A novel laboratory method for determining the diffusive mass flux of dense non-aqueous phase liquids (DNAPLs) into natural clays

In this study we developed a novel laboratory method, which aimed to robustly determine the in-situ aqueous diffusive mass fluxes of solutes derived from an organic solvent DNAPL into a natural clayey deposit. The method included the emplacement of an undisturbed clay core in a diffusion cell apparatus and the position of 1 L dichloromethane (DCM) as DNAPL on top of the water-saturated clay core. The DCM dissolution and diffusive mass flux into the clay was then determined in two different ways by 1) directly measuring DCM solvent volume loss with a capillary tube installation and 2) measuring and simulating the resulting DCM concentration profile after 15.07 days of diffusion transport with sorption into the undisturbed clay core. At the end of the experiment, the DCM penetrated the clay core a distance of 9 cm showing a typical transient diffusion concentration profile. The DCM diffusion profile was then simulated by using the ratio of the effective diffusion coefficient and the retardation factor (D_e/R) as fitting parameter. Additionally, R was determined from the best-fit source concentration (m_w,0) and the known solubility limit of DCM (C_S) using the relation R = m_w,0/C_S resulting in a value of 1.19. This enabled the calculation of the effective diffusion coefficient (D_e) based on the best-fit D_e/R, yielding D_e = 3.5·10^-10 m²/s. Moreover, based on the estimated best-fit parameters, a DCM mass of 445 mg was estimated to have diffused into the clay core that was consistent with the capillary tube readings and literature-derived parameter values. The consistency between capillary tube reading and diffusion profile shows that diffusive fluxes into a water-saturated clayey sediment could be reliably determined by capillary tube readings only, facilitating the experimental procedure for estimating diffusive fluxes into clayey deposits. Overall, the novel developed laboratory method allows for the determination of the effective diffusion coefficient and retardation factor with concentration gradients typical of DNAPL-contaminated sites as well as their verification under in-situ conditions in independent ways (diffusion profile and capillary tube reading). This enhances the robustness of the estimated effective diffusion coefficient under in-situ field conditions.