Using Machine Learning to Predict and Understand Complex Self‐Assembly Behaviors of a Multicomponent Nanocomposite

Blends of nanoparticles, polymers, and small molecules can self‐assemble into optical, magnetic, and electronic devices with structure‐dependent properties. However, the relationship between a multicomponent nanocomposite's formulation and its assembled structure is complex and cannot be predicted by theory. The blends can be strongly influenced by processing conditions, which can introduce non‐equilibrium states. Currently, nanocomposite devices are designed through cycles of experimental trial and error. Machine learning (ML) methods are a compelling alternative because they can use existing datasets to map high‐dimensional spaces. These methods do not rely on known relationships between parameters, so they are suited to complex systems without a solid theoretical foundation. Here, a dataset of 595 microscopy images of nanocomposite thin films is used to train a series of ML models. Correlations between the input and output parameters are examined, providing new insights into the system. Finally, the most successful ML model is used to predict the structures of new nanocomposite compositions. The results confirm that ML techniques can be used to improve the efficiency of nanocomposite device design. More broadly, the current study suggests some of the advantages and challenges associated with applying ML to complex systems. The phase behavior of multicomponent nanocomposites is complex and cannot be predicted by theory alone. Machine learning (ML) methods are a compelling alternative to experimental trial and error. In this work, 595 microscopy images of nanocomposites are used to train a series of ML models. The predictions provide insights into the system and suggest ML can significantly streamline the design of nanocomposites.