Defect physics, delithiation mechanism, and electronic and ionic conduction in layered lithium manganese oxide cathode materials

Abstract

Layered LiMnO$_2$ and Li$_2$MnO$_3$ are of great interest for lithium-ion battery cathodes because of their high theoretical capacities. The practical application of these materials is, however, limited due to poor electrochemical performance. We herein report a comprehensive first-principles study of defect physics in LiMnO$_2$ and Li$_2$MnO$_3$ using hybrid-density functional calculations. We find that manganese antisites have low formation energies in LiMnO$_2$ and may act as nucleation sites for the formation of impurity phases. The antisites can also occur with high concentrations in Li$_2$MnO$_3$; however, unlike in LiMnO$_2$, they can be eliminated by tuning the experimental conditions during preparation. Other intrinsic point defects may also occur and have an impact on the materials' properties and functioning. An analysis of the formation of lithium vacancies indicates that lithium extraction from LiMnO$_2$ is associated with oxidation at the manganese site, resulting in the formation of manganese small hole polarons; whereas in Li$_2$MnO$_3$ the intrinsic delithiation mechanism involves oxidation at the oxygen site, leading to the formation of bound oxygen hole polarons $η_{\rm O}^{+}$. The layered oxides are found to have no or negligible bandlike carriers and they cannot be doped n- or p-type. The electronic conduction proceeds through hopping of hole and/or electron polarons; the ionic conduction occurs through lithium monovacancy and/or divacancy migration mechanisms. Since $η_{\rm O}^{+}$ is not stable in the absence of negatively charged lithium vacancies in bulk Li$_2$MnO$_3$, the electronic conduction near the start of delithiation is likely to be poor. We suggest that the electronic conduction associated with $η_{\rm O}^{+}$ and, hence, the electrochemical performance of Li$_2$MnO$_3$ can be improved through nanostructuring and/or ion substitution.