Simulation\nStudy on the All-Inorganic CsSn<sub><i>x</i></sub>Ge<sub>1–<i>x</i></sub>I<sub>3</sub>‑Based Perovskite\nSolar Cells Using Isotypic Perovskites as\nHole Transport Layers

All-inorganic Sn–Ge-based perovskite solar cells\n(PSCs)\nhave made great progress in recent years. Furthermore, they can be\nused as promising lead-free absorbers for PSCs, and <i>p</i>-type-doped CsSnI<sub>3</sub>, CsGeI<sub>3</sub>, and CsSn<sub>0.5</sub>Ge<sub>0.5</sub>I<sub>3</sub> could also be used as good hole transport\nlayers (HTLs). In this simulation work, CsSnI<sub>3</sub>, CsGeI<sub>3</sub>, and CsSn<sub>0.5</sub>Ge<sub>0.5</sub>I<sub>3</sub> are\nused as both absorbers and HTLs. The effects of the dopant concentration\nof HTLs, the thickness of absorbers, and HTLs on the photovoltaic\nperformance of PSCs were studied to optimize the device structures.\nThe maximum efficiencies from high to low are 28.35%, 26.35%, 25.84%,\n25.23%, 18.83%, 17.49%, and 11.79% for the TiO<sub>2</sub>/<i>i</i>-CsSnI<sub>3</sub>/<i>p</i>-CsSnI<sub>3</sub>, TiO<sub>2</sub>/<i>i</i>-CsSn<sub>0.5</sub>Ge<sub>0.5</sub>I<sub>3</sub>/<i>p</i>-CsSn<sub>0.5</sub>Ge<sub>0.5</sub>I<sub>3</sub>, TiO<sub>2</sub>/<i>i</i>-CsSn<sub>0.5</sub>Ge<sub>0.5</sub>I<sub>3</sub>/<i>p</i>-CsSnI<sub>3</sub>, TiO<sub>2</sub>/<i>i</i>-CsSnI<sub>3</sub>/<i>p</i>-CsGeI<sub>3</sub>, TiO<sub>2</sub>/<i>i</i>-CsSn<sub>0.5</sub>Ge<sub>0.5</sub>I<sub>3</sub>/<i>p</i>-CsGeI<sub>3</sub>, TiO<sub>2</sub>/<i>i</i>-CsGeI<sub>3</sub>/<i>p</i>-CsGeI<sub>3</sub>, and TiO<sub>2</sub>/<i>i</i>-CsGeI<sub>3</sub>/<i>p</i>-CsSnI<sub>3</sub>, respectively. The TiO<sub>2</sub>/<i>i</i>-CsGeI<sub>3</sub>/<i>p</i>-CsSnI<sub>3</sub> cell exhibits the lowest efficiency of 11.79% in all of the\nsimulated PSCs due to the spike-like band offset at the <i>i</i>-CsGeI<sub>3</sub>/<i>p</i>-CsSnI<sub>3</sub> interface\nand high recombination rate in the <i>p</i>-CsSnI<sub>3</sub> region. It is found that the <i>n</i>-<i>p</i> structures could have better photovoltaic performance (thickness\nof <i>i</i>-film approaching zero) than the conventional <i>n</i>-<i>i</i>-<i>p</i> structures for the\nTiO<sub>2</sub>/<i>i</i>-CsSnI<sub>3</sub>/<i>p</i>-CsSnI<sub>3</sub>, TiO<sub>2</sub>/<i>i</i>-CsGeI<sub>3</sub>/<i>p</i>-CsGeI<sub>3</sub>, and TiO<sub>2</sub>/<i>i</i>-CsSn<sub>0.5</sub>Ge<sub>0.5</sub>I<sub>3</sub>/<i>p</i>-CsSn<sub>0.5</sub>Ge<sub>0.5</sub>I<sub>3</sub> PSCs if the defects in HTLs created by high doping can be effectively\ncontrolled. The efficiencies of PSCs are sensitive to the defect density\nand defect level position, and the influence of defect density on\nthe PV performance is larger than that of the defect level position.\nThe solar cells could maintain high power conversion efficiency for\ndefect density below about 5 × 10<sup>17</sup> cm<sup>–3</sup>. Furthermore, the increase of the interface trap density is found\nto reduce the photovoltaic performance of PSCs. Our study provides\ninsight into the optimal design of CsSn<sub><i>x</i></sub>Ge<sub>1–<i>x</i></sub>I<sub>3</sub>-based PSCs.