The Newton-X Platform for Mixed Quantum–Classical Dynamics
Speaker: Rafael Souza Mattos
Affiliation: Aix-Marseille Université
Type: Contributed talk
Abstract
Mixed quantum–classical dynamics (MQCD) simulations require tight integration between dynamics algorithms, electronic-structure methods, analysis tools, and increasingly machine-learning workflows. This presentation introduces Newton-X 26, a new generation of the Newton-X platform that reorganizes two decades of development into a modular ecosystem for spectrum generation, initial-condition sampling, dynamics propagation, analysis, postprocessing, and data management. A central component of this redesign is a reusable interface layer connecting Newton-X to multiple quantum chemistry programs through a unified architecture decoupled from specific electronic-structure backends. This approach enables interoperability across different electronic-structure methods while facilitating external reuse of the interface infrastructure in other applications. The platform also incorporates tools for automated data curation, machine-learning-assisted workflows, and FAIR-oriented reporting and archiving. Together, these developments provide an open-source environment for routine MQCD simulations and continued method development across heterogeneous electronic-structure ecosystems.