Statistical Physics-Based Model of Solid Electrolyte Interphase Growth in Lithium Ion Batteries (original) (raw)

The article presents a statistical physics-based model for the growth of the solid electrolyte interphase (SEI) in the negative electrode of lithium ion batteries. During battery operation, the SEI thickness grows by the reaction between lithium ions, electrons and solvent species on the surface of active particles at the negative electrode. The growth of the SEI layer causes a loss of lithium ions that induces capacity fade. In addition, it increases the ion transport resistance and decreases the total porosity. Our model employs a population balance formalism based on the Fokker-Planck Equation to describe the propagation of the particle density distribution function in the electrode. Structure-transforming processes at the level of individual particles are accounted for by using a set of kinetic and transport equations. These processes alter the particle density distribution function, and cause changes in battery performance. A parametric study of the model singles out the first moment of the initial SEI thickness distribution as the determining factor in predicting the capacity fade. The model-based treatment of experimental data allows analyzing processes that control SEI growth and extracting their controlling parameters. Lithium ion batteries (LIBs) are highly touted energy storage devices for portable electronics and electric vehicles. 1 The LIB system consists of a negative electrode, a positive electrode, a separator, an electrolyte, and two current collectors. The most commonly used elec-trolytes are comprised of lithium salts, such as LiPF 6 in a solution of ethylene carbonate (EC) and dimethyl carbonate (DMC). 1 The electrodes consist of randomly distributed and interconnected particles of active material, which store and release the lithium ions. Aging and degradation of LIBs have become major concerns for the operation of electric vehicles (EVs), which must fulfill exacting requirements in terms of durability, cyclability and overall lifetime. 2 Battery aging is linked to the (electro-)chemical and mechanical degradation of the electrodes and the electrolyte. 2 Three main degradation mechanisms prevail at the particle level during the cycling of LIBs: (1) growth of the solid electrolyte interphase (SEI) at the negative electrode, 2,3 (2) formation of cracks in the SEI layer at the negative electrode, 4 and (3) dissolution and isolation of nanocrystalline particles at the positive electrode. 5 These processes lead to capacity fade and power loss. At the beginning of the battery life, particularly during the first cycle, the electrolyte undergoes reduction at the electrode/electrolyte interface, because the negative electrode operates at potentials that are outside of the electrochemical stability window of electrolyte components. This reduction is accompanied by the irreversible consumption of lithium ions. 6 This process forms a passivating surface layer known as the solid electrolyte interphase. The SEI layer grows further during charging, thereby reducing the amount of active lithium ions. Moreover , this layer penetrates into the electrode pores, reducing the overall porosity and decreasing ion access to the active surface area of the electrode. 2 Understanding the mechanism of SEI thickness growth and the resulting composition of the layer is a vital topic of LIB research, because of its great practical significance and the intricate interplay of underlying processes. 2,6–14 The growth of the SEI involves a complex reaction network 15,16 that is sensitive to operating conditions as well as the battery cycling protocol. A mechanistic understanding of the processes involved in SEI formation and growth is vital for designing LIB systems with high performance and long cycle life. The topic therefore garners significant interest in the experimental and theoretical research community. Recent efforts in modeling of the electrochemical processes during cycling of LIBs include particle level models 17,18 and porous electrode theory. 16 In 1976, Bennion developed the first model of the SEI for lithium metal electrodes. 19 Peled further derived a parabolic growth law for the SEI considering the electron transfer as the rate-determining step in the growth of the SEI layer. 20 Ploehn et al. developed a model for SEI growth, in which the diffusion of solvent through the SEI is the rate-determining step. 21 Interestingly , this assumption also leads to a parabolic growth law. Newman et al. developed a model to predict the SEI growth based on porous electrode theory, including the detailed chemistry of SEI formation. 16 This model was the first attempt to link chemical reactions leading to SEI formation to irreversible capacity loss. However, it comprises a large number of model parameters. Xie et al. expanded Newman's model to incorporate thermal effects. 22 They found the battery skin temperature to significantly affect the growth of the SEI. Several subsequent works are based on Newman's model. Their common feature is that they couple a one dimensional model of charge and mass transport through the electrode to a one dimensional model of spherical diffusion inside active particles, resulting in a pseudo two-dimensional porous electrode (P2D) model. 23–26 Under low to moderate working conditions, the P2D model can be reduced to a single particle (SP) model. 27,28 In this approach, the electrode is represented by identical spherical particles, whose accumulated surface area is equivalent to the total active area of the solid phase in the porous electrode. This model assumes that the mass and charge transport resistances in solution can be neglected. Moreover, the faradaic current across the porous electrode exhibits a linear profile , which is valid in the limit of small applied currents in thin and highly conductive electrodes. 18 The modeling work presented in this article strives to establish relations among structural properties and electrochemical performance of the battery in order to rationalize the influence of SEI growth upon them. The central component of the statistical physics-based modeling framework is the Fokker-Planck Equation (FPE). The porous battery electrode is treated as a statistical distribution of interconnected particles. The FPE governs the temporal evolution of this distribution. The developed formalism allows leading causes of structural degradation , driven by (electro-)chemical and mechanical stressors, to be incorporated. Capacity and power fade as well as the battery cycle life can be analyzed in dependence of the initial structure, the external conditions and the cycling protocol applied, to predict the propagation of the particle density distribution function due to SEI growth. Eikerling and coworkers introduced a statistical modeling framework to study the degradation of platinum nanoparticles in the) unless CC License in place (see abstract). ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 134.169.25.245 Downloaded on 2017-12-10 to IP