Guest talk: Prof. Sergei Tretiak, USA: In the quest for excited states, from machine learning to non-adiabatic dynamics

When: Thu, 20.10.22, 9:30 Uhr

Where: Institut für Theoretische Chemie, Seminarraum, Währinger Straße 17, 4. Stock

Prof. Dr. Sergei Tretiak

Theoretical Division, Center for Nonlinear Studies, and Center for Integrated Nanotechnologies, Los Alamos National Laboratory, Los Alamos NM, 87545. E-mail: serg@lanl.gov

Machine learning (ML) is a premier tool for modeling chemical processes and materials. Designing high-quality training data sets is crucial to overall model accuracy. I will describe the active learning strategy, in which new data are automatically collected for atomic configurations that produce large ML uncertainties. The locality approximation is another severe limitation that fails to capture long-range effects. I will discuss how ML models can overcome nonlocality. These advances are exemplified by applications to molecules and materials. Explosive growth of user-friendly ML frameworks, designed for chemistry, demonstrate that the field is evolving towards physics-based models augmented by data science. I will further overview some applications of Non-adiabatic EXcited-state Molecular Dynamics (NEXMD) framework developed at several institutions and released to public. The NEXMD code is able to simulate tens of picoseconds photoinduced dynamics in large molecular systems. As an application, I will exemplify ultrafast coherent excitonic dynamics guided by intermolecular conical intersections (CoIns). In the second example, we use NEXMD simulations to compute X-ray Raman signals, which are able to sensitively monitor the coherence evolution. These spectroscopic signals are possible to measure at XFEL facilities. Our modeling results allow us to understand and potentially manipulate excited state dynamics and energy transfer pathways toward applications.

1.       J. S. Smith, B. Nebgen, N. Mathew, J. Chen, N. Lubbers, L. Burakovsky, S. Tretiak, H. Ah Nam, T. Germann, S. Fensin, K. Barros, “Automated discovery of a robust interatomic potential for aluminum” Nature Comm. 12, 1257 (2021).

2.       R. Zubatyuk, B. Nebgen, J. S. Smith, S. Tretiak and O. Isayev, “Teaching neural network to attach and detach electrons from molecules” Nature Comm. 12, 4870 (2021).

3.       N. Fedik, R. Zubatyuk, N. Lubbers, J. S. Smith, B. Nebgen, R. Messerly, Y. W. Li, M. Kulichenko, A. I. Boldyrev,  K. Barros, O. Isayev, and S. Tretiak “Extending machine learning beyond interatomic potentials for predicting molecular properties” Nature Rev. Chem. 6, 653 (2022).

4.       T. Nelson, A. White, J. Bjorgaard, A. Sifain, Y. Zhang, B. Nebgen, S. Fernandez-Alberti, D. Mozyrsky, A. Roitberg and S. Tretiak, “Non-adiabatic Excited-State Molecular Dynamics: Theory and Applications for Modeling Photophysics in Extended Molecular Materials” Chem. Rev. 120, 2215 (2020).

5.       A. De Sio, E. Sommer, X. T. Nguyen, L. Gross, D. Popović, B. Nebgen, S. Fernandez-Alberti, S. Pittalis, C. A. Rozzi, E. Molinari, E. Mena-Osteritz, P. Bäuerle, T. Frauenheim, S. Tretiak, C. Lienau, “Intermolecular conical intersections in molecular aggregates” Nature Nanotech., 16, 63 (2021).

6.       D. Keefer, V. M. Freixas, H. Song, S. Tretiak, S. Fernandez-Alberti, and S. Mukamel, “Monitoring Molecular Vibronic Coherences in a Bichromophoric Molecule by Ultrafast X–Ray Spectroscopy” Chem. Sci. 12, 5286 (2021).  


Sergei Tretiak is a Laboratory Fellow of Los Alamos National Laboratory (LANL). He received his Chemistry doctorate in 1998 from the University of Rochester. He was then a Postdoctoral Fellow (1999-2001), and subsequently became a staff scientist at LANL. He became an American Physical Society Fellow (APS) in 2014 and a Fellow of the Royal Society of Chemistry, (FRSC) in 2019. He has also received the Humboldt Research Award (2021), the LANL Postdoctoral Distinguished Mentor Award (2015) and the LANL Fellow's Prize for Research (2010). His research interests include development of electronic structure methods for molecular optical properties, non-adiabatic dynamics of electronically excited states, optical response of excitons in conjugated polymers, carbon nanotubes, semiconductor nanoparticles, mixed halide perovskites and molecular aggregates, and the use of Machine Learning for chemistry and materials. Tretiak has published nearly 400 articles cited more than 23,000 times (h-index=74, WebOfSci) and presented more than 300 invited talks. 

Organised by the Institute of Theoretical Chemistry: González Research Group (univie.ac.at)