Spray reaction prepared FA_1−xCs_xPbI₃ solid solution as a light harvester for perovskite solar cells with improved humidity stability

Abstract

FA_1−xCs_xPbI₃ solid solution films are prepared as light harvesters for perovskite solar cells by a spray reaction. Cs⁺ ions can improve the quality of the light harvester film, and the average PCE of devices increases from 11.3% for FAPbI₃ to 14.2% for FA_0.9Cs_0.1PbI₃. Moreover, FA_0.9Cs_0.1PbI₃-based devices exhibit remarkable improvement on humidity stability.

---

Inorganic–organic lead halide is a kind of solution processable semiconductor with excellent photovoltaic performance.^1,2 This kind of material has a perovskite structure of AMX₃, and small organic ammonium cations occupy the A sites (M represents lead and X represents halogen). Methylammonium (CH₃NH₃⁺, MA⁺) is a wildly used organic ammonium cation in this material. Since the previous reports define the basic structure of perovskite solar cells,^3,4 the devices based on perovskite materials with MA⁺ ions have been extensively investigated^5–14 and their best power conversion efficiency (PCE) reaches about 20% under one sun condition.^15,16

This kind of perovskite materials can also employ an alternative organic ammonium cation of formamidinium (HC(NH₂)₂⁺, FA⁺), which is promising for broadening the wavelength range of light absorption.^17–29 And it is noted that the band gap of FAPbI₃ (1.48 eV) is closer to the ideal band gap of solar cells than that of MAPbI₃ (1.57 eV).^27,28 However, the devices based on pure FAPbI₃ usually have lower PCE than the devices based on MAPbI₃.^27,30,31 Although the α-phase FAPbI₃ is the desirable material for photovoltaic application, the δ-phase FAPbI₃, which seriously damages the performance of the device, emerges easily in the preparation process.²⁷ Thermal treatment is necessary for translating the δ-phase FAPbI₃ into its α-phase, but thermal decomposition usually arises at the elevated temperature.^18,32 These factors lead to the difficulty for fabricating FAPbI₃ films with high quality. Moreover, α-phase FAPbI₃ easily translates into δ-phase at room temperature in the wet air condition, which sincerely decreases the stability of perovskite solar cells.²⁷ Therefore, the simultaneous improvement on both the quality and the humidity stability of perovskite films becomes desirable.

The solid solution of perovskite materials usually preserves the photovoltaic property, while the detailed physical properties can be adjusted. Several works about solid solutions of perovskite materials are reported, where MA⁺ ions can be partially replaced by FA⁺ or cesium (Cs⁺) ions and I⁻ ions can be partially replaced by Br⁻ or Cl⁻ ions.^30–37 It is found that the doping atoms can affect the photovoltaic performance of the corresponding devices remarkably. Recently, it is also reported that the quality of α-phase FAPbI₃ films can be improved by the formation of solid solutions, such as FA_1−xMA_xPbI_3−xBr_x and FA_1−xMA_xPbI₃.^32,38 As a consequence, highly efficient perovskite solar cells based on FA_1−xMA_xPbI_3−xBr_x is produced by the intramolecular exchange method, and the best PCE of 20.2% is achieved.³⁹ To form the perovskite structure of AMX₃, organic ammonium cations must fit the space composed of four adjacent corner sharing MX₆ octahedra. The relatively large size of FA⁺ ions leads to the extension of PbI₆ octahedra matrices in FAPbI₃. Solid solution can be a method for relaxing the PbI₆ matrices by partially replacing large FA⁺ ion with small Cs⁺ ion, which can also occupy the A site in several perovskite materials with AMX₃ structure.^40,41

Spray technique is a wildly used method in industry process, and it is also an enlargeable technique for preparing perovskite light harvester films.⁴² In the preparation of FA_1−xCs_xPbI₃ solid solution, conventional solution based method has the possibility of phase separation in the process of solvent evaporation. For the spray reaction, most solvent converts to vapor before arriving at the iodide surface. Therefore, spray reaction method is likely to facilitate the formation of FA_1−xCs_xPbI₃ solid solution films with pure phase structure.

In this study, we prepare a series of FA_1−xCs_xPbI₃ light harvester films by an interface reaction method based on the spray technique. The crystal structure and light absorption of these light harvester films are characterized, and the solar cells with these films are investigated. Pure α-phase structure can be obtained for the Cs⁺ incorporation range up to 30%, which permits us to give a systematic investigation on this solid solution material.

Firstly, the fluorine doped tin oxide (FTO) glass substrate was etched and cleaned. 30 nm compact TiO₂ film and 200 nm mesoporous TiO₂ film were sequentially deposited by spray pyrolysis and spin-coating. Then FA_1−xCs_xPbI₃ perovskite film was prepared on the TiO₂ film by an interface reaction method based on the spray technique. Certain amounts of CsI were added in 462 mg mL⁻¹ PbI₂ N,N-dimethylformamide (DMF) solution following the Cs/Pb ratio of FA_1−xCs_xPbI₃ (x = 0, 0.1, 0.2 and 0.3). This CsI/PbI₂ mixed solution or pure PbI₂ solution were spin-coated on the mesoporous TiO₂ film. After preheating at the reaction temperature for 5 min, 50 mg mL⁻¹ FAI in isopropanol solution was sprayed on the top of the PbI₂ film at 160 °C for the interface reaction. Finally, the perovskite film was washed with isopropanol and annealed at the reaction temperature for 10 min. FA_1−xCs_xPbI₃ films were also prepared at different reaction temperatures, the X-ray diffraction (XRD) and scanning electron microscopy (SEM) analyses were shown in the ESI (Fig. S1 and S2†). To fabricate perovskite solar cell, a hole-conductor layer and a gold top electrode were deposited on the FA_1−xCs_xPbI₃ perovskite film sequentially by spin-coating and thermal evaporation. The cross-sectional morphology of a perovskite solar cell is shown in Fig. S3.† Experimental detail is provided in the ESI.†

Schematic illustration of the crystal structure of FAPbI₃ with Cs⁺ ions incorporation is shown in Scheme 1. Four kinds of FA_1−xCs_xPbI₃ films (x = 0, 0.1, 0.2 and 0.3) are prepared, and the amounts of Cs⁺ in the films are confirmed by energy dispersive X-ray spectroscopy (EDS) measurement (Table S1†) and X-ray photoelectron spectroscopy (XPS) measurement (Fig. S5 and Table S2†). Fig. 1(a) is the XRD patterns of these four FA_1−xCs_xPbI₃ films with different Cs⁺ concentrations. Except three diffraction peaks from the substrate, all diffraction peaks can be assigned to the trigonal crystal structure of FAPbI₃ film. It means that our spray reaction method can produce pure α-phase FAPbI₃ films at the optimized reaction condition. In Fig. 1(a), the incorporation of Cs⁺ ions in the FAPbI₃ films does not change the XRD patterns of pure α-phase structure. It means that spray reaction method avoids phase separation in the formation of these solid solution films. Fig. 1(b) shows the (110) diffraction peaks of four FA_1−xCs_xPbI₃ films with different Cs⁺ concentrations. The 2θ angle of diffraction peak increases with the increase of Cs⁺ concentration. We calculate the lattice parameters of FA_1−xCs_xPbI₃ by the corresponding XRD patterns, which are summarized in Table S3.† It indicates the lattice constant can be modified by controlling the concentration of Cs⁺ ions, and the more Cs⁺ ions incorporate the smaller the lattice constant is. This relationship implies that Cs⁺ ions replace FA⁺ ions by occupying the A site of AMX₃ structure than the interstitial site. And the substitutional solid solution of FA_1−xCs_xPbI₃ can be produced in the Cs⁺ concentration up to 30% by the spray reaction method. The decrease of lattice constant in the FA_1−xCs_xPbI₃ films is related to relax the extension in the PbI₆ matrices. Furthermore, the full width of half maximum (FWHM) of the (110) diffraction peaks in these four films are also different, which is summarized in Fig. 1(c). The FWHM of FAPbI₃ film (0.337°) is much larger than that of FA_0.9Cs_0.1PbI₃ (0.199°), while the FWHMs are 0.251° and 0.262° for FA_0.8Cs_0.2PbI₃ and FA_0.7Cs_0.3PbI₃, respectively. The broadening of XRD peaks are usually induced by the low quality of crystallization in the films. Then the smallest FWHM of FA_0.9Cs_0.1PbI₃ film reflects its highest quality among these four kinds of films. The incorporation of Cs⁺ ions also induces the variation of light absorption as show in Fig. 1(d). The absorption edge blue shifts by increasing Cs⁺ concentration. The energy band gaps of FA_1−xCs_xPbI₃ films are calculated by the absorption edge according to Kubelka–Munk equation, which are summarized in Fig. S6 and Table S4.† It indicates Cs⁺ ions incorporation increases the band gap of FA_1−xCs_xPbI₃. It is noted that light absorption intensity is also different for FA_1−xCs_xPbI₃ films with different Cs⁺ concentrations. Although large Cs⁺ concentration (FA_0.7Cs_0.3PbI₃) induces remarkable decrease in the absorption intensity, FA_1−xCs_xPbI₃ with small Cs⁺ concentration (FA_0.9Cs_0.1PbI₃) exhibits higher absorption intensity than FAPbI₃ in the wavelength range of 400–800 nm.

Scheme 1 Schematic illustration of the crystal structure of FAPbI₃ with Cs⁺ ions incorporation.

Fig. 1 (a) XRD patterns of FA_1−xCs_xPbI₃ (x = 0, 0.1, 0.2 and 0.3) films. The XRD peaks assigned to perovskite and FTO substrate are marked with solid circles and asterisks, respectively. (b) (110) diffraction peaks of FA_1−xCs_xPbI₃ films. (c) Full width of half maximum (FWHM) of the (110) diffraction peaks. (d) Absorption spectra of FA_1−xCs_xPbI₃ films.

We fabricate perovskite solar cells based on these four kinds of FA_1−xCs_xPbI₃ materials. The photocurrent density–photovoltage characteristics of corresponding devices are shown in Fig. 2(a), and the photovoltaic parameters are summarized in Table 1. By the Cs⁺ ion incorporation, the short circuit current density (J_SC) of corresponding devices increases from 17.7 mA cm⁻² at x = 0 to 20.3 mA cm⁻² at x = 0.1, whereas further increasing the amount of Cs⁺ ions decreases J_SC to 17.4 mA cm⁻² at x = 0.2 and 13.4 mA cm⁻² at x = 0.3. Moreover, the open circuit voltage (V_OC) is also increased from 0.95 V at x = 0 to 1.05 V at x = 0.1, and then decrease to 1.01 V at x = 0.2 and 0.99 V at x = 0.3. As a result, the PCE of corresponding devices is enhanced from 11.0% to 14.9% by replacing 10% FA⁺ ions of FAPbI₃ with Cs⁺ ions. The statistics comparison between PCEs of both FAPbI₃-based and FA_0.9Cs_0.1PbI₃-based devices is shown in Fig. 2(b). FA_0.9Cs_0.1PbI₃-based devices show an average efficiency of 14.2% with narrow distribution range, which is obviously better than that of FAPbI₃-based devices (11.3%). This result indicates that Cs⁺ ion incorporation can improve the performance of perovskite solar cells based on FAPbI₃. We measure the variation of current density with the light soaking time at the constant bias voltage of maximum power condition (V_opt is about 0.8 V). As shown in Fig. 2(c), both FAPbI₃-based and FA_0.9Cs_0.1PbI₃-based devices exhibit stabilized power conversion efficiency after 100 seconds light soaking, but the transient time is different. FA_0.9Cs_0.1PbI₃-based device reaches 90% of its stabilized PCE in 3.4 seconds, while FAPbI₃-based device needs 8.5 seconds for this transient process. It implies that Cs⁺ ions incorporation can accelerate to establish the balance state in the devices.

Fig. 2 (a) Photocurrent density–photovoltage characteristics of the FA_1−xCs_xPbI₃-based devices (x = 0, 0.1, 0.2 and 0.3). (b) Statistics of efficiencies for both FAPbI₃-based and FA_0.9Cs_0.1PbI₃-based devices. (c) Photocurrent density and PCE as a function of light soaking time measured at maximum power point (0.793 V for FAPbI₃-based devices and 0.804 V for FA_0.9Cs_0.1PbI₃-based device).

Table 1 The photovoltaic parameters of the devices based on FA_1−xCs_xPbI₃ (x = 0, 0.1, 0.2 and 0.3)

FA1−xCsxPbI3	JSC (mA cm^−2)	VOC (V)	Fill factor	PCE (%)
x = 0	17.7	0.95	0.65	11.0
x = 0.1	20.3	1.05	0.70	14.9
x = 0.2	17.4	1.01	0.67	11.8
x = 0.3	13.4	0.99	0.66	8.7

---

Fig. 3(a) is the incident photon-to-electron conversion efficiency (IPCE) spectra of devices based on FA_1−xCs_xPbI₃ films. At short wavelength range (below 500 nm), all devices based on these four FA_1−xCs_xPbI₃ films exhibit good IPCE values. This can be related to the large absorption coefficient of FA_1−xCs_xPbI₃ films at the short wavelength range. For devices based on FAPbI₃, the onset wavelength of IPCE is at 850 nm. After the incorporation of Cs⁺ ions, the onset wavelength blue shifts to 841 nm, this is in accordance with the blue shifting of absorption edge. It is noted that the onset wavelength of devices based on FA_0.9Cs_0.1PbI₃ films is still obviously longer than that of devices based on MAPbI₃ films (at about 800 nm).³² At the middle wavelength range (600 to 700 nm), the IPCE spectra of devices based on these four FA_1−xCs_xPbI₃ films have remarkable different IPCE values. The incorporation of 10% Cs⁺ ions can increase the IPCE value of corresponding devices, while further increasing the concentration of Cs⁺ ions leads to the decrease of IPCE value in the middle wavelength range. Although the trend of IPCE value is in accordance with the absorption spectra, the remarkable difference on IPCE values cannot be fully attributed to the minor difference of light absorption. SEM images of FA_1−xCs_xPbI₃ with four different Cs⁺ ion concentrations are shown in Fig. S7.† Although these FA_1−xCs_xPbI₃ films have similar morphology with compact structure, the grain sizes are different. The grain size distributions are analyzed by Nano Measurer software. The average grain size of FA_0.9Cs_0.1PbI₃ film is 257 nm, which is the largest one in these four samples. As previous reports,^7,19 the perovskite film with large crystal grains has improved crystallization and reduced crystal interfaces, which are beneficial to the recombination and transportation process. These improvements can increase the IPCE value of devices. The SEM analysis is accordance with the smallest FHWM of FA_0.9Cs_0.1PbI₃ film in the XRD measurement. We perform impedance spectroscopy (IS) experiments to investigate the recombination and transportation process.

Fig. 3 (a) IPCE spectra of the FA_1−xCs_xPbI₃-based devices (x = 0, 0.1, 0.2 and 0.3). (b) Nyquist plots of FA_1−xCs_xPbI₃-based devices measured at 0.8 V applied voltage. (c) Recombination resistance (R_rec) at different applied voltage. (d) Selective contact resistance (R_sc) at different applied voltage.

Impedance spectroscopy of perovskite solar cells measured at 0.8 V applied voltage are shown in Fig. 3(b). The obtained impedance data can be fitted with the equivalent circuit (Fig. S8†) to extract the impedance parameters, such as the recombination resistance in the perovskite layer (R_rec) and the charge transfer resistance at the interfaces of selective contacts (R_sc). As previous reports,^43–45 the low frequency arc of the impedance spectra can be attributed to the recombination process, and the corresponding resistance (R_rec) is inversely related to the recombination rate of photo-generated electrons. The high frequency arc of the impedance spectra is influenced by the charge transfer resistance at the selective contact layer/perovskite interfaces, and also by the transport resistance in the selective contact layers.^43,44 As the selective contact layers do not change in this experiment, the observed differences of high frequency impedance can be attributed to the charge transfer process at the selective contact layer/perovskite interfaces.

R_rec obtained at different applied potentials are summarized in Fig. 3(c). The R_rec with large applied potential is inversely proportion to the recombination rate. It can be observed that the device based on FAPbI₃ exhibits the smallest R_rec. The fast recombination leads to the low carrier density, which is accordance with the low open circuit voltage (0.95 V) of the FAPbI₃-based device in Fig. 2(a). By incorporation of 10% Cs⁺ ions in the FAPbI₃ films, R_rec increases remarkably and a V_OC of 1.05 V is obtained for the corresponding device. Further increasing the concentration of Cs⁺ ions to 20% and 30%, the R_rec of corresponding devices reduce gradually, which is accordance with the decrease of V_OC in Fig. 2(a). The trend of recombination process is also accordance with the quality of crystallization in the corresponding FA_1−xCs_xPbI₃ films in Fig. 1(c). It means that improving the quality of light harvester films is an effective way to decease the recombination rate in the perovskite solar cells.

Fig. 3(d) shows the R_sc obtained at different applied potentials. R_sc almost does not change at various applied potentials, and it decreases with the increase of Cs⁺ concentration in FA_1−xCs_xPbI₃. It implies that Cs⁺ incorporation is beneficial for the charge transfer at the selective contact layer/perovskite interfaces. R_sc is related with the series resistance of solar cells, while R_rec at low applied voltage reflects the shunt resistance. Small R_sc and large R_rec lead to good fill factor of solar cells.⁴³ As a result, the fill factor of FAPbI₃-based device (0.65) is enhanced to 0.70 by the 10% incorporation of Cs⁺ ions. Although FA_0.8Cs_0.2PbI₃-based and FA_0.7Cs_0.3PbI₃-based devices have lower R_sc than FA_0.9Cs_0.1PbI₃-based device, the decrease of R_rec at low applied voltage lead to the slightly decrease in fill factor (0.67 and 0.66).

Moreover, the stability of unsealed devices based on FAPbI₃ and FA_0.9Cs_0.1PbI₃ films are investigated in the environment of 50% humidity at 20 °C for 100 hours. The PCE decay is summarized in Fig. 4(a). In the first five hours, the PCE of device with FAPbI₃ light harvester film decreases from 11.4% to 3.58%. It further decreases to 0.35% after the aging of 100 hours. On the contrary, the PCE of device with FA_0.9Cs_0.1PbI₃ light harvester film decreases from 14.5% to 12.4% in the first five hours, and no obviously change is detected in the following time of 100 hours aging (12.5% at 100 hours). The photographs of these devices with and without aging process are shown in the ESI (Fig. S9†). After 100 hours aging, the FAPbI₃-based device becomes yellow in color, while the FA_0.9Cs_0.1PbI₃-based device remains its dark brown color. This result indicates 10% Cs⁺ ions incorporation leads to remarkable improvement on the humidity stability of devices.

Fig. 4 (a) PCE of both FAPbI₃-based and FA_0.9Cs_0.1PbI₃-based devices (unsealed) with aging time in 50% humidity and 20 °C environment. (b) XRD patterns of FAPbI₃ film and FA_0.9Cs_0.1PbI₃ film before and after aging in 50% humidity and 20 °C environment for 100 h. The XRD peaks assigned to α-phase and δ-phase are marked with solid circles and δ, respectively.

We also age naked FAPbI₃ and FA_0.9Cs_0.1PbI₃ films without hole conductor layers and Au electrode layers in the same condition. Fig. 4(b) shows the corresponding XRD patterns of these films before and after the aging process. Before humidity aging, both FAPbI₃ and FA_0.9Cs_0.1PbI₃ films give a set of strong peaks, which indicate an α-phase crystal structure of halide perovskite (marked with solid circles). After 100 hours aging, the peaks of δ-phase FAPbI₃ appear in the XRD patterns of the FAPbI₃ film (marked with δ), and the peaks of α-phase FAPbI₃ greatly attenuate. It indicates that most of α-phase FAPbI₃ changes into δ-phase FAPbI₃ in the aging process. As δ-phase FAPbI₃ cannot act as an effective light harvester for perovskite solar cell, the remarkable performance decay of FAPbI₃-based device in Fig. 4(a) can be attributed to the damage of α-phase FAPbI₃ film. On the contrary, FA_0.9Cs_0.1PbI₃ film still shows strong diffraction peaks of α-phase after the aging process, and no peak for δ-phase is observed. The undamaged FA_0.9Cs_0.1PbI₃ light harvester film leads to the stable performance of device in the aging process. This result indicates the Cs⁺ ions incorporation can enhance the stability of α-phase structure, which is the major factor for improving the humidity stability of perovskite photovoltaic devices.

Conclusions

In summary, we prepare a series of FA_1−xCs_xPbI₃ solid solution films as the light harvester in perovskite solar cells by the spray reaction method. XRD analysis reflects that solid solution with pure α-phase structure is obtained in the Cs⁺ concentration up to 30%, and its lattice constant can be adjusted by controlling Cs⁺ concentration. By optimizing the composition of FA_1−xCs_xPbI₃ films, the average PCE of devices is enhanced from 11.3% for FAPbI₃ to 14.2% for FA_0.9Cs_0.1PbI₃. This improvement can be attributed to the increase of light absorption intensity and the decrease of recombination process. Furthermore, FA_0.9Cs_0.1PbI₃ also exhibits better humidity stability than FAPbI₃. In the 50% humidity environment, α-phase FAPbI₃ rapidly translates into its δ-phase, and the PCE of FAPbI₃-based device decreases to 0.35% after the aging of 100 hours. On the contrary, FA_0.9Cs_0.1PbI₃ remains its α-phase structure in the humidity aging process, and the PCE of corresponding device keeps constant at about 12.5% after the initially minor decrease.

Acknowledgements

This work was partially supported by National Natural Science Foundation of China (Grant No. 51273079, 11374121), National Basic Research Program of China (Grant No. 2011CB808204), the Science Development Program of JiLin Province (Grant No. 20150519021JH), and the Fundamental Research Funds for Central Universities at Jilin University.

Notes and references

1. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050.
2. J. Im, C. Lee, J. Lee, S. Park and N. Park, Nanoscale, 2011, 3, 4088.
3. H. Kim, C. Lee, J. Im, K. Lee, T. Moehl, A. Marchioro, S. Moon, R. Humphry-Baker, J. Yum, J. Moser and M. Grätzel, Sci. Rep., 2012, 2, 591.
4. M. Lee, J. Teuscher, T. Miyasaka, T. Murakami and H. Snaith, Science, 2012, 338, 643.
5. M. Liu, M. Johnston and H. Snaith, Nature, 2013, 501, 395.
6. J. Burschka, N. Pellet, S. Moon, R. Humphry-Baker, P. Gao, M. Nazeeruddin and M. Grätzel, Nature, 2013, 499, 316.
7. J. Im, I. Jang, N. Pellet, M. Grätzel and N. Park, Nat. Nanotechnol., 2014, 9, 927.
8. J. Yan and B. Saunders, RSC Adv., 2014, 4, 43286.
9. Y. Yang, J. Xiao, H. Wei, L. Zhu, D. Li, Y. Luo, H. Wu and Q. Meng, RSC Adv., 2014, 4, 52825.
10. L. Etgar, P. Gao, Z. Xue, Q. Peng, A. Chandiran, B. Liu, M. Nazeeruddin and M. Grätzel, J. Am. Chem. Soc., 2012, 134, 17396.
11. M. Cai, V. Tiong, T. Hreid, J. Bell and H. Wang, J. Mater. Chem. A, 2015, 3, 2784.
12. J. Xiao, J. Shi, H. Liu, Y. Xu, S. Lv, Y. Luo, D. Li, Q. Meng and Y. Li, Adv. Energy Mater., 2015, 5, 1401943.
13. X. Yin, Y. Guo, Z. Xue, P. Xu, M. He and B. Liu, Nano Res., 2015, 8, 1997.
14. T. Liu, Q. Hu, J. Wu, K. Chen, L. Zhao, F. Liu, C. Wang, H. Lu, S. Jia, T. Russell, R. Zhu and Q. Gong, Adv. Energy Mater., 2015 DOI:10.1002/aenm.201501890.
15. N. Ahn, D. Son, I. Jang, S. Kang, M. Choi and N. Park, J. Am. Chem. Soc., 2015, 137, 8696.
16. 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.
17. J. Lee, D. Seol, A. Cho and N. Park, Adv. Mater., 2014, 26, 4991.
18. S. Pang, H. Hu, J. Zhang, S. Lv, Y. Yu, F. Wei, T. Qin, H. Xu, Z. Liu and G. Cui, Chem. Mater., 2014, 26, 1485.
19. F. Wang, H. Xu, H. Xu and N. Zhao, Adv. Funct. Mater., 2015, 25, 1120.
20. M. Hu, L. Liu, A. Mei, Y. Yang, T. Liu and H. Han, J. Mater. Chem. A, 2014, 2, 17115.
21. D. Seol, J. Lee and N. Park, ChemSusChem, 2015, 8, 2414.
22. G. Eperon, D. Bryant, J. Troughton, S. Stranks, M. Johnston, T. Watson, D. Worsley and H. Snaith, J. Phys. Chem. Lett., 2015, 6, 129.
23. F. Hanusch, E. Wiesenmayer, E. Mankel, A. Binek, P. Angloher, C. Fraunhofer, N. Giesbrecht, J. Feckl, W. Jaegermann, D. Johrendt, T. Bein and P. Docampo, J. Phys. Chem. Lett., 2014, 5, 2791.
24. D. Yuan, A. Gorka, M. Xu, Z. Wang and L. Liao, Phys. Chem. Chem. Phys., 2015, 17, 19745.
25. M. Leyden, M. Lee, S. Raga and Y. Qi, J. Mater. Chem. A, 2015, 3, 16097.
26. S. Aharon, A. Dymshits, A. Rotem and L. Etgar, J. Mater. Chem. A, 2015, 3, 9171.
27. 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.
28. C. Stoumpos, C. Malliakas and M. Kanatzidis, Inorg. Chem., 2013, 52, 9019.
29. H. Zhang, J. Shi, J. Dong, X. Xu, Y. Luo, D. Li and Q. Meng, J. Energy Chem., 2015, 24, 707.
30. G. Eperon, S. Stranks, C. Menelaou, M. Johnston, L. Herz and H. Snaith, Energy Environ. Sci., 2014, 7, 982.
31. N. Pellet, P. Gao, G. Gregori, T. Yang, M. Nazeeruddin, J. Maier and M. Grätzel, Angew. Chem., Int. Ed., 2014, 53, 3151.
32. N. Jeon, J. Noh, W. Yang, Y. Kim, S. Ryu, J. Seo and S. Seok, Nature, 2015, 517, 476.
33. J. Liu, Y. Shirai, X. Yang, Y. Yue, W. Chen, Y. Wu, A. Islam and L. Han, Adv. Mater., 2015, 27, 4918.
34. Y. Chen, T. Chen and L. Dai, Adv. Mater., 2015, 27, 1053.
35. H. Choi, J. Jeong, H. Kim, S. Kim, B. Walker, G. Kim and J. Kim, Nano Energy, 2014, 7, 80.
36. C. Bi, Y. Yuan, Y. Fang and J. Huang, Adv. Energy Mater., 2015, 5, 1401616.
37. J. Noh, S. Im, J. Heo, T. Mandal and S. Seok, Nano Lett., 2013, 13, 1764.
38. A. Binek, F. Hanusch, P. Docampo and T. Bein, J. Phys. Chem. Lett., 2015, 6, 1249.
39. W. Yang, J. Noh, N. Jeon, Y. Kim, S. Ryu, J. Seo and S. Seok, Science, 2015, 348, 1234.
40. L. Protesescu, S. Yakunin, M. Bodnarchuk, F. Krieg, R. Caputo, C. Hendon, R. Yang, A. Walsh and M. Kovalenko, Nano Lett., 2015, 15, 3692.
41. Q. Akkerman, V. D'lnnocenzo, S. Accornero, A. Scarpellini, A. Petrozza, M. Prato and L. Manna, J. Am. Chem. Soc., 2015, 137, 10276.
42. X. Xia, H. Li, W. Wu, Y. Li, D. Fei, C. Gao and X. Liu, ACS Appl. Mater. Interfaces, 2015, 7, 16907.
43. E. Juarez-Perez, M. Wuβler, F. Fabregat-Santiago, K. Lakus-Wollny, E. Mankel, T. Mayer, W. Jaegermann and I. Mora-Sero, J. Phys. Chem. Lett., 2014, 5, 680.
44. J. Zhang, E. Juárez-Pérez, I. Mora-Seró, B. Viana and T. Pauporté, J. Mater. Chem. A, 2015, 3, 4909.
45. M. Lv, X. Dong, X. Fang, B. Lin, S. Zhang, X. Xu, J. Ding and N. Yuan, RSC Adv., 2015, 5, 93957.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details. See DOI: 10.1039/c5ra23359c

---