Research papersNon-Fickian dispersive transport of strontium in laboratory-scale columns: Modelling and evaluation

•Integrated methodology for coupling release and transport of strontium.•Experimental data with Fickian and non-Fickian transport re-evaluated.•Advection-dispersion equation and continuous time random walk examined.•ADE suitable for Fickian transport only; CTRW captures both regimes.•Effects of chemical and physical heterogeneities on transport analysed.

In the context of environmental remediation of contaminated sites and safety assessment of nuclear waste disposal in the near-surface zone, we investigate the leaching and non-Fickian dispersive migration with sorption of strontium (mocking strontium-90) through columns packed with sand and clay. Analysis is based on breakthrough curves (BTCs) from column experiments, which simulated rainfall infiltration and source term release scenario, rather than applying constant tracer solution at the inlet as commonly used. BTCs are re-evaluated and transport parameters are estimated by inverse modelling using two approaches: (1) equilibrium advection-dispersion equation (ADE); and (2) continuous time random walk (CTRW). Firstly, based on a method for calculating leach concentration, the inlet condition with an exponential decay input is identified. Secondly, the results show that approximately 39%-58% of Br− and 16%-49% of Sr2+ are eluted from the columns at the end of the breakthrough experiments. This suggests that trapping mechanisms, including diffusion into immobile zones and attachment of tracer on mineral surfaces, are more pronounced for Sr2+ than for Br−. Thirdly, we demonstrate robustness of CTRW-based truncated power-law (TPL) model in capturing non-Fickian reactive transport with 0 < β < 2, and Fickian transport with β > 2. The non-Fickian dispersion observed experimentally is explained by variations of local flow field from preferential flow paths due to physical heterogeneities. Particularly, the additional sorption process of strontium on clay minerals contributes to the delay of the peak concentration and the tailing features, which leads to an enhanced non-Fickian transport for strontium. Finally, the ADE and CTRW approaches to environmental modelling are evaluated. It is shown that CTRW with a sorption term can describe non-Fickian dispersive transport of strontium at laboratory scale by identifying appropriate parameters, while the traditional ADE with a retardation factor fails to reproduce the complex non-Fickian transport of strontium with strong sorption on clay surface.