Multivariate Analysis of Mixed Ternary and Quaternary A-Site Organic Cations in Tin Iodide Perovskite Solar Cells

Connecting data science (machine learning) with experimental methods is critical for accelerating material science research. Herein, we report a multivariate analysis for exploring A-site organic cation mixing in tin iodide perovskite (ASnI3) solar cells (PSCs), which are the most suitable Pb-free PSC candidates. To address the common drawbacks of Sn perovskites (facile oxidation of Sn2+ to Sn4+ and large degree of mixing), we proposed an efficient experimental screening method using 133 types of environmentally stable A2Sn(IV)I6 zero-dimensional pseudoperovskites to predict the power conversion efficiency (PCE) of ASn(II)I3, in which A is a ternary or quaternary mixed organic cation (namely, metylammonium, formamidinium (FA), dimethylammonium, guanidinium, ethylammonium, acetamidinium, trimethylammonium, imidazolium, or phenylethylammonium (PEA)). The high correlation coefficient of our model (0.953) and experimental validation (0.982) allowed us to identify a new (FA0.92IM0.08)0.9PEA0.1SnI3 Sn-PSC with a PCE of 7.22%. Our results provide a basis for exploring A-site cation mixing in Sn-PSCs for improving their performance.