High-Throughput Virtual Screening of Novel Ternary Phosphosulfides for Photovoltaic Applications

In order to improve the efficiency of solar cells, stable materials with suitable optoelectronic properties must be found. For this purpose, phosphosulfides offer a large and varied range of semiconductors with very diverse compositions, structures, and therefore, optoelectronic properties. This is mainly due to the wide range of oxidations states that phosphorus can take in this class of materials (from -3 to +5). We are currently studying phosphosulfides for solar energy by a multi-faceted approach, including first-principles calculations, high-throughput experiments, and machine learning.<br/>It is estimated that the number of chemically-plausible ternary phosphosulfides is close to 1800, yet just around 100 have been synthesized, and only a handful have been investigated for PV or PEC applications. A high-throughput screening is reported here for over 800 of these materials, obtained from metal substitution in 67 structural prototypes reported in the Materials Project, resulting in some completely unknown potential candidates as wide-gap solar cell absorbers in tandem solar cells. On top of these simple elemental substitutions, we have also explored some previously unreported phosphosulfide stoichiometries, based on analogy with other pnictogens (N, As, Sb) and chalcogens (Se, Te). Unlike the case of most of the already-known phosphosulfides, these structures have phosphorus-rich stoichiometries, which appears to be correlated with favorable optoelectronic properties.<br/>The initial criteria for the screening process are thermodynamic stability, intermediate band gaps and low effective masses. As a result, around 100 new phosphosulfide materials potentially suitable for solar cell applications are obtained from this virtual screening. Most of these compounds are not present in computational materials databases and have never been synthesized. More detailed calculations on a selected subset of these new materials suggest that a few of them deserve experimental investigation as PV absorbers.<br/>The large volume of materials screened gives us a unique opportunity to identify composition-structure-property trends in phosphosulfides with special emphasis on optoelectronic energy conversion applications. We will show how these trends can be used to design new optoelectronic phosphosulfides beyond ternary systems.