Recent progress on stability issues of organic–inorganic hybrid lead perovskite-based solar cells

Abstract

Over the past few years, substantial progress has been made in research on organic–inorganic halide perovskite solar cells. The power conversion efficiency (PCE) of perovskite solar cells has been boosted from 3.8% in 2009 to 22.1% in 2016 due to the excellent intrinsic properties of organic–inorganic hybrid lead perovskite, such as a high light absorption coefficient and long carrier diffusion length and the intensive efforts made to optimize film deposition and device fabrication techniques. In addition to the high PCE, the perovskite solar cells take advantage of low cost and solution processability, making them a promising photovoltaic technology to date. However, a number of obstacles limit their commercialization, such as long-term device stability and the toxicity of lead components. In this mini-review, we highlight the instability issues of organic–inorganic hybrid lead perovskite solar cells and then turn to a discussion of corresponding solutions.

---

1. Introduction

Solar energy in all its facets raises the greatest of public expectations. Currently, the conversion of sunlight into energy can be achieved through solar photovoltaic cells, via concentrated photovoltaic and solar thermal technologies. Photovoltaics (PV) is a promising technology that takes direct advantage of our planet's ultimate source of power, i.e., the sun. When exposed to light, solar cells are capable of producing electricity without any harmful effect to the environment, which means they can generate power for many years, only requiring minimal maintenance and operational costs. Presently, PV only has a relatively small share of total global electricity generation (∼1%) but the use of solar PV is expanding rapidly due to its reduced cost.

Currently, the widespread use of photovoltaics is limited by its relatively high cost and low power conversion efficiency (PCE). Solar cells made of crystalline silicon have dominated the PV market over the past half-century. However, due to the high cost and sophisticated manufacturing steps, “third generation” solar cell technologies have been developed to pursue high power conversion efficiency, solution processing and low cost, including light-concentrator cells, organic photovoltaics (OPV), dye-sensitized solar cells (DSSC), organic–inorganic hybrid perovskite solar cells (PSC), and so on.^1,2

Since their first application as visible-light absorbers in DSSCs in 2009,² organic–inorganic hybrid lead perovskite materials have evoked a tide of research in this field.^3–9 In the past six years, the PCE of solution-processed PSC has increased explosively to 22.1% from its initial value of 3.8%.² This achievement obtained for PSCs should be attributed to the unique and excellent properties of the organic–inorganic hybrid lead perovskite materials, including suitable bandgap (∼1.5 eV), high absorption coefficient (∼10⁴ cm⁻¹), long charge carrier lifetime (∼300 ns) and diffusion length (>1 μm).^4,5,10,11 Organic–inorganic hybrid lead perovskite materials have a generic chemical formula of AMX₃ that exhibits a cubic unit cell (shown in Fig. 1).¹² A represents an organic cation which typically is CH₃NH₃⁺ (MA), C₂H₅NH₃⁺ (EA) or HC(NH₂)₂⁺ (FA) located at the corners of the cubic unit cell.^12–15 M is a divalent metal (Pb²⁺, Sn²⁺) residing at the body center.^16–19 X is a halogen anion (Cl⁻, Br⁻, I⁻) located at the face centers.^20,21 Among various perovskite compositions, methylammonium lead iodide (MAPbI₃) has been the prototypical and most-researched compound. Life cycle analysis and environmental impact analysis have shown that a stable perovskite solar module possesses a short energy payback time of 0.22 years.²² These results have made such solar cells viable members of next generation photovoltaics that can adopt a scalable fabrication and low-cost solution process.

Fig. 1 Crystalline structure of inorganic–organic hybrid lead perovskite AMX₃.

Despite the indisputable and impressive progress in laboratory conversion efficiency, the remaining issue inhibiting the commercialization of PSCs is stability. Compared to the counterpart of liquid electrolyte-based perovskite solar cells,^2,23 the stability of solid-state devices has been improved remarkably.²⁴ However, the stability of organic–inorganic hybrid lead perovskite solar cell devices is still unsatisfactory, particularly in the presence of moisture and light illumination. Several reports have indicated that the instability of PSCs is mainly caused by the degradation of inorganic–organic hybrid lead perovskite materials and charge transport materials as well as interfaces between them.^25–32 The inorganic–organic hybrid lead perovskite materials are sensitive to moisture, oxygen, UV-light, thermal stresses, and some reactive dopants in charge transport layers.^26,28,30 To achieve good reproducibility and long lifetimes for PSCs with excellent efficiency, reducing the degradation rate of perovskite must be urgently addressed. This review summarizes the factors that influence the PSC device's stability and provides some strategies for improving device stability, which may give some insights into stability enhancement for PSCs, and lead to an understanding of how far we have to go to achieve really stable devices.

2. The development of perovskite solar cells

Up to now, several device architectures have been introduced for PSCs with which excellent photovoltaic performance can be achieved. Fig. 2 illustrates the most popular PSCs device structures in the literature. Their corresponding progress and representative achievements are also presented. In 2009, Miyasaka et al. provided strong evidence that organic–inorganic lead perovskite can be used as a light absorber for DSSCs and a PCE of 3.8% is achieved by using this new light absorber. The devices showed a high photovoltage of 0.96 V when MAPbBr₃ was used.² In those devices, a conventional dye-sensitized mesoporous n-type titanium oxide TiO₂ electrode and liquid electrolytes (typically the I⁻/I₃⁻ redox couple) were employed. In 2011, the device PCE was increased to 6.54% by optimizing the mesoporous TiO₂ film thickness and the loading of perovskite materials.²³ In this case, a 3.6 μm-thick TiO₂ film was modified by Pb(NO₃)₂, followed by the deposition of ca. 2–3 nm sized MAPbI₃ quantum dots. The stability of these devices was poor in a liquid electrolyte cell configuration. However, the high absorption coefficient of MAPbI₃ quantum dots renders them one of the best absorbers in solid-state solar cells. In 2012, Park and Grätzel et al. demonstrated ∼10% efficiency for perovskite-based solid-state solar cells by adopting small molecular 2,2′,7,7′-tetrakis-(N,N-di-p-methoxyphenylamine)-9,9′-spirobifluorene (spiro-OMeTAD) as the hole transport material (HTM) to replace the liquid electrolyte.²⁴ At the same time, Snaith and Miyasaka reported similar work in which an insulating Al₂O₃ mesoporous layer and spiro-OMeTAD were used as scaffold and HTM, respectively. The device exhibited a PCE of 10.9%.³³ Since then, this kind of device has been called a solid-state mesoporous perovskite solar cell (Fig. 2a), in which the perovskite layer was sandwiched between a compact or mesoporous TiO₂ and spiro-OMeTAD. The application of spiro-OMeTAD dramatically increases both device efficiency and device stability although the stability issue is still under debate.²⁴ The device performance has been further improved by controlling the perovskite morphology. For example, a simple one-step spin-coating method has been widely utilized to deposit perovskite from a single precursor solution. MAPbI₃ precursor solution can be obtained by dissolving stoichiometric quantities of MAI and PbI₂ in polar solvents such as g-butyrolactone (GBL) or N,N-dimethylformamide (DMF).^24,34 By optimizing the precursor concentration and spin-coating procedure parameters, the MAPbI₃ can be formed within the pores of the TiO₂ mesoporous layer. However, the infiltration of perovskite into the mesoporous layers would critically depend on the solution concentration and spin-coating speed as well on the solvents.^35a This is quite an old topic for solid-state DSSCs, in which the HTMs need to be deposited within the mesoporous TiO₂ layers.^35b–d For example, rough surfaces were reported due to the crystallization tendency of the perovskite, which causes multiple shunt pathways in the solar cells.³⁴ Later on, a two-step sequential deposition method was introduced to efficiently form the perovskite pigment, which led to a PCE of 15% for the solid-state mesoscopic solar cells and increased batch-to-batch reproducibility.³⁶ This method involves a spin-coating step with PbI₂ solution and a dipping step in MAI solution of the as-prepared PbI₂ film. The device PCE was further increased to 17% by using a two-step spin-coating procedure, which involves two spin-coating steps with PbI₂ solution and MAI solution.³⁷ The size of the cuboid MAPbI₃ can be easily controlled by varying the MAI concentration, which is essential to device performance. In 2015, Seok et al. reported on PSCs with an efficiency of over 20% by mixing narrow band gap formamidinium lead iodide (FAPbI₃) with MAPbI₃ perovskite as light absorber, which has a broader absorption spectrum compared with pure MAPbI₃.³⁸

Fig. 2 Different device configurations for perovskite solar cells and the corresponding progress and representative achievements: (a) solid-state mesoscopic solar cell; (b) planar heterojunction solar cell; (c) inverted planar heterojunction solar cell; (d) full-printable HTM-free mesoporous solar cell.

During the same period, another device configuration, a planar heterojunction (PHJ) perovskite solar cell, has also achieved significant progress. Snaith et al. firstly adopted the insulating mesoporous Al₂O₃ as a scaffold to replace the mesoporous n-type TiO₂ layer. The device showed a PCE of 10.9% with an increased open-circuit voltage.³³ The bipolar conductivity and long electron–hole diffusion length of perovskite film indicate that the mesoporous TiO₂ layer (i.e., electron transport layer ETL) is unnecessary. This leads to the use of PHJ perovskite solar cells, instead of cells with a mesoporous structure (Fig. 2b).^34,39,40 This type of device architecture excludes the utilization of a mesoporous scaffold and simplifies device processing. However, it should be noted that perovskite morphology plays a critical role in PHJ device performance, which can be influenced by underlying substrates, annealing temperature and perovskite deposition time.³⁹ A device with a PCE of 15.4% was further achieved with vapour deposition of the perovskite layer, yielding a short circuit photocurrent of 21.5 mA cm⁻², an open-circuit voltage of 1.07 V and a fill factor of 0.68.⁴¹ Meanwhile, interfacial engineering is another efficient way to enhance device efficiency.^42–45 By doping the TiO₂ with yttrium and modifying indium-tin oxide (ITO) with polyethyleneimine ethoxylated (PEIE), Yang and co-workers reported PHJ perovskite devices with a PCE of 19.3%.⁴⁵ The modification of TiO₂ and ITO electrodes achieves favourable energy alignment and facilitates efficient electron transport between ITO and perovskite layers, which suppresses excessive interface recombination and thus enhances device performance.

Meanwhile, an inverted PHJ PSC (Fig. 2c) with a similar device structure in organic solar cells has attracted growing attention.^46–48 In 2013, Chen and Guo firstly reported a device architecture of poly(3,4-ethylenedioxythiophene):polystyrene sulphonate (PEDOT:PSS)/MAPbI₃/(6,6)-phenyl-C₆₁-butyric acid methyl ester (PCBM) or indene-C₆₀ bisadduct (ICBA)/bathocuproine (BCP)/Al, in which PEDOT:PSS acted as a hole transport layer and PCBM or ICBA is employed as an electron transport layer, achieving a PCE of 3.9%.⁴⁹ A very promising PCE of over 15% has been demonstrated in this type of solar cell by optimizing the perovskite morphology and the interface.^48,50,51 Several other inorganic counterparts with appropriate work functions have been tested as efficient alternatives, such as PEDOT:PSS, including CuSCN,^52,53 NiO_x,^54–58 and V₂O₅ ⁵⁹ In particular, NiO_x has shown the greatest promise due to its low cost, superior stability, high optical transmittance, and appropriate energy levels. The suitable work function of NiO_x facilitates hole transport and electron blocking as well as aligning well with the highest occupied molecular orbital (HOMO) level of MAPbI₃.^54,60–62 A nanostructured NiO_x film prepared by a pulsed laser deposition method or low-temperature combustion process possessed good optical transparency and electrical conductivity, resulting in a PCE of 15–18%.^57,61

Apart from the above-mentioned device configurations, fully screen-printed HTM-free mesoscopic PSCs which employ a carbon counter electrode instead of a metal electrode (Fig. 2d) also arouse great interest.⁶³ Very recently, we demonstrated a ∼15% efficient mesostructured PSC based on an inorganic scaffold, which employs a quadruple-layer architecture of mesoporous TiO₂/Al₂O₃/NiO/C.⁶⁴ The application of a carbon electrode has made the device preparation easier via screen printing which is favourable for achieving large-scale production.

3. Stability of organic–inorganic hybrid lead perovskites and their solar cells

High power conversion efficiency and long-term stability are two key requirements for the commercialization of solar cells. At present, excellent device performance has been achieved for solar cells based on inorganic–organic hybrid lead perovskites, which are comparable with silicon solar cells. However, device stability has still been a serious obstacle toward their practical application. The instability of devices comes mainly from the degradation of perovskite materials.^26,28,30 Organic–inorganic perovskite is susceptible to degradation by moisture, UV light, and temperature.

3.1 Effects of moisture

Among the different factors, moisture has been considered one of the biggest challenges. So far, several mechanisms have been reported for perovskite degradation correlated to moisture. For example, an acid–base reaction mechanism was proposed to explain the degradation of MAPbI₃, as shown in Fig. 3.²⁵ In this mechanism, a single water molecule is sufficient to trigger the perovskite degradation process. The H₂O molecule firstly reacts with MAPbI₃ and forms an intermediate of [(CH₃NH₃⁺)_n−1(CH₃NH₂)_nPbI₃][H₃O]. Then this intermediate decomposes into HI, CH₃NH₂, PbI₂, and H₂O. An excess of water is required to dissolve the by-products of HI and CH₃NH₂. Once the degradation process starts to happen, this reaction will keep going until the H₂O is saturated with HI or the vapour pressure of CH₃NH₂ reaches equilibrium.

Fig. 3 Possible decomposition pathway of hybrid halide perovskites in the presence of water.

Some researchers have different opinions on explaining the degradation of perovskite films. They suggested that in the presence of H₂O, MAPbI₃ firstly decomposes into MAI solution and PbI₂ (eqn (1)), and then the MAI solution decomposes into CH₃NH₂ and HI (eqn (2)).^26,65 The resulting HI either reacts with oxygen in the air or undergoes a photochemical reaction under light illumination to generate I₂ (eqn (3-1)) or eqn (3-2)). X-ray diffraction (XRD) spectra of MAPbI₃ film before and after degradation assuredly verify the proposed degradation process. All the diffraction peaks for MAPbI₃ disappeared, while the peaks for hexagonal 2H polytypic PbI₂ and orthorhombic I₂ appeared.²⁶

CH₃NH₃PbI(aq) ↔ CH₃NH₂(aq) + HI(aq) (2)

4HI(aq) + O₂ ↔ 2I₂(s) + 2H₂O (3-1)

Kamat's group and Kelly's group suggested another different degradation mechanism.^66,67 The MAPbI₃ could complex with H₂O, forming a similar hydrate product to MA₄PbI₆·2H₂O according to eqn (4), rather than simply reverting to PbI₂ under H₂O exposure.⁶⁶

CH₃NH₃PbI₃ + H₂O ↔ (CH₃NH₃)₄PbI₆·2H₂O (4)

An experimental setup (shown in Fig. 4a) was designed to precisely control the humidity to quantitatively and systematically investigate the MAPbI₃ degradation process.⁶⁷ In situ absorption spectroscopy could be performed to monitor perovskite phase changes in the degradation process (Fig. 4b). Fig. 4c shows the powder XRD spectra of bulk MAPbI₃, demonstrating that the water-added perovskite sample consists of a mixture of MA₄PbI₆·2H₂O and residual MAPbI₃. This result indicates a hydrated intermediate containing isolated PbI₆⁴⁻ octahedra as the first step in the perovskite degradation (eqn (5)).

4CH₃NH₃PbI₃ + H₂O → (CH₃NH₃)₄PbI₆·2H₂O + 3PbI₂ (5)

Fig. 4 (a) Sample holder for in situ UV-vis spectroscopy; (b) UV-vis spectra of a MAPbI₃ film exposed to N₂ gas with RH = 98 ± 2%, acquired at 15 min intervals. (c) Powder XRD patterns for MAPbI₃ powder, before and after the addition of the trace amount of water. The calculated powder patterns for tetragonal MAPbI₃, MA₄PbI₆·2H₂O, and PbI₂ are shown for comparison.

The moisture-assisted interaction between metal electrodes (such as Ag, Al, and Au) and perovskite could also destroy the stability of perovskite and perovskite devices.^68–70 It was reported that the PCE of perovskite solar cells dropped drastically in humid conditions when the Ag back-contact electrode was used.⁷¹ The formation of AgI observed by X-ray diffraction could be one of the most plausible causes. Qi and co-workers had carried out systematic studies to understand the origin of AgI and confirmed that ion migration could be responsible for Ag electrode corrosion. This process can be significantly accelerated in a moist atmosphere.⁶⁹ This was confirmed by Lee's investigation, showing that migration of ionic defects in the perovskite layer plays a critical role in the corrosion of metal electrodes, and thus in device degradation.⁷⁰

3.2 Effects of temperature

Thermal annealing is an effective route to form the perovskite crystal structure in a typical solution-processed PSC, which consequently ultimately affects device efficiency.^72–74 It was found that perovskite fabricated with a larger fraction of organic cations and annealed at a lower temperature cannot fully form a crystalline film, suggesting that organic cations slow down the perovskite formation. In this case, a higher thermal annealing temperature is required for perovskite crystallization. However, a high temperature may induce a crystal structure transition in the organic–inorganic hybrid perovskite.⁷⁵ A tolerance factor (t) has been proposed as an important index to evaluate the stability of AMX₃ perovskite. The tolerance factor can be described by eqn (6):where r_A, r_M and r_X are the ionic radii for the ions in the A, M and X sites, respectively. Stable perovskites can be formed when t values are between 0.8 and 1.^76,77 Therefore, in order to get a stable perovskite structure, ions with a particular radius size should be carefully selected. In addition to ionic radius, there are other aspects that significantly affect the perovskite phase transition and structural stability, such as temperature.^12,77–79 As early as 1987, Weber et al. investigated the relationship between temperature and the structure of methylammonium trihalide MAPbX₃ (X = Cl, Br, I).⁷⁷ At room temperature, both MAPbBr₃ and MAPbCl₃ present a cubic structure. However, MAPbI₃ becomes a tetragonal structure when the temperature increases to 327.4 K. The phase transformation from tetragonal phase to cubic phase is accompanied by a slight distortion of the PbI₆ octahedra around the c axes. Whereas lowering the temperature would result in a phase transition from the tetragonal phase to the orthorhombic phase.¹² Baikie et al. investigated the temperature-dependent phase transition for MAPbX₃ perovskite in detail and gave a description of the preparation, structural characterization and physical characteristics of MAPbI₃.¹² In situ powder XRD and single crystal XRD characteristics were employed to monitor the phase transition as a function of temperature. The variation in lattice parameters with temperature change indicates that MAPbI₃ undergoes a tetragonal to orthorhombic phase transition at approximately 161 K and a tetragonal to cubic phase transition at a temperature of 329.15 K. These findings agree well with the previous result reported by Weber et al.⁷⁷ The phase structure of MAPbX₃ perovskites strongly influences their electronic and optical properties.⁷⁸ For PSCs which are working under illumination by the sun, the elevated temperature by long-time light illumination would cause the phase transition of perovskite materials and then affect the device performance of the solar cells. We have investigated the photovoltaic metrics of MAPbI₃-based PSCs over a wide temperature range of 80–360 K.⁷⁹ We found that the photovoltaic metrics were strongly affected by the phase transition from the tetragonal to the orthorhombic phase. A maximum value of open-circuit voltage was observed at about 200 K which is close to the phase transition temperature (from the tetragonal to the orthorhombic phase). The photocurrent is remarkably stable down to 240 K but drops precipitously upon approaching and below the phase transition temperature at 160 K. The temperature-dependent thermal conductive properties of MAPbI₃ were also reported,⁸⁰ showing that MAPbI₃ has a very low thermal conductivity in both large single crystals and polycrystalline films.

3.3 Effects of light exposure

TiO₂ is the most commonly used electron transport material in PSCs, in either compact or mesoporous form.^{64,79,81–83} However, some experiments have indicated that when TiO₂ was used as the electron transport layer, the PSC device performance suffered from rapid decay even though it was encapsulated in an inert atmosphere.^27,28,84,85 After carefully excluding the influence of MAPbI₃ and spiro-OMeTAD on the instability of PSCs under illumination, oxygen vacancies in the TiO₂ particles were proposed to serve as deep electronic trap sites.²⁷ They can bind with photo-induced electrons from perovskite and cause the degradation in device performance. Fig. 5 illustrates a hypothesized degradation mechanism. It is well-known that TiO₂ particles contain oxygen vacancies, particularly at the particle surface.⁸⁶ Those oxygen vacancies are deep electron-donating sites and can easily adsorb oxygen (Fig. 5a). UV illumination could excite the electron in TiO₂ to form an electron–hole pair. The hole in the valence band of TiO₂ would recombine with the electron at the oxygen adsorption site, resulting in desorption of oxygen (Fig. 5b). In this case, a free electron in the conduction band of TiO₂ and a positively charged, unfilled oxygen vacancy site at the TiO₂ surface are left behind. The excess holes in heavily p-doped spiro-OMeTAD will readily recombine with free electrons. In addition, the photo-induced electrons from pigments are trapped by the oxygen vacancy sites (Fig. 5c). Finally, the holes in the HTM will recombine with the immobile trapped electrons (Fig. 5d). The presence of oxygen could decrease the number of empty deep trap sites because they could adsorb on the TiO₂ to pacify these sites.

Fig. 5 Degradation mechanism under UV illustration. (a) Oxygen adsorption on the vacancy. (b) Recombination of holes in TiO₂ with electrons on oxygen site. (c) Recombination of holes in HTM with the free electron left on TiO₂. (d) Recombination of holes in HTM with electrons reduced by dye.

Another degradation mechanism for the PSCs under light illumination was proposed by Hitoshi et al.²⁸ They examined the light durability of MAPbI₃-based solar cells without encapsulation. The device architectures used for study are (A) FTO/TiO₂/MAPbI₃/CuSCN/Au and (B) FTO/TiO₂/Sb₂S₃/MAPbI₃/CuSCN/Au. As shown in Fig. 6a, the PCE of device B was maintained at 65% of its initial efficiency without encapsulation after 12 h of light irradiation. However, the stability of device A was very poor, with a drastic decrease within 5 h, approaching to zero within 12 h. Therefore, the Sb₂S₃ layer at the TiO₂/MAPbI₃ interfaces significantly affects the device PCE. After a combination of characterizations such as reflectance absorption spectra, XRD patterns, incident photon-to-current conversion efficiency (IPCE) spectra and Fourier transform infrared spectroscopy (FTIR) spectra, a reaction scheme to explain the degradation effect of the MAPbI₃ layer against light exposure was put forward and is shown in Fig. 6b and c. After overnight light exposure, the MAPbI₃ layer without Sb₂S₃ was degraded into PbI₂ by the removal of CH₃NH₂ and HI, as shown in eqn (7) and Fig. 6b. On the other hand, the MAPbI₃ layer can be protected and made durable against light exposure with the insertion of a Sb₂S₃ layer (Fig. 6c). Therefore, the decomposition of MAPbI₃ occurs at the interface between MAPbI₃ and TiO₂. As a typical n-type semiconductor, TiO₂ is widely used as a photocatalyst for environmental purification, (e.g. reduction of CO₂, decomposition of organic compounds and splitting of water).^35,87–90 TiO₂ has a strong ability to extract electrons from electron-rich materials. Hence, electron extraction from iodide anions by TiO₂ facilitates the decomposition of MAPbI₃. eqn (8)–(10) explain possible reactions happening at the TiO₂ surface. Firstly, the valence electron of I⁻ is extracted by the TiO₂. Consequently, the structure of MAPbI₃ is deconstructed, resulting in I₂ (Fig. 6b and eqn (8)). Then, reaction (9) can be moved forward by a continuous elimination of H⁺ through reaction (10) and evaporation of CH₃NH₂ (bp 17 °C). Finally, I₂ is reduced by extracted electrons at the interface between TiO₂ and MAPbI₃ (Fig. 6b). Inserting a blocking layer of Sb₂S₃ between TiO₂ and perovskite MAPbI₃ can significantly enhance device stability under light illumination.

CH₃NH₃PbI₃ ↔ PbI₂ + CH₃NH₂↑ + HI↑ (7)

2I⁻ ↔ I₂ + 2e⁻ (8)

CH₃NH₂⁺ ↔ 3CH₃NH₂↑ + 3H⁺ (9)

I⁻ + I₂ + 3H⁺ + 2e⁻ ↔ 3HI↑ (10)

Fig. 6 (a) Variation of photo-energy conversion efficiencies of solar cells during light exposure (AM1.5, 100 mW cm⁻²) without encapsulation in air for 12 h: (up) FTO/TiO₂/Sb₂S₃/MAPbI₃/CuSCN/Au and (down) FTO/TiO₂/MAPbI₃/CuSCN/Au; degradation scheme of MAPbI₃ perovskite solar cells during light exposure test: (b) without Sb₂S₃ layer and (c) with Sb₂S₃ layer.

In addition, the phase separation of mixed halide perovskite, for example MAPbI_3−xBr_x, under light illumination has been observed. Light exposure causes phase separation into iodide-rich minority and bromide-enriched majority domains,^91–93 in which the iodide-rich minority can serve as a recombination center trap, leading to a photo-instability of PSC devices. A new crystalline phase could form upon light illumination of MAPbI_3−xBr_x and other mixed halide perovskites.⁹²

4. Stability of charge transport materials

Apart from the effects of environmental factors, the degradation of charge transport materials also significantly influences the long-term stability of PSCs.^29,94–96 Charge transport layers play an important role both in transporting charges to the electrode and in preventing charge recombination. A wide variety of HTMs ranging from classical semi-conducting polymers to small molecules have been investigated in PSCs.^34,96–100 Spiro-OMeTAD has hitherto been the most effective HTM in PSCs. However, pristine spiro-OMeTAD suffers from low hole mobility due to the large intermolecular distance.^101–103 4-tert-Butylpyridine (tBP) and bis(trifluoromethane)sulfonimide lithium salt (Li-TFSI) are widely used as dopants^104,105 to improve the electronic property. The use of this lithium salt also plays an important role in tuning the polarity of spiro-OMeTAD which is able to increase the contact between MAPbI₃ and the hole transport layer. However, as the additive tBP is a polar solvent similar to GBL, it can dissolve hybrid perovskites, further causing the instability of the solar cell. In addition, the deliquescent behavior of Li-TFSI may tend to introduce moisture into the devices, thus aggravating perovskite structure degradation. Upon the dropping of tBP, the MAPbI₃ films faded out in nitrogen atmosphere due to the corrosion of MAPbI₃ with the existence of tBP.³⁰ The absorption intensity was nearly close to zero after the spin-coating of tBP directly on the top of the MAPbI₃ layer. It was found that PbI₂ reacts with tBP to form a transparent-yellow liquid solution at room temperature. A light yellow powder was obtained by drying the solution under vacuum. The binding energy of Pb 4f_7/2 for tBP-treated powder shifted to a lower energy level from 139.02 eV to 138.47 eV, observed in XPS spectra. Those results indicate the formation of a complex (i.e. [PbI₂·xtBP]) by the interaction between tBP and the center lead metal.

In addition to the wide use of spiro-OMeTAD in the normal architecture, PEDOT:PSS and PCBM are the most commonly used organic charge transport materials in inverted PHJ perovskite solar cells.^50,106,107 However, long-term exposure in ambient air could damage the PEDOT:PSS and PCBM layer through the adsorption of oxygen or water.^29,108 This could break the hydrogen bonds between individual PEDOT:PSS grains within the layer, resulting in a significant loss in cohesion.^31,109 The LUMO of PCBM could increase after air exposure due to the water–PCBM interaction.²⁹ The degradation of PCBM would result in a large contact resistance which decreases the device performance.

5. Solutions to increase the stability of perovskite solar cells

5.1 Strategies for reduction of inorganic/organic halide lead perovskite degradation

Material engineering has been considered one of the most promising strategies to enhance the stability of inorganic/organic halide lead perovskites. It was reported that iodide–chloride mixed halide perovskite MAPbI_xCl_3−x was more stable than single-halide perovskite MAPbI₃ in ambient atmosphere.³³ The stability of MAPbI₃ in humid air could be significantly improved by doping with bromide.¹¹⁰ Tuning the tolerance factor is also an effective way to stabilize the perovskite structure for solar cell applications. Extensive researches have been done on formanidinium methylammonium lead iodide (FAPbI₃) due to its broad light absorption and good thermal stability.^111–113 The A cation in FAPbI₃ is larger than that in MAPbI₃, which would normally result in a higher tolerance factor t.¹¹⁵ The stability of FAPbI₃ perovskite can be further enhanced by alloying with CsPbI₃.¹¹⁴ The FA_1−xCs_xPbI₃ alloy shows a higher stability than the neat FAPbI₃ under humid conditions.

Inorganic perovskites presumably exhibiting higher thermal stability than organic–inorganic hybrid counterparts have great potentialities to fabricate stable solar cells. Two representative inorganic perovskites, cesium lead trihalides (CsPbX₃) and cesium tin trihalides (CsSnX₃),^{113,115–117} have been reported as light absorbers in photovoltaics. The CsPbI₃-based planar structured solar cell showed a maximum 2.9% PCE.¹¹⁵ The thermal stability of CsPbBr₃ and MAPbBr₃ were compared by thermogravimetric analyses,¹¹⁸ showing that the first weight loss onset for CsPbBr₃ was at about 580 °C; however, the temperature was as low as 220 °C for MAPbBr₃. Therefore, organic cations might be unnecessary for high performance PSCs.^113,119 The CsPbBr₃-based device shows a comparable photovoltaic performance to that of an MAPbBr₃-based device, but with much improved stability. The work of inorganic PSCs paves the way for further developments, likely to lead to much more thermally stable solar cells and other optoelectronic devices.

Two-dimensional (2D) perovskite may be another good choice for making long-term stable PSCs.^120–122 2D layered perovskite with a general formula of A₂A′_n−1M_nX_3n+1 is derived from 3D perovskite AMX₃ by introducing a larger A cation into the crystal structure.^120–123 In the 2D structure, the small A′ cation is separated by a layer of larger A cation and the distance between inorganic sheets is increased. Compared with 3D perovskite, the 2D perovskite has a larger band gap and an increased binding energy, which can be modulated by varying the stacking of the metal halide lattices (n) but is more stable when subjected to moisture and light. Recently, solar cells based on 2D perovskite also have achieved considerable efficiencies with excellent humility stability.^120,121 Mhaisalkar et al. reported on the fabrication of nanostructuring mixed-dimensional perovskite [(IC₂H₄NH₃)₂(CH₃NH₃)_n−1Pb_nI_3n+1] and its application in mesoscopic solar cells, achieving an overall PCE of 9.03%.¹²⁰ More importantly, the 2D perovskite is more stable in a humid atmosphere than MAPbI₃. Mohite and co-workers produced 2D perovskite films with near-single-crystalline quality.¹²¹ Solar cells based on the as-prepared solution-processed 2D layered perovskite films showed a photovoltaic efficiency of 12.52%. The aggressive long-term stability measurements demonstrated that devices exhibit greatly improved stability compared with their 3D equivalents when subjected to light, humidity and heat stress tests. After constant illumination of 100 mW cm⁻² over 2250 h, the un-encapsulated devices based on 2D perovskite still retained over 60% of their initial efficiency and exhibited a much greater tolerance to high humidity than the 3D counterparts.

5.2 Strategies for reduction of UV-light-induced degradation

As discussed above, under UV-light illumination the degradation of PSCs originates from the oxygen vacancies in the TiO₂ layer, which can trap photo-induced electrons in perovskite and/or extract electrons at I⁻, and thus deconstruct the perovskites. Therefore, effective routes to prevent UV degradation include isolating the TiO₂ from UV light, and even replacing the TiO₂ with other electron transport materials. For example, a device which adopts an insulating mesoporous Al₂O₃ scaffold showed significantly improved stability.²⁷ It is evident that the Al₂O₃-based devices are considerably more stable than the TiO₂-based devices under UV-light illumination. Contrary to the evolution of TiO₂-based cells under UV light exposure, Al₂O₃-based devices remained stable over a 1000 h exposure period and the photocurrent remains closed to 15 mA cm⁻².

5.3 Strategies for charge transport materials

Another choice to enhance the stability of PSCs is to find stable charge transport materials. In this regard, new HTMs without additives have drawn intensive attention. For instance, an efficient pristine HTM, tetrathiafulvalene derivative (TTF-1) without using dopants was introduced for PSCs.⁹⁶ The dopant-free TTF-1 based device exhibits slower degradation in device performance than a control device based on doped spiro-OMeTAD at a relative humidity of ≈40% in air without encapsulation. We suggested dopant-free 3,3′-bithiophene derivatives as hole transport materials (coded as DHPT-SC, DOPT-SC, and DEPTSC) for PSCs, which can significantly improve the cell stability and device performance.¹²⁴ Polymer poly(methylmethacrylate) (PMMA)-functionalized single-walled carbon nanotubes (SWNTs) embedded in an insulating polymer matrix were used as the hole transport layer to replace the organic hole transport material.¹²⁵ PMMA is nonhygroscopic and therefore could prevent the penetration of moisture into the perovskite structure while simultaneously enhancing the device stability. Other strategies such as hydrophobic surface modifications have also been applied to improve solar cell stability. For example, an oligothiophene derivative named DR3TBDTT has been used as HTM for PSC devices.¹²⁶ The authors found the perovskite layer was efficiently protected from moisture by the hydrophobic DR3TBDTT layer, and therefore devices with excellent stability could be achieved. Furthermore, the functionalized nanographene TSHBC was employed as HTM in PSCs. The resulting devices showed improved stability in a humidity of about 45% without encapsulation under AM1.5 illumination. They could also benefit from the hydrophobic nature of TSHBC.¹²⁷

5.4 Perovskite solar cells using inorganic oxide frameworks

Fully-printable HTM-free mesoscopic PSCs which employ a triple-layer of mesoscopic TiO₂/Al₂O₃ or ZrO₂/carbon as a scaffold infiltrated with perovskites has been reported to show excellent stability even in harsh conditions, such as high humidity and high temperature.^63,64 The removal or replacement of HTMs, especially organic HTMs and additives, not only simplifies the device fabrication process but also benefits from a reduction in device cost and enhancement in device stability. A significant change in perovskite solar cell configuration stems from the work of Etgar et al., who demonstrated a new type of HTM-free device.¹²⁸ This device can be also called a depleted heterojunction perovskite solar cell. The successful function of such devices indicates that hybrid chloride lead perovskite MAPbX₃ can act not only as light absorbers but also as effective HTM themselves. Carbon material based counter electrodes have been employed in the HTM-free perovskite solar cell.¹²⁹ Devices were fabricated with a one-step deposition method, which collocated with a double layer of mesoporous TiO₂ and ZrO₂ as a scaffold, achieving a maximum PCE of 12.8%. We further improved device efficiency to over 15% by the insertion of NiO as a hole transport layer in combination with a two-step-deposition method.^130–132 A quadruple-layer architecture of mesoscopic TiO₂/Al₂O₃/NiO/carbon was formed to increase charge collection (Fig. 7a).⁶⁴ Extensive stability tests proved that the PCE of a quadruple-layer device with NiO hole selective contact reduced from the initial 14.7% to the final 13.7% when they were stored in ambient atmosphere at room temperature (humidity ∼ 40%) without light for 1000 h (Fig. 7b). A photo-stability test was also performed with the quadruple-layer device being soaked in constant light with a white light LED array emitting visible light at an intensity of 100 mW cm⁻². Prior to the photo-stability test, the devices were hermetically sealed, as shown in Fig. 7c. A sheet of YC-AR130 (an anti-reflecting UV cut-off film, λ < 420 nm, Japan) was attached to each cell surface for the stability test in order to remove the effect of UV light. The light stability test results showed that the device maintained 82% of its initial PCE after light soaking for about 500 h at room temperature, and more than 75% of its initial values after exposure for 1000 h (Fig. 7d). During this study, we also found that, after aging, a recovery of original voltage and photovoltaic performance of the solar cell upon storage in the dark for several hours was observed. These results demonstrate that no serious photo-induced decomposition, or an irreversible process of photo-induced decomposition, had happened in the device under low light intensity illumination at room temperature. All these results suggest that the PSC device could be stable with a particular configuration and architecture, which provides a promising path towards realizing efficient and stable perovskite devices. For example, metal oxides with a high charge carrier mobility and matched energy level are used to increase the inverted PHJ devices' stability. p-Type NiO_x and n-type ZnO nanoparticles were used as hole and electron transport layers to improve device stability against water and oxygen.²⁹ These experimental results highlight the potential application of inorganic metal oxides in PSCs.

Fig. 7 (a) Schematic representation of the quadruple-layer PSC device. (b) Stability investigation on quadruple-layer PSC device stored at room temperature in the dark (blue), and at 60 °C in the dark (red). (c) Photo stability test at 100 mW cm⁻² (with white light emitting diodes) and a photograph of the sealed device. (d) Evolution of PCE during the light soaking test (green).

As mentioned above, it was supposed that the mixed halide MAPb(I_1−xCl_x)₃ or MAPb(I_1−xBr_x)₃ are chemically more stable in conjunction with humidity than the most commonly studied perovskite MAPbI₃.^33,133 Among these analogues, MAPbI₂Br has some appealing features, such as a high absorption coefficient in the range of 400–700 nm, prolonged charge carrier lifetime, and ideal conduct band position, which all make it a promising candidate for PSCs. We combined a MAPbI₂Br light absorber with a triple-layer of mesoscopic TiO₂/Al₂O₃/carbon scaffold, achieving an overall energy conversion of over 11% under standard AM1.5 illumination at 100 mW cm⁻² with superior stability.⁸³ The stability test of the unsealed devices was carried out under constant light soaking at a light intensity of 25 mW cm⁻² in an ambient atmosphere at room temperature. The efficiency of the MAPbI₂Br devices is well retained after exposure for 300 h to air.

The thick carbon counter electrode seems to be one of the best choices due to its protection from humidity. However, PSCs employing a carbon black/graphite composite CE face the dilemma that charge collection may become less efficient due to severe recombination, entailing a loss in the photocurrent. As far as we know, only a few groups can successfully fabricate efficient carbon black/graphite CE-based mesoporous PSCs in the range of 12–15%, which largely limits their scientific investigation. So far, most of the reported carbon black/graphite CE-based mesoporous perovskite solar cells were fabricated with a drop coating method to assure good penetration of the perovskite precursors into the thick-CEs. Therefore, carbon black/graphite CE-based mesoporous PSC devices always presented poorer batch-to-batch efficiency reproducibility than others. Therefore, a highly conductive thin film CE with an excellent hole transport ability would be expected to avoid the above-mentioned problems.

6. Current and future developments

Due to their unique optical and electrical properties, inorganic–organic hybrid halide perovskite materials have attracted much attention in the field of photovoltaics. Perovskite crystal structure is sensitive to subtle variation during film formation, which plays a critical role in photophysics. Although it is agreed that a crystalline film results in limited electron–hole recombination, the scientific community has not yet had a clear understanding of how microstructures affect exciton and free carrier dynamics macroscopically. A prediction of the photophysics and novel properties of new materials based on a perovskite structure is not yet possible. Therefore, it is urgent to study the structure–photophysics–function relationship for organic–inorganic hybrid lead perovskite and other types of perovskite material, on which an intermediate photophysics bridge to better connect structure and emerging optoelectronic properties with the goal of understanding the structure–function relationship should be built. Numerous research efforts have been undertaken to improve the conversion efficiency of PSCs and the highest conversion efficiency of over 22.1% was reached in 2016. However, the long-term stability of organic–inorganic hybrid lead-based PSCs has become an obstacle for their practical application. Moisture, UV-light, temperature, the organic components, phase separation, and metal contact electrode in the devices have important influences on device stability. Several degradation mechanisms have been proposed for PSCs, which provide fundamental understanding and some insights for stability enhancement. To overcome device stability issues and produce stable PSC devices, various approaches have been explored, including utilization of a protecting layer, alternative organic or inorganic charge transport materials, component engineering, interfacial engineering and so on. Meanwhile, the great promise of perovskite-based solar cells is being challenged by their lead content, and sensitivity to water. Therefore, for the large-scale scenario, we must consider the possible implications of implementing lead halide perovskite solar cells. Proper material processing and device handling technologies in perovskite solar cells will become more and more mature with a deep understanding of perovskite material degradation and device performance decay. Therefore, perovskite devices with high efficiency and long stability are expected to be realized for practical application in the near future.

Acknowledgements

Financial support from the 973 Program of China (2014CB643506 and 2013CB922104), the China Scholarship Council (no. 201506165038), the Natural Science Foundation of Hubei Province (No. ZRZ2015000203), Technology Creative Project of Excellent Middle & Young Team of Hubei Province (No. T201511), the Director Fund of the WNLO and the Wuhan National High Magnetic Field Center (2015KF18) are acknowledged.

Notes and references

1. (a) C. Tang, Appl. Phys. Lett., 1986, 48, 183; (b) M. Wang, N. Chamberland, L. Breau, J. Moser, R. Humphry-Baker, B. Marsan, S. Zakeeruddin and M. Grätzel, Nat. Chem., 2010, 2, 385.
2. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050.
3. G. Eperon, V. Burlakov, P. Docampo, A. Goriely and H. Snaith, Adv. Funct. Mater., 2014, 24, 151.
4. Q. Dong, Y. Fang, Y. Shao, P. Mulligan, J. Qiu, L. Cao and J. Huang, Science, 2015, 347, 967.
5. W. Nie, H. H. Tsai, R. Asadpour, J. Blancon, A. Neukirch, G. Gupta, J. Crochet, M. Chhowalla, S. Tretiak, M. Alam, H. Wang and A. Mohite, Science, 2015, 347, 522.
6. D. Shi, V. Adinolfi, R. Comin, M. Yuan, E. Alarousu, A. Buin, Y. Chen, S. Hoogland, A. Rothenberger, K. Katsiev, Y. Losovyj, X. Zhang, P. A. Dowben, O. Mohammed, E. Sargent and O. Bakr, Science, 2015, 347, 519.
7. K. Lim, H. Kim, J. Jeong, H. Kim, J. Kim and T. Lee, Adv. Mater., 2014, 26, 6461.
8. H. Yu, F. Wang, F. Xie, W. Li, J. Chen and N. Zhao, Adv. Funct. Mater., 2014, 24, 7102.
9. M. Saliba, T. Matsui, J. Y. Seo, K. Domanski, J. Correa-Baena, M. Nazeeruddin, S. Zakeeruddin, W. Tress, A. Abate, A. Hagfeldt and M. Grätzel, Energy Environ. Sci., 2016, 9, 1989.
10. G. Xing, N. Mathews, S. Sun, S. S. Lim, Y. M. Lam, M. Grätzel, S. Mhaisalkar and T. Sum, Science, 2013, 342, 344.
11. T. Sum and N. Mathews, Energy Environ. Sci., 2014, 7, 2518.
12. T. Baikie, Y. Fang, J. Kadro, M. Schreyer, F. Wei, S. Mhaisalkar, M. Grätzel and T. White, J. Mater. Chem. A, 2013, 1, 5628.
13. T. Koh, K. Fu, Y. Fang, S. Chen, T. Sum, N. Mathews, S. Mhaisalkar, P. Boix and T. Baikie, J. Phys. Chem. C, 2014, 118, 16458.
14. J. Im, J. Chung, S. Kim and N. Park, Nanoscale Res. Lett., 2012, 7, 353.
15. G. Kieslich, S. Sun and A. Cheetham, Chem. Sci., 2014, 5, 4712.
16. C. Stoumpos, C. Malliakas and M. Kanatzidis, Inorg. Chem., 2013, 52, 9019.
17. F. Hao, C. Stoumpos, R. Chang and M. Kanatzidis, J. Am. Chem. Soc., 2014, 136, 8094.
18. N. Noel, S. Stranks, A. Abate, C. Wehrenfennig, S. Guarnera, A. Haghighirad, A. Sadhanala, G. Eperon, S. Pathak, M. Johnston, A. Petrozza, L. Herz and H. Snaith, Energy Environ. Sci., 2014, 7, 3061.
19. F. Hao, C. Stoumpos, D. Cao, R. Chang and M. Kanatzidis, Nat. Photonics, 2014, 8, 489.
20. B. Cai, Y. Xing, Z. Yang, W. Zhang and J. Qiu, Energy Environ. Sci., 2013, 6, 1480.
21. E. Edri, S. Kirmayer, D. Cahen and G. Hodes, J. Phys. Chem. Lett., 2013, 4, 897.
22. N. Espinosa, L. Serrano-Luján, A. Urbina and F. Krebs, Sol. Energy Mater. Sol. Cells, 2015, 137, 303.
23. J. H. Im, C. Lee, J. Lee, S. Park and N. Park, Nanoscale, 2011, 3, 4088.
24. H. Kim, C. Lee, J. Im, K. Lee, T. Moehl, A. Marchioro, S. Moon, R. Humphry-Baker, J. Yum, J. Moser, M. Grätzel and N. Park, Sci. Rep., 2012, 2, 591.
25. J. Frost, K. Butler, F. Brivio, C. Hendon, M. Schilfgaarde and A. Walsh, Nano Lett., 2014, 14, 2584.
26. G. Niu, W. Li, F. Meng, L. Wang, H. Dong and Y. Qiu, J. Mater. Chem. A, 2014, 2, 705.
27. T. Leijtens, G. E. Eperon, S. Pathak, A. Abate, M. Lee and H. Snaith, Nat. Commun., 2013, 4, 2885.
28. S. Ito, S. Tanaka, K. Manabe and H. Nishino, J. Phys. Chem. C, 2014, 118, 16995.
29. J. You, L. Meng, T. Song, T. Guo, Y. Yang, W. Chang, Z. Hong, H. Chen, H. Zhou, Q. Chen, Y. Liu, N. Marco and Y. Yang, Nat. Nanotechnol., 2015, 11, 75.
30. W. Li, H. Dong, L. Wang, N. Li, X. Guo, J. Li and Y. Qiu, J. Mater. Chem. A, 2014, 2, 13587.
31. H. Yip and A. Jen, Energy Environ. Sci., 2012, 5, 5994.
32. N. Alia, A. Hussaina, R. Ahmeda, M. Wang, C. Zhao, B. Haq and Y. Fu, Renewable Sustainable Energy Rev., 2016, 59, 726.
33. M. Lee, J. Teuscher, T. Miyasaka, T. Murakami and H. Snaith, Science, 2012, 338, 643.
34. J. Heo, S. Im, J. Noh, T. Mandal, C. Lim, J. Chang, Y. Lee, H. Kim, A. Sarkar, M. Nazeeruddin, M. Grätzel and S. Seok, Nat. Photonics, 2013, 7, 486.
35. (a) J. Yu, J. Low, W. Xiao, P. Zhou and M. Jaroniec, J. Am. Chem. Soc., 2014, 136, 8839; (b) M. Wang, J. Liu, N. Cevey-Ha, S. Moon, P. Liska, R. Humphry-Baker, J. Moser, C. Grätzel, P. Wang, S. Zakeeruddin and M. Grätzel, Nano Today, 2010, 5, 169; (c) M. Wang, S. Moon, D. Zhou, F. Formal, N. Cevey-Ha, R. Humphry-Baker, C. Grätzel, P. Wang, S. Zakeeruddin and M. Grätzel, Adv. Funct. Mater., 2010, 20, 1821; (d) M. Wang, S. Moon, M. Xu, K. Chittibabu, P. Wang, N. Cevey-Ha, R. Humphry-Baker, S. Zakeeruddin and M. Grätzel, Small, 2010, 6, 319.
36. J. Burschka, N. Pellet, S. Moon, R. Humphry-Baker, P. Gao, K. Nazeeruddin and M. Grätzel, Nature, 2013, 499, 316.
37. J. Im, I. Jang, N. Pellet, M. Grätzel and N. Park, Nat. Nanotechnol., 2014, 9, 927.
38. W. Yang, J. Noh, N. Jeon, Y. Kim, S. Ryu, J. Seo and S. Seok, Science, 2015, 348, 1234.
39. G. Eperon, V. Burlakov, P. Docampo, A. Goriely and H. Snaith, Adv. Funct. Mater., 2014, 24, 151.
40. B. Conings, L. Baeten, C. Dobbelaere, J. Haen, J. Manca and H. Boyen, Adv. Mater., 2014, 26, 2041.
41. M. Liu, M. Johnston and H. Snaith, Nature, 2013, 501, 395.
42. L. Zuo, Z. Gu, T. Ye, W. Fu, G. Wu, H. Li and H. Chen, J. Am. Chem. Soc., 2015, 137, 2674.
43. H. Zhang, H. Azimi, Y. Hou, T. Ameri, T. Przybilla, E. Spiecker, M. Kraft, U. Scherf and C. Brabec, Chem. Mater., 2014, 26, 5190.
44. J. Min, Z. Zhang, Y. Hou, C. Quiroz, T. Przybilla, C. Bronnbauer, F. Guo, K. Forberich, H. Azimi, T. Ameri, H. Spiecker, Y. Li and C. Brabec, Chem. Mater., 2015, 27, 227.
45. H. Zhou, Q. Chen, G. Li, S. Luo, T. Song, H. Duan, Z. Hong, J. You, Y. Liu and Y. Yang, Science, 2014, 345, 542.
46. D. Li, J. Cui, H. Li, D. Huang, M. Wang and Y. Shen, Sol. Energy, 2016, 131, 176.
47. D. Li, J. Cui, H. Zhang, H. Li, M. Wang and Y. Shen, Chem. Lett., 2016, 45, 89.
48. C. Sun, Z. Wu, H. Yip, H. Zhang, X. Jiang, Q. Xue, Z. Hu, Z. Hu, Y. Shen, M. Wang, F. Huang and Y. Cao, Adv. Energy Mater., 2016, 6, 1501534.
49. J. Jeng, Y. Chiang, M. Lee, S. Peng, T. Guo, P. Chen and T. Wen, Adv. Mater., 2013, 25, 3727.
50. Q. Wang, Y. Shao, Q. Dong, Z. Xiao, Y. Yuan and J. Huang, Energy Environ. Sci., 2014, 7, 2359.
51. 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.
52. S. Ye, W. Sun, Y. Li, W. Yan, H. Peng, Z. Bian, Z. Liu and C. Huang, Nano Lett., 2015, 15, 3723.
53. A. Subbiah, A. Halder, S. Ghosh, N. Mahuli, G. Hodes and S. Sarkar, J. Phys. Chem. Lett., 2014, 5, 1748.
54. J. Cui, F. Meng, H. Zhang, K. Cao, H. Yuan, Y. Cheng, F. Huang and M. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 22862.
55. J. Jeng, K. Chen, T. Chiang, P. Lin, T. Tsai, Y. Chang, T. Guo, P. Chen, T. Wen and Y. Hsu, Adv. Mater., 2014, 26, 4107.
56. 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.
57. J. Kim, P. Liang, S. Williams, N. Cho, C. Chueh, M. Glaz, D. Ginger and A. Jen, Adv. Mater., 2015, 27, 695.
58. J. Cui, H. Yuan, J. Li, X. Xu, Y. Shen, H. Lin and M. Wang, Sci. Technol. Adv. Mater., 2015, 16, 36004.
59. P. Docampo, J. Ball, M. Darwich, G. Eperon and H. Snaith, Nat. Commun., 2013, 4, 2761.
60. X. Yin, M. Que, Y. Xing and W. Que, J. Mater. Chem. A, 2015, 3, 24495.
61. J. Park, J. Seo, S. Park, S. Shin, Y. Kim, N. Jeon, H. Shin, T. Ahn, J. Noh, S. Yoon, C. Hwang and S. Seok, Adv. Mater., 2015, 27, 4013.
62. G. Alemu, J. Li, J. Cui, X. Xu, B. Zhang, K. Cao, Y. Shen, Y. Cheng and M. Wang, J. Mater. Chem. A, 2015, 3, 9216.
63. 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.
64. K. Cao, Z. Zuo, J. Cui, Y. Shen, T. Moehl, S. Zakeeruddin, M. Grätzel and M. Wang, Nano Energy, 2015, 17, 171.
65. G. Niu, X. Guo and L. Wang, J. Mater. Chem. A, 2015, 3, 8970.
66. J. Christians, P. Herrera and P. Kamat, J. Am. Chem. Soc., 2015, 137, 1530.
67. J. Yang, B. Siempelkamp, D. Liu and T. Kelly, ACS Nano, 2015, 9, 1955.
68. A. Guerrero, J. You, C. Aranda, Y. Kang, G. Garcia-Belmonte, H. Zhou, J. Bisquert and Y. Yang, ACS Nano, 2016, 10, 218.
69. Y. Kato, L. Ono, M. Lee, S. Wang, S. Raga and Y. Qi, Adv. Mater. Interfaces, 2015, 2, 1500195.
70. H. Back, G. Kim, J. Kim, J. Kong, T. Kim, H. Kang, H. Kim, J. Lee, S. Lee and K. Lee, Energy Environ. Sci., 2016, 9, 1258.
71. Y. Han, S. Meyer, Y. Dkhissi, K. Weber, J. Pringle, U. Bach, L. Spiccia and Y. Cheng, J. Mater. Chem. A, 2015, 3, 8139.
72. Z. Xiao, Q. Dong, C. Bi, Y. Shao, Y. Yuan and J. Huang, Adv. Mater., 2014, 26, 6503.
73. Q. Wang, Y. Shao, Q. Dong, Z. Xiao, Y. Yuan and J. Huang, Energy Environ. Sci., 2014, 7, 2359.
74. H. Hsua, C. Chen, J. Changc, Y. Yu and Y. Shen, Nanoscale, 2014, 6, 10281.
75. S. Aharon, A. Dymshits, A. Rotem and L. Etgar, J. Mater. Chem. A, 2015, 3, 9171.
76. A. Amat, E. Mosconi, E. Ronca, C. Quarti, P. Umari, M. Nazeeruddin, M. Grätzel and F. Angelis, Nano Lett., 2014, 14, 3608.
77. A. Poglitsch and D. Weber, J. Chem. Phys., 1987, 87, 6373.
78. C. Stoumpos, C. Malliakas and M. Kanatzidis, Inorg. Chem., 2013, 52, 9019.
79. H. Zhang, X. Qiao, Y. Shen, T. Moehl, S. Zakeeruddin, M. Grätzel and M. Wang, J. Mater. Chem. A, 2015, 3, 11762.
80. A. Pisoni, J. Jaćimović, O. Barišić, M. Spina, R. Gaál, L. Forró and E. Horváth, J. Phys. Chem. Lett., 2014, 5, 2488.
81. S. Dharani, H. Mulmudi, N. Yantara, P. Trang, N. Park, M. Grätzel, S. Mhaisalkar, N. Mathews and P. Boix, Nanoscale, 2014, 6, 1675.
82. H. Kim, J. Lee, N. Yantara, P. Boix, S. Kulkarni, S. Mhaisalkar, M. Grätzel and N. Park, Nano Lett., 2013, 13, 2412.
83. K. Cao, J. Cui, H. Zhang, H. Li, J. Song, Y. Shen, Y. Cheng and M. Wang, J. Mater. Chem. A, 2015, 3, 9116.
84. R. Misra, S. Aharon, B. Li, D. Mogilyansky, I. Visoly-Fisher, L. Etgar and E. Katz, J. Phys. Chem. Lett., 2015, 6, 326.
85. S. Guarnera, A. Abate, W. Zhang, J. Foster, G. Richardson, A. Petrozz and H. Snaith, J. Phys. Chem. Lett., 2015, 6, 432.
86. K. Schwanitz, U. Weiler, R. Hunger, T. Mayer and W. Jaegermann, J. Phys. Chem. C, 2007, 111, 849.
87. X. Zhang, B. Zhang, Z. Zuo, M. Wang and Y. Shen, J. Mater. Chem. A, 2015, 3, 10020.
88. X. Zhang, B. Zhang, K. Cao, J. Brillet, J. Chen, M. Wang and Y. Shen, J. Mater. Chem. A, 2015, 3, 21630.
89. S. Habisreutinger, L. Schmidt-Mende and J. Stolarczyk, Angew. Chem., Int. Ed., 2013, 52, 7372.
90. K. Matsumoto, T. Makino, T. Ebara and J. Mizuguchi, J. Chem. Eng. Jpn., 2008, 41, 51.
91. Y. Yuan and J. Huang, Acc. Chem. Res., 2016, 49, 286.
92. E. Hoke, D. Slotcavage, E. Dohner, A. Bowring, H. Karunadasa and M. McGehee, Chem. Sci., 2015, 6, 613.
93. Z. Xiao, Y. Yuan, Y. Shao, Q. Wang, Q. Dong, C. Bi, P. Sharma, A. Gruverman and J. Huang, Nat. Mater., 2014, 14, 193.
94. J. Christians, R. Fung and P. Kamat, J. Am. Chem. Soc., 2014, 136, 758.
95. L. Zhang, X. Zhang, Z. Yin, Q. Jiang, X. Liu, J. Meng, Y. Zhao and H. Wang, J. Mater. Chem. A, 2015, 3, 12133.
96. 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.
97. S. Ma, H. Zhang, N. Zhao, Y. Cheng, M. Wang, Y. Shen and G. Tu, J. Mater. Chem. A, 2015, 3, 12139.
98. H. Li, K. Fu, A. Hagfeldt, M. Grätzel, S. Mhaisalkar and A. Grimsdale, Angew. Chem., Int. Ed., 2014, 53, 4085.
99. 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.
100. P. Qin, S. Tanaka, S. Ito, N. Tetreault, K. Manabe, H. Nishino, M. Nazeeruddin and M. Grätzel, Nat. Commun., 2014, 5, 3834.
101. T. Leijtens, J. Lim, J. Teuscher, T. Park and H. Snaith, Adv. Mater., 2013, 25, 3227.
102. B. Cai, Y. Xing, Z. Yang, W. Zhang and J. Qiu, Energy Environ. Sci., 2013, 6, 1480.
103. J. Burschka, A. Dualeh, F. Kessler, E. Baranoff, N. Cevey-Ha, C. Yi, M. Nazeeruddin and M. Grätzel, J. Am. Chem. Soc., 2011, 133, 18042.
104. A. Abate, T. Leijtens, S. Pathak, J. Teuscher, R. Avolio, M. Errico, J. Kirkpatrik, J. Ball, P. Docampo, I. Phersonc and H. Snaith, Phys. Chem. Chem. Phys., 2013, 15, 2572.
105. G. Niu, W. Li, F. Meng, L. Wang, H. Dong and Y. Qiu, J. Mater. Chem. A, 2014, 2, 705.
106. Q. Xue, Z. Hu, J. Liu, J. Lin, C. Sun, Z. Chen, C. Duan, J. Wang, C. Liao, W. Lau, F. Huang, H. Yip and Y. Cao, J. Mater. Chem. A, 2014, 2, 19598.
107. H. Choi, C. Mai, H. Kim, J. Jeong, S. Song, G. Bazan, J. Kim and A. Heeger, Nat. Commun., 2015, 6, 7348.
108. Q. Bao, X. Liu, S. Braun and M. Fahlman, Adv. Energy Mater., 2014, 4, 1301272.
109. S. Dupont, F. Novoa, E. Voroshazi and R. Dauskardt, Adv. Funct. Mater., 2014, 24, 1325.
110. J. Noh, S. Im, J. Heo, T. Mandal and S. Seok, Nano Lett., 2013, 13, 1764.
111. G. Eperon, S. Stranks, C. Menelaou, M. Johnston, L. Herz and H. Snaith, Energy Environ. Sci., 2014, 7, 982.
112. T. Koh, K. Fu, Y. Fang, S. Chen, T. Sum, N. Mathews, S. Mhaisalkar, P. Boix and T. Baikie, J. Phys. Chem. C, 2014, 118, 16458.
113. A. Binek, F. Hanusch, P. Docampo and T. Bein, J. Phys. Chem. Lett., 2015, 6, 1249.
114. Z. Li, M. Yang, J. Park, S. Wei, J. Berry and K. Zhu, Chem. Mater., 2016, 28, 284.
115. G. Eperon, G. Paterno, R. Sutton, A. Zampetti, A. Haghighirad, F. Caciallib and H. Snaith, J. Mater. Chem. A, 2015, 3, 19688.
116. Z. Chen, J. Wang, Y. Ren, C. Yu and K. Shum, Appl. Phys. Lett., 2012, 101, 093901.
117. M. Kumar, S. Dharani, W. Leong, P. Boix, R. Prabhakar, T. Baikie, C. Shi, H. Ding, R. Ramesh, M. Asta, M. Grätzel, S. Mhaisalkar and N. Mathews, Adv. Mater., 2014, 26, 7122.
118. M. Kulbak, S. Gupta, N. Kedem, I. Levine, T. Bendikov, G. Hodes and D. Cahen, J. Phys. Chem. Lett., 2016, 7, 167.
119. M. Kulbak, D. Cahen and G. Hodes, J. Phys. Chem. Lett., 2015, 6, 2452.
120. T. Koh, V. Shanmugam, J. Schlipf, L. Oesinghaus, P. Müller-Buschbaum, N. Ramakrishnan, V. Swamy, N. Mathews, P. Boix and S. Mhaisalkar, Adv. Mater., 2016, 28, 3653.
121. H. Tsai, W. Nie, J. Blancon, C. Stoumpos, R. Asadpour, B. Harutyunyan, A. Neukirch, R. Verduzco, J. Crochet, S. Tretiak, L. Pedesseau, J. Even, M. Alam, G. Gupta, J. Lou, P. Ajayan, M. Bedzyk, M. Kanatzidis and A. Mohite, Nature, 2016, 536, 312.
122. I. Smith, E. Hoke, D. Solis-Ibarra, M. McGehee and H. Karunadasa, Angew. Chem., 2014, 126, 11414.
123. D. Mitzi, J. Mater. Chem., 2004, 14, 2355.
124. G. Gong, N. Zhao, D. Ni, J. Chen, Y. Shen, M. Wang and G. Tu, J. Mater. Chem. A, 2016, 4, 3661.
125. S. Habisreutinger, T. Leijtens, G. Eperon, S. Stranks, R. Nicholas and H. Snaith, Nano Lett., 2014, 14, 5561.
126. L. Zheng, Y. Chung, Y. Ma, L. Zhang, L. Xiao, Z. Chen, S. Wang, B. Qu and Q. Gong, Chem. Commun., 2014, 50, 11196.
127. J. Cao, Y. Liu, X. Jing, J. Yin, J. Li, B. Xu, Y. Tan and N. Zheng, J. Am. Chem. Soc., 2015, 137, 10914.
128. L. Etgar, P. Gao, Z. Xue, Q. Peng, A. Chandiran, B. Liu, M. Nazeeruddin and M. Grätzel, J. Am. Chem. Soc., 2012, 134, 17396.
129. Z. Wei, H. Chen, K. Yan, X. Zheng and S. Yang, J. Mater. Chem. A, 2015, 3, 24226.
130. Z. Liu, M. Zhang, X. Xu, L. Bu, W. Zhang, W. Li, Z. Zhao, M. Wang and H. He, Dalton Trans., 2015, 44, 3967.
131. Z. Liu, M. Zhang, X. Xu, F. Cai, H. Yuan, L. Bu, W. Li, A. Zhu, Z. Zhao, M. Wang, Y. Cheng and H. He, J. Mater. Chem. A, 2015, 3, 24121.
132. X. Xu, Z. Liu, Z. Zuo, M. Zhang, Z. Zhao, Y. Shen, H. Zhou, Q. Chen, Y. Yang and M. Wang, Nano Lett., 2015, 15, 2402.
133. S. Stranks, G. Eperon, G. Grancini, C. Menelaou, M. Alcocer, T. Leijtens, L. Herz, A. Petrozza and H. Snaith, Science, 2013, 342, 341.

---