Recently, perovskite solar cells showed power conversion efficiencies (PCE) >23%. However, commercialization of perovskite solar cells is hampered by limited long term stability and environmental concerns due to the toxicity of the used lead. Hence, there is a high interest to substitute the lead by more environmentally friendly elements with similar electronic properties. Bismuth is a promising candidate,[1,2] which can form a ternary A₃Bi₂X₉ (A= MA, FA, Cs; X=Cl, Br, I) perovskite. However, these Bi-based perovskite absorbers mainly suffer from poor film morphology[1,3] and band gaps > 2 eV[2,4], resulting in low PCE of only 3 %[1,5] up to now.

To gain more insight into the layer and solar cell formation process, we studied the film morphology, in particular the layer coverage and crystallinity of Cs₃Bi₂I₉ absorber by the influence of different processing methods and post-treatments. In detail, the impact of gas quenching and annealing temperature was analyzed. The quality of the layers was determined by scanning-electron-microscopy, X-ray diffraction, absorption, photoluminescence measurements and IV-curves. The morphology and crystallinity could be improved. For example, by higher annealing temperatures the crystallinity is improved and the crystal grows in a preferred orientation as measured by XRD which leads to higher currents in the solar cell. But still the short circuit current is very low (j_sc= 0.15 mA/cm²) which results in very low 0.04 % PCE values; an already reasonable open circuit voltage of 0.6 V and a fill factor above 40 % could be demonstrated.

In order to further enhance the power efficiency of the Bi-based perovskites the band-gap should be lowered. Sun et al. showed that for the similar system of MA₃Bi₂I₉ the band-gap can be lowered to 1.38 or 1.29 eV [6], respectively by incorporation of Selenium or Sulfur into the perovskite. Here different methods for the selenization and sulfurization of precursor layers are analyzed and considered in terms of their feasibility.

For a selenization the layers are post-treated with selenium-vapor. This method is challenging because high temperatures are needed for high selenium concentration in the vapor. But in those high ranges of temperatures, the perovskite decomposes. Selenium incorporation remains challenging by this direct method.

A solution-processed method for sulfurization is more promising. First results showed black layers, which suggest a lower band-gap. Additionally, those solar cells show a characteristic diode behavior. Further optimization has to be made to improve film morphology and solar cell parameters.