Detailed Approach to Calculate the Heat Generation of Lithium-Ion Cells during Heat-Wait-Seek Tests in Accelerating Rate Calorimetry

Abstract

By means of the Heat-Wait-Seek test in Accelerating Rate Calorimeters (ARC) the safety performance of Lithium-ion cells can be estimated. The most interesting measured parameters that can be extracted from these tests are the critical temperatures, such as the onset of self-heating, the venting or thermal runaway temperature. Another essential safety factor is the generated heat during cell failure. In general, the generated heat is calculated by an average heat capacity of the cell and the temperature difference in the experiment, which is not accurate enough especially when facing the challenges in specific volumetric energy density and safety of battery packs.In this study a more accurate way to calculate the generated heat is proposed, which consists of using temperature dependent heat capacity of the cell, based on its individual components. This leads to a significant difference in the calculated generated heat, for instance the specific heat capacity of NMC as a cathode material increases in the temperature range from 298 K to 398 K from 0.83 to 0.96 J/g·K, which is an increase of about 16 %. This leads to a difference in the calculated generated heat in this small range of 7 J/g (or 8 %), if a static specific heat capacity of 0.83 J/g·K and on the other hand a temperature dependent specific heat capacity is used. Moreover, the influence of additional heating steps, venting of the cell, melting of the separator and the overall accuracy of the measurement will be discussed. The investigation of these effects is in the focus of the project AnaLiBa (Analytics of Lithium-ion-Batteries), funded by the German Federal Ministry of Education and Research (BMBF) in the framework of the competence cluster Analytics/quality assurance (AQua). In this project the approach is applied to commercial type 21700 cells. The 21700 type cells are extensively used in battery packs, for instance in electric cars, due to their 20 % higher energy density with respect to the former used type 18650 cells. The cells were measured in an ES-ARC from Thermal Hazard Technologies, UK, using the Heat-Wait-Seek test. The generated heat was compared for either fresh cells and cells after cyclic aging. The cyclic aging was performed at 0 °C and 20 °C with a charging rate of 0.3C for 0 °C and 0.7C for 20 °C and a discharging rate of 2C for 20 °C and 0.5C for 0 °C until the state of health was at 80 % ±2 %. The two different temperatures studied in the cyclic aging were chosen as 0 °C to cause lithium plating and as 20 °C to cause a higher internal resistance due to a thicker solid-electrolyte-interface, which has a high influence on the measured properties in the Heat-Wait-Seek test and therefore on the generated heat.With this approach the generated heat in abuse tests can be calculated more accurately to simulate the propagation of heat during a single cell failure in a pack or to calculate the necessary thickness of a heat barrier for safer battery packs. This helps manufacturers to improve their battery pack design with respect to the specific volumetric energy density and safety.