Abstract
Quaternary chalcogenide semiconductors (I2-II-IV-X4) are key materials for thin-film photovoltaics (PVs) to alleviate the energy crisis. Scaling up of PVs requires the discovery of I2-II-IV-X4 with good photoelectric properties; however, the structure search space is significantly large to explore exhaustively. The scarcity of available data impedes even many machine learning (ML) methods. Here, we employ the unsupervised learning (UL) method to discover I2-II-IV-X4 that alleviates the challenge of data scarcity. We screen all the I2-II-IV-X4 from the periodic table as the initial data and finally select eight candidates through UL. As predicted by ab initio calculations, they exhibit good optical conversion efficiency, strong optical responses, and good thermal stabilities at room temperatures. This typical case demonstrates the potential of UL in material discovery, which overcomes the limitation of data scarcity, and shortens the computational screening cycle of I2-II-IV-X4 by ~12.1 years, providing a research avenue for rapid material discovery.
Highlights
Solar energy is the most important basic energy among all types of renewable energy[1,2]
Workflow of material discovery The workflow for unsupervised discovery of I2-II-IV-X4 chalcogenides for thin-film PVs is illustrated in Fig. 1, including four modules: determination of crystal structures (Fig. 1a–d)[38], element selection from the periodic table (Fig. 1e), the establishment of the machine learning (ML) model (Fig. 1f), and ab initio calculation (Fig. 1g)
Our achievements are as follows: (1) We propose an accessible descriptor of sums and differences of elemental properties (SDEPs) based on the isolated elemental properties to obtain the feature vector of I2-II-IV-X4 compounds, which can be expanded to other material systems. (2) eight I2-II-IVX4 compounds (Ag2BaTiS4, Ag2BaTiSe4, Ag2BaCrS4, Ag2BaSiSe4, Ag2BaZrS4, Ag2BaZrSe4, Ag2BaHfSe4, and Cu2BaMnSe4) with optimal band gaps, desired optical absorptions, and practical thermal stabilities at room temperatures were selected out of 2700 original structures by unsupervised learning (UL)
Summary
Solar energy is the most important basic energy among all types of renewable energy[1,2]. Their battery power conversion efficiency (PCE) exceeds 20% at present[6,7] These materials require expensive or rare elements (In, Te), or even toxic (Cd), severely limiting their largescale development. CZTSSe, where the smaller Zn of is replaced by Ba with a larger ionic radius (Cu2BaSnS4-xSex(CBTSSe)) to ease the antisite disorder, has demonstrated better performance in PV in comparison with CZTSSe9–12.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
