Predicting Composition\nEvolution for a Sulfuric Acid-Dimethylamine\nSystem from Monomer to Nanoparticle Using Machine Learning

Experimental and theoretical studies\non the compositional\nchanges\nof new particle formation in the nucleation and initial growth stages\nof acid–base systems (2 and 5 nm) are extremely challenging.\nThis study proposes a machine learning method for predicting the composition\nchange of the sulfuric acid-dimethylamine system in the transformation\nfrom monomer to nanoparticle by learning the structure and composition\ninformation on small-sized sulfuric acid (SA)–dimethylamine\n(DMA) molecular clusters. Based on this method and changes in components,\nwe found that the sulfuric acid-dimethylamine growth was mainly through\nthe alternate adsorption of (SA)<sub>1</sub>(DMA)<sub>1</sub>, (SA)<sub>1</sub>(DMA)<sub>2</sub>, and (SA)<sub>1</sub> clusters at the early\nstage of nucleation, which accounted for about 70, 20, and 10%, respectively.\nThis can explain the nature of possible changes in cluster acidity\nduring the initial nucleation stage for the sulfuric acid-dimethylamine\nsystem. This method can also predict the base-stabilization mechanism\nof the sulfuric acid-dimethylamine system without relying on any experimental\ndata, thereby yielding results that are consistent with those of previous\nexperimental measurement.