Font Size: 

The effects of modulus and partial molar volume on curvature of a bilayer electrode in Li-ion battery are analyzed by using the finite bilayer plate theory. In order to obtain the closed form solution of the relationship between curvature and material properties for this finite bilayer plate, three steps are applied. To apply the curvature solution, we consider a bilayer electrode, in which a layer of graphite (or other active material) is fabricated on the surface of a copper foil current collector (or some other substrate not permeable to lithium). The bending of the bilayer plate structure is controlled by the stress due to the diffusion of lithium into the layer of graphite. New experiment setups which can in situ online measure the deformation of electrode during charge/discharge are developed. In these experiment setups, the deformation is calculated by the finite bilayer plate theory and measured by a high speed high resolution CCD camera which could also measure large deformation. LiFePO₄ acted as the positive electrode, while a porous graphite electrode made by Shenzhen Kejing Star technology Co. served as the negative electrode. Copper was used as the current collector for the anode and aluminum was used as the current collector for the cathode. The thickness of graphite active material and copper current collector are 91µm and 9µm respectively. The LiFePO₄ electrode was separated from the graphite electrode by a woven Celgard 2325 separator (thickness:25μm). 1M lithium hexafluorophosphate in 1:1 (vol%) ethylene carbonate:diethyl carbonate was used as the electrolyte. The cell was assembled in a glove box in ultra-high pure Argon atmosphere at 25⁰C (+1⁰C). The deformation of electrode was recorded in situ on line by a high speed high resolution CCD camera. The modulus variation has a significant effect on curvature of electrode. Theoretical prediction of κ is almost twice as much as the experimental value without modulus change and only 5%-10% deviation with modulus change. The theoretical predictions of κ and its experiment value are agreed very well when both modulus and partial molar volume variation are taken into consideration. The prediction value of partial molar volume which determined by the experiment is very closed to the value used by most literatures.