CAREER: Multiscale Photodynamics Simulations in Solvated and Crystalline Environments

点击翻译按钮获取中文摘要

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2).With support from the Chemical Structure, Dynamics & Mechanisms-B Program of the Chemistry Division, Steven A. Lopez of Northeastern University is using computational techniques to discover new light-promoted (photochemical) reactions. Photochemical reactions are attractive to several research sectors because they avoid the need for extensive heating and expensive catalysts to access complex, high-energy molecules. Photochemical reactions occur on very fast timescales, typically less than one-millionth of a second, making observation of intermediate structures with experiments very challenging. This project aims to use computational and machine learning techniques to understand the reactivities and selectivities of these photochemical reactions. Research in the Lopez group will enable computational predictions in realistic, complex chemical environments (e.g., solvent and crystalline phase) towardsmore accurate predictions and design principles. This work is at the intersection of data science, organic, and physical chemistry and will support the interdisciplinary training of young scientists at all levels. Dr. Lopez and his team will create "pandemic-proof" Summer Research Experiences for community college students across the United States and engage with the Alliance for Diversity in Science and Engineering to parallelize the outreach impact. Steven A. Lopez and his research group plan to apply and develop computational and machine learning techniques to predict photochemical reaction outcomes, mechanisms, and stereoselectivities in complex environments (e.g., molecular solids and solvated systems). Unlike thermal reactions, structure-property relationships are more complex and difficult to understand for photochemical reactions. The general lack of excited-state structural information has limited structure-reactivity relationships and slowed the discovery of high-yielding, selective reactions. Experimental and computational techniques cannot resolve dynamic excited-state structures of short-lived molecular excited states (nano- to femtosecond scale). This project will enable the comprehensive exploration of the reactivities and stereoselectivities of gas-evolving reactions with multiconfigurational quantum chemical calculations and machine-learning-accelerated non-adiabatic molecular dynamics simulations. The research group will focus on parent and substituted triazolines, pyrazolines, and diazirines. This project aims to resolve the mechanisms and structures of molecular excited states, thusly targeting a knowledge gap towards structure-reactivity relationships. A second phase will evaluate the role of the chemical environment on the excited- and ground-state components of these reactions, enabled by an open-access machine learning code Python Rapid Artificial Intelligence Ab Initio Molecular Dynamics (PyRAI2MD). The anticipated mechanistic insights have the potential to enable future design of light-responsive frameworks (e.g. covalent organic frameworks and metal-organic frameworks, COFs and MOFs) and molecular machines in the longer term.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.