Fig. 5.

Further testing of the terms in Eq. (6a), (6b). a $\tau ∕L_{\mathrm{E}}$ versus L_E for electrodes with 5% and 10% acetylene black, and so different conductivities (extracted from ref. ¹⁶). This results in different a parameters (slopes) but the same b parameter (intercept), consistent with Eq. (6a), (6b). b $\tau ∕L_{\mathrm{E}}^2$ versus porosity extracted from ref. ¹⁹. The line is a fit to Eq. (9) and yields a value of σ_BL close to the expected value (see panel). c $\tau ∕L_{\mathrm{E}}^2$ versus inverse electrolyte concentration extracted from ref. ¹⁶. The line is a fit to Eq. (10) and yields D_BL close to the expected value (see panel). d $\tau ∕L_{\mathrm{E}}^2$ versus separator thickness (this work). The line is a fit to Eq. (11) and yields σ_BL close to the expected value (see panel). e Characteristic time versus the thickness of a thin active layer (TiO₂) extracted from ref. ¹². The line is a fit to Eq. (12) and yields a diffusion coefficient for Li ions in anatase TiO₂ close to the expected value⁸⁵. f 1/Θ plotted versus the intrinsic volumetric electrode capacity, Q_V, for cohorts I and II showing the scaling predicted by Eq. (13a). All errors in this figure are fitting errors combined with measurement uncertainty