Perovskites (of the ABX3-stochiometry) are excellent candidates for novel solar cells due to low cost fabrication (solution processing) and tunable bandgap by ion-replacement, which opens the door for targeted design of tandem solar cells. Most successful perovskites are based on organic, lead-halide perovskites like MaPbI3 or FaPbI3. Despite their excellent performance in lab settings, their instablities and toxicity prevent large scale production of such solar cells. Recent research aims the design of novel lead-free, inorganic perovskites. We develop machine learning methods in combination with Density Functional Theory to allow a high throughput sampling of possible materials with desired optoelectronic characteristics.
Covered Topics: Density Functional Theory, Machine Learning, Data science
