Machine learning-guided synthesis of advanced inorganic materials

Abstract

Synthesis of materials with minimum number of trials is of paramount importance towards the acceleration of advanced materials development. The enormous complexity involved in existing multi-variable synthesis methods leads to high uncertainty, numerous trials and exorbitant cost. Recently, machine learning (ML) has demonstrated tremendous potential for material discovery and property enhancement. Here, we extend the application of ML to guide material synthesis process through the establishment of the methodology including model construction, optimization, and progressive adaptive model (PAM). Two representative multi-variable systems are studied. A classification ML model on chemical vapor grown MoS2 is developed, capable of optimizing the synthesis conditions to achieve a higher success rate. And a regression model is constructed on the hydrothermal-grown carbon quantum dots, to enhance the process-related properties such as the photoluminescence quantum yield. The importance of synthesis parameters on experimental outcomes is particularly extracted from the constructed ML models. Furthermore, off-line analysis shows that enhancement of the experimental outcome with minimized number of trials can be achieved with the effective feedback loops in PAM, suggesting the great potential of involving ML to guide new material synthesis at the beginning stage. This work serves as a proof of concept for using ML in facilitating the synthesis of inorganic materials, thereby revealing the feasibility and remarkable capability of ML in opening up a new promising window for accelerating material development.