Battery Material Comparisons Should Refocus on Diffusivity with Best Practices

Abstract

The continuous demand for improved batteries motivates the discovery and advancement of materials with improved transport. Ionic diffusivity is the relevant material property where its measurement depends on accurate assessment of the active material length-scale, generally from the mass-specific surface area. In this perspective, we argue for renewed focus on diffusivity comparisons. A procedural review of 303 recent open-access publications about battery material development revealed two aspects: (1) 49% of publications support structure-property transport claims using diffusivity values and (2) of those reporting diffusivity values, 15% clearly stated that the length scale was measured after grinding-alone or stated that grinding was not used at all. Diffusivity assessment rationally requires length scale measurement after grinding (grind-measure), rather than the reverse. A range of measurement methods are compared, including SEM, BET, and SAXS as well as the resulting apparent diffusivities. Common errors and pitfalls of each of these approaches are described. As an example, datasets are presented for TiNb2O7 (TNO1) and Ti2Nb10O29 (TNO2) made from sol-gel (SOL) and solid state (SS) techniques using rigorous quantitative measurements to separately compare material diffusivities and galvanostatic performance. Here, the SOL samples had shorter length-scales and lower diffusion coefficients. Galvanostatic cell measurements, however, revealed that the shorter length-scales more than compensated for the lower diffusivities with better overall high-rate capacity retention. This example shows how cell level metrics often differ from underlying diffusivities. We argue that materials development needs renewed focus on property measurements like diffusivity where best-practices are important to derive meaningful insights towards structure-property relationships.