Home | Legals | KIT

Nuclear waste disposal – Examples of services

Zr K edge XAFS investigation on Zr-doped hydrotalcite
Nuclear waste
Figure 1: Perspective view of the hydrotalcite structure. AL/Mg are green, O red and the CO3 are located in the grey plane.

From leaching experiments with metallic uranium-aluminium research reactor fuel elements in repositoryrelevant MgCl2-rich salt brines, a Mg-Al-layered double hydroxide with chloride as interlayer anion is identified as a crystalline secondary phase component. In view of the final disposal of irradiated research reactor fuel elements, the ability of this component to retard mobilized radionuclides is under investigation. In addition to sorption studies, the incorporation behavior of zirconium into the lattice structure of the Mg-Al-Cl-hydrotalcite is investigated. Zr is chosen in this study as a non radioactive analogue for tetravalent actinides.
The Zr-doped hydrotalcite is structurally (XRD) and analytically characterized. The elemental analysis of the sample reveals the presence of Cl and the following formula can be derived: Mg3Al0,96Zr0,09 (OH8)] Cl0,9(CO3)0,050 2,69 H2O. Cl and H2O are supposed to be located between the MgAlZr (OH)8 layers (Fig.1). Zr K X-ray absorption fine structure (XAFS) spectra are recorded at the ANKA-INE-Beamline. Spectra are energy calibrated to the first inflection point in the XANES of a Zr foil (17.998 keV) measured simultaneously. The Zr-doped hydrotalcite XAFS is recorded at room temperature in transmission-mode using Ar-filled ionization chambers at ambient pressure. Ge<422> crystals are used in the monochromator, operating in fixedexit mode. The incident intensity is held constant by means of a piezo-driven feedback system. The parallel alignment of the crystal faces is detuned to ~70% of the maximum beam intensity.
EXAFS fits are performed with Artemis, part of the Ifeffit package, using phase and amplitude data calculated for a 9 atom cluster derived from the undoped hydrotalcite structure. The cell parameters are increased to account for the larger size of the Zr cation. Single path scattering files for phase and amplitude are used for the second coordination sphere. The k-range used for the fit is [2.33 – 13.93Å-1] and fits are performed in the R-space on the k2-weighted data.
The data are well reproduced using 3 shells within the FT range [1.1 – 3.4Å] as shown Fig.2. The first coordination sphere contains 5 O atoms at 2.13Å with a Debye-Waller factor (σ2) of 7.0 10-3 Å2 and 1 Cl atom at 2.45Å with a σ2 of 9.8 10-3 Å2. The presence of a Cl atom was not expected in the first coordination shell but no satisfactory fit can be achieved without adding the chlorine atom.

Understanding the difference between Ce (III) and Ce (IV) retention on clay minerals: a Polarized EXAFS study
XANES spectra for Ce in references (black, solid) and sorbed on hectorite (blue, dotted) or montmorillonite (red, dashed).XANES spectra for Ce in references (black, solid) and sorbed on hectorite (blue, dotted) or montmorillonite (red, dashed).
Figure 1: XANES spectra for Ce in references (black, solid) and sorbed on hectorite (blue, dotted) or montmorillonite (red, dashed).

Scientific background and Aims of the experiment Cerium exists under oxidation states of +III or +IV in the stability domain of water, and the coexistence of these oxidation makes this element of central interest in both nanotechnology and radioactive management. For example, oxidized Ce (IV) can form small colloids or nanoparticles used in nanotechnology, but serious concern exists as to the release and dispersion of these particles in the environment, e.g. as a result of material breakdown, or washing. Nanoparticles may then be hazardous to bacteria and soil living organisms [1], but this hazard may be mitigated by Ce (IV) retention or reduction by soil reactive particles such as clay minerals. Also, Ce (III) can be used as an analogue for trivalent actinides, but in addition the Ce (III)/Ce (IV) redox couple is a unique surrogate for Pu (III)/Pu (IV). Cerium can thus be used to understand the specific interactions between natural solids and Pu at distinct oxidation states, and its impact on the migration of this radioelement in natural systems.
Hence, for both nanotechnological and nuclear environmental reasons, a fundamental understanding of the oxidation state impact on the interaction between Ce (III), Ce (IV) and natural solids is warranted.
Preliminary chemical experiments were performed to quantify the relative affinity of Ce (III) and Ce (IV) on montmorillonite to clarify the direct role played by oxidation state of Ce. Cerium (III) sorption was observed only at pH > 6, likely resulting from the interaction between Ce (III) and layer edges of clay particles. In contrast, Ce (IV) is retained at pH as low as 4, and the sorption increases with pH, even in the presence of carbonate forming dissolved complexes of the lanthanide. This distinct macroscopic behaviour hints at distinct molecular mechanisms of Ce (III) and Ce (IV) retention at the solid surface, but these mechanisms are unknown. The purpose of the study was to identify these mechanisms as a preliminary step of investigating Ce (III) oxydoreduction on prepared (i.e., pre-oxidized and prereduced) suspensions.
Cerium LIII-edge XAFS spectra collected on Ce (III)- and Ce (IV)- reacted hectorite and montmorillonite revealed unexpected redox behavior, as obvious from the shapes of the X-ray absorption near-edge structure (XANES) spectra (fig. 1). The XANES spectra collected for Ce (III) and Ce (IV) sorbed on hectorite all lacked the strong white line at 5226 eV which is a signature of trivalent lanthanides, as obvious for aqueous Ce°(III). Instead, they all yielded the characteristic double-peak profile of Ce tetravalent state. This suggests that Ce interaction with hectorite promote oxidation of the lanthanide. Also, the XANESspectra of sorbed Ce appears to be distinct from that of CeO2(cr), suggesting distinct environments. This inference is confirmed by the absence of strong Ce contributions on Fourier transforms. In contrast, Ce sorbed on montmorillonite remains at the oxidation state of +III, and the apparent angular dependence of polarized EXAFS spectra (Fig 2) revealed a seemingly anostropic environment suggesting direct coordination to montmorillonite platelets. Surprisingly, however, the XANES for Ce (IV)-sorbed montmorillonite is intermediate between that of Ce (III) and Ce (IV) (fig. 1), suggesting a partial reduction of sorbed Ce (IV). Simulation of the absorption edge using Ce(III)-sorbed montmorillonite and solvated Ce(IV) cation showed that a significant fraction (30% mol.) of sorbed Ce (IV) was reduced. This reduction was clearly unexpected as no reducing species other than, perhaps, montmorillonite structural cations, were known to be present in our system. It may however compare with oxidoreduction behaviors observed for Fe (II)/Fe (III) cations sorbed at the edges of montmorillonite. To better understand the origin of these spectral dissimilarities, the (pH, Eh) values of clay suspensions were monitored under anoxic conditions. These measurements confirmed that Ce (IV) was at equilibrium under the hectorite redox conditions, whereas Ce (III) was stable under the conditions
measured for montmorillonite. Future solution chemical experiments carefully following the (pH, Eh) upon introduction of the sorbent species will help to fully understand the redox process in these two systems. The combination of solution measurement and possibly in situ spectral characterization will clarify the nature of the unexpected redox reactions observed in these very simple clay suspensions and preclude the possibility of artifacts due to sample preparation. Once these reactions are identified, the next step will be to check whether the redox properties of clay minerals can be modified prior to Ce addition. Ultimately, these experiences will help to understand the fate of Ce (IV) nano particles in soil systems, and to answer key questions about cerium biodisponibility and toxicity in soils and natural environments.