Computational molecular science · Machine learning · Spectroscopy

Connecting events across scales: from ultrafast vibrations to slow nucleation.

I am an R&D staff scientist at Oak Ridge National Laboratory. My research combines molecular simulation, statistical mechanics, machine learning, and spectroscopic modeling to understand complex chemical systems and accelerate materials discovery.

Explore my research Google Scholar
60+ publicationsacross chemistry, physics & materials science
DOE PI & Co-PIfundamental and mission-driven research
2024 awardUT–Battelle Research Accomplishment

Theory & computational methods

New methods for dynamics, spectra, and chemical complexity

My research begins with method development: describing rare events and rates, translating molecular dynamics into experimental spectra, and extending first-principles accuracy across length and time scales with deep learning.

Research overview connecting rare-event rate theory, computational spectroscopy, and deep-learning interatomic potentials to molten salts, carbon capture, polymer electrolytes, and bond-breaking chemistry.
One computational framework, three method families, and a broad set of molecular and materials applications.
01Statistical mechanics

Rare events & rate theory

Developing multidimensional reaction coordinates, free-energy descriptions, transition-state theory, reactive-flux corrections, and Marcus-like frameworks to reveal mechanisms and rates that direct simulation cannot readily access.

Key papers

Nucleation Rate Theory for Coordination Number ACS AMI, 2022 ↗ Resolving Heterogeneous Dynamics of Excess Protons in Aqueous Solution with Rate Theory J. Phys. Chem. B, 2020 ↗ On the Relation between Marcus Theory and Ultrafast Spectroscopy of Solvation Kinetics Chem. Phys. Lett., 2018 ↗ Marcus Theory of Ion-Pairing JCTC, 2017 ↗ Reaction Rate Theory in Coordination Number Space J. Phys. Chem. C, 2016 ↗
02Theory meets experiment

Computational spectroscopy

Creating molecular-level interpretations of IR, SFG, 2DIR and 2DSFG spectra, and connecting atomistic simulations to X-ray absorption, X-ray scattering, Raman, and neutron observables.

Key papers

Hidden Features in the OH-Stretching Spectra of Amino-Acid-Decorated Air–Water Interfaces J. Chem. Phys., 2026 ↗ A Holistic Approach for Elucidating Local Structure, Dynamics, and Speciation in Molten Salts JACS, 2021 ↗ Vibrational Spectroscopic Maps & Interactions Chem. Rev., 2020 ↗ Theoretical SFG of Interfacial Water J. Chem. Phys., 2014 ↗ Analysis of 2D Spectra for Non-Gaussian Dynamics J. Phys. Chem. B, 2011 ↗
03AI for atomistic science

Ab initio deep-learning potentials

Training accurate and transferable interatomic potentials on electronic-structure data to reach system sizes, compositions, phases, and time scales beyond conventional ab initio molecular dynamics.

Key papers

Concerted Electron–Ion Transport by Polyacrylonitrile Elucidated with Reactive Deep-Learning Potentials JACS, 2026 ↗ ML Potentials for Molten AlCl₃ Phase Equilibrium J. Phys. Chem. B, 2025 ↗ Deep-Learning Potential for Polyacrylonitrile ACS AMI, 2024 ↗ Transferable Deep-Learning Potential for Molten Salt JACS Au, 2022 ↗

Leadership & impact

Building research programs and scientific communities

I lead and contribute to multidisciplinary programs spanning molten salts, carbon capture, machine-learned potentials, automated discovery, soft materials, and quantum-enabled sensing.

Program leadership

DOE research

Principal investigator in the DOE EFRC Molten Salts in Extreme Environments and co-principal investigator on DOE programs in carbon capture and machine-learned potentials for molten salts.

Emerging directions

Laboratory-directed R&D

Leadership roles in automated materials discovery, transient polyelectrolyte assemblies, critical-materials sensing, phase-change materials, and molecular digital twins.

People

Mentoring

Mentored five ORNL postdoctoral researchers across BES and EERE projects and advised a Georgia Tech doctoral student through the DOE SCGSR program.

Professional community

Scientific service

Organizer and session chair for ACS and APS meetings, DOE proposal reviewer, and referee for leading journals across chemistry, physics, and materials science.

Applications & discoveries

Methods translated into high-impact chemical insight

View all publications
Heavy-element chemistryNature · 2024

Observation of a promethium complex in solution

Synchrotron X-ray absorption and quantum chemistry establish the first coordination structure of promethium in solution and complete the lanthanide series.

About

A computational perspective grounded in physical chemistry

Santanu Roy joined Oak Ridge National Laboratory in 2017 as a postdoctoral research associate in the Chemical Separations Group. He became an R&D associate scientist in 2020 and an R&D staff scientist in 2023. His research brings together molecular dynamics, machine learning, statistical mechanics, and experimentally connected simulation.

Before ORNL, he held postdoctoral appointments at Pacific Northwest National Laboratory and the University of Wisconsin–Madison. He earned his Ph.D. in Natural Sciences from the Center for Theoretical Physics at the University of Groningen's Zernike Institute for Advanced Materials.

ORCID ORNL publications Curriculum vitae
2023–present

R&D Staff Scientist

Oak Ridge National Laboratory

2020–2023

R&D Associate Scientist

Oak Ridge National Laboratory

2017–2020

Postdoctoral Research Associate

Oak Ridge National Laboratory

2014–2017

Postdoctoral Research Associate

Pacific Northwest National Laboratory

2012–2014

Postdoctoral Research Associate

University of Wisconsin–Madison

Ph.D. 2012

Natural Sciences

University of Groningen, The Netherlands

M.Sc. 2006

Physics

University of Pune, India

B.Sc. 2004

Physics Honours

University of Calcutta, India

Connect

Interested in rare-event theory, computational spectroscopy, molecular simulation, machine learning, or collaborative science?

roys@ornl.gov LinkedIn