Data and physics driven atomic modeling of electrochemical interfaces

VISION

Formulate innovative design principles to overcome parasitic side reactions and overpotential loses to develop scalable, efficient, and selective electrochemical technologies.

Utilize novel workflows, incorporating graph-theory based data mining and machine learning algorithms along with advanced atomistic simulation techniques, to discover governing interactions and prevalent reactions at the electrode-electrolyte interface.

Thrust 1: Data driven methods for atomic modeling

Elucidate experimentally relevant atomic configurations

  • Effectively navigate large phase space of possible configurations.

  • Develop efficient workflows:

    • Systematic data sampling

    • Estimate important interactions

    • Model range of materials

Thrust 2: Capturing dynamics at electrocatalytic interfaces

Identify novel electrocatalysts for carbon-neutral reactants

  • Benchmark subset of atomic configurations to experimental characterization data.

  • Identify descriptors to inverse design and experimentally probe promising electrocatalyst.

Thrust 3: Atomic modeling at battery electrodes

Identify solvents and electrodes for metal-chalcogen batteries

  • Algorithmic approaches to identify selectivity and activity descriptors.

  • Generate and gather descriptor-based information for materials.

  • Pareto analysis to identify promising materials