JRP CALL information
Supported By

European Commission

Short description of the work

Under HLW repository relevant conditions (i.e., high temperature, anoxic), the corrosion of steel canister will lead to the formation of iron (hydr)oxides such as green rust and magnetite. Such neoformed phases may represent a sink for RN released upon waste matrix corrosion, delaying their migration out of the repository. RN retention may occur either by surface adsorption and/or by structural incorporation. Unfortunately, only limited information is available on the retention of long-lived and radiotoxic actinides by such phases. This project focuses on the trivalent actinides retention by green rust and magnetite.


In a former Actinet JRP (C2-06: Trivalent actinides binding to magnetite) green rust and magnetite were synthesized in the presence of lanthanides and actinides. Whereas the synthesis of magnetite is well reproducible, that of chloride green rust is more complex and by-products were observed. In the first part of this Talisman JRP, extensive lab work was devoted to optimize the synthesis conditions of the highly reactive green rust chloride. For example, an increase in Fe(II):Fe(III) ratio was found to favor the formation of green rust and reduce the amount of Fe2(OH)3Cl formed concomitantly. Furthermore, an increase in pH also favors the formation of magnetite.

The goal of the August stay was the spectroscopic characterization of the local chemical environment of Lu/Am associated to various iron (hydr)oxide phases. Synthesis and sorption operations were conducted under anoxic conditions (Ar filled glove box). Green rust was prepared in the presence of trace amounts of Lu(III) or Am(III) to mimic conditions relevant for a HLW disposal site. Samples were prepared by slowly increasing the pH of a solution containing Fe(II) and Fe(III) considering a ratio of Fe(II):Fe(III) = 7 and either Lu (sample LuCopGR) or Am (sample AmCopGR). The corresponding suspensions were subsequently centrifuged, the supernatant replaced by water before increasing the pH to values >8 to form magnetite (samples LuCopGRtransMag and AmCopGRtransMag). In separate experiments, magnetite was synthesized directly in the presence of Lu (sample LuCopMag) or Am (sample AmCopMag) by increasing the pH of a solution containing Fe(II) and Fe(III) with a ratio of Fe(II):Fe(III) = 0.5. Finally, Lu and Am were adsorbed in separate experiments on pre-existing magnetite (samples LuAdsMag and AmAdsMag, respectively) in suspension (0.1 M NaCl, pH ~ 6). For all samples, pH and Eh values were recorded in situ after an equilibration time of at least 14 h (overnight). The suspensions were centrifuged and the samples placed in sealed containers. The samples were transported to the INE-Beamline at ANKA in order to perform XAS experiments on wet pastes under anoxic conditions. Data were collected at the Lu L2-edge or at the Am L3-edge. The solid phase characterization (i.e., XRD, SEM) and analysis of the supernatant (dissolved Fe and Lu/Am concentrations) is still on-going.

The samples containing Lu all have distinct spectroscopic signatures, meaning that Lu environments in the two samples are dissimilar. The Fourier transform (FT) of the spectra indicate the presence of cationic neighbors beyond the first ligand shell. For example, small differences can be seen between the data of LuCopGR and that of LuCopMag, indicating distinct crystallochemical environments. At first sight, this is consistent with the different structures of green rust (layer double hydroxide or LDH) and magnetite (inverse spinel). Furthermore, the differences in first ligand (i.e., oxygen) and cationic shells indicate that whether adsorbed on magnetite or coprecipitated with magnetite Lu is located at a different site. On-going analysis and fits to the data will provide more insight into the actual binding mechanism(s). No time was available to measure the sample LuCopGRtransMag.

The samples containing Am also exhibit differences in their binding environment. The FT reveals the presence of cationic neighbors upon coprecipitation with green rust in AmCopGR. After conversion to magnetite (sample AmCopGRtransMag), the Am chemical environment evolved: contributions from neighboring Fe atoms increased, as expected for the more compact inverse spinel structure (magnetite) compared to the LDH structure (green rust). Furthermore, the data indicate that the magnetite formation pathway influences the nature of the chemical environment because the cationic contributions in AmCopMag are different from that in AmCopGRtransMag. Specifically, the FT peak corresponding to Fe atoms has a higher amplitude contribution in AmCopMag. Finally, the highest FT peak contribution for neighboring Fe is obtained in AmAdsMag, but also the highest amplitude in first O shell. For surface sorbed Am, more neighboring O atoms can be expected in the first shell than for Am located in the bulk, in agreement with observations. However, more cationic neighbors would be expected for the actinide substituting for Fe, contrary to observations. Yet, Am located in the bulk is located in a more constrained environment and the structural disorder may dampen the Fe contribution compared to surface sorbed Am.

Fits to the data will provide valuable information with regard to the binding environment and thus on the nature of the retention site(s).
These preliminary results are based on the experimental FT, quantitative information will be provided by fits to the data. They will also allow comparing the data of the Lu series to that of the Am series and enable to draw conclusions regarding the influence of the size on structural incorporation.


Main visitor contact data
Name: Dr. Knud Dideriksen
Organisation: University of Copenhagen

JRP Identification
JRP nr: TALI-C01-13
JRP title: Structural elucidation of trivalent actinides retention by iron (hydr)oxides
JRP scope: Scope 2: Actinide in the geological environment

Visited Associated Pooled Facility
Visited APF during the stay: KIT-INE - Beamline
Name of the APF Contact Person: Thorsten Schäfer

Other APF and organisation involved in the JRP
Other organisations involved:
Other APF involved in the project: KIT-INE - Laboratories

Description of the work done at the associated pooled facility
Start date of the stay: 8/17/2014
End date of the stay: 8/23/2014
Quantity of access: 12
Access Unit: Days
Misc.: Work during week-ends

Other APF visitors of the JRP during the stay
Visitor 2: Dr. Michel Schlegel (CEA Saclay)
Visitor 3:
Visitor 4:
Visitor 5: