The electronic structure of a solid determines virtually all of its physical properties, excluding those nuclear in nature. It is extremely important to understand the character of the electrons near the Fermi energy (outermost) of the material because these electrons in particular are involved in bonding or conduction and collectively form the valence band of the solid. We have an ongoing effort to measure and understand the electronic structure of plutonium metal including all of its thermal allotropes (a, b, g, d, d', e), and impurity-stabilized d-plutonium. The methods we use include electron photoemission spectroscopy (PES), Auger electron spectroscopy, electron energy loss spectroscopy (EELS), and photon absorption techniques in the visible and x-ray regimes. This article describes the progress of our photoemission and EELS studies of Pu metal with the objective of understanding the electronic structure of this material and the unique changes it undergoes from low temperature to the liquid phase.
Figure 1. UHV scanning Auger spectroscopy/EELS instrument located in the Chemistry and Metallurgy Research building at Los Alamos. This instrument is used for fundamental actinide surface science research, plutonium physical properties measurements, and analytical surface characterization.
The methods we have used are in general sensitive only to the outermost surface (3 to 10 atomic layers) of the solid under investigation. Therefore, it is very important to understand the nature of the surface relative to the bulk of the material, in particular to insure that the surfaces are atomically clean. It is also necessary to distinguish the presence of possible surface reconfigurations that intrinsically alter the surface electronic structure from that of the bulk material. To that end, the sample preparation and spectroscopic measurements were made in an ultra high vacuum (UHV) environment (~10-10 Torr) in order to preserve the integrity of the surface for the duration of the measurements. Figure 1 shows the UHV Auger spectrometer/EELS instrument where some of these experimental measurements take place.
The electron energy loss measurements have shed light on two aspects of the plutonium metal allotropes: electronic reconfigurations at the surface of each allotrope, and changes in electronic environment induced through temperature changes of the pure Pu metal. We have made careful measurements of the EELS spectra from the surfaces of ion-sputter-cleaned and annealed Pu metal. An exhaustive number of sputter-cleaning and annealing cycles were required to attain a surface that was free of contamination (Figure 2). This effort resulted in a surface that would remain pristine, in the area of analysis, to less than 10% of a monolayer of oxygen impurity for a period of two hours, allowing us sufficient time to make high-quality measurements.
Figure 2. A pure plutonium sample in the UHV EELS instrument. The sample is mounted on a 1-inch puck, which sits in the hot/cold stage of the sample manipulator. The sample is illuminated by a green fluorescent glow in the early stages of ion sputter cleaning as a result of the presence of Pu oxide. The snout of the electron gun and the electron energy analyzer can be seen directly above the sample.
The EELS measurements involve bombardment of the Pu surface with 150 to 1000 eV electrons, with subsequent energy analysis of the scattered electrons. The scattered electrons have lost energy through interaction of the surface by a variety of different processes including excitation of plasmons (Figure 3) in the material, and promotion of electrons in atomic plutonium core levels to unoccupied states in the valence band of the solid. Plasmons are collective excitations of the electron gas (conduction electrons) in the material. Analysis of the plasmon response of the material over the range of primary electron energies that were used shows substantial changes as the depth of material sampled by the spectroscopy changes from approximately 8 to 18 Å. This behavior is consistent for each of the Pu thermal allotropic phases and the Ga-stabilized d-Pu phase. This response indicates that the surface is being reconfigured in an electronic, and probably a structural nature, to an estimated depth of 10 Å (about 3 atomic layers) in each case. We believe that this reconfiguration involves a relaxation of atomic positions to something that is approximately a factor of three less dense atomically than the bulk structure. The precise nature of this reconfiguration is not yet understood.
Figure 3. Schematic representation of an incident primary electron undergoing energy losses due to excitation of plasmons. In the general case the electron may lose energy by exciting a surface plasmon upon entering and exiting the solid (maximum of 2 surface plasmon excitations possible), and may lose energy by exciting any integral number of bulk plasmon losses.
In our measurements of plutonium, the intention is to obtain a value for the free electron count per atom in the solid and to examine how this value differs for each allotropic phase. In the case of EELS measurements at 700 eV primary electron energy, these measurements probe the electronic environment in the bulk (not surface layers) of the Pu metal. We observe up to three bulk plasmon loss events spectroscopically, with peaks evident at approximately 11, 22, and 33 eV loss energy. A compilation of the 700 eV EELS spectra for each thermal allotrope exhibits subtle shifts in the energy of the plasmon resonance peaks as a result of changes in the free electron volume density with the allotrope.
The bulk plasmon resonance energy is used to extract the free electron volume density. The free electron count per atom is calculated by correcting the volume density using the atomic volume density specific for each allotrope. The free electron count per atom appears to be nearly invariant over the range of allotropes, but the subtle changes, however, track exactly with changes in other measured physical properties such as absolute resistivity and magnetic susceptibility. This pattern indicates that the changes in the free electron density measured here are identically responsible for these changes in electronic properties over the allotropic series. These changes, however, do not reflect the traditional view of a shift in 5f electron nature from itinerant to localized with increasing temperature across the allotropic series. Alternatively, the measurements made with this technique probe free electron behavior not associated with the Pu 5f valence electrons, suggesting that the 5f electrons largely do not participate in charge conduction or magnetic properties of Pu. Future measurements and ongoing analysis may clarify these details.
Our photoemission measurements probe the occupied density of states (DOS) in the plutonium and include core level states as well as valence band states (see also Actinide Quarterly, first quarter 1999). This measurement involves absorption of an x-ray or ultraviolet photon by an atom in the solid with corresponding ejection of a photoelectron from the surface of the material.
Through an energy balance, the binding energy of the electron in the solid (as part of the electronic density of states) is related to the measured kinetic energy of the photoelectron and the original photon energy. This is the first step in mapping out the electronic structure of Pu metal, with an ultimate goal of measuring the band structure of each allotrope. In this way we hope to understand the electronic band structure changes that plutonium undergoes with each thermal phase transformation, as well as the electronic mechanism for impurity stabilization of the d-Pu phase. In addition, the unoccupied density of states above the Fermi energy can be examined through the use of near edge x-ray absorption fine structure (NEXAFS) and inner shell EELS (ISEELS) measurements. These are analogous processes in which a core-level electron is promoted into an unoccupied state through absorption of a photon (NEXAFS) or through an inelastic energy loss of a primary electron in the EELS measurements. To a first approximation, within the constraints of the orbital selection rule, the spectroscopic shape of the absorption curve or the loss feature is a convolution of the narrow core level and the shape of the unoccupied states. Thus as shown in Figure 4, we have been able to assemble a complete experimental representation of the electronic density of states from the atomic core levels to the occupied and unoccupied states of the solid valence band.
Figure 4. Experimental electronic structure DOS representation for a-plutonium metal. The occupied valence band DOS was probed by valence band photoemission spectroscopy (VBPES), and the unoccupied DOS was probed by NEXAFS and ISEELS measurements. Some of the deeper core levels below the Pu 6s state are accessible using soft x-ray photoemission (experimental spectrum not shown).
Fundamental studies involving photon and electron interactions with plutonium surfaces continue in our laboratories in an attempt to understand the electronic structure of the metal. An understanding is of paramount importance with regard to the electronic contribution to changes in crystal structure and physical properties with thermal phase changes in plutonium metal, and the nature of the impurity stabilization of the d-Pu phase. An accurate description will help us to understand materials stability, solid-state reactions, and surface chemistry of plutonium metal.
This article was contributed by Roland Schulze (NMT-16). Other researchers on the project are J. Doug Farr, Jeff Terry, and Jeffrey Archuleta (NMT-16); John Joyce and Al Arko (MST-10); Jim Tobin (LLN); and David Shuh and Eli Rotenberg (LBNL).
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
Copyright © UC
- For conditions of use, see Disclaimer