DELFIN
Digital Electrode Design for Solid-State Batteries
All-solid-state batteries (ASSBs) are considered promising candidates for the next generation of batteries in the automotive and aerospace industries due to their improved energy density and safety compared to conventional batteries. The 3D structure of ASSB electrodes significantly affects their performance.
The main goal of our collaborative project is to develop a validated software tool for digital design of ASSB materials to understand the relationships between microstructure and properties and provide recommendations for electrode design. We use stochastic microstructure modeling and machine learning to generate "digital twins" of ASSB electrodes as input to performance simulations and simplify the design process.
This reduces the need to fabricate and test numerous physical electrodes, which saves costs and shortens time in experimental material design. This accelerates the optimization of ASSBs and contributes to cost reduction. The combination of 3D stochastic microstructure modeling and performance analysis has the potential to advance material design for ASSBs in Germany, especially in improving the electrochemical properties of ASSB electrodes to support the decarbonization of the energy sector. In addition, the developed software platform will contribute to the ongoing digitization of materials science.
The development of the project is divided into different work packages (WP) as described below:
- WP1: Fabrication and electrochemical characterization of ASSB electrodes with different microstructures.
- WP2: Reconstruct microstructures using high-resolution 3D imaging to understand their effects on electrochemical performance.
- WP3: Enhance the database with virtual, yet realistic microstructures using image processing, machine learning, and stochastic geometry.
- WP4: Develop models to simulate electrochemical performance and apply them to virtual microstructures to establish microstructure-property relationships and design recommendations.
- WP5: Perform comprehensive validations involving electrochemical simulations of the structures from WP3 using the model from WP4 and comparing the results with experimental characterizations from WP1.
The Math2Market subproject focuses on GeoDict workflows, including segmentation of ASSB electrodes, calculation of parameters, creation of statistical digital twins, and scripting for stochastic microstructure modeling in WP3. A major task is to implement simulation and mathematical models for ASSBs in WP4, determine relevant parameter values for the user interface using real ASSBs, and perform comprehensive validations in WP5.
Our project partners:
Universität Ulm - Institute of Stochastics
Justus Liebig Universität Gießen - Center for Materials Research
Start: October 2023
End: September 2026
Latest developments
Project meeting at the Justus-Liebig University in Giessen
On March 18 2024, the members of the publicly funded project DELFIN met at Justus Liebig University Giessen.
Kick-Off meeting for the project DELFIN
The first meeting of all partners of the DELFIN project took place in Kaiserslautern. During this meeting, the project partners provided updates to each other and defined the next steps in the project.
DELFIN project start
The DELFIN project has officially begun and aims to revolutionize the design of battery materials through advanced simulation methods. We look forward to working with our partners on this exciting three-and-a-half-year journey.
