Tritiated Water Diffusion in Fresh and Carbonated Cement Pastes: Effect of a drying/resaturation cycle
Résumé
In the French radioactive waste management context, cementitious materials will be widely used as engineered barriers, backfill materials or confinement matrices for some Intermediate-Level Long-Lives waste. In a waste disposal facility, during the construction and the operation phases, the ventilation of underground galleries/shafts will impose specific conditions (relative humidity, presence of CO2) that will modify the degree of water saturation (Sw) of cement materials (surfacial drying) and induce a potential carbonation. The aim of this study is to investigate the effect of a drying/resaturation cycle on the effective diffusion coefficient of tritiated water (HTO) in fresh and carbonated cement pastes.
Samples used are disks (2 mm thick) of a CEM V/A hardened cement paste (HCP) prepared with a water to cement ratio of 0.43 and cured at least 27 months in an artificial cement pore water (ACW ; pH~13.5) [1]. The drying stage consisted of storing HCP samples (during several months at 20±1°C under Ar atmosphere) at fixed relative humidity (RH) imposed by saturated saline solutions. Thus, initial Sw values obtained range from 0.63 to 0.85. Afterward, HCP samples were resaturated with ACW (under vaccum). Additionally, a fully saturated sample was used as a reference. Carbonated samples were prepared in conditions representative of atmospheric carbonation [2]. An accelerated carbonation process was performed in a climatic chamber (PCO2 = 3% vol ; RH=55 % in order to obtain a maximal carbonation rate). Through-diffusion experiments were performed, in at least duplicate, using standard two-reservoir set-ups. The effective diffusion coefficient, De(HTO) and material capacity factor, were obtained by modelling the HTO cumulative activity curves (in the downstream reservoir) with a routine based on an analytical solution of Fick laws (with fixed initial and boundary conditions).
Deref(HTO) value, obtained for the reference sample is 3±1 10-13 m2.s-1 which is in agreement with literature data for this materials [1,3]. For fresh HCP samples, values are similar to water accessible porosity (0.33±0.02). No significant HTO retention is then assumed in HCP samples.
For fresh HCP, the drying/resaturation cycle results in a slight non-linear increase of De(HTO) values as initial Sw values diminish. This result is consistent with the reported description of the drying effect on HCP microstructure [4]. For high initial Sw (0.85), water is mostly drained from the capillary porosity with no specific effect on HCP microstructure. As this process is reversible the final De(HTO) value (after resaturation) is similar to Deref(HTO). For intermediate initial Sw (0.63), the water molecules, distributed as multilayers at C-S-H particles surfaces, partly desorb and cause small changes in the mesopore distribution by opening/closing pores (MIP measurements) which ease HTO diffusion through the pore network. The impact on De(HTO) value is an increase by a factor of 2 (5.4±0.4 10-13 m2.s-1).
For carbonated HCP samples, the drying/resaturation cycle results in decreasing the water accessible porosity (0.21±0.01) and increasing of De(HTO) value (12±2 10-13 m2.s-1). These results are consistent with a combination of porosity clogging (precipitation of CaCO3 in pores) and presence of microcracks (decalcification of C-S-H) as described in Auroy et al study [2]. The increase of De(HTO) value supports the conclusion that for CEM V HCP, microcracks could be the main pathway for diffusion of HTO in carbonated samples.
Comparing to the reference sample (fully saturated non-carbonated HCP), these results suggest that HTO diffusion is favoured during a drying/resaturation cycle. For a blended cement, this effect may be enhanced if coupled with carbonation.
Acknowlegment
The research leading to these results has received funding from the European Union’s Horizon 2020 Research and Training Programme of the European Atomic Energy Community (EURATOM) (H2020-NFRP-2014/2015) under grant agreement n°662147 (CEBAMA)
[1] Savoye S., Macé N., Lefèvre S., Spir G., Robinet J.C., Applied Geochemistry 96 (2018) 78-86
[2] Auroy M., Poyet S., Le Bescop P., Torrenti J.M., Charpentier T., Moskura M., Bourbon X., Cement and Concrete Research, 74 (2015) 44-58
[3] Landesman C, Macé, N., Radwan J., Ribet S., Bessaguet N., David K., Page J., Henocq P., 3rd International Symposium on Cement-based Materials for Nuclear Wastes (NUWCEM), October 24-26, 2018 (Avignon, France)
[4] Roosz C., Gaboreau S., Grangeon S., Prêt D., Montouillout V., Maubec N., Ory S., Blanc P., Vieillard P., Henocq P., Langmuir, 2016, 32, 6794-6805