. Yang and coworkers also prepared high-quality planar MAPbI₃ films by VASP and achieved a PCE of 12.1% (bl-TiO₂/MAPbI₃/spiro-OMeTAD).⁴⁸ Yella et al. introduced a low-temperature route for the fabrication of FTO/bl-TiO₂(rutile)/MAPbI₃/spiro-OMeTAD/Au solar cell and obtained a higher PCE (13.7%) compared to the device with a compact TiO₂ (anatase) layer (3.7%).⁶⁶ A noteworthy feature was that a relatively lower J_sc (<20 mA cm⁻²) was observed for pure iodide-based perovskite solar cells, which could be attributed to the shorter electron–hole diffusion lengths (on the order of 100 nm) than that of Cl-mixed perovskite.^115,116 Han and coworkers recently optimized the SDP using a strongly coordinative solvent, DMSO, instead of the commonly used DMF to dissolve PbI₂ and fabricate PbI₂ films, and obtained a highly modified MAPbI₃ perovskite film with relatively uniform distributions of crystal sizes.⁴¹ PHJ devices demonstrated good reproducibility and yielded the highest PCE of 13.5% (J_sc = 20.71 mA cm⁻², V_oc = 1.02 V, and FF = 0.64), with an average efficiency of 12.5%. Spiccia and coworkers also reported a one-step, solvent-induced, fast crystallization method to control the dynamics of nucleation and grain growth of MAPbI₃.²⁷ The device based on the FTO/bl-TiO₂/MAPbI₃/spiro-OMeTAD/Ag structure yielded the maximum PCE of 16.2% under standard AM 1.5 condition (J_sc = 21.1 mA cm⁻², V_oc = 1.04 V, and FF = 0.74).

In 2013, Kelly and coworkers reported the use of a thin ZnO nanoparticle film (∼25 nm) as an electron-transport layer in an ITO/bl-ZnO/MAPbI₃/spiro-OMeTAD/Ag structure PHJ solar cell.⁴⁷ The higher electron mobility of ZnO than TiO₂ (ref. 92) and the large crystallite size on the surface of the ITO/bl-ZnO/MAPbI₃ layer resulted in a higher J_sc (20.4 mA cm⁻²) with a PCE of 15.7%. Considering that no sintering or annealing step is required for fabricating the ZnO layer, the device on a flexible ITO/PET substrate was prepared, yielding a V_oc of 1.03 V, J_sc of 13.4 mA cm⁻², FF of 0.739 and PCE of 10.2%. Also, the PCE of 10.2% was the highest value in the reported flexible perovskite devices. Later, Lee et al. used a simple PC60BM modified, sol–gel processed ZnO layer as the electron conductor.¹¹⁸ Through this interfacial engineering, higher V_oc values were observed in devices with bl-ZnO/PC60BM substrates compared to devices with only ZnO, and a PCE of 12.2% was achieved in ITO/bl-ZnO/PC60BM/MAPbI₃/spiro-OMeTAD/MoO₃/Ag-based devices. Bai and coworkers also reported low-temperature magnetron sputtered ZnO nanorod film as the cathode interlayer for ITO/bl-ZnO/MAPbI₃/spiro-OMeTAD/MoO₃/Ag-based device and achieved PCEs of 13.4% and 8.03% respectively in glass/ITO- and PET/ITO-based substrates.¹¹⁹

Besides the above-mentioned TiO₂ or ZnO-based positive devices using spiro-OMeTAD as the HTM, other types of p–i–n junctions with different p-type HTMs such as organic P3HT,^63,64PTB7-Th,¹¹⁸DR3TBDTT¹²⁰ and inorganic CuSCN¹²¹ were also reported, with favorable performance achieved.

Very recently, through a method of Lewis base passivation,¹²² in which the crystal surfaces were treated with the Lewis bases thiophene and pyridine, a significant decrease in the rate of nonradiative recombination in perovskite films was measured by Snaith et al. The authors thought that the under-coordinated Pb ions in the perovskite crystal could be bound with Lewis base molecules, thus passivating these defect sites. At last, PCEs of 15.3% and 16.5% were achieved using thiophene and pyridine treated MAPbI_3−xCl_x respectively in the FTO/bl-TiO₂/MAPbI_3−xCl_x/Lewis base/spiro-OMeTAD/Au-based device. Yang's group fabricated a modified planar configuration with an ITO/PEIE/bl-TiO₂(Y)/MAPbI_3−xCl_x/spiro-OMeTAD/Au structure, in which ITO was coated with polyethyleneimine ethoxylated (PEIE) while TiO₂ was doped with yttrium (Y) (see Fig. 8).²³ The PEIE-modified ITO effectively reduces the work function of ITO from 4.6 eV to 4.0 eV, which is beneficial to the efficient electron transport between the TiO₂ and ITO layers. Also, like the Al-doped ZnO mentioned above,⁹⁸ the doping of TiO₂ with Y not only improves the conductivity (from 6 × 10⁻⁶ S cm⁻¹ to 2 × 10⁻⁵ S cm⁻¹) but also raises the Fermi level in the Y–TiO₂ layer. At last, this all-low-temperature processed device yielded a record PCE of 19.3% with an average PCE of 16.6%.

Fig. 8 (a) SEM cross-sectional image of the device. (b) Diagram of energy levels (relative to the vacuum level) of each functional layer in the device. (c) J–V curves for the champion cell without antireflective coating. Reprinted with permission.²³

Table 4 A summary of published representative results of positive PHJPSC performance parameters with different device fabrication methods and configurations

PL	PL thickness (nm)	DP	Device structure	J sc (mA cm^−2)	V oc (V)	FF (%)	PCE (%)	Ref.
MAPbI3−xClx	400–800	OSPD	FTO/bl-TiO2/PL/spiro/Au	20.3	0.89	64	11.4	31
MAPbI3−xClx	∼330	DSVD	FTO/bl-TiO2/PL/spiro/Ag	21.5	1.07	67	15.4	19
MAPbI3−xClx	∼400	SDP	FTO/bl-TiO2/PL/spiro/Au	22.9	0.98	69	14.82	43 and 117
MAPbI3	∼350	VASP	FTO/bl-TiO2/PL/spiro/Ag	19.8	0.924	66.3	12.1	48
MAPbI3	∼300	SDP	FTO/bl-TiO2(rutile)/PL/spiro/Au	19.8	1.05	64	13.7	66
MAPbI3	∼300	SDP	FTO/bl-TiO2/PL/spiro/Au	20.71	1.02	64	13.5	41
MAPbI3	∼350	OSPD	FTO/bl-TiO2/PL/spiro/Ag	21.1	1.04	74	16.2	27
MAPbI3	∼300	SDP	ITO/bl-ZnO/PL/spiro/Ag	20.4	1.03	74.9	15.7	47
MAPbI3	∼300	SDP	PET/ITO/bl-ZnO/PL/spiro/Ag	13.4	1.03	73.9	10.2
MAPbI3	NA	SDP	ITO/bl-ZnO/PC60BM/PL/spiro/MoO3/Ag	18.18	1.00	67	12.2	118
MAPbI3	NA	SDP	ITO/bl-ZnO/PL/spiro/MoO3/Ag	22.4	1.04	57.4	13.4	119
MAPbI3	NA	SDP	PET/ITO/bl-ZnO/PL/spiro/MoO3/Ag	18.4	0.87	49.7	8.03
MAPbI3−xClx	NA	OSPD	ITO/bl-TiO2/PL/P3HT/Ag	21	0.936	69.1	13.6	63 and 64
MAPbI3	NA	SDP	ITO/bl-ZnO/PC60BM/PL/PTB7-Th/MoO3/Ag	15.1	1.03	71	11.04	118
MAPbI3−xClx	∼350	SDP	FTO/bl-TiO2/PL/DR3TBDTT/Au	15.3	0.95	60	8.8	120
MAPbI3	∼400	OSPD	FTO/bl-TiO2/PL/CuSCN/Au	14.4	0.727	61.7	6.4	121
MAPbI3−xClx	NA	OSPD	FTO/bl-TiO2/MAPbI3−xClx/thiophene/spiro/Au	21.3	1.02	68	15.3	122
MAPbI3−xClx	NA	OSPD	FTO/bl-TiO2/MAPbI3−xClx/pyridine/spiro/Au	24.1	1.05	72	16.5
MAPbI3−xClx	∼350	OSPD	ITO/PEIE/bl-TiO2(Y)/PL/spiro/Au	22.75	1.13	75.01	19.3	23

---

4.3.2 Device based on an inverted configuration (Table 5). In 2013, inspired by the structure of bulk-heterojunction (BHJ) organic solar cells, Guo and coworkers firstly reported a series of bilayer inverted devices based on a PHJ of MAPbI₃ perovskite/fullerene-derivative structure, in which MAPbI₃ acted as a “donor” material while C60, PC60BM or an indene–C₆₀ bisadduct (IC60BA) acted as the “acceptor” material (consisting of glass/ITO/PEDOT:PSS as the positive electrode, a thin bathocuproine (BCP) film as the hole-blocking layer, and an Al negative electrode).¹²³ Under standard 1 sun AM1.5 simulated solar irradiation (100 mW cm⁻²), V_oc was varied with the LUMO levels of acceptors, proving the formation of a donor–acceptor interface. Different solvents (GBL or DMF) were used to control the MAPbI₃ perovskite thin film morphologies and optimal devices containing MAPbI₃/C60, MAPbI₃/PC60BM and MAPbI₃/IC60BA showed PCE of 3.0%, 3.9% and 3.4%, respectively. Subsequently, Lam et al. reported a similar system with an ITO/PEDOT:PSS/MAPbI₃/PC60BM/Al structure.¹²⁴ High photovoltaic performance and an IQE close to 100% were observed, suggesting the highly efficient exciton diffusion, charge transfer and charge collection. Devices fabricated by the OSPD method yielded a J_sc of 8.2 mA cm⁻², V_oc of 0.82 V, FF of 0.77 and PCE of 5.2% while a PCE of 7.4% was obtained using SDP, with a J_sc of 10.8 mA cm⁻², V_oc of 0.91 V, and FF of 0.76 under AM1.5G illumination (100 mW cm⁻²).

Around the same time, Snaith, Bolink, and Yang reported independently several perovskite PHJ solar cells with high PCEs up to or over 10%. Snaith et al. demonstrated a device structure consisting of FTO/PEDOT:PSS/MAPbI_3−xCl_x/PC60BM/TiO_x/Al in which a bilayer of PC60BM and compact-TiO_x were employed as the n-type charge collection layer.¹²⁵ A PCE of 9.8% was obtained, with a J_sc of 15.8 mA cm⁻², V_oc of 0.94 V, and FF of 0.66. Also, a PCE of 6.4% was achieved for the same configuration on an ITO-coated PET plastic foil. Interestingly, when fabricating devices on ITO-covered glass, an obvious decrease of J_sc and FF was observed which arises likely from poorer perovskite film formation and lower surface coverage upon the PEDOT:PSS-coated ITO than on the PEDOT:PSS-coated FTO. Bolink et al. reported a MAPbI₃ perovskite solar cell in which a highly oriented pure MAPbI₃ film was sandwiched between an electron-blocking poly-TPD layer and a hole-blocking PC60BM layer (ITO/PEDOT:PSS/poly-TPD/MAPbI₃/PC60BM/Au).¹²⁶ A high J_sc of 16.12 mA cm⁻² and V_oc of 1.05 V revealed that very few electrons and holes recombine and at last, a PCE of 12.04% (cell area = 0.09 cm²) was achieved at the standard solar AM1.5G intensity of 100 mW cm⁻² (the PCE was further improved to 14.8% with a 0.065 cm² solar cell¹²⁷). Yang and coworkers also fabricated two PHJ devices based on ITO/PEDOT:PSS/MAPbI_3−xCl_x/PC60BM/Al and ITO/PEDOT:PSS/MAPbI_3−xCl_x/ZnO/Al, yielding PCEs of 11.5% and 10.53% in a rigid substrate, respectively (a 9.2% efficiency was achieved in PET/ITO/PEDOT:PSS/MAPbI_3−xCl_x/PC60BM/Al-based flexible devices).¹²⁸ Later, Jen et al. reported a simple way to enhance the crystallization of solution-processed perovskite by incorporating additives into its precursor solution to modulate thin film formation.³⁵ The device derived from the 1% DIO solution (FTO/PEDOT:PSS/MAPbI_3−xCl_x/PC60BM/Bis-C60/Ag) exhibited a PCE of 11.8%, with a J_sc of 17.5 mA cm⁻², V_oc of 0.92 V, and FF of 0.73, compared to the 9.0% PCE of the device without using DIO. Similar enhancement was observed in the case of ITO substrates. Also, the relatively lower performance of the ITO-based device than the FTO-based device was consistent with Snaith's experiment. Lee et al. used a new self-organized hole extraction layer, which was composed of PEDOT:PSS and a perfluorinated ionomer (PFI) to modify the interface work function and then reduce the potential energy loss.¹²⁹ The device based on an ITO/PEDOT:PSS/PFI/MAPbI₃/PC60BM/Al structure yielded an enhanced performance (PCE = 11.7%, J_sc = 16.7 mA cm⁻², V_oc = 0.982 V, FF = 0.705) compared to that of an ITO/PEDOT:PSS/MAPbI₃/PC60BM/Al-based device (PCE = 8.1%, J_sc = 14.1 mA cm⁻², V_oc = 0.835 V, FF = 0.685). Also, 8.0% efficiency was achieved in flexible perovskite solar cells with this self-organized hole extraction layer on a PET/ITO substrate. Seok and coworkers also used a solvent-engineering technology, which employed a mixture solution of DMSO : GBL (3 : 7, v/v), to fabricate the uniform and dense perovskite layer, and further improved the PCE to 14.1% in ITO/PEDOT:PSS/MAPbI₃/PC60BM/LiF/Al-based devices.²⁶

Huang and coworkers fabricated two types of PHJ perovskite solar cells with an ITO/PEDOT:PSS/MAPbI₃/PC60BM/C60/BCP/Al or ITO/PEDOT:PSS/MAPbI₃/IC60BA/C60/BCP/Al structure (see Fig. 9).¹³⁰ A unique double fullerene layer was adopted in these devices which could effectively reduce dark current leakage by forming a Schottky junction with the anode. By varying the ratio and concentration of the precursor solutions, the morphology, absorption and crystallization of perovskite films could be tuned and at last, an optimal device with an IC60BA acceptor layer exhibited the best PCE of 12.2%, with a J_sc of 15.7 mA cm⁻², V_oc of 0.97 V, and FF of 0.80. Later, using a new film forming method by interdiffusion of spin-coated stacking layers of PbI₂ and MAI (via TSSD), a high quality film was achieved by Huang et al.⁴⁹ Perovskite based on an ITO/PEDOT:PSS/MAPbI₃/PC60BM/C60/BCP/Al system gave a PCE of 15.3%, with a relatively high J_sc of 20.59 mA cm⁻² under AM 1.5 simulated one sun illumination (further optimization using solvent annealing yielded a PCE of 15.6% with a perovskite thickness of 630 nm (ref. 131). Recently, Lin⁵⁶ and Wu⁵⁰ also obtained high PCEs of 15.4% and 16.3% in ITO/PEDOT:PSS/MAPbI_3−xCl_x/C60/Bphen/Ca/Ag-based and ITO/PEDOT:PSS/MAPbI₃/PC70BM/Ca/Al-based devices using SVD or TSSD.

Fig. 9 (a) The schematic of the device structure. (b) Top view SEM images. (c) Absorption spectra. (d) Photoluminescence spectra and (e) XRD patterns of the iodine perovskite films spun from solutions with a precursor ratio from 0.35 to 1. Reprinted with permission.¹³⁰

Apart from the most widely used PEDOT:PSS, in early 2014, Guo and coworkers replaced the PEDOT:PSS layer with a thin NiO_x interlayer (∼10 nm).¹⁰² The device containing ITO/NiO_x/MAPbI₃/C60/BCP/Al and ITO/NiO_x/MAPbI₃/PC60BM/BCP/Al showed PCEs of 5.7% and 7.8% respectively with enhanced V_oc compared to devices using the PEDOT:PSS layer. The authors attributed this to the low energy loss for holes in the NiO_x/MAPbI₃ junction and a better surface coverage of MAPbI₃ film on the glass/ITO/NiO_x substrate. Sarkar and coworkers also reported two types of MAPbI_3−xCl_x perovskite solar cells with a FTO/NiO/MAPbI_3−xCl_x/PC60BM/Ag or FTO/CuSCN/MAPbI_3−xCl_x/PC60BM/Ag structure.¹³² Devices with this electrodeposited NiO film and CuSCN film exhibited a PCE of 7.26% and 3.8%, respectively. Later, Yang et al. used a sol–gel process to fabricate a thin NiO nanocrystalline film (∼40 nm) for the development of a FTO/NiO/MAPbI₃/PC60BM/Au-based solar cell, and further improved the PCE to 9.11%.¹³³ Recently, by using a thin graphene oxide (GO) (∼2 nm)¹³⁴ or polythiophene (PT) (∼18 nm)¹³⁵ layer as a new p-type conductor instead of PEDOT:PSS, PCEs of 12.4% and 11.8% were achieved in ITO/GO/MAPbI_3−xCl_x/PC60BM/ZnO/Al and ITO/PT/MAPbI₃/C60/BCP/Ag-based devices, respectively.

Table 5 A summary of published representative results of the performance parameters of inverted PHJPSC with different device fabrication methods and configurations

PL	PL thickness (nm)	DP	Device structure	J sc (mA cm^−2)	V oc (V)	FF (%)	PCE (%)	Ref.
a With 1% DIO.
MAPbI3	20–30	OSPD	ITO/PEDOT:PSS/PL/C60/BCP/Al	9.02	0.55	61	3.0	123
MAPbI3	20–30	OSPD	ITO/PEDOT:PSS/PL/PC60BM/BCP/Al	10.32	0.60	63	3.9
MAPbI3	20–30	OSPD	ITO/PEDOT:PSS/PL/IC60BA/BCP/Al	10.03	0.58	58	3.4
MAPbI3	50 ± 5	OSPD	ITO/PEDOT:PSS/PL/PC60BM/Al	8.2	0.82	77	5.2	124
MAPbI3	110 ± 5	SDP	ITO/PEDOT:PSS/PL/PC60BM/Al	10.8	0.91	76	7.4
MAPbI3−xClx	300–400	OSPD	FTO/PEDOT:PSS/PL/PC60BM/TiOx/Al	15.8	0.94	66	9.8	125
MAPbI3−xClx	300–400	OSPD	ITO/PEDOT:PSS/PL/PC60BM/TiOx/Al	14.4	0.92	47	6.3
MAPbI3−xClx	300–400	OSPD	PET/ITO/PEDOT:PSS/PL/PC60BM/TiOx/Al	14.4	0.88	51	6.4
MAPbI3	∼285	DSVD	ITO/PEDOT:PSS/poly-TPD/PL PC60BM/Au	16.12	1.05	67	12.04	126
MAPbI3−xClx	∼340	OSPD	ITO/PEDOT:PSS/PL/PC60BM/Al	18.5	0.87	72	11.5	128
MAPbI3−xClx	∼340	OSPD	ITO/PEDOT:PSS/PL/ZnO/Al	17.29	0.886	68.76	10.53
MAPbI3−xClx	∼340	OSPD	PET/ITO/PEDOT:PSS/PL/PC60BM/Al	16.5	0.86	64	9.2
MAPbI3−xClx	∼400	OSPD	ITO/PEDOT:PSS/PL/PC60BM/Bis-C60/Ag	15	0.90	58	7.9	35
MAPbI3−xClx	∼400	OSPD	ITO/PEDOT:PSS/PL/PC60BM/Bis-C60/Ag	15.6	0.92	71	10.3
MAPbI3−xClx	∼400	OSPD	FTO/PEDOT:PSS/PL/PC60BM/Bis-C60/Ag	16	0.90	62	9
MAPbI3−xClx	∼400	OSPD	FTO/PEDOT:PSS/PL/PC60BM/Bis-C60/Ag	17.5	0.92	73	11.8
MAPbI3	NA	TSSD	ITO/PEDOT:PSS/PFI/PL/PC60BM/Al	16.7	0.982	70.5	11.7	129
MAPbI3	NA	TSSD	PET/ITO/PEDOT:PSS/PFI/PL/PC60BM/Al	15.5	1.04	49.9	8.0
MAPbI3	∼290	OSPD	ITO/PEDOT:PSS/PL/PC60BM/LiF/Al	20.7	0.886	78.3	14.1	26
MAPbI3	∼140	OSPD	ITO/PEDOT:PSS/PL/PC60BM/C60/BCP/Al	15.9	0.88	72.2	10.1	130
MAPbI3	∼140	OSPD	ITO/PEDOT:PSS/PL/IC60BA/C60/BCP/Al	15.7	0.97	80.1	12.2
MAPbI3	270–300	TSSD	ITO/PEDOT:PSS/PL/PC60BM/C60/BCP/Al	19.6	0.99	79.3	15.4	49
MAPbI3	∼630	TSSD	ITO/PEDOT:PSS/PL/PC60BM/C60/BCP/Al	21.0 ± 0.5	0.96 ± 0.02	76.0 ± 1.5	15.6	131
MAPbI3−xClx	∼430	SVD	ITO/PEDOT:PSS/PL/C60/Bphen/Ca/Ag	20.9	1.02	72.2	15.4	56
MAPbI3	∼360	TSSD	ITO/PEDOT:PSS/PL/PC70BM/Ca/Al	19.98	1.05	78	16.31	50
MAPbI3	∼60	OSPD	ITO/NiOx/PL/C60/BCP/Al	12.95	0.74	60	5.7	102
MAPbI3	∼60	OSPD	ITO/NiOx/PL/PC60BM/BCP/Al	12.43	0.92	68	7.8
MAPbI3−xClx	NA	DSVD	FTO/NiO/PL/PC60BM/Ag	14.2	0.786	65	7.26	132
MAPbI3−xClx	NA	DSVD	FTO/CuSCN/PL/PC60BM/Ag	NA	NA	NA	3.8
MAPbI3	∼250	SDP	FTO/NiO/PL/PC60BM/Au	16.27	0.882	63.5	9.11	133
MAPbI3−xClx	∼170	OSPD	ITO/GO/PL/PC60BM/ZnO/Al	17.46	1.00	71	12.40	134
MAPbI3	250 ± 20	SDP	ITO/PT/PL/C60/BCP/Ag	16.2	1.03	70.7	11.8	135

---

5. Hole-transporting materials (HTMs) for perovskite solar cells

Despite the rapid increase in efficiency associated with the evolution of different types of perovskites and device fabrication techniques, the development of HTM is very limited and mainly focused on organic compounds. In particular, the bulky 3-D spiro-OMeTAD (see Fig. 10) with the twisted spirobifluorene center has been proven as the most potential hole conductor for state-of-the-art perovskite-based devices since it was firstly introduced in 1998 for ss-DSSCs.¹⁰ So far, based on spiro-OMeTAD HTM, MMOPSCs,²¹ MSSCs²² and PHJPSCs²³ have respectively showed their best device performance. However, the tedious synthesis of spiro-OMeTAD represents a potential hurdle for future commercialization due to its high cost. So looking for new kinds of alternative HTMs to spiro-OMeTAD with a simpler synthetic route, lower production cost and comparable device performance is very necessary. Also, a compatible HOMO energy level relative to perovskites and high charge-carrier mobility should also be considered. In the past two years, many new-type HTMs, including organic small molecules, polymers and inorganic materials, have been used for perovskite solar cells and promising performance has been achieved for the corresponding devices.

Fig. 10 The chemical structure of spiro-OMeTAD.

5.1 HTMs based on small molecules (Table 6)

5.1.1 Small molecules based on phenylamine derivatives (Fig. 11 and 12). Due to the great success achieved in spiro-OMeTAD-based devices, at present, small molecule HTMs containing phenylamine derivatives have been widely investigated for application in perovskite solar cells.

Early in 2013, Hodes and coworkers employed a small molecule TPD as the HTM for perovskite solar cells with MAPbBr₃-coated alumina scaffolds and achieved a low PCE of 0.67%.¹⁰⁵ Later, Johansson et al. reported a small molecule hole conductor DEH for fabricating MAPbI₃ perovskite solar cells, and found that, compared with the spiro-OMeTAD-based device, the device with DEH showed a faster recombination rate (∼100 times higher), leading to a decreased performance with a low PCE of 1.6%.¹³⁶ The authors suggested that for perovskite solar cells, the molecular structure of the HTM should be designed to block close contact between the perovskite and the hole on the HTM, which favors reduction of electronic coupling and charge recombination. Thus molecules with a bulky, twisty 3-D structure (e.g. phenylamine) may be a potential choice for future HTMs.

In the end of 2013, by replacing the spirobifluorene core of spiro-OMeTAD with a pyrene core, three HTMs based on pyrene-core arylamine derivatives, Py-A, Py-B and Py-C, were investigated by Seok et al.¹³⁷ Comparable J_sc (∼20 mA cm⁻²) and PCE (∼12%) were observed in the cells fabricated with Py-B and Py-C, while the Py-A-based device showed low performance (J_sc = 10.8 mA cm⁻², PCE = 3.3%), which was due to the insufficient driving force for hole injection from MAPbI₃ (−5.44 eV) to Py-A (−5.41 eV). The optimal device based on Py-B and Py-C demonstrated PCEs of 12.3% and 12.4%, respectively, showing the possibility of using HTMs with the new structure instead of spiro-OMeTAD for efficient and low-cost organometal halide perovskite solar cells. Subsequently, a series of dumbbell-shaped or star-shaped molecules with phenylamine derivatives, especially triphenylamine (TPA), as the terminal group or core were synthesized and some of them exhibited comparable device performance to that of spiro-OMeTAD-based perovskite solar cells.

In 2014, Mhaisalkar and coworkers developed two thiophene/TPA-based HTMs H101¹³⁸ and KTM3,¹³⁹ incorporating 3,4-ethylenedioxythiophene or swivel-cruciform 3,3-bithiophene as the core unit terminated with two or four TPAs. Through appropriate chemical doping with tris(2-(1H-pyrazol-1-yl)pyridine)cobalt(III) tris(hexafluorophosphate) (FK102) (for H101) or bis(2,6-di(1H-pyrazol-1-yl)pyridine)cobalt(III) tris(bis(trifluoromethylsulfonyl)-imide) (FK269) (for KTM3), optimal mesoporous devices based on H101 or KTM3 yielded PCEs of 13.8% or 11%, respectively under AM1.5G solar simulation (100 mW cm⁻²). After that, the authors reported three star-shaped HTMs (T101, T102 and T103) based on a rigid triptycene central core.¹⁴⁰ The HOMO levels of T101, T102 and T103 calculated from CV are −5.29, −5.35 and −5.33 eV, respectively, which are lower than that of spiro-OMeTAD (−5.22 eV). By modifying the linkage form between triptycene and diphenylamine groups, devices based on T101, T102 and T103 yielded PCEs of 8.42%, 12.24% and 12.38%, respectively with relatively high V_oc. Ko et al. also reported two star-shaped HTMs OMeTPA-TPA and OMeTPA-FA with an incorporated fused quinolizino acridine or TPA as a core unit.¹⁴¹ Devices comprising FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/OMeTPA-TPA/Au and FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/OMeTPA-FA/Au demonstrated PCEs of 12.31% and 13.63% using three dopants composed of 4-tert-butylpyridine (TBP), bis(trifluoromethane)sulfonimide lithium salt (LiTFSI), and tris(2-(1H-pyrazol-1-yl)-4-tert-butylpyridine)cobalt(III) tris(bis(trifluoromethylsulfonyl)imide) (FK209) into the HTMs. A noteworthy feature is that even without any p-type additives, a high PCE of 11.7% was still achieved in the OMeTPA-FA-based device. Later, Meng and coworkers developed two simple HTMs TPBS and TPBC by introducing electron-donating groups asymmetrically into the N,N,N′,N′-tetraphenyl-benzidine core, and achieved impressive PCEs of 10.29% and 13.10% respectively in TPBS- and TPBC-based devices without any doping.¹⁴²

Recently, Sun and coworkers fabricated FTO/bl-TiO₂/mp-TiO₂/MAPbI_3−xCl_x/HTM/Ag-based perovskite solar cells using two carbazole-based small molecules X19 and X51 as HTMs.¹⁴³ Almost three times higher hole mobilities than that of spiro-OMeTAD (X19: 1.19 × 10⁻⁴ cm² V⁻¹ s⁻¹, X51: 1.51 × 10⁻⁴ cm² V⁻¹ s⁻¹, spiro-OMeTAD: 5.31 × 10⁻⁵ cm² V⁻¹ s⁻¹) were measured for these two materials. Optimal devices with X19 and X51 HTMs yielded PCEs of 7.6% and 9.8% under AM1.5G (100 mW cm⁻²) illumination. Lee et al. reported the synthesis and characterization of three carbazole-based HTMs with two-arm and three-arm type structures SGT-404. SGT-405 and SGT-407, which are linked through phenylene, diphenylene or triphenyl amine derived core units.¹⁴⁴ FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/HTM/Au-based devices yielded high efficiencies of 13.28% (SGT-404), 14.79% (SGT-405) and 13.86% (SGT-407), which were comparable to that of the device employing commercial spiro-OMeTAD (15.23%).

Xiao and coworkers synthesized a series of HTMs based on a diphenyl core with TPA as the terminal group bridged with olefinic bonds of different lengths, which is the first case of adoption of small molecule HTMs with a linear π-conjugated structure.¹⁴⁵ Without any dopant for HTMs, the FTO/TiO₂/MAPbI₃/2TPA-2-DP/Au solar cell exhibits an encouraging PCE of 9.1%, with a V_oc of 0.94 V, J_sc of 16.3 mA cm⁻², and FF of 0.597. Meng et al. reported two TPA-based hole conductors HTM1 and HTM2 containing butadiene derivatives.¹⁴⁶ Both of them exhibited high hole mobilities, which were 2.98 × 10⁻³ cm² V⁻¹ s⁻¹ for HTM1 and 1.27 × 10⁻³ cm² V⁻¹ s⁻¹ for HTM2. MAPbI₃ perovskite solar cells showed a PCE of 11.34% for HTM1 and 11.63% for HTM2, respectively. Later, the authors employed another low-cost, non-traditional TPA-based HTM PNBA in FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/PNBA/Au-based devices and achieved a PCE of 11.4%.¹⁴⁷

Besides looking for alternative HTMs with new structures, studies based on the classic spiro-OMeTAD, such as structure optimization or energy level engineering, are still needed. In 2014, Seok et al. synthesized and reported two more spiro-OMeTAD isomers by replacing pareb (p)-OMe groups in spiro-OMeTAD (pp-spiro-OMeTAD) with ortho (o)- and meta (m)-OMe groups (which are named as po-spiro-OMeTAD and pm-spiro-OMeTAD, respectively; see Fig. 11).⁶⁰ Similar J_sc and V_oc were observed for the three pp-, pm-, and po-spiro-OMeTAD derivatives while po-spiro-OMeTAD showed the highest FF value which can be attributed to its low series resistance (R_s) and shunt resistance (R_sh). As a result, a higher PCE of 16.7% was obtained for the po-spiro-OMeTAD-based device compared with the classic pp-spiro-OMeTAD with a PCE of ∼15%. Leo and coworkers used five TPA-based HTMs (including MeO-TPD, spiro-MeO-TPD, spiro-TTB, spiro-TAD and BPAPF) with spiro-OMeTAD to correlate their HOMO energy levels with the V_oc of the inverted planar MAPbI_3−xCl_x-based devices, and highlighted the delicate energetic balance between the driving force for hole-extraction and maximizing the photovoltage (see Fig. 12).¹⁴⁸

Fig. 11 A summary of small molecule HTMs based on phenylamine derivatives.

Fig. 12 (a) Chemical structures of HTMs reported by Leo et al. (b) General device architecture and corresponding energy level diagram. Reprinted with permission.¹⁴⁸

5.1.2 Small molecules without phenylamine derivatives (Fig. 13). Compared with the rapid development of small molecule HTMs containing phenylamine derivatives, studies on non-phenylamine derivatives-based HTMs are relatively scarce. Earlier studies in 2013 using PDI,¹⁰⁵PC60BM¹⁰⁵ or CBP¹⁴⁹ as HTMs reported very low device performances (PCE < 3%). Nowadays, studies on the phenylamine-free small molecule HTMs are mainly focused on the linear or branched conjugated structures. These classes of materials always show high hole mobility and good light harvesting ability, and some of them even afford additional contribution to the photocurrent generation of the perovskite solar cells.^150,151

Fig. 13 A summary of small molecule HTMs without phenylamine derivatives.

In 2014, Han and coworkers employed an alkyl-substituted tetrathiafulvalene TTF-1 as the HTM for an MAPbI₃-based device.¹⁵² The long alkyl chains in TTF-1 not only improve its solubility for solution processes, but also maintain a certain amount of intermolecular stacking due to the fastener-effect of the long alkyl chains. No p-type dopants were added, and FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/TTF-1/Ag-based devices yielded a PCE of 11.03%, with a J_sc of 19.9 mA cm⁻², V_oc of 0.86 V, and FF of 0.644.

Xiao et al. reported a low band gap oligothiophene HTM named DR3TBDTT, containing benzodithiophene as the central block and ethylrhodanine as the end group.¹²⁰ No ion additive was mixed into DR3TBDTT but for the addition of a small amount of insulated polydimethylsiloxane (PDMS), which acted as a flow agent to improve film formation of HTM on perovskites. Optimal PHJ devices with a FTO/bl-TiO₂/MAPbI_3−xCl_x/DR3TBDTT/Au structure yielded a PCE of 8.8% with excellent stability. Grätzel et al. also developed two more S,N-heteroacene-based dopant-free oligothiophenes G1 and G2 as HTMs for FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/HTM/Au-based perovskite solar cells, and achieved PCEs of 10.5% and 9.5%, respectively.¹⁵⁰ Nazeeruddin and coworkers synthesized a flattened star-shaped molecule Fused-F with quinolizino acridine as the core.¹⁵¹ The device based on Fused-F achieved a high PCE of 12.8% under the illumination of 98.8 mW cm⁻², with a V_oc of 1.04 V, J_sc of 17.9 mA cm⁻², and FF of 0.68 without any additives.

Table 6 A summary of published representative results of small molecule HTM-based device performance parameters with different device fabrication methods and configurations

HTM	E HOMO (eV)	p-Doping	Device structure	DP	J sc (mA cm^−2)	V oc (V)	FF (%)	PCE (%)	Ref.
a1	−5.3	NA	FTO/bl-TiO2/mp-Al2O3/MAPbBr3/a1/Au	OSPD	1.22	1.20	46	0.67	105
a2	NA	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/a2/Au	OSPD	NA	NA	NA	1.6	136
a3	−5.41	LiTFSI/TBP/FK209	FTO/bl-TiO2/mp-TiO2/MAPbI3/a3/Au	OSPD	10.8	0.89	34.6	3.3	137
a4	−5.25	LiTFSI/TBP/FK209	FTO/bl-TiO2/mp-TiO2/MAPbI3/a4/Au	OSPD	20.4	0.95	63.7	12.3
a5	−5.11	LiTFSI/TBP/FK209	FTO/bl-TiO2/mp-TiO2/MAPbI3/a5/Au	OSPD	20.2	0.89	69.4	12.4
a6	−5.16	LiTFSI/TBP/FK102	FTO/bl-TiO2/mp-TiO2/MAPbI3/a6/Au	SDP	20.5	1.04	65	13.8	138
a7	−5.29	LiTFSI/TBP/FK269	FTO/bl-TiO2/mp-TiO2/MAPbI3/a7/Au	SDP	13.0	1.08	78.3	11.0	139
a8	−5.29	LiTFSI/TBP/FK102	FTO/bl-TiO2/mp-TiO2/MAPbI3/a8/Au	SDP	13.5	0.996	62.6	8.42	140
a9	−5.35	LiTFSI/TBP/FK102	FTO/bl-TiO2/mp-TiO2/MAPbI3/a9/Au	SDP	17.2	1.03	69.1	12.24
a10	−5.33	LiTFSI/TBP/FK102	FTO/bl-TiO2/mp-TiO2/MAPbI3/a10/Au	SDP	20.3	0.985	61.9	12.38
a11	−5.13	LiTFSI/TBP/FK209	FTO/bl-TiO2/mp-TiO2/MAPbI3/a11/Au	SDP	20.88	0.946	62	12.31	141
a12	−5.14	LiTFSI/TBP/FK209	FTO/bl-TiO2/mp-TiO2/MAPbI3/a12/Au	SDP	20.98	0.972	67	13.63
a13	−5.30	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3/a13/Au	SDP	15.75	0.932	70	10.29	142
a14	−5.33	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3/a14/Au	SDP	19.32	0.942	72	13.10
a15	NA	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3−xClx/a15/Ag	OSPD	17.14	0.76	58	7.6	143
a16	NA	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3−xClx/a16/Ag	OSPD	16.79	0.88	66	9.8
a17	NA	LiTFSI/TBP/FK209	FTO/bl-TiO2/mp-TiO2/MAPbI3/a17/Au	SDP	19.76	0.963	69.8	13.28	144
a18	NA	LiTFSI/TBP/FK209	FTO/bl-TiO2/mp-TiO2/MAPbI3/a18/Au	SDP	20.28	1.023	71.3	14.79
a19	NA	LiTFSI/TBP/FK209	FTO/bl-TiO2/mp-TiO2/MAPbI3/a19/Au	SDP	20.35	0.993	68.6	13.86
a20	−4.96	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3/a20/Au	OSPD	16.3	0.94	59.7	9.1	145
a21	−5.35	NA	FTO/bl-TiO2/mp-TiO2/MAPbI3/a21/Au	SDP	18.1	0.921	68	11.34	146
a22	−5.23	NA	FTO/bl-TiO2/mp-TiO2/MAPbI3/a22/Au	SDP	17.9	0.942	69	11.63
a23	−5.42	NA	FTO/bl-TiO2/mp-TiO2/MAPbI3/a23/Au	SDP	17.5	0.945	68.9	11.4	147
a24	−5.31	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/a24/Au	OSPD	21.1	1.01	65.2	13.9	60
a25	−5.22	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/a25/Au	OSPD	21.2	1.02	77.6	16.7
spiro	−5.0	F6-TCNNQ	ITO/dopant/spiro/MAPbI3−xClx/C60/Ag	DSVD	14.4	0.795	69	7.8	148
a26	−5.1	F6-TCNNQ	ITO/dopant/a26/MAPbI3−xClx/C60/Ag	DSVD	14.9	0.863	69	8.7
a27	−5.1	F6-TCNNQ	ITO/dopant/a27/MAPbI3−xClx/C60/Ag	DSVD	16	1.03	66	10.9
a28	−5.3	F6-TCNNQ	ITO/dopant/a28/MAPbI3−xClx/C60/Ag	DSVD	16.1	0.968	70	10.9
a29	−5.4	F6-TCNNQ	ITO/dopant/a29/MAPbI3−xClx/C60/Ag	DSVD	12.4	0.820	58	6.7
a30	−5.6	NDP9	ITO/dopant/a30/MAPbI3−xClx/C60/Ag	DSVD	0.72	0.835	13	0.08
b1	−5.8	NA	FTO/bl-TiO2/mp-Al2O3/MAPbBr3/b1/Au	OSPD	1.08	1.30	40	0.56	105
b1	−5.8	NA	FTO/bl-TiO2/mp-TiO2/MAPbBr3/b1/Au	OSPD	1.14	1.00	41	0.47
b2	−6.1	NA	FTO/bl-TiO2/mp-Al2O3/MAPbBr3/b2/Au	OSPD	1.57	1.06	43	0.72
b3	−6.23	LiTFSI/TBP	FTO/bl-TiO2/mp-Al2O3/MAPbBr3−xClx/b3/Au	OSPD	4.0	1.50	46	2.7	149
b4	−5.05	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3/b4/Ag	SDP	19.9	0.86	64.4	11.03	152
b5	−5.39	PDMS	FTO/bl-TiO2/MAPbI3−xClx/b5/Au	SDP	15.3	0.95	60	8.8	120
b6	−5.26	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3/b6/Au	SDP	16.4	0.992	65	10.5	150
b7	−5.10	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3/b7/Au	SDP	15.2	0.90	68	9.5
b8	−5.23	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3/b8/Au	SDP	17.9	1.04	68	12.8	151

---

5.2 HTMs based on polymers (Table 7 and Fig. 14)

As spiro-OMeTAD is the model material in small molecule HTMs, P3HT is the model material in polymer HTMs. In previous reports, low PCEs less than 1% were obtained by Hodes and Qiu et al. in the mp-TiO₂/MAPbBr₃/P3HT/Au- or mp-Al₂O₃/MAPbBr₃/P3HT/Au-based devices.^105,153 Later, by using MAPbI₃ or MAPbI_3−xCl_x instead of MAPbBr₃ as the active layer, PCEs from 6.7% to over 13% were achieved by several groups,^{20,63,64,154–157} showing that P3HT can be a suitable HTM for efficient and low cost perovskite based solar cells. Bian and coworkers also used a thin polythiophene film prepared via electrochemical polymerization as the HTM in an ITO/PT/MAPbI₃/C60/BCP/Ag-based device, and observed a PCE of 11.8%.¹³⁵

In 2013, Seok and coworkers reported the use of PTAA as the hole conductor for perovskite solar cells.²⁰ A pillared structure consisting of 3-D composites of TiO₂/MAPbI₃ with partially infiltrated PTAA was observed and the optimized solar cell device based on FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/PTAA/Au exhibited a J_sc of 16.5 mA cm⁻², V_oc of 0.997 V and FF of 0.727, yielding a PCE of 12.0% under standard AM 1.5 conditions. Subsequently, two more triarylamine polymer derivatives containing fluorene and indenofluorene, named PF8-TAA, and PIF8-TAA, combined with PTAA, were studied to understand and optimize the V_oc of the resulting perovskite solar cells with an architecture consisting of FTO/TiO₂/MAPbBr₃/HTM/Au or FTO/TiO₂/MAPbI₃/HTM/Au.¹⁵⁸ Both the perovskite materials and the HOMO level of the HTM determine the high voltage output and at last, a PCE of 6.7% with a high V_oc of 1.40 V was obtained in PIF8-TAA-based MAPbBr₃ perovskite solar cells, which is the highest PCE reported for MAPbBr₃ perovskite solar cells. Also, the PCE of the PTAA/MAPbI₃-based device was further improved to 16.2%. Later, Yang and coworkers reported three polyfluorene derivatives, named PFO, TFB and PFB, as the HTMs in MAPbI₃-based devices. The highest hole extraction rate of TFB was observed and the optimal device (FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/TFB/Au) yielded a PCE of 10.92% and 12.8% via the OSPD and SDP techniques.¹⁵⁹ Grätzel et al. also developed a novel 2,4-dimethoxy-phenyl substituted triarylamine oligomer S197 as the HTM for FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/S197/Au-based device, and achieved comparable PCE (12%) with PTAA-based devices.¹⁶⁰

Conjugated donor–accepter (D–A) copolymers are widely used in OPVs. D–A polymers with high hole mobility and an appropriate HOMO level can be potential HTM materials for perovskite solar cells. Seok et al. fabricated MAPbI₃ perovskite solar cells with D–A copolymers PCPDTBT or PCDTBT as the HTM, in which benzothiadiazole was used as the acceptor and carbazole or cyclopentadithiophene as the donor.²⁰ A PCE of 4.2% (J_sc = 10.5 mA cm⁻², V_oc = 0.92 V, FF = 0.437) was observed for PCDTBT and a PCE of 5.3% (J_sc = 10.3 mA cm⁻², V_oc = 0.77 V, FF = 0.667) was observed for PCPDTBT in FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/HTM/Au-based devices.

Qiu and Park et al. reported two diketopyrrolopyrrole-containing polymer HTMs PCBTDPP and PDPPDBTE. PCEs of 5.55% and 9.2% were obtained respectively by using a FTO/TiO₂/MAPbI₃/HTM/Au structure at 100 mA cm⁻² illumination (AM1.5G);^153,161 it is worth noting that without encapsulation, devices containing PCBTDPP or PDPPDBTE showed excellent long-term stability in air at room temperature, guaranteeing the practical applicability of these solid-state hybrid solar cells under outdoor working conditions.

Park and coworkers also investigated two D–A polymer HTMs PTB-BO and PTB-DCB21 and found that,¹⁶² compared with PTB-BO without a 3,4-dichlorobenzyl group, the introduction of functionalized PTB-DCB21 was more effective in accelerating electron transport and retarding charge recombination. As a result, under the condition of the absence of any additives, higher PCE of 8.7% was achieved for PTB-DCB21-based devices while PTB-BO-based devices showed a PCE of 7.4%. Recently, Lee et al. fabricated a PHJ perovskite solar cell using PTB7-Th as the HTM. Devices using an ITO/bl-ZnO/PC60BM/MAPbI₃/PTB7-Th/MoO₃/Ag structure yielded a PCE of 11.04%, with a J_sc of 15.1 mA cm⁻², V_oc of 1.03 V, and FF of 0.71.¹¹⁸

Fig. 14 A summary of polymer HTMs.

However, despite the wide absorption of low band gap polymers, it has been proven that the role of polymers is only limited to a charge transporting layer rather than as a light harvester because no obvious contribution to external quantum efficiency is found in the long wavelength range. Recently, Snaith and coworkers combined a MAPbI_3−xCl_x perovskite with a fullerene selfassembled monolayer (C60SAM) functionalized TiO₂ to produce a dual absorbing, perovskite–polymer HTM hybrid solar cell (see Fig. 15).¹⁶³ In this case, electron transfer from the perovskite to the TiO₂ can be blocked and the V_oc loss reduced. C60SAM acts as a very effective electron acceptor from the perovskite and the polymer HTM, additionally providing polymer photoactivation. With C60SAM fullerene functionalization, an increased PCE from 0.17% to 0.43% was observed in no-perovskite devices containing P3HT. MAPbI_3−xCl_x perovskite solar cells based on P3HT or PCPDTBT showed increased PCE from 3.8% to 6.7% or 0.58% to 6.84%, respectively.

Fig. 15 (a) Absorption spectra of P3HT and perovskite films with and without C60SAM fullerene functionalization. (b) The schematic of the device structure. Reprinted with permission.¹⁶³

Table 7 A summary of published representative results of polymer HTM-based device performance parameters with different device fabrication methods and configurations

HTM	E HOMO (eV)	p-Doping	Device structure	DP	J sc (mA cm^−2)	V oc (V)	FF (%)	PCE (%)	Ref.
c1	−5.2	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c1/Au	OSPD	16.5	0.997	72.7	12	20
c1	−5.2	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbBr3/c1/Au	OSPD	6.6	1.29	70	5.9	158
c2	−5.44	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbBr3/c2/Au	OSPD	6.3	1.36	70	6.0
c2	−5.44	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c2/Au	OSPD	8.9	0.92	56	4.6
c3	−5.51	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbBr3/c3/Au	OSPD	6.1	1.40	79	6.7
c3	−5.51	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c3/Au	OSPD	19.0	1.04	46	9.1
c4	−5.8	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c4/Au	OSPD	3.6	0.61	56	1.22	159
c5	−5.3	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c5/Au	OSPD	17.5	0.96	65	10.92
c5	−5.3	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c5/Au	SDP	NA	NA	NA	12.8
c6	−5.1	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c6/Au	OSPD	13.8	0.91	64	8.03
c7	−5.12	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c7/Au	SDP	17.6	0.967	70	12	160
c8	−5.3	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c8/Au	OSPD	10.3	0.77	66.7	5.3	20
c9	−5.45	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c9/Au	OSPD	10.5	0.92	43.7	4.2
c10	−5.4	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbBr3/c10/Au	OSPD	4.47	1.16	59	3.04	153
c10	−5.4	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3/c10/Au	OSPD	13.86	0.83	48	5.55
c11	−5.4	LiTFSI/TBP	FTO/bl-TiO2/mp-TiO2/MAPbI3/c11/Au	OSPD	14.4	0.855	74.9	9.2	161
c12	−5.22	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3/c12/Au	OSPD	14.35	0.827	62	7.4	162
c13	−5.25	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3/c13/Au	OSPD	15.35	0.888	64	8.7
c14	−5.22	No dopant	ITO/bl-ZnO/PC60BM/MAPbI3/c14/MoO3/Ag	SDP	15.1	1.03	71	11.04	118
P3HT	−5.2	No dopant	FTO/bl-TiO2/mp-TiO2/P3HT/Ag	—	0.69	0.42	58	0.17	163
P3HT	−5.2	No dopant	FTO/bl-TiO2/mp-TiO2/C60SAM/P3HT/Ag	—	1.6	0.5	55	0.43
P3HT	−5.2	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3−xClx/P3HT/Ag	OSPD	10.1	0.68	55.3	3.8
P3HT	−5.2	No dopant	FTO/bl-TiO2/mp-TiO2/C60SAM/MAPbI3−xClx/P3HT/Ag	OSPD	14.9	0.81	55.5	6.7
c8	−5.3	No dopant	FTO/bl-TiO2/mp-TiO2/MAPbI3−xClx/c8/Ag	OSPD	5.02	0.3	40	0.58
c8	−5.3	No dopant	FTO/bl-TiO2/mp-TiO2/C60SAM/MAPbI3−xClx/c8/Ag	OSPD	15.6	0.88	51	6.84

---

5.3 HTMs based on inorganics

Compared with organic HTMs, inorganic p-type semiconductors appear to be an ideal choice due to their high mobility, stability, ease of synthesis and low cost. However, until now, the study on inorganic HTMs for perovskite solar cells is very limited. In 2013, Kamat and coworkers reported a FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/CuI/Au structure device using CuI as the HTM and achieved a PCE of 6%, with a relatively lower V_oc of 0.55 V compared to that (0.79 V) of the device with spiro-OMeTAD.¹⁶⁴ The authors attributed this to the higher recombination in the CuI-based device and this research highlighted the direction to develop all-inorganic materials for perovskite-based photovoltaic devices. Subsequently, Sarkar,¹³² Ito,^165,166 Tena-Zaera¹²¹ and Nazeeruddin¹⁶⁷et al. independently fabricated a conventional or inverted device based on CuSCN HTM and achieved a PCE from 3.8% to 12.4%.

Inspired by the application of a NiO layer in polymer bulk-heterojunction solar cells,^168,169 Guo and Sarkar et al. developed a series of perovskite-based mesoscopic or PHJ solar cells using spin-coated NiO_x thin film,¹⁰² nanocrystalline NiO^101,104 or electrodeposited NiO film¹³² on a glass/ITO or FTO electrode as the HTM, and the PCE increased from 5.7% to 11.6% was obtained finally.

5.4 HTM-free perovskite solar cells

Apart from the widely used TiO₂/perovskite/HTM configuration, another configuration, in which n-type metal oxides combine with perovskites to form a p–n junction without additional HTMs, was developed. Organometal halide perovskite acts both as a light harvester and as a hole conductor simultaneously. Elimination of the hole conductor can improve the stability, lower the cost and simplify the fabrication process.

In 2012, Etgar and coworkers firstly reported a HTM-free perovskite solar cell with a FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/Au structure using TiO₂ (anatase) nanosheets as the electron collector.¹⁴ The device based on this simple MAPbI₃ perovskite/TiO₂ heterojunction showed a promising PCE of 5.5%, with a J_sc of 16.1 mA cm⁻², V_oc of 0.63 V, and FF of 0.57 under standard AM 1.5 solar light of 100 mW cm⁻² intensity. Subsequently, the authors further improved the photovoltaic performance to 8% (ref. 170) then to 10.85% (ref. 171) through the optimization of the MAPbI₃ perovskite film formation.⁵⁴ In 2014, Etgar et al. introduced Br ions into the perovskite structure to construct MAPbI_nBr_3−n (where 0 ≤ n ≤ 3) as a hole conductor and light harvester in the solar cell.¹⁷² Compared to the pure MAPbI₃ which yielded a PCE of 7.2%, perovskite film with an MABr to MAI molar ratio of 1 : 2 (in the dip solution) exhibited a PCE of 8.54% with improved stability.

Considering the Schottky contact at the metal–semiconductor interface between MAPbI₃ and Au, in 2013, Meng et al. deposited an ultrathin Al₂O₃ insulator layer on a MAPbI₃ layer to construct a metal–insulator–semiconductor back contact and an enhanced PCE from 3.30% to 5.07% was obtained.¹⁷³ Later, they developed a simple solution process to engineer the M–S interface using a thin wide band gap organic semiconductor N,N,N′,N′-tetraphenyl-benzidine layer, yielding an enhanced PCE from 5.26% to 6.71%.¹⁷⁴ In 2014, using an ideal model for a single heterojunction solar cell, Meng and coworkers confirmed the heterojunction nature of the TiO₂/MAPbI₃/Au cell and improved PCEs to over 10% with high V_oc over 0.9 V in a HTM-free perovskite system.^42,175 Recently, by the in situ preparation of a perovskite sensitized photoanode, Xiao et al. also achieved PCEs of 9.03% (ref. 87) and 10.03% (ref. 176) in a TiO₂ nanofiber- and TiO₂ nanoparticle-based device. Zhang et al. deposited MAPbI₃ on the oriented rod-type TiO₂ by a solvothermal process and a PCE of 4.2% was obtained with good stability.⁶⁵ Kanatzidis et al. achieved a PCE of over 10.6% with a negligible standard deviation by depositing MAPbI₃via a facile low-temperature (<150 °C), gas–solid crystallization process.¹⁷⁷

The inexpensive and abundantly available carbon (work function: −5.0 eV) may be an ideal material to substitute Au or Ag as a counter electrode (CE) in perovskite solar cells. In 2013, Han and coworkers initially developed a HTM-free fully printable mp-TiO₂/MAPbI₃ heterojunction solar cell with low-cost carbon CE.¹⁷⁸ In this device, a double layer of mesoporous TiO₂ and ZrO₂ is firstly deposited on the top of FTO/bl-TiO₂ substrates, followed by the printing of a porous carbon black/graphite composite as the CE (see Fig. 16a). The inserted ZrO₂ layer acts as an insulating layer to prevent short circuit. Then the MAPbI₃ perovskite is infiltrated into the porous TiO₂/ZrO₂ scaffold by drop-casting a solution through the printed carbon layer. The authors selected two types of carbon CEs including a carbon black/flaky graphite (FG) composite and carbon black/spheroidal graphite (SG) composite to fabricate devices, and achieved PCEs of 4.08% (J_sc = 10.6 mA cm⁻², V_oc = 0.825 V, FF = 0.46) and 6.64% (J_sc = 12.4 mA cm⁻², V_oc = 0.878 V, FF = 0.61), respectively. The main difference in FF can be attributed to the different morphologies of FG- and SG-based CEs. In the FG-based device, large graphite sheets were stacked on the top of ZrO₂, while a loose structure was observed in SG-based CE, which is more beneficial to the pore-filling of MAPbI₃ in the TiO₂ films. Later, via optimization of the carbon CEs¹⁷⁹ (e.g. ordered mesoporous carbon/FG), TiO₂ nanostructure⁸⁰ (e.g. TiO₂ nanosheets), perovskite deposition method⁸⁰ or perovskite structure^180,181 (e.g. (5-AVA)_x(MA)_1−xPbI₃ or (FA)_x(MA)_1−xPbI₃), higher PCE up to ∼13% was achieved, which is the highest reported value for HTM-free perovskite solar cells. Zhao et al. also replaced the insulating ZrO₂ with the p-type mesoscopic NiO, and observed enhanced performances in the TiO₂/mp-NiO(MAPbI₃)/carbon-based device (J_sc = 18.2 mA cm⁻², V_oc = 0.89 V, FF = 0.71, PCE = 11.4%) compared with the TiO₂/mp-ZrO₂(MAPbI₃)/carbon-based device (J_sc = 16.4 mA cm⁻², V_oc = 0.818 V, FF = 0.60, PCE = 8.2%), which was attributed to the enlarged electron lifetime and the augmented interfacial charge transfer process on the carbon counter electrode.¹⁸²

Fig. 16 (a) The schematic structure of a carbon based monolithic device. Reprinted with permission.¹⁸¹ (b) Device architecture and energy levels (relative to vacuum) of various device components. Reprinted with permission.¹⁸³

Yang¹⁸³ and Ma¹⁸⁴ also reported a low-temperature-processed carbon CE-based device with a conventional FTO/bl-TiO₂/mp-TiO₂/MAPbI₃/carbon CE structure at the same time. Similar PCE of 8.31% and 9.08% were achieved and both devices demonstrated good stability (see Fig. 16b). Later, slightly higher PCE of 10.2% was reported by Meng et al. in a similar device, in which the counter electrode contained a composition of graphite and carbon black.¹⁸⁵ Recently, by using the inkjet printing technique, the first example of planar carbon-based perovskite solar cell FTO/bl-TiO₂/MAPbI₃/carbon was fabricated.¹⁸⁶ Unlike the traditional perovskite deposition via OSPD or SDP, a reactive ink mixing carbon black and MAI in isopropanol was directly printed on the top of FTO/bl-TiO₂/PbI₂-based devices, leading to a quick chemical transformation in situ and significantly improving the carbon/MAPbI₃ interface. Also, another noteworthy feature is the precisely controlled pattern of the carbon electrodes. As a result, a PCE of 11.60% was achieved, with a J_sc of 17.2 mA cm⁻², V_oc of 0.95 V, and FF of 0.71.

6. Perovskite structure engineering

6.1 Bromine (Br)-based organic–inorganic halide perovskite

For methylammonium lead halide perovskite (MAPbX₃), the tuning of the halide in the X position from Cl to Br and I can effectively broaden the absorption spectrum, with a decreased E_g from 3.11 (ref. 187) to 2.3 (ref. 188) and 1.50 eV.¹² At present, two types of perovskites including the single halide perovskite (typically MAPbI₃ or MAPbBr₃) and the mixed halide perovskite (typically MAPbI_3−xCl_x, MAPbI_3−xBr_x or MAPbBr_3−xCl_x) have been applied in perovskite solar cells. Especially, devices with MAPbI₃ or MAPbI_3−xCl_x as the active layer have attracted significant attention due to their relatively low band gaps (∼1.5 eV) and large charge carrier diffusion lengths.^115,116 Excellent efficiencies (∼10%) were obtained in 2012 from the two initial reports on solid organometal trihalide perovskite solar cells based on MAPbI₃(ref. 12) or MAPbI_3−xCl_x.¹³ So far, via the optimization of device architecture and perovskite morphology, cells incorporating MAPbI₃ or MAPbI_3−xCl_x perovskite have yielded PCE of over 17% (ref. 21) and 19%.²³

While great progress has been achieved in MAPbI₃-based devices, MAPbBr₃-based devices have not exhibited satisfactory performance. Initial studies on porous TiO₂ or Al₂O₃-based devices using PDI, TPD, PC60BM or P3HT as the HTM demonstrated low PCEs of less than 1%,^105,153 which then improved to 6.7% efficiency in a FTO/bl-TiO₂/mp-TiO₂/MAPbBr₃/PIF8-TAA/Au based device (J_sc = 6.1 mA cm⁻², V_oc = 1.4 V, FF = 0.79).^153,158 The main reason for this low device performance can be attributed to the wider band gap of MAPbBr₃ (2.3 eV) than MAPbI₃ (1.5 eV), which limits the light harvesting and then decreases the short current. However, higher V_oc was always observed in MAPbBr₃-based devices compared to the referenced MAPbI₃-based devices (with the same device architecture), which is due to the lower valence band edge (−5.6 eV) and the higher conduction band edge (−3.4 eV) of MAPbBr₃ compared to that of MAPbI₃ (−5.4/−3.9), leading to a larger difference between the quasi-Fermi level of the electron and the quasi-Fermi level of the hole in MAPbBr₃-based devices. Later, by doping chloride ions in MAPbBr₃ films with CBP as the HTM, a V_oc as high as 1.5 V was obtained in a FTO/bl-TiO₂/mp-Al₂O₃/MAPbBr_3−xCl_x/CBP/Au-based device, with a J_sc of 4 mA cm⁻², a FF of 0.46 and a PCE of 2.7%.¹⁴⁹

In 2013, Seok and coworkers introduced band gap engineering by the chemical management of perovskites MAPb(I_1−xBr_x)₃ (0 ≤ x ≤ 1).¹⁸⁸ With the increase of Br concentration (x), the tetragonal phase (I4/mcm) of MAPbI₃ gradually transited to a cubic phase (Pm3̄m) due to the enhanced symmetry (x > 0.2), along with the decrease of the band gap, which was clearly demonstrated in the perovskite films where the color changed from dark brown to brown/red, then to yellow (see Fig. 17). In the J−V characteristics of the resultant devices with a FTO/bl-TiO₂/mp-TiO₂/MAPb(I_1−xBr_x)₃/PTAA/Au structure, with increasing x from 0 to 1, regularly decreased J_sc and increased V_oc were obtained, corresponding to the variation of absorption and energy levels. Interestingly, an increase in the fill factor from 0.66 to 0.74 was also measured, indicating the better charge transport properties in Br-doped cells. At last, an average of more than 10% with maximum PCEs of 12.3% were achieved when x varied in the range of 0 to 0.2, and a low sensitivity to humidity when x ≥ 0.2 was observed (the authors further achieved a higher PCE of up to 16.2% in later reports using solvent engineering in MAPb(I_1−xBr_x)₃ (x = 0.1–0.15)-based devices²⁴).

Fig. 17 (a) Crystal structures and unit lattice vectors on the (00l) plane of the tetragonal (I4/mcm) (top) and cubic (Pm3̄m) (bottom) phases are represented. (b) UV-vis absorption spectra of FTO/bl-TiO₂/mp-TiO₂/MAPb(I_1−xBr_x)₃/Au cells measured using an integral sphere. (c) Photographs of TiO₂/MAPb(I_1−xBr_x)₃ bilayer nanocomposites on FTO glass substrates. (d) A quadratic relationship of the band-gaps of MAPb(I_1−xBr_x)₃ as a function of Br composition (x). Reprinted with permission.¹⁸⁸

Yang and coworkers also fabricated a MAPbI₂Br-based mesoporous device by spin-coating a precursor solution of equimolar MABr and PbI₂ on the top of one-dimensional TiO₂ nanowire arrays, and achieved a PCE of 4.87%, with a J_sc of 10.12 mA cm⁻², V_oc of 0.82 V, and FF of 0.59.⁸³ In 2014, Zhu et al. prepared a high-quality MAPbI₂Br film on a planar device by thermal decomposition from a film deposited using a precursor containing PbI₂, MABr, and MACl.³⁷ The amount of Cl in the film decreased with annealing duration and no trace of Cl was observed at the latter stage of annealing.³⁶ The incorporation of MACl not only effectively adjusted the crystallization process for MAPbI₂Br but also enhanced the absorption of the resultant film. Finally, a compact packing of nanosheets completely covering the substrate was achieved and the device based on FTO/bl-TiO₂/MAPbI₂Br/spiro-MeOTAD/Ag showed a PCE of 10.03%, with a J_sc of 14.81 mA cm⁻², V_oc of 1.09 V, and FF of 0.62 under the simulated AM1.5G illumination (100 mW cm⁻²).

6.2 Formamidinium (FA)-based organic–inorganic halide perovskite (Table 8)

The first formamidinium (FA) lead trihalide-based perovskite solar cell was reported by Boix and coworkers in the end of 2013,¹⁸⁹ in which the FA cation (HC(NH₂)²⁺) was employed to substitute the conventional MA cation to synthesize FAPbI₃. The incorporation of the larger FA cation can effectively lower the band gap of the commonly used MAPbI₃ perovskite towards the optimum value of ∼1.4 eV (E_g = 1.47 eV), which is beneficial in the extended absorption of light and increasing the short-circuit current density. The primary device based on FTO/bl-TiO₂/mp-TiO₂/FAPbI₃/spiro-OMeTAD/Au yielded a PCE of 4.3% (J_sc = 6.45 mA cm⁻², V_oc = 0.97 V, FF = 0.687),¹⁸⁹ which was further improved to 7.5% (ref. 190) in a mesoscopic FTO/bl-TiO₂/mp-TiO₂/FAPbI₃/P3HT/Au device and 14.2% (ref. 191) in a planar FTO/bl-TiO₂/FAPbI₃/spiro-OMeTAD/Au device. Unlike MAPbI₃, the complete conversion of FAPbI₃via OSPD always needs a high temperature (>140 °C), which may result in a poor crystal quality FAPbI₃ perovskite layer. Thus a SDP method is suitable, as the film fabrication can be performed at a low temperature (∼100 °C). Snaith et al. also reported a method to prepare uniform and continuous films by adding a small amount of hydroiodic acid (HI) to the precursor solution (FAI : PbI₂ = 1 : 1).¹⁹¹ The authors thought that the presence of HI could help to solubilize the inorganic component, then slow down perovskite film crystallization, enabling a smoother film to be formed, without influencing the crystal structure. Another noteworthy feature is that FAPbI₃ has two polymorphs, including a black perovskite-type material (α-phase) with trigonal symmetry (P3m1), and a yellow hexagonal nonperovskite (γ-phase) counterpart (P63mc).¹⁹² In a humid atmosphere at room temperature, the black α-phase can quickly and fully convert to the yellow γ-phase. For a FAPbI₃-based perovskite solar cell, the presence of the yellow γ-phase is an adverse factor to the device performance; however, in most cases, the yellow γ-phase disappears at a higher annealing temperature (>100 °C). Later, considering the weak light absorption of FAPbI₃ at long wavelengths (>700 nm), Park and coworkers introduced a thin MAPbI₃ overlayer on top of FAPbI₃.¹⁹³ Enhanced IPCE at long wavelengths and a modified perovskite/HTM interface resulted in increased average photocurrent (∼2.8%) and photovoltage (∼1.9%) compared to the non-MAPbI₃ overlayer devices, yielding a high PCE of 16.01%, with an average PCE of 15.56%.

Grätzel and coworkers fabricated a mixed-cation perovskite (MA)_x(FA)_1−xPbI₃ (x = 0–1) as the active layer via SDP.¹⁹⁴ Devices based on FTO/bl-TiO₂/mp-TiO₂/MA_0.6FA_0.4PbI₃/spiro-OMeTAD/Au yielded the highest PCE of 14.9%, with an average PCE of 13.4%, which is higher than pure FAPbI₃- (11%) and MAPbI₃- (12.5%) based devices. The mixed-cation perovskite MA_0.6FA_0.4PbI₃ could avoid the formation of the yellow phase and, interestingly, exhibited the same band gap as FAPbI₃. Also, by optimizing the ratio of FA and MA cations, a PCE of 12.9% was achieved in a hole-conductor-free-based device with a FTO/bl-TiO₂/mp-TiO₂/ZrO₂/carbon CE/MA_0.4FA_0.6PbI₃ structure.¹⁸⁰ Besides the modification of the organic cation, Cui et al.¹⁹⁵ and Docampo et al.¹⁹⁶ reported two devices based on FAPbI_3−xCl_x (FTO/bl-TiO₂/mp-TiO₂/FAPbI_3−xCl_x/P3HT/Au) and FAPbBr₃ (FTO/bl-TiO₂/FAPbBr₃/spiro-OMeTAD/Au) perovskites, and achieved PCEs of 7.51% and 6.5%.

Table 8 A summary of published representative results of FA-based perovskite solar cell performance parameters with different device fabrication methods and configurations

PL	DP	Device structure	J sc (mA cm^−2)	V oc (V)	FF (%)	PCE (%)	Ref.
FAPbI3	SDP	FTO/bl-TiO2/mp-TiO2/PL/spiro/Au	6.45	0.97	68.7	4.3	189
FAPbI3	SDP	FTO/bl-TiO2/mp-TiO2/PL/P3HT/Au	18.3	0.84	50	7.5	190
FAPbI3	OSPD	FTO/bl-TiO2/PL/spiro/Au	23.3	0.94	65	14.2	191
FAPbI3	SDP	FTO/bl-TiO2/mp-TiO2/PL/MAPbI3/spiro/Au	20.97	1.032	74	16.01	193
MA0.6FA0.4PbI3	SDP	FTO/bl-TiO2/mp-TiO2/PL/spiro/Au	21.2	1.003	70	14.9	194
MA0.4FA0.6PbI3	SDP	FTO/bl-TiO2/mp-TiO2/ZrO2/carbon CE/PL	20.9	0.921	67	12.9	180
FAPbI3−xClx	OSPD	FTO/bl-TiO2/mp-TiO2/PL/P3HT/Au	19.24	0.73	54	7.51	195
FAPbBr3	SDP	FTO/bl-TiO2/PL/spiro/Au	6.6	1.35	73	6.5	196

---

6.3 Tin (Sn)-based organic–inorganic halide perovskite

Sn could be a potential alternative to Pb for organic–inorganic halide perovskites due to the reasons that both of them belong to the IVA group with similar ionic radii. Also, the low toxicity of Sn compared to Pb and the low optical band gap of MASnI₃ (ref. 92 and 197) (∼1.3 eV) compared to MAPbI₃ (∼1.55 eV) provide an opportunity to develop lead-free solar cells with high efficiencies. In 2014, Ogomi and coworkers first fabricated a series of mixed Pb/Sn alloy MASn_xPb_1−xI₃ perovskites in a FTO/bl-TiO₂/mp-TiO₂/MASn_xPb_1−xI₃/P3HT/Ag/Au-based device.¹⁹⁸ The optimal photovoltaic performance was observed when x was 0.5 (MASn_0.5Pb_0.5I₃), exhibiting a PCE of 4.18%, with a J_sc of 20.04 mA cm⁻², V_oc of 0.42 V, and FF of 0.50 at AM1.5G one sun illumination. The wide absorption area up to 1060 nm leads to this high J_sc while the low V_oc can be attributed to the poor perovskite morphology with flowerlike crystals and the obvious observation of the Sn²⁺ oxidation in the Sn perovskites.¹⁹² Later, Kanatzidis et al. also reported ∼7% efficiency, with a ∼20 mA cm⁻²J_sc in a FTO/bl-TiO₂/mp-TiO₂/MASn_0.5Pb_0.5I₃/spiro-OMeTAD/Au device (see Fig. 18).¹⁹⁹ Higher PCE of 10.1% was achieved by Jen and coworkers in a planar ITO/PEDOT:PSS/MAPb_0.85Sn_0.15I_3−xCl_x/PC60BM/C60-bis/Ag-based device, with a J_sc of 19.5 mA cm⁻², V_oc of 0.77 V, and FF of 0.67.²⁰⁰ Compared with the referential MAPbI_3−xCl_x film with a coverage of 87%, the MAPb_0.85Sn_0.15I_3−xCl_x film demonstrated an increased coverage (∼97%) and excellent continuity due to the Sn's effect on nucleation and growth, which is beneficial to suppress charge recombination and improve transport.

Fig. 18 Crystal structure (a–c) and X-ray diffraction pattern (d) of the MASn_1−xPb_xI₃ solid solutions. Simulated X-ray diffraction patterns of the two end compositions of MAPbI₃ and MASnI₃ are also shown in (d). Reprinted with permission.¹⁹⁹

While the good semiconducting behaviour has been observed in the Pb/Sn alloy perovskite-based device, completely lead-free organic–inorganic tin halide perovskites have not exhibited inspiring photovoltaic performance, and have always yielded a relatively low fill factor (<60%), which is mainly because of the more significant self-doping in the Sn perovskites. So far, devices based on the different lead-free perovskites MASnI₃ (ref. 199 and 201) (FTO/bl-TiO₂/mp-TiO₂/MASnI₃/spiro-OMeTAD/Au), MASnI_3−xBr_x(ref. 202) (FTO/bl-TiO₂/mp-TiO₂/MASnI_3−xBr_x/spiro-OMeTAD/Au), MASnI_3−xCl_x (ref. 200) (ITO/PEDOT:PSS/MASnI_3−xCl_x/PC60BM/C60-bis/Ag) have yielded PCEs of 6.4%, 5.73%, and 0.04%, respectively.

7. Conclusions and prospects

In this review, we summarize and discuss recent developments in organometal halide perovskite-based all-solid-state solar cells. With extensive research and accumulated understanding, the unique physical and optoelectronic properties of organometal halide perovskites have been explored and elucidated. The organometal halide perovskites can not only act as light absorbers but also work efficiently in various device architectures going from a dye sensitized concept due to the bipolar transport of both holes and electrons.

For the core part of ABX₃ perovskites (a) reasonable structure engineering based on A (e.g. MA and FA), B (e.g. Pb and Sn), and/or X (e.g. Cl, Br and I) can effectively modify the resultant optical band gaps, the HOMO/LUMO energy levels as well as the charge diffusion length, which finally affect the resultant photovoltaic performance. For example, long electron–hole diffusion lengths exceeding 1 μm for the mixed halide perovskite of MAPbI_3−xCl_x were measured, while it was only 100 nm for the triiodide perovskite of MAPbI₃; (b) the morphology and crystals in perovskite films are not only related to the nature of materials, but also can be optimized by carefully modifying the deposition methods (e.g. solution-based deposition and vapor-based deposition) and/or controlling the device fabrication conditions (e.g. thermal annealing, additive treatment and solvent engineering); (c) since in perovskite layers, as well as for most ionic crystals, the coordination number for ions at the crystal surfaces is always lower than in the bulk material, disorderness with respect to energy (i.e., charge traps and structural defects) always exists at the perovskite surface and/or grain boundaries; thus a passivation technique (e.g. IPFB, Lewis base and PbI₂) could be employed to lower carrier recombination rate and improve device performance.

For perovskite solar cells based on a porous structure (MMOPSC and MSSC), the porous scaffold can be conductive metal oxides including n-type TiO₂ (17.01%, ITO/bl-TiO₂/mp-TiO₂/MAPbI₃/spiro-OMeTAD/Au), ZnO (12%, FTO/bl-ZnO(Al)/mp-ZnO(Al)/MAPbI₃/HTM/Ag) and p-type NiO (11.6%, ITO/bl-NiO_x/mp-NiO/MAPbI₃/PC60BM/BCP/Al); or insulating materials including Al₂O₃ (15.9%, FTO/bl-TiO₂/mp-Al₂O₃/MAPbI_3−xCl_x/spiro-OMeTAD/Ag), ZrO₂ (10.8%, FTO/bl-TiO₂/mp-ZrO₂/MAPbI₃/spiro-OMeTAD/Ag) and SiO₂ (11.45%, FTO/bl-TiO₂/mp-SiO₂/MAPbI_3−xCl_x/spiro-OMeTAD/Au). The contents in parentheses are the highest efficiencies and the corresponding device architectures based on different porous structures reported so far. At present, bl-TiO₂/mp-TiO₂/perovskite/HTM-based MMOPSC and bl-TiO₂/mp-Al₂O₃/perovskite/HTM-based MSSC are the two most widely studied types of device architectures. Optimization of the following factors, (a) “scaffold” thickness and porosity modification, (b) interface/electrode engineering, (c) doping, and (d) HTM adjustment can effectively improve the device performance. Also, exploring on some new types of “scaffold” substances is still necessary (e.g. MO₃,²⁰³ Zn₂SnO₄,²⁰⁴ SrTiO₃ (ref. 205)). However, for fabricating these porous scaffolds as well as the thin blocking layer, especially employed in state-of-the-art perovskite solar cells, the process of sintering at high temperatures is always needed, which increases production cost and energy consumption, and also limits the possibility of fabricating large-area, flexible devices. Thus, in the future, developing low-temperature fabricated devices would be a promising direction and trend.

For PHJPSC, two configurations including a positive or inverted structure, in which a n-type (e.g. TiO₂ and ZnO) or p-type conductor (e.g. PEDOT:PSS, NiO, CuSCN, graphene oxide and polythiophene) was coated on a conductive glass, have been widely studied with the highest PCEs of 19.3% (ITO/PEIE/bl-TiO₂(Y)/MAPbI_3−xCl_x/spiro-OMeTAD/Au) and 16.31% (ITO/PEDOT:PSS/MAPbI₃/PC70BM/Ca/Al), respectively. Apart from the above-mentioned optimization methods of (b), (c) and (d) for MMOPSC and/or MSSC, the PHJ device performance is more dependent on the perovskite morphology, in particular the film thickness, roughness, and coverage. Also, due to the similar device architecture of the inverted planar perovskite solar cells (typically devices based on an ITO/PEDOT:PSS substrate) and the OPVs, some device engineering methods for OPVs can be used as a reference to this inverted perovskite device for further improving the device performance.

Although great success in the photovoltaic field has been achieved for organometal halide perovskites, the extremely high sensitivity of organometal halide perovskites to elevated temperature and moisture is still the main limiting factor for further practical application. Also, the low reproducibility, high deviation of the device performance and environmental pollution are problems exigent to be solved. Optimizing the perovskite material and structure of devices (e.g. large-area/flexible/durable/semitransparent/tandem device) will benefit its real application finally.^26,206–214 We believe that, in the near future, the development of perovskite solar cells will open a new chapter on solving the problem of energy crisis.

Nomenclature

3-D	Three-dimensional
ALD	Atomic layer deposition
AZO	Al-doped mesoporous ZnO
BCP	Bathocuproine
BHJ	Bulk-heterojunction
Bphen	Bathophenanthroline
C60	Fullerene
CBD	Chemical bath deposition
CE	Counter electrode
DIO	1,8-Diiodooctane
DMF	N,N-Dimethylformamide
DMSO	Dimethylsulfoxide
DP	Deposition process
DSSC	Dye sensitized solar cell
DSVD	Dual source vacuum deposition
ETM	Electron-transporting material
FA	Formamidinium
FF	Fill factor
FG	Flaky graphite
FK102	Tris(2-(1H-pyrazol-1-yl)pyridine)cobalt(III) tris(hexafluorophosphate)
FK209	Tris(2-(1H-pyrazol-1-yl)-4-tert-butylpyridine)cobalt(III) tris(bis(trifluoromethylsulfonyl)imide)
FK269	Bis(2,6-di(1H-pyrazol-1-yl)pyridine)cobalt(III) tris(bis(trifluoromethylsulfonyl)-imide)
FTO	Fluorine-doped tin oxide
GBL	γ-Butyrolactone
GO	Graphene oxide
HI	Hydroiodic acid
HMTA	Hexamethylenetetramine
HTM	Hole-transporting material
IC60BA	Indene–C60 bisadduct
IPCE	Incident photon-to-electron conversion efficiency
IPFB	Iodopentafluorobenzene
IQE	Internal quantum efficiency
ITO	Indium tin oxide
J sc	Short-circuit current density
LiTFSI	Bis(trifluoromethane)sulfonimide lithium salt
MMOPSC	Mesoporous metal oxide perovskite solar cell
MSSC	Meso-superstructured solar cell
NRA	Nanorod array
OPV	Organic photovoltaic cell
OSPD	One step precursor deposition
PA	Phenylamine
PC60BM	[6,6]-Phenyl C61-butyric acid methyl ester
PC70BM	[6,6]-Phenyl C71-butyric acid methyl ester
PCE	Power conversion efficiency
PDMS	Polydimethylsiloxane
PECVD	Plasma-enhanced chemical vapor deposition
PEIE	Polyethyleneimine ethoxylated
PET	Polyethylene terephthalate
PFI	Perfluorinated ionomer
PHJ	Planar heterojunction
PHJPSC	Planar heterojunction perovskite solar cell
PL	Perovskite layer
PMMA	Poly(methyl methacrylate)
Poly-TPD	Poly(N,N′-bis(4-butylphenyl)-N,N′-bis(phenyl)benzidine)
PT	Polythiophene
PTAA	Poly(triarylamine)
QDSC	Quantum dot solar cell
R s	Series resistance
R sh	Shunt resistance
SCD	Spray-coating deposition
SDP	Sequential deposition process
SG	Spheroidal graphite
Spiro-OMeTAD or spiro	2′,7,7′-Tetrakis(N,N-di-p-methoxyphenylamine)-9,9′-spirobifluorene
ss-DSSCs	Solid-state dye-sensitized solar cells
SVD	Sequential vapour deposition
SWNTs	Single-walled carbon nanotubes
TAA	Triarylamine
TBP	4-tert-Butylpyridine
TPA	Triphenylamine
TSSD	Two step spin-coating deposition
VASP	Vapor assisted solution process
V oc	Open-circuit voltage

Notes and references

1. http://news.sciencemag.org/breakthrough-of-the-year-2013 .
2. http://www.nature.com/news/365-days-nature-s-10-1.14367 .
3. P. V. Kamat, J. Am. Chem. Soc., 2014, 136, 3713.
4. Z. Cheng and J. Lin, CrystEngComm, 2010, 12, 2646.
5. C. Li, X. Lu, W. Ding, L. Feng, Y. Gao and Z. Guo, Acta Crystallogr., Sect. B: Struct. Sci., 2008, 64, 702.
6. M. A. Green, A. Ho-Baillie and H. J. Snaith, Nat. Photonics, 2014, 8, 506.
7. I. Borriello, G. Cantele and D. Ninno, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 77, 235214.
8. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737.
9. A. Yella, H. W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W. Diau, C. Y. Yeh, S. M. Zakeeruddin and M. Grätzel, Science, 2011, 334, 629.
10. U. Bach, D. Lupo, P. Comte, J. E. Moser, F. Weissörtel, J. Salbeck, H. Spreitzer and M. Grätzel, Nature, 1998, 395, 583.
11. J. Burschka, A. Dualeh, F. Kessler, E. Baranoff, N. L. Cevey-Ha, C. Yi, M. K. Nazeeruddin and M. Grätzel, J. Am. Chem. Soc., 2011, 133, 18042.
12. H. S. Kim, C. R. Lee, J. H. Im, K. B. Lee, T. Moehl, A. Marchioro, S. J. Moon, R. Humphry-Baker, J. H. Yum, J. E. Moser, M. Grätzel and N. G. Park, Sci. Rep., 2012, 2, 591.
13. M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami and H. J. Snaith, Science, 2012, 338, 643.
14. L. Etgar, P. Gao, Z. Xue, Q. Peng, A. K. Chandiran, B. Liu, M. K. Nazeeruddin and M. Grätzel, J. Am. Chem. Soc., 2012, 134, 17396.
15. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050.
16. J. H. Im, C. R. Lee, J. W. Lee, S. W. Park and N. G. Park, Nanoscale, 2011, 3, 4088.
17. J. H. Im, J. Chung, S. J. Kim and N. G. Park, Nanoscale Res. Lett., 2012, 7, 353.
18. J. Burschka, N. Pellet, S. J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin and M. Grätzel, Nature, 2013, 499, 316.
19. M. Liu, M. B. Johnston and H. J. Snaith, Nature, 2013, 501, 395.
20. J. H. Heo, S. H. Im, J. H. Noh, T. N. Mandal, C.-S. Lim, J. A. Chang, Y. H. Lee, H.-j. Kim, A. Sarkar, M. K. Nazeeruddin, M. Grätzel and S. I. Seok, Nat. Photonics, 2013, 7, 486.
21. J. H. Im, I. H. Jang, N. Pellet, M. Grätzel and N. G. Park, Nat. Nanotechnol., 2014, 9, 927.
22. K. Wojciechowski, M. Saliba, T. Leijtens, A. Abate and H. J. Snaith, Energy Environ. Sci., 2014, 7, 1142.
23. H. Zhou, Q. Chen, G. Li, S. Luo, T. B. Song, H. S. Duan, Z. Hong, J. You, Y. Liu and Y. Yang, Science, 2014, 345, 542.
24. N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. Ryu and S. I. Seok, Nat. Mater., 2014, 13, 897.
25. H. B. Kim, H. Choi, J. Jeong, S. Kim, B. Walker, S. Song and J. Y. Kim, Nanoscale, 2014, 6, 6679.
26. J. Seo, S. Park, Y. Chan Kim, N. J. Jeon, J. H. Noh, S. C. Yoon and S. I. Seok, Energy Environ. Sci., 2014, 7, 2642.
27. M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A. Gray-Weale, U. Bach, Y. B. Cheng and L. Spiccia, Angew. Chem., Int. Ed., 2014, 53, 9898.
28. S. Paek, N. Cho, H. Choi, H. Jeong, J. S. Lim, J.-Y. Hwang, J. K. Lee and J. Ko, J. Phys. Chem. C, 2014, 118, 25899.
29. J. W. Jung, S. T. Williams and A. K. Y. Jen, RSC Adv., 2014, 4, 62971.
30. R. Kang, J.-E. Kim, J.-S. Yeo, S. Lee, Y.-J. Jeon and D.-Y. Kim, J. Phys. Chem. C, 2014, 118, 26513.
31. G. E. Eperon, V. M. Burlakov, P. Docampo, A. Goriely and H. J. Snaith, Adv. Funct. Mater., 2014, 24, 151.
32. H. L. Hsu, C. P. Chen, J. Y. Chang, Y. Y. Yu and Y. K. Shen, Nanoscale, 2014, 6, 10281.
33. M. Saliba, K. W. Tan, H. Sai, D. T. Moore, T. Scott, W. Zhang, L. A. Estroff, U. Wiesner and H. J. Snaith, J. Phys. Chem. C, 2014, 118, 17171.
34. A. Dualeh, N. Tétreault, T. Moehl, P. Gao, M. K. Nazeeruddin and M. Grätzel, Adv. Funct. Mater., 2014, 24, 3250.
35. P. W. Liang, C. Y. Liao, C. C. Chueh, F. Zuo, S. T. Williams, X. K. Xin, J. Lin and A. K. Jen, Adv. Mater., 2014, 26, 3748.
36. Y. Zhao and K. Zhu, J. Phys. Chem. C, 2014, 118, 9412.
37. Y. Zhao and K. Zhu, J. Am. Chem. Soc., 2014, 136, 12241.
38. C. Zuo and L. Ding, Nanoscale, 2014, 6, 9935.
39. D. B. M. Kangning Liang and T. P. Michael, Chem. Mater., 1998, 10, 403.
40. Y. Ma, L. Zheng, Y.-H. Chung, S. Chu, L. Xiao, Z. Chen, S. Wang, B. Qu, Q. Gong, Z. Wu and X. Hou, Chem. Commun., 2014, 50, 12458.
41. Y. Wu, A. Islam, X. Yang, C. Qin, J. Liu, K. Zhang, W. Peng and L. Han, Energy Environ. Sci., 2014, 7, 2934.
42. J. Shi, Y. Luo, H. Wei, J. Luo, J. Dong, S. Lv, J. Xiao, Y. Xu, L. Zhu, X. Xu, H. Wu, D. Li and Q. Meng, ACS Appl. Mater. Interfaces, 2014, 6, 9711.
43. P. Docampo, F. C. Hanusch, N. Giesbrecht, P. Angloher, A. Ivanova and T. Bein, APL Mater., 2014, 2, 081508.
44. L. Zheng, Y. Ma, S. Chu, S. Wang, B. Qu, L. Xiao, Z. Chen, Q. Gong, Z. Wu and X. Hou, Nanoscale, 2014, 6, 8171.
45. D. H. Cao, C. C. Stoumpos, C. D. Malliakas, M. J. Katz, O. K. Farha, J. T. Hupp and M. G. Kanatzidis, APL Mater., 2014, 2, 091101.
46. D. Bi, S.-J. Moon, L. Häggman, G. Boschloo, L. Yang, E. M. J. Johansson, M. K. Nazeeruddin, M. Grätzel and A. Hagfeldt, RSC Adv., 2013, 3, 18762.
47. D. Liu and T. L. Kelly, Nat. Photonics, 2013, 8, 133.
48. Q. Chen, H. Zhou, Z. Hong, S. Luo, H. S. Duan, H. H. Wang, Y. Liu, G. Li and Y. Yang, J. Am. Chem. Soc., 2014, 136, 622.
49. Z. Xiao, C. Bi, Y. Shao, Q. Dong, Q. Wang, Y. Yuan, C. Wang, Y. Gao and J. Huang, Energy Environ. Sci., 2014, 7, 2619.
50. C.-H. Chiang, Z.-L. Tseng and C.-G. Wu, J. Mater. Chem. A, 2014, 2, 15897.
51. J.-H. Im, H.-S. Kim and N.-G. Park, APL Mater., 2014, 2, 081510.
52. A. T. Barrows, A. J. Pearson, C. K. Kwak, A. D. F. Dunbar, A. R. Buckley and D. G. Lidzey, Energy Environ. Sci., 2014, 7, 2944.
53. D. T. Moore, H. Sai, K. Wee Tan, L. A. Estroff and U. Wiesner, APL Mater., 2014, 2, 081802.
54. B.-E. Cohen, S. Gamliel and L. Etgar, APL Mater., 2014, 2, 081502.
55. H. Hu, D. Wang, Y. Zhou, J. Zhang, S. Lv, S. Pang, X. Chen, Z. Liu, N. P. Padture and G. Cui, RSC Adv., 2014, 4, 28964.
56. C. W. Chen, H. W. Kang, S. Y. Hsiao, P. F. Yang, K. M. Chiang and H. W. Lin, Adv. Mater., 2014, 26, 6647.
57. Q. Chen, H. Zhou, T. B. Song, S. Luo, Z. Hong, H. S. Duan, L. Dou, Y. Liu and Y. Yang, Nano Lett., 2014, 14, 4158.
58. T. Supasai, N. Rujisamphan, K. Ullrich, A. Chemseddine and T. Dittrich, Appl. Phys. Lett., 2013, 103, 183906.
59. L. Wang, C. McCleese, A. Kovalsky, Y. Zhao and C. Burda, J. Am. Chem. Soc., 2014, 136, 12205.
60. N. J. Jeon, H. G. Lee, Y. C. Kim, J. Seo, J. H. Noh, J. Lee and S. I. Seok, J. Am. Chem. Soc., 2014, 136, 7837.
61. Y. Bai, I. Mora-Sero, F. De Angelis, J. Bisquert and P. Wang, Chem. Rev., 2014, 114, 10095.
62. G. Murugadoss, G. Mizuta, S. Tanaka, H. Nishino, T. Umeyama, H. Imahori and S. Ito, APL Mater., 2014, 2, 081511.
63. B. Conings, L. Baeten, T. Jacobs, R. Dera, J. D'Haen, J. Manca and H. G. Boyen, APL Mater., 2014, 2, 081505.
64. B. Conings, L. Baeten, C. De Dobbelaere, J. D'Haen, J. Manca and H. G. Boyen, Adv. Mater., 2014, 26, 2041.
65. W. Liu and Y. Zhang, J. Mater. Chem. A, 2014, 2, 10244.
66. A. Yella, L. P. Heiniger, P. Gao, M. K. Nazeeruddin and M. Grätzel, Nano Lett., 2014, 14, 2591.
67. J. M. Ball, M. M. Lee, A. Hey and H. J. Snaith, Energy Environ. Sci., 2013, 6, 1739.
68. Y. Wu, X. Yang, H. Chen, K. Zhang, C. Qin, J. Liu, W. Peng, A. Islam, E. Bi, F. Ye, M. Yin, P. Zhang and L. Han, Appl. Phys. Express, 2014, 7, 052301.
69. A. K. Chandiran, A. Yella, M. T. Mayer, P. Gao, M. K. Nazeeruddin and M. Grätzel, Adv. Mater., 2014, 26, 4309.
70. P. Qin, A. L. Domanski, A. K. Chandiran, R. Berger, H. J. Butt, M. I. Dar, T. Moehl, N. Tetreault, P. Gao, S. Ahmad, M. K. Nazeeruddin and M. Grätzel, Nanoscale, 2014, 6, 1508.
71. J. Xia, N. Masaki, K. Jiang and S. Yanagida, J. Phys. Chem. B, 2006, 110, 25222.
72. W. Ke, G. Fang, J. Wang, P. Qin, H. Tao, H. Lei, Q. Liu, X. Dai and X. Zhao, ACS Appl. Mater. Interfaces, 2014, 6, 15959.
73. J. J. Choi, X. Yang, Z. M. Norman, S. J. Billinge and J. S. Owen, Nano Lett., 2014, 14, 127.
74. E. Edri, S. Kirmayer, A. Henning, S. Mukhopadhyay, K. Gartsman, Y. Rosenwaks, G. Hodes and D. Cahen, Nano Lett., 2014, 14, 1000.
75. Q. Shen, Y. Ogomi, J. Chang, S. Tsukamoto, K. Kukihara, T. Oshima, N. Osada, K. Yoshino, K. Katayama, T. Toyoda and S. Hayase, Phys. Chem. Chem. Phys., 2014, 16, 19984.
76. Y. Zhao, A. M. Nardes and K. Zhu, Faraday Discuss., 2014 10.1039/c4fd00128a.
77. R. Lindblad, D. Bi, B.-w. Park, J. Oscarsson, M. Gorgoi, H. Siegbahn, M. Odelius, E. M. J. Johansson and H. Rensmo, J. Phys. Chem. Lett., 2014, 5, 648.
78. J. W. Lee, T. Y. Lee, P. J. Yoo, M. Grätzel, S. Mhaisalkar and N. G. Park, J. Mater. Chem. A, 2014, 2, 9251.
79. M. I. Dar, F. J. Ramos, Z. Xue, B. Liu, S. Ahmad, S. A. Shivashankar, M. K. Nazeeruddin and M. Grätzel, Chem. Mater., 2014, 26, 4675.
80. Y. Rong, Z. Ku, A. Mei, T. Liu, M. Xu, S. Ko, X. Li and H. Han, J. Phys. Chem. Lett., 2014, 5, 2160.
81. H. S. Kim, J. W. Lee, N. Yantara, P. P. Boix, S. A. Kulkarni, S. Mhaisalkar, M. Grätzel and N. G. Park, Nano Lett., 2013, 13, 2412.
82. K. Manseki, T. Ikeya, A. Tamura, T. Ban, T. Sugiura and T. Yoshida, RSC Adv., 2014, 4, 9652.
83. J. Qiu, Y. Qiu, K. Yan, M. Zhong, C. Mu, H. Yan and S. Yang, Nanoscale, 2013, 5, 3245.
84. H. Chen, Z. Wei, K. Yan, Y. Yi, J. Wang and S. Yang, Faraday Discuss., 2014 10.1039/c4fd00155a.
85. M. Yang, R. Guo, K. Kadel, Y. Liu, K. O'Shea, R. Bone, X. Wang, J. He and W. Li, J. Mater. Chem. A, 2014, 2, 19616.
86. X. Gao, J. Li, J. Baker, Y. Hou, D. Guan, J. Chen and C. Yuan, Chem. Commun., 2014, 50, 6368.
87. Y. M. Xiao, G. Y. Han, Y. P. Li, M. Y. Li and J. H. Wu, J. Mater. Chem. A, 2014, 2, 16856.
88. S. Dharani, H. K. Mulmudi, N. Yantara, P. T. Thu Trang, N. G. Park, M. Graetzel, S. Mhaisalkar, N. Mathews and P. P. Boix, Nanoscale, 2014, 6, 1675.
89. Q. Jiang, X. Sheng, Y. Li, X. Feng and T. Xu, Chem. Commun., 2014, 50, 14720.
90. E. J. Crossland, N. Noel, V. Sivaram, T. Leijtens, J. A. Alexander-Webber and H. J. Snaith, Nature, 2013, 495, 215.
91. G. S. Han, S. Lee, J. H. Noh, H. S. Chung, J. H. Park, B. S. Swain, J. H. Im, N. G. Park and H. S. Jung, Nanoscale, 2014, 6, 6127.
92. Q. Zhang, C. S. Dandeneau, X. Zhou and G. Cao, Adv. Mater., 2009, 21, 4087.
93. J. A. Anta, E. Guillén and R. Tena-Zaera, J. Phys. Chem. C, 2012, 116, 11413.
94. D. Bi, G. Boschloo, S. Schwarzmuller, L. Yang, E. M. Johansson and A. Hagfeldt, Nanoscale, 2013, 5, 11686.
95. D. Y. Son, J. H. Im, H. S. Kim and N. G. Park, J. Phys. Chem. C, 2014, 118, 16567.
96. M. H. Kumar, N. Yantara, S. Dharani, M. Graetzel, S. Mhaisalkar, P. P. Boix and N. Mathews, Chem. Commun., 2013, 49, 11089.
97. F. J. Ramos, M. C. Lopez-Santos, E. Guillen, M. K. Nazeeruddin, M. Grätzel, A. R. Gonzalez-Elipe and S. Ahmad, ChemPhysChem, 2014, 15, 1148.
98. K. Mahmood, B. S. Swain and H. S. Jung, Nanoscale, 2014, 6, 9127.
99. K. Mahmood, R. Munir, B. S. Swain, G.-S. Han, B.-J. Kim and H. S. Jung, RSC Adv., 2014, 4, 9072.
100. J. Dong, Y. Zhao, J. Shi, H. Wei, J. Xiao, X. Xu, J. Luo, J. Xu, D. Li, Y. Luo and Q. Meng, Chem. Commun., 2014, 50, 13381.
101. K. C. Wang, J. Y. Jeng, P. S. Shen, Y. C. Chang, E. W. G. Diau, C. H. Tsai, T. Y. Chao, H. C. Hsu, P. Y. Lin, P. Chen, T. F. Guo and T. C. Wen, Sci. Rep., 2014, 4, 4756.
102. J.-Y. Jeng, K.-C. Chen, T.-Y. Chiang, P.-Y. Lin, T.-D. Tsai, Y.-C. Chang, T.-F. Guo, P. Chen, T.-C. Wen and Y.-J. Hsu, Adv. Mater., 2014, 26, 4107.
103. L. Hu, J. Peng, W. Wang, Z. Xia, J. Yuan, J. Lu, X. Huang, W. Ma, H. Song, W. Chen, Y.-B. Cheng and J. Tang, ACS Photonics, 2014, 1, 547.
104. K. C. Wang, P. S. Shen, M. H. Li, S. Chen, M. W. Lin, P. Chen and T. F. Guo, ACS Appl. Mater. Interfaces, 2014, 6, 11851.
105. E. Edri, S. Kirmayer, D. Cahen and G. Hodes, J. Phys. Chem. Lett., 2013, 4, 897.
106. W. Zhang, M. Saliba, S. D. Stranks, Y. Sun, X. Shi, U. Wiesner and H. J. Snaith, Nano Lett., 2013, 13, 4505.
107. M. J. Carnie, C. Charbonneau, M. L. Davies, J. Troughton, T. M. Watson, K. Wojciechowski, H. Snaith and D. A. Worsley, Chem. Commun., 2013, 49, 7893.
108. A. Abate, M. Saliba, D. J. Hollman, S. D. Stranks, K. Wojciechowski, R. Avolio, G. Grancini, A. Petrozza and H. J. Snaith, Nano Lett., 2014, 14, 3247.
109. S. K. Pathak, A. Abate, P. Ruckdeschel, B. Roose, K. C. Godel, Y. Vaynzof, A. Santhala, S. I. Watanabe, D. J. Hollman, N. Noel, A. Sepe, U. Wiesner, R. Friend, H. J. Snaith and U. Steiner, Adv. Funct. Mater., 2014, 24, 6046.
110. S. N. Habisreutinger, T. Leijtens, G. E. Eperon, S. D. Stranks, R. J. Nicholas and H. J. Snaith, Nano Lett., 2014, 14, 5561.
111. J. T. Wang, J. M. Ball, E. M. Barea, A. Abate, J. A. Alexander-Webber, J. Huang, M. Saliba, I. Mora-Sero, J. Bisquert, H. J. Snaith and R. J. Nicholas, Nano Lett., 2014, 14, 724.
112. X. Dong, H. Hu, B. Lin, J. Ding and N. Yuan, Chem. Commun., 2014, 50, 14405.
113. H. S. Kim, I. Mora-Sero, V. Gonzalez-Pedro, F. Fabregat-Santiago, E. J. Juarez-Perez, N. G. Park and J. Bisquert, Nat. Commun., 2013, 4, 2242.
114. S. H. Hwang, J. Roh, J. Lee, J. Ryu, J. Yun and J. Jang, J. Mater. Chem. A, 2014, 2, 16429.
115. S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou, M. J. Alcocer, T. Leijtens, L. M. Herz, A. Petrozza and H. J. Snaith, Science, 2013, 342, 341.
116. G. Xing, N. Mathews, S. Sun, S. S. Lim, Y. M. Lam, M. Grätzel, S. Mhaisalkar and T. C. Sum, Science, 2013, 342, 344.
117. P. Docampo, F. C. Hanusch, S. D. Stranks, M. Döblinger, J. M. Feckl, M. Ehrensperger, N. K. Minar, M. B. Johnston, H. J. Snaith and T. Bein, Adv. Energy Mater., 2014, 4 DOI:10.1002/aenm.201400355.
118. J. Kim, G. Kim, T. K. Kim, S. Kwon, H. Back, J. Lee, S. H. Lee, H. Kang and K. Lee, J. Mater. Chem. A, 2014, 2, 17291.
119. L. Liang, Z. Huang, L. Cai, W. Chen, B. Wang, K. Chen, H. Bai, Q. Tian and B. Fan, ACS Appl. Mater. Interfaces, 2014, 6, 20585.
120. L. Zheng, Y. H. Chung, Y. Ma, L. Zhang, L. Xiao, Z. Chen, S. Wang, B. Qu and Q. Gong, Chem. Commun., 2014, 50, 11196.
121. S. Chavhan, O. Miguel, H. J. Grande, V. Gonzalez-Pedro, R. S. Sanchez, E. M. Barea, I. Mora-Sero and R. Tena-Zaera, J. Mater. Chem. A, 2014, 2, 12754.
122. N. K. Noel, A. Abate, S. D. Stranks, E. S. Parrott, V. M. Burlakov, A. Goriely and H. J. Snaith, ACS Nano, 2014, 8, 9815.
123. J. Y. Jeng, Y. F. Chiang, M. H. Lee, S. R. Peng, T. F. Guo, P. Chen and T. C. Wen, Adv. Mater., 2013, 25, 3727.
124. S. Sun, T. Salim, N. Mathews, M. Duchamp, C. Boothroyd, G. Xing, T. C. Sum and Y. M. Lam, Energy Environ. Sci., 2014, 7, 399.
125. P. Docampo, J. M. Ball, M. Darwich, G. E. Eperon and H. J. Snaith, Nat. Commun., 2013, 4, 2761.
126. O. Malinkiewicz, A. Yella, Y. H. Lee, G. M. Espallargas, M. Graetzel, M. K. Nazeeruddin and H. J. Bolink, Nat. Photonics, 2013, 8, 128.
127. O. Malinkiewicz, C. Roldán-Carmona, A. Soriano, E. Bandiello, L. Camacho, M. K. Nazeeruddin and H. J. Bolink, Adv. Energy Mater., 2014, 4 DOI:10.1002/aenm.201400345.
128. J. You, Z. Hong, Y. M. Yang, Q. Chen, M. Cai, T. B. Song, C. C. Chen, S. Lu, Y. Liu, H. Zhou and Y. Yang, ACS Nano, 2014, 8, 1674.
129. K.-G. Lim, H.-B. Kim, J. Jeong, H. Kim, J. Y. Kim and T.-W. Lee, Adv. Mater., 2014, 26, 6461.
130. Q. Wang, Y. C. Shao, Q. F. Dong, Z. G. Xiao, Y. B. Yuan and J. S. Huang, Energy Environ. Sci., 2014, 7, 2359.
131. Z. Xiao, Q. Dong, C. Bi, Y. Shao, Y. Yuan and J. Huang, Adv. Mater., 2014, 26, 6503.
132. A. S. Subbiah, A. Halder, S. Ghosh, N. Mahuli, G. Hodes and S. K. Sarkar, J. Phys. Chem. Lett., 2014, 5, 1748.
133. Z. Zhu, Y. Bai, T. Zhang, Z. Liu, X. Long, Z. Wei, Z. Wang, L. Zhang, J. Wang, F. Yan and S. Yang, Angew. Chem., Int. Ed., 2014, 53, 12571.
134. Z. Wu, S. Bai, J. Xiang, Z. Yuan, Y. Yang, W. Cui, X. Gao, Z. Liu, Y. Jin and B. Sun, Nanoscale, 2014, 6, 10505.
135. W. Yan, Y. Li, W. Sun, H. Peng, S. Ye, Z. Liu, Z. Bian and C. Huang, RSC Adv., 2014, 4, 33039.
136. D. Bi, L. Yang, G. Boschloo, A. Hagfeldt and E. M. J. Johansson, J. Phys. Chem. Lett., 2013, 4, 1532.
137. N. J. Jeon, J. Lee, J. H. Noh, M. K. Nazeeruddin, M. Grätzel and S. I. Seok, J. Am. Chem. Soc., 2013, 135, 19087.
138. H. Li, K. Fu, A. Hagfeldt, M. Grätzel, S. G. Mhaisalkar and A. C. Grimsdale, Angew. Chem., Int. Ed., 2014, 53, 4085.
139. T. Krishnamoorthy, F. Kunwu, P. P. Boix, H. Li, T. M. Koh, W. L. Leong, S. Powar, A. Grimsdale, M. Grätzel, N. Mathews and S. G. Mhaisalkar, J. Mater. Chem. A, 2014, 2, 6305.
140. A. Krishna, D. Sabba, H. R. Li, J. Yin, P. P. Boix, C. Soci, S. G. Mhaisalkar and A. C. Grimsdale, Chem. Sci., 2014, 5, 2702.
141. H. Choi, S. Paek, N. Lim, Y. H. Lee, M. K. Nazeeruddin and J. Ko, Chem.–Eur. J., 2014, 20, 10894.
142. Y. Song, S. Lv, X. Liu, X. Li, S. Wang, H. Wei, D. Li, Y. Xiao and Q. Meng, Chem. Commun., 2014, 50, 15239.
143. B. Xu, E. Sheibani, P. Liu, J. Zhang, H. Tian, N. Vlachopoulos, G. Boschloo, L. Kloo, A. Hagfeldt and L. Sun, Adv. Mater., 2014, 26, 6629.
144. S. D. Sung, M. S. Kang, I. T. Choi, H. M. Kim, H. Kim, M. Hong, H. K. Kim and W. I. Lee, Chem. Commun., 2014, 50, 14161.
145. J. Wang, S. Wang, X. Li, L. Zhu, Q. Meng, Y. Xiao and D. Li, Chem. Commun., 2014, 50, 5829.
146. S. Lv, L. Han, J. Xiao, L. Zhu, J. Shi, H. Wei, Y. Xu, J. Dong, X. Xu, D. Li, S. Wang, Y. Luo, Q. Meng and X. Li, Chem. Commun., 2014, 50, 6931.
147. J. Xiao, L. Han, L. Zhu, S. Lv, J. Shi, H. Wei, Y. Xu, J. Dong, X. Xu, Y. Xiao, D. Li, S. Wang, Y. Luo, X. Li and Q. Meng, RSC Adv., 2014, 4, 32918.
148. L. E. Polander, P. Pahner, M. Schwarze, M. Saalfrank, C. Koerner and K. Leo, APL Mater., 2014, 2, 081503.
149. E. Edri, S. Kirmayer, M. Kulbak, G. Hodes and D. Cahen, J. Phys. Chem. Lett., 2014, 5, 429.
150. P. Qin, H. Kast, M. K. Nazeeruddin, S. M. Zakeeruddin, A. Mishra, P. Bäuerle and M. Grätzel, Energy Environ. Sci., 2014, 7, 2981.
151. P. Qin, S. Paek, M. I. Dar, N. Pellet, J. Ko, M. Grätzel and M. K. Nazeeruddin, J. Am. Chem. Soc., 2014, 136, 8516.
152. J. Liu, Y. Wu, C. Qin, X. Yang, T. Yasuda, A. Islam, K. Zhang, W. Peng, W. Chen and L. Han, Energy Environ. Sci., 2014, 7, 2963.
153. B. Cai, Y. Xing, Z. Yang, W.-H. Zhang and J. Qiu, Energy Environ. Sci., 2013, 6, 1480.
154. F. Di Giacomo, S. Razza, F. Matteocci, A. D'Epifanio, S. Licoccia, T. M. Brown and A. Di Carlo, J. Power Sources, 2014, 251, 152.
155. Y. Guo, C. Liu, K. Lnoue, K. Harano, H. Tanaka and E. Nakamura, J. Mater. Chem. A, 2014, 2, 13827.
156. J. H. Heo and S. H. Im, Phys. Status Solidi RRL, 2014, 8, 816.
157. M. Zhang, M. Lyu, H. Yu, J.-H. Yun, Q. Wang and L. Wang, Chem.–Eur. J., 2014, 21, 434.
158. S. Ryu, J. H. Noh, N. J. Jeon, Y. C. Kim, S. Yang, J. W. Seo and S. I. Seok, Energy Environ. Sci., 2014, 7, 2614.
159. Z. Zhu, Y. Bai, H. K. H. Lee, C. Mu, T. Zhang, L. Zhang, J. Wang, H. Yan, S. K. So and S. Yang, Adv. Funct. Mater., 2014, 24, 7357.
160. P. Qin, N. Tetreault, M. I. Dar, P. Gao, K. L. McCall, S. R. Rutter, S. D. Ogier, N. D. Forrest, J. S. Bissett, M. J. Simms, A. J. Page, R. Fisher, M. Grätzel and M. K. Nazeeruddin, Adv. Energy Mater., 2014 DOI:10.1002/aenm.201400980.
161. Y. S. Kwon, J. Lim, H.-J. Yun, Y.-H. Kim and T. Park, Energy Environ. Sci., 2014, 7, 1454.
162. J. W. Lee, S. Park, M. J. Ko, H. J. Son and N. G. Park, ChemPhysChem, 2014, 15, 2595.
163. A. Abrusci, S. D. Stranks, P. Docampo, H. L. Yip, A. K. Jen and H. J. Snaith, Nano Lett., 2013, 13, 3124.
164. J. A. Christians, R. C. Fung and P. V. Kamat, J. Am. Chem. Soc., 2014, 136, 758.
165. S. Ito, S. Tanaka, K. Manabe and H. Nishino, J. Phys. Chem. C, 2014, 118, 16995.
166. S. Ito, S. Tanaka, H. Vahlman, H. Nishino, K. Manabe and P. Lund, ChemPhysChem, 2014, 15, 1194.
167. P. Qin, S. Tanaka, S. Ito, N. Tetreault, K. Manabe, H. Nishino, M. K. Nazeeruddin and M. Grätzel, Nat. Commun., 2014, 5, 3834.
168. M. D. Irwin, D. B. Buchholz, A. W. Hains, R. P. H. Chang and T. J. Marks, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 2783.
169. K. X. Steirer, P. F. Ndione, N. E. Widjonarko, M. T. Lloyd, J. Meyer, E. L. Ratcliff, A. Kahn, N. R. Armstrong, C. J. Curtis, D. S. Ginley, J. J. Berry and D. C. Olson, Adv. Energy Mater., 2011, 1, 813.
170. W. A. Laban and L. Etgar, Energy Environ. Sci., 2013, 6, 3249.
171. S. Aharon, S. Gamliel, B. E. Cohen and L. Etgar, Phys. Chem. Chem. Phys., 2014, 16, 10512.
172. S. Aharon, B. El Cohen and L. Etgar, J. Phys. Chem. C, 2014, 118, 17160.
173. J. J. Shi, W. Dong, Y. Z. Xu, C. H. Li, S. T. Lv, L. F. Zhu, J. Dong, Y. H. Luo, D. M. Li, Q. B. Meng and Q. Chen, Chin. Phys. Lett., 2013, 30, 128402.
174. Y. Xu, J. Shi, S. Lv, L. Zhu, J. Dong, H. Wu, Y. Xiao, Y. Luo, S. Wang, D. Li, X. Li and Q. Meng, ACS Appl. Mater. Interfaces, 2014, 6, 5651.
175. J. Shi, J. Dong, S. Lv, Y. Xu, L. Zhu, J. Xiao, X. Xu, H. Wu, D. Li, Y. Luo and Q. Meng, Appl. Phys. Lett., 2014, 104, 063901.
176. Y. M. Xiao, G. Y. Han, Y. P. Li, M. Y. Li, Y. Z. Chang and J. H. Wu, J. Mater. Chem. A, 2014, 2, 16531.
177. F. Hao, C. C. Stoumpos, Z. Liu, R. P. Chang and M. G. Kanatzidis, J. Am. Chem. Soc., 2014, 136, 16411.
178. Z. Ku, Y. Rong, M. Xu, T. Liu and H. Han, Sci. Rep., 2013, 3, 3132.
179. M. Xu, Y. Rong, Z. Ku, A. Mei, T. Liu, L. Zhang, X. Li and H. Han, J. Mater. Chem. A, 2014, 2, 8607.
180. M. Hu, L. Liu, A. Mei, Y. Yang, T. Liu and H. Han, J. Mater. Chem. A, 2014, 2, 17115.
181. A. Mei, X. Li, L. Liu, Z. Ku, T. Liu, Y. Rong, M. Xu, M. Hu, J. Chen, Y. Yang, M. Grätzel and H. Han, Science, 2014, 345, 295.
182. Z. Liu, M. Zhang, X. Xu, L. Bu, W. Zhang, W. Li, Z. Zhao, M. Wang, Y. B. Cheng and H. He, Dalton Trans., 2014 10.1039/c4dt02904f.
183. F. Zhang, X. Yang, H. Wang, M. Cheng, J. Zhao and L. Sun, ACS Appl. Mater. Interfaces, 2014, 6, 16140.
184. H. Zhou, Y. Shi, Q. Dong, H. Zhang, Y. Xing, K. Wang, Y. Du and T. Ma, J. Phys. Chem. Lett., 2014, 5, 3241.
185. Y. Yang, J. Xiao, H. Wei, L. Zhu, D. Li, Y. Luo, H. Wu and Q. Meng, RSC Adv., 2014, 4, 52825.
186. Z. Wei, H. Chen, K. Yan and S. Yang, Angew. Chem., Int. Ed., 2014, 53, 13239.
187. N. Kitazawa, Y. Watanabe and Y. Nakamura, J. Mater. Sci., 2002, 37, 3585.
188. J. H. Noh, S. H. Im, J. H. Heo, T. N. Mandal and S. I. Seok, Nano Lett., 2013, 13, 1764.
189. T. M. Koh, K. Fu, Y. Fang, S. Chen, T. C. Sum, N. Mathews, S. G. Mhaisalkar, P. P. Boix and T. Baikie, J. Phys. Chem. C, 2014, 118, 16458.
190. S. Pang, H. Hu, J. Zhang, S. Lv, Y. Yu, F. Wei, T. Qin, H. Xu, Z. Liu and G. Cui, Chem. Mater., 2014, 26, 1485.
191. G. E. Eperon, S. D. Stranks, C. Menelaou, M. B. Johnston, L. M. Herz and H. J. Snaith, Energy Environ. Sci., 2014, 7, 982.
192. C. C. Stoumpos, C. D. Malliakas and M. G. Kanatzidis, Inorg. Chem., 2013, 52, 9019.
193. J. W. Lee, D. J. Seol, A. N. Cho and N. G. Park, Adv. Mater., 2014, 26, 4991.
194. N. Pellet, P. Gao, G. Gregori, T. Y. Yang, M. K. Nazeeruddin, J. Maier and M. Grätzel, Angew. Chem., Int. Ed., 2014, 53, 3151.
195. S. Lv, S. Pang, Y. Zhou, N. P. Padture, H. Hu, L. Wang, X. Zhou, H. Zhu, L. Zhang, C. Huang and G. Cui, Phys. Chem. Chem. Phys., 2014, 16, 19206.
196. F. C. Hanusch, E. Wiesenmayer, E. Mankel, A. Binek, P. Angloher, C. Fraunhofer, N. Giesbrecht, J. M. Feckl, W. Jaegermann, D. Johrendt, T. Bein and P. Docampo, J. Phys. Chem. Lett., 2014, 5, 2791.
197. C. Bernal and K. Yang, J. Phys. Chem. C, 2014, 118, 24383.
198. Y. Ogomi, A. Morita, S. Tsukamoto, T. Saitho, N. Fujikawa, Q. Shen, T. Toyoda, K. Yoshino, S. S. Pandey, T. Ma and S. Hayase, J. Phys. Chem. Lett., 2014, 5, 1004.
199. F. Hao, C. C. Stoumpos, R. P. Chang and M. G. Kanatzidis, J. Am. Chem. Soc., 2014, 136, 8094.
200. F. Zuo, S. T. Williams, P. W. Liang, C. C. Chueh, C. Y. Liao and A. K. Jen, Adv. Mater., 2014, 26, 6454.
201. N. K. Noel, S. D. Stranks, A. Abate, C. Wehrenfennig, S. Guarnera, A.-A. Haghighirad, A. Sadhanala, G. E. Eperon, S. K. Pathak, M. B. Johnston, A. Petrozza, L. M. Herz and H. J. Snaith, Energy Environ. Sci., 2014, 7, 3061.
202. F. Hao, C. C. Stoumpos, D. H. Cao, R. P. H. Chang and M. G. Kanatzidis, Nat. Photonics, 2014, 8, 489.
203. K. Mahmood, B. S. Swain, A. R. Kirmani and A. Amassian, J. Mater. Chem. A, 2015 10.1039/c4ta04883k.
204. L. S. Oh, D. H. Kim, J. A. Lee, S. S. Shin, J.-W. Lee, I. J. Park, M. J. Ko, N.-G. Park, S. G. Pyo, K. S. Hong and J. Y. Kim, J. Phys. Chem. C, 2014, 118, 22991.
205. A. Bera, K. Wu, A. Sheikh, E. Alarousu, O. F. Mohammed and T. Wu, J. Phys. Chem. C, 2014, 118, 28494.
206. G. E. Eperon, V. M. Burlakov, A. Goriety and H. J. Snaith, ACS Nano, 2014, 8, 591.
207. P.-Y. Chen, J. Qi, M. T. Klug, X. Dang, P. T. Hammond and A. M. Belcher, Energy Environ. Sci., 2014, 7, 3659.
208. E. Horvath, M. Spina, Z. Szekrenyes, K. Kamaras, R. Gaal, D. Gachet and L. Forro, Nano Lett., 2014, 14, 6761.
209. S. M. Kang, N. Ahn, J.-W. Lee, M. Choi and N.-G. Park, J. Mater. Chem. A, 2014, 2, 20017.
210. B. J. Kim, D. H. Kim, Y.-Y. Lee, H.-W. Shin, G. S. Han, J. S. Hong, K. Mahmood, T. K. Ahn, Y.-C. Joo, K. S. Hong, N.-G. Park, S. Lee and H. S. Jung, Energy Environ. Sci., 2014 10.1039/c4ee02441a.
211. F. Matteocci, S. Razza, F. Di Giacomo, S. Casaluci, G. Mincuzzi, T. M. Brown, A. D'Epifanio, S. Licoccia and A. Di Carlo, Phys. Chem. Chem. Phys., 2014, 16, 3918.
212. L. Qiu, J. Deng, X. Lu, Z. Yang and H. Peng, Angew. Chem., Int. Ed., 2014, 53, 10425.
213. C. Roldán-Carmona, O. Malinkiewicz, R. Betancur, G. Longo, C. Momblona, F. Jaramillo, L. Camacho and H. J. Bolink, Energy Environ. Sci., 2014, 7, 2968.
214. C. Chen, S. Hsiao, C. Chen, H. Kang, Z. Huang and H. Lin, J. Mater. Chem. A 10.1039/c4ta05237d.

---