The safe disposal of high-level nuclear waste relies on its confinement in deep geological formations. Whether as vitrified waste or as spent fuel, the waste matrix is foreseen to be encapsulated in thick-walled steel containers and surrounded by various barriers. Several countries, such as United States and Germany, are considering saline formations as a potential rock formation to host such a repository. In the case ground water moves through the barriers and reaches the steel containers, corrosion processes will take place. Corrosion will limit container integrity, and will also produce secondary Fe-phases which can provide an additional barrier against radionuclide migration to the far field. Knowing the corrosion rate and mechanism will greatly help improving the Safety Case. Detailed knowledge on metallic corrosion processes under highly saline conditions is rather limited [1]. The goal of this study is to precisely determine the corrosion rate and mechanism under anaerobic, highly saline conditions of an alloyed steel (AST 309S) used to construct canisters containing vitrified waste.

Polished coupons were contacted with saturated NaCl brine for 6 months at 90°C under strictly anoxic conditions in a closed vessel. At the end of the experiment, the analysis of the brine showed a slight increase in pH whereas E_h hardly changed, and the experimentally determined mean weight loss (8 µm/a) is typical of stainless steel (e.g. [2]). Analysis of the surface by SEM and XPS revealed low and localized corrosion damage, with the presence of pits surrounded by distinct zones. The inner zone is enriched in Cr₂O₃, Ni(OH)₂ and chromite (FeCr₂O₄) whereas Cr and Ni containing Fe hydroxide phases (i.e., green rust) along with trevorite (NiFe₂O₄) are present in the outer zone. The presence of these phases at the corroded surface is further supported by µ-Raman spectroscopy. Results agree with the separation of anodic surface within the cavity and cathodic part outside, and the formation of hydroxide phases (e.g., green rust) at the edge of the pit cavity where dissolved metallic cations from the interior mix with the alkali part from outside. Data also consistently point at a decrease in Cr(III) and an enrichment in Fe(III) in the corrosion products when moving away from the pit. Information on elemental distribution and speciation of Cr, Fe and Ni as a function of depth was provided by synchrotron-based µXRF and µXAS on polished cross cut samples. These elements are not distributed homogeneously within the corrosion layer: regions of higher Cr content have lower Fe and Ni contents, and vice versa. µXANES spectra are consistent with the presence of phases detected at the surface by SEM, namely Cr₂O₃, chromite, green rust containing Cr and Ni substituting for Fe, and trevorite. Unfortunately, the polishing caused smearing thereby rendering very complicated the determination of the precise location of specific phases within the corrosion layer. Similar steel corrosion experiments were conducted in saturated MgCl₂ brine, but hardly any corrosion damage could be detected on these coupons. This finding confirms the importance of the type of salt on steel corrosion.

The corrosion rate is non-negligible but the container integrity will be more severely limited by pit formation. Furthermore, the formed corrosion products contain substantial amounts of Cr and Ni substituting for Fe. The presence of these elements may possibly affect the retention properties of corrosion phases compared to the pure Fe substances. Uptake studies with radionuclides on Cr and/or Ni containing steel corrosion products are planned and will help improving the Safety Case of deep repositories.

This work received funding from the German Federal Ministry for Economic Affairs and Energy (BMWi) under contract number no.02E11496B. We acknowledge the synchrotron light source SOLEIL (Saint-Aubin, France) for provision of synchrotron radiation beam time and Solenn Réguer for assistance during measurements at the DiffAbs beamline.

References

[1] B. Kienzler, F&E-Arbeiten zur Korrosion von Endlager-Behälterwerkstoffen im INE. KIT Scientific Report 7729, KIT Scientific Publishing, Karlsruhe, Germany (2017).

[2] Mundhenk N., Huttenloch P., Kohl T., Steger H., Zorn R. Geothermics, 46, 14-21.