Influence of constraints on axial growth reduction of cylindrical Li-ion battery electrode particles

TL;DR

This paper investigates how physical constraints can reduce axial growth and prevent buckling in cylindrical silicon anode particles of lithium-ion batteries, using theoretical models and phase diagrams.

Contribution

It introduces simplified models and phase diagrams to analyze the effectiveness of different physical constraints on axial growth reduction in cylindrical silicon particles.

Findings

Axial growth reduction is limited in annular cylinder geometries.

High yield stress and thickness of constraints improve growth reduction.

Phase diagrams identify optimal conditions for constraints.

Abstract

Volumetric expansion of silicon anode particles in a lithium-ion battery during charging may lead to the generation of undesirable internal stresses. For a cylindrical particle such growth may also lead to failure by buckling if the expansion is constrained in the axial direction due to other particles or supporting structures. To mitigate this problem, the possibility of reducing axial growth is investigated theoretically by studying simple modifications of the solid cylinder geometry. First, an annular cylinder is considered with lithiation either from the inside or from the outside. In both cases, the reduction of axial growth is not found to be significant. Next, explicit physical constraints are studied by addition of a non-growing elasto-plastic material: first, an outer annular constraint on a solid silicon cylinder, and second a rod-like inner constraint for an annular silicon cylinder. In both cases, it is found that axial growth can be reduced if the yield stress of the constraining material is significantly higher than that of silicon and/or the thickness of the constraint is relatively high. Phase diagrams are presented for both the outer and the inner constraint cases to identify desirable operating zones. Finally, to interpret the phase diagrams and isolate the key physical principles two different simplified models are presented and are shown to recover important qualitative trends of the numerical simulation results.