Publications

The surface area dependence of the decomposition reaction between lithiated graphites and electrolytes for temperatures above 100◦C up to ~200◦C is explored through comparison of model predictions to published calorimetry data. The initial rate of the reaction is found to scale super-linearly with the particle surface area. Initial reaction rates are suggested to scale with edge area, which has also been measured to scale super-linearly with particle area. As in previous modeling studies, this work assumes that electron tunneling through the solid electrolyte interphase (SEI) limits the rate of the reaction between lithium and electrolyte. Comparison of model predictions to calorimetry data indicates that the development of the tunneling barrier is not linear with BET surface area; rather, the tunneling barrier correlates best with the square root of specific surface area. This result suggests that tunneling though the SEI may be controlled by defects with linear characteristics. The effect of activation energy on the tunneling-limited reaction is also investigated. The modified area dependence results in a model that predicts with reasonable accuracy the range of observed heat-release rates in the important temperature range from 100◦C to 200◦C where transition to thermal runaway typically occurs at the cell level.

The surface topology of a solid subjected to destructive environments is often difficult to quantify. In thermal environments, the size and shape of the solid changes as it pyrolyzes, ablates, warps, or chars. Quantitative descriptions of such responses are valuable for data reporting and model validation. In this work, a three-dimensional scanner is evaluated for non-destructive material analysis. The scans spatially resolve the response of materials to a high-heat-flux environment. To account for the effect of distortion induced in thin materials, back-side scans of the sample are used to characterize the displacement of the bulk material. Data spanning the area of the sample, rather than using a net or average quantity, enhances the evaluation of the crater formed by the incident flux. The 3D reconstruction of the sample also provides the ability to perform volumetric calculations. The data obtained from this methodology may be useful for characterizing materials exposed to a variety of destructive environments.

Intense, dynamic radiant heat loads damage and ignite many common materials, but are outside the scope of typical fire studies. Explosive, directed-energy, and nuclear-weapon environments subject materials to this regime of extreme heating. The Solar Furnace at the National Solar Test Facility simulated this environment for an extensive experimental study on the response of many natural and engineered materials. Solar energy was focused onto a spot (∼10 cm2 area) in the center of the tested materials, generating an intense radiant load (∼100 kW m−2 –1000 kW m−2) for approximately 3 seconds. Using video photography, the response of the material to the extreme heat flux was carefully monitored. The initiation time of various events was monitored, including charring, pyrolysis, ignition, and melting. These ignition and damage thresholds are compared to historical ignition results predominantly for black, α-cellulose papers. Reexamination of the historical data indicates ignition behavior is predicted from simplified empirical models based on thermal diffusion. When normalized by the thickness and the thermal properties, ignition and damage thresholds exhibit comparable trends across a wide range of materials. This technique substantially reduces the complexity of the ignition problem, improving ignition models and experimental validation.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

When exposed to a strong radiant heat source (>1,000 kW/m2), combustible materials pyrolyze and ignite under certain conditions. Studies of this nature are scarce, yet important for some applications. Pyrolysis models derived at lower flux conditions do not necessarily extrapolate well to high-heat-flux conditions. The material response is determined by a complex interplay of thermal and chemical transport phenomena, which are often difficult to model. To obtain model validation data at high-heat-flux conditions (up to 2500 kW/m2), experiments on a variety of organic and engineered materials were performed at the National Solar Thermal Test Facility at Sandia National Laboratories. Mass loss during the short duration (2-4 sec) heat pulse was determined using the pre- and post-test weight. The mass-loss data were fairly linear in the fluence range of 200-6000 kJ/m2. When divided into subsets based on material types, the mass loss was similar at the peak flux/fluence condition for engineered polymers (≈1 g) and organic materials (≈2.5 g), although some exceptions exist (PMMA, dry pine needles). Statistical correlations were generated and used to evaluate the significance of the observed trends. These results contribute to the validation data for simulating fires and ignition resulting from very high incident heat flux.