Multiscale Microstructure Modeling and Design of Electrochemomechanics in Lithium-ion Batteries
This thesis investigates the electro-chemo-mechanics behavior of electrodes in lithium ion batteries. One of the critical challenges in advanced lithium ion batteries is preventing fracture and mechanical failure of electrodes during lithium insertion and deinsertion. The large volume expansion, phase transition, and the associated Li diffusion induced stresses within electrode materials can lead to their fracture and failure. Numerical simulations have the potentials for design optimization of lithium ion batteries. For traditional Newman's model, it is limited by the assumption that the electrodes are constructed with uniformly distributed spherical particles of equal size. To consider the influence of real microstrucutural effects, a 3D microstructure resolved model has been developed. The developed model considers the electrochemical reactions, Li transport in electrodes and electrolyte, lithiation induced volume change, mechanical strains and stresses, and the electro-chemo-mechanical coupling. Important parameters such as Young's modulus were investigated by ab initio tensile tests. The Young's modulus and ultimate strength of LixCoO2 under various lithium ion concentration were calculated. The observed Li concentration dependent mechanical properties and anisotropy are due to the changes of the Co-O bond strength during Li intercalation. The Young's modulus and ultimate strength of Li xCoO2 have a linear relationship with both the Li concentration and the charge transfer. The obtained concentration dependent expressions was used as input to the continuum model. The model was used to study the diffusion and mechanics behaviors in polycrystalline microstructures with varying grain size, grain boundaries and crystallographic orientations. It is found that the chemical diffusion coefficients increase with increasing grain orientation angle and decrease with the decrease of the grain boundary diffusivity. For small grain boundary diffusivity, the stress increases with increasing grain orientation angle. In contrast, for large grain boundary diffusivity, the stress decreases with increasing grain orientation angle due to reduced concentration gradients in grain boundary regions. The developed model was also applied to study the diffusion induced stress in realistic microstructures reconstructed from FIB-SEM and CT. By simulating discharge processes, the results show that microstructure has a significant influence on the lithium ion diffusion and voltage response. The polarization was studied to explain the significant voltage drop at high C rate. It is obtained that the activation overpotential is the major contribution to the total polarization, and it is about 4-5 times larger than the concentration polarization. It is obtained that the stress generation inside lithium ion battery is highly dependent on microstructure. The maximum stress is more likely to occur at concave regions rather than convex regions. The study shows the maximum stresses in the concave region can be 32% larger than the convex region. The model was extended to study the phase separation and stress generation. Th extended model can track the phase boundary implicitly and can be used in complex geometries. Compared with the elliptical and spherical particles, the stress in phase-separating LiFePO4 reconstructed from nano-CT is about 1.4 times higher. In summary, a multi-scale multi-physics microstructure based model was developed to evaluate the electro-chemo-mechanics in lithium ion batteries. The model can provide a computational tool for battery materials design.
Son, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our