Atomic-Scale Insights into the Structural Transformations in Cathodes for Multivalent Metal-Ion Batteries

Multivalent-metal ion batteries are promising and safer alternatives to lithium-ion batteries for next-generation energy storage devices. Most of the multivalent-metal ions, such as calcium (Ca) and magnesium (Mg), and their respective cathode material systems can be sourced from earth-abundant minerals. Moreover, Ca and Mg-ion batteries, based on high-voltage oxide cathodes show high energy density compared to lithium-ion batteries. This makes multivalent-metal ion batteries a cost-effective proposition for a variety of battery technologies. However, multivalent-metal ion batteries often suffer from poor electrochemical stability. A key challenge in achieving higher performance and stability in these material systems is to understand the fundamental structural changes to the cathode at the atomic scale. Such structural changes during electrochemical cycling include cation disorder, planar defects, phase separation, and (de)intercalation mechanisms. In this work, we combine aberration-corrected scanning transmission electron microscopy (STEM), first-principles density functional theory (DFT) calculations, and STEM simulations to develop a fundamental understanding of the structural changes occurring during electrochemical cycling of Ca and Mg-ions into FePO₄ and α-V₂O₅ as cathodes respectively.<br/><br/>During electrochemical cycling of Ca into FePO₄, we observe that the large particles (&gt; 100 nm) show Ca intercalation at the surface with the bulk of the particle showing little to no activity. Moreover, the smaller particles ( &lt; 50 nm) exhibit a tendency to phase-separate into CaPO₄. STEM imaging reveals the direct atomic-scale evidence of Ca intercalation into the FePO₄ olivine structure, where Ca-ions show a preferential arrangement along the [010] orientation. For the Mg-ion batteries, we observe that the Mg intercalation into α-V₂O₅ is heterogeneous and is accompanied by the formation of a network of low-angle grain boundaries. We hypothesize that there is significant phase separation between various phases of Mg_xV₂O₅ and V₂O₅, such as the fully intercalated δ-MgV₂O₅ and fully deintercalated α-V₂O₅ phases. Our atomic-scale findings for these two model systems can be leveraged by optimizing the materials synthesis to achieve theoretical operating potentials and performance.<br/><br/><b>Acknowledgments</b><b>:</b> This work is supported by the Joint Center for Energy Storage Research (JCESR) and Energy Innovation Hub funded by the U.S. Department of Energy (DOE), Office of Science, Basic Energy Sciences.