No abstract

No abstract

Oven exposure testing is a standard benchmark that Li-ion cells must pass in order to be approved for sale by regulating bodies. In order to test the safety of new cell designs or electrode materials, manufacturers must make small test batches of cells. This can be both costly and time consuming. Using reaction kinetics that have been developed for electrode materials with electrolyte exposed to high temperature, and thermal properties of cells from the literature, a predictive model for oven exposure testing has been developed. The model predictions are compared to oven exposure test results for E-One/Moli Energy, Canada, 18650 LiCoO2 /graphite cells and shown to be in good agreement. The model can predict the response of new cell sizes and electrode materials to oven exposure testing without actually producing any cells. This is illustrated with a number of examples: (i) increasing the specific surface area of the graphite electrode; (ii) using LiMn2O4 or other cathode substitutes instead of LiCoO2 ; (iii) varying the diameter of cylindrical cells; and (iv) varying the thickness of prismatic cells. © 2001 The Electrochemical Society. All rights reserved.

In a recent paper, Lu et al. showed that normalLifalse[Ni0.25Co0.5Mn0.25false]O2 and normalLifalse[Ni0.375Co0.25Mn0.375false]O2 are attractive electrode materials for Li-ion cells. Most interesting, they appear to be much less reactive with electrolyte at high temperatures than LiCoO2. Since these two materials are part of the solid-solution series normalLifalse[NixCo1−2xMnxfalse]O2 with 0⩽x⩽1/2, which also contains LiCoO2 when x=0, it is possible to study the reasons for the safety advantage by a careful study of electrochemical, structural, and thermal properties as a function of x. This paper reports the structural and electrochemical properties of normalLifalse[NixCo1−2xMnxfalse]O2 for 0⩽x⩽1/2. The materials all show good specific capacity (between 110 and 130 mAh/g between 3.0 and 4.2 V vs. Li at 40 mA/g) and very little capacity loss over 50 cycles. Careful consideration of differential capacity measurements proves that Ni2+ and Mn4+ do, indeed, substitute for Co3+, at least for x<0.15, as was hypothesized by Lu et al. The thermal stability of the charged cathode materials improves rapidly compared to LiCoO2, then saturates as x in normalLifalse[NixCo1−2xMnxfalse]O2 increases. Materials with x⩾0.075 show approximately equal thermal stability. The reason for the variation of the thermal stability with x is the subject of a companion paper. © 2002 The Electrochemical Society. All rights reserved.

No abstract