Abstract
Use of highly-enriched uranium (HEU) at research reactors or the exporting of HEU for humanitarian purposes, such as for the production of medical isotopes, leads to a risk of nuclear proliferation. This thesis will discuss efforts to improve the understanding of low-enriched uranium (LEU) to enable more accurate predictions of the material behavior during manufacturing and use as an irradiation target for the production of medical isotope Tc-99m and as a reactor fuel.
The α-phase (orthorhombic crystal structure) LEU foils proposed for use in the production of medical isotope Tc-99m have anisotropic properties due to the crystallographic texture which is introduced during the foil rolling process. This was previously demonstrated using physics-based viscoplastic self-consistent (VPSC) modeling, which used the deformation-induced texture as an input [1]. A phenomenological, analytical model for the anisotropic yield stress behavior of orthotropic, hexagonal metals was developed by Cazacu, Plunkett, and Barlat [2], denoted CPB06. A MATLAB optimization routine was used to determine values for the anisotropy coefficients used in the model, by fitting to the VPSC predictions. CPB06 was implemented as a user-subroutine (VUMAT) in ABAQUS/Explicit, a commercial finite element analysis (FEA) software, which allowed for finite element simulation of the irradiation target manufacturing process. FEA simulations ultimately revealed that while the plastic anisotropy of the foil could potentially change the strength of the material relative to the isotropic case under certain loading conditions, anisotropy did not noticeably affect the foil strength when under internal pressure, and the performance of the Tc-99m target was not affected.
The -phase stabilized LEU-10Mo (wt%) alloy has been identified as a candidate fuel for high performance research reactors, though there are concerns regarding phase and mechanical stability under reactor conditions, especially during transient conditions. In-situ neutron diffraction performed at the Los Alamos Neutron Science Center (LANSCE) was used to investigate phase decomposition behavior in U-10Mo and U-9.8Mo with 0.2 wt% ternary additions of Cr, Ni, or Co, thus maintaining the total alloy content in all four alloys at 10 wt%. Since the metastable BCC phase γ-U is optimal, it is critical to understand whether or not such alloying additions delay or promote phase decomposition. These alloying additions were chosen for research since they are readily available, and come in different unit-cell structures (BCC, FCC, and HCP).
During the in-situ experiments performed on the Spectrometer for Materials Research at Temperature and Stress (SMARTS), the samples were first heated at a rate of 50 ºC/min to ~650 ºC, which is above the γ-phase solvus line, and held for 1 hour to dissolve any fine, second phase particles which may have precipitated during prior heating and homogenization steps. Then, the samples were cooled at a rate of 50 ºC/min to the isothermal hold temperatures of interest, 490 or 500 ºC, and held for 20 hours to observe the kinetics of decomposition of the metastable γ-U-Mo phase toward the equilibrium α-U and γ’ (U2Mo) phases. Finally, the samples were cooled to room temperature at a rate of 50 ºC and remeasured ex-situ, both in SMARTS and in the High Intensity Pressure and Preferred Orientation (HIPPO) instrument. Rietveld analysis using the GSAS-II and MAUD software packages was employed to determine the phase fractions, lattice parameters, and crystallographic texture of all the observed phases.
Experiments conducted on U-10Mo and U-9.8Mo-0.2Cr did not exhibit measurable phase decomposition. However, some phase evolution was observed in the U-9.8Mo-0.2Ni and U-9.8Mo-0.2Co alloys, which included development of the orthorhombic α-U phase along with a corresponding molybdenum-rich, and perhaps ordered, version of the BCC γ-phase, here denoted γb. Hence, it is concluded that Ni and Co ternary additions degrade the thermal stability of U-10Mo, while Cr additions do not have an observable effect. It is hypothesized that the more rapid phase evolution in the Ni and Co containing alloys is due to heterogeneous nucleation associated with the presence of small grain boundary precipitate phases, including U6Co and U6Ni, which were previously observed by SEM [3], and whose presence is corroborated by the presence of a small, solitary neutron diffraction peak in the experiments performed on U-9.8Mo-0.2Co.
