Solvent Dependence\nof Ionic Liquid-Based Pt Nanoparticle\nSynthesis: Machine Learning-Aided In-Line Monitoring in a Flow Reactor

Colloidal platinum nanoparticles (Pt NPs) possess a myriad\nof technologically\nrelevant applications. A potentially sustainable route to synthesize\nPt NPs is <i>via</i> polyol reduction in ionic liquid (IL)\nsolvents; however, the development of this synthetic method is limited\nby the fact that reaction kinetics have not been investigated. In-line\nanalysis in a flow reactor is an appealing approach to obtain such\nkinetic data; unfortunately, the optical featurelessness of Pt NPs\nin the visible spectrum complicates the direct analysis of flow chemistry\nproducts <i>via</i> ultraviolet–visible (UV–vis)\nspectrophotometry. Here, we report a machine learning (ML)-based approach\nto analyze in-line UV–vis spectrophotometric data to determine\nPt NP product concentrations. Using a benchtop flow reactor with ML-interpreted\nin-line analysis, we were able to investigate NP yield as a function\nof residence time for two IL solvents: 1-butyl-1-methylpyrrolidinium\ntriflate (BMPYRR-OTf) and 1-butyl-2-methylpyridinium triflate (BMPY-OTf).\nWhile these solvents are structurally similar, the polyol reduction\nshows radically different yields of Pt NPs depending on which solvent\nis used. The approach presented here will help develop an understanding\nof how the subtle differences in the molecular structures of these\nsolvents lead to distinct reaction behavior. The accuracy of the ML\nprediction was validated by particle size analysis and the error was\nfound to be as low as 4%. This approach is generalizable and has the\npotential to provide information on various reaction outcomes stemming\nfrom solvent effects, for example, differential yields, orders of\nreaction, rate coefficients, NP sizes, <i>etc.</i>