Tailor-Made Design of Three-Dimensional Batteries Using a Simple, Accurate Geometry Optimization Scheme

• Published in: ACS Physical Chemistry Au Vol. 4; no. 5; pp. 546 - 554
• Main Author: Miyamoto, Kaito
• Format: Journal Article
• Language: English
• Published: United States American Chemical Society 25.09.2024
• Subjects: multiobjective optimization; lithium-ion batteries; optimization; 3D microbatteries; electrode design; machine learning; regression models
• ISSN: 2694-2445; 2694-2445
• DOI: 10.1021/acsphyschemau.4c00039

In the rapidly evolving Internet of Things (IoT) society, the demand for microbatteries with high areal energy density is surging. As a promising strategy to enhance areal energy density, three-dimensional (3D) batteries have attracted attention. The feature of 3D batteries is the decoupling of the electrode thickness from the ion-transport distance through the modification of the spatial arrangement of the positive and negative electrodes beyond the conventional parallel plates configuration. This allows for the accommodation of a larger amount of active materials without increasing internal resistance. However, identifying the optimal 3D geometry is a complex task, as it depends on printable materials, the resolution of the fabrication equipment, as well as battery usage, which constitutes a multiobjective optimization problem. To overcome this challenge, we propose a novel approach to determine the optimal 3D microbattery geometry. Our innovative method involves a 3D battery optimization system, which integrates an automatic geometry generator with a quick and accurate performance simulator. This approach allows, for the first time, the discovery of material- and discharge-current-dependent optimal geometries. We successfully apply this optimization scheme to two standard electrode pairs (LiFePO4/Li4Ti5O12 and LiNi0.5Mn0.3Co0.2O2/graphite), demonstrating a significant increase in energy density (30%–40% greater than the current state-of-the-art geometry), particularly under high current conditions. These findings underscore the importance of tailor-made batteries for diverse IoT applications and showcase the potential of our approach in realizing such designs.