Figure 1(a-c) shows the XRD reflections of the grown films as a function of reaction time (at 220oC). Crystal reflection indices and secondary phases have been allocated to the corresponding films. All of the films' XRD data designate that the crystal structure is in good conformity with that of the Cs2SnI6 structure (JCPDS No. 51-0466, space group Fm3m). The primary intensity peaks corresponds to the atomic planes of (222), (004), and (044), respectively. The Cs2SnI6 film grown for 60 minutes also contains a tiny amount of CsI and SnI2 with secondary phases at 24.5 and 27.5o. At higher reaction temperatures, the CsI impurity peak is often generated by the decomposition of Cs2SnI6 perovskite. However, the amount of this impurity is quite small, implying that the presented approach is capable of producing highly crystalline Cs2SnI6 films.
The lattice strain factor (ε) and average crystallite size of the fabricated films were calculated using the Williamson-hall plot method, which is given as follows
Here, the ε is calculated by the slope of fitted lines while the average crystallite size is calculated from the y-intercept of the plot, as depicted in Fig. 1(d-f). The average crystallite size of the film decreased substantially with the increase in the reaction time. Furthermore, the calculated values of the crystallite size and lattice strain are finely consistent with the theoretical model and literature reports on Cs2SnX6 [19].
The optical characteristics of the Cs2SnI6 films were determined using the UV-Vis-NIR absorption spectrum. As shown in the Fig. 2(a), the absorption edge of Cs2SnI6 shifted to higher wavelengths as the reaction time decreased. The optical bandgap (Eg) of films was determined using the Tauc plots in Fig. 2(b), where (hv) is plotted along the X-axis and (αhν)2 is plotted along the Y-axis. To determine the bandgap value, the straight-line component of the plot was extended to span the X-axis. The samples had Eg values of 1.34, 1.27, and 1.22 eV for Cs2SnI6 films made at 80, 60, and 40 minutes, respectively. These values are consistent with the range of values previously reported for Cs2SnI6 perovskite materials (1.2-1.7 eV) [10-11]. The Photoluminescence (PL) spectra at room temperature stimulated by a 633 nm laser are depicted in Fig. 2(c). A strong PL peak extending from 800 to 900nm was seen, with the peak centered around 860-880 nm, indicating that the films' bandgaps match the values derived from the Tauc plot. With decreasing reaction time, we found a minor change in peak position toward the higher wavelength, which was also detected in absorption spectra.
The morphological examination of Cs2SnI6 films fabricated under various reaction time is shown in Fig. 3(a-c). The primary objective of this investigation was to determine the impact of reaction time on the growth and morphology of the crystallites. The morphology of the fabricated films is dominated by homogeneous and truncated octahedrons and tetrahedrons with uniform surface coverage and grain sizes ranging from a few micrometers to several micrometers. The films are highly dense, uniform, and free of pinholes, which is crucial for photovoltaic applications. At 80 minutes, the film reveals microscopic grain sizes and a significantly rougher morphology. By reducing the reaction time to 60 and 40 minutes, the grain size increased dramatically, implying that the films have a lower overall surface to volume ratio, and hence should exhibit increased resistance to the ambient atmosphere and stability. Additionally, small white pores can be detected in the corresponding SEM images, which may be generated by residual SnI2 vapor bound to the outermost surface of the Cs2SnI6 films in the closed reactor. Additional investigation of the grain size indicated that as the reaction time changes, the grain size fluctuate, as illustrated in Fig. 3(d-e). As seen in Fig. 3(f), the EDS spectra reveal the existence of Cs, Sn, and I elements with no significant impurity peaks.
We further investigated the role of reaction temperature on the properties of Cs2SnI6 films. To investigate the films' vibrational properties, Raman spectra were recorded using a 633 nm excitation wavelength, as illustrated in Fig. 4. The Raman spectrum of Cs2SnI6 films synthesized at various reaction temperatures provides more evidence for the perovskite modified structure. Six distinct symmetric stretching vibration modes (m1–m6) exist for [SnI6]2- octahedral in Cs2SnI6, among which m1, m2, and m5 are Raman-active. Three prominent peaks in the spectra at 126 cm-1, 92 cm-1, and 78 cm-1 correspond to Cs2SnI6 Raman-active modes. There are some secondary phase peaks seen in the 160oC samples, indicating that the low reaction temperature may not result in the formation of the pure phase Cs2SnI6. At 249 cm-1, an additional higher order mode was detected, which is ascribed to the strong resonance circumstances.
The absorbance spectrum of Cs2SnI6 thin film was examined in the wavelength range of 500-1000 nm in order to regulate its optical characteristics (Fig. 5a).As shown in (Fig. 5b), the absorption edge maintained a blue shift as the reaction temperature decreased. From Fig. 5b, the absorption onsets revealed that the films had Eg values of 1.27, 1.28, and 1.25 eV for Cs2SnI6 films formed at 220oC, 190oC, and 160oC minutes, respectively. The optical band gap was also confirmed using the room temperature PL spectrum, as depicted in Fig. 5c. A prominent PL peak was observed spanning from 800 to 900 nm, with the peak positioned around 860-880 nm, showing that the films' bandgaps complement the values estimated from the Tauc plot. The PL and absorption analyses show that the fabricated Cs2SnI6 has a direct band gap, which is in agreement with findings in the literature [10].
To characterize the photovoltaic device, we constructed it using the Cs2SnI6 film (60 minutes at 220oC) in the planar device design represented in Fig. 6a. Fig. 6(b) shows the energy band diagram and a characteristic cross-sectional view of the perovskite layers. Unfortunately, the device exhibited low PCE and a low current voltage (Voc) (Fig. 6c). The basic explanation for the low PCE is believed to be a combination of factors. To begin, Cs2SnI6 phase stability and crystallinity should be enhanced further. Second, by optimizing the electron and hole transporting layers, it is possible to increase the efficiency of power conversion by lowering the series resistance and increasing the shunt resistance. Additionally, the high electrical resistance of Cs2SnI6 might result in low FF and current density values. Our findings, we hope, will open up new avenues for the fabrication of efficient and stable Cs2SnI6 devices.