Home
Multi-Scale Modeling
Data Science in Catalysis
This research program is dedicated to (1) increasing the reliability of computational catalysis research by providing a systematic Bayesian uncertainty quantification framework for integrating experimental data into microkinetic reaction models and (2) accelerating computational catalysis studies with the help of machine learning methods. To ensure usage of the most advanced data science tools, we often collaborate in our method developments with computer science experts.
References:
"A multiple filter based neural network approach to the extrapolation of adsorption energies on metal surfaces for catalysis applications," A. J. Chowdhury, W. Yang, K.E. Abdelfatah, M. Zare, A. Heyden, G. Terejanu, J. Chem. Theory Comput. 16, 1105-1114 (2020).
"Prediction of transition state energies of hydrodeoxygenation reactions on transition metal surfaces based on machine learning," K. Abdelfatah, W. Yang, R.V. Solomon, B. Rajbanshi, A. Chowdhury, M. Zare, S. Kundu, A. Yonge, A. Heyden, G. Terejanu, J. Phys. Chem. C 123, 29804-29810 (2019).
"Prediction of adsorption energies for chemical species on metal catalyst surfaces using machine learning," A. Chowdhury, W. Yang, E. Walker, O. Mamun, A. Heyden, G. Terejanu, J. Phys. Chem. C 122, 28142-28150 (2018).
"Identifying active sites of the water-gas shift reaction over titania supported platinum catalysts under uncertainty," M.A. Walker, D. Mitchell, G.A. Terejanu, A. Heyden, ACS Catalysis 8, 3990-3998 (2018).
"Uncertainty quantification framework applied to the water-gas shift reaction over Pt-based catalysts," E. Walker, S.C. Ammal, G. Terejanu, A. Heyden, J. Phys. Chem. C 120, 10328-10339 (2016).
Mixed Resolution Modeling
This research program is dedicated to developing mixed-resolution models that combine the efficiency of coarse-grained models, that lump a group of atoms into a pseudo-atom whose motion is governed by a simplified potential, with the accuracy of atomistic models for systems that require atomistic resolution only locally, for example at a reactive group, defect, or interface.
Our method is called adaptive partitioning of the Lagrangian (APL) which is currently the only method for which symplectic integrators have been developed for mixed-resolution systems in which the resolution of selected groups of atoms change during a simulation.
References:
"Solving the equations of motion for mixed atomistic and coarse-grained systems," S. Pezeshki, C. Davis, A. Heyden, H. Lin, J. Chem. Theory Comput. 10 (11), 4765-4776 (2014).
"Solving the equations of motion for mixed atomistic and coarse-grained systems," J. H. Park, A. Heyden, Mol. Sim. 35, 962-973 (2009).
"Conservative algorithm for an adaptive change of resolution in mixed atomistic / coarse-grained multiscale simulations," A. Heyden and D. G. Truhlar, J. Chem. Theory Comp. 4, 217-221 (2008).
"Adaptive partitioning in combined quantum mechanical and molecular mechanical calculations of potential energy functions for multiscale simulations," A. Heyden, H. Lin, and D. G. Truhlar, J. Phys. Chem. B 111, 2231-2241 (2007).
