SOMAS: a platform for data-driven material discovery in redox flow battery development.

Area of Science:

• Chemistry
• Materials Science
• Energy Storage

Background:

• Aqueous organic redox flow batteries are promising for grid-scale energy storage due to their safety and tunability.
• High energy density in these batteries relies on the aqueous solubility of redox-active organic molecules.
• Predicting molecular solubility in water is a significant challenge in chemistry.

Purpose of the Study:

• To create a comprehensive, open-access database of organic molecules and their aqueous solubility.
• To provide essential data for developing accurate machine learning models for solubility prediction.
• To support advancements in aqueous organic redox flow battery technology.

Main Methods:

• Compiled a database (SOMAS) of approximately 12,000 organic molecules.
• Included experimental solubility data for each molecule.
• Generated quantum and molecular descriptors using high-throughput density functional theory calculations.

Main Results:

• Established the "Solubility of Organic Molecules in Aqueous Solution" (SOMAS) database.
• SOMAS contains diverse molecules covering wide chemical and solubility ranges.
• The database includes experimental solubility, optimized geometries, and 14 molecular descriptors per molecule.

Conclusions:

• The SOMAS database provides a critical foundation for AI-driven solubility prediction.
• This resource will accelerate the development of effective machine learning models for organic redox flow batteries.
• Facilitates improved design of molecules for efficient large-scale aqueous energy storage.