Polar molecules modify perovskite surface to reduce recombination in perovskite solar cells

Abstract

Thin film solar cells can work efficiently by successful interfacial charge separation/collection. The solution-processed perovskite (CH₃NH₃PbI₃) film carries many trap states on the surface, which is detrimental to the high performance of solar cells. Therefore, it is of great urgency to control the interface in the device. In this study, polar silane molecules with amino end groups are self-assembled at the interface of the perovskite/hole transport materials, which works efficiently for the cells even without enough thermal annealing. It reforms the surface of the insufficiently annealed perovskite film, which leads to a normally performing solar cell without the S-shaped current density–voltage curve. For sufficiently annealed perovskite film, the small amount of PbI₂ formed and a Si–O–Si network at the interface passivates the surface traps and acts as an energy barrier to reduce recombination in the perovskite solar cells. With the amino-ended silane modification, the optimized performance of the perovskite solar cell reaches 11.8%, which shows great advantages over the original device with a performance of 8.25% (0.92 Sun, AM1.5).

---

Introduction

Organic–inorganic halide perovskite solar cells (PSC) with rapidly increased performance are very promising as highly efficient and cost-effective solar energy technologies and have been predicted as the “next big thing in photovoltaics”.^1–4 In principle, carriers are created in the perovskite absorber after absorption of incident photons and travel through transport pathways, including the electron or hole transport layer, the electrodes and each interface in between .⁵ Therefore, the electron transport at each part and the interface are critically important for high efficiency solar cells.⁶ The imperfect crystal passivation and undesirable interface properties result in fill factor (FF), open circuit voltage (V_oc) and short circuit current density (J_sc) losses.⁷ Efforts have been made on the interface modification in both planar and mesoporous types of PSCs. The efficiency is improved remarkably. However, most of the study emphasises on the interface of TiO₂/perovskite, aiming at facilitating the electron transport from the perovskite to TiO₂.^8–11 Few studies are carried out on the interface of perovskite/hole transport material (HTM).^12–14 For perovskite crystals, undercoordinated metal cations, organic cations and halide anions become the defect sites at the perovskite/HTM interface,⁷ deteriorating the electron transport property and thus the performance of the solar cells. Therefore, the perovskite/HTM interface is also very important in the device.

In our previous study, non-polar silane molecules with long alkyl chain groups are engineered at the perovskite/HTM interface. The long alkyl chains are insulating and act as an energy barrier effectively passivating the defect on the perovskite surface.¹³ This results in reduced interface recombination from TiO₂/perovskite to HTM. It is noticed that the long alkyl chain is non-polar, thus it will not change the stoichiometry of the perovskite film, which always deteriorates under polar solvents or polar molecules.^4,15

In contrast to the previous study, a type of polar silane molecule with amino-end groups is self-assembled at the surface of the perovskite/HTM. This is the first time interface modification has been used to change the stoichiometry of the perovskite and to check the effect of the method. The amino-ended silane changes the chemical composition of the perovskite surface, forming small amounts of PbI₂ and a Si–O–Si network at the interface, which acts as an energy barrier to enhance the performance of the perovskite solar cells even without sufficient thermal annealing time.

Experimental section

Preparation of perovskite solar cells

A mesoscopic solar cell is prepared as follows: a 60 nm thick dense blocking layer of TiO₂ was deposited onto an F-doped SnO₂ (FTO, Nippon, Japan) substrate by spin coating the precursor solution.^16,17 The compact TiO₂ layer is obtained after sintering at 450 °C for 30 min. A TiO₂ mesoporous layer with a thickness of about 450 nm was deposited onto the compact layer by spin coating the diluted paste of TiO₂ nanoparticles in ethanol with a weight ratio of 1 : 3.5, and sintered at 450 °C for 30 min. Furthermore, TiO₂ films were treated with 40 mM TiCl₄ solution at 70 °C for 30 min and finally sintered at 500 °C for 30 min. The perovskite solution was synthesized as reported in literature.¹ After spin coating the perovskite solution (CH₃NH₃I (synthesized) and PbCl₂ (Aldrich) with a 3 : 1 molar ratio in dimethylformamide (DMF)) onto TiO₂ film, the films on FTO glass are carefully baked on the hot plate to form a uniform perovskite film. The final annealing temperature is 95 °C. For sufficient annealing, the thermal procedure lasts 120 min, whereas the insufficiently annealed perovskite film received 100 min annealing. With very little Cl⁻ remaining in the final product, the perovskite is treated as CH₃NH₃PbI₃.²

The self-assembly monolayer of silane molecules is formed as follows: 3-aminopropyltriethoxysilane (H₂N–(CH₂)₃–Si(OC₂H₅)₃) or ‘amino-silane’ for short, is diluted in isopropanol. Different molar ratios (10, 7.5, and 5 mM) of the solution are designed to determine the optimal concentration. The prepared perovskite films were dipped in the abovementioned solution for 10 min to let the alkylalkoxysilane molecules self-assembled on the film surface. Subsequently, the films were baked for another 10 min at 80 °C to evaporate the solvent.

Subsequently, the hole transporting material (HTM) was deposited by spin-coating (at 4000 rpm for 20 s) 60 mM 2,2′,7,7′-tetrakis(N,N-di-p-methoxyphenylamine)-9,9′-spirobifluorene (spiro-MeOTAD) in chlorobenzene with an added 80 mol% tert-butylpyridine (tBP) and 30 mol% lithium bis(trifluoromethanesulfonyl)imide (Li-TFSI). Finally, 100 nm thick silver electrodes were evaporated onto the devices through a shadow mask.

Functional characterizations

The perovskite films are fabricated on a plastic substrate. They are then adhered to the Ge window for attenuated total reflectance-Fourier transform infrared (ATR-FTIR, Nicolet 560 FT-IR) measurement from 4000 to 675 cm⁻¹, with a 2 cm⁻¹ resolution over 64 scans. The surface morphologies were studied by a scanning electron microscope (SEM, Hitachi, SU-70, Japan). X-ray diffraction (XRD) patterns were collected using a Bruker D8 Advance instrument (Bruker, Germany) using Cu-K_α radiation at a scan rate of 4° min⁻¹. The abovementioned SEM, XRD, FT-IR and UV-IR results are based on the sufficiently annealed perovskite films.

Current–voltage (I–V) characteristics were measured with a 2400 source meter (Keithley) together with a sunlight simulator (Newport 3A Solar Simulator, AM 1.5, 92 mW cm⁻²) calibrated with a standard silicon reference cell. The solar cells were masked with a black aperture to define the active area of 0.07 cm². Incident photo-to-electron conversion efficiency (IPCE) spectra were obtained with monochromatic incident light in DC mode. Impedance spectra were obtained with an electrochemical workstation (Zennium, Germany) in the dark under different bias voltages. The scanning frequency was set between 10⁶ Hz and 0.1 Hz with amplitude of 10 mV as the sine perturbation bias.

Results and discussion

Fig. 1 shows the surface morphologies of the perovskite film with and without amino-silane modification. The original perovskite film (Fig. 1(a)) exhibits large crystallites and square holes. The unsuccessful coverage of perovskite with holes results in low resistance shunting paths, causing electron recombination from the electron transport material (ETM) to the hole transport material (HTM), which is detrimental to good device performance.^17–20 After amino-silane modification, the holes diminished and the films change to be more compact and smoother with increasing amino-silane concentration (Fig. 1(b)–(d)), indicating the perovskite surface structure is changed by the amino-silane. In perovskite structure, 6-fold coordinated Pb²⁺ cations occupy the corners of the unit cell, surrounded by an octahedron of X⁻ anions, with CH₃NH₃⁺ cations filling the space in the middle of the unit cell.¹² For surface modification molecules, amino-silane with an amino end group, which has a lone electron pair, may interact with the proton in the CH₃NH₃⁺ cations. Therefore, the surface stoichiometry and morphology are changed by the surface modification.

Fig. 1 The SEM of perovskite film on porous (a) TiO₂/compact TiO₂/FTO and the (b) 5 mM, (c) 7.5 mM and (d) 10 mM amino-silane modified film.

The FT-IR spectra are presented to examine the chemical modification of amino-silane on the surface of perovskite. The absorption bands at 2960 and 2880 cm⁻¹ can be assigned to the asymmetric and symmetric stretches of –CH₂ groups on amino-silane.^21,22 The increased intensity of the peaks indicates the amino-silane is anchored on the perovskite surface and the intensity increase as the modification amount increases. The vibration band at 1000–1100 cm⁻¹ is broader for amino-silane modified samples compared to the unmodified perovskite film. The broader band is ascribed to the Si–O–Si network formed by hydrolysis and condensation of the silane groups.^21,22 Interestingly, the N–H stretching in the range about 3200–3450 cm⁻¹ does change with a decreased intensity after increased amino-silane modification. It is ascribed to a loss of CH₃NH₃⁺ groups following amino-silane modification. It is known that amino-silane is a polar molecule with a lone electron pair on the amino end-group. When the perovskite film is immersed in the amino-silane/isopropanol solution, the CH₃NH₃⁺ group in perovskite may be displaced from the lattice structure. From the FT-IR spectroscopy, it is deduced that the amino-silane is assembled on the perovskite surface and forms a Si–O–Si network; moreover, the CH₃NH₃⁺ groups in the perovskite surface structure may interact with the amino groups in silane and be desorbed from the surface (Fig. 2).

Fig. 2 ATR-FTIR spectra of amino-silane modified perovskite film.

The XRD pattern of perovskite film on glass with and without amino-silane modification was checked to reveal any crystalline structure change (Fig. 3). Without the surface modification, the peaks at 14.06°, 20.04°, 23.54°, 28.40° and 31.89° are assigned to the (110), (112), (211), (220) and (312) planes of CH₃NH₃PbI₃, respectively. After amino-silane surface modification, the small diffraction peak at 12.5° emerged (noted by black circle, Fig. 3) and is ascribed to a PbI₂ phase.²³ Moreover, other diffraction peaks belonging to crystalline perovskite are not changed compared with the unmodified sample. The small diffraction peak from PbI₂ indicates that a small amount of PbI₂ is generated after surface modification. It was proposed from the FT-IR spectra that the CH₃NH₃⁺ groups (especially the excess and unstable ones) may be desorbed from the lattice structure in amino-silane polar solution, therefore leaving a PbI₂ phase on the surface of the perovskite.

Fig. 3 The XRD spectra of perovskite with 0, 5, 7.5 and 10 mM amino-silane solution modification.

The change in surface composition was further checked by UV-IR spectroscopy. The absorption spectrum of unmodified and 5 mM amino-silane modified film show a typical perovskite absorption onset at 770 nm. For higher modification amounts (7.5 and 10 mM), a new peak at around 410 nm appears (arrows in Fig. 4), corresponding to the PbI₂ absorbance.²⁴ This indicates that after amino-silane modification, PbI₂ emerged at the surface, which is consistent with the XRD spectra.

Fig. 4 UV-IR spectra of perovskite film on glass with amino-silane modification.

It is concluded from the abovementioned characterization that the amino-silane is assembled on the surface forming a Si–O–Si network. At the same time, the amino end-group may interact with the CH₃NH₃⁺ groups and desorb them to form a PbI₂ phase on the surface of the perovskite film. The interface modification mechanism is shown in Scheme 1.

Scheme 1 The interface modification mechanism of amino-silane on perovskite film (the green part represents the PbI₆ octahedra and the blue circles are CH₃NH₃⁺ cations).

The morphology and composition of the resulting perovskite thin film plays a central role in the PSC's performance. Compared with the vacuum deposition method in the solution-processing method, it is more difficult to control the perovskite crystallization and film uniformity at the annealing stages. The situations of insufficient annealing and over-annealing often result in nonstoichiometric perovskite films. The consequent nonstoichiometric material is comprised a local excess of positive or negative ions. Together with the holes, they serve as the traps at the interface, which deteriorates the carrier transport and accelerates interface recombination in working solar cells. With insufficient annealing time, the perovskite solar cells are reported to exhibit very low fill factors (even S-shape of the current density versus voltage curves), which indicates the severe interface recombination and poor charge transport properties.^14,25 The composition of the insufficiently annealed perovskite film deviates from stoichiometry (with residuals of unevaporated solvent and CH₃NH₃⁺), thus bearing lots of trap states. Fig. 5 shows the performance of a perovskite solar cell with insufficient annealing time. It exhibits J_sc of 14.47 mA cm⁻², V_oc of 0.941 V, a low FF of 0.41 and efficiency of 6.1%. By contrast, with an amino-silane modified perovskite/HTM interface, the performance increased to 9.1% with J_sc of 15.76 mA cm⁻², V_oc of 0.91 V and a much enhanced FF of 0.58. The enhanced FF and performance are due to the polar amino-silane modification of the nonstoichiometric perovskite film, which may desorb the excess CH₃NH₃⁺ groups and the residual solvent.

Fig. 5 The J–V curves of unmodified and 10 mM amino-silane modified insufficiently annealed perovskite solar cells.

The amino-silane is also applied on sufficiently annealed perovskite film to examine the interface modification effects. Current density versus voltage (J–V) characteristics of the perovskite photovoltaic cells are shown in Fig. 6(a) and the detailed photovoltaic parameters are shown in Table 1. A batch of more than 35 solar cells was tested and the V_oc and FF distribution is indicated in Fig. S1.† The original perovskite solar cell exhibits J_sc of 16.77 mA cm⁻², V_oc of 0.856 V, FF of 0.529 and efficiency of 8.25%. After amino-silane modification, both V_oc and FF clearly improve. V_oc improves by 46 mV at 5 mM concentration and 70 mV at 10 mM modification. A change in V_oc correlates with the rate of charge carrier extraction and recombination.²⁶ After interface modification, the improved V_oc is ascribed to decreased interface recombination. The PbI₂ is reported to possess a larger bandgap than perovskite, CH₃NH₃PbI₃;^14,27,28 moreover, the Si–O–Si network is insulating, both of which serve as energy barriers at the interface, blocking the interface recombination. Thus, the V_oc is improved. This energy barrier also helps to increase the shunt resistance of the solar cells; the FF is also enhanced at the same time. With the increase of V_oc and FF, the performance of the modified solar cell is remarkably improved to 11.8%, with J_sc of 16.55 mA cm⁻², V_oc of 0.902 V and FF of 0.725. The cells with amino-silane modification measured in the dark exhibit smaller current compared with the original perovskite solar cell under the same bias voltage, which also indicates the blocking effect of the interface. The energy barrier property of the amino-silane network formed at the interface is indicated in Fig. 6(b). It is noticed that the J_sc is decreased slightly with higher amino-silane amounts (7.5 and 10 mM). From the XRD and UV-IR characterization, it is known that the formation of PbI₂ at the interface occurs at the price of desorbing the CH₃NH₃⁺ groups from the surface. Due to the destroyed perovskite structure at the interface (especially for the higher amino-silane modification amounts), the electron generation rate is impaired and the photocurrent is thus reduced. IPCE spectra are presented in Fig. 6(c), showing the same tendency for change as the J_sc of the solar cells. The hysteresis effect is tested for the modified and original device (Fig. 6(d)). It is found that the amino-silane treated device shows a smaller hysteresis phenomenon compared to the unmodified device. This reduced hysteresis might be due to the improved interface property²⁹ that decreases electron recombination and improves the ability of electrons to flow out the device.

Fig. 6 (a) J–V curves of perovskite photovoltaic cells; (b) the interface blocking property of the formed amino-silane network; (c) IPCE spectra of the devices; and (d) I–V curves of the devices under forward and reverse scan (scan delay time: 30 ms).

Table 1 The photovoltaic parameters of the perovskite solar cells with amino-silane modifications

Concentration [mM]	Jsc [mA cm^−2]	Voc [V]	FF	Eff [%]
0	16.77	0.856	0.529	8.25
10	16.0	0.926	0.55	8.79
7.5	16.55	0.902	0.725	11.8
5	18.6	0.902	0.60	10.9

---

Electrochemical impedance spectroscopy (EIS) is used to monitor the interfacial changes of the perovskite solar cell system.^30,31 Two arcs appear in the Nyquist plots in Fig. 7(a): high frequency arc (left part) is ascribed to the contact resistance (R_H) at the HTM/Ag electrode; whereas the lower frequency one (right part) is associated with the recombination resistance (R_rec) and chemical capacitance (C_μ) of the interface.³² The R_rec, which is the diameter of the middle frequency arc, is related to the recombination of electrons in perovskite/TiO₂ with hole transport layer (HTM). It is clear that the R_rec is much larger for the 7.5 mM amino-silane modified device in the dark at a bias of 0.8 V compared with the original device, which means that the charge recombination is blocked. The electron lifetime, τ, plotted with the bias voltage is presented in Fig. 6(b). It is clear that the 7.5 mM amino-silane modified perovskite solar cell exhibits higher electron lifetime compared with the original device. The EIS results are consistent with the photovoltaic property of the devices, which indicates that the amino-silane modification induces an energy barrier at the interface, decreasing the interface recombination.

Fig. 7 (a) Nyquist plots of the perovskite solar cells. The inset is the equivalent circuit of the solar cell; (b) the electron lifetime versus bias voltage of the original and 7.5 mM amino-silane modified solar cells.

Conclusions

In this study, a polar molecule of amino-ended silane has modified the surface of perovskite. FT-IR, XRD and UV-IR analyses indicate that a small portion of CH₃NH₃⁺ groups are desorbed, thus forming a phase of PbI₂ at the interface. Together with the Si–O–Si network formed, the interface layer acts as an energy barrier to reduce interface recombination in the perovskite solar cell. The insufficiently annealed perovskite solar cells, which show S-shaped J–V curves, perform normally after amino-silane modification, whereas sufficiently annealed solar cells show prominent improvement in FF, V_oc and their performance. The amino-silane polar molecules are verified to be effective in interface modification in perovskite solar cells.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (contact grant number: 51302137, 11174163 and 11374168), China; the K.C. Wong Magna Fund in Ningbo University, China; and the Qiangjiang Talented Person Project of Zhejiang (No. QJD1302008), China.

Notes and references

1. J. Burschka, N. Pellet and S. J. Moon, et al., Sequential deposition as a route to high-performance perovskite-sensitized solar cells, Nature, 2013, 499(7458), 316–319.
2. M. Gratzel, The light and shade of perovskite solar cells, Nat. Mater., 2014, 13(9), 838–842.
3. W. S. Yang, J. H. Noh and N. J. Jeon, et al., High-performance photovoltaic perovskite layers fabricated through intramolecular exchange, Science, 2015, 348(6240), 1234–1237.
4. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, Organometal Halide Perovskites as Visible-Light Sensitizers for Photovoltaic Cells, J. Am. Chem. Soc., 2009, 131(17), 6050–6051.
5. H. Zhou, Q. Chen and G. Li, et al., Interface engineering of highly efficient perovskite solar cells, Science, 2014, 345(6196), 542–546.
6. J. Min, Z.-G. Zhang and Y. Hou, et al., Interface Engineering of Perovskite Hybrid Solar Cells with Solution-Processed Perylene–Diimide Heterojunctions toward High Performance, Chem. Mater., 2015, 27(1), 227–234.
7. X. Wu, M. T. Trinh and D. Niesner, et al., Trap States in Lead Iodide Perovskites, J. Am. Chem. Soc., 2015, 137(5), 2089–2096.
8. L. Zuo, Z. Gu and T. Ye, et al., Enhanced photovoltaic performance of CH₃NH₃PbI₃ perovskite solar cells through interfacial engineering using self-assembling monolayer, J. Am. Chem. Soc., 2015, 137(7), 2674–2679.
9. J. C. Yu, D. B. Kim and G. Baek, et al., High-Performance Planar Perovskite Optoelectronic Devices: A Morphological and Interfacial Control by Polar Solvent Treatment, Adv. Mater., 2015, 27, 3465.
10. J. T.-W. Wang, J. M. Ball and E. M. Barea, et al., Low-Temperature Processed Electron Collection Layers of Graphene/TiO₂ Nanocomposites in Thin Film Perovskite Solar Cells, Nano Lett., 2014, 14(2), 724–730.
11. Z. Wu, S. Bai and J. Xiang, et al., Efficient planar heterojunction perovskite solar cells employing graphene oxide as hole conductor, Nanoscale, 2014, 6(18), 10505–10510.
12. A. Abate, M. Saliba and D. J. Hollman, et al., Supramolecular Halogen Bond Passivation of Organic–Inorganic Halide Perovskite Solar Cells, Nano Lett., 2014, 14, 3247–3254.
13. J. Zhang, Z. Hu and L. Huang, et al., Bifunctional alkyl chain barriers for efficient perovskite solar cells, Chem. Commun., 2015, 51(32), 7047–7050.
14. Q. Chen, H. Zhou and T.-B. Song, et al., Controllable Self-Induced Passivation of Hybrid Lead Iodide Perovskites toward High Performance Solar Cells, Nano Lett., 2014, 14(7), 4158–4163.
15. W. Li, H. Dong and L. Wang, et al., Montmorillonite as bifunctional buffer layer material for hybrid perovskite solar cells with protection from corrosion and retarding recombination, J. Mater. Chem. A, 2014, 2(33), 13587–13592.
16. E. Edri, S. Kirmayer, D. Cahen and G. Hodes, High Open-Circuit Voltage Solar Cells Based on Organic–Inorganic Lead Bromide Perovskite, J. Phys. Chem. Lett., 2013, 4(6), 897–902.
17. G. E. Eperon, V. M. Burlakov, P. Docampo, A. Goriely and H. J. Snaith, Morphological Control for High Performance, Solution-Processed Planar Heterojunction Perovskite Solar Cells, Adv. Funct. Mater., 2014, 24(1), 151–157.
18. N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. Ryu and S. I. Seok, Solvent engineering for high-performance inorganic–organic hybrid perovskite solar cells, Nat. Mater., 2014, 13(9), 897–903.
19. M. Xiao, F. Huang and W. Huang, et al., A Fast Deposition–Crystallization Procedure for Highly Efficient Lead Iodide Perovskite Thin-Film Solar Cells, Angew. Chem., 2014, 126(37), 10056–10061.
20. Y. Wu, A. Islam and X. Yang, et al., Retarding the crystallization of PbI₂ for highly reproducible planar-structured perovskite solar cells via sequential deposition, Energy Environ. Sci., 2014, 7(9), 2934–2938.
21. L. Liu, A. Mei and T. Liu, et al., Fully Printable Mesoscopic Perovskite Solar Cells with Organic Silane Self-Assembled Monolayer, J. Am. Chem. Soc., 2015, 137(5), 1790–1793.
22. J. Zhang, Y. Yang and S. J. Wu, et al., Improved photovoltage and performance by aminosilane-modified PEO/P(VDF-HFP) composite polymer electrolyte dye-sensitized solar cells, Electrochim. Acta, 2008, 53(16), 5415–5422.
23. A. Dualeh, N. Tétreault, T. Moehl, P. Gao, M. K. Nazeeruddin and M. Grätzel, Effect of Annealing Temperature on Film Morphology of Organic–Inorganic Hybrid Pervoskite Solid-State Solar Cells, Adv. Funct. Mater., 2014, 24(21), 3250–3258.
24. D. Liu and T. L. Kelly, Perovskite solar cells with a planar heterojunction structure prepared using room-temperature solution processing techniques, Nat. Photonics, 2014, 8(2), 133–138.
25. J. J. Shi, H. Y. Wei and L. F. Zhu, et al., S-shaped current–voltage characteristics in perovskite solar cell, Acta Biochim. Biophys. Sin., 2015, 64(3), 38402–038402.
26. J.-H. Im, I.-H. Jang, N. Pellet, M. Grätzel and N.-G. Park, Growth of CH₃NH₃PbI₃ cuboids with controlled size for high-efficiency perovskite solar cells, Nat. Nanotechnol., 2014, 9(11), 927–932.
27. D. Bi, A. M. El-Zohry, A. Hagfeldt and G. Boschloo, Unraveling the Effect of PbI₂ Concentration on Charge Recombination Kinetics in Perovskite Solar Cells, ACS Photonics, 2015, 2(5), 589–594.
28. L. Wang, C. McCleese, A. Kovalsky, Y. Zhao and C. Burda, Femtosecond Time-Resolved Transient Absorption Spectroscopy of CH₃NH₃PbI₃ Perovskite Films: Evidence for Passivation Effect of PbI₂, J. Am. Chem. Soc., 2014, 136(35), 12205–12208.
29. H. J. Snaith, A. Abate and J. M. Ball, et al., Anomalous Hysteresis in Perovskite Solar Cells, J. Phys. Chem. Lett., 2014, 5(9), 1511–1515.
30. V. Gonzalez-Pedro, E. J. Juarez-Perez and W.-S. Arsyad, et al., General Working Principles of CH₃NH₃PbX₃ Perovskite Solar Cells, Nano Lett., 2014, 14(2), 888–893.
31. B. Suarez, V. Gonzalez-Pedro, T. S. Ripolles, R. S. Sanchez, L. Otero and I. Mora-Sero, Recombination Study of Combined Halides (Cl, Br, I) Perovskite Solar Cells, J. Phys. Chem. Lett., 2014, 5(10), 1628–1635.
32. D. Liu, J. Yang and T. L. Kelly, Compact Layer Free Perovskite Solar Cells with 13.5% Efficiency, J. Am. Chem. Soc., 2014, 136(49), 17116–17122.

---

Footnote

† Electronic supplementary information (ESI) available: The V_oc (a) and FF (b) distribution of the perovskite photovoltaic cells. See DOI: 10.1039/c5ra21698b

---