Related Experiment Video
Updated: Sep 23, 2025
10:03
Characterization of Electrode Materials for Lithium Ion and Sodium Ion Batteries Using Synchrotron Radiation Techniques
Published on: November 11, 2013
25.6K
Application of DFT-based machine learning for developing molecular electrode materials in Li-ion batteries.
Omar Allam1,2, Byung Woo Cho1,2, Ki Chul Kim1,3
1Computational NanoBio Technology Laboratory, School of Materials Science and Engineering, Georgia Institute of Technology Atlanta GA 30332-0245 USA SeungSoon.Jang@mse.gatech.edu.
RSC Advances
|May 13, 2022
Summary
This study introduces a machine learning framework for designing molecular electrode materials. It accurately predicts redox potentials using key electronic and structural properties, accelerating materials discovery.
Area of Science:
- Computational Chemistry
- Materials Science
- Machine Learning
Background:
- Developing novel molecular electrode materials is crucial for energy storage applications.
- Predicting material properties computationally can accelerate the design process.
Purpose of the Study:
- To develop a high-throughput screening method for designing new molecular electrode materials.
- To utilize a density functional theory-machine learning framework for predicting redox potentials.
Main Methods:
- Density functional theory (DFT) modeling to predict quantum mechanical quantities and electronic properties.
- Machine learning, specifically artificial neural networks, trained on DFT data.
- Linear correlation analysis to reduce input variables to six core features: electron affinity, highest occupied molecular orbital (HOMO), lowest unoccupied molecular orbital (LUMO), HOMO-LUMO gap, and counts of oxygen and lithium atoms.
Main Results:
- The artificial neural network accurately estimates redox potentials for organic materials.
- Electron affinity was identified as the most influential feature for predicting redox potential.
- Other significant features include the number of oxygen atoms, HOMO-LUMO gap, number of lithium atoms, LUMO, and HOMO.
Conclusions:
- The developed DFT-machine learning framework is effective for high-throughput screening of molecular electrode materials.
- The study highlights key descriptors for predicting redox potentials, aiding in rational material design.
- This approach accelerates the discovery of advanced materials for electrochemical applications.
