Automated screening of precipitation polymerizations and evaluation using image recognition for divinylbenzene and methacrylic acid

Abstract

AbstractBy applying automated high‐throughput experimentation, 63 precipitation polymerizations of divinylbenzene and methacrylic acid were performed with a total of 1638 samples analyzed by gas chromatography (GC), nuclear magnetic resonance (NMR) spectroscopy, and scanning electron microscopy (SEM). The conversion of each reaction was investigated revealing the best substrate concentrations within the current setup. The GC evaluation was performed automatically via a new custom‐made Python script significantly reducing the time to evaluate the results. Furthermore, the particle growth was monitored by utilizing an innovative image recognition tool to identify particles and their respective sizes using SEM images. Furthermore, a statistical particle size distribution analysis was performed, which is hardly achievable in reasonable time by classical evaluation methods. Using this new procedure, the highest conversion (70%) as well as the largest particles (3700 nm) have been obtained utilizing a high initial monomer (5 vol%) and initiator (5 mol%) concentration. Accordingly, the smallest particles (245 nm) yielded from the lowest starting concentration (1 vol% monomer and 1 mol% initiator).