Understanding the phase preferences of transition metal dichalcogenides using machine learning

Transition metal dichalcogenides (TMDs) are layered 2D quantum materials that preferentially adopt one of the several possible crystal structures. Since the crystal structure dictates electronic structure properties of a material, one of the natural questions that arises is: What dictates the observed structural preference of TMDs? An answer to this question can help to understand composition-structure-property relationships, paving the way for engineering properties of TMDs. In a recent work, we combined high-throughput quantum mechanical computations with machine learning tools to address this six-decade old problem. Our work demonstrates how machine learning can be used to tackle old problems in Condensed Matter Physics.