Quantifying Battery Rate Performance

Abstract

The global energy demand is quickly increasing and sustainable energy solutions are needed to keep pace. Batteries have emerged as the premier energy storage solution for increasing the adoption of renewable energy sources, and for reducing CO2 emission by replacing polluting technologies. The success of this transition depends on the energy storing capabilities of batteries, requiring continuous improvements to aspects like fast-charging.Addressing the fast-charging limitations of batteries is made more difficult by experimentalists lacking simple tools for analysing rate performance data. In this study, a simple model based on solid- and liquid-phase diffusion as well as electrical and electrochemical effects is used to analyse battery rate performance in terms of common cell parameters. Thicker electrodes are needed to maximise the energy density of batteries, but the reduced fast-charging capability of thick electrodes means that there is a balance to be found between stored energy and charging speed. The above-mentioned mechanistic rate model predicts such a quantitative relationship between electrode thickness and rate performance, serving as the basis for a systematic study between these two parameters. The electrode thickness was varied (10–242 μm) through standard slurry casting methods where segregated network composites of lithium nickel cobalt aluminium oxide (NCA) and carbon nanotubes (CNT) prevented cracking in thicker films. Rate performance data was collected via chronoamperometry (CA) and analysed using a characteristic charging time, τ, which was extracted from specific capacity vs rate curves using a semi-empirical fitting equation. Analysing thickness dependent τ data revealed a quadratic dependence of τ on electrode thickness, a trend that was readily explained and predicted by the mechanistic rate model. Impedance spectroscopy data was also recorded as a function of electrode thickness, showing that the electrochemical and solid-state diffusion contributions of rate performance are in line with the outputs of the iii model. Separators are often neglected battery components, but their properties have an impact on electrochemical performance. In particular, the thickness of the separator can be a key parameter for rate performance if the ion transport properties of the separator are significantly limiting. For these reasons a quantitative study between separator thickness and rate performance was done, where separator thickness was varied (16–144 µm) by stacking multiple separators in one cell. The model system was based on lithium nickel manganese cobalt oxide (NMC) and CNT composites, and rate performance was quantified in the same way as before. Plotting τ as a function of separator thickness showed τ to linearly increase until it saturated at a separator thickness of ∼ 100 µm. The pre-saturation region was analysed using the mechanistic rate model to determine that rate performance is limited by the electrolyte resistance effects rather than the diffusion effects. Increasing the volumetric energy density of batteries is highly topical since it allows electric vehicle manufacturers to achieve higher cruising ranges. However, it was not considered before whether volumetric capacity has an impact on rate performance. The mechanistic rate model of this study predicts such a relation, specifically that higher volumetric capacities result in diminished rate performance. This prediction was investigated by manufacturing CNT composites based on graphite and boron-nitride (BN). Since graphite is a known Li storing active material and BN is not, mixing them in different ratios allowed the volumetric capacity to be varied between 25–280 mAh/cm3 . The rate performance of these systems was measured once again with CA, and fitted with a semi-empirical equation to obtain τ. A linear scaling was observed between τ and volumetric capacity, verifying the negative impact of volumetric capacity on charging time. This anti-correlation was in perfect agreement with the model, and it was attributed to the RC charging times of the electrodes.