Calculation of expected pit volumes and exsolved metal particle sizesThe maximum radius, R (nm), of a particle forming sub-surface composed of Ni exsolved from a rectangular parallelepiped perovskite of area A e (nm 2 ) and depth d e (nm), under the constraint that d e /2 < 3 R (required for the particle to escape the bulk; see modeling section) can be calculated using the following equation:

We studied the mechanical damage within a lithium-ion graphite-based porous electrode during electrochemical cycling. The effects of charging-discharging rate and the variation in graphite diffusivity on average stress in the electrode cell were investigated. In particular, differences between spatial and average stress evolution in graphite particles were explored. We considered two different microstructures: a) graphite particles connected together with binder bridges and b) graphite particles encased in binder shells. Electrochemical charging-discharging in a composite electrode was simulated by spatially resolving the electrode and electrolyte phases. As indicated by experimental measurements, the binder is assumed to follow an elastic-plastic stress-strain relation. Average stress developed in the electrode was calculated for different binder yield-stress levels and an appropriate yield-stress value was chosen on the basis of experimental findings of the literature. We find the stress in the particles can be of the order of 43 MPa, and can be particularly large in regions where the particles come in close contact with their neighbors. The average stress in the electrodes, however, is the range of 10 MPa and is largely determined by the mechanical properties, in particular the yield stress of the binder. Computed stress profiles were compared qualitatively with experimental measurements using the wafer-curvature method. Elastic stresses and plastic strains predicted by 3D models are shown to be close to those predicted using simpler 2D models of the microstructure.Lithium-ion batteries have gained prominence in electronic devices such as cell phones and laptop computers owing to their high energy density that reduces their weight by half and their volume by 20% compared to other batteries, such as nickel-cadmium and nickel-metal-hydride. Lithium-ion batteries also present an attractive solution as renewable energy sources for the transportation industry. Lithium diffuses into the electrode, and hence they undergo volumetric expansion resulting in stress and large plastic deformations, which can potentially lead to the failure of electrodes and of the binder that holds the particles together. [1][2][3][4][5] To address the stability of lithium-ion batteries and to study the stress response during the electrochemical intercalation and deintercalation processes, theoretical and numerical investigations were conducted at particle and electrode levels.The mechanical properties of different binder types have been extensively studied. 6-10 Experimental investigation shows that in a lithium-ion battery electrode, binders form thin non-continuous layers on the particle surface. 11,12 Modeling the realistic morphology of binders is an extremely difficult task, which requires using image processing techniques. 13,14 To avoid this difficulty, Awarke et al. 11 assumed that the electrode pore volumes are fully occupied by binders that could overstate the stiffness of the electrode. They presented a 3D mesoscale model of th...

Most models of battery charging assume a shrinking core model with a Lithium-rich shell and Li-poor core. In this study, we provide direct experimental evidence from in situ optical measurements and that suggests that the lithiation of (spherical) MCMB (meso carbon micro beads) particles do not follow the shriking-core model, even for particles which eventually become uniformly lithiated. We observe "hot spots" which expand until the particles are fully charged. The observations are explained using a microstructural model of MCMB particles that 1) accounts for their polycrystalline nature, 2) anisotropy in diffusion and 3) dependence of the intercalation rate at the surface on the orientation of the graphene planes. We show that our model reproduces the key features of the charging process observed in experiments. Implications of these results to Li-plating are discussed.