Sandia_Natl_Labs_FY19_LDRD_Annual_SAND2020-3752 R_2_S


Design/fabrication of reliable, reactive multilayer coatings/foils to release stored chemical energy as light/heat. A Sandia-led team discovered the thermal and chemical factors underlying stable deflagration in reactive nanomaterials. These discoveries enabled the design and fabrication of new, more reliable reactive multilayer coatings and foils that can be stimulated by an external source to promptly release stored chemical energy as light and heat. The developed heterogenous solids have tailorable ignition thresholds and calorific output – important for envisioned heat source applications such as joining and power sources. Critical to function, these materials also avoid unstable reaction modes including 2-dimensional spin instabilities. The predictive, 3D reaction-diffusion model established for this LDRD project provides opportunity for rapid assessment of emerging reactive materials and optimization for future applications. Details were also published in the Journal of Applied Physics .

Snapshots showing propagating reaction waves in Co/Al nanolaminates after point ignition. Unstable and stable propagating waves develop in thick and thin period multilayers, respectively. Temperature maps on top are results from predictive thermal model simulations with the ignition zone positioned in the left, upper corner (in orange). The gray-scale, high-speed videographic images, included at the bottom, show similar behavior in experimental test samples. Multiple spin waves are evident in the unstable example on the left. Reacting material is bright in these images due to light emission.

Statistical uncertainty quantification for multivariate physical parameter estimation with multivariate outputs. Sandia recently developed algorithms quantifying uncertainty in computer model inputs by diagnosing and reducing overfitting in model calibration when the input is functional in nature. These methods were applied to the calibration

of dynamic material properties in Z-machine experiments. Elastic shape analysis is used to calibrate misaligned functional data, and power-likelihood models are used

to discount the statistical information associated with model discrepancy. Additionally, this method has found interest in the calibration of sea ice models by the monitoring of cracks in the ice shelf. The method is being actively applied in that domain. Above: Improved uncertainty quantification and data fitting is achieved using elastic functional data calibration and power likelihood models in the Tantalum Equations of State (EOS) exemplar problem. (Figure by James Tucker)



Made with FlippingBook Learn more on our blog