Concluding Remarks

Catalysis is a key technology in the production of value-added chemicals from widely abundant feedstocks with applications in fields such as in renewable energy, pollution management, ammonia production and carbon sequestration. Much of the research in this field has been driven by a search for the most optimally active catalyst. For over a century, the field of heterogeneous catalysis has been dictated by the Sabatier principle which states that the binding strength between an optimally active catalyst and reactant is neither too strong nor too weak. This implicitly caps the activity of any optimal catalyst, theoretical or otherwise, to what is known as the Sabatier limit. A focal point for many in this field therefore seeks to defy these limitations set by this age-old adage.

Research Interests

Dynamic catalysis can provide a viable strategy to do so by optimizing the adsorption strength of the surface and reactant followed by a modification of the properties of the catalyst to optimize the desorption strength of the surface and product. Such a catalytic system can also potentially be used to modulate product selectivity and other catalytic properties that are inaccessible through static systems. However, doing so requires viable catalysts with functional properties that can by varied under an applied stimulus. Quantum materials exhibit such functionality with Mott insulators having the unique ability to transition between metals and insulators. Here, I present my plans to explore this class of materials and their potential application in dynamic catalysis using computational methods with experimental validation.

Tuning Mott transition: Precision control of the metal-insulator transition (M-IT) of Mott insulators is key to the viability of this class of quantum materials. Often when the critical temperature (T_C) for Mott transition is unfavorable under applicable conditions with materials such as NbO₂ requiring temperatures above 810°C. Nanoscale stability can provide an avenue to tune T_C by controlling the relative stability of the insulating and metallic phases through particle size. It is well known that the polymorphs of materials become more metastable at the nanoscale relative to the ground state due to the greater contributions of surface thermodynamics. With the aid of density functional theory (DFT), my group will explore nanoparticles exhibiting intrinsic M-IT at varying quantities of T_C as a function of particle size. Our goal in this study is to eventually provide a pathway to making this phenomenon more easily tailored in device manufacturing by avoiding extrinsically (defect) induced M-IT in exchange for intrinsic M-IT.

Mott insulators as dynamic catalysts: More recently, Mott insulators have also found application as support materials in electrocatalysts with its "switchability" between electronic phases being applied towards conditional activation of catalytic reactions. It has demonstrated that Mott transition (as a result of extrinsic H-intercalation) in WO₃ catalysts can indeed result in an enhancement of catalytic activity for hydrogen evolution reaction due to a shift in the d-band center. Furthermore, the change in electronic properties were shown to have time scales similar to catalytic turnover. This allows for potential applications as dynamic catalysts which can modify reaction pathways to exceed typical rate and activity limitations. Encouraged by these findings, my group will explore the possibility of Mott transistors as viable dynamic catalysts in order to optimize activity and selectivity and to expose new reaction pathways.

Research experience

My unique research experience as a graduate student and postdoctoral researcher provides me with the domain knowledge required to approach this novel solution to the Sabatier limit. From my past research, I have been exposed to a variety of topics including the study of nanoparticle engineering, general surface properties of crystalline solids, simulations of structural materials for extreme environments including a variety of ceramics and refractory metals, investigation of grain boundary properties, bulk defects, quantum materials for neuromorphic computing, and more recently the application of machine learning in catalysis. In addition, I have also fostered a variety of skill sets such as scientific programming, high-throughput workflow construction, database and API development, website and application management and development, machine learning, and the use of several in-silico techniques for materials research with a specific focus on density functional theory.

As a PhD student under Shyue Ping Ong at UCSD, I explored a wide variety of subjects such as the surface and grain boundary properties of crystalline solids, the effect of atomic doping at material interfaces, the modulation of mechanical properties in ceramic materials, and the properties of Mott insulators for the application of neuromorphic computing. During my time here, I was able to construct fully automated high-throughput workflows capable of calculating the surface energy, work function and detailed Wulff shapes of elemental crystalline solids. This experience has since forged my technique in scientific programming and data management. Doing so exposed me to collaborations with several experimental groups and their methods in the lab.

My post-doctoral work with Zachary Ulissi at Carnegie Mellon University focused on applying machine learning models to discover catalytic materials for a variety of different reactions such as a the chemical reduction of nitrate pollutants, water electrolysis, and the storage and dehydrogenation of liquid organic hydrogen carriers. In particular, I had also worked on developing datasets for the adsorption of simple molecules on unary and binary metal oxide surfaces in order to train machine learning models that can predict the total DFT energy of these systems with a long term focus on applying our models to aid in the discovery of new oxides for OER. Through this project, I had the opportunity to work with machine learning experts from the Facebook AI Research group as well as with experimentalists from the University of Toronto.

Currently my career as a post-doctoral researcher has seen me working under the tutelage of Lar Grabow at the University of Houston. Prof Lars Grabow is one of the few groups that specialize in the computational exploration of dynamic heterogeneous catalysts. Here I am further honing my knowledge in the field of computational catalysis by using the OC22 framework to explore catalysts for static and dynamic water electrolysis. I am also exploring new classes of materials such as single atom catalysts and working closely with the Center for Programmable Energy Catalysis led by Paul Dauenhauer who proposed the concept of dynamic and programmable catalysts.

Teaching Interests

During my career as a graduate student, I was a teaching assistant (TA) for the core undergraduate NanoEngineering course “NANO106: Crystallography of Materials” taught by Professor Shyue Ping Ong. This experience was particularly influential to me, as Prof. Ong had taught that course for several years by that time and always held his TAs to the highest standards. Each week we discussed the construction of homework assignments and we would also work together to develop and grade exam questions which was a sobering experience from the perspective of someone who has only been a student up to that point. The teaching process, including one-hour review sessions each week, was extremely rewarding and was my primary focus for the semester. As a result, I had also made efforts to accommodate ad-hoc discussions with students outside of my usual office hours.

My background is well-suited to the instruction of a range of courses, including both core chemical and materials engineering subjects and ones outside of the curriculum. Among the core courses I would be most interested in teaching are crystallography and thermodynamics, both of which are important aspects to my day-to-day research activity. I would be comfortable teaching at the undergraduate or graduate level. Other core undergraduate courses would also be of interest, including introductory chemical engineering, courses on programming for scientific applications, and project-based junior and senior undergraduate courses. The directions described in my research statement lend themselves to a range of physical examples that could be used in these chemical engineering courses.

In addition to the core chemical engineering courses, I would be interested in developing and teaching several electives. I would like to develop a course on the application of computational simulations for the cross validation and interpretation of experimental results, a key challenge throughout my career as a computational researcher. This course will focus on the applications of density functional theory to assess more physically tangible results such as the use of the Wulff construction in comparing to nanoparticle shapes, the analysis of nanoscale stability, the interpretation of computational results using ab-initio thermodynamics with quantities like pH, applied potentials, temperature, pressure and chemical potential, and the modelling of defects. This course will culminate in a group project to be presented at the end of the quarter where students will select from existing experimental literature lacking computational validation and perform the validation themselves using the knowledge obtained in this class. This course would also make for excellent preparation for students looking to join research groups with a simulation focus, including my own.