Diffusion of dissolved gases in saturated cementitious materials: Does the size matter?

The same diffusion coefficient is commonly used for modelling the transport and durability of cementitious materials regardless the type of diffusing species (i.e. whatever ions, dissolving gases). The question arising is to what extent this assumption is still reasonable without resulting in an over/underestimation of concrete structure performance. This work aims at partly answering this question by measuring the diffusivity of diffusing species with a different molecular size in various cementitious materials with different microstructures. The investigated materials include intact, leached and carbonated saturated cement pastes with varying water to powder ratios (w/p = 0.325-0.425) and limestone filler replacements (0, 10%, 20%). Furthermore, lining concrete and high porous backfill materials (w/c = 1.5) to be used in nuclear waste repositories was also investigated. A recent developed diffusion setup was used to measure the diffusivity of two dissolved gases in a single experiment. The effect of molecular size is quantified by measuring diffusion coefficients of noninteracting dissolved gases with different molecular sizes, including CH4, He, Xe, and O2 (0.280 to 0.432 nm in diameter). Experimental data shows a rather poor relation between the diffusivity and porosity due to the fact that other microstructural parameters (e.g. tortuosity, constrictivity) play a role. Results also prove that the molecular size of the diffusing species influences significantly the constrictivity and thereby the diffusion process, but this impact depends strongly on the microstructure of cementitious materials. The influence of the diffusing species size is less important for high porous cementitious materials.