M.J. McDermott
Newfound Materials,
United States
Keywords: materials synthesis, discovery, informatics, modeling, AI, ML
Summary:
The scalable synthesis of novel inorganic materials remains a key bottleneck across materials R&D, from energy storage to advanced manufacturing. We present a new software platform that automates retrosynthetic design for inorganic solids, bridging the gap between computational materials prediction and experimental realization. Our system combines solid-state reaction networks, selectivity-driven ranking algorithms, machine-learning models, and cellular automata simulations to predict viable synthesis pathways with high probability of success. Designed for practical integration into R&D workflows, the platform offers an interactive interface for exploring synthesis options, simulating reaction dynamics, and generating optimized reaction recipes. We showcase recent applications of our platform in discovering and optimizing the synthesis of ceramics, energy materials, ferroelectrics, and catalysts. By transforming inorganic synthesis planning into a computationally driven process, our platform aims to guide autonomous exploration of new chemical spaces, accelerate innovation cycles, and improve the manufacturability of next-generation inorganic compounds.