Thermally evaporated methylammonium tin triiodide thin films for lead-free perovskite solar cell fabrication

Abstract

We report on the synthesis of methylammonium tin triiodide (MASnI₃) thin films at room temperature by a hybrid thermal evaporation method and their application in fabricating lead (Pb)-free perovskite solar cells. The as-deposited MASnI₃ thin films exhibit smooth surfaces, uniform coverage across the entire substrate, and strong crystallographic preferred orientation along the 〈100〉 direction. By incorporating this film with an inverted planar device architecture, our Pb-free perovskite solar cells are able to achieve an open-circuit voltage (V_oc) up to 494 mV. The relatively high V_oc is mainly ascribed to the excellent surface coverage, the compact morphology, the good stoichiometry control of the MASnI₃ thin films, and the effective passivation of the electron-blocking and hole-blocking layers. Our results demonstrate the potential capability of the hybrid evaporation method to prepare high-quality Pb-free MASnI₃ perovskite thin films which can be used to fabricate efficient Pb-free perovskite solar cells.

---

Introduction

Organic–inorganic lead (Pb) halide perovskite solar cells have progressed dramatically over the past few years,^1–10 with a record efficiency of 22.1% being reported recently.¹¹ This incredible progress is largely ascribed to the superior photovoltaic properties of the Pb-based perovskite absorbers, including high absorption coefficients, long carrier diffusion length, suitable band gaps, and low non-radiative recombination rates.^12–16 Despite this rapid increase in efficiency, the Pb-based perovskite solar cells continue to face challenges such as their instability against heat and moisture and the toxicity of their most significant component element, Pb.¹² As reported, the Pb-based perovskites decompose upon contacting with some polar solvents, such as water, leading to the resultant lead iodide (PbI₂) and environmental concerns.¹⁷ This combined environmental and public health issue raises serious concerns of the viability of Pb-based perovskites, which might hinder large-scale adoption of solar cells and similar devices based on these materials. Therefore, the development of efficient and Pb-free perovskite devices is of great interest and extensive efforts have been recently made towards Pb-free perovskites.^18,19

A straightforward approach for developing Pb-free perovskites is to replace the Pb²⁺ ion with isovalent tin (Sn²⁺), thus forming Sn-based perovskites such as ASnX₃ (with A = cesium (Cs) or organic cations such as methylammonium (MA) and formamidinium (FA), and X = halogen atoms such as chlorine (Cl), bromine (Br), or iodine (I)). Pb-free perovskite solar cells using CsSnI₃, MASnI₃, FASnI₃, CsSnI_3−xBr_x, and MASnBr₃ as absorbers have already been reported.^13,20–25 In contrast to Pb-based perovskite solar cells, there are two major obstacles hindering the fabrication of efficient Sn-based perovskite solar cells. Firstly, the most commonly used deposition method, i.e. spin-coating, cannot easily produce smooth Sn perovskite thin films with uniform coverage.^21,26 Secondly, due to the ease of oxidation of Sn(II), the synthesized Sn perovskite thin films usually exhibit high hole concentration, which is undesirable for efficient solar cell fabrication. To prevent the formation of such large hole densities, tin(II) fluoride (SnF₂) is typically used as an additive in the spin-coating process.²⁴ Compared to the spin-coating process, the thermal evaporation technique usually produces films with smoother surfaces, full surface coverage and high uniformity. Thermal evaporation has been used successfully to fabricate efficient MAPbI₃ perovskite solar cells in our and other groups.^4,27–30 Recently, smooth MASnBr₃ thin films have been synthesized by vapor deposition.³¹ However, so far, there is no report on successful synthesis of MASnI₃ thin films by vacuum deposition, which has a more suitable band gap for more efficient single-junction solar cells than MASnBr_3.

Here we report on the first thermal evaporation of high-quality MASnI₃ thin films at room temperature. Without post-deposition annealing, the as-deposited MASnI₃ thin films exhibited high crystallinity, smooth surface, full coverage, and strong preferred crystallographic orientation along the 〈100〉 direction as revealed by various characterizations. Additionally, the as-deposited MASnI₃ thin films showed hole densities suitable for solar cell applications even without the use of any SnF₂ additives, due to the ability of accurate film stoichiometry control via the hybrid thermal evaporation method. The band gap of our MASnI₃ thin film is about 1.3 eV. Sn-based perovskite solar cells were fabricated with the inverted device architecture of ITO/PEDOT:PSS/Poly-TPD/MASnI₃/C₆₀/BCP/Ag, where PEDOT:PSS, Poly-TPD, C₆₀, and BCP stands for poly(3,4-ethylenedioxythiophene):polystyrene sulfonate, poly[N,N′-bis(4-butylphenyl)-N,N′-bis(phenyl)benzidine], fullerene, and 2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline, respectively. The PEDOT:PSS layer acts as both the anode buffer layer and hole transporting layer and while the Poly-TPD acts as another hole transporting layer and an electron blocking layer. C₆₀ is the passivation layer and electron transporting layer, while BCP is a hole blocking layer. The thickness of the MASnI₃ absorber layer is typically about 200 nm. Our MASnI₃ perovskite solar cells have achieved open-circuit voltages (V_oc's) as high as 494 mV. The champion cell showed a power conversion efficiency (PCE) of 1.7% with a V_oc of 377 mV, a short-circuit current density (J_sc) of 12.1 mA cm⁻², and a fill factor (FF) of 36.6%. Further solar cell efficiency improvements could be expected upon the optimization of the MASnI₃ thickness and the interface engineering between the perovskite and the electron and/or hole blocking layers. Our results suggest that hybrid thermal evaporation is suitable for fabricating efficient MASnI₃ perovskite solar cells.

Our deposition process is a hybrid thermal evaporation. As depicted in Fig. S1,† two ceramic crucibles containing tin iodide (SnI₂) and methylammonium iodide (MAI) are used as the inorganic and organic precursor sources, respectively, and are heated to maintain at their respective temperature set points before the deposition is allowed to begin. The entire vacuum chamber is thus filled with MAI vapor before the beginning of the actual MASnI₃ deposition due to the high vapor pressure of MAI and lack of confinement of the MAI vapor within the crucible beyond the blocking shutter. The existence of the MAI vapor is confirmed by the increased pressure from ∼10⁻⁷ to ∼10⁻⁵ Torr in the vacuum chamber. The readouts from the quartz crystal microbalances (QCMs) placed above the SnI₂ and MAI crucibles before and after opening their individual shutter for SnI₂ and MAI confirm the thermal co-evaporation of SnI₂ and MAI in the presence of MAI vapor, which we refer as to hybrid thermal evaporation. Due to the complexity of the MAI deposition, instead of monitoring the deposition rate via the QCMs, the vapor pressure is monitored as mentioned in our earlier work.⁴ The deposition rate of the SnI₂ source is simply monitored by a QCM placed above the SnI₂ crucible. The stoichiometry of MASnI₃ thin films is controlled by the evaporation rate of SnI₂, while the MAI vapor pressure remains constant.

Experimental

Material synthesis

The MASnI₃ thin films were co-evaporated by a dual-source evaporation system (Kurt J. Lesker Mini Spectros) equipped in a nitrogen-filled glove box. Two different substrates were used for the depositions: soda-lime glass (SLG) substrates for the material characterization and pre-patterned indium tin oxide (ITO) coated SLG substrates for the device fabrication and characterization. Prior to their utilization, the substrates were sequentially ultrasonicated in diluted Micro-90 detergent, deionized water, acetone, and isopropanol for 15 min respectively, and then dried with flowing nitrogen gas. Two ceramic crucibles containing tin iodide (SnI₂) and methylammonium iodide (MAI) were used as the inorganic and organic precursor sources respectively and were heated to their respective temperature set points before the deposition was started. The deposition rate of the SnI₂ source was monitored by a quartz crystal microbalance (QCM) placed close to the substrate, while the vapor phase pressure of MAI was monitored instead of the deposition rates, since directly monitoring the deposition rate of MAI was relatively difficult as mentioned earlier in our work.⁴ The co-evaporation parameter used was 1.0 Å s⁻¹ for SnI₂ and 8 × 10⁻⁵ Torr for MAI, roughly corresponding to a 1 : 1 molar ratio between the two precursor sources. During the entire deposition, the substrate was rotated at a constant speed of 10 rpm and was not heated intentionally. The as-deposited CH₃NH₃SnI₃ thin films could be stored in a nitrogen-filled glove box in the dark for up to two months and no obvious degradation was observed.

Device fabrication

The CH₃NH₃SnI₃ thin films were fabricated into solar cells with an inverted planar structure. Before any depositions, the ITO substrates were cleaned with the aforementioned process followed by a UV/O₃ treatment for 10 min to improve the wetting ability of the substrates. A layer of PEDOT:PSS with or without Poly-TPD was used as the hole transporting layer. The PEDOT:PSS layer was deposited by spin coating at 4000 rpm from an aqueous solution. The PEDOT:PSS solution was filtered before use. The Poly-TPD layer was deposited by spin-coating at 4000 rpm from a 10 mg mL⁻¹ chlorobenzene solution. Both layers were baked on a hot plate for 30 min to remove the residual solvent. The substrate was then loaded into the co-evaporation system and the CH₃NH₃SnI₃ layer was deposited by co-evaporation. A combination of C₆₀ and BCP was used as the electron transporting and hole blocking layer. The C₆₀ and the BCP layers were deposited by thermal evaporation at a rate of 0.1–0.2 Å s⁻¹ followed by a layer of Ag thermal evaporated through a mask to produce 0.04 cm² cells. Except for the PEDOT:PSS, all other layers were deposited in a nitrogen-filled glove box and afterwards the substrates were not exposed to the ambient until shortly before the device characterization. All materials used in this work were commercially available and were used as received.

Material characterization

The morphological, topological, structural, optical, and electrical properties of the MASnI₃ thin films were evaluated by different characterization methods. The morphology and topology was characterized by scanning electron microscopy (SEM), and atomic force microscopy (AFM) respectively. The SEM images were taken with a Hitachi S-4800 High Resolution SEM. The AFM image and surface roughness analysis was performed with a Veeco Nanoscope V instrument. The structural properties were characterized by X-ray diffraction (XRD). The XRD pattern was acquired with a Rigaku Ultima III high resolution X-ray diffractometer, with the Cu-Kα line (λ = 0.154 nm) at 44 kV and 40 mA source excitation. The optical properties were characterized by UV-vis spectroscopy. The transmittance spectrum was obtained with a UV-vis spectrophotometer (PerkinElmer Lambda 1050) and transferred into the absorption coefficient spectrum using the equation α = −log(T)/d, where α, T, and d is the absorption coefficient, transmittance, and the film thickness, respectively. The electrical properties were characterized by 4-point probe and Hall effect measurements. The Hall measurements were performed at room temperature in a four-probe configuration by a commercialized Hall effect measurement system (MMR Technologies), which is equipped with a designed software to calculate the carrier type and the carrier concentration.

Device characterization

The performance of the CH₃NH₃SnI₃ perovskite solar cells was evaluated by current density–voltage (J–V) and external quantum efficiency (EQE) measurements. The J–V curves were measured using a Keithley 2400 sourcemeter under AM1.5G (100 W cm⁻²) solar illumination (PV Measurements Inc.) by sweeping between −0.5 V and 1.5 V with a voltage scan rate of 157 mV s⁻¹. The EQE spectra were measured using a commercialized QE system (PV Measurements Inc.), the monochromatic light intensity of which was calibrated with a standard silicon diode at each wavelength. All device characterizations were performed in the ambient with the device unencapsulated.

Results and discussion

The as-deposited MASnI₃ thin films show highly reflective surfaces, a representative photo of which is shown in Fig. S2.† The top-view scanning electron microscopy (SEM) images of our Sn-based perovskite thin film show densely packed grains over large areas, as seen in Fig. 1(a) and S3.† No apparent pin-holes are observed in the thin films. The grain size distribution is obtained by using the software Image J. The results are shown in Fig. 1(b). The average grain size is about 110 nm, which is slightly smaller than the thickness of the films (typically about 200 nm). The grain size distribution is found to be comparable with that obtained in the atomic-force microscopy (AFM) image, shown in Fig. 1(c). Fig. 1(d) shows the 3D AFM image of the surface morphology, with an RMS roughness of 10.7 nm from the AFM measurements. It is also worth noting that the choice of substrates, e.g., bare soda lime glass (SLG), indium doped tin oxide (ITO) coated SLG, PEDOT:PSS- or Poly-TPD-coated substrates, does not affect the smoothness and the morphology of the MASnI₃ thin films deposited by this hybrid thermal evaporation method. This feature allows us to conduct UV-vis spectroscopy measurements using MASnI₃ thin films grown on SLG substrates to calculate the band gap.

Fig. 1 The morphological and topographic properties of the MASnI₃ thin films grown by the hybrid thermal evaporation: (a) top-view SEM image, (b) histogram of the grain size, (c) top-view AFM image, and (d) projected AFM 3D plot, of the MASnI₃ thin film. The RMS roughness is 10.7 nm.

The hybrid evaporated MASnI₃ thin films deposited at room temperature without any post-deposition annealing exhibit strong crystallinity. Fig. 2(a) shows the X-ray diffraction (XRD) pattern of the MASnI₃ thin film. The observed characteristic peaks correspond to the (100), (200), (220), (221), and (400) reflections at 2θ = 14.1°, 28.5°, 40.8°, 43.4°, and 59.1°, respectively. All of these peaks belong to the cubic structure of MASnI₃ perovskite and the calculated XRD pattern is shown in Fig. 2(a). Compared with our calculated XRD pattern or the measured XRD patterns reported in literature, including solution-processed^20,21 or vapor-assisted solution-processed MASnI₃,³² our hybrid thermally evaporated MASnI₃ films show a strong preferred orientation along the 〈100〉 direction. The mechanism of the preferred orientation is still under investigation.

Fig. 2 (a) XRD pattern and (b) absorption and absorption coefficient (α) spectrum of the MASnI₃ thin film by the hybrid thermal evaporation method. The inset in figure is the Tauc plot and the derived band gap is about 1.3 eV.

The optical properties of our MASnI₃ thin films grown by the hybrid thermal evaporation are examined by UV-vis spectroscopy. Fig. 2(b) shows the absorption spectrum of the MASnI₃ film. The band gap derived from the Tauc plot, shown in the inset of Fig. 2(b), is about 1.3 eV. This value is in good agreement with the previously reported band gap of the MASnI₃ from both experiment and theoretical calculations.^20,21

Electrical properties of our MASnI₃ thin films are characterized by Hall effect measurement. Our MASnI₃ thin films show hole densities in the range of 10¹⁷–10¹⁸ cm⁻³. It has been proposed that due to the ease of oxidation of Sn(II), Sn vacancies can easily form during film deposition, leading to excessively large hole densities and low shunt resistances and therefore poor solar cell performance. The hole density of our as-deposited MASnI₃ thin films is suitable for solar cell applications, although they are not optimal yet.

The above characterization results indicate that our as-deposited MASnI₃ thin films fabricate by hybrid thermal evaporation exhibit structural and optoelectronic properties suitable for solar cell fabrication: excellent surface morphology with full area coverage, high crystallinity, a suitable band gap, and a reasonable carrier density. Therefore, solar cells are fabricated by using these MASnI₃ thin films as the light absorbers. An inverted device architecture of ITO/PEDOT:PSS/Poly-TPD/MASnI₃/C₆₀/BCP/Ag is explored in this work, which enables successful fabrication of MASnI₃ perovskite solar cells.

Fig. 3(a) shows the cross-sectional SEM image of our MASnI₃ perovskite solar cell with artificial color added to highlight the layer structure. The thickness of the MASnI₃ layer is confirmed to be about 200 nm. Even though the MASnI₃ film with such a low thickness might not be sufficient to fully absorb the photons with energy larger than the band gap of MASnI₃, this thickness was chosen to reduce the total MASnI₃ deposition time and to evaluate the suitability of the hybrid thermal evaporation of MASnI₃ thin films for solar cell application. Fig. 3(b) shows the schematic energy band diagram of our MASnI₃ solar cells with the inverted architecture. The energy levels of the conduction and valence band edges of MASnI_3, electron and hole blocking layers, and the work functions of electrodes are taken from literature.^5,20 The Poly-TPD layer can efficiently block the electrons from the MASnI₃ perovskite and transfer holes to the PEDOT:PSS-coated ITO electrode.⁵ In a similar manner, the C₆₀ and BCP layers work together to block holes. Since electrons need to tunnel through the BCP layer to reach the Ag electrode, the BCP layer is thinner (6 nm) than the C₆₀ (20–30 nm), Poly-TPD (20 nm), and PEDOT:PSS (40 nm) layers. The selective nature of these interlayers is the working mechanism of our solar cells.

Fig. 3 (a) Cross-sectional SEM image of our MASnI₃ perovskite solar cell. (b) Schematic energy band diagram of the devices.

During the solar cell fabrication, we've found that the use of PEDOT:PSS/Poly-TPD hole selective bi-layers results in significantly improved device performance compared with only PEDOT:PSS. Fig. 4(a) shows the current density–voltage (J–V) curves of the MASnI₃ perovskite solar cells under both reverse and forward voltage scans by using PEDOT:PSS or PEDOT:PSS/Poly-TPD as the hole selective layers. Except for the hole selective layers, all other layers in our work, i.e. MASnI₃, C₆₀, BCP, and Ag, are taken in the same deposition. Therefore, different device performance could be ascribed to the effect of different hole selective layers. It is seen that the MASnI₃ perovskite solar cell using PEDOT:PSS/Poly-TPD hole selective bi-layers shows much higher J_sc (5.1 mA cm⁻²) and V_oc (494 mV) than the cell using PEDOT:PSS-only hole selective layer (J_sc = 3.5 mA cm⁻² and V_oc = 470 mV). We have fabricated at least 30 MASnI₃ perovskite solar cells using each hole selective layer and the device performance improvement by PEDOT:PSS/Poly-TPD hole selective bi-layers has been statistically confirmed. The efficiency histogram of Sn-based perovskite solar cells with PEDOT:PSS-only and PEDOT:PSS/Poly-TPD hole selective layer is shown in Fig. S4.† It is worth noting that all the MASnI₃ perovskite solar cells with the inverted architecture present very little J–V hysteresis.

Fig. 4 (a) J–V curves and (b) EQE spectra of MASnI₃ perovskite cells using PEDOT:PSS and PEDOT:PSS/Poly-TPD as the electron selective layers. The scan rate in J–V measurements is 157 mV s⁻¹.

The J_sc difference is further understood from the measured external quantum efficiency (EQE) spectra of the MASnI₃ perovskite solar cells with PEDOT:PSS-only and PEDOT:PSS/Poly-TPD hole selective bi-layers, shown in Fig. 4(b). Both EQE spectra reflect spectral response in the range from 350 to 950 nm. The EQE spectral response onset is 950 nm, which indicated a band gap value of 1.3 eV. This result is consistent with the optical band gap of 1.3 eV measured by UV-vis spectroscopy, mentioned earlier in our discussion. The integrated photocurrents from these EQE spectra using the AM1.5G solar photon flux spectrum are 3.2 mA cm⁻² for the cell using PEDOT:PSS-only and 5.4 mA cm⁻² for the cell using PEDOT:PSS/Poly-TPD, which are in good agreement with the J_sc's measured from the J–V curves. It is obvious that by adding the Poly-TPD layer on PEDOT:PSS, the EQE value is enhanced in the entire wavelength range, suggesting a reduced charge recombination due to the introduction of Poly-TPD layer. A unique feature of these EQE spectra is that the spectral response is stronger in the longer wavelength region than in the shorter wavelength region, which is different from most spectral response of MASnI₃ perovskite solar cells reported.^20,32 This feature may be ascribed to the use of C₆₀ electron selective layer, which is known to facilitate electron transfer from Pb-based perovskites to the conductive electrodes.^4,9 Therefore, the recombination rate at the MASnI₃/C₆₀ interface is expected to be lower than that at the PEDOT:PSS/MASnI₃ or PEDOT:PSS/Poly-TPD/MASnI₃ interfaces.

The performance of MASnI₃ perovskite solar cells is further improved by optimizing the thickness of the C₆₀ electron selective layer. We found a tradeoff between V_oc and J_sc by varying the C₆₀ thickness. The highest V_oc of 494 mV is achieved with a 30 nm thick C₆₀. However, the J_sc (5.4 mA cm⁻²) of this cell is lower than that of the other cells with a thinner C₆₀ layer. The solar cell with a 20 nm-thick C₆₀ layer achieved the best efficiency of 1.7%, with a V_oc of 377 mV, a J_sc of 12.1 mA cm⁻², and an FF of 36.6%. The J–V curves under reverse and forward voltage scans and the EQE spectrum of this best-performing MASnI₃ device are shown in Fig. 5. It is worth highlighting that the best cell shows no obvious J–V hysteresis, which should be owed to the inverted device architecture.^13,33 Its corresponding EQE spectrum has a similar shape shown in Fig. 4(b), but it has much higher values in the entire wavelength range. This indicates that improvement of the J_sc can be expected in the future by optimizing all the interlayers. The efficiency histogram of Sn-based perovskite solar cells with a 20 nm C₆₀ layer is shown in Fig. S5.† It is also noted that all MASnI₃ perovskite solar cells suffer from low FFs, which could be due to the low shunt resistance (R_sh) and high series resistance (R_s). For the best-performing device, the R_sh and R_s are estimated to be 151.1 Ω cm² and 396.9 Ω cm², respectively. The low R_sh could result from the high hole density in MASnI₃, which could be caused by the easy formation of Sn⁴⁺ from the oxidation of Sn²⁺.²⁰ The high R_s could be ascribed to the charge recombination. These issues will be addressed in our future studies.

Fig. 5 (a) J–V curves under reverse and forward voltage scans and (b) EQE spectrum of our champion MASnI₃ perovskite solar cell.

To understand the limiting characteristics of our MASnI₃ perovskite solar cells, we have conducted J–V measurement with varying light intensity. Fig. 6 shows the dependence of the J_sc and V_oc on the light intensity ranging from 0.8 to 100 mW cm⁻². The power law dependence of the J_sc on the light intensity is J ∝ I^α. Generally, a space charge limited solar cell has α ∼ 0.75. Here, our hybrid thermally evaporated MASnI₃ perovskite solar cell exhibits an α value of about 0.83, shown in Fig. 6(a). This could be ascribed to the charge carrier imbalance resulting from the interfacial barriers formed due to the non-optimal thickness of the electron/hole transporting layers or the energy level alignment between these interlayers and the perovskite.^4,5,34–36 In Fig. 6(b), the V_oc of our MASnI₃ perovskite solar cell shows a linear relationship with the light intensity (natural logarithmic scale). A trap-free relationship should have a slope of δV_oc = kT/q.^37,38 The slope of δV_oc = 1.97 kT/q in our devices implies significant Shockley–Read–Hall recombination prevalent in our MASnI₃ perovskite solar cells.^39–41

Fig. 6 Light intensity-dependence of the MASnI₃ perovskite solar cells: (a) J_sc versus light intensity and (b) V_oc versus light intensity.

It is known that the facile oxidation of Sn²⁺ could cause instability for MASnI₃ perovskites, which could be avoided by encapsulation.²¹ For this work, our MASnI₃ perovskite solar cells were not encapsulated and the cells were measured in ambient air. We observed gradual degradation if the cells were kept in the ambient, similar to other works on Sn halide perovskite solar cells.^13,21 Parallel degradation mechanisms such as the oxidation of the Ag electrode and/or the oxygen doping of the fullerene layer have also been proposed.¹³ However, the hybrid evaporated MASnI₃ perovskite thin films could survive for more than two months without obvious degradation if stored in a nitrogen-filled glove box, as shown in Fig. S6.† There is no obvious performance degradation of the solar cells fabricated with these aged MASnI₃ perovskite thin films as shown in Fig. S6.†

Conclusions

In summary, high-quality MASnI₃ thin films have been successfully fabricated by the hybrid thermal evaporation method. The as-deposited films have very smooth surface, densely packed grains, excellent surface coverage, and preferred crystallographic orientation along the 〈100〉 direction. The carrier density of holes could be controlled by varying the stoichiometry of the films, allowing for favorable electronic properties for use in solar cells. Our as-deposited MASnI₃ thin films exhibited hole densities of 10¹⁷–10¹⁸ cm⁻³, which are suitable for solar cell application. The inverted planar MASnI₃ perovskite solar cells have achieved a V_oc as high as 494 mV. Our best-performing cell has a PCE of 1.7%, with a V_oc of 377 mV, a J_sc of 12.1 mA cm⁻², and an FF of 36.6%. Further performance enhancement could be expected by optimizing the MASnI₃ perovskite and interfacial engineering between the perovskite and the electron/hole selective layers. Our results demonstrate that hybrid thermal evaporation is a viable technique to prepare high-quality MASnI₃ thin films, beneficial for efficient Sn halide perovskite solar cells.

Acknowledgements

This work is financially supported by the U.S. Department of Energy (DOE) SunShot Initiative under the Next Generation Photovoltaics 3 program (DE-FOA-0000990), National Science Foundation under contract no. CHE–1230246 and DMR–1534686, and the Ohio Research Scholar Program. The work at the National Renewable Energy Laboratory is supported by the U.S. Department of Energy SunShot Initiative under the Next Generation Photovoltaics 3 program under Contract No. DE-AC36-08-GO28308. This research uses the resources of the Ohio Supercomputer Center and the National Energy Research Scientific Computing Center, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. H.M.Z thanks the support from the 973 Program of China (2015CB932203).

Notes and references

1. J. Burschka, N. Pellet, S.-J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin and M. Graetzel, Nature, 2013, 499, 316–319.
2. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050–6051.
3. J. H. Heo, S. H. Im, J. H. Noh, T. N. Mandal, C.-S. Lim, J. A. Chang, Y. H. Lee, H.-j. Kim, A. Sarkar, M. K. Nazeeruddin, M. Graetzel and S. I. Seok, Nat. Photonics, 2013, 7, 487–492.
4. D. Zhao, W. Ke, C. R. Grice, A. J. Cimaroli, X. Tan, M. Yang, R. W. Collins, H. Zhang, K. Zhu and Y. Yan, Nano Energy, 2016, 19, 88–97.
5. D. Zhao, M. Sexton, H.-Y. Park, G. Baure, J. C. Nino and F. So, Adv. Energy Mater., 2015, 5, 1401855.
6. W. Ke, G. Fang, Q. Liu, L. Xiong, P. Qin, H. Tao, J. Wang, H. Lei, B. Li and J. Wan, J. Am. Chem. Soc., 2015, 137, 6730–6733.
7. W. Ke, G. Fang, J. Wan, H. Tao, Q. Liu, L. Xiong, P. Qin, J. Wang, H. Lei, G. Yang, M. Qin, X. Zhao and Y. Yan, Nat. Commun., 2015, 6, 6700.
8. Z. Xiao, Y. Yuan, Y. Shao, Q. Wang, Q. Dong, C. Bi, P. Sharma, A. Gruverman and J. Huang, Nat. Mater., 2015, 14, 193–198.
9. E. M. Sanehira, B. J. Tremolet de Villers, P. Schulz, M. O. Reese, S. Ferrere, K. Zhu, L. Y. Lin, J. J. Berry and J. M. Luther, ACS Energy Lett., 2016, 38–45,  DOI:10.1021/acsenergylett.6b00013.
10. W.-J. Yin, H. Chen, T. Shi, S.-H. Wei and Y. Yan, Adv. Electron. Mater., 2015, 1 DOI:10.1002/aelm.201500044.
11. NREL Best Research-Cell Efficiencies Chart, http://www.nrel.gov/ncpv/images/efficiency_chart.jpg, (accessed on August 2016).
12. M. A. Green, A. Ho-Baillie and H. J. Snaith, Nat. Photonics, 2014, 8, 506–514.
13. K. P. Marshall, R. I. Walton and R. A. Hatton, J. Mater. Chem. A, 2015, 3, 11631–11640.
14. W.-J. Yin, T. Shi and Y. Yan, Adv. Mater., 2014, 26, 4653–4658.
15. W.-J. Yin, T. Shi and Y. Yan, Appl. Phys. Lett., 2014, 104, 063903.
16. W.-J. Yin, T. Shi and Y. Yan, J. Phys. Chem. C, 2015, 119, 5253–5264.
17. M. Graetzel, Nat. Mater., 2014, 13, 838–842.
18. B. Saparov, F. Hong, J.-P. Sun, H.-S. Duan, W. Meng, S. Cameron, I. G. Hill, Y. Yan and D. B. Mitzi, Chem. Mater., 2015, 27, 5622–5632.
19. F. Hong, B. Saparov, W. Meng, Z. Xiao, D. B. Mitzi and Y. Yan, J. Phys. Chem. C, 2016, 120, 6435–6441.
20. F. Hao, C. C. Stoumpos, D. H. Cao, R. P. H. Chang and M. G. Kanatzidis, Nat. Photonics, 2014, 8, 489–494.
21. N. K. Noel, S. D. Stranks, A. Abate, C. Wehrenfennig, S. Guarnera, A.-A. Haghighirad, A. Sadhanala, G. E. Eperon, S. K. Pathak, M. B. Johnston, A. Petrozza, L. M. Herz and H. J. Snaith, Energy Environ. Sci., 2014, 7, 3061–3068.
22. S. J. Lee, S. S. Shin, Y. C. Kim, D. Kim, T. K. Ahn, J. H. Noh, J. Seo and S. I. Seok, J. Am. Chem. Soc., 2016, 138, 3974–3977.
23. T. M. Koh, T. Krishnamoorthy, N. Yantara, C. Shi, W. L. Leong, P. P. Boix, A. C. Grimsdale, S. G. Mhaisalkar and N. Mathews, J. Mater. Chem. A, 2015, 3, 14996–15000.
24. M. H. Kumar, S. Dharani, W. L. Leong, P. P. Boix, R. R. Prabhakar, T. Baikie, C. Shi, H. Ding, R. Ramesh, M. Asta, M. Graetzel, S. G. Mhaisalkar and N. Mathews, Adv. Mater., 2014, 26, 7122–7127.
25. F. Hao, C. C. Stoumpos, P. Guo, N. Zhou, T. J. Marks, R. P. H. Chang and M. G. Kanatzidis, J. Am. Chem. Soc., 2015, 137, 11445–11452.
26. E. Edri, S. Kirmayer, D. Cahen and G. Hodes, J. Phys. Chem. Lett., 2013, 4, 897–902.
27. M. Liu, M. B. Johnston and H. J. Snaith, Nature, 2013, 501, 395–398.
28. O. Malinkiewicz, A. Yella, Y. H. Lee, G. M. Espallargas, M. Graetzel, M. K. Nazeeruddin and H. J. Bolink, Nat. Photonics, 2014, 8, 128–132.
29. L. K. Ono, S. Wang, Y. Kato, S. R. Raga and Y. Qi, Energy Environ. Sci., 2014, 7, 3989–3993.
30. Q. Lin, A. Armin, R. C. R. Nagiri, P. L. Burn and P. Meredith, Nat. Photonics, 2015, 9, 106–112.
31. M.-C. Jung, S. R. Raga and Y. Qi, RSC Adv., 2016, 6, 2819–2825.
32. T. Yokoyama, D. H. Cao, C. C. Stoumpos, T.-B. Song, Y. Sato, S. Aramaki and M. G. Kanatzidis, J. Phys. Chem. Lett., 2016, 7, 776–782.
33. Y. Shao, Y. Yuan and J. Huang, Nat. Energy, 2016, 1, 15001.
34. V. D. Mihailetchi, J. Wildeman and P. W. M. Blom, Phys. Rev. Lett., 2005, 94, 126602.
35. L. J. A. Koster, V. D. Mihailetchi, H. Xie and P. W. M. Blom, Appl. Phys. Lett., 2005, 87, 203502.
36. D. Bi, L. Yang, G. Boschloo, A. Hagfeldt and E. M. J. Johansson, J. Phys. Chem. Lett., 2013, 4, 1532–1536.
37. S. R. Cowan, A. Roy and A. J. Heeger, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 82, 245207.
38. M. M. Mandoc, F. B. Kooistra, J. C. Hummelen, B. de Boer and P. W. M. Blom, Appl. Phys. Lett., 2007, 91, 263505.
39. S. R. Cowan, W. L. Leong, N. Banerji, G. Dennler and A. J. Heeger, Adv. Funct. Mater., 2011, 21, 3083–3092.
40. R. N. Hall, Phys. Rev., 1952, 87, 387.
41. W. Shockley and W. T. Read, Phys. Rev., 1952, 87, 835–842.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19476a

---