(27g) An Adaptive Learning Data Driven Method Based on Molecular Descriptors for Solvent Selection of Mitsunobu Esterification

Optimization of a synthetic reaction with respect of solvent choice and operating conditions was implemented as a machine learning-based workflow. The approach is exemplified on the case study of selection of a promising solvent to maximise the yield of a Mitsunobu esterification reaction producing isopropyl benzoate. A solvent was defined with 15 molecular descriptors and a library of solvent descriptors was built. The descriptors were converted into a reduced dimensionality form using an Autoencoder. Experimental yields were used to train a multi-layered Artificial Neural Network (ANN) surrogate model, which was used for the optimisation and design of experiments (DoE). DoE was performed in an active learning mode in order to reduce the number of experiments required for the reaction optimization. The final surrogate model identified 1-chloropentane as a promising solvent which resulted in the experimental yield of 93%.