Multiscale Modelling of Battery Systems at Low Temperatures: Impact of Microscale on Performance

Abstract

Lithium-ion batteries (LIB) are currently the most widely spread all-purpose energy storage technology. Although improved energy and power densities continue to widen and foster their adoption in multiple application domains, such as electric mobility, LIBs operating temperatures continue to be typically in a range of 15-35 °C. Particularly, at temperatures around and below 0 °C, the performance of LIBs is dramatically reduced, which prohibits deployment of LIBs in cold climates, high latitudes, mountainous regions, or aerospace applications. Low temperatures impact the cycling performance, safety and, most dramatically, decrease the battery capacity. This degradation in performance is a result of several microscale physical mechanisms affecting both electron and Li-ion transport, as well as reaction rates at electrodes and in electrolyte of batteries. In particular, at low temperatures, the charge transfer at the electrode/electrolyte, lithium diffusion in the solid material and Li-ion transport in electrolyte are diminished when compared to room temperature, yielding a reduction in capacity or even battery inoperability in some applications.To improve the low temperature performance, it becomes important to trace down the microscale mechanisms and identify dominant limiting phenomena to address them individually. Despite the fact that vast literature exists for many cathode and anode materials at low temperatures [1-2], identifying the dominant limiting physical mechanisms is still a topic of debate [3]. Different works attribute the low-temperature performance limitations to different aspects of the battery system, such as sluggish transport in the electrolyte [1], slow solid diffusion processes in the active materials [2], charge transfer disruption in electrode or electrolyte composite media or at the solid-electrolyte interface [4], leaving a multitude of possibilities for a particular battery system. Additionally, the complexity of the microscale phenomena in battery systems makes it difficult to trace the root causes from the macro to the microscale, and thus to tackle this problem effectively.In this context, modelling and simulation of the microscale charge and mass transport phenomena is a useful tool in providing insight into the microscale aspects dictating the macroscale operation. Typically, the Doyle-Fuller-Newman (DFN) multiscale model is applied to predict macroscale performance from microscale properties of the battery system [5]. In this model charge and mass conservation equations are solved for the input physical parameters describing the battery system such as geometry and parameters describing the battery materials comprising electronic conductivities of the electrodes, ionic diffusivities and conductivity of the electrolyte and electrochemical reaction rates at the electrode’s active particles. The balancing between these parameters with respect to each other will influence (and limit) the response of the LIB. Each of these parameters is typically considered to follow an Arrhenius relationship with temperature, where an activation energy controls the coefficients for the transport and reaction processes, such that at lower temperatures an exponential decrease in electronic and ionic transport as well as reaction rates leads to strongly reduced performance. However, the particular effects of low temperatures on these coefficients describing the battery microstructure phenomena and its impact on the macroscale performance remain poorly explored.In this paper, we report low temperature effects on microscale parameters and their influence on the macroscale performance with particular examples targeting low temperature effects in LiFePO4 (LFP) and LiNi1/3Mn1/3Co1/3O2 (NMC) based batteries. By implementing a multiscale DFN model in the finite element modeling software COMSOL Multiphysics®, we perform numerical simulations of these battery systems within a range of temperatures and operating conditions and compare them with experimental data. Our results allow to assess the effect of temperature based on microstructural parameters of the system and corresponding macroscale output. Figure 1 exemplifies typical simulation results of the temperature effect on discharge curves for LFP/Graphite (Figure 1a) and NMC/Graphite cells (Figure 1b). These results hold great potential for understanding battery’s microstructure impact on the overall performance at low temperatures and to inform electrode choice, battery design and manufacturing processes for low temperature battery systems. Acknowledgement This research has received funding from the European Union’s Horizon Europe research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 101062008.