Autonomous phase diagram mapping via real-time self-driving experiment-CALPHAD closed-loop interaction

Iterative cycles of theoretical prediction and experimental validation are the cornerstone of the modern scientific method. However, the proverbial “closing of the loop” in experiment-theory cycles in practice are usually ad hoc and often inherently difficult, beset by the scale or time constraint of computation or phenomena. Here, we demonstrate Autonomous MAterials Search Engine (AMASE), where self-driving continuous cyclical interaction of experiments and computational predictions is performed for materials exploration. We have applied this formalism to rapid mapping of a temperature–composition phase diagram. Experimental determination of phase boundaries in thin films are autonomously interspersed with real-time updating of phase diagram prediction using CALPHAD. AMASE was able to accurately determine the eutectic phase diagram of the Sn-Bi thin-film system from a self-guided campaign covering just a small fraction of the phase space, translating to a 6-fold reduction in the number of experiments. This study demonstrates real-time, autonomous, and iterative interactions of experiments and theory carried out without any human intervention.