An acceleration search method of higher Tc superconductors by a machine learning algorithm

We propose a method to efficiently search for superconductors with higher critical temperature T-c by machine learning based on a superconductor database. The T-c prediction and the search for new superconductors are still difficult problems. With the progress of computer power and calculation algorithms, the possibility of finding new materials with higher T-c at high throughput is emerging. Using the obtained T-c prediction model, the scope is expanded to the search space of multielement materials that have never been searched, and candidates for superconductors with higher T-c that can be synthesized are proposed. (C) 2019 The Japan Society of Applied Physics