Molecule Design for Non-Aqueous Wide-Temperature Electrolytes via the Intelligentized Screening Method

Area of Science:

• Materials Science
• Electrochemistry
• Artificial Intelligence

Background:

• Wide-temperature electrolytes (WTEs) are critical for stable lithium-ion battery (LIB) performance under varying thermal conditions.
• Current electrolytes often specialize for either low or high temperatures, complicating WTE design due to diverse property requirements.

Purpose of the Study:

• To develop an AI-assisted workflow for designing WTEs with a broad operating temperature range.
• To identify novel electrolyte components and formulations for enhanced battery durability.

Main Methods:

• An AI-driven workflow involving stepwise parameterization and calculations was employed for WTE design.
• Linear mono-nitriles were investigated as potential solvents for their wide liquidus range and weak lithium-ion solvation properties.
• Explainable AI modules were used to analyze structure-property relationships, identifying similarities between cyanide and fluorine.

Main Results:

• Linear mono-nitriles were identified as effective solvents for WTEs.
• 3-methoxypropionitrile (MPN) was determined as a key solvent, enabling battery operation from -60°C to 120°C.
• A LiCoO2/Li cell with the developed WTE demonstrated 72.3% capacity retention after 50 cycles at 100°C.

Conclusions:

• The AI-assisted workflow successfully designed a novel WTE for wide-temperature LIB applications.
• MPN-based electrolytes show significant promise for enhancing battery stability and reliability across extreme temperatures.
• This approach facilitates the rational design of advanced electrolytes for demanding energy storage applications.