Single Source Thermal Evaporation of Two-dimensional Perovskite Thin Films for Photovoltaic Applications

Hybrid two-dimensional (2D) halide perovskites has been widely studied due to its potential application for high performance perovskite solar cells. Understanding the relationship between microstructural and opto-electronic properties is very important for fabricating high-performance 2D perovskite solar cell. In this work, the effect of solvent annealing on grain growth was investigated to enhance the efficiency of photovoltaic devices with 2D perovskite films based on (BA)2(MA)3Pb4I13 prepared by single-source thermal evaporation. Results show that solvent annealing with the introduction of solvent vapor can effectively enhance the crystallization of the (BA)2(MA)3Pb4I13 thin films and produce denser, larger-crystal grains. The thin films also display a favorable band gap of 1.896 eV, which benefits for increasing the charge-diffusion lengths. The solvent-annealed (BA)2(MA)3Pb4I13 thin-film solar cell prepared by single-source thermal evaporation shows an efficiency range of 2.54–4.67%. Thus, the proposed method can be used to prepare efficient large-area 2D perovskite solar cells.

properties of 2D MA 2 Pb (SCN) 2 I 2 perovskite, which can be used as an absorber layer for the top cell of a tandem solar cell 16 . Although superior device performance has not yet been achieved, this 2D layered mixtures have been demonstrated as effective new perovskite film with adjustable photoelectric properties and enhanced air stability 17,18 . In contrast to 3D perovskite, 2D perovskite [CH 3 (CH 2 ) 3 NH 3 ) 2 (CH 3 NH 3 ) n-1 Pb n I 3n+1 (BA) 2 (MA) n-1 Pb n I 3n+1 , n = 1, 2, 3, 4, …, ∞] have better optoelectronic property tunability because of their greater degree of freedom in quantum mechanics and chemistry, and, more importantly, higher environmental stability. Therefore, the development of 2D perovskite thin films will directly aid in improving the stability of perovskite solar cells.
Based on vacuum preparation method, the dual-source or single-source thermal evaporation methods, are also available to deposit perovskite thin films [19][20][21] . However, the dual-source thermal evaporation requires precise simultaneous control of the evaporation source of organic and inorganic materials, but the effective control of the film compounding process is very difficult. The easy deviation from the stoichiometric ratio directly leads to a decrease in film quality and repeatability 22 . To our knowledge, single-source thermal evaporation is a effective method for preparing large-area, high-efficiency perovskite solar cells 23 . In this study, 2D perovskite (BA) 2 (MA) 3 Pb 4 I 13 thin film was prepared by single-source thermal evaporation, and the effects of solvent annealing on the microstructural and optoelectronic properties of the thin film were investigated. Figure 1a shows the crystal structure of the (BA) 2 (MA) 3 Pb 4 I 13 powder and the thin films including the as-deposited and solvent-annealed thin films. The prepared (BA) 2 (MA) 3 Pb 4 I 13 powders have characteristic diffraction peaks of 2D perovskite. However, the as-deposited thin film shows broad peaks, which indicates low crystallinity. After the solvent annealing, the stronger characteristic diffraction peaks of the (060), (080), (111), (131), and (222) planes usually refer to the 2D (BA) 2 (MA) 3 Pb 4 I 13 perovskite crystal structure. These results indicate that the perovskite crystallinity is increased, with fewer low-dimensional defects and/or larger perovskite grain sizes, and less scattering of internal grain boundaries (Fig. 1b) [24][25][26][27] . Figure 1c displays the FWHM of the 2D perovskite (060), (080), and (111) peaks. The FWHM of the solvent-annealed thin film is significantly smaller, indicating better crystallization 28 . Based on the Debye-Scherrer formula, D = Kλ/(β cosθ), (D is the grain size of crystals, K is a constant, λis the wavelength of the X-ray, βis the FWHM, and θ is the diffraction angle 29 ), the grain sizes of the as-prepared and solvent-annealed thin films were calculated and the results are shown in Fig. 1d. After solvent annealing, the 2D perovskite grain size becomes markedly larger, suggesting that the solvent annealing can improve the crystallinity of the thin film, which might lead to a higher efficiency for device. www.nature.com/scientificreports www.nature.com/scientificreports/ The composition of the (BA) 2 (MA) 3 Pb 4 I 13 film is an important factor affecting the structural, electrical, and optical properties of the light-absorber. Figure 2 and Table 1 show the composition of the (BA) 2 (MA) 3 Pb 4 I 13 thin films measured by EDS. Two typical peaks located at 2.48 and 3.98 keV, corresponding to the Pb and I elements. The atomic ratio of Pb to I of the as-deposited thin film is approximately 0.392. It decreases to 0.365 for the solvent annealed films, which is much close to the stoichiometry of the (BA) 2 (MA) 3 Pb 4 I 13 film, indicating the formation of pure-phase 2D perovskite thin films. Figure 3 displays the elemental distribution of the thin film after solvent annealing, and shows that the Pb and I have uniform distribution in the entire plane without element enrichment or deficiency.     Figure 4a illustrates that the as-deposited thin film exhibits complete surface coverage but with small grains on the substrate. The cross-section in the inset of Fig. 4a shows no distinct grains which may easily lead to poor reproducibility and photocurrent hysteresis of the 2D perovskite solar cells [30][31][32][33] . Solvent vapor of γ-butyrolactone introduced during the annealing of the 2D perovskite causes the recrystallization of (BA) 2 (MA) 3 Pb 4 I 13 . Precise control of recrystallization can improve the quality of perovskite film [34][35][36] . After the γ-butyrolactone vapor annealing treatment (Fig. 4b), the (BA) 2 (MA) 3 Pb 4 I 13 thin film shows denser and larger grain distribution, and the defects are significantly reduced. Therefore, more photogenerated charges can successfully reach the electrode instead of recombining in the grain boundary.

Results
Time-resolved PL (TRPL) decay measurements were performed to study the influence of the charge transfer process in the (BA) 2 (MA) 3 Pb 4 I 13 thin film. Figure 5 displays the patterns and the lifetimes of the charge carriers in the thin films were estimated by fitting the data 37    The optical transmittance properties were obtained by a UV/visible/near-IR spectrophotometer in the wavelength range of 300-1000 nm. Figure 6 shows the transmittance spectra for the as-deposited and solvent-annealed (BA) 2 (MA) 3 Pb 4 I 13 thin films prepared by single-source thermal evaporation. As shown in the previous SEM image, more defect states due to smaller grains, the absorption edge is clearly moving toward the IR region after solvent annealing, indicating wide range of light absorption caused by enhanced crystallinity. The absorption range of the as-deposited thin film was lower than that of the annealed thin film due to the improvement of the film's crystallinity as we mentioned above. Compared with perovskites with multiple nano-grains, the solvent-annealed perovskite film has fewer grain boundaries, this facilitates a greater range of light absorption by the absorbing layer. The band-gap energy can be calculated as α ν where α is the absorption coefficient, hv is the photon energy, A is the constant, n depends on the nature of transition, and E is the band-gap energy 41 . Figure 7 shows that the band gap of the as-deposited thin film is 2.40 eV and decreases to 1.89 eV after solvent annealing, which is close to the theoretical value 42 . Hydrogen bond exists in perovskite, the presence of hydrogen bonds may affect the optical band gap of (BA) 2 (MA) 3 Pb 4 I 13 , Similarly, Filip et al. 43,44 have experimentally shown perovskite tunable optical bandgaps.
The perovskite solar cells with a device structure of ITO/PEDOT: PSS/2D perovskite (BA) 2 (MA) 3 Pb 4 I 13 / PC 61 BM/Ag (Fig. 8a) were fabricated. PEDOT: PSS and PCBM were the hole and electron transport layers, respectively. Figure 8b shows the J-V curves of the (BA) 2 (MA) 3 Pb 4 I 13 perovskite solar cells based on the as-deposited and solvent-annealed (BA) 2 (MA) 3 Pb 4 I 13 thin film. The J sc , V oc , FF, PCE, R s , and R sh of the corresponding devices are summarized in Table 2. In the 2D perovskite solar cell, the (BA) 2 (MA) 3 Pb 4 I 13 thin films without γ-butyrolactone vapor treatment are presented as the as-deposited PSCs, while the films treated with γ-butyrolactone are presented as the solvent-annealed PSCs. The as-deposited PSCs exhibit J sc of 6.51 mA/cm 2 ,  54%. Compared with the as-deposited PSCs, when the γ-butyrolactone solvent vapor is introduced during annealing, the performance of the 2D perovskite solar cell is significantly enhanced. J sc substantially increases to 10.98 mA/cm 2 , V oc increases to 0.95 V, R s is reduced to 26.55 Ω, and PCE increases to 4.67%. The γ-butyrolactone vapor during annealing can lead to enhanced crystallinity and larger grain size of (BA) 2 (MA) 3 Pb 4 I 13 , passivate defects, and improve device performance. Hence, the solvent annealing produces a high quality non-porous 2D perovskite film with a high purity phase, complete surface coverage, and good crystallinity. These characteristics can suppress internal recombination and leakage currents and promote photoelectric conversion of 2D perovskite solar cells.

conclusions
The effect of solvent annealing on grain growth is investigated to enhance the photovoltaic-device efficiency of 2D perovskite (BA) 2 (MA) 3 Pb 4 I 13 thin film prepared by single-source thermal evaporation. Solvent annealing can effectively enhance the crystallization of (BA) 2 (MA) 3 Pb 4 I 13 thin film with denser and larger crystal grains. The element ratio of Pb/I is close to the ideal stoichiometric ratio. The films show a favorable band gap of 1.896 eV and long electron and hole diffusion lengths of 314 nm and 266 nm, respectively. The performance of the (BA) 2 (MA) 3 Pb 4 I 13 perovskite solar cell is significantly enhanced, that is, J sc remarkably increases to 10.98 mA/ cm 2 , V oc increases to 0.95 V, and R s is reduced to 26.55 Ω. The solvent-annealed (BA) 2 (MA) 3 Pb 4 I 13 thin-film solar cell prepared by single-source thermal evaporation shows an efficiency of 4.67%. Thus, the proposed method is promising for preparing large-area and efficient 2D perovskite solar cells.

Methods
(BA) 2 (MA) 3 pb 4 i 13 crystal and powder preparation. PbI 2 (7.38 g, 99.99%, Xi'an Polymer Light Technology), MAI (1.91 g, 99.5%, Xi'an Polymer Light Technology), and BAI (1.61 g, 99.5%, Xi'an Polymer Light Technology) were mixed in γ-butyrolactone (150 ml, 99%, TCI) in the beaker for 24 h with constant magnetic stirring. The 2D perovskite solution was then transferred onto a glass culture dish and maintained at 150 °C on a hot plate. Until all 2D perovskite solutions were evaporated, (BA) 2 (MA) 3 Pb 4 I 13 crystals can be obtained as shown in Fig. 9. Then, the prepared (BA) 2 (MA) 3 Pb 4 I 13 crystals were ground into powders as the film evaporation material.
(BA) 2 (MA) 3 pb 4 i 13 thin-film preparation. Figure 9 shows the fabrication of the crystals, including the (BA) 2 (MA) 3 Pb 4 I 13 crystals preparation, single-source thermal evaporation, and solvent annealing. Prior to deposition, the ITO glass substrate was cleaned, and 0.8 g of (BA) 2 (MA) 3 Pb 4 I 13 perovskite powder was weighed. The powder was placed in the evaporation boat. The distance from the evaporation source to the substrate was 25 cm, and the substrate speed was 40 rpm. Once the chamber pressure was pumped down to below 1 × 10 −3 Pa, the working current of the evaporation source was rapidly raised to 150 A, and then the film was deposited. Until the powder was completely evaporated, the as-deposited 2D perovskite (BA) 2 (MA) 3 Pb 4 I 13 thin films has a thickness   Table 2. Photovoltaic performances and fitting parameters used for the impedance spectra of the (BA) 2 (MA) 3 Pb 4 I 13 -based perovskite solar cells before and after solvent annealing.