Experimental Characterization and Modeling of the Aging Behavior of the Lithium-ion Batteries Considering Kinetic-diffusion Limitation and Graphite-silicon Blended Anode

Abstract

Fundamental understanding of the electrochemical, thermal and mechanical behaviors of the lithium ion batteries during aging is crucial for securing lifespan and safety by the cost-effective and efficient design of cell and the system thereof. The emergence of silicon (Si) and graphite-silicon (C-Si) blended anodes impact cell-level performance which adds the difficulty in experimental and modeling characterization. In this work, we have firstly developed a highly accurate and dynamic isothermal calorimeter that enables the measurement of heat generation rate (HGR) of the cylindrical and pouch type cell. The electrical and thermal behaviors as a function of the aging cycle are recorded for different temperature conditions. Then, a reduced-order electrochemical-thermal model (ETM) platform is developed, which is highlighted by three microcells considering the geometry of a cylindrical cell. The numbers of microcells and numerical mesh elements are optimized with respect to the simulation accuracy and computational speed. During operations, it can be observed that a temperature gradient arises in the radial direction, resulting in a decrease in local resistance and an increase in reaction rate at the high-temperature core location. To model the electrochemical and thermal behaviors in a C-Si blended anode, a mechanical stress-driven composite anode model implemented on the ETM platform is developed, which considers hydrostatic diffusion-induced stress on particle level and biaxial stress on electrode level. The competing lithiation and delithiation mechanisms between C and Si particles are described by the Butler-Volmer equation, driven by the stress. The effects of particle-level mechanical stress on the electrode potential, hysteresis and cell-level HGR are further justified. The results show that the stress within anode particles is dependent upon not only Li+ concentration but also concentration gradient. The hydrostatic stress within silicon particles is notably larger than graphite, which drives a silicon-dominated reaction in low-SOC range, and consequently causes a voltage hysteresis and a HGR peak majorly at low SOC. Based on the established ETM, two major challenges in characterizing battery aging mechanisms are addressed. The first one is the kinetic-diffusion limitation of the solid electrolyte interphase (SEI) generation. A physics-based methodology is proposed considering a two-stage process, with a Piecewise Kinetic-Diffusion (PKD) control mechanism of the SEI formation in the electrochemical degradation model. The kinetic and diffusion limits are separately determined by calculating the molar fluxes of Li+ and ethylene carbonate (EC) solvent as two reactant species for SEI, which are compared at the reaction interphase to identify the limiting mechanism. The simulation results are validated with both calendar and cycle life data, under different SOCs, temperatures, and charging profiles. The PKD method more accurately captures the temperature and SOC dependency of capacity and voltage fade, as compared to the conventional methods. The second challenge is the degradation mechanism of the C-Si two-particle anode system. In this work, the stress-induced overpotential (SIO) is considered to be the factor that differentiates the electrochemical degradation rate between C and Si particles, which leads to a faster aging of the Si particles due to its mechanical properties and high utilization during lithiation. The proposed model is validated against the experimental aging data, which provide detailed analysis on the individual contribution of C and Si component on the overall cell-level aging and thermal behavior. The experimentally validated modeling analysis is dedicated to develop a deeper physical understanding of reaction kinetics, mass and charge transport within the battery. This work may provide guidelines for the development of battery management system, cooling circuit, and the design of electrode materials and fast-charging algorithms, all of which are closely linked to the electrical, thermal, and mechanical behavior of batteries over their lifespan.