There is overwhelming demand for the design of new high performance cathode materials for the next generation of rechargeable batteries for both Li and Na ion technology. The olivine structured orthophosphates Li/NaMPO4 (M = Fe, Mn, Ni and Co) have garnered significant interest due to their good thermal stability and high voltage vs Li+/Li / Na+/Na couple. This interest manifested in the relative recent commercialization of LiFePO4. LiMnPO4 remains a promising alternative candidate to LiFePO4 due to its higher operational voltage (~4.1 V vs Li+/Li), however it has been shown to be less electrochemically active than its iron counterpart, as it specifically exhibits lower ionic diffusion due to its Jahn-Teller activity. Investigating and understanding ionic diffusion at an atomic level is therefore a priority for battery materials innovation. Computational methods continue to provide insight on an atomic level and Molecular Dynamics (MD) simulation remains one of the most powerful tools. For a typical diffusive process, there are numerous energy barriers due to the highly corrugated potential energy surface. Therefore a standard MD simulation will frequently visit probable configurations whilst reactive states of Li/Na particle hopping will be less or not represented. Instead of just increasing the overall temperature of the system, our approach selectively ‘warm up’ the Li/Na ions by transferring a variable amount of kinetic energy from the slower frequencies of framework dynamics (MPO4) to the lighter and more mobile Li/Na ions. The justification of this novel method is based upon the fact that upon activation of Li/Na mobility the Li/Na sublattice behaves resembles a liquid hosted in a solid framework (MPO4), this implies a natural separation of velocities and frequencies between the Li/Na sublattice and framework. Using the ‘Shooter’ algorithm it is possible to simulate and visualize diffusion on a suitable timeframe and analyze the mechanism and dimensionality of the diffusive process. Based on a detailed elucidation of mass transport mechanisms, diffusion coefficients, and subsequently free energies and reaction rates cab be assessed for each cathode candidate material. This step is propaedeutic to a deep understanding of the coupling of electrochemical redox processes with mass and ion motion within materials.

The Leoni group focuses on the study of activated processes in the solid state by means of advanced computational tools. Structural and electronic phase transitions, chemical reactions, formation mechanisms, reactive intermediates, structure prediction and the rules behind polymorphism in general are relevant research areas. Understanding processes like crystallization, nucleation and growth, diffusion of impurities or defects, or electrochemical reactions are crucial factors for the development of better materials. Despite major advances in device resolution, experiments can only provide a coarse-grained view of such processes. Theory can now integrate the experimental data by implementing the missing length and time resolution, thanks to novel strategies of numerical simulations. At the interface of inorganic and material sciences, theoretical chemistry, computational chemistry and  physics, physical chemistry, materials for energy and sustainability, this area offers fascinating opportunities to leverage computational tools in the design of innovative materials.