A Data-driven Framework to Accelerate the Discovery of Hybrid Cathode Materials for Metal-based Batteries (2410.19204v1)

Abstract: Selecting materials for hybrid cathodes for batteries, which combine intercalation and conversion materials, has gained interest due to their unique synergistic properties, which are not achievable by homogeneous materials. Here, we present a data-driven, chemistry-agnostic, inverse material design framework to discover hybrid cathode materials (HCMs) for metal-based batteries. This framework systematically explores the material space for any working ion, evaluates the stability of candidates, and identifies the growth modes and adsorption of components for a stable hybrid cathode. To demonstrate the framework's application, we conducted a case study aimed at discovering HCMs with an average gravimetric energy density that exceeds the widely used high-energy NMC333 cathode material. The framework identified LiCr$_4$GaS$_8$-Li$_2$S as a promising HCM, achieving an average energy density of 1,424 Wh/kg (on a lithiated cathode basis), surpassing NMC333's theoretical maximum of 1,028 Wh/kg. Additional features of the identified material include: (1) thermodynamically stable lithiated and delithiated intercalation and conversion phases; (2) minimal volume change upon (de)lithiation, which mitigates the conversion material's high volume change; (3) high energy density, compensating for the lower energy density of the intercalation material; (4) the intercalation component acting as both a conductive additive and a sulfur immobilizer while contributing to the total energy density; (5) serving as an ideal support for soft sulfur species; and (6) potential improvements in lifespan, self-discharge, mechanical integrity, and capacity fading compared to conventional Li-S batteries. The developed framework was crucial in exploring materials within the vast potential HCM space with predefined battery material design objectives.