Retention von Radionukliden (Nb(V), Zr(IV), Ra(II), Sn(II/IV)) an zementartigen Materialien: Quantitative und mechanistische Interpretation

Safety concepts for underground and near-surface nuclear waste repositories rely on a multibarrier
system to limit radionuclide release. Cementitious materials play a central role in both waste
stabilization and construction. Cement undergoes degradation through dissolution and
precipitation of different phases. In the context of near-surface repositories, cement carbonation
by atmospheric CO2 can also impact the long-term stability and retention properties of this
material. The final stage of cement degradation is dominated by calcite (CaCO3) as aggregate
material.
In repositories for low and intermediate level waste (L/ILW), solubility and sorption phenomena
of radionuclides (RN) such as niobium-94 (from the neutron activation of alloyed reactor
components), zirconium-93 (from neutron activation of zircaloy and a fission product of 238U),
radium-226 (from the decay chain of 238U) and tin-126 (a fission product of 235U or 239Pu), are
considered as key processes to accurately assess the retention / mobilization of these RN.
Understanding the behavior of these radionuclides in complex systems consisting of cementitious
phases, pore water, waste matrix, and other components, is of great relevance to assess their longterm
... mehr
behavior. This includes complexing ligands such as isosaccharinic acid (ISA, main
degradation product of cellulose), or high chloride content as groundwater component or from
evaporator concentrates. These components may alter RN speciation or surface properties of
cement, leading to changes in RN retention.
This work focuses on these four radionuclides and investigates their sorption and solubility under
conditions relevant in cement-based repositories. The impact of ISA and chloride on RN retention
is assessed using systematic series of sorption experiments. Experiments were conducted in
carbonated systems (pH ≈ 8.3) for Nb(V), Zr(IV) and Ra(II), and in anoxic conditions with C-SH phases
(pH ≈ 9-13) for Sn(II/IV), to account for conditions relevant to different cement
degradation stages and repository concepts.
The characterization of cement pastes formed by carbonation or alteration revealed a complete
phase transformation to (mostly) calcite, as well as a significantly increased specific surface area
compared to unreacted cement. These solid phases and the associated equilibrated aqueous phases
form the basis of subsequent solubility and sorption studies.
Very low Nb(V) concentrations (≤ 10-8 M) were measured in oversaturation solubility
experiments under weakly alkaline, Ca-containing systems. These values are well below the
solubility of Nb2O5(s), hinting towards the formation of a new solid phase. XRD and Raman
confirmed the predominance of low-crystallinity, small particle-size pyrochlore-type secondary
phases (e.g., Ca2Nb2O7), for which no thermodynamic data is available.
The retention of Nb(V) in calcite systems showed distribution ratios (Rd) ≈ 103 L·kg-1
at t = 3 days. At longer times, Rd values increased to > 2·104 L·kg-1, suggesting a fast adsorption followed
by slower incorporation into the calcite structure. Carbonated cement systems exhibited stronger
uptake (Rd ≈ 105-106 L·kg-1), consistently with literature on fresh cement and C-S-H phases. The
impact of ISA and chloride on Nb(V) retention is relatively minor under near-neutral pH
conditions. The more pronounced effect of ISA previously reported in hyperalkaline conditions
(pH ≈ 13.6) is possibly explained by the formation of stronger Nb(V)-ISA complexes in this pHregime.
Sorption experiments with Zr(IV) targeted carbonated cement materials and fresh cement pastes.
Rd values determined in carbonated cement systems were ≈ 5·105–106 L·kg-1. Similarly, the
uptake of Zr(IV) by cement exhibited Rd values of ≈ 105 L·kg-1. This strong uptake is consistent
with the strong hydrolysis of Zr(IV) and the behavior reported for other hard Lewis’ acids and
strongly hydrolyzing ions, e.g., Th(IV), Sn(IV) or Nb(V). Although similar estimated values were
previously reported in the literature, this is the first experimental confirmation of the strong uptake
of Zr(IV) under these conditions.
The uptake of Ra(II) by calcite exhibited low-to-moderate Rd values (4–6 L·kg-1). Distribution
ratios were monitored for up to 237 days and a constant retention was observed already after 3
days. This can be explained by the rapid surface adsorption or exchange with Ca2+ ions in the
surface layer. Unlike Nb(V), there was no evidence of the possible incorporation of Ra(II) into
the calcite structure. Consistent with these observations, theoretical calculations indicate that
deeper incorporation of Ra2+ into the calcite structure is energetically very unfavorable. Neither
ISA nor NaCl influenced Ra(II) retention, which is explained by the weak interaction of both
anions with Ra(II).
Tin was studied under anoxic conditions in the oxidation states +II and +IV. Solubility in the pH
range ≈ 9–12.8 confirmed agreement between the Sn(II) data and NEA-TDB calculations,
whereas a worse agreement was obtained for Sn(IV). Thermodynamic calculations based on
standard databases predict the full predominance of Sn(IV) within the stability field of water, in
disagreement with experimental observations in this work. This discrepancy aligns with recent
developments by Rai, who claimed a gross error in selected ΔfG0(Sn4+) in reference databases [1].
Sorption experiments on C-S-H phases result in strong uptake even under reducing conditions.
As confirmed by XAFS, this observation can be explained by the oxidative adsorption of Sn(II)
by C-S-H. Adjusting Sn(IV) thermodynamic constants is required to match available sorption
data under anoxic and reducing conditions with the developed CASH+ model. This shifts also the
Sn(IV)/Sn(II) redox borderline towards less reducing conditions, in line with experimental
observations obtained in this study. These results inherently imply that Sn(IV) and Sn(II) could
both be relevant in the reducing, alkaline conditions expected in L/ILW repositories.
This work provides insight into key retention processes (solubility, sorption) for relevant
radionuclides (94Nb, 93Zr, 226Ra and 126Sn) in the context of cement-bases repository systems.
Particular attention is paid to the degradation stage IV of cement, as well as to the influence of
ISA and high chloride concentrations, and the redox behavior of tin under repository relevant
conditions.