Machine-learning approach for discovery of conventional superconductors
Tran Doan Huan, Vu Ngoc Tuoc
Abstract
First-principles computations are the driving force behind numerous discoveries of hydride-based superconductors, mostly at high pressures, during the last decade. Machine-learning (ML) approaches can further accelerate the future discoveries if their reliability can be improved. The main challenge of current ML approaches, typically aiming at predicting the critical temperature ${T}_{\mathrm{c}}$ of a solid from its chemical composition and target pressure, is that the correlations to be learned are deeply hidden, indirect, and uncertain. In this paper, we show that predicting superconductivity at any pressure from the atomic structure is sustainable and reliable. For a demonstration, we curated a diverse data set of 584 atomic structures for which $\ensuremath{\lambda}$ and ${\ensuremath{\omega}}_{log}$, two parameters of the electron-phonon interactions, were computed. We then trained some ML models to predict $\ensuremath{\lambda}$ and ${\ensuremath{\omega}}_{log}$, from which ${T}_{\mathrm{c}}$ can be computed in a postprocessing manner. The models were validated and used to identify two possible superconductors whose ${T}_{\mathrm{c}}\ensuremath{\simeq}10--15$ K at zero pressure. Interestingly, these materials have been synthesized and studied in some other contexts. In summary, the proposed ML approach enables a pathway to directly transfer what can be learned from the high-pressure atomic-level details that correlate with high-${T}_{\mathrm{c}}$ superconductivity to zero pressure. Going forward, this strategy will be improved to better contribute to the discovery of new superconductors.