Department Seminar Series: Luke Achenie, Ph.D.

This event is part of the CBEE DEPARTMENT SEMINAR SERIES

SAVE THE DATE - 

CBEE Department Seminar Series

Luke Achenie, Ph.D. 

Professor of Chemical Engineering
Virginia Tech Faculty of Health Sciences 

Location: ENG 336

Title: What a background in chemical engineering and applied mathematics can lead to: an interesting academic career

Abstract: 

My research is in multiscale and multi-physics scientific computing/modeling with optimization as a core technology. I have done modeling in (a) molecular systems (molecular dynamics of gas permeation through organic-inorganic membranes and DFT modeling of solvent effects in Kolbe-Schmidt reactions, (b) biological systems (pharmacokinetic modeling in oral drug delivery, drug transport across blood-brain barrier), (c) digital twins for health and (d) machine learning/AI applications.

In this talk I will cover some topics during my academic journey as follows.

Organic-Inorganic Membranes: I will discussour research in molecular dynamics (MD) simulation of gas permeation in organic-inorganic membranes. Specifically, we have investigated the separation of CO2/CH4 gas mixture with a hybrid inorganic-organic membrane. We studied how the radius of pores in the structure and the number of inserted phenyl groups affected the overall separation efficiency. For a membrane with pore size larger than 3.5 Å, the separation followed Knudsen diffusion. For small pore radius the inserted phenyl groups helped the selectivity but caused the permeances to decrease at the same time. With the computational membrane structure we created, the separation showed close selectivity to the experimental data. Later we developed agent-based models to speed up the MD simulations.

Machine Learning Accelerated Catalyst Design: Currently we have employed ML for computational catalysis. Here we employed a machine-learning-augmented chemisorption model that enables fast and accurate prediction of surface reactivity of metal alloys within a broad chemical space. Specifically, used neural networks, trained with a set of ab initio adsorption energies on idealized bimetallic surfaces and their electronic fingerprints to capture complex, non-linear adsorbate-substrate interactions with ~ 0.1 eV error (e.g., CO adsorption), outperforming the standard d-band theory of chemisorption in prediction. By leveraging scaling relations between adsorption energies of similar adsorbates, we illustrate that this integrated approach greatly facilitates high-throughput catalyst screening and, as a specific case, suggests promising {100}-terminated multimetallic alloys with improved efficiency and selectivity for CO2 electroreduction to C2 species. Statistical analysis of the network response to perturbations of input features further consolidates our understanding of chemical bonding on metal surfaces.

I will devote most of my lecture on computational oral drug delivery in which we track the transport of the drug active (which is encapsulated in the drug excipients) as it makes its way through the gastrointestinal tract into general blood circulation. We couple the transport model with pharmaco-kinetics (what the body does to the drug) and pharmaco-dynamics (what the drug does to the body). Finally, I will discuss digital twins for health which is a new frontier for us.

Biography: 

Dr. Luke Achenie is Professor of Chemical Engineering at the Virginia Polytechnic Institute and State University (Virginia Tech), in Blacksburg, Virginia. Luke obtained his undergraduate degree at MIT (Cambridge, MA).  Prior to joining Virginia Tech in August of 2007, he was at the University of Connecticut (Storrs, CT) from 1991 until 2007. He has an M.S. degree in Applied Math (Carnegie Mellon University, Pittsburg, PA) and graduated in 1988 with a PhD in Chemical Engineering also at Carnegie Mellon. Before joining academia, he did a three-year stint as Associate Research Engineer at Shell Development Company in Houston, Texas. At Shell he developed advanced process control and optimization algorithms for oil refineries. He served as a CBET/Program Officer (rotator) at the National Science Foundation in 2012/2013.

Achenie’s primary research interests are in Machine Learning, Molecular Modeling, Agent-Based Multi-Scale Modeling, and Drug Delivery Modeling. He has published over 260 archival journal articles together with over 140 proceedings and technical presentations. His research activities led to an induction in the Connecticut Academy of Science and Engineering (CASE) in 2007. He serves as editorial board member of three journals. He is also an AIChE Fellow. Achenie has served in many roles with the American Institute of Chemical Engineers (AIChE) including service as a programming chair in the process systems engineering and as  a member of the Board of Trustees of AIChE. He is currently the in-coming chair of AIChE/PAIC Committee. He is a MAC Eminent Award winner.  Achenie is a very active member of the ACS (American Chemical Society) and has held a number of executive positions within the COMP Division and the ACS/MPPG.