PbCl₂-assisted film formation for high-efficiency heterojunction perovskite solar cells

Abstract

In this work, PbCl₂ is used as an additive to assist organolead trihalide perovskite film formation in a two-step sequential deposition process. PbCl₂ inhibits PbI₂ crystallization and contributes to the full conversion of PbI₂ and enhanced perovskite film morphology control. Cl⁻ incorporation into the perovskite improves charge transport within the film, as confirmed by the resulting prolonged photoluminescence lifetime observed. A reaction temperature of approximately 50 °C between the PbI₂/PbCl₂ film and CH₃NH₃I isopropanol solution is essential for synthesizing high-performance perovskite solar cells. Addition of PbCl₂ results in a perovskite solar cell energy efficiency of 14% and achieves an average efficiency enhancement of approximately 30% compared with that obtained from the control group.

---

Introduction

Organic–inorganic hybrid perovskite solar cells are an emerging photovoltaic technology. These devices have gained much attention since the first solid perovskite solar cells were first reported in 2012 to have power conversion efficiencies (PCEs) of 10%.^1,2 Perovskite solar cells have exhibited rapid development within the last three years, and a remarkable high efficiency of over 19% has been achieved.³ The early perovskite solar cells were mostly solid mesoporous solar cells in which perovskite materials, such as CH₃NH₃PbI₃ and CH₃NH₃PbI_3−xCl_x, were often deposited on mesoporous scaffolds, such as TiO₂ and Al₂O₃. Subsequent studies showed that organolead trihalide perovskite can function as exceptional panchromatical-light absorbing materials,^4–7 efficiently facilitate charge separation, and react as both electron and hole transporters. These characteristics are due to the ambipolar semiconducting nature of the perovskite, which enables fabrication of efficient planar heterojunction devices. Compared with mesoporous structured devices, planar heterojunction devices simplify the fabrication processes and are more attractive for developing flexible and tandem solar cells.⁷ Several methods have been developed to achieve high-performance planar heterojunction perovskite solar cells. The earliest report of such a method adopted a dual-source evaporation technique to prepare a high-quality perovskite layer by co-evaporation of PbCl₂ and CH₃NH₃I and obtained a PCE of as high as 15.4%.⁸ However, the requirements of high vacuum conditions and sophisticated equipment are unfavorable for the large-scale fabrication of these devices.

Two-step sequential deposition of perovskite has also been used to fabricate planar heterojunction devices. This technique features improved control of the perovskite film morphology. Two-step sequential deposition of (RNH₃)₂(CH₃NH₃)_n−1M_nI_3n+1 (R = butyl, phenethyl; M = Pb, Sn) perovskite film was first reported by Mitzi et al.⁹ and successfully developed by Grätzel et al.¹⁰ for CH₃NH₃PbI₃ synthesis. This deposition technique was first applied to mesoporous structured devices with a scaffold layer in the early reach stages, such as TiO₂ or Al₂O₃. During two-step deposition, a high-concentration PbI₂ solution is first spin-coated onto the substrate and then reacted with CH₃NH₃I. Two-step sequential deposition can be categorized into two according to the treatment in the second step. Thus, deposition can either be a vapor-assisted solution process (VASP) or a solution-based process that includes dipping¹⁰ and interdiffusion¹¹ methods.

VASP was first developed by Yang et al.¹² A PbI₂ thin film prepared by spin-coating was reacted with CH₃NH₃I vapor at 150 °C to convert the thin film into perovskite, leading to a PCE of 12.1%. The resulting PCE was further enhanced to 16.8% (ref. 13) by lowering the reaction temperature and CH₃NH₃I vapor pressure. In this work the beneficial effect of using mixed PbI₂/PbCl₂ precursor in two step process has also been demonstrated.¹³ Except that the use of mixed PbI₂/PbBr₂ precursor solution for a sequential deposition process has also been demonstrated.¹⁴ Compared with VASP, the second type of deposition solution process is easier to operate and considerably more accessible to diverse perovskite devices. However, the good crystallinity of PbI₂ produces undesirable effects. The PbI₂ crystal size is confined by the space between nanoparticles in the mesoporous structure but preferably forms large crystals in the planar heterojunction structure. Considering that the second reaction with CH₃NH₃I is a heterogeneous reaction, large PbI₂ crystals will feature reduced accessibility to small organic molecules, thereby resulting in incomplete PbI₂ conversion. This phenomenon can severely impact device efficiency and reproducibility. Additionally, variations in PbI₂ crystal size lead to changes in the perovskite crystal size, which probably contributes to uncontrollable film morphologies. When N,N-dimethylformamide (DMF) solvent is replaced with dimethyl sulfoxide (DMSO), the molecules of which have stronger coordination with Pb²⁺ than DMF, PbI₂ crystallization can be effectively inhibited, leading to full conversion of PbI₂.¹⁵ Consequently, high-efficiency CH₃NH₃PbI₃ perovskite solar cells with enhanced reproducibility are obtained.

To the best of our knowledge, the use of Cl⁻ as a dopant agent can result in formation of CH₃NH₃PbI_3−xCl_x and greatly improve the transport properties of perovskite, especially its hole/electron diffusion length. Cl⁻ can also induce lattice distortion during perovskite crystallization, which slows down the crystal growth rate.¹⁶ This phenomenon is mainly inferred from the longer annealing time and higher annealing temperature required for full conversion into perovskite compared with pure PbI₂-based materials. In this work, we presented PbCl₂ as an additive and adopted interdiffusion of spun stacking layers of PbCl₂/PbI₂ and CH₃NH₃I to fabricate perovskite solar cells.¹¹ The PbCl₂ additive effectively inhibited crystallization of PbI₂ to complete its conversion of PbI₂ and controlled the resulting CH₃NH₃PbI_3−xCl_x film morphology. Introduction of PbCl₂ dramatically prolonged the fluorescence lifetime of perovskite, indicating improved charge transport properties within the perovskite layer. The system with PbCl₂ additive was more sensitive to the spin-coating temperature of the CH₃NH₃I/isopropanol solution than that without the additive. Higher temperatures benefited formation of larger perovskite crystal sizes, which remarkably enhanced fill factor (FF) and J_sc. Adjusting the amount of PbCl₂ and further optimizing the spin-coating temperature resulted in enhancements in the average PCE of the corresponding perovskite solar cells by over 30% compared with that of pure PbI₂-based devices.

Results and discussion

The preparation of perovskite film on glass substrate is illustrated in Fig. 1a. Various molar ratios of PbI₂ and PbCl₂ were dissolved in DMF, spin-coated onto cleaned FTO at 70 °C, and then annealed at 100 °C for 60 min to exclude the DMF molecules. Then, 40 mg mL⁻¹ CH₃NH₃I in isopropanol solution was hot-casted onto the as-prepared PbI₂/PbCl₂ thin film, followed by solvent annealing treatment for 60 min at 100 °C. Eventually, spiro-MeOTAD was spin-coated onto the top of the substrate as a hole-transporting material. The Au counter electrode was formed via magnetron sputtering. The device structure adopted in the current study was FTO/cTiO₂/CH₃NH₃PbI_3−xCl_x/spiro-MeOTAD/Au. A cross-sectional view of the device is shown in Fig. 1b.

Fig. 1 (a) Schematic of the preparation process of perovskite film (cTiO₂ referred to the TiO₂ compact layer). (b) Scanning electron micrograph of the cross-sectional view of the perovskite solar cell.

We found that the temperature of the FTO substrate and PbI₂–PbCl₂/DMF solution influenced the final morphology of the perovskite film during spin-coating of the PbI₂–PbCl₂/DMF solution. Fig. S1† shows a scanning electron micrograph of the film spin-coated from PbI₂–PbCl₂/DMF solution with 33% molar ratio of PbCl₂ at room temperature and then annealed at 100 °C for 60 min. The morphology of this film remarkably differed from that of the film spin-coated at higher temperatures (70 °C; Fig. S1b†). Lower temperatures induced the formation of smaller crystal sizes and higher porosity, which are not beneficial to the morphological control of the perovskite film. Warmer substrates and precursors may contribute to PbI₂ crystallization, consistent with other reports that adopted a temperature of 70 °C to spin-coat PbI₂/DMF solution during the two-step sequential deposition process.^8,17 The morphologies of PbI₂/PbCl₂ films with varied molar ratios of PbCl₂, which are shown in Fig. S2a–c,† are clearly distinguishable from those of pure PbI₂ film, which is composed of large flaky crystals, as illustrated in Fig. S2d.† Addition of PbCl₂ apparently changed the film morphology by diminishing the PbI₂/PbCl₂ crystal size and smoothing the film surface to decrease roughness. When the molar ratio was further increased to 50%, film coverage decreased and porosity evidently increased. Well-regulated PbI₂/PbCl₂ films generally favor the formation of high-quality perovskite film.^8,17 Therefore, addition of an appropriate molar ratio of PbCl₂ is necessary for morphological control of subsequent perovskite film.

When the amount of PbCl₂ added was increased from 0% to 50%, the color of the prepared PbI₂/PbCl₂ film on prepared substrate changed from yellow to white (Fig. S3†). This change was further demonstrated by the UV-vis absorption spectra of PbI₂/PbCl₂ films on substrate with different molar ratios of PbCl₂. The pure PbI₂-based film revealed an absorption peak centered at 500 nm (Fig. 2a), which is attributed to the band-gap excitation of crystallized PbI₂. The absorption intensity of the characteristic peak weakened with increasing amount of added PbCl₂. The absorption peak at 500 nm disappeared at 50% PbCl₂ molar ratio. This phenomenon basically proves the amorphous nature of PbI₂. XRD patterns of the PbI₂/PbCl₂ films (Fig. 2b) also verified the inhibition effect of PbCl₂ on PbI₂ crystallization. PbI₂ films exhibited a strong diffraction peak at 2θ = 12.6°, which corresponds to the (001) crystal plane of crystallized PbI₂. The intensity of this characteristic peak significantly decreased when PbCl₂ was introduced to the system and completely disappeared in the XRD pattern of PbI₂/PbCl₂ film when the PbCl₂ molar ratio was increased to 50%. The absence of this characteristic diffraction peak indicates that the film is in an amorphous state. The inhibitory effect of PbCl₂ may be ascribed to the formation of a new phase (PbClI) which can be identified by XRD pattern of the film. By comparing the XRD of FTO/cTiO₂/PbI₂:PbCl₂ and FTO/cTiO₂ (Fig. 2c), we can found that peaks 2θ = 26.4°, 37.8°, 51.4° belongs to the substrate of FTO/cTiO₂. And according to previous report^13,18 relatively weak peaks at 21.34°, 28.9°, 30.7°, and 34.1°, corresponding to the (120), (121), (211), and (310) crystal planes of PbClI(Fig. 2c). The new double peaks at 2θ = 23.6° and 24.0° also appeared at the XRD pattern of PbI₂/PbCl₂ based on other PbCl₂ of other molar ratios which we believe might be assign to the intermediate phase of PbCl₂/PbI₂ formed during preparation process. Taken together, the UV-vis absorption spectra and XRD patterns confirm the inhibitory effects of PbCl₂ on PbI₂ crystallization; such inhibition may accelerate the reaction between PbI₂/PbCl₂ film and organic CH₃NH₃I molecules.¹⁶

Fig. 2 UV-vis absorption spectra (a) and XRD patterns (b) of PbI₂/PbCl₂ films based on different PbCl₂ molar ratios (adopted structure, FTO/cTiO₂/PbI₂–PbCl₂). (c) XRD patterns of 50% PbCl₂ molar ratio PbI₂/PbCl₂ film and the control sample. (d) XRD patterns of perovskite films obtained from 33% PbCl₂ molar ratio PbI₂/PbCl₂ films obtained at different annealing times.

We compared the evolutional processes of the XRD patterns of perovskite films based on PbI₂/PbCl₂ (33% molar ratio of PbCl₂) (Fig. 2d) and pure PbI₂-based (Fig. S4†) films at different annealing times. The perovskite peak intensity was fairly weak immediately after spin-coating, especially in the pure PbI₂-based sample. This finding suggests that both pure PbI₂ and PbI₂/PbCl₂ films without annealing cannot completely react; such a phenomenon is supported by the scanning electron micrograph showing unreacted PbI₂/PbCl₂ phase in Fig. S5.† After annealing at 100 °C for 15 min, the pure PbI₂ sample still showed diffraction peaks at 2θ = 12.6°, indicating residual unreacted PbI₂. The peak intensity of PbI₂ continued to decrease until an annealing time of 30 min. The small PbI₂ peak was maintained even after 60 min, thereby implying incomplete conversion of PbI₂. In the PbI₂/PbCl₂ samples, the small PbI₂ peak disappeared immediately after spin-coating, and the intensity of the perovskite film remained nearly unchanged after 15 min; no characteristic peak of PbI₂ was observed at 2θ = 12.6°. The peak of the PbClI phase completely disappeared, likely because of reaction between PbClI phase and MAI that convert into perovskite. These findings indicate that addition of PbCl₂ facilitates the reaction between PbI₂ and CH₃NH₃I.

PbCl₂ inhibited PbI₂ crystallization and assisted in the completion of the PbI₂ reaction. Hence, we supposed that addition of PbCl₂ may play an important role in perovskite formation and device performance. Thus, we examined the property of perovskite films based on PbI₂/PbCl₂ films with varying amounts of PbCl₂. We found that all of the PbI₂/PbCl₂ films formed perovskite after solvent annealing treatment. Fig. 3a–d show the scanning electron micrographs of perovskite based on PbI₂/PbCl₂ films of different molar ratios. The resulting perovskite demonstrated larger crystal sizes and enhanced smoothness compared with the thermally annealed sample as a result of the solvent annealing process (Fig. S6†). The morphology of the perovskite films showed no distinct difference, probably because of solvent annealing. When the added amount of PbCl₂ was excessively high or no PbCl₂ was added, the grain size and smoothness decreased to a certain extent, likely because of morphological differences in the PbI₂/PbCl₂ films discussed above.^1,2,13 The XRD patterns of the perovskite films (Fig. 4e) formed from PbI₂/PbCl₂ films of different PbCl₂ molar ratios showed peaks at 2θ = 14.08°, 28.5°, 31.8°, and 43.2°, respectively corresponding to the (110), (220), (310), and (330) crystal planes of perovskite. No apparent differences were found between the peak positions of CH₃NH₃PbI_3−xCl_x and CH₃NH₃PbI₃, in agreement with other reports.^13,19

Fig. 3 Scanning electron micrographs of perovskite films based on PbI₂/PbCl₂ films with PbCl₂ molar ratios of (a) 50%, (b) 33%, (c) 25%, and (d) 0%. (e) XRD patterns of perovskite films based on PbI₂/PbCl₂ films with different PbCl₂ molar ratios. (f) Time-resolved photoluminescence decay plot of perovskite films based on PbI₂/PbCl₂ films with different PbCl₂ molar ratios. The excitation wavelength was 440 nm and the probe wavelength was 780 nm.

Fig. 4 Photovoltaic parameters of solar cells with different PbCl₂ molar ratios. Dark cyan, red, brown, and blue boxes represent the photovoltaic J_sc, PCE, V_oc, and FF, respectively. Each box presents the parameter distribution of 40 devices under similar working conditions. The parameter values are presented as box plots for distributions of the different PbCl₂ molar ratios. Box edges represent the 25/75 percentile. The small square symbol (□) inside the boxes represents the mean, whereas the line across the boxes represents the median. The furcation symbols (×) represent the maximum and minimum values.

EDS (Energy Dispersive X-ray Spectrometer) results show that CH₃NH₃PbI_3−xCl_x also exhibits no detection signal of Cl because of content of Cl below the LOD (limit of detection which is about 1% atomic percentage) of EDS (Fig. S7†).^17,20,21 According the previous report that the measurement result delivered by XPS (X-ray photoelectron spectroscopy) of Cl atomic percentage in CH₃NH₃PbI_3−xCl_x-based perovskite is about 0.7%.^17,21 Compared with the XRD patterns of perovskite films based on PbI₂/PbCl₂ films with different molar ratios (Fig. 4e), the pure PbI₂-based perovskite film showed the residual characteristic peak of PbI₂ at 2θ = 12.6°, indicating incomplete conversion of PbI₂. When PbCl₂ was introduced to the system, new diffraction peak at 2θ = 15.5° appeared. According to the previous work,^19,22,23 this diffraction peak corresponds to the CH₃NH₃PbCl₃ phase with clear identification, and this which also appears in other CH₃NH₃PbI_3−xCl_x-based perovskite XRD patterns.^5,13,17,23 When the amount of added PbCl₂ was increased, the characteristic peak intensity of CH₃NH₃PbCl₃ was enhanced. At 50% PbCl₂, an obvious peak at 2θ = 15.5° appeared in the XRD pattern (Fig. 4e). The scanning electron micrographs and XRD patterns prove that PbI₂/PbCl₂ films with different PbCl₂ molar ratios could form perovskite films. Excessively high or too low PbCl₂ molar ratios resulted in decreased grain sizes and smoothness. Addition of PbCl₂ also led to the formation of the CH₃NH₃PbCl₃ phase.

We investigated the photovoltaic parameters of solar cells based on different PbCl₂ molar ratios, and the results are summarized in Table 1 and Fig. 4; statistical data derived from 40 devices are listed under each condition. Addition of PbCl₂ showed no apparent influence on the open-circuit voltage, which was approximately 1 V. When the amount of PbCl₂ added was at a certain range (PbCl₂ molar ratio from 0% to 33%), J_sc was not evidently affected. J_sc declined distinctly only when the molar ratio of PbCl₂ reached 50%. This phenomenon may have been caused by incomplete conversion of PbCl₂. The reaction between PbCl₂ and CH₃NH₃I usually requires longer reaction time and higher reaction temperatures compared with that of PbI₂.^2,24 However, the reaction time and temperature were limited during the spin-coating process. The amount of CH₃NH₃I left over on top of the PbI₂/PbCl₂ film after spin-coating was also limited. These characteristics led to incomplete conversion of PbCl₂. The observed phenomena could also be supported by the fact that the optimized reaction temperature for PbI₂/PbCl₂-based devices is higher than that of pure PbI₂-based ones. Addition of PbCl₂ evidently improved FF and contributed to the enhancement of PCE. The cells presented the highest average PCE at a PbCl₂ molar ratio of 33%. The devices demonstrated the following characteristics: average V_oc, 0.98 ± 0.02 V; J_sc, 20.70 ± 1.00 mA cm⁻²; FF, 0.59 ± 0.03; and PCE, 12.1% ± 0.98%. The average PCE of the devices obtained was enhanced by 30%, from 9.3% to 12.1%, compared with that of the control group.

Table 1 Device performance of perovskite solar cells^a

PbCl2 molar ratio (%)	Voc (V)	Jsc (mA cm^−2)	FF	PCE (%)
a Under each condition, the listed result is the standard deviation calculated from 40 devices.
0	0.96 ± 0.03	20.1 ± 1.33	0.48 ± 0.03	9.3 ± 0.78
25	1.01 ± 0.02	20.3 ± 0.84	0.57 ± 0.03	11.7 ± 0.90
33	0.98 ± 0.02	20.7 ± 1.00	0.59 ± 0.03	12.1 ± 0.98
50	1.00 ± 0.03	18.1 ± 1.21	0.53 ± 0.03	9.7 ± 0.92

---

We performed time-resolved photoluminescence (PL) measurements on the different perovskite films to explore the mechanism by which PbCl₂ affects device performance. Time-resolved PL measurements provide important information on the lifetime of the excited species in the photoactive layer, which can be used to deduce the diffusion length of the charge carriers. All of the test samples adopted the structure of glass/CH₃NH₃PbI₃ or glass/CH₃NH₃PbI_3−xCl_x. The decay plots obtained are shown in Fig. 3f. All the testing results fit the biexponential decay dynamics, and the fitted parameters are presented in Table S2.† Biexponential decay maybe attributed to variations in the perovskite grain size.¹⁷ Slower decay time represent longer diffusion lengths, indicating higher charge collection efficiencies. Addition of PbCl₂ effectively prolonged the PL lifetime τ_slow. At a PbCl₂ molar ratio of 33%, the lifetime τ_slow reached 172.63 ns, which is far longer than the 75.17 ns of τ_slow of the pure PbI₂-based sample. A PbCl₂ molar ratio higher than 33% decreased the lifetime. This variation in the PL lifetime basically agrees with the variation in device performance. Lifetime extension can be explained by Cl⁻ doping, which has been reported to remarkably improve transport property.^16,25 Thus, addition of PbCl₂ can enhance device performance by improving perovskite film transport properties.

The spin-coating temperature of the CH₃NH₃I/isopropanol solution plays an important role in device efficiency during device fabrication, especially when PbI₂/PbCl₂-based devices are produced. This sensitivity to temperature may be derived from the fact that PbCl₂ requires higher reaction temperature and longer reaction time to convert into perovskite.^2,24 Excess Cl⁻ may sublime in the form of CH₃NH₃Cl and leave only trace amounts of chloride in the perovskite. The sublimation process would therefore influence the phase transformation process, wherein lattice distortion caused by Cl⁻ doping and exclusion of organic molecules could improve perovskite crystallization^5,12,26,27. Fig. 5e shows the XRD patterns of CH₃NH₃PbI_3−xCl_x prepared at different temperatures. When the spin-coating temperature was enhanced from 30 °C to 60 °C, the peak intensity of the formed perovskite was reinforced and the half-peak width of the characteristic peak at 2θ = 14.08° became smaller, indicating improved perovskite crystallinity. The characteristics of the perovskite films prepared at different temperatures shown in Fig. 5a–d indicate that larger crystalline grains are preferred at higher temperatures, further supporting the enhancement of perovskite crystallization. Compared with the XRD patterns of perovskite films at spin-coating temperatures of 30 °C (Fig. S8†) and 50 °C (Fig. 2d), the intensity of the characteristic perovskite peak at 2θ = 14.08° obtained immediately after 50 °C hot casting without annealing was much stronger than that observed at 30 °C. This phenomenon suggests that larger amounts of PbI₂/PbCl₂ convert into the perovskite phase before annealing treatment. Higher conversion ratios of PbI₂/PbCl₂ may benefit the formation of larger perovskite grain sizes. When the hot spin-coating temperature of the CH₃NH₃I/isopropanol solution reached 60 °C, severe damage of the perovskite film surface is observed (Fig. S9†). This development is mainly due to the restriction of the boiling temperature (82 °C) of isopropanol. High spin-coating temperatures accelerate the evaporation of isopropanol and result in rapid separation of CH₃NH₃I to form visible crystals. Excessive CH₃NH₃I residues may also affect device performance. Thus, reaction temperatures of no more than 60 °C may ensures device performance.

Fig. 5 Scanning electron micrographs of perovskite films prepared from 33% PbCl₂ molar ratio PbI₂/PbCl₂ films based on MAI spin-coating temperatures of (a) 30 °C, (b) 40 °C, (c) 50 °C, and (d) 60 °C. (e) XRD patterns of perovskite films based on 33% PbCl₂ molar ratio PbI₂/PbCl₂ films with different MAI spin-coating temperatures.

The photovoltaic parameters of 40 devices for each spin-coating temperature are shown in Fig. S10.† No apparent difference in V_oc was observed under the different spin-coating temperatures, and values of approximately 1 V were consistently obtained. The different spinning temperatures mainly affected the J_sc and FF values of the devices. When the temperature was increased from 30 °C to 50 °C, the average J_sc increased from 16.68 mA cm⁻² to 20.65 mA cm⁻², and the average FF increased from 0.54 to 0.59. The average J_sc values of devices based on different spin-coating temperatures corresponded to the IPCE integral values at varied spin-coating temperatures (Fig. 6a). This phenomenon can be explained by the formation of larger grain sizes of perovskite at higher temperatures, as confirmed by the XRD patterns and scanning electron micrographs described earlier. Larger grain sizes of perovskite decrease the impact of grain boundaries on charge separation and transmission. Grain boundaries, especially those on the longitudinal transport pathway, induce charge recombination, which severely affected the device performance.¹¹ Fewer grain boundaries provide effective charge transport, inhibit interface charge recombination, and improve J_sc and FF. When the temperature was excessively high, CH₃NH₃I precipitated before reacting with PbI₂/PbCl₂ during spin-coating, and the residual CH₃NH₃I negatively affected device performance. The optimized spin-coating temperature was 50 °C, at which point the average PCE was enhanced by 34% compared with that at 30 °C. The best device achieved a maximum PCE of 14% as well as V_oc, J_sc, and FF values of 0.99 V, 21.8 mA cm⁻², and 0.64, respectively (Fig. 6d).

Fig. 6 (a) IPCE of perovskite solar cells based on 33% PbCl₂ molar ratio PbI₂/PbCl₂ film spin-coated CH₃NH₃I solution at different temperatures. (b) J_sc of solar cells based on 33% PbCl₂ molar ratio (black) films and pure PbI₂ films (red) prepared from different CH₃NH₃I isopropanol solution spin-coating temperatures. (c) PCE of solar cells based on 33% PbCl₂ molar ratio film (black) and pure PbI₂ films (red) prepared from different CH₃NH₃I isopropanol solution spin-coating temperatures. Each box presents the parameter distribution of 40 devices under similar working conditions. The parameter values are presented as box plots of distributions for a range of different PbCl₂ molar ratios. Box edges represent the 25/75 percentile. The small square symbol (□) inside the boxes represents the mean, whereas the line across the boxes represents the median. The furcation symbols (×) represent the maximum and minimum values. (d) J–V curve of the device with highest PCE reverse scan. The device was spin-coated with 33% molar ratio PbCl₂ at 50 °C.

We further compared the device performances of pure PbI₂ with the PbI₂/PbCl₂ system obtained at different CH₃NH₃I isopropanol solution spin-coating temperatures (Fig. S11†). Similar to PbI₂/PbCl₂, the V_oc of the devices showed no apparent difference under various spin-coating temperatures. However, the J_sc of the PbI₂-based device was not sensitive to the temperature, and the average J_sc was approximately 20 mA cm⁻², as shown in Fig. 6b. The influence of temperature was mainly observed in terms of FF. The optimal spin-coating temperature of the pure-PbI₂ device was 40 °C, which is lower than that of the PbI₂/PbCl₂-based device (Fig. 6c), and the distribution of the PCE of the former was wider compared with that of the latter. Under optimized fabrication conditions, the average PCE based on PbI₂/PbCl₂ was enhanced by 21% compared with that of the control group.

Photocurrent hysteresis is one of the major issues hindering accurate measurement of device efficiency, especially for planar heterojunction devices, which often present obvious hysteresis.²⁸ We found that both PbI₂/PbCl₂ and PbI₂-based device show a certain degree of hysteresis. The reverse scanning efficiency was higher than the forward scanning efficiency, and the scanning rate impacted hysteresis. Fig. S12† shows that the device demonstrates the smallest hysteresis at a scanning rate of 0.1 V s⁻¹. Hysteresis was aggravated when the scanning rate was higher or lower than this value. This phenomenon may be related to the device structure and fabrication process.²⁸ Independent of the PCE of the device, all of the J–V curves presented crosses, which are caused by overestimation of J_sc during forward scanning or underestimation of J_sc during reverse scanning (Fig. S13†). This characteristic may be relevant to the grain size of the perovskite according to a previous report.²⁹ The variation in PCE at 0.1 V s⁻¹ scanning rate was generally within 2%. Thus, this condition was employed in typical characterizations. In order to better evaluate our device performance, we took the steady-state measurements and the results is given in the Fig. S14.† No apparent difference PCE acquired from the steady-state measurement and from J–V plot was found. During the steady-state measurement we found the current fast raised when the device was first exposure to the light and declined about 3.2% during the 180 s measurement process. The decline of the photocurrent might because of the inferior stability under the ambient measurement condition.

In summary, PbCl₂ is an effective additive for organolead trihalide perovskite thin films. PbCl₂ inhibited crystallization of PbI₂, resulting in complete conversion of PbI₂ and improvement of the perovskite morphology. Addition of PbCl₂ resulted in the formation of CH₃NH₃PbI_3−xCl_x and remarkably prolonged the photoluminescence lifetime, which implies enhancements in transport property. A suitable spin-coating temperature enabled enlargement of the perovskite grain size, thereby enhancing both FF and J_sc, which are crucial for high-efficiency devices. Thus, addition of PbCl₂ to the PbI₂ DMF precursor solution is a simple and potent method to achieve highly efficient heterojunction perovskite solar cells.

Acknowledgements

This study is jointly supported by Ministry of Science and Technology of China (No. 2011CB933300), National Natural Science Foundation of China (No. 91333107, No. 51573004) and the fund from Shenzhen City (No. CXZZ20120618162051603).

References

1. H. Kim, C. Lee, J. Im, K. Lee, T. Moehl, A. Marchioro, S. Moon, R. Humphry-Baker, J. Yum, J. E. Moser, M. Grätzel and N. Park, Sci. Rep., 2012, 2, 591.
2. M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami and H. J. Snaith, Science, 2012, 338, 643–647.
3. H. Zhou, Q. Chen, G. Li, S. Luo, T. Song, H. Duan, Z. Hong, J. You, Y. Liu and Y. Yang, Science, 2014, 345, 542–546.
4. L. Etgar, P. Gao, Z. Xue, Q. Peng, A. K. Chandiran, B. Liu, M. K. Nazeeruddin and M. Grätzel, J. Am. Chem. Soc., 2012, 134, 17396–17399.
5. J. M. Ball, M. M. Lee, A. Hey and H. J. Snaith, Energy Environ. Sci., 2013, 6, 1739.
6. W. A. Laban and L. Etgar, Energy Environ. Sci., 2013, 6, 3249–3253.
7. M. Peng and D. Zou, J. Mater. Chem. A, 2015, 3, 20435–20458.
8. M. Liu, M. B. Johnston and H. J. Snaith, Nature, 2013, 501, 395–398.
9. K. Liang, D. B. Mitzi and M. T. Prikas, Chem. Mater., 1998, 10, 403–411.
10. J. Burschka, N. Pellet, S. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin and M. Grätzel, Nature, 2013, 499, 316–319.
11. Z. Xiao, Q. Dong, C. Bi, Y. Shao, Y. Yuan and J. Huang, Adv. Mater., 2014, 26, 6503–6509.
12. Q. Chen, H. Zhou, Z. Hong, S. Luo, H. Duan, H. Wang, Y. Liu, G. Li and Y. Yang, J. Am. Chem. Soc., 2013, 136, 622–625.
13. Y. Li, J. K. Cooper, R. Buonsanti, C. Giannini, Y. Liu, F. M. Toma and I. D. Sharp, J. Phys. Chem. Lett., 2015, 6, 493–499.
14. M. Ibrahim Dar, M. Abdi Jalebi, N. Arora, T. Moehl, M. Grätzel and M. K. Nazeeruddin, Adv. Mater., 2015, 27, 7221–7228.
15. Y. Wu, A. Islam, X. Yang, C. Qin, J. Liu, K. Zhang, W. Peng and L. Han, Energy Environ. Sci., 2014, 7, 2934–2938.
16. S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou, M. J. Alcocer, T. Leijtens, L. M. Herz, A. Petrozza and H. J. Snaith, Science, 2013, 342, 341–344.
17. P. Docampo, F. C. Hanusch, S. D. Stranks, M. Döblinger, J. M. Feckl, M. Ehrensperger, N. K. Minar, M. B. Johnston, H. J. Snaith and T. Bein, Adv. Energy Mater., 2014, 4, 1400355.
18. L. H. Brixner, H. Chen and C. M. Foris, J. Solid State Chem., 1981, 40, 336–343.
19. S. Colella, E. Mosconi, P. Fedeli, A. Listorti, F. Gazza, F. Orlandi, P. Ferro, T. Besagni, A. Rizzo, G. Calestani, G. Gigli, F. De Angelis and R. Mosca, Chem. Mater., 2013, 25, 4613–4618.
20. H. Yu, F. Wang, F. Xie, W. Li, J. Chen and N. Zhao, Adv. Funct. Mater., 2014, 24, 7102–7108.
21. M. I. Dar, N. Arora, P. Gao, S. Ahmad, M. Grätzel and M. K. Nazeeruddin, Nano Lett., 2014, 14, 6991–6996.
22. A. Poglitsch and D. Weber, J. Chem. Phys., 1987, 87, 6373.
23. W. Nie, H. Tsai, R. Asadpour, J. C. Blancon, A. J. Neukirch, G. Gupta, J. J. Crochet, M. Chhowalla, S. Tretiak, M. A. Alam, H. L. Wang and A. D. Mohite, Science, 2015, 347, 522–525.
24. J. Yan and B. R. Saunders, RSC Adv., 2014, 4, 43286–43314.
25. G. Xing, N. Mathews, S. Sun, S. S. Lim, Y. M. Lam, M. Grätzel, S. Mhaisalkar and T. C. Sum, Science, 2013, 342, 344–347.
26. G. E. Eperon, V. M. Burlakov, P. Docampo, A. Goriely and H. J. Snaith, Adv. Funct. Mater., 2014, 24, 151–157.
27. A. Abrusci, S. D. Stranks, P. Docampo, H. Yip, A. K. Y. Jen and H. J. Snaith, Nano Lett., 2013, 13, 3124–3128.
28. E. L. Unger, E. T. Hoke, C. D. Bailie, W. H. Nguyen, A. R. Bowring, T. Heumüller, M. G. Christoforo and M. D. McGehee, Energy Environ. Sci., 2014, 7, 3690–3698.
29. H. J. Snaith, A. Abate, J. M. Ball, G. E. Eperon, T. Leijtens, N. K. Noel, S. D. Stranks, J. T. Wang, K. Wojciechowski and W. Zhang, J. Phys. Chem. Lett., 2014, 5, 1511–1515.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra23062d

---