Search

Author(s): Chang, Chun; Kneafsey, Timothy J; Wan, Jiamin; Tokunaga, Tetsu K; Nakagawa, Seiji

How the wettability of pore surfaces affects supercritical (sc) CO2 capillary trapping in geologic carbon sequestration (GCS) is not well understood, and available evidence appears inconsistent. Using a high-pressure micromodel-microscopy system with image analysis, we studied the impact of wettability on scCO2 capillary trapping during short-term brine flooding (80 seconds, 8 to 667 pore volumes). Experiments on brine displacing scCO2 were conducted at 8.5 MPa and 45°C in water-wet (static contact angle θ = 20° ± 8°) and intermediate-wet (θ = 94° ± 13°) homogeneous micromodels under four different flow rates (capillary number Ca ranging from 9 × 10−6 to 8 × 10−4) with a total of eight conditions (four replicates for each). Brine invasion processes were recorded and statistical analysis was performed for over two thousand images of scCO2 saturations, and scCO2 cluster characteristics. The trapped scCO2 saturation under intermediate-wet conditions is 15% higher than under water-wet conditions under the slowest flow rate (Ca ∼ 9 × 10−6). Based on the visualization and scCO2 cluster analyses, we show that the scCO2 trapping process in our micromodels is governed by bypass trapping that is enhanced by the larger contact angle. Smaller contact angles enhance cooperative pore filling and widen brine fingers (or channels), leading to smaller volumes of scCO2 being bypassed. Increased flow rates suppress this wettability effect.

We present an experimental study of pore-scale flow dynamics of liquid CO2 and water in a two-dimensional heterogeneous porous micromodel, inspired by the structure of a reservoir rock, at reservoir-relevant conditions (80 bar, 21 °C). The entire process of CO2 infiltration into a water-saturated micromodel was captured using fluorescence microscopy and the micro-PIV method, which together reveal complex fluid displacement patterns and abrupt changes in velocity. The CO2 front migrated through the resident water in an intermittent manner, forming dendritic structures, termed fingers, in directions along, normal to, and even opposing the bulk pressure gradient. Such characteristics indicate the dominance of capillary fingering through the micromodel. Velocity burst events, termed Haines jumps, were also captured in the heterogeneous micromodel, during which the local Reynolds number was estimated to be ∼21 in the CO2 phase, exceeding the range of validity of Darcy's law. Furthermore, these drainage events were observed to be cooperative (i.e., across multiple pores simultaneously), with the zone of influence of such events extending beyond tens of pores, confirming, in a quantitative manner, that Haines jumps are non-local phenomena. After CO2 completely breaks through the porous section, shear-induced circulations caused by flowing CO2 were also observed, in agreement with previous studies using a homogeneous porous micromodel. To our knowledge, this study is the first quantitative measurement that incorporates both reservoir-relevant conditions and rock-inspired heterogeneity, and thus will be useful for pore-scale model development and validation.

A new algorithm for imposing the contact angle on solid surfaces is proposed in the Lattice Boltzmann color-gradient model. The capability and accuracy of this algorithm are validated by simulation of contact angles for a droplet resting on a flat surface and on a cylindrical surface. The color-gradient model with the proposed contact angle algorithm is then used to study the capillary valve effect in porous media. As a preliminary study, the capillary valve effect is explained by simulating immiscible two-phase displacement within a single-pore geometry. It is shown that the capillary valve effect is accurately captured by the present simulations. Further simulations of drainage and imbibition are also conducted to understand the capillary valve effect in an experiment-matched pore network micromodel. The simulated results are found to agree quantitatively with the experimental results reported in literature, except for a few differences which result from the exclusion of contact angle hysteresis in the proposed algorithm.

The experimental measurement of water retention curve in hydrate-bearing sediments is critically important to understand the behavior of hydrate dissociation and gas production. In this study, tetrahydrofuran (THF) is selected as hydrate former. The pore habit of THF hydrates is investigated by visual observation in a transparent micromodel. It is confirmed that THF hydrates are not wetting phase on the quartz surface of the micromodel and occupy either an entire pore or part of pore space resulting in change in pore size distribution. And the measurement of water retention curves in THF hydrate-bearing sediments with hydrate saturation ranging from Sh = 0 to Sh = 0.7 is conducted for excess water condition. The experimental results show that the gas entry pressure and the capillary pressure increase with increasing hydrate saturation. Based on the experimental results, fitting parameters for van Genuchten equation are suggested for different hydrate saturation conditions.

Immiscible displacement in a porous medium is important in many applications such as soil remediation and enhanced oil recovery. When gravitational forces are negligible, two-phase immiscible displacement at the pore level is controlled by capillary and viscous forces whose relative importance is quantified through the dimensionless capillary number Ca and the viscosity ratio M between liquid phases. Depending on the values of Ca and M, capillary fingering, viscous fingering, or stable displacement may be observed resulting in a variety of patterns affecting the phase entrapment. The Capillary Desaturation Curve (CDC), which represents the relationship between the residual oils saturation and Ca, is an important relation to describe the phase entrapment at a given Ca. In the present study, we investigate the CDC as influenced by the viscosity ratio. A comprehensive series of experiments using a high-resolution microscope and state-of-the-art micromodels were conducted. The CDCs were calculated and the effects of Ca and M on phase entrapments were quantified. The results show that CDCs are not necessarily monotonic for all M.

This paper presents a novel methodology for capturing instantaneous, temporally and spatially resolved velocity fields in an immiscible multiphase flow of liquid/supercritical CO2 and water through a porous micromodel. Of interest is quantifying pore-scale flow processes relevant to geological CO2 sequestration and enhanced oil recovery, and in particular, at thermodynamic conditions relevant to geological reservoirs. A previously developed two-color microscopic particle image velocimetry approach is combined with a high-pressure apparatus, facilitating flow quantification of water interacting with supercritical CO2. This technique simultaneously resolves (in space and time) the aqueous phase velocity field as well as the dynamics of the menisci. The method and the experimental apparatus are detailed, and the results are presented to demonstrate its unique capabilities for studying pore-scale dynamics of CO2-water interactions. Simultaneous identification of the boundary between the two fluid phases and quantification of the instantaneous velocity field in the aqueous phase provides a step change in capability for investigating multiphase flow physics at the pore scale at reservoir-relevant conditions.

The goal of this research was to investigate the difference in behavior of hydrophilic and hydrophobic colloids during transport in two-phase flow, in general, and their attachment and remobilization characters, in particular. Experiments were performed in a hydrophobic polydimethylsiloxane (PDMS) micromodel. Water and fluorinert-FC43 were used as the two immiscible liquids. Given the fact that PDMS is a hydrophobic material, fluorinert was the wetting phase and water was the nonwetting phase in this micromodel. As model colloids, we used hydrophilic polystyrene carboxylate-modified microspheres (dispersible in water) and hydrophobic fluorous-modified silica microspheres (dispersible in fluorinert) in separate experiments. Using a confocal laser scanning microscope, we directly observed fluid distribution and colloid movement within pores of the micromodel. We also obtained concentration breakthrough curves by measuring the fluorescent intensities in the outlet of the micromodel. The breakthrough curves during steady-state flow showed that the colloid attachment rate is inversely related to the background saturation of the fluid in which the colloids were dispersed. Our visualization results showed that the enhanced attachment of hydrophilic colloids at lower water saturations was due to the retention at the fluorinert-water interface and fluorinert-water-solid contact lines. This effect was observed to be much less in the case of hydrophobic colloids (dispersed in fluorinert). In order to explain the colloids behavior, we calculated interaction potential energies of colloids with PDMS surfaces, fluid-fluid interfaces, and fluid-fluid-solid contact lines. Also, balance of forces that control colloid, including DLVO, hydrodynamic, and surface tension forces, were determined. Our calculations showed that there is a stronger repulsive energy barrier between hydrophobic colloids and fluorinert-water interface and solid-fluid interface, compared with the hydrophilic colloids. Moreover, hydrophobic colloids were seen to aggregate due to strong attractive forces among them. These aggregates had even less tendency to attach to various interfaces, due to an increase in the corresponding energy barrier. For the colloid retention at fluid-fluid-solid contact lines, we found that the role of DLVO interactions was less important than capillary forces. During transient events, we found that attached colloids become remobilized. The colloids deposited on the solid-fluid interface were detached by the moving fluid-fluid-solid contact lines. But, this happened only when the liquid containing colloids was being displaced by the other liquid. We simulated breakthrough curves using a model based on a coupled system of equations for two-phase flow, advection-dispersion of colloids, adsorption to and desorption from fluid-fluid interfaces and fluid-solid interfaces. Very good agreements were obtained among measured breakthrough curves, visualization results, and numerical modeling.

Capillarity and both gas and water permeabilities change as a function of gas saturation. Typical trends established in the discipline of unsaturated soil behavior are used when simulating gas production from hydrate-bearing sediments. However, the evolution of gas saturation and water drainage in gas invasion (i.e., classical soil behavior) and gas nucleation (i.e., gas production) is inherently different: micromodel experimental results show that gas invasion forms a continuous flow path while gas nucleation forms isolated gas clusters. Complementary simulations conducted using tube networks explore the implications of the two different desaturation processes. In spite of their distinct morphological differences in fluid displacement, numerical results show that the computed capillarity-saturation curves are very similar in gas invasion and nucleation (the gas-water interface confronts similar pore throat size distribution in both cases); the relative water permeability trends are similar (the mean free path for water flow is not affected by the topology of the gas phase); and the relative gas permeability is slightly lower in nucleation (delayed percolation of initially isolated gas-filled pores that do not contribute to gas conductivity). Models developed for unsaturated sediments can be used for reservoir simulation in the context of gas production from hydrate-bearing sediments, with minor adjustments to accommodate a lower gas invasion pressure Po and a higher gas percolation threshold.

[1] In this work, we study colloid transport through a porous medium under steady-state and transient two-phase flow conditions. The porous medium was a PDMS micromodel and the immiscible fluids were water and fluorinert-FC43. Given that the micromodel was hydrophobic, fluorinert was the wetting phase, and water was the nonwetting phase. We used hydrophilic fluorescent microspheres (dispersed in water) with mean diameter of 300 nm. We directly observed colloid movement and fluids distribution within pores of the micromodel using a confocal laser scanning microscope. We also obtained concentration breakthrough curves by measuring the fluorescence intensity in the outlet of the micromodel. The breakthrough curves showed that, under steady-state flow at different water saturations, more colloids were retained at lower saturations. Our visualization results suggested that the enhanced attachment was due to the retention of colloids onto fluorinert-water interfaces (FWIs) and fluorinert-water-solid contact lines (FWSCs). At the end of a steady-state two-phase flow experiment, we changed the micromodel saturation by injecting either water (drainage) or fluorinert (imbibition). We found remobilization of colloids during imbibition events, but no remobilization was observed during drainage. Visualization showed that colloids deposited on solid-water interfaces (SWIs) were dislodged by moving FWSCs during imbibition. We simulated breakthrough curves by modeling colloids interactions with SWIs and FWIs separately. Remobilization of colloids attached to SWI was modeled as a first-order kinetic process and the rate coefficient was assumed proportional to temporal rate of change of saturation. Colloids attachment to and detachment from FWIs were modeled as an equilibrium process. Generally, good agreements between experimental results and simulation were obtained. This is the first study of colloid transport in two-phase flow, where pore-scale visualization, breakthrough concentration measurement, and modeling of results are combined.

Micromodels have been increasingly employed in various ways in porous media research, to study the pore-scale behavior of fluids. Micromodels have proven to be a valuable tool by allowing the observation of flow and transport at the micron scale in chemical, biological, and physical applications. They have helped to improve our insight of flow and transport phenomena at both microscale and macroscale. Up to now, most micromodels that have been used to study the role of interfaces in two-phase flow were small, square, or nearly square domains. In this work, an elongated PDMS micromodel, bearing a flow network with dimensions 5×30 mm2 was manufactured. The pore network was designed such that the REV size was around 5×7 mm2. So, our flow network was considered to be nearly four times the REV size. Using such micromodels, we established that the inclusion of interfacial area between the wetting and the nonwetting fluids models the hysteretic relationship between capillary pressure and saturation in porous media. In this paper, we first present the procedure for manufacturing PDMS micromodels with the use of soft lithography. Then, we describe an innovative and novel optical setup that allows the real-time visualization of elongated samples. Finally, we present the results obtained by quasi-static, two-phase flow experiments.

[1] The abiotic precipitation of uranium (U(VI)) was evaluated in a microfluidic pore network (i.e., micromodel) to assess the efficacy of using a phosphate amendment to immobilize uranium in groundwater. U(VI) was mixed transverse to the direction of flow with hydrogen phosphate (HPO42−), in the presence or absence of calcium (Ca2+) or sulfate (SO42−), in order to identify precipitation rates, morphology and types of minerals formed, and effects of mineral precipitates on pore blockage. Precipitation occurred over the time scale of hours to days. Relative to when only U(VI) and HPO42− were present, precipitation rates were 2.3 times slower when SO42− was present, and 1.4 times faster when Ca2+ was present; larger crystals formed in the presence of SO42−. Raman backscattering spectroscopy and micro-X-ray diffraction results both showed that the only mineral precipitated was chernikovite, UO2HPO4 · 4H2O; energy dispersive X-ray spectroscopy results indicate that Ca and S are not incorporated into the chernikovite lattice. A pore-scale model was developed, and simulation results of saturation ratio (SR = Q/Ksp) suggest that chernikovite is the least thermodynamically favored mineral to precipitate (0 105). Fluorescent tracer studies and laser confocal microscopy images showed that densely aggregated precipitates blocked pores and reduced permeability. The results suggest that uranium precipitation with phosphate as chernikovite is rapid on the time scale of remediation for the conditions considered and can block pores, alter fluid flow paths, and potentially limit mixing and precipitation.

Surfactant-enhanced aquifer remediation is an emerging technology for aquifers contaminated with nonaqueous phase liquids (NAPLs). A two-dimensional micromodel and image capture system were applied to observe microscale NAPL mobilization and solubilization phenomena. In each experiment a common residual NAPL field was established, followed by a series of mobilization and solubilization experiments. Mobilization floods included pure water floods with variable flow rates and surfactant floods with variations in surfactant formulations. At relatively low capillary numbers (Nca < 10−3) the surfactant mobilization floods resulted in higher NAPL saturations than for similar Nca pure water floods. These differences in macroscopic saturations are explained by differences in microscale mobilization processes. Sqlubilization of the residual NAPL remaining after the mobilization stage was dominated by the formation of microscale dissolution fingers, which produced nonequilibrium macroscale NAPL solubilization. A macroemulsion phase also was observed to form spontaneously and persist during the solubilization stage of the experiments.

Microscale experiments can provide mechanistic insights into larger-scale flow and transport phenomena. Studies of the microscale mechanics involved in preferential flow in general, and unsaturated fast flow paths in particular, require the development of new experimental techniques. A new method for constructing glass micromodels has been developed which permits direct visualization and quantification of flow and transport phenomena in fractured porous media. In the fracture-matrix micromodels a sequential etching procedure was developed in order to provide the necessary contrast of depths between matrix pores and fracture apertures. This high contrast in etching depths ensures that very different capillary properties are associated with micromodel "fractures" and "matrix" blocks. Improved techniques were also developed for reducing the pore sizes of the matrix to a natural fine-grained sandstone pore scale. The improved micromodel pattern designs allow for previously unachievable control of boundary conditions. Various saturated and unsaturated fracture flow and transport processes can be visually and quantitatively studied with these micromodels. A method for directly measuring pore-scale flow velocity distribution through tracing trajectories of suspended fluorescent microspheres was also developed. Examples of applications include measurements of velocity profiles in fractures, imbibition, fracture-matrix transient flow, and matrix diffusion. In general, the improved micromodel method provides a unique tool for exploring some of the previously unrecognized flow and transport processes in fractured porous media. This research is directed at providing microscale explanations to some currently unresolved flow and transport issues important in predicting the larger-scale flow processes.

[1] We develop a 2-D pore scale model of coupled fluid flow, reactive transport, and calcium carbonate (CaCO3) precipitation and dissolution. The model is used to simulate transient experimental results of CaCO3 precipitation and dissolution under supersaturated conditions in a microfluidic pore network (i.e., micromodel) in order to improve understanding of coupled reactive transport systems perturbed by geological CO2 injection. In the micromodel, precipitation is induced by transverse mixing along the centerline in pore bodies. The reactive transport model includes the impact of pH upon carbonate speciation and a CaCO3 reaction rate constant, the effect of changing reactive surface area upon the reaction, and the impact of pore blockage from CaCO3 precipitation on diffusion and flow. Overall, the pore scale model qualitatively captured the precipitate morphology, precipitation rate, and maximum precipitation area using parameter values from the literature. In particular, we found that proper estimation of the effective diffusion coefficient (Deff) and the reactive surface area is necessary to adequately simulate precipitation and dissolution rates. In order to match the initial phase of fast precipitation, it was necessary to consider the top and bottom of the micromodel as additional reactive surfaces. In order to match a later phase when dissolution occurred, it was necessary to increase the dissolution rate compared to the precipitation rate, but the simulated precipitate area was still higher than the experimental results after ∼30 min, highlighting the need for future study. The model presented here allows us to simulate and mechanistically evaluate precipitation and dissolution of CaCO3 observed in a micromodel pore network. This study leads to improved understanding of the fundamental physicochemical processes of CaCO3precipitation and dissolution under far-from-equilibrium conditions.

This paper exposes the significant role played by the gas-water interface in the fate and transport of colloids in porous media and also introduces a micromodel method to allow direct visualization of colloid behavior in pore networks. The gas-water interface was created by trapping air in the pore space. Various types of latex and clay particles, as well as bacteria, were studied. The results suggest that the gas-water interface sorbs not only hydrophobic but also hydrophilic particles. The degree of sorption is controlled by particle surface hydrophobicity, solution ionic strength, and particle charge sign. Sorption increases with increasing particle hydrophobicity and solution ionic strength, while positively charged particles have a very strong affinity for the gas-water interface. The sorption on the gas-water interface is essentially irreversible, in that the capillary free energy provides a large attractive force to hold particles on the gas-water interface after its rupture. These findings reveal a mechanism of vadose zone transport: A static gas-water interface behaves as a sorbent phase retarding the transport of particulate contaminants. The visualization method developed in this research is very useful for the investigation of particulate contaminant behavior and interface-related transport, especially in the context of bioremediation. 32 refs., 8more » figs., 1 tab.« less

[1] A new method to resolve biofilms in three dimensions in porous media using high-resolution synchrotron-based X-ray computed microtomography (CMT) has been developed. Imaging biofilms in porous media without disturbing the natural spatial arrangement of the porous medium and associated biofilm has been a challenging task, primarily because porous media generally preclude conventional imaging via optical microscopy; X-ray tomography offers a potential alternative. Using silver-coated microspheres for contrast, we were able to differentiate between the biomass and fluid-filled pore spaces. The method was validated using a two-dimensional micromodel flow cell where both light microscopy and CMT imaging were used to image the biofilm.

[1] Micromodel technologies are a useful and important method to study pore-scale fluidic processes, using two-dimensional formats that enable direct visualization of processes within patterned microstructures. In this technical note, Nile red, 9-diethylamino-5H-benzo[alpha]phenoxazine-5-one, is demonstrated as a single dye whose solvatochromism enables selective visualization of two immiscible liquid fluids in a pore network micromodel containing a homogeneous array of pillars. Nile red dissolves in, and partitions between, hexadecane as a nonwetting fluid and polyethylene glycol 200 (PEG200) as a hydrophilic wetting fluid in a micromodel with silicon oxide surfaces. Both the absorption spectra and fluorescence emission spectra are sensitive to the solvent environment, such that the two phases can be distinguished by the observed color or the fluorescence emission band. Bright field, epifluorescence, confocal fluorescence, and hyperspectral microscopy methods were used to image the micromodel after displacing PEG200 in the model with hexadecane. Using a single solvatochromic dye facilitates direct visualization and identification of both phases anywhere in the micromodel on the basis of color and also enables collection of complementary fluorescent images for each phase. The use of Nile red with these imaging methods facilitates selective visualization of phase identity at specific locations; the interfaces between the two immiscible liquid phases; wetting behavior of the wetting phase within the pore network; and retention of the wetting phase as thin films around pillars and as bridges across the pore throats. The pillars and wetting phase bridges create a network of obstacles defining a tortuous flow path for the displacing nonwetting phase.

Development of pore network models based on detailed topological data of the pore space is essential for predicting multiphase flow in porous media. In this work, an unstructured pore network model has been developed to simulate a set of drainage and imbibition laboratory experiments performed on a two-dimensional micromodel. We used a pixel-based distance transform to determine medial pixels of the void domain of micromodel. This process provides an assembly of medial pixels with assigned local widths that simulates the topology of the porous medium. Using this pore network model, the capillary pressure-saturation and capillary pressure-interfacial area curves measured in the laboratory under static conditions were simulated. On the basis of several imbibition cycles, a surface of capillary pressure, saturation and interfacial area was produced. The pore network model was able to reproduce the distribution of the fluids as observed in the micromodel experiments. We have shown the utility of this simple pore network approach for capturing the topology and geometry of the micromodel pore structure.

[1] Colloids are considered as potential carriers of otherwise immobile contaminants. In this framework the contaminant-colloid interaction is assumed to be permanent, or the contaminant release from the colloids is not explicitly assessed. In this study, we report on micromodel experiments to visualize and quantify the behavior of colloids in a chemically reactive environment. Fluorescent polystyrene latex beads were injected into a micromodel pore network containing water, air, and 1-octanol. The colloids were found to attach to the hydrophobic 1-octanol phase with slightly higher attachment rates compared to the solid interface. Colloids attached to the 1-octanol sometimes disintegrated because of a swelling of the polystyrene polymer. The hydrophobic fluorescent dye was then found to partition into the 1-octanol. While the 1-octanol droplets were slowly dissolving, the fluorescent dye was staying in the droplet at first. At high concentrations and very small droplets the dye was dissolving with 1-octanol. Although one might be tempted to neglect the repartitioning of contaminants from colloids to other phases because of a limited mass of colloid associated contaminants, this effect has to be included into colloid transport models at contaminated sites with nonaqueous phase liquids, which are more attractive to the contaminant than the colloid or which change the stability of the colloid.

[1] The transport and deposition behavior of pathogenic Escherichia coli O157:H7 was studied under unfavorable electrostatic conditions in saturated quartz sands of various sizes (710, 360, 240, and 150 μm) and at several flow rates. At a given velocity, column effluent breakthrough values for E. coli tended to decrease in magnitude, and concentration curves became more asymmetric with decreasing sand size. In a given sand, experiments conducted at a higher velocity tended to produce higher effluent concentrations, especially for finer (240 and 150 μm) textured sands. The shape of the deposition profiles for E. coli was also highly dependent on the sand size and velocity. Coarser-textured sands and higher flow rates were associated with less deposition and gradually decreasing concentrations with depth. Conversely, finer-textured sands and lower flow rates tended to produce greater deposition and nonmonotonic deposition profiles that exhibited a peak in retained concentration. This deposition peak occurred nearer to the column inlet for finer-textured sands and at low flow rates. Microscopic observations of E. coli retention in these finer-textured sands (micromodel experiments) clearly indicated that straining was the dominant mechanism of deposition. Batch experiments also indicated minor amounts of E. coli attachment for the selected sands and solution chemistry. A conceptual and numerical model was developed and successfully used to describe the observed E. coli transport and deposition data. Our conceptual model assumes that E. coli can aggregate when large numbers of monodispersed E. coli are deposited at pore constrictions or straining sites. When the deposited E. coli reach a critical concentration at the straining site, the aggregated E. coli O157:H7 can be released into aqueous solution as a result of hydrodynamic shearing forces.

[1] Considerable research suggests that colloid deposition in porous media is frequently not consistent with filtration theory predictions under unfavorable attachment conditions. Filtration theory was developed from an analysis of colloid attachment to the solid-water interface of a single spherical grain collector and therefore does not include the potential influence of pore structure, grain-grain junctions, and surface roughness on straining deposition. This work highlights recent experimental evidence that indicates that straining can play an important role in colloid deposition under unfavorable attachment conditions and may explain many of the reported limitations of filtration theory. This conclusion is based upon pore size distribution information, size exclusion, time- and concentration-dependent deposition behavior, colloid size distribution information, hyperexponential deposition profiles, the dependence of deposition on colloid and porous medium size, batch release rates, micromodel observations, and deposition at textural interfaces. The implications of straining in unsaturated and heterogeneous systems are also discussed, as well as the potential influence of system solution chemistry and hydrodynamics. The inability of attachment theory predictions to describe experimental colloid transport data under unfavorable conditions is demonstrated. Specific tests to identify the occurrence and/or absence of straining and attachment are proposed.

[1] Colloid transport was studied at the pore scale in order to gain insight into the microscale processes governing particle removal. Monodisperse suspensions of colloids and water-saturated micromodels were employed. Experiments were carried out for different particle sizes, grain surface roughness, solution ionic strength, and flow rates. Straining and attachment were observed and measured by tracking the trajectory and fate of individual colloids using optical microscopy. Classical filtration theory proved appropriate for throat to colloid ratios (T/C) larger than 2.5 but did not take into account the possibility of straining that becomes an important capture mechanism for smaller T/C ratios. Spatially within the porous medium, straining occurred within the first 1–2 pore throats, while interception and attachment was seen from the inlet to the first 6–10 pore spaces, depending on particle size. Once a particle passed the initial region, the probability of attachment was very small. Colloid attachment increased with increasing solution ionic strength or decreasing flow rate, whereas straining was mainly independent of flow rate. Surface roughness of the grains also played a significant role in colloid capture, increasing collision efficiency by a factor of 2–3. The mechanisms of removal and the spatial distribution of colloid retention differed noticeably as a function of the T/C ratio. Micromodel visualizations clearly showed that physical straining and the effect of surface roughness should be taken into account when predicting the transport of colloids in saturated porous media.

[1] Micromodels employing a two-dimensional representation of pore space were used to observe directly (via microscope) water imbibition into a matrix and matrix-fracture interactions between wetting and nonwetting fluids. Within a single field of view, some pores are responsible for the uptake of water, whereas immediately adjacent pores expel nonwetting phase into the fracture. When water flow through fractures is relatively slow and fluid transfer from the fracture is relatively rapid, imbibition is microscopically cocurrent and micromodel observations teach that uptake of the wetting phase by the matrix correlates directly with the volume of water injected. This mode of transfer is coined a filling fracture. On the other hand, when fractures fill with water quickly relative to the rate of matrix-fracture transfer, the mass of water imbibed scales with the square root of time. Here imbibition is found to be countercurrent at the pore level. In the countercurrent mode, significant channeling of the nonwetting phase through the continuous wetting phase is observed that reduces the efficiency of water infiltration. Overall, it is found that the rate of water uptake from a fracture into an unsaturated matrix and the pore-level pattern of water infiltration depend critically on the rate of water infiltration through fractures.

[1] Although natural soil and aquifer systems often contain layers and lenses of contrasting soil texture, relatively little research has focused on the mechanisms of colloid deposition at textural interfaces. Saturated column studies were undertaken to characterize the straining behavior of negatively charged latex colloids (1.1 and 3.0 μm) at textural interfaces. Mechanisms of colloid transport and retention were deduced from measured effluent concentration curves, final spatial distributions in the columns, mass balance information, microscopic examination of deposition behavior in micromodel experiments, and numerical modeling. Transport and deposition of colloids were found to be highly dependent upon the textural interface. Deposition of colloids in a given sand was always most pronounced at the sand (inlet) surface. Here colloids enter a new pore network and are more likely to encounter smaller pores or dead-end regions of the pore space that contribute to straining. Less deposition occurred at textural interfaces within the column than at the sand surface. We believe that this is due to the fact that advection, dispersion, and size exclusion tend to confine colloid transport to the larger pore networks, thus limiting accessibility to straining sites. Increasing the textural contrast at an interface produced greater colloid deposition when water flowed from coarser- to finer-textured sands. Conversely, when water flowed from finer- to coarser-textured sands, little deposition occurred. Numerical modeling indicates the need to account for blocking (filling) and accessibility of straining sites in layered systems. A previously developed straining model was modified to account for this behavior.

[1] Intermittent filtration through porous media used for water and wastewater treatment can achieve high pathogen and colloid removal efficiencies. To predict the removal of bacteria, the effects of cyclic infiltration and draining events (transient unsaturated flow) were investigated. Using physical micromodels, we visualized the intermittent transport of bacteria and other colloids in unsaturated porous media. Column experiments provided quantitative measurements of the phenomena observed at the pore scale. Tagged Escherichia coli and a conservative tracer (NaI) were introduced in an initial pulse into a 1.5 m sand column. Subsequent hydraulic flushes without tagged bacteria or tracer were repeated every 4 hours for the next 4 days, during which outflow concentrations were monitored. Breakthrough behavior between colloids and dissolved tracer differed significantly, reflecting the differences in transport processes. Advancement of the wetting front remobilized bacteria which were held in thin water films, attached to the air-water interface (AWI), or entrapped in stagnant pore water between gas bubbles. In contrast, the tracer was only remobilized by diffusion from immobile to mobile water. Remobilization led to successive concentration peaks of bacteria and tracer in the effluent but with significant temporal differences. Observations at the pore-scale indicated that the colloids were essentially irreversibly attached to the solid-water interface, which explained to some extent the high removal efficiency of microbes in the porous media. Straining, cluster filtration, cell lysis, protozoa grazing, and bacteriophage parasitism could also contribute to the removal efficiency of bacteria.

[1] The success of in situ bioremediation projects depends on the mixing of contaminants and nutrients in the presence of microbes. In this work, a pore-scale model is developed to simulate biomass growth that is controlled by the mixing of an electron donor and acceptor. A homogeneous packing of cylinders representing solid grains is used as the model two-dimensional porous medium. The system is initially seeded with microbes in computational cells located at grain-water interfaces. The solutes enter the system completely unmixed; each solute is input over one half of the inlet boundary. Solute mixing is controlled by molecular diffusion transverse to the flow direction, and solutes are biotransformed according to dual Monod kinetics only where biomass is present. Simulation of biomass growth requires calculation of the water flow field as well as transport and reaction of solutes. The lattice Boltzmann method is used to obtain the flow field. Transport and reaction of the solutes is modeled by a finite volume discretization of the advection-diffusion-reaction equation. Biomass is allowed to grow and spread by means of a cellular automata algorithm. Model parameters are systematically varied to understand their effects on biomass development. Base case parameter values are obtained from batch experiments reported in the literature and are modified to achieve agreement between simulation results and previously reported micromodel experimental results. The most significant mechanisms that control biomass development are shear strength of new biomass and solute degradation rates. The biomass growth model achieves good qualitative agreement with experimental results.

[1] Elastic waves have been observed to increase productivity of oil wells, although the reason for the vibratory mobilization of the residual organic fluids has remained unclear. Residual oil is entrapped as ganglia in pore constrictions because of resisting capillary forces. An external pressure gradient exceeding an “unplugging” threshold is needed to carry the ganglia through. The vibrations help overcome this resistance by adding an oscillatory inertial forcing to the external gradient; when the vibratory forcing acts along the gradient and the threshold is exceeded, instant “unplugging” occurs. The mobilization effect is proportional to the amplitude and inversely proportional to the frequency of vibrations. We observe this dependence in a laboratory experiment, in which residual saturation is created in a glass micromodel, and mobilization of the dyed organic ganglia is monitored using digital photography. We also directly demonstrate the release of an entrapped ganglion by vibrations in a computational fluid-dynamics simulation.

[1] We present results from pore-scale observations of colloid transport in an unsaturated physical micromodel. The experiments were conducted separately using three different sizes of carboxylate polystyrene latex spheres and Bacteriophage MS2 virus. The main focus was to investigate the pore-scale transport processes of colloids as they interact with the air-water interface (AWI) of trapped air bubbles in unsaturated porous media, as well as the release of colloids during imbibition. The colloids travel through the water phase but are attracted to the AWI by either collision or attractive forces and are accumulated at the AWI almost irreversibly, until the dissolution of the air bubble reduces or eliminates the AWI. Once the air bubbles are near the end of the dissolution process, the colloids can be transported by advective liquid flow, as colloidal clusters. The clusters can then attach to other AWI down-gradient or be trapped in pore throats that would have allowed them to pass through individually. We also observed small air bubbles with attached colloids that traveled through the porous medium during the gas dissolution process. We used Derjaguin-Landau-Verwey-Overbeek (DLVO) theory to help explain the observed results. The strength of the force that holds the colloids at the AWI was estimated, assuming that the capillary force is the major force that holds the colloids at the AWI. Our calculations indicate that the forces that hold the colloids at the AWI are larger than the energy barrier between the colloids. Therefore it is quite likely that the clusters of colloids are formed by the colloids attached at the AWI as they move closer at the end of the bubble dissolution process. Coagulation at the AWI may increase the overall filtration for colloids transported through the vadose zone. Just as important, colloids trapped in the AWI might be quite mobile when the air bubbles are released at the end of the dissolution process, resulting in increased breakthrough. These pore-scale mechanisms are likely to play a significant role in the macroscopic transport of colloids in unsaturated porous media. INDEX TERMS: 1831 Hydrology: Groundwater quality; 1832 Hydrology: Groundwater transport; 5139 Physical Properties of Rocks: Transport properties; KEYWORDS: air-water interface, DLVO, hydrophobic, colloid, unsaturated, micromodel

[1] Soil wettability has been recently recognized as a factor that can dramatically influence the dissolution behavior of residual nonaqueous phase liquids (NAPL). A NAPL that wets the solid surface is trapped within the smaller pores and along the corners of pores invaded by water (the nonwetting phase). We present a two-dimensional network simulator of wetting NAPL dissolution, inspired by observations of this process in transparent glass micromodels. The network model idealizes the pore space as a network of cubic pores connected by square tubes, following respective distributions. In accordance with micromodel observations, capillary equilibrium is assumed to exist between NAPL-water interfaces along pore corners and within pores. Advection and diffusion of the organic dissolved in the aqueous phase, as well as dissolution mass transfer from residual NAPL, are explicitly accounted for in the model. Pores filled with NAPL are invaded at a rate which is controlled by mass transfer from dissolving thick NAPL films in pore corners and in order of increasing entry capillary pressure, resulting in quasi-static drainage and fingering of the aqueous phase. Loss of NAPL continuity due to rupture of thick NAPL films and heterogeneity are found to affect profoundly the dissolution behavior, resulting in concentration tailing. The network simulator reproduces qualitatively the behavior observed in column experiments with oil-wet media.

[1] We present experimental evidence of the effect of colloid exclusion from areas of small aperture sizes, using direct observations at the pore-scale using a realistic micromodel of porous media. Four sizes of hydrophobic latex spheres in aqueous suspension, from 0.05 to 3 μm, were introduced into the micromodel at three different pressure gradients. We observed the frequency of occurrence of the size exclusion effect and the influence of relative size of pore throats and colloids (T/C ratio) and flow velocity. From our observations the smallest T/C ratio entered by these different colloids was 1.5. We also observed certain preferential pathways through the pore space for different colloid sizes, such that size exclusion eventually results in distinct pathways. These preferential paths become more important for larger colloids and for greater pressure gradients. Measured colloid velocities were 4–5.5 times greater than estimated pore water velocities. Acceleration factors (ratio of colloid to water velocity) increased for all colloid sizes with increasing pore-scale Pe. Smaller particles appeared to travel along faster streamlines in pore throats, while larger particles travel along with a number of streamlines, thus at a slightly lower velocity than the small colloids. At larger scales the acceleration factor is decreased owing to Brownian motion, adsorption, colloid and straining filtration, and other factors, but these pore-scale results shed light on the size exclusion effect and its role in determining early colloid breakthrough.