Research
High-Precision Single-Step Printing of Nanostructures using Crystal-Structure-Directed Growth:
Electrochemical additive manufacturing of materials combined with crystal structure-directed growth mechanisms has the potential to generate precisely engineered surfaces and 3D architectures with nanoscale resolution in a template-free process. This would enable printing of a limitless number of precisely structured functional materials for applications as battery electrodes and current collectors, surfaces with spatially resolved and dynamically controllable wettability, and nanophotonic arrays. Through in situ tuning of the electrochemical deposition parameters, it would additionally allow for high-throughput synthesis of a range of spatially resolved nanomaterials, enabling application of data-driven techniques to electrochemistry. This thrust will establish a blueprint for atom-precise electrochemical manufacturing bridging across length scales.
Mapping Active Site Transformations and Degradation Pathways in CO2 Reduction Electrocatalysts to Decipher Mechanistic Understanding and Enable Robust Catalyst Design:
Generation of spatially resolved maps of electrocatalytic activity within single particles/grains using scanning electrochemical cell microscopy (SECCM) and co-mapping crystallographic orientation using electron backscatter diffraction (EBSD) or transmission electron microscopy (TEM) and electronic structure through scanning transmission X-ray microscopy (STXM) will enable construction of crystal-structure—electronic structure—electrocatalytic activity relationships. Initial studies will investigate the evolution of structure and activity for catalytic sites on polycrystalline Cu electrodes upon extended exposure to CO2 electrocatalytic reduction conditions to gain detailed understanding of site stability and degradation pathways. Separately, spatially resolved mapping of Cu nanocrystals deposited onto TEM/STXM electrodes will enable identification of characteristic spectroscopic signatures for catalytic sites. The distinctive effects of factors influencing bulk performance such as grain boundaries, defects, and facet-specific catalytic sites will be deconvoluted to establish a clear roadmap towards engendering improvements in catalyst designs for the robust electrochemical reduction of CO2 to multi-carbon products.
Accelerating the Exploration of Synthetic Design Spaces through use of Machine Learning:
Exploration of materials landscapes through design of experiment methods coupled with data-driven modeling and sequential learning will accelerate discovery of new materials and elucidate the underlying correlations in synthesis that govern emerging fields of study. As a model system that will serve as the initial platform for discovery, manganese oxides form numerous metastable structures accessible through subtle changes in reaction trajectories which are poorly understood, leaving the literature dominated by ‘guess and check’ synthetic methods. Work will involve generation of training data using design of experiment methods, followed by modeling using machine learning algorithms in an iterative fashion. Our ultimate goal will be to converge on autonomous systems with the ability to formulate, test, and pivot hypotheses and that are capable of implementing physics-based learning.
