Bonan Zhu

Beijing Institute of Technology, Beijing, 100081, China 

University College London, London,WClE 6BT,UK 

EXTENDED ABSTRACT: Developing large-scale energy storage is essential for achieving the carbon peak and carbon neutrality targets. Li-ion batteries has been widely used and is currently driving the transition towards electric vehicles in the transport sector. The discovery of a series of key cathode materials has been pivotal for the development of Li-ion batteries, and cathode material is still the determining factor for performance and cost even today. Hence, finding new high-capacity, low-cost cathode materials is essential for scaling up grid-storage applications. High-throughput calculations based on elemental substitution is a widely used approach. However, the scarcity of experimental structural data limits its effectiveness in chemical spaces that are currently underexplored. On the other hand, first-principles structure prediction does not rely on any pre-existing data. Our trials have shown that ab initio random structure searching (AIRSS) can be used to obtain ground state structures of LiCo02 and LiFeP04, two commercial cathode materials, with high efficiency. It was found that using constraints such as rigid polyanionic units, symmetry operations, and chemically motivated specie-wise minimum separations are essential for efficient searching. In the search for LiFeS04F, a new polymorph was found in addition to the known tavorite and triplite structures. Further calculations suggested that this new phase combines the high mobility and high voltage features of the known phases. Oxalate-based cathodes have been reported to exhibit anion redox. We found new polymorphs of LiFe(C204)2 as well as Li2FeC204F2, which have not been reported previously. Interestingly, not all low­energy structures exhibit anionic redox, making these theoretical structures ideal platforms for studying factors controlling anionic redox. Lastly, a multi-composition search was performed to look for oxysulphides in the Li-Fe-S-0 chemical spaces. In addition to the previously known compound Li2FeSO, many new stable and metastable phases have been found. Based on the obtained structures, calculations of cycling properties suggested that two of the new phases could potentially offer higher energy density than LiFeP04. In summary, we have used AIRSS to search for novel cathode materials for specific compositions or specific chemical spaces and found a variety of new phases. This approach can also be combined with traditional elemental substitution for cross-validation.
Keywords: first-principles; Li-ion battery cathode materials; crystal structure prediction;