Continuous flow synthesis of pyridinium salts accelerated by multi-objective Bayesian optimization with active learning

Abstract

Human-in-the-loop experimentation enables interactive machine learning for continuous flow chemistry reaction planning and optimization.

Adaptive mixed variable Bayesian self-optimisation of catalytic reactions

“…S4†). 33 Remarkably, the size of the initial dataset had only a small effect on the subsequent speed of convergence towards the optimum in most cases. As a result, optimisations using CP initialisation required on average 21 and 28 fewer total experiments compared to LHC sizes of three and five respectively.…”

Trendbericht Organische Chemie 2024

“…52) Dass Flowsynthese in mehrstufigen Synthesen nützlich ist, zeigen die effizienten Synthesen von Dithiothreitol (DTT), 53) des HIV-Integrase-Inhibitors Dolutegravir (11) 54) und des CB2-Agonisten RG7774 (12). [55][56] Mehrstufige automatisierte Synthesen komplexer Zielmoleküle, die durch Bayes'sche Optimierung gesteuert werden, stellten Baldwin 57) und Bourne 58) vor.…”

A Review of the Applications of Artificial Intelligence in the Process Analysis and Optimization of Chemical Products

“…BO platform (EDBO þ ). 49 They used nmrglue, an open-source Python module, to compare semi-automated NMR spectroscopy analysis data and manual processing methods of spectra collected on low-field (60 MHz) and high-field (400 MHz) NMR spectrometers. The optimization objects, reaction yield, and STY were optimized by changing three continuous variables: residence time (T res ), temperature (T), and the mole fraction of pyridine (C pyr ).…”