H53E-1454
Tomography of underground target elements using CWR
Severe environmental, health, and national security impacts posed by underground contamination and buried explosive devices have created the need to develop cost effective technology to detect and monitor underground target elements (UTE). Cross well radar (CWR) has been applied for detection of target objects that exhibit significant contrast of dielectric properties in soils. Its application for detection of single or distributed elements in heterogeneous underground conditions has yet to be developed. This research addresses the development of a physics-based tomographic model for in situ soil characterization and detection and imaging of UTEs in unsaturated soils by CWR. Experimental work involves transmitting and receiving electromagnetic (EM) signals by an array of antennas in a 2D SoilBED subject to variable water content conditions. The EM response is analyzed using Visual Basic and Matlab Codes based on obtain the S-parameters generated by the network analyzer, estimate the dielectrical permittivities, and create a tomogram of the soil and its characteristics. These tomograms are generated using simultaneous algebraic reconstruction techniques (SART), with enhance convergence parameters. Embedded UTEs are distinguished through comparative simulations of tomographic dielectric properties. Preliminary results indicate that good tomograms can be generated using an array of 5 antennas on each side of a 30 cm x 30 cm two-dimensional setup, and an array of 17 antennas for a 90 cm 90 cm two-dimensional setup.
H53E-1455
Synchrotron X-Ray Microtomography and Interfacial Partitioning Tracer Test Measurements of Napl-Water Interfacial Areas
Interfacial areas between an immiscible organic liquid (NAPL) and water were measured for two natural porous media using two methods, aqueous-phase interfacial partitioning tracer tests and synchrotron X-ray microtomography. The interfacial areas measured with the tracer tests were similar to previously reported values obtained with the method. The values were, however, significantly larger than those obtained from microtomography. Analysis of microtomography data collected before and after introduction of the interfacial tracer solution indicated that the surfactant tracer had minimal impact on fluid-phase configuration and interfacial areas under conditions associated with typical laboratory application. The disparity between the tracer-test and microtomography values is attributed primarily to the inability of the microtomography method to resolve interfacial area associated with microscopic surface heterogeneity. This hypothesis is consistent with results recently reported for a comparison of microtomographic analysis and interfacial tracer tests conducted for an air-water system. The tracer-test method provides a measure of effective, total (capillary and film) interfacial area, whereas microtomography can be used to determine separately both capillary-associated and film-associated interfacial areas. Both methods appear to provide useful information for given applications. A key to their effective use is recognizing the specific nature of the information provided by each, as well as associated limitations.
H53E-1456
Evaluation of Local and Global Segmentation Techniques for Quantitative Analysis of X-Ray CT Images of Geological Materials
Advanced 3-D imaging techniques such as X-ray Computed Tomography (CT) in association with mathematical morphology based analysis can be employed to quantify the geometrical and topological features of pore networks. The first and critical step in image analysis is segmentation to separate pores from the surrounding matrix. Results of all subsequent two- and three-dimensional analyses is strongly dependent on the accuracy of the segmentation step, hence the choice of thresholding technique is of crucial importance. In this study we tested several global and local thresholding techniques for binarization of CT images of geological materials. Geometrical and topological pore network features such as porosity, specific surface area, pore size distribution, tortuosity, and connectivity were determined based on mathematical morphology operations.
H53E-1457
Macropore Flow in Soil Columns: Investigations with Computer Tomography and Lattice Boltzmann Simulations
Soil structure greatly affects the ability of soil to transmit and to retain water, chemicals, and colloidal particles that can carry contaminants or be contaminants themselves, e.g. pathogenic microorganisms. No theory or empirical relationships have been developed to date to quantitatively relate parameters of soil structure and parameters of the contaminant transport in soils. The absence of theoretical advances in this area seriously hampers the ability to address issues of public concern, e.g. spread of contaminants introduced in the environment by agricultural activities. Recently, computer tomography of soils has become available to generate detailed images of soil pore space with high resolution and density. Successful applications of computer tomography in medical and material sciences show the great potential of this technique to create an exhaustive characterization of soil structure heterogeneity. In this presentation we investigate saturated flow through twelve undisturbed macroporous soil columns (7.62- cm sample diameter and 18-cm length) with lattice Boltzmann simulations. Saturated flow was measured for the complete columns, as well as on 2 cm sections for selected columns. Computed X-Ray tomography was performed on each of the columns, using the 420 kV X-ray source of a HYTEC FlashCT high-speed industrial CT scanner. The resolution was 116 microns per voxel, yielding a final tomography image of 656x656x1482 (~ 6.3 10E8) voxels. X-Ray CT observations typically provide "gray-scale" representations of the imaged object that must be segmented to yield discrete pore and particle geometry. Many segmentation algorithms are available, each yielding different final pore geometries thus potentially creating uncertainties in subsequent flow analyses. Lattice Boltzmann (LB) simulations will be presented only for some of the columns as the simulations are extremely computationally intensive (each simulation requires ~ 60 GB of computer RAM at the observed resolution). The main objectives of this presentation are: 1) to define an optimal resolution for the LB simulations in terms of simulated saturated hydraulic conductivity and computational requirements, 2) study the effect of different types of segmentation algorithms on simulated flow, and 3) compare simulated to observed hydraulic conductivities. Successful LB simulations may lead to improved insights into flow patterns within macroporous structures and lead to better understanding of solute and colloid transport.
H53E-1458
The Effect of Roughness on Fluid Droplets between Parallel Plates
Available theories predict that the kinematics of fluid droplets between thinly spaced parallel flat plates is dependent on gravitational, capillary, and viscous forces. Varying the surface roughness of the plates may affect the capillary and viscous forces. It is known that statically, roughness changes the fluid-solid-gas contact angle. Roughness may also affect viscous forces by modifying the internal flow. However, the precise nature of the overall effect of roughness on the parallel plate system is unknown. Experiments were designed to test current theories for predicting droplet speed. The terminal speed of water droplets was measured while varying parameters such as incline and glass plate roughness. Here, it is shown that standard theoretical models do not adequately predict the behavior of fluid droplets in a rough parallel plate system. The purpose of these experiments is to further the understanding of the physical mechanism governing behavior of multiphase flow between rough parallel plates; this is necessary before fluid flow in comparable situations, such as rock fractures, can be correctly modeled.
H53E-1459 [WITHDRAWN]
Numerical and experimental studies of drying and shrinkage induced microcracking in concrete
The aim of this research is to investigate moisture flow in cement paste and Interface Transition Zone, around aggregate, as well as associated shrinkage induced strains, and subsequent microcracking. Two coupled methods are used: numerical simulations and experiments. The moisture flow is numerically simulated by Lattice Gas Automata, while drying experiments are performed in Environmental Scanning Electron Microscope (ESEM) in order to determine drying deformations and shrinkage coefficient. The moisture movement simulation by Lattice Gas Automata, a type of cellular automata, involves different drying collision rules, which results in different density (moisture content) at each node. Special attention is given to the effect of the presence of aggregate particles on the moisture flow in concrete, which are considered rigid obstacles. The shrinkage deformation is considered a linear function of the moisture content if the relative humidity, RH does not exceed 40%. The calculated moisture gradient from the numerical analysis and an assumed shrinkage coefficient are used to calculate drying shrinkage induced strains and stresses using the basic equations. To determine the drying shrinkage coefficient for the RH range 100% to 20%, new experimental techniques of drying in ESEM are used. Small paste samples are cast in a specially developed mould with 2 mm thickness and afterwards carefully grounded and polished to a thickness of required 1 mm. The used variables in the tests are: various cement types, w/c ratio, sample age and curing conditions. Drying shrinkage displacements and strains are determined in order to calculate the drying shrinkage coefficient. These experimentally obtained values of the coefficient are compared with the literature data and used in numerical simulations of shrinkage induced strains, stresses and cracking in drying concrete.
H53E-1460
Salinity Influence on Interfacial Area, Wettability, and NAPL Recovery
Wettability, the tendency of rock or sediment particle surfaces to be preferentially wet by one fluid phase, has a strong influence on the distribution and flow of immiscible fluids in oil reservoirs or aquifers. The efficiency of oil and non-aqueous phase liquid (NAPL) recovery processes and the displacement and production of oil/NAPL by fluids injected into the reservoir or aquifer depend on the wetting properties of the rock/sediment particle surfaces. Effects of salinity on wettability and residual oil saturation during water flooding are of particular interest in the petroleum industry with some reservoirs. It was indicated that the residual oil saturation may be reduced significantly by flooding with low salinity water instead of seawater or brine. This observation may be also true in NAPL recovery from contaminated aquifers. NAPL recovery enhancement may be achieved by manipulating the salinity of the remedial fluid. Two sets of 8 core-flooding column experiments have been completed, using decane and Alaska North Slope (ANS) crude oil as surrogate NAPLs. Unconsolidated sand packs were used as representative porous media. NAPL removal was conducted by flushing column at residual NAPL saturation using water with salinity ranging from 0% to 8% wt of NaCl. The NAPL-water interfacial area (anw, cm-1) was measured and used as an indicator for the wettability characteristics of the packed sand. Sodium Dodecyl Benzene Sulfonate (SDBS) was used as an interfacial partitioning tracer and Pentafluoro Benzoic acid (PFBA) was used as a non-reactive and non-partitioning tracer. NAPL was imbibed into an initially water saturated column, using positive displacement methods. NAPL was then flushed out using water at certain salinity. When the column attained a residual NAPL saturation after each water flushing displacement, the partitioning and conservative tracer experiments were conducted separately, to characterize the specific NAPL-water interfacial areas, and the wettability status. Water with 8%, 4%, 2%, 0% wt NaCl salinity was used to displace NAPL from the sand column sequentially. The interfacial tension (IFT) between the salinity water and the ANS oil was monitored. The residual oil saturations indicated that the fraction of NAPL retained in the column increased after water flushing as the salinity in the displacing water increased from 0 to 8%, clearly confirming the earlier findings that lower salinity may cause additional oil to be released. The NAPL-water interfacial area, anw, does not show a monotonic dependence on salinity; instead, anw shows an increasing trend with increasing salinity in the lower salinity range, and the opposite trend at high salinity values. The maximum anw was obtained in systems flushed with 2% salinity water. This trend appears to be consistent with a similar nonlinear dependence of interfacial tension on salinity, and might be an indication of wettability alternation. The observation of this research shread lights on the optimum operation in NAPL removal. The IFT change between NAPL and the salinity water might be attributed to the enhanced NAPL recovery.
H53E-1461
The Effects of a Shear Fracture on Multi-phase Transport in a Layered Sandstone
This research studies multi-phase transport in a sandstone core in the presence of a shear fracture. The cylindrical sample was cored parallel to bedding and placed in a Hoek cell that provided 10-Mpa confining stress on the sample. The sample was fractured in shear using offset shims to produce a fracture parallel to the axial bedding planes. After the initial breakdown, the fracture was displaced by about one percent of its initial length. The absolute permeability was measured by the Klinkenberg gas permeability method and by vacuum-saturated water flow. The absolute permeability of the fractured rock increased three fold when measured at a hydrostatic stress equal to the confining stress used for fracturing. A seqence of oil floods, water floods and fractional flow experiments were used to obtain relative permeability data in the fractured rock. The effective permeability to oil at residual water was high and the effective permeability to water at residual oil was very low. This indicates that once the large fracture volumes are filled with oil, they contribute only to oil flow and not to water flow. The distribution of oil and water in the large fractured region was established using Micro-Computed Tomography. This distribution confirms that most of the oil cannot be displaced out of the fracture and forces the water to flow in the matrix, thus yielding very low effective water permeabilities. Water-oil steady-state fractional flow trajectories were stable and confirmed a limited mobile saturation range in the fracture.
H53E-1462
Measurement of Thin Film Characteristics in Porous Media
Liquid film formation in unsaturated subsurface porous media is commonly not accounted for in flow and transport models. Characteristics of this type of film flow are complex to quantify and are therefore often overlooked, despite the apparent need for a more complete low saturation flow model in fields such as hazardous waste disposal, enhanced oil recovery, and environmental contamination situations. The most important effects of film flow in porous media relate to available interfacial areas, meniscus curvature development, and permeability; greater understanding of these effects is necessary to accurately model these low saturation flow situations. In order to examine this topic further, we have created two-dimensional, pore scale flow cells containing crushed Yucca Mountain tuff. Tuff is characterized by large surface areas and intra-granular porosity which encourages thin film formation. Using these flow cells, relative humidity sensors, and a digital microscope and image processing software, the saturation and capillary pressure (as relative humidity) inside of the cells are measured. The measured values allow us to estimate the saturation point at which Van der Waals forces no longer dominate and capillary forces become the controlling factor. This point is characterized by the formation of pendular rings from condensing thin films. By introducing water into the cell and subsequently drying it, the formation of thin films and pendular rings can be observed while relative humidity and temperature are recorded. Relative humidity and temperature are then related to capillary pressure and saturation using the Kelvin and Young-LaPlace equations. Using this data and imaging analysis, the transition point from pendular rings to film- dominated flow can be related to the saturations and capillary pressures.
H53E-1463
Effects of Low-Conductivity Regions Near Grain-to-Grain Contacts on Early Colloid Breakthrough at the Column Scale
Effects of fluid velocity (2 to 60 m day-1) on bulk transport of carboxylate-modified microspheres (5.0μm) was investigated using saturated sand (450μm)-packed columns and pore-scale colloid movement was observed in glass bead (3mm)-packed flow-cells. At all flow rates, the colloid peak emerged from columns approximately 0.1 pore volume earlier than the dissolved conservative tracer, suggesting a consistent pore volume from which colloids were excluded. Retention of colloids within columns increased with decreasing fluid velocity. Visualization of colloid movement in pseudo-3D pore networks did not indicate presence of preferential transport paths, as would be expected for size exclusion. Direct observation, however, revealed zones of reduced hydraulic conductivity near grain-to-grain contacts. Fewer than expected colloids entered zones of reduced conductivity and subsequently returned to faster flow streams or meandered until becoming associated with a solid-water surface. We investigate whether colloid exclusion or removal from low-conductivity regions and advection of remaining colloids in reduced pore volume can explain the observed early breakthrough of colloids.
H53E-1464
Quantifying thick liquid films and their role in evaporative drying of porous media
Evaporation rate from porous media is determined by a combination of driving forces and external conditions interacting with liquid and vapor within complex pore spaces. The evaporation rate during first stage of a drying process is controlled primarily by atmospheric demand and generally is not limited by medium transport properties. Liquid connections between the evaporation surface and the receding drying front sustain sufficient water supply to maintain a constant evaporation rate. When hydraulic connections are disrupted, water transport to the surface becomes limited to rates supported by vapor diffusion. To improve our understanding of the properties of stage-one supporting liquid connections, we delineated pore geometry and liquid configuration in sand samples imaged using synchrotron X-rays tomography. As air invades large pores, remaining liquid in crevices and grain contacts form a network of thick films. We compared evaporation rates with liquid phase configuration above the drying front in an attempt to relate the end of the first stage of drying with liquid films connectivity. The dependency of high evaporation rates on residual liquid continuity implies sensitivity to surface wettability properties. These effects were examined using different mixtures of hydrophobic and hydrophilic particles and their impact on drying rates and liquid phase distribution. Improved understanding of relationships between pore scale effects, liquid configuration and evaporation processes enhances predictability of drying rates and may enable alterations or design of porous media with prescribed drying behavior.
H53E-1465
Analytical and Lattice Boltzmann Predictions of Intrinsic Permeability for Deterministic and Randomized Fractal Porous Media
The Menger sponge is a three-dimensional mass fractal that provides a good approximation of the pore space geometry of natural porous media. We present new analytical expressions for the intrinsic permeability (k) of both classical (deterministic) and randomized sponges. The analytical expressions are compared with estimates of k derived from Lattice Boltzmann simulations of saturated water flow in classical and randomized Menger sponges. Analytical model results compared favorably with Lattice Boltzmann simulations of k for deterministic and random Menger Sponges. However, Lattice Boltzmann simulations showed that the mean k from 100 realizations of randomized sponges was 50% less than the k of the deterministic sponge. Likewise, the analytical model predicted the k of randomized sponges were 87% less than that of deterministic sponges.
H53E-1466
Spatial Distribution of Biomass in Porous Media
Current understanding of subsurface microbial biofilm formation and their impact on fluid hydrodynamics is limited by our ability to observe the microscale geometry of developed biofilms. Biomass distribution in porous media has been observed previously in only two dimensional systems; currently, no high-resolution 3- dimensional datasets exist that give sufficient information about microbial distribution such that the impact on flow and transport at the microscale can be directly computed. Three dimensional biofilms can significantly alter pore flow velocities and overall mass transfer between the aqueous and biological phases. We are currently developing new methods to resolve high-resolution 3-dimensional tomographic images of biofilms in porous media using synchrotron based x-ray microtomography. Imaging biofilms without disturbing their natural spatial arrangement has been a challenging task, primarily because most conventional dopants that dissolve in water also easily diffuse into biofilms. One method that we have developed to overcome this problem is the addition of silver nanoparticles to the fluid phase. Using this approach, we have been able to differentiate between the biomass filled pore space and fluid filled pore space. To date, the images that we have collected have yielded good representations of the geometry and qualitative information about structures of biomass. Ultimately, we intend to combine this kind of experimental measurement with upscaling (via volume averaging) to determine how biofilms might alter the physical properties of the porous media. Ultimately, by quantifying the spatial distribution of biofilms we will gain a greater understanding of how changes in physical parameters may impact the rate at which microbes degrade contaminants or produce products, and therefore this research has applications to bioremediation and bioprocessing.
H53E-1467
Using optical index matching to visualize solute transport and perform particle tracking in a synthetic porous medium
Optical visualization techniques are heavily used in experimental fluid mechanics. In flow studies addressing natural objects (fractured or porous rocks), they are difficult to utilize as they often require transparent solid objects. Even if the solid objects in the experimental setup happen to be transparent to light, the existence of complex boundaries between the fluid and the surrounding solid phase leads to distorsion of the images through light refraction at the interfaces, unless the solid and fluid are nearly perfectly matched in optical index. This is in particularly true for porous media, due to the many optical interfaces present in the system. We present here an experimental setup addressing the visualization of solute transport and particle advection in a synthetic porous medium consisting of glass beads. The fluid is a mixture of benzol and ethanol mixed in a proportion that ensures a good an optical index matching as possible between the fluid and the beads. At fluid preparation, we monitor the index matching by measuring the transmission of a laser beam through a box containing a sample porous medium. The matched fluid allows monitoring of solute dispersion inside the porous medium. The influence of the index mismatch on the uncertainty of the measurements has been estimated using available theories. Successful attempts at tracking colloidal particles advected inside the porous medium are also presented. The simple tracking algorithm requires that particles not too close to each other be monitored at a large enough acquisition frequency For analog experiments in which a transparent porous medium can be used, this technique can be a reasonable choice with respect to more elaborate and expensive techniques such as X-ray tomography or MRI.
H53E-1468
Complete Immiscible Liquid Dissolution in Natural Porous Media
Immiscible liquids have proven to be a lasting source of subsurface contamination at many hazardous waste sites. Understanding the transport and fate behavior of these contaminants will allow for better site characterization and determination of applicable remediation technologies. Experiments were conducted to investigate immiscible liquid dissolution behavior in several natural porous media, comprising different particles size distributions. Immiscible liquid dissolution was investigated at the column scale by conducting experiments using homogeneously packed columns (7cm long) containing a residual saturation of trichloroethene. Nonideal dissolution behavior was observed, wherein secondary steady state dissolution regions occurred. Such nonideal behavior was not observed for a well-sorted silica sand. Imaging studies using synchrotron X-ray microtomography were conducted at Argonne National Lab, Argonne, IL to characterize the pore-scale morphology of the immiscible liquid as a function of dissolution.
H53E-1469
Characterizing Air-Water Interfacial Area for Variably Saturated Porous Media
The critical role of the air-water interface in variably saturated porous media systems has received increased attention in recent years. The air-water interface is recognized to influence multi-phase flow, control interfacial retention of chemical contaminants, pathogens, and colloids, and mediate various mass-transfer processes. Unfortunately, examination of air-water interfacial phenomena and the potential impact of system properties and conditions has been constrained by a lack of means by which to measure air water interfacial areas. Imaging methods combined with computer-assisted tomography have been used for some time to examine the structure of porous media, as well as the configuration and distribution of fluids residing in porous media. The resolutions associated with these methods have, until recently generally been insufficient to quantitatively characterize fluid-fluid interfacial areas, especially for natural porous media. However, this impediment has been resolved by the availability of synchrotron X-ray sources. In this study, air-water interfacial areas as a function of water saturation were measured for several natural porous media using synchrotron X-ray microtomography. As expected, total (capillary-and film-associated) interfacial area increased with decreasing water saturation for all media. In addition, the magnitude was greater for the media with smaller median particle diameters. Capillary-associated interfacial area was also examined.
H53E-1470
Comparing Pore-scale and Macro-scale Capillary Pressure Measurements Using a Two- dimensional Micromodel
Capillary pressure plays a critical role in multiphase flow and transport in porous media. At the pore scale, capillary pressure is defined by Laplace's law which states that capillary pressure is a function of surface tension, contact angle and curvature. This study focuses on imaging and estimating pore scale properties that determine capillary pressure. Drainage and imbibition experiments for a NAPL-water system are conducted in a two-dimensional micro-scale porous medium. High resolution images of the phase distributions and associated interfaces within the pores are collected during the experiments. Images are taken at a rate of approximately 50 frames per second with a resolution between 1-10 ėm per pixel. In addition, the pressure in each phase is measured with a transducer outside the porous medium, and pressure-saturation curves are plotted from the data. We will attempt to use Laplace's Law to estimate the average pressure inside the porous medium based on measured curvatures. The two pressure values (measured outside the system versus calculated via Laplace's Law) will be compared. The images will allow for investigation of pore scale properties during dynamic flow conditions, as well as static conditions, and importantly, allow for comparison among the two situations. Specifically, relaxation of menisci interfaces and resulting changes in interface curvature, and thus capillary pressure, will be correlated to variations in system properties such as fluid-fluid viscosities and flow rates.
H53E-1471
Numerical Modeling of Unsaturated Flows in Variable Gravity During Parabolic Flight
Parabolic flight experiments were conducted to study effects of reduced gravity on multiphase fluid distribution and transport. Notwithstanding the limited duration of microgravity (~20 s), measurements of porous-media fluid behavior have been successful in demonstrating significant differences between μ- and 1-g. Further understanding of reduced gravity effects can be gained through numerical modeling of hydrodynamic data. The gravitational acceleration during parabolic flight cycles between hypergravity (1.8-g) and microgravity (~10-6-g). Impacts of variable gravity on measurements focusing on the microgravity portion of the flight were ambiguous and difficult to interpret. One-dimensional numerical modeling using the Richards equation with a variable gravity term was compared with matric potential and water content measurements obtained during several parabolic flights. Introducing a time-dependent variable gravity term facilitated modeling of the hypergravity phase, which extends to 1.8-g and precedes each microgravity cycle. This complete' treatment of flight data allowed more accurate modeling of secondary water retention scanning curves. This is important because during parabolic flight, wetting and draining processes occur simultaneously in different volumes of the porous medium. Both baked clay aggregates and glass beads were packed into containers with heights varying from 1 to 7 cm. Hydrostatic and matric potentials were measured using micro-tensiometers and water content was determined either volumetrically or using TDR. Hydrus-1D was used to model the hydrodynamics with time- dependent gravity input in sub-second increments of time. Our results suggest that the impact of a preceding hypergravity-phase on microgravity hydrodynamics during parabolic flight should not be ignored and requires due attention for adequate modeling of matric potential and water content measurements in porous media.
H53E-1472
A new method of generating a pore-network model for representing porous media
In this study, we have developed a new method of generating a pore-network model for representing a porous medium. The method is based on the standard cubic lattice network and thus it has two elements: pore bodies located at the lattice points and pore throats connecting the pore bodies. One of the main features of our network is that pore throats can be oriented in 13 different directions. As a result, the coordination number of pore bodies may vary from 0 to 26, with a pre-specified average value for the whole network. We have applied this method to reconstruct a real sandstone utilizing information on coordination number distribution. Various network properties, such as number of pore bodies and pore throats, average coordination number, and coordination number distribution agree very well with observation data. This method can be especially useful in studying the effect of structure and coordination number distribution of pore-network models of solute transport and multiphase flow in porous media systems.