Probing the origins of photodegradation in organic–inorganic metal halide perovskites with time-resolved mass spectrometry

Abstract

Operational and long-term stability of perovskite solar cells are critical for their commercialization on a large scale. To mitigate stability issues, a fundamental understanding of the physicochemical processes associated with the degradation of perovskite materials is needed. Here, we perform time resolved mass spectrometry of the gas species evolved during the photoinduced degradation of organic–inorganic lead trihalide perovskites made with commonly used monovalent cations, including methylammonium (MA), formamidinium (FA), and Cs. Our results indicate that the hot-carrier-induced deprotonation of MA⁺ cations is the fundamental origin of the photodegradation, which inevitably leads to the release of volatile species such as ammonia (NH₃), aminocarbyne fragments (CNH₂), hydrogen (H₂), and iodine/hydrogen iodide (I/HI) from methylammonium lead iodide (MAPbI₃) at different rates under simulated one sun solar illumination. Photodegradation processes can be mitigated by applying ultra-violet (UV) filters with suitable cutoff wavelengths to the light source. Additionally, we demonstrate that the incorporation of FA reduces the release of organic species but does not prevent the formation of I/HI. However, the addition of Cs effectively suppresses the release of all volatile gases. The best photostability is obtained with the FA/Cs mixed perovskites, showing that the complete removal of MA from mixed-cation perovskites is preferred for more photostable perovskites.

---

Introduction

In the past few years, organic–inorganic metal halide perovskites have emerged as an exciting class of photovoltaic (PV) materials.^1,2 With high certified power conversion efficiencies that exceed 23.3% (ref. 3) and low-cost solution-based fabrication processes, perovskite solar cells (PVSCs) are projected to rival other more established PV technologies on the market, especially in the aspects of levelized cost of energy (LCOE),⁴ life-cycle environmental impact,^5,6 and energy payback time (EPBT).^7,8 Recent progress in large-area deposition techniques,^9–11 combined with advances in materials engineering to improve device performance and durability,^12–17 makes PVSCs attractive for more practical, commercial applications. However, major obstacles remain before high volume deployment of perovskite PV modules becomes feasible.¹⁸ The most critical one is the lack of operational and long-term stability. To be commercially viable, PVSCs need to retain robust stability throughout their lifetimes. The fact that most established PV companies provide a 25-year warranty for their products sets a high bar for perovskite PV technology to enter the energy production market.¹⁹

Over the last decade, lifetimes of PVSCs reported in the literature have extended from a few hours to over a year.²⁰ However, most high efficiency perovskite solar cells based on triple-cation (methylammonium, MA; formamidinium, FA; Cs) lead trihalide perovskites exhibit a certain degree of degradation under continuous operating conditions.²¹ In an attempt to understand and eliminate degradation, the origins and processes of device degradation are being intensively investigated.²² Degradation mechanisms of metal halide perovskite materials and devices under different environmental conditions have been extensively explored.^23–35 In general, a variety of extrinsic factors, including light,^36–42 heat,^43–45 atmosphere (e.g., oxygen and moisture),^42,46–48 and electric field,^49,50 are considered to contribute to the degradation of perovskite materials and devices. Among them, heat, moisture, and electric field related degradation can be mitigated by employing appropriate encapsulation and interface protection,^16,51–53 while solar radiation that contains a substantial quantity of energetic photons is an unavoidable stress during the operation of a solar cell.

A recent literature review that surveyed 50 publications shows that the stability of PVSCs is strongly reduced under illumination,³⁶ which hinders practical application of PVSCs. To mitigate this photoinstability issue, a fundamental understanding of the physicochemical processes associated with the photodegradation of perovskite materials is needed. Thus far, only a few studies have investigated the degradation processes of perovskite films under illumination via X-ray photoelectron, diffraction, and absorption spectroscopies,^40–42 with an emphasis on monitoring the chemical and structural changes in the perovskite films. Despite these efforts, little is known about the formation and escape rates of the volatile decomposition products during the photodegradation processes. Most recently, Nickel et al. reported an investigation on the photodegradation of MAPbI₃ by analyzing the residual gas composition above the sample,³⁹ providing insights for understanding the photodegradation mechanisms. However, the measurements of gas effusion were limited to a few selected low atomic weight (<32) gas molecules, and volatile species across the full spectrum of the photodegradation products, especially iodine or iodide vapors, have not yet been monitored. The missing information may hinder accurate understanding of the degradation mechanism and underestimate the impact of MA on the stability of state-of-the-art triple-cation (FA/MA/Cs) perovskites, which have demonstrated much better photostability than mixed FA/MA perovskites.^13,14,54 Therefore, more detailed work is needed to develop a comprehensive understanding of the underlying degradation mechanisms and pathways.⁵⁵

Herein, we address this need by employing high-resolution time resolved mass spectrometry (MS) to monitor the gas emission during the photodegradation of organic–inorganic halide perovskites in vacuum and under different illumination conditions. Starting with the archetypal perovskite, methylammonium lead iodide (CH₃NH₃PbI₃ or MAPbI₃), we measure the escape rates of volatile photodecomposition products and identify the photon energy threshold for the photodegradation of MAPbI₃ using a filtered light source. For the first time, we observed that MA gas is released faster than HI under illumination, indicating that photodegradation likely originates from the hot carrier-induced deprotonation of MA⁺ cations. Moreover, we study the photostability of a series of FA/MA/Cs mixed-cation perovskites, demonstrating that the incorporation of FA does not fully stop the photodegradation, but the incorporation of nonvolatile Cs⁺ into perovskite films leads to improved photostability. The results show that the complete removal of MA from mixed-cation perovskites is preferred for more photostable perovskites.

Results and discussion

We start our investigation with the archetypal perovskite material, i.e., MAPbI₃. An in-house-built temperature programmed MS spectroscopy system⁵⁶ (Fig. S1†) was used to determine the chemical compositions of the gases released from MAPbI₃ under simulated solar irradiation. Fig. 1a shows the evolution of the mass-to-charge ratio (m/z) spectrum recorded as a function of light exposure time (see the Experimental section for details). A variety of gas species including H₂ (m/z = 2), NH₃ (m/z = 17), H₂O (m/z = 18), CH₃NH₂ (m/z = 28–31), HI (m/z = 64 or 128), and I₂ (m/z = 127) were detected. By measuring these partial pressures as a function of time, the photodegradation dynamics of MAPbI₃ were determined (vide infra). A comparison of the relevant peaks at the initial stage and at a time during the photodegradation is shown in Fig. 1b–f. The significant increase in the partial pressures of the gas signals clearly indicates the presence of the volatile decomposition products through photodecomposition of MAPbI₃. It is worth noting that MS signals are comprised of the fragments of molecules due to the electron bombardment during the ionization process in the quadrupole ion analyzer. Multiple fragments from a gas can be detected, and one fragment species may originate from different parent molecules. Because of this, the chemical compositions revealed by MS are manifold, and chemical complexity needs to be considered while analyzing the data.

Fig. 1 (a) Evolution of the mass spectrum (m/z) of the gases released during the photoinduced decomposition of CH₃NH₃PbI₃. Comparison of the characteristic m/z peaks of the pristine and degraded samples, including (b) H⁺, (c) NH₃⁺/H₂O⁺, (d) CNH₂⁺/CH₂NH₂⁺/CH₃NH₂⁺, (e) HI²⁺, and (f) I⁺/HI⁺.

Turning to a detailed analysis of the photodegradation process, the release of H₂O was initiated after MAPbI₃ was exposed to light for ∼5 min (300 s). Due to the hygroscopic nature of perovskites,⁴⁶ water molecules could either be adsorbed on the surface of the perovskite⁵⁷ or infiltrate into grain boundaries to form the hydrated perovskite phases.⁵⁸ The observation of photo-induced desorption of water indicates the high physisorption affinity of water molecules to perovskites, which leads to substantial water adsorption even for a short duration of exposure (2 to 3 min) to the air during the sample transfer. However, the amount of water desorbed from the perovskite is challenging to quantify because water contributes significantly to the base pressure of a typical high vacuum (∼10⁻⁷ torr) chamber (Fig. S2†), and its partial pressure is sensitive to subtle chamber condition changes and degassing of other gas species. Therefore, water is not used as the primary indicator for the photodegradation analysis in this study.

The release of methylamine (CH₃NH₂ or MA) and HI signals commenced at ∼20 min (1200 s), signifying the beginning of photodegradation of MAPbI₃. The peaks observed at m/z = 2 (Fig. 1b) and m/z = 17 and 18 (Fig. 1c), corresponding to H₂ and NH₃/H₂O, respectively, could be directly produced by the photodecomposition of MAPbI₃ or from the ionization byproducts of MA. The released MA was characterized by three aminocarbyne fragments at m/z ratios of 28 (CNH₂), 30 (CH₂NH₂), and 31(CH₃NH₂), as shown in Fig. 1d, which is consistent with the reference MS of MA (Fig. S3†). The release of NH₃ and MA gas molecules can break the phase equilibrium homogeneity in MAPbI₃ and worsen the decomposition of the pristine perovskite phase by triggering a phase transfer from the pure corner-sharing Pb–I octahedral perovskite network into an inhomogeneous combination of MAPbI₃ perovskite, layered PbI₂, and low-dimensional perovskites.^59,60 This phase transition creates more vulnerable defect sites on the grain boundaries with loose bonding or dangling bonds that are prone to decomposition.

The release of HI gas was evident in the single and double charged HI with m/z ratios of 128 and 64, respectively (Fig. 1e and f), which was not found in the previous photodegradation study.³⁹ The signal of m/z = 127 could be attributed to HI as well as the single charged I or double charged I₂, indicating that an equilibrium between I₂ and I⁻ exists in the perovskite films during photodegradation. The condensation of I₂ residues on the inner wall of the sample conditioning tube (Fig. S4†) reveals the simultaneous formation of I₂ gas with HI, consistent with a recent report on measuring the optical absorption spectrum of I₂ release from MAPbI₃ in toluene under illumination.⁶¹ Both I₂ and HI are detrimental to the stability of perovskite films. Iodine vapor could induce severe degradation of MAPbI₃ due to chemical chain reactions.⁶² HI is a highly corrosive acid which could cause severe damage to the perovskite crystals upon contact and accelerate the degradation of perovskites.

The aforementioned results reveal that the MA derivatives and I₂/HI gases are the primary volatile products during the photodegradation of MAPbI₃. Interestingly, the photodegradation products of MAPbI₃ differ from the thermal degradation products in which CH₃I and NH₃ were detected at above 300 °C by MS measurements.⁴⁴ Although photo- and thermal degradation result in different products, there should be a chemical equilibrium between these gasses, which favors the formation of MA and HI at low temperatures (∼60 °C under illumination) but leads to the formation of CH₃I and NH₃ at high temperatures (>300 °C). Nonetheless, for both degradation pathways, the decomposition of perovskites is associated with the release of volatile gases, especially MA and HI. It is noteworthy that while some studies have demonstrated that photo-induced superoxide formed in the presence of oxygen or an oxide interface leads to the photodecomposition of perovskites,^37,63–65 the results here show that oxygen is not necessary for the photodecomposition since no oxygen or superoxide signals were detected.

To further elucidate the degradation mechanisms, we measured the photon energy dependences of the photodegradation of MAPbI₃. Long pass filters with various cutoff wavelengths were used to remove the high energy photons that are known to be particularly detrimental to PVSCs.^39,64Fig. 2 shows the evolution of the MS traces of MA and HI from MAPbI₃ films exposed to illumination with the spectra filtered with short wavelength photons being cut off at 280, 320, 370, 395, 420, and 455 nm. The transmittance spectra of the filters and full MS spectral data are shown in Fig. S5 and S6,† respectively.

Fig. 2 Mass spectroscopy traces of CH₃NH₂ (28–31) and HI (126–127) during the photodecomposition of MAPbI₃ films under illumination from a xenon lamp equipped with a long pass optical filter with a cutoff wavelength of (a) 280, (b) 320, (c) 370, (d) 395, (e) 420, and (f) 455 nm.

Insights into the dynamics and kinetics of photodegradation of MAPbI₃ can be obtained from analyzing the gas emission processes. First, the MS traces of MA and HI (Fig. 2a–d) show a general trend of an increase in MA and HI partial pressures to the maxima followed by a decline to a residual level, indicating that the photodegradation of MAPbI₃ is a self-limiting process that likely initiates on the surface of the film or at grain boundaries, reaches the pressure maxima when most surface bonds are cleaved, and then slows down due to the formation of more photostable PbI₂ on the surface. These findings are consistent with the report that photodegradation of MAPbI₃ became saturated after the decomposition of the surface layer.³⁸ Beyond that, the escape rates of the residual gas determined by the time derivative of the partial pressure decreased with increasing the cutoff wavelength (Fig. S7†), likely due to the retarded photodegradation caused by removing highly energetic photons. The residual gas signals were in the minimum detectable range and were negligible when the 420 (Fig. 2e) and 455 nm (Fig. 2f) filters were applied, respectively, indicating that a minimum photon energy of 2.95 eV (∼420 nm) is needed for triggering light-induced degradation of MAPbI₃. Considering the bandgap of 1.55 eV for MAPbI₃,⁴⁵ the excess kinetic energy of photoexcited charge carriers (∼1.4 eV) is close to the energy associated with the deprotonation of MA⁺ cations (1.49 eV) reported in the literature.⁶⁶ It is worth noting that our result is more accurate than the minimum energy of 2.72 eV required for the light-induced dissociation of MAPbI₃ perovskite reported by Nickel et al.³⁹ To verify that the mitigation of photodecomposition is merely due to removal of high energy photons rather than reduction of light intensity, we performed the MS measurements under reduced illumination intensities (25%) achieved by using a neutral density filter with an optical density (OD) of 0.6. Fig. S8† shows evidence of photodecomposition even under this weak illumination condition (∼25 mW cm⁻²), confirming that the high energy photons are responsible for the photodegradation of MAPbI₃.

The detailed analysis of in situ MS measurement data provides insights into the kinetics of the photodecomposition process. Interestingly, MA exhibits higher partial pressure than HI at the early stage of the photodecomposition process and reaches the peak prior to HI in most cases (Fig. 2a–d), indicating that MA is released faster than HI when photodegradation occurs. The result reveals that MA⁺, i.e., CH₃NH₃⁺ (resulting from the protonation of MA molecules), is more vulnerable to high energy photons than I⁻ and is more likely to be the origin of the photodegradation of the perovskite. It is worth noting that a clear initial spike in the MA signal was observed in some MAPbI₃ films (Fig. 2a–c), likely due to the existence of MA⁺ cations on film surfaces and in grain boundaries.⁶⁷ It is known that MAPbI₃ surfaces contain under-coordinated MA⁺ cations.⁶⁸ Photoexcitation could release those loosely bounded MA gas molecules by a deprotonation process, leading to the initial spike in the MA signal.

Applying a filter to eliminate UV photons is an effective strategy to improve the photostability of perovskite materials and devices. Additionally, when the perovskite films are fully encapsulated, the volatile photodecomposition products cannot escape and thus persist in the films, retaining the perovskite phase. However, these approaches do not solve the intrinsic instability issue of MAPbI₃. Alternatively, many studies suggest that incorporation of less volatile large organic (e.g., FA) and monovalent metal cations (e.g., Cs⁺) can significantly enhance photostability.^1,14,69 To explore the effect of the A-site cation on the photostability of perovskite materials, we performed MS measurements on a variety of single- and mixed-cation perovskites. Fig. 3a shows the accumulated MS signals of organic species (i.e. MA and FA derivatives) and HI for perovskite films with various (Cs_xFA_yMA_1−x−y)PbI₃ compositions, including MAPbI₃, FAPbI₃, MA_0.7FA_0.3PbI₃, FA_0.95Cs_0.05PbI₃, FA_0.9Cs_0.1PbI₃, FA_0.8Cs_0.2PbI₃, and (MA_0.7FA_0.3)_0.8Cs_0.2PbI₃, after 7 h of light exposure. Detailed traces of the corresponding curves are shown in Fig. S9.† As expected, we found that MAPbI₃ had the most severe photodecomposition. Interestingly, gravimetric analysis shows that MAPbI₃ lost about ∼30% of its initial weight, more than the weight ratio (25.6%) of MAI (159 g mol⁻¹) to MAPbI₃ (620 g mol⁻¹), indicating the further decomposition of PbI₂ into metallic lead (Pb⁰) and I₂ gas.^41,70 FAPbI₃ shows significantly lower emission of organic species (mainly in terms of CNH and CNH₂) but comparable HI emission compared with MAPbI₃ (Fig. 3b). Although improved photostability of FAPbI₃ was reported,⁶⁹ the significant release of HI could still fundamentally limit the durability of FAPbI₃. In contrast, incorporating Cs into FAPbI₃ significantly suppressed the photodecomposition of perovskite materials. The photostability of the perovskite increases with increasing Cs proportion, with FA_0.8Cs_0.2PbI₃ showing two orders of magnitude less loss of materials during continuous illumination (Fig. 3c). Interestingly, the addition of MA to the mixed-cation perovskites (FA/MA and FA/MA/Cs) shows worse photostability than that of their pure FA and FA/Cs counterparts, indicating that the UV-photon induced degradation of the perovskites is mainly attributed to MA cations.

Fig. 3 (a) Total amounts of organic species and HI gases released from various perovskite films, including MAPbI₃, FAPbI₃, MA_0.7FA_0.3PbI₃, FA_0.95Cs_0.05PbI₃, FA_0.9Cs_0.1PbI₃, FA_0.8Cs_0.2PbI₃, and (MA_0.7FA_0.3)_0.8Cs_0.2PbI₃ under a simulated irradiance of 100 mW cm⁻² for 7 h. Time-resolved mass spectra of (b) FAPbI₃ and (c) FA_0.8Cs_0.2PbI₃ under a simulated irradiance of 100 mW cm⁻².

The above observation leads us to hypothesize that the photodegradation mechanism of MA-containing perovskites proceeds through a sequence of steps involving (1) photoexcitation of hot carriers; (2) deprotonation of MA cations and formation of MA gas near surface or grain boundary regions; and (3) formation of HI gas and iodine vacancies (V_I) (deterioration of the PbI₆ octahedron), as described by eqn (1)–(3) and shown in Fig. 4.

hν(>3 eV) → e⁻(hot) + h⁺(hot) (1)

CH₃NH₃⁺ + e⁻(hot) → CH₃NH₂(g) + H⁺ + e⁻ (2)

H⁺ + I⁻ → HI(g) + V_I (3)

Fig. 4 Schematic representation of the photodegradation mechanism. (a) Hot carrier generation, (b) deprotonation and release of MA gas, and (c) formation of HI gas and V_I.

The generation of long-lived hot electrons is a requisite for the observed photodegradation of MAPbI₃ (eqn (1) and Fig. 4a). Recent studies reported that hot carriers in perovskites have a relatively long lifetime (∼100 ps) and can migrate a long distance (up to 600 nm) in perovskite crystals.^71–73 A hot electron may interact with a MA cation via coulombic coupling. The hot electron may transfer its excess energy to the MA cation and deprotonate it, releasing free protons and volatile MA gas (eqn (2) and Fig. 4b). The free protons then pair with under-coordinated I atoms on the surface, forming HI gas and iodine vacancies (i.e., deteriorated PbI₆ octahedra) (eqn (3) and Fig. 4c). We speculate that HI is the most detrimental species to the stability of the perovskite because the continuous loss of HI can cause the breakdown of the Pb–I framework via an equilibrium phase transfer from the corner-shared to the face-shared Pb–I octahedra, leading to the catastrophic decomposition of MAPbI₃ into PbI₂. The loss of HI completely changes the morphology of MAPbI₃ films, turning the compact and smooth perovskite films into discrete PbI₂ crystals (Fig. S10†). Although the primary gas products of the photodegradation are MA and HI, other gas species, such as H₂, NH₃, and I₂, may coexist via a photochemical equilibrium in the atmosphere. Nonetheless, preventing the formation of volatile gases could be an effective strategy to mitigate the photoinstability issue of perovskites.

The mechanism of improved stability of Cs containing perovskites is attributed to the low volatility of Cs. Due to its small ionic size (167 pm) compared with MA⁺ (217 pm) and FA⁺ (253 pm),⁷⁴ it is reasonable to expect Cs⁺ ions to segregate into surface and grain boundaries, a process that reduces strain and, therefore, is energetically favorable. As discussed earlier, organic MA and FA cations tend to deprotonate and vaporize after receiving energy from hot carriers generated by high energetic photons. In contrast, Cs has much lower vapor pressures and has no proton. Preserving monovalent cations limits the formation and release of HI gas, locking iodine atoms in the perovskite framework. Therefore, the intrinsic decomposition of iodine-based perovskites can be eliminated, showing potential for long-term stability. However, the triple cation (MA_0.7FA_0.3)_0.8Cs_0.2PbI₃, unlike its FA/Cs counterpart, still released substantial amounts of organic species and HI gases under continuous illumination (Fig. 3), indicating the intrinsic instability of perovskite materials incorporating MA cations. To overcome the instability issue, the MA content in the perovskites should be reduced or eliminated.

Conclusion

In summary, we have investigated the photodecomposition of MAPbI₃ under different illumination conditions using MS. We demonstrated that photodecomposition is a self-limited process, identified the photon energy threshold of ∼3 eV to trigger the decomposition of MAPbI₃, and showed that the release of MA is prior to HI. Lastly, we found that incorporating Cs and removing MA can improve the photostability of perovskite materials. Our findings suggest that compositional engineering of perovskite materials and UV filtering can prevent the photodegradation of the organic components and thus increase the operational and long-term stability of organic–inorganic halide perovskite solar cells.

Experimental

Synthesis of perovskite thin films

The perovskite films were synthesized following a previously published method. Details are included in the ESI.†

Time-resolved mass spectroscopy measurements

The mass spectra of perovskite films were obtained using a custom dynamically pumped temperature programmed desorption system. For each degradation measurement, a perovskite film (1 inch by 0.25 inch) was placed in a quartz tube that was connected to the primary chamber of the system. The system was pumped down for ∼20 h to allow degassing of the specimen prior to the photon induced degradation measurement. The chamber was evacuated to <10⁻⁷ torr before each measurement. The degradation process was monitored using a quadrupole mass spectrometer (Stanford Research System, RGA 300). The mass spectra with a range of 1 to 150 atomic mass units (AMU) were recorded at 5 s intervals. Prior to the illumination, background signals were first collected for 10 min in the dark and the averages were subtracted from the main signals. After that, the sample was exposed to a simulated solar irradiance of ∼100 mW cm⁻² generated by a xenon arc lamp. For the spectroscopy-dependent measurements, long wavelength filters with the cutoff edge at 280, 320, 370, 395, 420, or 455 nm were used to eliminate the influence of high energy photons.

Author contribution

Z. S. and Y. F. Y. conceived the project. Z. S. carried out the MS measurements. C. W., Y. Y., C. C., X. Y. and C. L. synthesized the perovskite thin films and helped with characterization. C. R. G., A. B. P. and M. J. H. assisted with the MS measurements. Z. S. and Y. F. Y. analyzed the data and wrote the manuscript. D. Z., A. B. P., R. J. E. and M. J. H. helped with the manuscript preparation. All the authors discussed the results and commented on the manuscript. Y. F. Y. supervised the project.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

This work is financially supported by the Office of Naval Research under contract no. N00014-17-1-2223, the Air Force Research Laboratory, Space Vehicles Directorate (contract # FA9453-11-C-0253), the National Science Foundation under contract no. CHE-1230246 and DMR-1534686, the U.S. Department of Energy (DOE) SunShot Initiative under the Next Generation Photovoltaics 3 program (DE-FOA-0000990), and the Ohio Research Scholar Program.

References

1. J.-P. Correa-Baena, M. Saliba, T. Buonassisi, M. Grätzel, A. Abate, W. Tress and A. Hagfeldt, Science, 2017, 358, 739–744.
2. J. J. Berry, J. van de Lagemaat, M. M. Al-Jassim, S. Kurtz, Y. Yan and K. Zhu, ACS Energy Lett., 2017, 2, 2540–2544.
3. NREL, Solar Cell Efficiency Chart, https://www.nrel.gov/pv/assets/images/efficiency-chart-20180716.jpg, accessed 6 August 2018.
4. Z. Song, C. L. McElvany, A. B. Phillips, I. Celik, P. W. Krantz, S. C. Watthage, G. K. Liyanage, D. Apul and M. J. Heben, Energy Environ. Sci., 2017, 10, 1297–1305.
5. J. Gong, S. B. Darling and F. You, Energy Environ. Sci., 2015, 8, 1953–1968.
6. I. Celik, Z. Song, A. J. Cimaroli, Y. Yan, M. J. Heben and D. Apul, Sol. Energy Mater. Sol. Cells, 2016, 156, 157–169.
7. I. Celik, A. B. Phillips, Z. Song, Y. Yan, R. J. Ellingson, M. J. Heben and D. Apul, Energy Environ. Sci., 2017, 10, 1874–1884.
8. I. Celik, A. B. Phillips, Z. Song, Y. Yan, R. J. Ellingson, M. J. Heben and D. Apul, IEEE J. Photovolt., 2017, 8, 305–309.
9. Z. Li, T. R. Klein, D. H. Kim, M. Yang, J. J. Berry, M. F. A. M. van Hest and K. Zhu, Nat. Rev. Mater., 2018, 3, 18017.
10. H. Chen, F. Ye, W. Tang, J. He, M. Yin, Y. Wang, F. Xie, E. Bi, X. Yang, M. Grätzel and L. Han, Nature, 2017, 550, 92.
11. M. Yang, Z. Li, M. O. Reese, O. G. Reid, D. H. Kim, S. Siol, T. R. Klein, Y. Yan, J. J. Berry, M. F. A. M. van Hest and K. Zhu, Nat. Energy, 2017, 2, 17038.
12. M. Saliba, T. Matsui, K. Domanski, J.-Y. Seo, A. Ummadisingu, S. M. Zakeeruddin, J.-P. Correa-Baena, W. R. Tress, A. Abate, A. Hagfeldt and M. Grätzel, Science, 2016, 354, 206–209.
13. M. Saliba, T. Matsui, J.-Y. Seo, K. Domanski, J.-P. Correa-Baena, M. K. Nazeeruddin, S. M. Zakeeruddin, W. Tress, A. Abate, A. Hagfeldt and M. Gratzel, Energy Environ. Sci., 2016, 9, 1989–1997.
14. H. Tsai, R. Asadpour, J.-C. Blancon, C. C. Stoumpos, O. Durand, J. W. Strzalka, B. Chen, R. Verduzco, P. M. Ajayan, S. Tretiak, J. Even, M. A. Alam, M. G. Kanatzidis, W. Nie and A. D. Mohite, Science, 2018, 360, 67–70.
15. W. S. Yang, B.-W. Park, E. H. Jung, N. J. Jeon, Y. C. Kim, D. U. Lee, S. S. Shin, J. Seo, E. K. Kim, J. H. Noh and S. I. Seok, Science, 2017, 356, 1376–1379.
16. J. A. Christians, P. Schulz, J. S. Tinkham, T. H. Schloemer, S. P. Harvey, B. J. Tremolet de Villers, A. Sellinger, J. J. Berry and J. M. Luther, Nat. Energy, 2018, 3, 68–74.
17. M. Abdi-Jalebi, Z. Andaji-Garmaroudi, S. Cacovich, C. Stavrakas, B. Philippe, J. M. Richter, M. Alsari, E. P. Booker, E. M. Hutter, A. J. Pearson, S. Lilliu, T. J. Savenije, H. Rensmo, G. Divitini, C. Ducati, R. H. Friend and S. D. Stranks, Nature, 2018, 555, 497.
18. A. Abate, J. P. Correa-Baena, M. Saliba, M. S. Su'ait and F. Bella, Chem.–Eur. J., 2018, 24, 3083–3100.
19. M. A. Green, Nat. Energy, 2016, 1, 15015.
20. G. Grancini, C. Roldán-Carmona, I. Zimmermann, E. Mosconi, X. Lee, D. Martineau, S. Narbey, F. Oswald, F. De Angelis, M. Graetzel and M. K. Nazeeruddin, Nat. Commun., 2017, 8, 15684.
21. K. Domanski, E. A. Alharbi, A. Hagfeldt, M. Grätzel and W. Tress, Nat. Energy, 2018, 3, 61–67.
22. T. Leijtens, G. E. Eperon, N. K. Noel, S. N. Habisreutinger, A. Petrozza and H. J. Snaith, Adv. Energy Mater., 2015, 5, 1500963.
23. D. Wang, M. Wright, N. K. Elumalai and A. Uddin, Sol. Energy Mater. Sol. Cells, 2016, 147, 255–275.
24. M. I. Asghar, J. Zhang, H. Wang and P. D. Lund, Renewable Sustainable Energy Rev., 2017, 77, 131–146.
25. X. Guo, C. McCleese, C. Kolodziej, A. C. S. Samia, Y. Zhao and C. Burda, Dalton Trans., 2016, 45, 3806–3813.
26. Y. Yao, G. Wang, F. Wu, D. Liu, C. Lin, X. Rao, R. Wu, G. Zhou and Q. Song, RSC Adv., 2017, 7, 42973–42978.
27. Z. Li, C. Xiao, Y. Yang, S. P. Harvey, D. H. Kim, J. A. Christians, M. Yang, P. Schulz, S. U. Nanayakkara, C.-S. Jiang, J. M. Luther, J. J. Berry, M. C. Beard, M. M. Al-Jassim and K. Zhu, Energy Environ. Sci., 2017, 10, 1234–1242.
28. K. Domanski, B. Roose, T. Matsui, M. Saliba, S.-H. Turren-Cruz, J.-P. Correa-Baena, C. R. Carmona, G. Richardson, J. M. Foster, F. De Angelis, J. M. Ball, A. Petrozza, N. Mine, M. K. Nazeeruddin, W. Tress, M. Gratzel, U. Steiner, A. Hagfeldt and A. Abate, Energy Environ. Sci., 2017, 10, 604–613.
29. Z. Song, J. Werner, S. C. Watthage, F. Sahli, N. Shrestha, S. D. Wolf, B. Niesen, A. B. Phillips, C. Ballif, R. J. Ellingson and M. J. Heben, IEEE J. Photovolt., 2017, 7, 1563–1568.
30. N. H. Tiep, Z. Ku and H. J. Fan, Adv. Energy Mater., 2016, 6, 1501420.
31. E. Smecca, Y. Numata, I. Deretzis, G. Pellegrino, S. Boninelli, T. Miyasaka, A. La Magna and A. Alberti, Phys. Chem. Chem. Phys., 2016, 18, 13413–13422.
32. A. Gomez, S. Sanchez, M. Campoy-Quiles and A. Abate, Nano Energy, 2018, 45, 94–100.
33. N. N. Shlenskaya, N. A. Belich, M. Grätzel, E. A. Goodilin and A. B. Tarasov, J. Mater. Chem. A, 2018, 6, 1780–1786.
34. F. U. Kosasih and C. Ducati, Nano Energy, 2018, 47, 243–256.
35. Y.-H. Kye, C.-J. Yu, U.-G. Jong, Y. Chen and A. Walsh, J. Phys. Chem. Lett., 2018, 9, 2196–2201.
36. F. Lang, O. Shargaieva, V. V. Brus, H. C. Neitzert, J. Rappich and N. H. Nickel, Adv. Mater., 2018, 30, 1702905.
37. D. Bryant, N. Aristidou, S. Pont, I. Sanchez-Molina, T. Chotchunangatchaval, S. Wheeler, J. R. Durrant and S. A. Haque, Energy Environ. Sci., 2016, 9, 1655–1660.
38. Y. Li, X. Xu, C. Wang, B. Ecker, J. Yang, J. Huang and Y. Gao, J. Mater. Chem. C, 2017, 121, 3904–3910.
39. N. H. Nickel, F. Lang, V. V. Brus, O. Shargaieva and J. Rappich, Adv. Electron. Mater., 2017, 3, 1700158.
40. R.-P. Xu, Y.-Q. Li, T.-Y. Jin, Y.-Q. Liu, Q.-Y. Bao, C. O'Carroll and J.-X. Tang, ACS Appl. Mater. Interfaces, 2018, 10, 6737–6746.
41. X. Tang, M. Brandl, B. May, I. Levchuk, Y. Hou, M. Richter, H. Chen, S. Chen, S. Kahmann, A. Osvet, F. Maier, H.-P. Steinruck, R. Hock, G. J. Matt and C. J. Brabec, J. Mater. Chem. A, 2016, 4, 15896–15903.
42. B.-A. Chen, J.-T. Lin, N.-T. Suen, C.-W. Tsao, T.-C. Chu, Y.-Y. Hsu, T.-S. Chan, Y.-T. Chan, J.-S. Yang, C.-W. Chiu and H. M. Chen, ACS Energy Lett., 2017, 2, 342–348.
43. B. Conings, J. Drijkoningen, N. Gauquelin, A. Babayigit, J. D'Haen, L. D'Olieslaeger, A. Ethirajan, J. Verbeeck, J. Manca, E. Mosconi, F. D. Angelis and H.-G. Boyen, Adv. Energy Mater., 2015, 5, 1500477.
44. E. J. Juarez-Perez, Z. Hawash, S. R. Raga, L. K. Ono and Y. Qi, Energy Environ. Sci., 2016, 9, 3406–3410.
45. Z. Song, S. C. Watthage, A. B. Phillips, B. L. Tompkins, R. J. Ellingson and M. J. Heben, Chem. Mater., 2015, 27, 4612–4619.
46. A. M. A. Leguy, Y. Hu, M. Campoy-Quiles, M. I. Alonso, O. J. Weber, P. Azarhoosh, M. van Schilfgaarde, M. T. Weller, T. Bein, J. Nelson, P. Docampo and P. R. F. Barnes, Chem. Mater., 2015, 27, 3397–3407.
47. A. J. Pearson, G. E. Eperon, P. E. Hopkinson, S. N. Habisreutinger, J. T.-W. Wang, H. J. Snaith and N. C. Greenham, Adv. Energy Mater., 2016, 6, 1600014.
48. Z. Song, A. Abate, S. C. Watthage, G. K. Liyanage, A. B. Phillips, U. Steiner, M. Graetzel and M. J. Heben, Adv. Energy Mater., 2016, 6, 1600846.
49. T. Leijtens, E. T. Hoke, G. Grancini, D. J. Slotcavage, G. E. Eperon, J. M. Ball, M. De Bastiani, A. R. Bowring, N. Martino, K. Wojciechowski, M. D. McGehee, H. J. Snaith and A. Petrozza, Adv. Energy Mater., 2015, 5, 1500962.
50. Y. Yuan, Q. Wang, Y. Shao, H. Lu, T. Li, A. Gruverman and J. Huang, Adv. Energy Mater., 2016, 6, 1501803.
51. F. Bella, G. Griffini, J.-P. Correa-Baena, G. Saracco, M. Grätzel, A. Hagfeldt, S. Turri and C. Gerbaldi, Science, 2016, 354, 203–206.
52. F. Matteocci, L. Cinà, E. Lamanna, S. Cacovich, G. Divitini, P. A. Midgley, C. Ducati and A. Di Carlo, Nano Energy, 2016, 30, 162–172.
53. R. Cheacharoen, N. Rolston, D. Harwood, K. A. Bush, R. H. Dauskardt and M. D. McGehee, Energy Environ. Sci., 2018, 11, 144–150.
54. C. Wang, Z. Song, Y. Yu, D. Zhao, R. A. Awni, C. R. Grice, N. Shrestha, R. J. Ellingson, X. Zhao and Y. Yan, Sustainable Energy Fuels, 2018 10.1039/c8se00200b.
55. J. A. McLeod and L. Liu, J. Phys. Chem. Lett., 2018, 9, 2411–2417.
56. K. E. Hurst, M. J. Heben, J. L. Blackburn, T. Gennett, A. C. Dillon and P. A. Parilla, Rev. Sci. Instrum., 2013, 84, 025103.
57. R. Long, W. Fang and O. V. Prezhdo, J. Phys. Chem. Lett., 2016, 7, 3215–3222.
58. Q. Wang, B. Chen, Y. Liu, Y. Deng, Y. Bai, Q. Dong and J. Huang, Energy Environ. Sci., 2017, 10, 516–522.
59. Y. Zhao and K. Zhu, Chem. Commun., 2014, 50, 1605–1607.
60. C. Li, S. Pang, H. Xu and G. Cui, Sol. RRL, 2017, 1, 1700076.
61. G. Y. Kim, A. Senocrate, T.-Y. Yang, G. Gregori, M. Grätzel and J. Maier, Nat. Mater., 2018, 17, 445–449.
62. S. Wang, Y. Jiang, E. J. Juarez-Perez, L. K. Ono and Y. Qi, Nat. Energy, 2016, 2, 16195.
63. N. Aristidou, C. Eames, I. Sanchez-Molina, X. Bu, J. Kosco, M. S. Islam and S. A. Haque, Nat. Commun., 2017, 8, 15218.
64. T. Leijtens, G. E. Eperon, S. Pathak, A. Abate, M. M. Lee and H. J. Snaith, Nat. Commun., 2013, 4, 3885.
65. Q. Sun, P. Fassl, D. Becker-Koch, A. Bausch, B. Rivkin, S. Bai, P. E. Hopkinson, H. J. Snaith and Y. Vaynzof, Adv. Energy Mater., 2017, 7, 1700977.
66. E. L. Øiestad, Å. M. L. Øiestad, H. Skaane, K. Ruud, T. Helgaker, E. Uggerud and T. Vulpius, Eur. Mass Spectrom., 1995, 1, 121–129.
67. D. W. deQuilettes, W. Zhang, V. M. Burlakov, D. J. Graham, T. Leijtens, A. Osherov, V. Bulović, H. J. Snaith, D. S. Ginger and S. D. Stranks, Nat. Commun., 2016, 7, 11683.
68. J. Haruyama, K. Sodeyama, L. Han and Y. Tateyama, J. Am. Chem. Soc., 2015, 137, 10048–10051.
69. J.-W. Lee, D.-J. Seol, A.-N. Cho and N.-G. Park, Adv. Mater., 2014, 26, 4991–4998.
70. J. Schoonman, Chem. Phys. Lett., 2015, 619, 193–195.
71. Z. Guo, Y. Wan, M. Yang, J. Snaider, K. Zhu and L. Huang, Science, 2017, 356, 59–62.
72. J. Huang, Y. Yuan, Y. Shao and Y. Yan, Nat. Rev. Mater., 2017, 2, 17042.
73. S. A. Bretschneider, F. Laquai and M. Bonn, J. Mater. Chem. C, 2017, 121, 11201–11206.
74. G. Han, H. D. Hadi, A. Bruno, S. A. Kulkarni, T. M. Koh, L. H. Wong, C. Soci, N. Mathews, S. Zhang and S. G. Mhaisalkar, J. Mater. Chem. C, 2018, 122, 13884–13893.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8se00358k

---