Fast method for calibrated self-discharge measurement of lithium-ion batteries including temperature effects and comparison to modelling
Nawfal Al‐Zubaidi R‐Smith, Manuel Moertelmaier, Georg Gramse, Manuel Kasper, Mykolas Ragulskis, Albert Groebmeyer, Mark Jurjovec, Ed Brorein, Bob Zollo, Ferry Kienberger
Abstract
The self-discharge rate is an important parameter to assess the quality of lithium-ion batteries (LIBs). This paper presents an accurate, efficient, and comprehensive method for measuring and understanding the self-discharge behaviour of LiB cells, considering factors such as temperature and cell to cell variability, as well as underlying electrochemical mechanisms. A method for precise potentiostatic self-discharge measurement (SDM) is demonstrated that is validated by measuring 21 commercial cylindrical 4.7 Ah cells at a state of charge (SoC) of 30%. The self-discharge current ranges between 3 and 6 μA at 23 °C, with an experimental noise level of 0.25 μA. At higher temperatures of 40 °C the self-discharge current increases to 97 μA. The temperature coefficient of voltage (TCV) is experimentally obtained by exposing the cells to a temperature profile with positive and negative step polarities and following the open circuit voltage (OCV) response. Observed TCVs range from +180 µV/K at 40% SoC to −320 µV/K at 0% SoC. For SDM temperature experiments, the cells were set to an SoC with a minimum TCV. From the SDM currents at different temperatures the Arrhenius kinetics and the electrochemical activation energy barrier is determined as 0.94 ± 0.14 eV, indicating chemical side reactions as source of self-discharge. For SDM modelling the electrochemical processes are coupled with a 3D temperature finite element model (FEM) and an electric circuit model resulting in a good overlap with the dynamics and time-constants of the experimental self-discharge curves. The primary challenges addressed are accurately measuring microampere (µA) discharge currents of high-quality cells, reducing measurement time, understanding the temperature dependence of self-discharge, determining activation energy, and demonstrating the applicability and generalization of SDM.