Grantee Research Project Results
Final Report: Theoretical Evaluation of the Interfacial Area between Two Fluids in Soil
EPA Grant Number: R827116Title: Theoretical Evaluation of the Interfacial Area between Two Fluids in Soil
Investigators: Bryant, Steven
Institution: The University of Texas at Austin
EPA Project Officer: Aja, Hayley
Project Period: October 1, 1998 through September 30, 2001 (Extended to November 30, 2002)
Project Amount: $246,378
RFA: Exploratory Research - Physics (1998) RFA Text | Recipients Lists
Research Category: Land and Waste Management , Air , Safer Chemicals
Objective:
All mass transfer between phases occurs through an interface; in particular, such transfer is the mechanism by which nonaqueous phase liquids (NAPLs) contaminate groundwater. Some remediation techniques also rely on interphase mass transfer to remove the NAPL, some of its components, or to introduce an agent to mobilize or degrade the NAPL. Certain allied technologies such as interwell partitioning tracer tests depend on interphase mass transfer of an oil- and water-soluble compound to assess the volume of NAPL residing in the subsurface. The rate at which chemical species move between phases is a critical factor in assessing contamination risks and in designing remediation strategies. This rate, in turn, is directly proportional to the surface area of the interface between the phases.
Interfacial area is clearly a fundamental parameter for making reliable, science-based assessments of risk and cleanup at NAPL-contaminated sites. However, direct measurement of the interfacial area between two immiscible fluids in a soil or rock is very difficult, especially when each fluid phase is connected (capable of flowing). Indeed, nondestructive measurement techniques, such as magnetic resonance imaging and interfacial tracers, have been introduced only very recently. Measurements reported to date exhibit considerable variation. Therefore, a means of assessing the interfacial area independently would be of both theoretical and practical interest.
The objective of this research project was to develop a novel mathematical modeling technique to predict the interfacial area from first principles. The key to this development was the use of an “ideal soil:” a random dense packing of equal spheres. The location of every sphere in the packing is accurately known. Though a highly simplified approximation, this model porous medium nevertheless is physically representative of a key geometrical feature of real soils; namely, the disordered but strongly spatially correlated locations of the soil grains. Because the interfacial area depends strongly on the geometry of the pore space confining the fluids, this approach allows the determination of physically realistic fluid configurations from which areas and volumes can be computed.
Summary/Accomplishments (Outputs/Outcomes):
We have obtained reasonable validation of a novel method for predicting the area of the interface between two immiscible fluids in soils. The method yields a priori predictions of the fluid phase configurations during capillary-controlled displacements, from which areas and volumes can be calculated. No adjustable parameters enter the calculations. In this work, the method was applied only to the ideal soil model (a dense random packing of equal spheres), but its success in this simple case indicates that it could be worthwhile to extend it to model more complicated soils (distribution of grain sizes, shapes, wettability, etc.).
Our calculations show that the area of the interface between macroscopic volumes of immiscible fluids in an ideal soil is relatively small. In particular, it is significantly less than the surface area of the grains comprising the soil. In many environmental processes, the grains will retain a film of wetting phase even when a nonwetting (NW) phase, such as a NAPL, is present in the pores. If a measurement technique is sensitive to the interface between a NW phase and a thin film of wetting phase, it will not be possible to distinguish the area of the film from the area of macroscopic fluid volumes. Depending on the mass-transfer process of interest, the distinction between the two types of interface can be important. Thus, the results of measurement techniques described in recent literature should be interpreted carefully before being applied.
A macroscopic volume of fluid need not be connected to the bulk phase, and this research project has provided the first estimates of the relative contributions of isolated (trapped) volumes and connected volumes to fluid/fluid interfacial area. Again, this distinction may be important depending on the application. For drainage processes, the relative contributions are strongly dependent on the degree of connectivity of the wetting phase. In essence, the connectivity corresponds to the time scale during which drainage occurs. In principle, wetting phase that is connected to the bulk only via thin films on grains can be displaced, given sufficient time for movement through the films. For poor connectivity (short time scales), isolated wetting phase makes a significant contribution to the total area even at moderate wetting phase saturations. For higher connectivity (long time scales), such that wetting phase can be isolated only as pendular rings at grain contacts, the contribution of isolated volumes to interfacial area is negligible except at irreducible wetting phase saturation.
Predictions of total wetting/NW phase interfacial area during drainage (i.e., the sum of contributions from isolated and bulk volumes and from the wetting film on grains in drained pores), agree with the trend of experiments reported in the literature from several sources and from different methods involving interfacial tracers. Because there are no adjustable parameters in the predictions, we conclude that, in general, aqueous interfacial tracer techniques applied in water-wet systems are sensitive to wetting films. Therefore, interfacial areas inferred from such measurements are likely to significantly overestimate the area between macroscopic volumes (any morphology other than films) of immiscible fluids. Similarly, predictions of total fluid/fluid interfacial area during imbibition agree with the trend of measurements reported in the literature, although the data exhibit more variability than the drainage experiments.
Values of interfacial area one or more orders of magnitude larger than the nominal grain surface area have been reported from some measurement techniques. The conclusion from our modeling is that such values must be dominated by grain surface roughness; at the capillary pressures prevailing during most experiments, there is no physically plausible configuration of fluid phases that can give such large values.
Predictions of the irreducible volume fraction of wetting phase (the endpoint of a drainage process) agree with laboratory measurements reported in the literature when a locally defined criterion for “intermediate” connectivity of the wetting phase is used. This criterion allows for the trapping of pendular rings and liquid bridges at grain contacts, of lenses in pore throats, and of “islands” in single pores. We conclude that this criterion gives a reasonable approximation of the conditions under which the experiments were conducted. Moreover, this provides an independent check of the physical relevance of the modeling approach, and increases our confidence in the predicted interfacial areas. Similarly, predictions of residual NW phase saturations are consistent with values reported for unconsolidated media.
Because coalescence of pendular rings leads to the closure of pore throats to NW phase, it diminishes NW phase connectivity and increases the likelihood of entrapment of NW phase. This qualitative argument has led to the conventional wisdom that “snap-off” of NW phase in pore throats is a key feature of the imbibition process and strongly affects the value of residual NW phase saturation. Our simulations with the geometrically determined coalescence criterion indicate that in an ideal soil, the influence of snap-off is almost negligible. This is readily explained by the fact that the values of coalescence curvatures are smaller than the values of curvature for imbibition of individual pores. Thus, by the time the capillary pressure (current curvature) has been reduced enough to initiate coalescence, much of the packing has already been imbibed. The geometry of grain contacts in consolidated media will be different than in the ideal soil; thus, determining whether this conclusion is relevant to sedimentary rocks requires further research.
An important overall conclusion is that we have obtained reasonable validation of a novel method for predicting the area of the interface between two immiscible fluids in soils. Its success in the simple cases considered in this research project indicates that it could be worthwhile to extend it to model more complicated soils (e.g., distributions of grain sizes and shapes). The predictions involve no adjustable parameters and use a physically representative model soil so that the method can provide quantitative insight into the complex phenomenon of pore-level fluid configurations.
The most significant practical implication of this research project is that the area of the interface between macroscopic volumes of immiscible fluids in simple porous media is relatively small compared to the surface area of the grains comprising the porous media. Thus, some measurement techniques, such as injection of interfacial tracers, may be insensitive to the area of macroscopic volumes of fluid. Therefore, measurements of interfacial area should be interpreted carefully before being using in mass-transfer models.
Journal Articles on this Report : 5 Displayed | Download in RIS Format
Other project views: | All 15 publications | 6 publications in selected types | All 5 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Bryant SL. Some mathematical and computational problems in reactive flow. Computational Geosciences 2001;5(3):203-223. |
R827116 (Final) |
not available |
|
Bryant SL, Anna J. Bulk and film contributions to fluid/fluid interfacial area in granular media. Chemical Engineering Communications 2004;191(12):1660-1670. |
R827116 (Final) |
not available |
|
Bryant S, Johnson A. Wetting phase connectivity and irreducible saturation in simple granular media. Journal of Colloid and Interface Science 2003;263(2):572-579. |
R827116 (Final) |
not available |
|
Gladkikh M, Bryant S. Prediction of interfacial areas during imbibition in simple porous media. Advances in Water Resources 2003;26(6);609-622. |
R827116 (Final) |
not available |
|
Jain V, Bryant S, Sharma M. Influence of wettability and saturation on liquid-liquid interfacial area in porous media. Environmental Science & Technology 2003;37(3):584-591. |
R827116 (Final) |
not available |
Supplemental Keywords:
remediation, nonaqueous phase liquid, NAPL, soil, chemical transport, physics, modeling, network modeling, porous media, multiphase flow., RFA, Scientific Discipline, Air, Toxics, Waste, Mathematics, Physics, Chemistry, HAPS, chemical mixtures, Groundwater remediation, Engineering, Chemistry, & Physics, fate and transport, soil , porus media, NAPL, chemical transport modeling, interfacial phenomena, mass transfer, interwall partitioning tracer tests, groundwater contamination, mathematical formulations, NAPLsProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.