Pore-Scale Modeling of the Surface Roughness Effect on Fluid-Fluid Interfacial Area for Contaminant Transport in Vadose Zone

Abstract

Interaction between solid, wetting fluids and nonwetting fluids frequently occurs in natural environmental processes. An ongoing concern for researchers is the fluid-fluid interfaces on rough solid surfaces inside natural porous media. Interfacial area between the immiscible fluids could be greatly affected by grain surface roughness, in which the adsorbed wetting-fluid films serve as the critical intermediary. It has been demonstrated that the configuration of wetting films is a combination of two competitive surface forces: DLVO adsorption and capillarity, whose effects on wetting fluids can lead to significant changes in the shape of films, and thus distinctive film area under different matric potentials. Therefore, the methodology of the research is to characterize the mechanism of surface roughness involved in the configuration of wetting film, and the resultant change of film area, with an explicit quantitative model. The main body of the present modeling approach is to use a bundle-of-cylindrical-capillaries (BCC) model for pore geometry that is modified with a surface roughness factor based on the solid surface area. Film-associated interfacial area in the model is represented by an interfacial area factor normalized with solid surface roughness, which is quantified by an explicit sigmoid function (logistic function) that defines the change of film area within the range of two limiting conditions: smooth-surface and maximum roughness. For a given porous medium, its inherent solid phase properties, especially the fractal-scale microstructures of surface roughness, will generate a characteristic profile of interfacial-area vs. wetting-fluid saturation, which can be fitted from measured data from interfacial partitioning tracer tests (IPTT). Following the development of modeling approach, simulations with both pre-determined input parameters and actual experimental data were conducted. Example calculations and sensitivity analyses of critical model parameters revealed the phenomenon of “surface roughness masking” that occurred in the interfacial-area vs. saturation curves. Simulation test on experimental data sets for multiple porous media demonstrated the excellent performance of the modeling approach, in which each medium can be explicitly quantified with five critical modeling parameters—two for pore size distribution, one for the sample-scale surface roughness, and two for micro-scale roughness. Inspection of the relationship between roughness-related parameters showed that the micro-scale surface roughness of natural porous media only partially correlate to soil texture. Studies on images from scanning electron microscopy (SEM) also illustrated the complexity of surface roughness. The complicated nature of the micro-scale surface roughness highlighted the potential of the proposed methodology in various environmental applications. It would be particularly useful for systems that comprise large magnitudes of interfacial domain, with energy or mass transport between solid, fluid, and atmosphere.