The microstructure in rechargeable Li-ion batteries is a complex ensemble of Li sources and sinks embedded in an electrolyte medium. Typically, the rate-determining processes during charging and discharging are a combination of: 1) Li diffusion in the electrolyte and separator, transport of Li-ion across the electrolyte/electrode-particle interface, 2) diffusion of Li atoms in the electrode, and 3) the nucleation and growth of Li-poor and Li-rich phases in the electrode particles.Macroscopic models have been developed for this technologically important system. Computational models for lithium-ion batteries were developed by West et al. [1][2][3][4] in which the electrode/interface/electrolyte was modeled as superimposed continuum, and the separator was not included. Doyle et al. developed a meanfi eld model which incorporated microstructural aspects by homogenizing a spatial distribution of particulate active material and porosity; [ 5 , 6 ] whereby the transfer of Li ions between the electrolyte and the active material was modulated by the Butler-Volmer (B-V) relation. Doyle's model has been employed by several authors [ 4 , 7-12 ] and additional effects have since been included into the model, such as, side reactions and capacity-limiting mechanisms, [ 13,14 ] effects of porosity gradients, [ 15 , 16 ] and thermal effects. [ 17,18 ] These methods shed insight for battery designers, but mean-fi eld approaches do not account for microstructural details, such as the particle geometry, distribution, and crystallographic orientation. These microstructural details were treated by Garcia et al. [ 19, , 20 ] who developed a Finite Element Method (FEM) that includes the time-evolution of the electrochemical potentials in representative microstructures and include particle-particle interactions and interface control.However, none of these macroscopic models treat the microscopic details of interface kinetics, Li diffusion, and phase evolution within the electrode particles. We believe that such microscopic models will provide further insight for battery design. Moreover, they could be coupled to mean-fi eld or FEM computations (as seen in the literature [ 19,21 ] ) to provide a better physical model of the charging/discharging process.We developed a kinetic Monte Carlo (KMC) method (Section 3.3) that treats the kinetics of Li-insertion and the structure and morphological evolution of the Li-rich/Li-poor phase boundary in electrode particles. This KMC model is coupled to a fi nite difference simulation (Section 3.1) of Li-ion diffusion into the surrounding electrolyte. This diffusion simulation couples the far-fi eld ion fl ux (imposed by galvanostatic boundary conditions) to the concentration fi elds in the proximity of electrode particles. These local concentration fi elds are coupled to particle adsorption via B-V interface kinetics (Section 3.2). In this paper, the model is developed and its results are related to microstructural and crystallographic
Kinetic Monte Carlo Simulations of Anisotropic Lithium Intercala...