Competition between delocalization and spin-orbit splitting in the actinide5fstates

Abstract

Synchrotron-radiation-based x-ray absorption, electron energy-loss spectroscopy in a transmission electron microscope, multielectronic atomic spectral simulations, and improved first-principles calculations ͑general-ized gradient approximation in the local density approximation͒ have been used to investigate the electronic structure of the light actinides: ␣-Th, ␣-U, and ␣-Pu. It will be shown that the spin-orbit interaction can be used as a measure of the degree of localization of valence electrons in a material. The spin-orbit interaction in the light actinide metals ␣-Th, ␣-U, and ␣-Pu, has been determined using the branching ratio of the white line peaks of the N 4,5 edges, which correspond to 4d → 5f transitions. Examination of the branching ratios and spin-orbit interaction shows that the apparent spin-orbit splitting is partially quenched in ␣-U, but is strongly dominant in ␣-Pu. These results are fully quantified using the sum rule. This picture of the actinide 5f electronic structure is confirmed by comparison with the results of electronic structure calculations for ␣-Th, ␣-U, and ␣-Pu, which in turn are supported by a previous bremsstrahlung isochromat spectroscopy experiment.

Nature of the5fstates in actinide metals

“…In other words, the delocalization of the 5f states in α-U reduces the angular part of the spin-orbit interaction, even though there is indeed a strong radial part of the spin orbit interaction (van der Laan et al, 2004;Tobin et al, 2005). It is important to note that if the spin-orbit value falls on the jj-curve, then the atom has jj-coupling, because this is the only way to create a large spin-orbit interaction.…”

On the electronic configuration in Pu: spectroscopy and theory

“…The return to an LS van der Laan and Thole [25], which has been used to explain the cross-sectional behavior of the lighter actinides. [12,18] Before going onto a discussion of photoelectron spectroscopy, it is useful to consider the possibility of phase specific variations of the XAS/EELS and thus the n value for Pu. As can be seen in Figures 3 and 4, within each technique, the results for a and d are quite similar for the 5d to 5f transitions.…”

“…This result is independent of any particular computational model of On the electronic configuration in Pu: spectroscopy and theory 10/10/06 schemes for spin-orbit splitting or the calculation of x-ray absorption cross section and thus does not depend upon the details of the branching ratio analysis presented previously. [12,18] II. Coupled with the result in I above, the absence of a pre-peak in the Pu 5d XAS and EELS indicates that n must be at least as great as 5.…”

“…The equivalence of the XAS and high-energy EELS measurements has already been demonstrated for Ce and Pu, as well as other actinides. [12,[16][17][18][19] Now, the argument against n = 6 will be discussed in a stepwise fashion.…”

Experimental Benchmarking of Pu Electronic Structure

“…Thus, we have embarked upon the utilization of unorthodox electron spectroscopies, to circumvent the problems caused by the absence of large single crystals of Pu with well-defined surfaces. Our approach includes the techniques of resonant photoelectron spectroscopy [1], x-ray absorption spectroscopy [1,2,3,4], electron energy loss spectroscopy [2,3,4], Fano Effect measurements [5], and Bremstrahlung Isochromat Spectroscopy [6], including the utilization of micro-focused beams to probe single-crystallite regions of polycrystalline Pu samples. [2,3,6] …”