Publications

Abstract not provided.

Polymer foam encapsulants provide mechanical, electrical, and thermal isolation in engineered systems. In fire environments, foams, such as polyurethanes, can liquefy and flow during thermal decomposition, and evolved gases and vapors can cause pressurization and failure of sealed containers. Liquefaction and flow of decomposing foam can cause serious modeling issues in systems safety and hazard analyses. To mitigate the issues resulting from liquefaction and flow, a hybrid polyurethane-cyanate-ester-epoxy foam was developed that has mechanical properties similar to currently used polyurethane foams. The hybrid foam behaves predictably, does not liquefy, and forms 40-50 percent by weight uniform char during decomposition in nitrogen. The char forms predictably and is a relatively uniform "participating medium." A previous paper discussed the experimental and modeling approach developed to predict radiation and conduction heat transfer through decomposing hybrid foam in vented containers. This paper discusses application of a similar approach to the more difficult problem of predicting heat transfer, foam decomposition, and pressure growth in sealed containers. Model predictions are compared with results from radiant heat transfer experiments involving foam encapsulated objects in sealed containers. All model parameters were evaluated from independent laboratory-scale experiments such as TGA and DSC. The time dependent-pressure in the container and the timedependent temperature near the surface of a foam-encapsulated object agreed well with experimental data. © (2010) by BCC Research All rights reserved.

Composite materials behave differently from conventional fuel sources and have the potential to smolder and burn for extended time periods. As the amount of composite materials on modern aircraft continues to increase, understanding the response of composites in fire environments becomes increasingly important. An effort is ongoing to enhance the capability to simulate composite material response in fires including the decomposition of the composite and the interaction with a fire. To adequately model composite material in a fire, two physical model development tasks are necessary; first, the decomposition model for the composite material and second, the interaction with a fire. A porous media approach for the decomposition model including a time dependent formulation with the effects of heat, mass, species, and momentum transfer of the porous solid and gas phase is being implemented in an engineering code, ARIA. ARIA is a Sandia National Laboratories multiphysics code including a range of capabilities such as incompressible Navier-Stokes equations, energy transport equations, species transport equations, non-Newtonian fluid rheology, linear elastic solid mechanics, and electro-statics. To simulate the fire, FUEGO, also a Sandia National Laboratories code, is coupled to ARIA. FUEGO represents the turbulent, buoyantly driven incompressible flow, heat transfer, mass transfer, and combustion. FUEGO and ARIA are uniquely able to solve this problem because they were designed using a common architecture (SIERRA) that enhances multiphysics coupling and both codes are capable of massively parallel calculations, enhancing performance. The decomposition reaction model is developed from small scale experimental data including thermogravimetric analysis (TGA) and Differential Scanning Calorimetry (DSC) in both nitrogen and air for a range of heating rates and from available data in the literature. The response of the composite material subject to a radiant heat flux boundary condition is examined to study the propagation of decomposition fronts of the epoxy and carbon fiber and their dependence on the ambient conditions such as oxygen concentration, surface flow velocity, and radiant heat flux. In addition to the computational effort, small scaled experimental efforts to attain adequate data used to validate model predictions is ongoing. The goal of this paper is to demonstrate the progress of the capability for a typical composite material and emphasize the path forward.

A thermal model is developed for the response of carbon-epoxy composite laminates in fire environments. The model is based on a porous media description that includes the effects of gas transport within the laminate along with swelling. Model comparisons are conducted against the data from Quintere et al. Simulations are conducted for both coupon level and intermediate scale one-sided heating tests. Comparisons of the heat release rate (HRR) as well as the final products (mass fractions, volume percentages, porosity, etc.) are conducted. Overall, the agreement between available the data and model is excellent considering the simplified approximations to account for flame heat flux. A sensitivity study using a newly developed swelling model shows the importance of accounting for laminate expansion for the prediction of burnout. Excellent agreement is observed between the model and data of the final product composition that includes porosity, mass fractions and volume expansion ratio.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Although smolder combustion has been extensively studied both computationally and experimentally, relatively few theoretical studies have examined the two-dimensional structure of the smolder wave. In this paper, two-dimensional smolder in polyurethane foam is modeled with a two-dimensional numerical formulation that includes a seven-step kinetic model of the polyurethane smolder reaction mechanism. The two-dimensional model formulation includes the effects of heat, mass, species, and momentum transfer of the porous solid and gas phase. The seven-step decomposition reaction mechanism, which includes a secondary char oxidation and an additional char pyrolysis step, was developed using genetic algorithm optimization. The mechanism is capable of modeling both forward and opposed smolder. The model was used to study the two-dimensionality of a forward propagating smolder wave. The model results show a two-dimensional structure in the temperature, species, and reaction profiles that agrees qualitatively with experimental observations. Oxygen is consumed at the reaction front, as expected, which leads to different reaction pathways governing the final products (i.e. thermal char and oxidative char). It was found that the model response is sensitive to boundary conditions, thermal properties, and heats of reaction for the char oxidation reaction. The incorporation of the secondary oxidation reaction step in the model paves the way to further analysis of the transition to flaming process. © 2009 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.