The physics and chemistry of oxidation

Defining the electronic structure of surface oxides

Understanding the physics and chemistry of oxidation is essential for long-term plutonium storage. Under normal atmospheric conditions oxide layers cover plutonium metal surfaces, which ultimately control chemical reactivity. Our experimental and theoretical efforts are aimed at revealing the electronic structure of surface oxides as well as the mechanisms of their formation. First, we describe the manner in which plutonium oxides develop in a highly controlled environment as seen by photoemission. We also describe this evolution using theoretical ab initio calculations of the electronic structure.

Clean and "nearly" clean delta-plutonium along with the difference spectrum between the

Photoemission spectroscopy (PES) provides an excellent direct probe of the electronic structure of surfaces and is extremely sensitive to surface contaminants so that even fractions of monolayers of adsorbed gases can be detected. Typical excitation sources include specific helium (He) plasma emission lines: He I alpha, 21.2 electronvolts (eV); He II alpha, 40.8 eV; and He II beta, 48.4 eV. Alternatively tunable sources in the vacuum ultraviolet (UV) region, such as those at synchrotron sources, have also been used. Los Alamos' Laser Plasma Light Source (LPLS) project has also provided tunable, high- intensity excitation light for PES studies.

We conducted our studies of the surface electronic structure following gas adsorption using the He II alpha 40.8-eV excitation energy because at that energy there is a sizeable cross section for both oxygen 2p and plutonium 5f and 6d electronic shells. Because of sensitivity to trace impurities, the sample was kept at a base pressure of 6x10^-11 torr during and following cleaning by laser ablation. Laser ablation has proven to be an excellent way to remove surface contaminants from actinides and other materials. The low base pressure ensures that any residual background adsorption of trace gases from the vacuum does not compromise the measurements over an extended timescale (hours).

The figure above shows PES data for a "clean" delta-plutonium surface and a plutonium surface with a small residual amount of oxygen. A difference spectrum between the two is provided and has the spectral characteristics of Pu₂O₃. The figure also demonstrates the need to obtain as clean a surface as possible to determine the true electronic structure of the base metal. The clean surface exhibits two main features typical of delta-plutonium: a localized peak at about 1 eV attributed to emission from the 5f level and characteristics at the Fermi level attributed to emission from 5f and conduction electrons. The subtle distinctions in features inherent to these clean surfaces can become obscured by trace contamination. In fact, earlier investigations on delta-plutonium sample surfaces purported to be "clean" exhibit low-level oxide contamination that may be comparable to our low-level doses.

Top: This view through a window on the side of the measurement chamber shows the copper sample holder, which is a central part of the apparatus. Samples mounted to the holder may be cooled to 10 kelvin for photoemission measurements.

Center: A close-up of the sample holder. The small square in the center is the plutonium sample. Surface cleaning is performed by focusing an infrared (1,064-nanometer) laser beam on the sample.

Bottom: The photoemission measurement chamber. The top and middle photos were taken through the inspection window in the front of the chamber. Pressure inside the chamber is approximately 10¹³ times lower than atmospheric pressure outside the chamber. To maintain such a high-vacuum environment, the chamber must occasionally be heated to temperatures above 120 degrees Celsius. The chamber is wrapped in aluminium foil to ensure uniform temperature distribution and thermal insulation during bakeout to establish operational pressure.

Photos by Tomasz Durakiewicz and Ela Guziewicz

Gas dosing was carried out by means of a needle valve from a gas line into the main vacuum chamber, which allowed for accurate surface exposures for O₂ gas measured in langmuirs (L). (A langmuir is a unit of measure used in surface science that is roughly equivalent to producing a surface saturated monolayer with a physically or chemically adsorbed species provided all incident gas molecules react and remain on the surface with unity probability.)

Given the reactive nature of plutonium and other actinides, this places a significant burden on the experimentalist's ability to work at an atmosphere where the surface is not compromised by spurious nondesirable adsorption. From the kinetic theory of gases, a 1-langmuir exposure is equivalent to 10^-6 torr gas pressure for one second at the surface. In this work measurements were made at temperatures of 10, 77, and 300 kelvin (K) to assess the relative role of kinetic formation and transformation rates among the plutonium oxides.

The figure on this page demonstrates the changes in surface electronic structure following sequential gas exposures at 77 K with 1, 5, 10, and 20 langmuirs of O₂. Following a 1-langmuir O₂ exposure an oxide is formed that we associate with Pu₂O₃ at the metal surface. In spectra recorded following a 5-langmuir O₂ exposure there is a superposition of two oxides, which we designate as Pu₂O₃ and PuO₂. The presence of two oxides is indicated by four peaks at approximately 1.6 and 5.5 eV (Pu₂O₃) and at 2.5 and 4.6 eV (PuO₂). Plutonium-related features, primarily 5f in character with some 6d, appear in the 0-3 eV energy interval, while oxygen features attributed to emission from 2p atomic levels appear in the 4-8 eV energy range.

Results of gas dosing of the delta-plutonium surface with oxygen (O₂) at 77 kelvin.

Further O₂ exposure causes the gradual disappearance of the Pu₂O₃ peaks and the gradual growth of the PuO₂ peaks. This suggests that the PuO₂ is in fact growing on top of the initial Pu₂O₃ layer, and even following an exposure of 20 langmuirs there is still some evidence of a residual Pu₂O₃ component. When the sample was allowed to recover to room temperature and left undisturbed in a vacuum of 6x10^-11 torr for one week, it appears that all of the oxide within the PES probe depth may have converted to Pu₂O₃. However, a Fermi edge (EF) reappeared following this latest reaction period in vacuum. This could also suggest that rather than a conversion of the oxide, the oxygen on the surface has either desorbed into the vacuum (not probable given the extremely high affinity of plutonium for oxygen) or diffused into the metal leaving only a thin Pu₂O₃ oxide layer (between 1 and 5 langmuirs) at the surface. It is also possible that the reduction to Pu₂O₃ resulted in the formation of oxide islands on the surface, therefore exposing areas of the metal substrate within our probe depth and resulting in spectral intensity at the Fermi edge.

PES results following O₂ gas dosing at a temperature of 77 K support an idealized model for the formation of Pu₂O₃ at the delta-plutonium metal surface with subsequent PuO₂ formation on the initial oxide. This is schematically indicated by the development and growth of an oxide layer on top of plutonium metal. Holding the sample at room temperature in a vacuum for an extended period of time (ten hours or more), results in the remaining surface oxide eventually converting to Pu₂O₃. The reappearance of a Fermi edge in the data after the sample was left in vacuum for one week and allowed to warm up to room temperature suggests that the oxygen at the sample surface diffused into the sample.

Oxidation of plutonium metal showing growth of a Pu₂O₃ overlayer. Plutonium atoms are shown in blue and oxygen atoms shown in red. The base plutonium metal is shown as a polycrystalline solid.

It is also possible that the reduction of the oxide caused island growth; thereby exposing some of the metal surface within the PES probe depth. The PuO₂ data at 77 K and the Pu₂O₃ at 300 K both demonstrate that these oxides are insulators with no photoemission intensity at the Fermi edge. Our previous PES results for O₂ gas dosing at room temperature support a model in which a Pu₂O₃ is formed immediately at the metal surface and persists with increasing exposure.

These systems have been studied theoretically using a new generation of density functional theory (DFT) termed hybrid DFT. The details of these approaches are beyond the scope of this article but basically involve combining the exact, nonlocal, Hartree-Fock exchange interaction with the traditional local density approximation (LDA) or generalized gradient approximation (GGA) and correlation interactions of density functional theory.

In their initial use, hybrid functionals were designed and mathematically constructed to faithfully predict the heats of formation of a set of small molecules, e.g. hydrogen (H₂), carbon dioxide (CO₂), ammonia (NH₃), water (H₂O), etc., but researchers quickly discovered that they performed quite well outside this set. In particular, they provide excellent bond energies and properties for molecules containing transition metal and actinide centers. These hybrid functionals have had a dramatic impact in molecular quantum chemistry. They have significantly improved our predictive capability for bond energies, our ability to predict reaction barriers and understand mechanisms, and our ability to predict excitation energies and oscillator strengths in molecules via linear response theory.

Only recently have they been applied to solids with full periodic boundary conditions. Preliminary results suggest they largely remedy two long-standing failures of conventional density functional theory: the prediction of metallic versus insulating behavior in the class of materials known as Mott insulators and the systematic underestimation of the band gap in semiconductors and insulators.

The calculations for UO₂, Pu₂O₃, and PuO₂ using the hybrid DFT provides a substantive improvement in the current state-of-the-art for calculating insulating actinide materials. Within the framework of the hybrid DFT we realize the desirable solution of an insulator for all three materials.

The total density of states for the PuO₂ and Pu₂O₃ ferromagnetic states computed in the PBEO approximation (black) compared with experiment (red).

The UO₂ and Pu₂O₃magnetic solutions are also in agreement with experiment. While the antiferromagnetic solution for PuO₂ in the hybrid DFT is not evident from the available susceptibility and neutron data, it has been suggested that the discrepancy between the low-lying states observed in the neutron scattering data and the temperature dependence of the susceptibility can be understood in terms of an antiferromagnetic interaction between the metal sites. We believe this common framework from which to compute and compare insulating actinide oxides provides a significant step forward in the understanding of these materials.

These improvements are of paramount importance for the study of actinide oxides because the conventional density functional theory approaches predict them to be metals, when in fact they are insulators with significant band gaps. On the other hand, we recently reported the first implementation of hybrid density functional theory capable of describing periodic solids containing f-elements and applied it successfully to the electronic structure of uranium dioxide (UO₂). This approach correctly predicted the anti-ferromagnetic, insulating ground state observed experimentally. In contrast, conventional LDA and GGA approaches both find a ferromagnetic, metallic ground state. The band gap, lattice constant, bulk modulus, photoemission spectrum, and optical spectrum were all in good agreement with experiment. Encouraged by this, we have extended this study to PuO₂ and Pu₂O₃.

In the ionic limit, formal charges for plutonium in PuO₂ and Pu₂O₃ are +4 and +3, respectively, corresponding to formal populations of f4 and f5. These configurations lead to local S=2 and S=5/2 plutonium moments, which can couple with other sites in either a ferromagnetic or anti-ferromagnetic manner. We find the anti-ferromagnetic solution to be the ground state in either case. However, the magnetic coupling is relatively weak and, aside from the energy, most properties are identical between the two states. In what follows, we therefore focus on the ferromagnetic solution.

The lattice constant of PuO₂ is predicted to be 5.39 angstroms (Å), which is in agreement with the experimental result of 5.40 Å. The unpaired spin density on the plutonium center is similar to the estimate from the ionic limit, yielding 4.1 and 5.1 electrons for PuO₂ and Pu₂O₃, respectively. Both actinide oxides are predicted to be insulators, with gaps of 2.4 and 2.5 eV, respectively.

The partial density of states for the PuO₂ and Pu₂O₃ ferromagnetic states computed in the PBEO approximation. Note that both oxygen 2p and plutonium 5f orbitals contribute significantly to the levels near the Fermi energy in PuO₂, implying significant mixing. By way of contrast, the peak near E_F in Pu₂O₃ is predominantly plutonium 5f, with much less oxygen 2p character.

The theoretical PuO₂ photoemission spectrum shows a weak feature near the Fermi edge followed by a strong peak at about 2.5 eV and a weaker one near 4.5 eV. It is encouraging that these peak positions are in good agreement with the experiment. There is, however, an additional feature near 6 eV is observed in the experiment, but is absent in the calculation. In addition, the relative intensity of the 2.5- and 4.5 eV peaks is not well reproduced by the calculation. It should also be noted that we have implicitly assumed identical photoemission cross sections for the oxygen 2p and uranium 5f orbitals in making this comparison. At 40.8 eV incident photon energy, the intensity ratio of the oxygen 2p photoemission peak to the plutonium 5f photoemission peak of approximately 1.5 is probably more appropriate. This adjustment would tend to increase the intensity of the 4.5 eV peak, but not sufficiently to make it stronger than the low-energy one.

For Pu₂O₃, the peak near the Fermi edge is principally plutonium 5f in character and the higher binding energy feature is mostly oxygen 2p. This spectrum is similar to that of UO₂, which also exhibits two distinct peaks of mostly uranium 5f and oxygen 2p parentage. On the other hand, the region near the Fermi edge in PuO₂ is nearly a 50:50 mixture of plutonium 5f and oxygen 2p, signifying much greater metal-ligand mixing.

This is a surprising result, given the smaller overlap anticipated in plutonium because of the smaller radius of the plutonium 5f orbital. In addition to the reduction in 5f radius in going from uranium to plutonium, one expects a stabilization of the plutonium 5f site orbital energy. In perturbation theory, the mixing between two levels is effectively given by the overlap integral divided by the difference in site energies. It may be that the origin of the stronger mixing is related to the stabilization of the plutonium 5f orbital and an unanticipated degeneracy with the oxygen 2p level. In this hypothesis, the small mixing in the Pu₂O₃ spectrum would be associated with the higher plutonium 5f site energy expected in the less highly charged, formally Pu³⁺, ion.

While there is clearly much to be done, we are encouraged by the argument between theory and experiment and believe the combination will tell us much about the electronic structure of these complex systems.

This article was contributed by Martin Butterfield, Los Alamos Neutron Science Center; Tomasz Durakiewicz, Materials Science and Technology Division; and Richard Martin, Theoretical Division.

Contributors to the experiment include Martin Butterfield, Los Alamos Neutron Science Center; John Joyce, Elzbieta Guziewicz, Aloysius Arko, and Kevin S. Graham, Materials Science and Technology Division; and David P. Moore and Luis Morales, Nuclear Materials Technology Division.

Contributors to theory include Ionut Prodan, José Sordo, Konstantin Kudin, and Gustavo Scuseria, Rice University, Houston.

For further reading: "Computational Studies of Actinide Chemistry," P. Jeffrey Hay and Richard L. Martin, Los Alamos Science Number 26 Volume II, 2000.

"Surface Chemistry of Pu Oxides," J.D. Farr, R.K. Schulze, and M.P. Neu, Journal of Nuclear Materials, 328 (2-3), 124-136, 2004.

"Photoemission of Surface Oxides and Hydrides of Delta Plutonium," M.T. Butterfield, T. Durakiewicz, E. Guziewicz, J.J. Joyce, A.J. Arko, K.S. Graham, D.P. Moore, D. P., and L.A. Morales, Surface Science, 571 (1-3), 74-82, 2004.

"Hybrid Density-Functional Theory and the Insulating Gap of UO2," Konstantin N. Kudin, Gustavo E. Scuseria, and Richard L. Martin, Physical Review Letters 89(26), 266402/1-266402/4, 2002.

NMT | LANL | DOE
Phone Book | Search | Help/Info

L O S  A L A M O S  N A T I O N A L   L A B O R A T O R Y
Operated by the University of California for the US Department of Energy