Machine-Learned Cathode Material Screening for Stable Sodium-Ion Batteries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional sodium-ion battery cathode materials face limitations in energy density and structural stability, necessitating the development of high-performance alternatives.
Innovation Solution
A method for designing cathode materials using machine learning techniques, including generating feature information and determining suitability through multiple models, such as graph neural networks, to identify materials with desirable properties like high energy density and stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If conventional lithium-ion battery materials are used, then high energy density is achieved, but resource scarcity and high cost occur
Solution Approach 1:
The patent transitions from lithium-ion to sodium-ion battery systems by changing the chemical composition parameter, specifically replacing lithium with sodium in the cathode material. This parameter change allows achieving similar energy density performance while using abundant and cost-effective sodium resources, directly resolving the contradiction between energy density and resource availability.
2Ease of manufacture
If sodium-ion battery cathode materials are used, then cost is reduced, but volume change and low structural stability occur
Solution Approach 1:
The patent employs composite material design by combining sodium superionic conductor (NASICON) framework with specific transition metal oxides (Fe3O4, Mn3O4, Co3O4). This composite structure provides both the cost advantage of sodium-ion technology and enhanced structural stability through the robust NASICON framework that accommodates volume changes during charge-discharge cycles, resolving the contradiction between cost and structural stability.
3Productivity
If machine learning models are used to design cathode materials, then material selection efficiency is improved, but computational complexity increases
Solution Approach 1:
The patent segments the material design process into distinct computational stages: first using graph neural networks for initial material screening and feature extraction, then applying Bayesian optimization for fine-tuning specific material compositions. This segmentation of the machine learning workflow reduces overall computational complexity by dividing the problem into manageable sub-tasks, each optimized for specific objectives, while maintaining high material selection efficiency.
Data Source
AI summary
A method for designing a cathode material includes determining whether a first electrode material corresponds to a first cathode material candidate, based on a cathode material candidate filtering model, and determining whether the first electrode material corresponds to a second cathode material candidate, when the first electrode material corresponds to the first cathode material candidate. The determining of whether the first electrode material corresponds to the first cathode material candidate includes generating cathode material feature information of the first electrode material from material feature information of the first electrode material, and determining whether the first electrode material corresponds to the first cathode material candidate, based on the cathode material feature information of the first electrode material. The material feature information includes chemical descriptor information and material characteristic information of an electrode material. The cathode material feature information includes composability information and cathode material core property information.