Failure Statistics for Commercial Lithium Ion Batteries

There are relatively few publications that assess capacity decline in enough commercial cells to quantify cell-to-cell variation (failure statistics), but those that do show a surprisingly wide variability in the rates of failure. Capacity curves cross each other often, a challenge for efforts to measure the state of health and predict the remaining useful life (RUL) of individual cells. We analyze capacity fade failure rates for 24 commercial pouch cells, providing an estimate for the time to 5% failure. Our statistical data indicate that RUL predictions based on remaining capacity or internal resistance are accurate only once the cells have already sorted themselves into “better” and“worse” ones. Analysis of our failure rate data, using maximum likelihood techniques, provide uniformly good fits for a variety of definitions of failure with normal and with 2- and 3-parameter Weibull probability density functions, but we argue against using a 3-parameter Weibull function for our data. pdf fitting parameters appear to converge after about 15 failures, although business objectives should ultimately determine whether data from a given number of batteries provides sufficient confidence to end lifecycle testing. Increased efforts to make batteries with more consistent lifetimes should lead to improvements in battery cost and safety.

Journal of Power Sources 342 (2017) 589-597

These represent almost the only published data on the statistics of Li battery failure for commercial cells. Note that in every case, all cells initially degrade at the same constant rate. At some point a knee develops. The actual failure time depends on when that knee occurs. This suggests that the second failure mode controls life.

Steve Harris Resume

The goal of much of present day Lithium battery research is to develop higher energy density batteries. Consider 4 approaches:
(1) Positive electrodes with greater capacity. Such electrodes can be made, but their durability is poor.
(2) Higher voltage positive electrodes, up to 5 V. Such electrodes can be made, but we do not have electrolytes that are stable at such high voltages. Neither approach addresses volumetric energy density.
(3) Higher mass density (lower porosity) electrodes. This could lead to higher tortuosity. A reduction in porosity from 40% to 25% would again increase energy density by 25%.
(4) More durable electrodes. The connection between durability and energy density comes from the fact that in order to achieve long life, much of the energy in Li-ion batteries is never accessed. If we could access 80% of the battery’s energy instead of, say, 65%, the energy density would increase by 25%. Importantly, both volumetric and gravimetric energy density would increase.
We believe that heterogeneity is the ultimate reason that many Li-ion batteries access only about 65% of their theoretical energy. That’s because when the average state of charge (SOC) in an electrode is 65%, parts of the electrode are already at 100% state of charge, and at such high values for SOC, either plating or electrolyte oxidation occurs. See Figure 6 of “Particle Size Polydispersity in Li-ion Batteries.”

Our work researching lithium battery problems is predicated on two hypotheses: First, that degradation and failure initiate at inhomogeneities (or heterogeneities) the the battery microstructure; and second, that these heterogeneities lead to an inhomogeneous transport of lithium ions. Inhomogeneities include any structures where there are rapidly varying spatial properties, such as the SEI lyer. The SEI film is an important site for generating lithium battery aging and failure. For this reason, we believe that a general study of degradation and failure can begin with identification and quantification of inhomogeneities (typically at the mesoscale) as well as measurements of Li transport and insertion into porous electrodes in the battery. These measurements could then guide researchers towards other experiments and models that provide fundamental knowledge of durability (aging, degradation and failure) using advanced diagnostic techniques. We are especially interested in research that connects with models that take into account microstructural and nanoscale properties and inhomogeneities.