Chemical reactions are generally governed by statistical effects that can be accurately described by transition state theory. However, several organic reactions have recently been identified that have nonstatisical behavior where transition state theory is insufficient to predict selectivity. This talk will describe our density-functional theory study using quasiclassical direct dynamics implement in DynSuite to model the nonstatistical thermal deazetization of 2,3-diazabicyclo[2.2.1]hept-2-ene. The success and failure of classification machine learning algorithms to predict the deazetization dynamical selectivity and examine the concept of dynamic matching will also be presented.

Machine learning analysis of quasiclassical direct dynamics trajectories for thermal deazetization

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.