Despite the obvious success of Li-ion technology over the last 20 years, safety concerns remain. Under suitable triggers, Li-ion cells can experience thermal runaway, i.e., a rapid increase in cell temperature accompanied by venting of combustible vapors, smoke, vent-with-flame, ejection of cell parts, fire and explosion.  Safety failures of lithium-ion cells can result from a variety of triggers; examples of which include overheating, overcharging, crushing, mechanical impact, external shorting and development of internal shorts. The underlying physics for these different failure mechanisms can be quite different, with different reaction kinetics and timing to failure, post trigger.

The internal short circuit trigger is the least studied but most dangerous trigger because it can result in violent cell failure with little warning and with the appearance that cell operation is "normal".  Most Li-ion safety failures that occur in the field take place due to the slow and rare development of such "grown-in" internal short circuits that mature to the point that they result in thermal runaway. An adequate safety test for grown-in internal short development, that replicates the conditions by which such failures occur in the field, has not previously been available and would be a significant improvement in battery safety both at the component level and the system level.

In pursuit of better understanding of these types of failures, we investigated the mechanism by which grown-in internal shorts develop from manufacturing defects, specifically, from the presence in cells of small foreign metal particles.  In parallel, we employed a finite element analysis (FEA) model to develop an enhanced understanding of how such shorts result in thermal runaway.  In this work, supported by the US Department of Energy, we successfully developed methods to implant metal particles in cells, without perturbing cell performance, in order to subsequently reproduce failures “similar” to those that occur in the field.  Following metal particle implantation, charge-discharge cycling of cells leads to shorting and/or thermal runaway.  Extensive cell post-mortems were carried out to confirm the mechanism of shorting.  Factors investigated include the influence of specific metallic contaminant, the electrochemical environment in which the short initiates and develops, chemical and physical characteristics of the surface of the metallic contaminant, and the electrical environment governing current flow to the area of the short.  Different metal contaminants clearly exhibit different electrochemical characteristics including different dissolution, plating, and short-formation behaviors. For example, a cell prepared with a single small nickel particle on the cathode developed an internal short circuit during charge-discharge cycling. A post-mortem clearly showed nickel deposition on the anode with additional nickel deposits in/on the separator. 

Based on insights gained from these investigations we have successfully developed technologies to manage internal short circuits in Li-ion.  CAMX Power (a wholly owned subsidiary of TIAX LLC) has developed two distinct, non-invasive and chemistry-agnostic technologies for sensitive early detection of internal shorts in Li-ion batteries before they pose a thermal runaway threat.  The technologies are non-invasive and easily implemented in Li-ion battery packs.  Detection of developing shorts using these technologies occurs at levels far below the point at which a thermal runaway occurs and, in the best cases, many charge-discharge cycles prior to thermal runaway, as will be illustrated using test data such as are captured in the attached Figure. One of these technologies will be demonstrated "live" during the presentation. Prototype test systems have been supplied to major automakers and are now in advanced stages of demonstration for automotive applications. Reliable early detection provides multiple opportunities for productive intervention.

Subsequently we employed a variety of experimental methods in pursuit of a better understanding of the contrast in physics and mechanism between grown-in failures (described above) and failures caused by "hard" shorts that result from crash/crush/penetration events that often result in instantaneous thermal events. Our experiments included development of custom equipment to create instantaneous "hard" shorts, as well as use of high speed data acquisition and high speed photography to monitor processes following instantaneous creation of a hard short.  This work revealed fundamental new findings regarding how hard shorts result in thermal runaway in contrast to "soft" shorts.  Use of widely considered safety technologies such as ceramic separators and non-flammable electrolytes were investigated using this approach.  The findings clearly illustrate that the underlying physics for grown-in shorts versus hard shorts are quite different, with different reaction kinetics and timing to failure, post trigger. Examples of high speed photographic resolution of thermal runaway processes will be shown and recommendations for safe Li-ion batteries will be summarized.