Statistical Thermodynamics & Molecular Simulations (STMS) Seminar Series

​These seminar series are aimed at providing a virtual platform for sharing scientific research in the area of statistical mechanics, molecular simulations, and computational materials science. Since early 2020, the coronavirus pandemic has disrupted many large in-person scientific gatherings, including conferences and department seminars, and it is not clear that the situation will improve any time soon. STMS is aimed at filling this gap, and provide a venue for dissemination of research findings and exchange of ideas in the age of COVID.  This model is being currently used by several other scientific communities, and can potentially continue even beyond the pandemic if successful. 

​Each seminar will be a 60-minute event and will comprise of a long-form (30-minute) talk by a principal investigator or a senior research scientists from academia or industry and a short-form (15-minute) presentation by a graduate student or a postdoc. The remainder of the event will be dedicated to Q&A (10 minutes for the PI, 5 minutes for the student/postdoc). Long-form speakers will be chosen by the STMS Organizing Committee, while we encourage suggestions from the community at large. Student and postdoctoral speakers can either be nominated by their advisors or can self-nominate themselves by sending a CV to the organizers.  During 2022 we expect to hold two seminar per month, and the events will take place in the last two Fridays of each month, from 10:45 AM-12:00 PM Eastern Standard Time (EST):

​This event's talks:

​Exploring the reactivity landscape of porous catalysts using molecular simulations and machine learning

​Prof. Peng Bai (University of Massachusetts, Amherst)

​Abstract: Porous materials such as zeolites and metal-organic frameworks are often described as enzyme-like industrial catalysts. They exhibit exquisite molecular shape selectivity due to the interactions offered by the extended reaction environments. This characteristic, however, present significant challenges to the predictive modeling of reactions occurring insides these materials, especially for large, articulated molecules. Using the example of hydrocarbon conversions in zeolites, this talk will present our recent work coupling zero-Kelvin quantum-chemical methods and forcefield-based molecular simulations to enable a consistent calculation of local activation energies for bulky reactants. Such site-dependent reactivity information further allows us to attempt at computing ensemble-averaged barriers. In a separate direction, I will present results investigating various computer vision techniques for characterizing the shape of pore space in porous materials. We are envisioning a converging future where forcefields and machine learning techniques address and guide the sampling of important states in computational catalysis modeling.

​Speaker Bio:  Peng Bai is an Assistant Professor of Chemical Engineering at the University of Massachusetts Amherst. He obtained his B.Sc. degree from Tsinghua University in China and received both his Ph.D. degree and postdoctoral training from the University of Minnesota. His group's research focuses on the development and application of molecular simulation, first-principles, and data science methods to study catalysis and electrochemistry in complex environments, and adsorption and diffusion phenomena. Specific applications include the catalytic upcycling of polymers, separation with nanoporous materials, and ion conduction in solid-state batteries. He is the recipient of the NSF CAREER Award and an ACS PRF Doctoral New Investigator.

​Inverse design of self-assembling soft and porous materials via free energy- and deep learning-based evolution strategies

​Dr. Alberto Perez de Alba Ortiz (Utrecht University)

​Abstract: Self-assembly provides a promising route to synthesize nanomaterials with desired physical and chemical properties. While precise control of such ordering and aggregation processes remains a daunting task, computational inverse design can guide and boost their realization. Here, we apply evolutionary computation—i.e., covariance matrix adaptation evolution strategy—to reverse-engineer the formation of various target phases. Our typical design parameters include temperature, volume, pressure, terms of the interaction potential and molar fractions. Our target phases include FCC, HCP and diamond, as well as porous phases with square and hexagonal pores, and zeolites. We successfully stabilize a given target phase by optimizing a fitness across few generations of Monte Carlo or Molecular Dynamics sampling. Employing a machine-learned fitness based on a convolutional neural network classifier of diffraction patterns, we can flexibly target different phases without previous knowledge about their order parameters. Alternatively, using a free energy-based fitness along an order parameter—e.g., environment similarity or permutation invariant vector—we can accelerate sampling and control the stability of a target phase. By including multiple order parameters, we can design minimum free-energy pathways between different metastable phases. Our design pipeline provides a robust tool to control not only stable phases, but also transition pathways and metastable states of complex soft matter systems.

​Speaker Bio: Alberto is currently a postdoctoral researcher in Prof. Marjolein Dijkstra’s Soft Condensed Matter group at Utrecht University, where he focuses on the inverse design of free-energy landscapes for self-assembling soft materials. Previously, he completed his PhD, cum laude, in Dr. Bernd Ensing’s Computational Chemistry group at the University of Amsterdam, where he developed and applied path-based free-energy methods for biomolecular transitions.  He obtained his MSc in Computational Science and Engineering at the Technical University of Munich, where he wrote his thesis on adaptive quantum mechanics/molecular mechanics simulations of proton transfers at Prof. Karsten Reuter’s Theoretical Chemistry group and worked on enhanced sampling of biomolecular conformations at Prof. Martin Zacharias’ Biomolecular Dynamics group. Before that, he studied his BSc in Engineering Physics at the Monterrey Institute of Technology and Higher Education and worked for the steel and cement industries in his native Mexico.