Deep Learning Accelerated Gold Nanocluster Synthesis

The understanding of inorganic reactions, especially those far from the equilibrium state, is relatively limited due to the inherent complexity. Poor understanding of the underlying synthetic chemistry constrains the design of efficient synthesis routes toward the desired final products, especially those at atomic precision. Using the synthesis of atomically precise gold nanoclusters as a demonstration platform, a deep learning framework for guiding material synthesis is successfully developed to accelerate the workflow. With only 54 examples, the graph convolutional neural networks (GCNN) plus siamese neural networks (SNN) classification model is trained. The prediction capability is demonstrated with the successful prediction of literature-reported protocols. In addition, understanding of the synthesis process can be acquired from a decision tree trained by plentiful generated data from a well-trained classification model. This study not only provides a data-driven method accelerating gold nanocluster synthesis, but also sheds light on understanding complex inorganic material synthesis with low data.