Publications

Abstract not provided.

An eXplosive CHEMical kinetics code, XCHEM, has been developed to solve the reactive diffusion equations associated with thermal ignition of energetic materials. This method-of-lines code uses stiff numerical methods and adaptive meshing to resolve relevant combustion physics. Solution accuracy is maintained between multilayered materials consisting of blends of reactive components and/or inert materials. Phase change and variable properties are included in one-dimensional slab, cylindrical and spherical geometries. Temperature-dependent thermal properties have been incorporated and the modification of thermal conductivities to include decomposition effects are estimated using solid/gas volume fractions determined by species fractions. Gas transport properties, including high pressure corrections, have also been included. Time varying temperature, heat flux, convective and thermal radiation boundary conditions, and layer to layer contact resistances have also been implemented.

The MELCOR code was used to simulate one of GRS`s (a reactor research group in Germany) core degradation experiments conducted in the CORA out-of-pile test facility. This test, designated CORA-13, was selected as one of the International Standard Problems, Number ISP31, by the Organization for Economic Cooperation and Development. In this blind calculation, only initial and boundary conditions were provided. The experiment consisted of a small core bundle of twenty-five PWR fuel elements that was electrically heated to temperatures greater than 2,800 K. The experiment composed three phases: a 3,000 second gas preheat phase, an 1,870 second transient phase, and a 180 second water quench phase. MELCOR predictions are compared both to the experimental data and to eight other ISP31 submittals. Temperatures of various components, energy balance, zircaloy oxidation, and core blockage are examined. Up to the point where oxidation was significant, MELCOR temperatures agreed very well with the experiment -- usually to within 50 K. MELCOR predicted oxidation to occur about 100 seconds earlier and at a faster rate than experimental data. The large oxidation spike that occurred during quench was not predicted. However, the experiment produced 210 grams of hydrogen, while MELCOR predicted 184 grams, which was one of the closest integral predictions of the nine submittals. Core blockage was of the right magnitude; however, material collected on the lower grid spacer in the experiment at an axial location of 450 mm, while in MELCOR the material collected at the 50 to 150 mm location. In general, compared to the other submittals, the MELCOR calculation was superior.

The MELCOR code was used to simulate PNL`s Ice Condenser Experiments 11-6 and 16-11. In these experiments, ZnS was injected into a mixing chamber, and the combined steam/air/aerosol mixture flowed into an ice condenser which was l4.7m tall. Experiment 11-6 was a low flow test; Experiment l6-1l was a high flow test. Temperatures in the ice condenser region and particle retention were measured in these tests. MELCOR predictions compared very well to the experimental data. The MELCOR calculations were also compared to CONTAIN code calculations for the same tests. A number of sensitivity studies were performed. It as found that simulation time step, aerosol parameters such as the number of MAEROS components and sections used and the particle density, and ice condenser parameters such as the energy capacity of the ice, ice heat transfer coefficient multiplier, and ice heat structure characteristic length all could affect the results. Thermal/hydraulic parameters such as control volume equilibrium assumptions, flow loss coefficients, and the bubble rise model were found to affect the results less significantly. MELCOR results were not machine dependent for this problem.