Anomalous solute transport in saturated porous media: Relating transport model parameters to electrical and nuclear magnetic resonance properties

Abstract

The advection-dispersion equation (ADE) fails to describe commonly observed non-Fickian solute transport in saturated porous media, necessitating the use of other models such as the dual-domain mass-transfer (DDMT) model. DDMT model parameters are commonly calibrated via curve fitting, providing little insight into the relation between effective parameters and physical properties of the medium. There is a clear need for material characterization techniques that can provide insight into the geometry and connectedness of pore spaces related to transport model parameters. Here, we consider proton nuclear magnetic resonance (NMR), direct-current (DC) resistivity, and complex conductivity (CC) measurements for this purpose, and assess these methods using glass beads as a control and two different samples of the zeolite clinoptilolite, a material that demonstrates non-Fickian transport due to intragranular porosity. We estimate DDMT parameters via calibration of a transport model to column-scale solute tracer tests, and compare NMR, DC resistivity, CC results, which reveal that grain size alone does not control transport properties and measured geophysical parameters; rather, volume and arrangement of the pore space play important roles. NMR cannot provide estimates of more-mobile and less-mobile pore volumes in the absence of tracer tests because these estimates depend critically on the selection of a material-dependent and flow-dependent cutoff time. Increased electrical connectedness from DC resistivity measurements are associated with greater mobile pore space determined from transport model calibration. CC was hypothesized to be related to length scales of mass transfer, but the CC response is unrelated to DDMT.

Continuum‐scale characterization of solute transport based on pore‐scale velocity distributions

“…In this context, Swanson et al . [] have shown that relying solely on information about the geometrical attributes of the pore space obtained through NMR and electric resistivity measurements does not lead to a reliable characterization of the full set of model parameters.…”

“…While this continuum-scale model is widely used, quantitative characterization of effective model parameters relying on pore-scale quantities associated with real threedimensional rock samples is still lacking. In this context, Swanson et al [2015] on information about the geometrical attributes of the pore space obtained through NMR and electric resistivity measurements does not lead to a reliable characterization of the full set of model parameters.…”

The Evaluation of Electrical Impedance of Three-Dimensional Fractal Networks Embedded in a Cube

“…33,34 The short relaxation time corresponds to large value of β and small absolute value of phase. [35][36][37] Besides, it seems that the selected values (10-1000 µm) of d 0 and l 0 do not have any remarkable effect on the value of phase. It is demonstrated again that the comparably large size of mother pore channel has minor effect on the ion relaxation.…”

Pore network modeling of the electrical signature of solute transport in dual‐domain media

“…In the PNM, we represent intergranular and intragranular porosity as different classes of pores, class 1 and 2, respectively. Whereas each intergranular connection occurs through a single bond between adjacent pores, intragranular connections are divided between a number ( N B ) of small bonds, consistent with CT scan visualizations of zeolite grains [ Swanson et al ., , ]. Grains are represented as regions of class 2 pores with diameters based on the α SRMT from Briggs et al [] and the formula for the diffusive length scale of a spherical grain, 0.5 cm [ Haggerty et al ., ].…”