Research Conducted Using Datachemical LAB Published in an International Peer-Reviewed Journal

A research group led by Professor Momiyama at the Institute for Molecular Science has published a paper demonstrating the data-driven design of additives that enable efficient halogenation of highly fluorinated naphthalenes at room temperature.

In this study, a machine learning model was built using Datachemical LAB. Molecular property descriptors derived from quantum chemical calculations (DFT) — including charge distribution and contributions to aromaticity — were used as inputs to develop a predictive model for identifying effective additives. The work demonstrates a shift in additive development from conventional trial-and-error approaches to an effective, machine learning-driven strategy.

The paper is available here.