Low-temperature processed high-performance flexible perovskite solar cells via rationally optimized solvent washing treatments

Abstract

High performance planar-heterojunction (PHJ) perovskite (CH₃NH₃PbI₃) solar cells fabricated through low-temperature annealing are demonstrated. Simple spin-coating with an optimized solvent washing process readily forms homogeneous and crystalline perovskite thin films. The perovskite films fabricated via this solvent washing process show a low dependence on annealing temperature in achieving high crystallinity and large grain size, prerequisites for high efficiency perovskite solar cells. The solar cell device fabricated by solvent washing and 100 °C annealing exhibited a high power conversion efficiency (PCE) over 14% with high short circuit current density (J_SC) of 19.3 mA cm⁻² and fill factor (FF) of 0.80. More importantly, the device annealed at low temperature (<90 °C) also yields high PCEs of over 12%. This enables us to fabricate flexible solar cells at low-temperatures with promising PCE as high as 9.43%. This study demonstrates that this optimized solvent washing process is highly relevant for low-cost roll-to-roll (R-2-R) processing of high performance perovskite solar cells.

---

Introduction

In the past few years, organometal halide perovskites (CH₃NH₃PbX₃, X = Cl⁻, Br⁻, and I⁻) have been vigorously investigated and considered as very efficient light-absorbing photoactive materials for next-generation solar cells.¹ In addition to their intense broadband absorption, this class of perovskites also possesses small exciton binding energies (∼kT at room temperature), long charge carrier diffusion lengths (100–1000 nm) and lifetimes (∼100 ns), decent ambipolar charge mobility, and low material-cost. These merits have attracted worldwide attention in the photovoltaic field,² but even more significant than these properties is the ease with which organometal halide perovskite materials can be solution processed at low temperatures. This makes the material readily translatable into large area, commercially viable production. Thus far, power conversion efficiencies (PCEs) of solution-processed perovskite solar cells (PSCs) have rapidly grown, currently exceeding 19%, partially as a result of the variety of processing techniques and device architectures that have been continuously developed.³

To date, most efficient PSCs with PCE over 15% are fabricated on meso-superstructured architectures.⁴ Although such perovskite meso-superstructured solar cells (MSSCs) have shown both high performance and decent stability, fabricating pure metal oxide (TiO₂) as an electron-transporting (hole blocking) scaffold requires high-temperature (>450 °C) processing.⁵ Such processing conditions may impede the future development of PSCs, particularly for application in soft electronic devices. In this respect, low-temperature processed TiO₂ or Al₂O₃ nanoparticle based scaffolds have been exploited for perovskite MSSCs; however, the device fabrication is still complicated and processing over 100 °C is still required.⁶

As an alternative to MSSCs, planar-heterojunction (PHJ) PSCs employing organic semiconductors as hole- and electron-transporting layers have also been widely investigated and appear to be an appealing system for high-throughput low-temperature manufacturing.⁷ With regard to the superior low-temperature solution processability of organic semiconductors, this device configuration is compatible for roll-to-roll (R-2-R) fabrication on flexible substrates. Moreover, low-temperature processing permits the use of a wide variety of interfaces, electrodes, and substrates. Although a few research groups have attempted to fabricate devices at low-temperatures on glass or flexible substrates, the perovskite layers still need to be annealed over 90 °C, which is higher than the glass-transition temperature (T_g) of typical flexible substrates such as polyethylene terephthalate (PET). As a result, it is desirable to develop an efficient low-temperature route (<90 °C) to form highly crystalline and morphologically uniform perovskite films to enable the R-2-R processing of high-performance flexible PSCs.⁸

In developing such a low-temperature route, an inevitable challenge is the slow evaporation of polar processing solvents like dimethyl sulfoxide (DMSO) or γ-butyrolactone (GBL) that have high boiling points (>190 °C). To completely evaporate these polar solvents, it is necessary to perform thermal treatment over 100 °C, which has a great influence on the resulting morphology leading to rough and inhomogeneous perovskite films.⁹ Very recently, Seok et al. reported that non-solvent (toluene) washing during spin-coating of the perovskite (CH₃NH₃PbI₃) precursor solution in a mixed solvent system (DMSO and GBL) significantly improves thin film morphology with high uniformity and coverage.¹⁰ By utilizing this solvent engineering strategy, a high-performance PSC with a certified PCE of 16.2% was successfully demonstrated. In spite of this high quality perovskite film, additional high temperature annealing to remove DMSO from the intermediate phase (CH₃NH₃I–PbI₂–DMSO) is still required. This step will increase processing cost and hinder extension to flexible plastic substrates. Therefore, it is critical to further optimize the solvent engineering to develop low-temperature processing for high-performance flexible PSCs.

In this study, we describe low-temperature processed high performance flexible PSCs by rationally optimizing the solvent washing process through advanced solvent engineering. We reveal that a non-polar washing solvent with larger dipole moment solvents than toluene (0.36 D) such as chlorobenzene (CB, 1.54 D) or o-dichlorobenzene (DCB, 2.14 D) can induce transformation directly to the crystalline perovskite phase (CH₃NH₃PbI₃), bypassing the intermediate phase (CH₃NH₃I–PbI₂–DMSO) observed when toluene is used as washing solvent.¹¹ As a result, an extremely uniform and highly crystalline perovskite film could be achieved in devices showing PCE as high as 14% with high fill factor (FF) of over 0.80. Furthermore, due to the direct formation of the highly crystalline perovskite phase after CB washing, the thermal annealing process can be reduced to <90 °C while still preserving high performance. Based on this achievement, we demonstrate a flexible solar cell with a promising PCE of 9.4% fabricated through low-temperature thermal annealing at 70 °C.

Results and discussion

To investigate the influence of the washing solvent's dipole moment on the spin-coated perovskite films, toluene, chlorobenzene (CB) and o-dichlorobenzene (DCB) were employed. During the spin-coating of the perovskite precursor solutions, the color of the perovskite film turned instantly to light brown and dark brown colors upon solvent washing by toluene and CB/DCB, respectively. The dark color of the perovskite film is an indication for the formation of crystalline CH₃NH₃PbI₃ phase. In contrast to the solvent-washed films, wet film deposited without solvent washing dried very slowly, resulting in a transparent and pale yellow film after 30 min. The morphologies of perovskite thin films were analyzed by scanning electron microscopy (SEM). As shown in Fig. 1 and S1,† the perovskite film without solvent washing exhibits the texture of elongated crystals structure with lengths of ∼2 μm. After thermal annealing at 100 °C, the film becomes dark and the morphology is dominated by fine crystallites (<100 nm) with poor coverage. In the case of the toluene-washed film, a mix of large, elongated crystals (length over 5 μm) and very small crystals (<100 nm) were observed. These crystals were converted to densely packed but small perovskite domains (∼100 nm) after thermal annealing at 100 °C. Interestingly distinct to the case of toluene, a remarkably different morphology with uniform and densely packed domains (200–300 nm) was observed when CB, a more polar solvent than toluene, was used as wash solvent.

Fig. 1 SEM images of surface morphology of CH₃NH₃PbI₃ thin films fabricated through solvent washing (top) and subsequent thermal annealing at 100 °C (bottom). (a) No solvent washing, (b) toluene washing, (c) CB washing, (d) DCB washing. The scale bar indicates 2 μm. Large images are in ESI.†

After thermal annealing, film morphology became smooth and grain size increased to ∼500 nm, which is much larger than that of the toluene-washed film. Similar to CB, DCB washing also generates smooth and uniform films with domain sizes of ∼100 nm, much smaller than that of the CB-washed film. While grain growth also occurs during annealing in this case, producing uniform domains with sizes of 150–200 nm, many pin-holes are generated as well. This might arise from the slow evaporation of DCB due to its high boiling point.

X-Ray diffraction (XRD) of the studied perovskite thin films was investigated to analyze phase composition after solvent washing. As shown in Fig. 2, the film without solvent washing shows major diffraction peaks at 6.45°, 7.10°, and 9.10° which are associated with the CH₃NH₃I–PbI₂–DMSO complex, an intermediate phase reported previously.¹⁰ In addition to this intermediate, weak diffraction signals at 13.95° and 28.35° belonging to (110) and (220) planes of the tetragonal CH₃NH₃PbI₃ phase are also observed, which indicates the coexistence of intermediate and CH₃NH₃PbI₃ crystallites after spin-coating. After thermal annealing, the signals associated with the intermediate phase diminished while the intensity of the characteristic perovskite peaks increases indicating transformation through DMSO loss. In the toluene-washed film, diffraction signals from both the intermediate and CH₃NH₃PbI₃ are apparent which is analogous to the phase composition of the unwashed film. This indicates that the intermediate phase rapidly precipitated as toluene washed out the solvents composing the perovskite precursor solution, resulting in large crystals as is apparent in the SEM data. As shown in Fig. 2(b), the intermediate phase completely converts to the crystalline perovskite phase upon thermal annealing at 100 °C. In stark contrast, thin films washed by CB and DCB only show diffraction signals characteristic of CH₃NH₃PbI₃. This signifies that CB and DCB washing change the transformation pathway through which the evolving perovskite film skipped the formation of large-scale perovskite–DMSO intermediate and form perovskite crystallites directly.

Fig. 2 XRD diffractograms of the CH₃NH₃PbI₃ thin films fabricated through solvent washing and subsequent thermal annealing at 100 °C for 15 min. (a) No solvent washing, (b) toluene washing, (c) CB washing, (d) DCB washing.

To gain more understanding of the way these solvent washes produce such differing transformations, we have further examined the solubility change of the precursor materials (CH₃NH₃I and PbI₂) caused by adding each washing solvent. When 30 vol% toluene was added to the precursor solution, a pale yellow precipitate formed immediately as shown in Fig. 3. This pale yellow solid was identified as the intermediate phase of CH₃NH₃I–PbI₂–DMSO by XRD as reported previously (Fig. S2†). Nevertheless, in the case of adding CB and DCB, this kind of precipitation is dramatically decreased. Interestingly, the solubility of the intermediate phase in the polar washing solvents improves with increased solvent dipole moment. This links solvent dipole moment and the stability of the intermediate phase, providing a simple means to retard or encourage its formation during solvent washing process.

Fig. 3 Photograph of the CH₃NH₃I and PbI₂ precursor solutions in DMSO and GBL after addition of non-polar solvents.

Based on the above observations, the underlying mechanism of film formation during deposition employing solvent washing is presented in Scheme 1. For as-cast perovskite thin films, the solvent still remains and the intermediate phase precipitates as the solvent slowly evaporates. After thermal annealing, the DMSO in the intermediate phase volatilizes resulting in the formation of crystalline CH₃NH₃PbI₃. However, as toluene was added in situ during spin-coating, solvents were washed out and the perovskite transformation is quickly quenched. This results in a mixture of the intermediate phase, crystalline perovskite, and precursor salts in the thin film. After annealing at 100 °C, the intermediates are converted into the crystalline CH₃NH₃PbI₃ phase, forming a uniform perovskite thin film. While in the case of CB, it also washes solvent out during spin-coating, however, CH₃NH₃PbI₃ readily develops without the formation of the intermediate phase due to its high solubility in the mixture of CB and polar solvents, yielding uniform and densely-packed highly crystalline perovskite films. After thermal annealing, crystallinity and grain size of perovskite increase and leads to a homogeneous, densely packed perovskite thin film which is critical for efficient charge generation and transport for high photovoltaic performance.

Scheme 1 Schematic illustration of the different morphology development with different solvent washing process.

To verify the effect of washing solvent polarity on film formation, PHJ devices with the configuration of (ITO/PEDOT:PSS/CH₃NH₃PbI₃ (300 nm)/PC₆₁BM (50 nm)/bis-C₆₀ (10 nm)/Ag (150 nm)) were fabricated as shown in Fig. 4.¹² Fig. 5 presents the J–V curves obtained from the studied devices with or without using the optimized solvent washing process (see Table 1), and corresponding incident photon-to-current efficiency (IPCE) spectra are shown in Fig. S3.† The devices fabricated without solvent washing exhibited a poor PCE of only 3.77%, mainly because of shunting, low light absorption, and inefficient charge transport arising from incomplete surface coverage and low crystallinity. In contrast, devices with toluene washing and subsequent annealing at 100 °C showed a much improved PCE of 11% with a high short circuit current density (J_SC) of 16.9 mA cm⁻², a good FF of 0.72 and a V_OC of 0.91 V. This huge difference in performance affirms that the formation of an ideal perovskite film morphology is very critical factor for achieve high photovoltaic performance.¹³

Fig. 4 Device architecture of the PSCs fabricated in this work.

Fig. 5 J–V curves of PSCs with different solvent washing process.

Table 1 Parameters of the PSCs with different solvent washing process

Non-polar solvent	VOC (V)	JSC (mA cm^−2)	FF	PCEbest (%)	PCEaverage^a (%)
a More than 20 solar cells were fabricated for each condition.
N/A	0.56	13.2	0.49	3.77	3.62
Toluene	0.91	16.9	0.72	11.4	11.1
CB	0.91	19.3	0.80	14.1	13.7
DCB	0.87	18.9	0.71	12.0	11.5

---

When CB was used as a washing solvent, the resultant device exhibited an even higher PCE of over 14% with a J_SC of 19.2 mA cm⁻², a FF of 0.80 and a V_OC of 0.91 V. The enhanced FF and J_SC apparently originate from the increased crystallinity of the perovskite thin film. It is well documented that CH₃NH₃PbI₃ crystallinity affects charge generation, transport, and recombination; and that the perovskite grain boundary is the major charge recombination site.¹³ Therefore, the increased FF and J_SC in the CB-washed film can be ascribed to the larger and more crystalline domains which also correlates with reduced grain boundary area. Note that the increased crystallinity of the CB-washed perovskite film also yields improved light absorption with respect to the toluene washed film as shown in Fig. S3.† However, DCB washing results in lower performance, J_SC, and FF owing to the small perovskite domain size. As mentioned previously, many pin-holes were observed in DCB-washed film possibly due to slow evaporation during thermal annealing. Therefore, DCB-washed devices exhibited inferior performance compared to CB-washed PSCs.

Recently, hysteresis in J–V curves under different bias sweep directions has emerged as an important issue in the accurate characterization of perovskite-based devices.¹⁴ When the photovoltaic performance of the CB-washed PSCs were measured with different sweep directions, no significant hysteresis was observed between forward and reverse bias sweeps as shown in Fig. S4,† indicating good reliability of the data obtained.

As shown above, CB washing readily induces the formation of crystalline perovskite, which inspires us to lower the annealing temperature to well below 100 °C to realize low-temperature processed solar cells on flexible substrates. The color of the CB-washed perovskite film instantly changes to black upon exposure to temperatures at or above 80 °C. Surprisingly, annealing at lower temperatures (60 and 70 °C) also affords a black perovskite film with a highly reflective and smooth surface after just 10 min, as shown in Fig. S5.† The crystallinity of low-temperature processed perovskite films was examined by XRD as shown in Fig. 6. As annealing time increases, the diffraction peaks of crystalline CH₃NH₃PbI₃ become more pronounced. In the case of low temperature annealed samples, intensity of the (110) reflection is dramatically enhanced after 60 min and further increases slightly after 120 min while higher annealing temperatures (80 and 90 °C) improve perovskite crystallinity more quickly. However, decomposition of CH₃NH₃PbI₃ occurs yielding pure PbI₂ (2θ at ∼12.7°) as shown in Fig. 6(c) and (d) upon extended annealing at higher temperatures.¹⁵ When the perovskite film was annealed at 90 °C, PbI₂ was observed just after 60 min.

Fig. 6 XRD diffractograms of the CB-washed CH₃NH₃PbI₃ thin films after different thermal annealing temperature and time.

This low-temperature annealing also enhances light absorption and crystal quality of the perovskite film. While annealing enhances the absorption of both low and high temperature processed films as shown in Fig. S6,† a slight decrease in absorbance was observed after 120 min of annealing at higher temperatures (80 and 90 °C), suggesting some perovskite decomposition.

The microstructure of the low-temperature annealed perovskite film was also observed by SEM as shown in Fig. S7.† As the annealing temperature and time increase, the grain size increases and the morphology gradually roughens. We fabricated devices via CB washing and subsequent thermal annealing at low temperature. The photovoltaic parameters are presented in Table 2 and corresponding J–V curves are shown in Fig. 7. The devices annealed at 60 °C for 10 min exhibit a low PCE of 4.20%. The low performance was attributed to small grain size and poor crystallinity, which limit efficient charge generation and transport and result in low V_OC, J_SC, and FF. As the annealing time increases to 120 min, the PCEs of devices also increase to ∼9%. However, when the devices were annealed at 70 °C for 10 min, much better V_OC (>0.88 V) and FF (>0.70) were observed, indicating charge carriers are efficiently transported to electrodes without significant recombination. Encouragingly, 120 min of annealing at 70 °C yielded 12% PCE with high J_SC and high FF, which is comparable to the performance of the device processed at 100 °C. Higher annealing temperature (80 and 90 °C) also gives high performance (PCE > 12%). The high performance of low-temperature processed perovskite devices is ascribed to the uniform and highly crystalline perovskite grains encouraged by the CB washing process.

Table 2 Parameters of the PSCs with different annealing temperature

Temperature (°C)	Time (min)	VOC (V)	JSC (mA cm^−2)	FF	PCEbest (PCEave) (%)
60	10	0.73	11.5	0.50	4.20 (3.95)
	30	0.78	13.9	0.64	6.94 (6.87)
	60	0.84	15.1	0.66	8.37 (8.13)
	120	0.86	15.6	0.67	8.99 (8.81)
70	10	0.88	11.7	0.74	7.62 (7.52)
	30	0.89	14.0	0.75	9.35 (9.11)
	60	0.90	15.8	0.76	10.8 (10.5)
	120	0.89	17.7	0.76	12.0 (11.8)
80	10	0.88	13.3	0.75	8.78 (8.71)
	30	0.89	15.6	0.76	10.6 (10.3)
	60	0.90	17.5	0.77	12.1 (11.7)
	120	0.91	17.6	0.78	12.5 (12.1)
90	10	0.88	16.0	0.77	10.8 (10.4)
	30	0.90	16.6	0.76	11.3 (11.0)
	60	0.89	18.5	0.77	12.7 (12.5)
	120	0.89	18.8	0.77	12.9 (12.7)

---

Fig. 7 J–V curves of the CB-washed devices with thermal annealing at (a) 60 °C, (b) 70 °C, (c) 80 °C and (d) 90 °C.

The charge lifetime of low-temperature processed perovskite films prepared via CB washing was characterized by time-resolved photoluminescence (TRPL). Detailed information regarding the preparation, measurement, and fitting methodology are described in the ESI.† The PL lifetime of the high-temperature (100 °C) annealed film was fitted with a bi-exponential decay function considering the fast and slow decay to be related to the quenching of free carriers by a quencher (PEDOT:PSS and PC₆₁BM) and radiative decay, respectively. The PL decay of the perovskite film annealed at high-temperature exhibits a time-constant of τ_ave = 125.5 ns which is reduced to 5.7 ns as the quencher layers are introduced, as is shown in Fig. S8.† The low-temperature processed perovskite film shows similar carrier lifetimes of 113.8 ns without quencher and 5.1 ns with quencher layers. Therefore, it can be concluded that this low-temperature annealed perovskite film supports comparable charge generation and transport to those derived from high-temperature processed films. In other words, CB-washed perovskite films show a weak dependence on annealing temperature.

Finally, a flexible provskite device on an ITO/plastic substrate (PET) was fabricated utilizing the CB washing process. Since the T_g of the PET is known to be ∼70–80 °C, the device was annealed at 70 °C for 120 min. As shown in Fig. 8, the flexible device showed a promising PCE of 9.43% with a high FF of 0.75, which is one of the highest performing flexible perovskite devices reported to date. After bending 20 times, this device still exhibited a PCE of 8.42% retaining 89% of its initial performance (see ESI†). This is the highest PCE for flexible PSCs reported so far (see ESI†). Moreover, this is the first demonstration of a low-temperature annealed device (<90 °C) which was fabricated via simple spin-coating process.

Fig. 8 J–V curve of the flexible perovskite solar cell device fabricated via CB-washing and thermal annealing at 70 °C. Inset is a photograph of the flexible device.

Conclusions

In this work, we optimize the solvent washing process during the spin-coating of a perovskite precursor solution utilizing non-polar solvents. We demonstrate that the polarity of the washing solvent has critical influence on transformation pathway via its effect on the formation of the intermediate phase, which proves to be a major obstacle to lowering annealing temperature while maintaining a highly crystalline perovskite film. Specifically, CB was found to be the optimum washing solvent for fabricating uniform, densely-packed, and highly crystalline perovskite films. The solar cells utilizing CB washing and high temperature annealing yielded a promising PCE of over 14% with high a J_SC and FF. Furthermore, low-temperature annealing facilitated by CB washing also produces highly crystalline and uniform perovskite films, which enables device PCE to be over 12% on glass substrates and 9% on plastic substrates. Our results provide a promising and effective methodology to achieve high quality perovskite thin films via a low-temperature processing route for high performance perovskite solar cells. Moreover, this approach demonstrates the practical fabrication of high performance flexible perovskite solar cells.

Acknowledgements

This work is supported by the Air Force Office of Scientific Research (FA9550-09-1-0426), the Office of Naval Research (N00014-11-1-0300), the Asian Office of Aerospace R&D (FA2386-11-1-4072) and the SunShot Initiative (DE-EE0006710). A. K.-Y. Jen thanks the Boeing-Johnson Foundation for financial support.

Notes and references

1. (a) M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami and H. J. Snaith, Science, 2012, 338, 643–647; (b) H.-S. Kim, C.-R. Lee, J.-H. Im, K.-B. Lee, T. Moehl, A. Marchioro, S.-J. Moon, R. Humphry-Baker, J.-H. Yum, J. E. Moser, M. Grätzel and N.-G. Park, Sci. Rep., 2012, 2, 591; (c) M. Liu, M. B. Johnston and H. J. Snaith, Nature, 2013, 501, 395–398; (d) D. Liu and T. L. Kelly, Nat. Photonics, 2013, 8, 133–138; (e) N.-G. Park, J. Phys. Chem. Lett., 2013, 4, 2423–2429.
2. (a) S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou, M. J. P. Alcocer, T. Leijtens, L. M. Herz, A. Petrozza and H. J. Snaith, Science, 2013, 342, 341–344; (b) H.-S. Kim, I. Mora-Sero, V. Gonzalez-Pedro, F. Fabregat-Santiago, E. J. Juarez-Perez, N.-G. Park and J. Bisquert, Nat. Commun., 2013, 4, 2242; (c) A. Marchioro, J. Teuscher, D. Friedrich, M. Kunst, R. van de Krol, T. Moehl, M. Gratzel and J. E. Moser, Nat. Photonics, 2014, 8, 250–255; (d) Y. Zhao and K. Zhu, J. Phys. Chem. Lett., 2013, 4, 2880–2884; (e) Y. Zhao, A. M. Nardes and K. Zhu, J. Phys. Chem. Lett., 2014, 5, 490–494.
3. (a) P. P. Boix, K. Nonomura, N. Mathews and S. G. Mhaisalkar, Mater. Today, 2014, 17, 16–23; (b) M. A. Green, A. H. Baillie and H. J. Snaith, Nat. Photonics, 2014, 8, 506–514; (c) T. C. Sum and N. Mathews, Energy Environ. Sci., 2014, 7, 2518–2534.
4. (a) H. Zhou, Q. Chen, G. Li, S. Luo, T.-B. Song, H.-S. Duan, Z. Hong, J. You, Y. Liu and Y. Yang, Science, 2014, 345, 542–546; (b) J.-H. Im, I.-H. Jang, N. Pellet, M. Grätzel and N.-G. Park, Nat. Nanotechnol., 2014, 9, 927–932.
5. J. T.-W. Wang, J. M. Ball, E. M. Barea, A. Abate, J. A. Alexander-Webber, J. Huang, M. Saliba, I. Mora-Sero, J. Bisquert and H. J. Snaith, Nano Lett., 2014, 14, 724–773.
6. (a) J. M. Ball, M. M. Lee, A. Hey and H. J. Snaith, Energy Environ. Sci., 2013, 6, 1739–1743; (b) M. J. Carnie, C. Charbonneau, M. L. Davies, J. Troughton, T. M. Watson, K. Wojciechowski, H. Snaith and D. A. Worsley, Chem. Commun., 2013, 49, 7893–7895; (c) A. Yella, L.-P. Heiniger, P. Gao, M. K. Nazeeruddin and M. Grätzel, Nano Lett., 2014, 14, 2591–2596.
7. (a) S. Sun, T. Salim, N. Mathews, M. Duchamp, C. Boothroyd, G. Xing, T. C. Sum and Y. M. Lam, Energy Environ. Sci., 2014, 7, 399–407; (b) O. Malinkiewicz, A. Yella, Y. H. Lee, G. M. Espallargas, M. Grätzel, M. K. Nazeeruddin and H. J. Bolink, Nat. Photonics, 2014, 8, 128–132; (c) Q. Wang, Q. Dong, Z. Xiao, Y. Yuan and J. Huang, Energy Environ. Sci., 2014, 7, 2359–2365; (d) Z. Xiao, C. Bi, Y. Shao, Q. Dong, Q. Wang, Y. Yuan, C. Wang, Y. Gao and J. Huang, Energy Environ. Sci., 2014, 7, 2619–2623; (e) J. Seo, S. Park, Y. C. Kim, N. J. Jeon, J. H. Noh, S. C. Yoon and S. I. Seok, Energy Environ. Sci., 2014, 7, 2642–2646.
8. (a) C. Roldán-Carmona, O. Malinkiewicz, A. Soriano, G. M. Espallargas, A. Garcia, P. Reinecke, T. Kroyer, M. I. Dar, M. K. Nazeeruddine and H. J. Bolink, Energy Environ. Sci., 2014, 7, 994–997; (b) J. You, Z. Hong, Y. Yang, Q. Chen, M. Cai, T.-B. Song, C.-C. Chen, S. Lu, Y. Liu, H. Zhou and Y. Yang, ACS Nano, 2014, 8, 1674–1680; (c) P. Docampo, J. M. Ball, M. Darwich, G. E. Eperon and H. J. Snaith, Nat. Commun., 2014, 4, 2761.
9. (a) G. E. Eperon, V. M. Burlakov, P. Docampo, A. Goriely and H. J. Snaith, Adv. Funct. Mater., 2014, 24, 151–157; (b) A. Dualeh, N. Tétreault, T. Moehl, P. Gao, M. K. Nazeeruddin and M. Grätzel, Adv. Funct. Mater., 2014, 24, 3250–3258.
10. N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. Ryu and S. I. Seok, Nat. Mater., 2014, 13, 897–903.
11. F. A. Carey and R. J. Sundberg, Advanced Organic Chemistry: Part A: Structure and Mechanisms, Springer, 5th edn, 2007.
12. (a) P. W. Liang, C. Y. Liao, C. C. Chueh, F. Zuo, S. T. Williams, X. K. Xin, J. Lin and A. K. Jen, Adv. Mater., 2014, 26, 3748–3754; (b) P. W. Liang, C. C. Chueh, X. K. Xin, F. Zuo, S. T. Williams, C. Y. Liao and A. K.-Y. Jen, Adv. Energy Mater., 2014 DOI:10.1002/aenm.201400960; (c) F. Zuo, S. T. Williams, P. W. Liang, C. C. Chueh, C. Y. Liao and A. K.-Y. Jen, Adv. Mater., 2014, 26, 6454–6460.
13. (a) E. Edri, S. Kirmayer, A. Henning, S. Mukhopadhyay, K. Gartsman, Y. Rosenwaks, G. Hodes and D. Cahen, Nano Lett., 2014, 14, 1000–1004; (b) C. Wehrenfennig, M. Liu, H. J. Snaith, M. B. Johnstona and L. M. Herz, Energy Environ. Sci., 2014, 7, 2269–2275; (c) Z. Xiao, Q. Dong, C. Bi, Y. Shao, Y. Yuan and J. Huang, Adv. Mater., 2014, 26, 6503–6509.
14. (a) M. Grätzel, Nat. Mater., 2014, 13, 838–842; (b) M. D. McGehee, Nat. Mater., 2014, 13, 845–846; (c) M. A. Green, A. Ho-Baillie and H. J. Snaith, Nat. Photonics, 2014, 8, 506–514.
15. Q. Chen, H. Zhou, T. Song, S. Luo, Z. Hong, H. Duan, L. Dou, Y. Liu and Y. Yang, Nano Lett., 2014, 14, 4158–4163.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, XRD of intermediate phase, UV-vis absorption spectra of perovskite films with different solvent washing and different annealing temperature, SEM images of perovskite films with different annealing temperature, J–V curves with different scan direction, time-resolved photoluminescence results. See DOI: 10.1039/c4ra13212b

---