A Computational Study of Lithium-Ion Insertion and Diffusion in the NiTi2O2(PO4)2 Anode Material for Lithium-Ion Batteries

Abstract

Polyanionic compounds currently attract considerable interest as electrode materials for Li and Na-ion batteries, owing to their apparent gain in electrochemical performance. Titanium oxyphosphate MTi2O2(PO4)2 (M = Ni, Fe, Co) compounds have been targeted as potential anode materials since NiTi2O2(PO4)2 was identified by Belharouak and Amine as a polyanionic host material for Li-ion battery electrodes [1]. Following this pioneering work, other groups have reported a number of structurally related compounds. Key issues regarding the electrochemical mechanisms and structural changes upon Li-ions insertion have been investigated through a combination of several experimental techniques. The main concern is that titanium oxyphosphate materials lose crystallinity and become amorphous during lithiation [2-7] with a typical mixed insertion-conversion mechanism during the first discharge. In this context, high-energy spectroscopic techniques have enabled a more careful analysis of the changes in composition and the oxidation states of the materials components at different intercalation steps [8]. However, despite the many efforts, the results found on MTi2O2(PO4)2 in literature are still incomplete and contradictory, and a complete picture of the complexity of the electrode reactions is missing. For instance, lithium ions insertion sites, lithium ions diffusion, and migration pathways are not yet known. Obtaining such insights for complex polyhedral structures from experiments remains a challenge. To date, most of the research on this subject has been devoted to the experimental analysis, and to our knowledge no theoretical studies have been reported. The present study uses state of the art computational tools based on density functional theory (DFT) in order to gain valuable insights into crystal structure, Li-ions insertion sites, phase stability upon Li-ions insertion and diffusion pathways and energetics in NiTi2O2(PO4)2, which are crucial for a complete understanding of the electrochemical behavior of this anode material. In order to probe the validity of our computational approach, structural optimization of NiTi2O2(PO4)2 was performed based on the bulk crystal structure observed experimentally [9]. The crystal structure provides three-dimensionally connected empty sites for intercalation of guest ions, and NiTi2O2(PO4)2 can thus be described as [Ni]2a [□]2d [□]2b [□]2c [Ti2 ]4e[O10 ]4e[P2 ]4e . Several lithiation scenarios are investigated in our calculations based on the model of pristine NiTi2O2(PO4)2. In these calculations, the evolution of pressure, volume, lattice parameters and local structure changes were identified upon the insertion of 1 to 6 lithium ions, using all possible combinations of occupation. Upon Li insertion, the pressure in the cell increases and can vary by an order of magnitude depending on the number of Li ions and the site occupation combination. Allowing a volume change of the cell relieves the pressure. We report on these pressure/volume changes for different Li concentrations and site occupations and put them into context of the electrochemical performance during battery cycling. We also report on diffusion barriers and pathways between high symmetry sites for different Li concentrations.