Lithium Metal Batteries: AI-Driven Discovery of 368 Non-Fluorinated Diluents

AI타임스 South Korea

Overview

Researchers at Seoul National University developed an AI algorithm that screened 1.32 million molecules to identify 368 promising non-fluorinated electrolyte diluents for lithium metal batteries. Three of these candidates were experimentally verified in pouch cells, demonstrating stable charge-discharge performance. This approach significantly reduces the time and cost of material discovery by replacing trial-and-error synthesis with physics-based machine learning.

In Depth

Key Findings

A research team led by Professor Jong-Woo Lim at Seoul National University has identified 368 promising non-fluorinated electrolyte diluent candidates from a database of approximately 1.32 million molecules. By applying a machine learning algorithm to the PubChem database, the team narrowed the search space to approximately 0.028% of the original pool. To validate these results, the researchers applied three of the selected materials to lithium metal single-layer pouch cells, confirming stable operation during repeated charging and discharging cycles.

Technical Details

The AI algorithm utilizes two specific physical and chemical descriptors to predict the suitability of a molecule as a diluent:

• BF₃ Affinity: Measures the electron-pair donating characteristics to evaluate the molecule’s tendency to bind directly with lithium ions.
• Molecular Polarity Index (MPI): Analyzes charge distribution and electrostatic interactions to determine how well the diluent interacts with other electrolyte components.

Unlike standard AI models that rely solely on statistical patterns, this system uses these chemical criteria to ensure the diluent does not bind strongly to lithium ions while maintaining appropriate interactions with the rest of the electrolyte.

Background and Context

Lithium metal batteries offer higher energy density than lithium-ion batteries but suffer from high chemical reactivity, leading to side reactions that degrade stability and lifespan. Localized High-Concentration Electrolytes (LHCE) address this by adding diluents to high-concentration electrolytes to lower viscosity without disrupting the lithium-ion coordination structure. While fluorinated compounds are commonly used, there is a demand for non-fluorinated alternatives due to high synthesis costs and environmental persistence concerns.

Significance and Outlook

This research demonstrates that physics-grounded AI can drastically compress the experimental timeline for material discovery. While only three candidates have been experimentally verified so far, the study expands the library of viable non-fluorinated alternatives. Future commercialization will require further evaluation of long-term cycle life, stability under various operating conditions, and mass-production feasibility.

]]

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.

• Your email and selected fields are used only to deliver the newsletter.
• We never share your information with third parties.
• You can unsubscribe anytime via the link in each email.

See our Privacy Policy for details.

Agree & Continue ▶

Takes about a minute · Unsubscribe anytime