AVS2011 Session AC+SS-ThM: The Surface Science of Actinides and Rare Earths

An understanding of the oxidation and corrosion processes of plutonium metal at room temperature is important to the safe, effective use and storage of this reactive metal. The oxidation rate for the δ-phase stabilized, plutonium/gallium alloy (a commonly employed alloy) can be significantly affected by a number of parameters including the gallium content and the composition of the oxidizing atmosphere (O₂, O₂/H₂O, H₂O). The nature of plutonium oxidation has typically been thought of as the growth of a dioxide (PuO₂) overlayer on the metal to a thickness at which the film begins to spallate (μ's). Based on thermodynamic arguments, it has been pointed out that a relatively thin layer of the sesquioxide (Pu₂O₃) should exist at the dioxide/metal interface for thick oxide films.

Historically, the oxidation/corrosion of plutonium has been studied by oxygen uptake of samples at elevated temperatures inferred from mass gain measurements. Accuracy of these experimental setups likely limited measurements to oxide films thicker than ~0.05 to 0.1 μ (50 - 100 nm). This is at the upper-limit of the thicknesses typically observed for plutonium oxide films. Little work has been published for studies of plutonium oxide thin-films (< 50 nm) on metal substrates. Additionally, very little is known about the role that gallium plays during the oxidation of the alloy other than it can significantly slow the rate.

The oxidation of uranium dioxide has received much experimental and theoretical attention over the last several decades in large part because of its relevance to the operation and storage of uranium-based nuclear fuel. The oxidation process is inherently complicated, involving the formation of multiple different phases via distinct mechanisms even at relatively low temperatures. In the range of a few hundred degrees centigrade oxidation is generally assumed to be a two step process[1]: UO₂ → U₃O₇/U₄O₉ → U₃O₈. At low pressures the intermediate phases adopt crystal structures that are modifications of the UO₂ fluorite structure and are slightly denser. By contrast, U₃O₈ forms a considerably less dense orthorhombic structure (by some 23%). The large volume expansion resulting from the oxidation of UO₂ to U₃O₈ is a potentially serious concern in the event of oxidation of a fuel element, with consequent splitting of protective sheaths and the spalling of powder.

While attention has been focused on the oxidation of UO₂ at elevated temperatures, the associated experiments have all been performed at low partial pressures of oxygen. It is unclear how pressure affects the oxidation process – particularly in the context of the formation of U₃O₈, with and its large volume change with respect to UO₂. We have examined the oxidation of a nominal single crystal of UO₂ in pure oxygen at elevated pressures up to approximately 0.9 GPa (9000 atm) and temperatures of up to 450 °C. In-situ Raman scattering measurements were made as a function of temperature in order to monitor the oxidation. Recovered material was examined using electron based techniques including SEM, TEM, and electron diffraction and also using x-ray photoelectron spectroscopy.

Material synthesized under high pressure has a Raman spectrum that is different from both the UO₂ starting material and the common form of U₃O₈. Also, compared with common U₃O₈, we find that it has fewer crystalline defects and mostly adopts a hexagonal rather than orthorhombic form. Figure 1 of the supplemental document compares Raman spectra of UO₂ in oxygen with synthesized material. Fig. 2 compares electron diffraction obtained from recovered material with U₃O₈ synthesized at ambient pressure.

Lawrence Livermore National Laboratory is operated by Lawrence Livermore National Security, LLC, for the U.S. Department of Energy, National Nuclear Security Administration under Contract DE-AC52-07NA27344. This work is funded by Laboratory Directed Research and Development (LDRD) Program (10-SI-016) of Lawrence Livermore National Laboratory.

Preferential uranium hydriding occurs frequently in narrow gaps. There are different hypotheses about its causes, one of which could be radiation-induced chemistry in gaps. Both ²³⁸U and ²³⁵U generate ionizing α, β, and γ radiation capable of vibrationally or translationally exciting, ionizing, or dissociating H₂, all of which increase the reactivity of H₂ with uranium. Dissociation of H₂ is necessary to initiate hydriding, and it has been shown that the sticking coefficient of H atom is about 1200 times higher than H₂’s[1]. Here we use the MCNPX radiation transport code to calculate the energy dependent electron flux generated from the ^234mPa β^- decay and from photoelectrons generated by brems-strahlung. We apply the code to gaps occurring in two ²³⁸U cylindrical pieces welded together and filled on the inside with LiD with a 100µm gap between ²³⁸U and LiD, and having a 100µm gap in ²³⁸U itself, typical for step-joint-welded uranium shells.

The MCNPX Monte Carlo Code - as configured now - tracks the life cycle of electrons throughout the material and calculates the electron flux as a function of energy, putting results into “energy bins” 1 keV wide. We find that at 2 keV (±.5 keV, the last energy bin) the calculated electron flux in the U-U gap is approximately 19 times larger than in the U-LiD gap, and fifty two times larger than in the U-air gap on the outside of the cylindrical shells. Cross-sections for electron-hydrogen collisions peak, however, below 1 keV energy. We establish the upper limit of the effect of electron-hydrogen collisions by extrapolating the MCNPX electron flux results from the last bin to energies as low as 1 eV by fitting a function to the flux between 2 and 20 keV. To calculate the fraction of H₂ vibrationally or translationally excited, ionized, or dissociated per cm²/s , we integrate the product of the energy dependent cross sections (listed in reference [2]) and the energy dependent electron flux over the relevant energy range. The fraction of H₂ molecules calculated to be dissociated is small, but significant during long-time exposure. Future work will extend the MCNPX code below 1keV (as is done for biological radiation damage), to avoid energy extrapolation.

This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.

References

1. Balooch M, Hamza AV. “Hydrogen and water vapor adsorption on and reaction with uranium.” Journal of Nuclear Materials, 230, 3, 259-270, 1996.

2. Yoon JS et al., “Cross sections for electron collisions with hydrogen molecules.” Journal of Physical and Chemical Reference Data, 37, 2, 913-931, 2008.

The development of rare-earth (RE) doped phosphors allows for the conversion of photons to energies that are more usable for the desired application. Additionally, these RE phosphors have long lifetimes, on the order of ms, which offer potential in many energy conversion and energy transfer devices. Currently, RE phosphors are used in fiber optics amplifiers, broad absorption solar cells and various other lighting applications. Energy transfer mechanisms of the excited RE states, such as defect quenching and sensitizer/emitter interactions, must be understood in order to achieve high efficiency energy conversion and propagation.

In order to increase the efficiency of solar cells, high efficiency phosphors need to synthesize in order to convert photons at the edge of the absorption band into higher energy photons while avoiding undesired quenching effects. In effort to produce these high efficiency phosphors, spatially controlled RE doped thin films are deposited around nanoparticles (NP) to produce a core-shell nanophosphor. Reduction in luminescent quenching can be achieved by increasing the distance between the surface quenching site and active ions through the deposition of a precisely controlled thin film around the NP. Secondly, the luminescent fingerprint can be further controlled through doping of the shell structure by increasing the absorption spectrum or the introduction of emission peaks. Deposition of a thin film around an Y₂O₃:Er³⁺ core NP is achieved via atomic layer deposition or sol-gel synthesis to compare the effect of a spatially controlled vs. a random dopant distribution. Primarily, this work focuses on the emission of visible photons through upconversion, making them ideal components in broad absorption solar cells. By spatially controlling the position and concentration of the RE ions within the nanostructure, increased luminescence is observed due to energy transfer between the dopant ions within a critical interatomic distance. Passitvation of surface sites with increasing film thickness was shown to increase luminescent lifetimes up to 53%, with a critical shell thickness of 8 nm, while lowering the theoretical lifetimes extracted from Judd-Ofelt parameters. The effect of the spatially controlled Yb ions was probed through the extraction of the upconversion photon requirement, showing a statistical decrease in photons from 2.16 to 1.43, or ~30 %. Finally, the effective energy transfer distance and energy transfer coefficients were studied as a spacer layer is added to the system, showing energy transfer up to ~3 nm, confirming the Föester-Dexter theory.

Recent studies showed a colossal enhancement in the ionic conductivity of multi-layer oxygen-ion conducting thin films compared to most commonly used solid oxide fuel cell (SOFC) electrolytes. It has been observed that, the oxygen ionic conductivity of nano-scale hetero-structures increases with the increase in number of layers. However, some of these findings were questioned due to the inability to distinguish electronic and ionic conductivity. Thus, here we investigated the scandia stabilized zirconia and samaria doped ceria (SDC/ScSZ) multi-layer system to understand the oxygen-ion conductance in multi-layer hetero-structures.

In this study, the growth of SDC and ScSZ multi-layer thin films was carried out using the optimized growth conditions, dopant concentrations and film properties established for single- layer SDC and ScSZ epitaxial thin films. The epitaxial SDC and ScSZ multi-layer thin films were grown on Al₂O₃(0001) substrates by oxygen plasma-assisted molecular beam epitaxy (OPA-MBE) at 650°C. The number of layers in the multi-layer hetero-structures was varied from 2 to 20 by keeping the total film thickness constant at 140 nm. Following the growth, thin films were characterized by various in-situ and ex-situ characterization techniques including reflection high energy electron diffraction (RHEED), x-ray diffraction (XRD), Rutherford backscattering spectrometry (RBS), and x-ray photoelectron spectroscopy (XPS). The initial streaky RHEED pattern indicates the epitaxial growth of SDC/ScSZ multi-layer thin films. Furthermore, RHEED patterns indicate the transition of the initial smooth surface to a rough surface with the increase in number of layers. The individual layer thickness was found to be approximately 7 nm for the twenty-layer film as confirmed by x-ray reflectivity data. RBS was also used to find the composition and thickness of the films. HRXRD patterns of ScSZ/SDC thin films exhibit only CeO₂(111) and ZrO₂(111) reflections, indicating the growth of epitaxial SDC(111) and ScSZ(111) multi-layers. XPS depth profile confirmed the uniform dopant concentration in both SDC and ScSZ layers, which was found to be 7 and 6 atom % for Sc and Sm, respectively. In the near future, the ionic conductivity of SDC and ScSZ multi-layer thin films will be measured by four probe conductivity method.