About this Abstract |
Meeting |
2020 TMS Annual Meeting & Exhibition
|
Symposium
|
Computational Discovery and Design of Emerging Materials
|
Presentation Title |
Exploring Van der Waals 2D Heterostructures using a Combined Machine Learning and Density Functional Theory Approach |
Author(s) |
Daniel Willhelm, Nathan Wilson, Tahir Cagin, Raymundo Arroyave, Ruth Pachter, Xiaofeng Qian |
On-Site Speaker (Planned) |
Daniel Willhelm |
Abstract Scope |
Van der Waals (vdW) heterostructures consisting of vertically stacked two-dimensional (2D) materials offer an exciting opportunity for the direct tailoring of electronic structure with potential applications in ultrathin electronics and optoelectronics. In many cases, the properties of these heterostructures are unique to the individual layers, which allows for precise tuning of properties by selective stacking orders. As the number of potential 2D monolayers grows, the number of possible combinations increases dramatically. Consequently, traditional material exploration of the vast heterostructure design space is costly and time-consuming, even with the use of modern high-throughput computing. Here we present a data-driven approach to predict the properties of vdW heterostructures including bandgap energy, band alignment type, interlayer distance, etc. using thousands of unique bilayers and trilayers constructed from several 2D material families. Our approach combines machine learning methods with density functional theory which significantly reduce computational costs and speed up materials discovery and design. |
Proceedings Inclusion? |
Planned: Supplemental Proceedings volume |