[en] A research effort was made to evaluate the usefulness of modern thermodynamic and diffusion computational tools, Thermo-Calc(copyright) and Dictra(copyright), in optimizing the parameters for diffusion welding of Alloy 800H. This would achieve a substantial reduction in the overall number of experiments required to achieve optimal welding and post-weld heat treatment conditions. This problem is important because diffusion welded components of Alloy 800H are being evaluated for use in assembling compact, micro-channel heat exchangers that are being proposed in the design of a high temperature gas-cooled reactor by the US Department of Energy. The modeling was done in close contact with experimental work. The latter included using the Gleeble 3500 System(reg sign) for welding simulation, mechanical property measurement, and light optical and Scanning Electron Microscopy. The modeling efforts suggested a temperature of 1150 C for 1 hour with an applied pressure of 5 MPa using a 15 μm Ni foil as a joint filler to reduce chromium oxidation on the welded surfaces. Good agreement between modeled and experimentally determined concentration gradients was achieved, and model refinements to account for the complexity of actual alloy materials are suggested.

[en] The NGNP Project is currently investigating the use of metallic, diffusion welded, compact heat exchangers to transfer heat from the primary (reactor side) heat transport system to the secondary heat transport system. The intermediate heat exchanger will transfer this heat to downstream applications such as hydrogen production, process heat, and electricity generation. The channeled plates that make up the heat transfer surfaces of the intermediate heat exchanger will have to be assembled into an array by diffusion welding.

[en] DOE is converting TREAT (Transient Reactor Test) facility from its existing highly enriched uranium (HEU) core to a low-enriched uranium (LEU) core. In order to assess the magnitude of chemical interaction between fuel and cladding materials during physical contact under expected TREAT operational limits, planned transients tests, and reactivity accident scenarios, a combination of experimental testing and thermodynamic modeling was performed to predict the expected chemical interactions among fuel and cladding chemical constituents. Thermodynamic calculations were then validated with empirical data from experiments that emulate TREAT's expected upper operational limits. Pellet samples composed of LEU oxide powder dispersed in a graphite matrix had intimate contact with zirconium-based alloy cladding. The samples were subjected to long-term isothermal heating under high vacuum. Specimen characterization consisted of scanning electron microscopy, x-ray diffraction analysis, and x-ray tomography. ThermoCalc software and Ellingham's diagrams were used for thermodynamic calculations. The X-ray tomography and SEM analysis of the pellets showed a homogeneous distribution of UO₂ particles within the graphite matrix with isolated larger UO₂ agglomerates. The SEM analysis showed the formation of a lamellar structure between the large UO₂ agglomerate and the graphite matrix. This is an indication that diffusion is taking place at a rather moderate temperature and reaction time during the pellet fabrication. Secondary phases, with high Al and Mg content, were detected inside a UO₂ agglomerate using SEM in SE/EBSD mode and chemical analysis with EDS. LEU feedstock impurities, as Al and Mg, seem to promote the chemical reactivity between UO₂ particles and graphite matrix. XRD analysis of a pellet sample showed the presence of UO₂ and graphite, but no other crystalline phases were detected with this technique. Thermodynamic analysis through the use of Ellingham diagrams showed that within TREAT's operational to reactivity accident temperature (278-820 Celsius degrees), UO₂ can be reduced to U metal by Zr due to a more favorable ΔG for the oxidation of Zr