Data Science Based Redox Molecule Design for Flow Batteries
EMSL Project ID
51219
Abstract
Data science methods using large data sets from experimental and computational sources are proposed to predict organic redox molecules that will greatly improve redox flow battery performance. Specifically, the aim of the proposed research is to establish a data science-based model to predict aqueous solubility of redox active molecules under various pH conditions by using machine learning techniques based on chemical and physical descriptors for complex structure-property relationships. In addition to experimental descriptors such as aqueous solubilities from the literature, quantum chemistry calculated descriptors will be generated from EMSLs NWChem computational chemistry (i.e., quantum chemistry) suite.
Project Details
Start Date
2019-10-31
End Date
2022-09-30
Status
Closed
Released Data Link
Team
Principal Investigator
Team Members