Goldsmith Lab

RESEARCH GOALS AND HIGHLIGHTS

The Goldsmith Lab is dedicated to training the next generation of researchers to solve major societal challenges pertaining to energy and the environment. In this lab, we welcome students of all ages, backgrounds, beliefs, ethnicities, gender identities and expressions, national origins, physical abilities, religious affiliations, sexual orientations, and other visible and nonvisible differences. Our lab has expertise in first-principles computational modeling and data science to understand and predict catalysts and materials for use in sustainable chemical production, energy storage, and pollution reduction. Our research is supported by external grants from the National Science Foundation, Department of Energy, Army Research Laboratory and American Chemical Society, as well as internal support through the University of Michigan. See the summaries below for more information about our current research thrusts and research highlights.

Catalysts and materials for a sustainable future

Click the poster for more infomation about our research group's thrusts, namely: (i) Thermo- and electro-catalysis for air and water pollution remediation; (ii) Machine learning to accelerate catalyst design and discovery; and (iii) Catalysis and materials for renewable energy generation, storage, and use. The poster highlights recent and ongoing research in our lab, as well as tools used and developed by our group. This work has been made possible by the efforts of many graduate, undergraduate, and postdoctoral scholars, as well as collaborators and agencies such as the NSF, DOE, ACS, and ARL.

Machine learning for heterogeneous catalysis

The Goldsmith lab is using machine learning to: (1) accelerate catalyst screening; (2) find descriptors and patterns in catalysis data; (3) create interatomic potentials for catalyst simulation; and (4) discover and analyze catalytic mechanisms. Ultimately, the power of machine learning has just begun to be exploited in heterogeneous catalysis research, with much room remaining for advancement. A perspective we wrote on this topic can be found here. Recent collaborative works with Prof. Linic's lab showed how interpretable machine learning can be used to understand chemisorption on alloys (read here and here).

Electrocatalytic reduction of wastewater pollutants

With the decreasing price of renewable electricity, electrocatalysis is becoming a promising approach to sustainably remediate pollutants in wastewater. One major wastewater pollutant is nitrate (NO₃⁻). Nitrate pollution of aquatic ecosystems and drinking water has become a serious problem around the world and poses a threat to human health and the environment. To help address this challenge, we are studying the reaction mechanism for nitrate reduction on a variety of electrocatalysts, with the goal to engender the design of improved nitrate reduction electrocatalysts. Our computational predictions are validated and refined through experimental collaborations. Read here, here, and here.

Single atom and nanocluster catalysis

Single atoms and nanoclusters can show dramatically different activity, selectivities, and stabilities for a variety of important reactions, e.g., the reverse water gas shift reaction, C-H activation, and CO oxidation. In collaboration with General Motors and Tianjin University, we recently reported the use of isolated Pt₁ atoms on ceria as “seeds” to develop a Pt-O-Pt ensemble, which is well-represented by a Pt₈O₁₄ model cluster that retains 100% metal dispersion. The Pt atom in the ensemble is 100–1000 times more active than their single-atom Pt₁/CeO₂ parent in catalyzing the low-temperature CO oxidation under oxygen-rich conditions. Read article here. Our lab has also recently published related articles in this thematic area (Read here and here).

Redox flow batteries for energy storage

Understanding the interfacial chemical physics and reactions on redox electrocatalysts will directly impact energy storage applications, enabling lower cost, higher energy/power density systems. A molecular level understanding of electrochemistry can be enabled by carefully combining controlled first-principles modeling with kinetic measurements and spectroscopy. We are working to understand the molecular structure of transiton metal ions in aqeuous-phase redox flow batteries and their electrocatalytic redox mechanism. The Goldsmith lab leads the computational modeling efforts of this collaborative project with Prof. Nirala Singh's group. Recently published works on the V²⁺/V³⁺ redox couple can be found here and here, and here and for the Ce³⁺/Ce⁴⁺ couple here.

Conversion of bio-oil to fuels and chemicals

Our world's addiction to fossil fuels for transportation is unsustainable and emits huge amounts of greenhouse gases like CO2. The need to supply energy for transportation in a sustainable manner has motivated ongoing efforts to replace fossil fuels with renewable and CO2-neutral transportation fuels. One promising strategy is to produce transportation-grade fuels from biomass waste (e.g., bio-oil consisting of oxygenated aromatic lignin byproducts) through electrocatalytic hydrogenation (ECH) using renewable electricity, with an accompanying oxidation reaction such as water oxidation. Our work aims to understand the electrocatalytic conversion of bio-oils at an atomistic level to help design improved catalysts (e.g., read here).