AVS2013 Session AC+AS+EN-TuA: Actinides and Rare Earths: The Nuclear Fuel Cycle and Critical Materials

Electronic structure of uranium materials have been studied by means of photoelectron spectroscopy (PES) as well as angle-resolved PES (ARPES) using synchrotron radiations. The results on the nuclear fuel material UN as well as uranium heavy Fermion superconductors UPd₂Al₃ will be presented.

Uranium monoride (UN) is a promising fuel material for the generation-IV advanced nuclear reactors since it has high melting point, a very good thermal conductivity at high temperatures as well as a high fuel density. We have revealed its electronic structure by ARPES, and found that they are well explained by the band structure calculation based on local density approximation (LDA) [1].

UPd₂Al₃ is a heavy Fermion superconductor. It shows transitions into the antiferromagnetic phase at T_N=14 K and into the superconducting phase at T_SC=2 K. Since U 5f electrons in these compounds show both pronounced localized and delocalized characters, the nature of U 5f electrons have been not well understood. Their detailed electronic structure was derived by ARPES by synchrotron radiation as well as very high resolution PES by leaser. It was found that quasi-particle bands with strong U 5f character form Fermi surface at low temperatures, suggesting that U 5f electrons have itinerant nature in this compound. Furthermore, we have succeeded to observe the superconducting gap of this compound by PES [2]. This is the first observation of the superconducting states of actinide materials by PES.

[1] Shin-ichi Fujimori et al., Phys. Rev. B 86, 235108 (2012).

[2] Y. Ohta et al., in preparation.

Uranium metal absorbs H readily at low pressures, forming a fine highly pyrophoric powder of β-UH₃, which is ferromagnetic with T_C ≈175 K. More technologically important than the orthorhombic α-U is the bcc γ-U, known as existing at high temperatures or stabilized by doping to low temperatures.

In the course of the study of fundamental properties of Mo-stabilized bcc U alloys [1] we have been testing their resistance to H. There was no interaction with H₂ at ambient pressure and room temperature. However, increasing pressure over approx. 4 bar a slow hydrogenation was observed after several days if incubation. Increasing the H₂ pressure makes the hydrogenation faster. The product, in which the H concentration corresponds to the 1U-3H ratio, does not turn into powder, but consist of large brittle fragments with metallic luster. X-ray diffraction indicates that the hydride is amorphous, with a basic structure pattern corresponding to β-UH₃, while the grain size is ≈1 nm only.

The hydrides obtained were subjected to studies of magnetic properties, electrical resistivity, and specific heat. The most striking fact is that the Curie temperature can be higher than in β-UH₃. It reaches T_C = 200 K for UH₃Mo_0.18, for both lower and higher Mo concentrations it weakly decreases. Also the magnetization values increase from 0.93 μ_B/f.u. to 1.2 μ_B/f.u. (both in μ₀H = 14 T). At low temperatures a striking high coercivity, reaching 4 T, is seen from magnetization loops. It can be understood as due to a high anisotropy randomly distributed over spatial directions.

The Sommerfeld coefficient of electronic specific heat γ = 29.2 mJ/mol K² is similar to β-UH₃ values (29-33.2 mJ/mol K² by various authors) and higher that pure UMo_0.18 (18.8 mJ/mol K²). Electrical resistivity exhibits the weakly negative slope known from the bcc U-Mo alloys [2] and other strongly disordered systems with a cusp-like anomaly at 200 K superimposed. The absolute values about 1 mΩcm are about an order of magnitude higher than for UMo_0.18, and even higher than that of UH₃, which exceeds 600 μΩcm at room T [3]. The most surprising fact is that the 5f magnetism, which is normally suppressed by disorder, is supported in the case of UH₃Mo_x, probably due to enhanced U-U spacing due to Mo included in the lattice.

This work was supported the Czech Science Foundation under the grant No. P204/12/0285.

[1] I. Tkach, N.-T.H. Kim-Ngan, S. Mašková, M. Dzevenko, L. Havela, J. Alloys Comp. 534 (2012) 101.

[2] B.S. Chandrasekhar, J.K. Hulm, J. Phys. Chem. Solids 7 (1958) 259.

[3] J.W. Ward, L.E. Cox, J.L. Smith, G.R. Stewart, J.H. Hood, J. Phys. 40 (1979) C4-15.

Thermal water splitting to hydrogen on reducible materials is one of the promising methods to secure renewable energy vectors for the future. the process relies on a redox cycle whereby steam is introduced to a prior reduced CeO₂ to generate hydrogen which is consequently oxidized. However to achieve this cycle CeO₂ needs be reduced at elevated temperatures (> 1500^oC) [1, 2]. Mixing CeO₂ with metal cations can lower the energy needed for the reduction process. Among the methods of changing CeO₂ are the incorporation of metal cations smaller in size than Ce⁴⁺ cations such as Zr⁴⁺ [3] or the addition of a metal cation that can transfer electrons to Ce⁴⁺ [4, 5]. In this work we present a core and valence level study of the extent of reduction of Ce⁴⁺ cations by the addition of Fe and U cations. We compare XPS Ce3d, Ce4d of stoichiometric CeO₂ to those of Ce_xFe_1-xO₂ and Ce_xU_1-xO₂ (x<1). We also investigate the reduction of these materials upon Ar ions sputtering. We find that the addition of Fe or U cations in any proportions increases the reduction of Ce⁴⁺ cations. In particular considerable enhancement of the reduction of CeO₂ with Ar ions was noticed (compared to CeO₂ alone) when x>0.5 in both Ce_xFe_1-xO₂ and Ce_xU_1-xO₂ series of the oxide materials. The reasons for the enhancement of CeO₂ reduction by Fe cations can be linked to both size and electron transfer effect while in the case of U cations it is mainly due to electron transfer from the U5f to Ce4f levels. Tests for hydrogen production by thermal water splitting for both series were conducted and indicated the enhancement of the activity of the mixed oxide when compared to CeO₂ alone.

[1] W.C. Chueh, C. Falter, M. Abbott, D. Scipio, P. Furler, S. M. Haile, A. Steinfeld, Science (2010) 330: 1797-1801.

[2] K-S Kang, C-H Kim, C-S Park, J-W Kim, J. Ind. Eng. Chem. (2007) 13: 657-663.

[3] S. Abanades, A. Legal, A. Cordier, G. Peraudeau, G. Flamant, A. Julbe, J Mater Sci (2010) 45:4163–4173.

[4] I. Al-Shankiti, F. Al-Otaibi, Y. Al-Salik, H. Idriss, Topics in Catalysis (2013) in press.

[5] B.E. Hanken, C.R. Stanek, N. Grønbech-Jensen, M. Asta, Phys. Rev. B (2011) 84: 085131-1 to 085131-9.

Developing a clear understanding of how metal oxide electronic structure changes for a range compounds and materials will greatly benefit a variety of existing and emerging energy technologies. Many of the technologically desirable chemical, magnetic, electronic, and thermal properties of metal oxides are derived from strongly covalent metal–oxygen multiple bonds (metal oxos). Among approaches explored previously, ligand K-edge X-ray absorption spectroscopy (XAS) has emerged as an effective method for quantitatively probing electronic structure and orbital mixing. The presence of covalent mixing is observed as a pre-edge feature in the ligand K-edge XAS, which only has transition intensity if the final state metal orbital contains a component of ligand p orbital character. Recent advances have shown that insights regarding the nature of orbital mixing in metal oxides can be obtained at the K-edge for oxygen through a combination of XAS with a scanning transmission X-ray microscope (STXM), non-resonant inelastic X-ray scattering (NIXS), and hybrid density functional theory calculations (DFT). The spectroscopic work in this study was performed at the ALS Molecular Environmental Sciences beamline 11.0.2 (STXM), beamline 6.2 at SSRL (NIXS), and the LERIX facility at the APS (NIXS).

Herein, a new effort is discussed that employs these techniques to understand bonding interactions in d- and f-block oxides. Oxygen K-edge XAS measurements and DFT studies began with a series of six tetrahedral oxyanions, MO₄^2- and MO₄^1- (M = Cr, Mo W and Mn, Tc, Re). Despite the similarities of the isoelectronic d⁰ MO₄^2- and MO₄^1- anions, unexpected differences in metal oxo orbital mixing were observed for adjacent metals in the periodic table. The lanthanide dioxides and sesquioxides, LnO₂ and Ln₂O₃ (Ln = Ce, Pr, Tb), were chosen for subsequent work because their electronic structures are well-established from hard X-ray spectroscopies. Features in the O K-edge XAS follow anticipated trends based on 4f and 5d orbital energies and occupancies. Taken together with L₃-edge intensities determined previously, a detailed picture of the electronic structure in lanthanide oxides emerges. Overall, the research shows that orbital composition is influenced by a complex interplay between periodic changes in both orbital energy and radial extension.

This work was funded by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences, under contracts DE-AC02-05CH11231 at LBNL and DE-AC52-06NA25396 at LANL. Operation of the ALS, SSRL, and the APS is supported by the U.S. Department of Energy, Office of Science.

References

[1] M. H. Mintz, J. Bloch, Prog. Solid St. Chem. Vol. 16, (1985) 163-194.

[2] K. H. Gayer, W. G. Bos, J. Phys. Chem. Vol. 68, No. 9, (1964) 2569- 2574.

[3] G. W. McGillivray, J. P. Knowles, I. M. Findlay, M. J. Dawes, J. Nucl. Mater. 385 (2009) 212-215.