H32A-01 10:20h
Characterization of Pore Structures Relevant for Phase Distribution and Water Flow
Water retention and flow velocity depend on the geometric properties of the pore scale. With synchrotron tomography we determined the pore structure of different sand materials with a resolution of 3.5 and 11 microns. The highly resolved images of the sand samples were analyzed with respect to transport relevant geometric properties. Different measures of pore size and connectivity were compared to the measured water retention curve and the water saturated conductivity. The distribution of water and air in the mapped pore structure was calculated with a morphological pore network model. The water saturated hydraulic conductivity was determined with a Lattice-Boltzmann approach. The predictions were in good agreement with the results of the laboratory measurements.
H32A-02 10:35h
Interfacial Area per Volume: The link between capillary pressure and saturation
Measurements were performed on micro-models to quantify interfacial area per volume for a known pore geometry as a function of fluid pressure and saturation. The micro-models are completely transparent and measure 600 m x 600 m with an aperture of 1.08 m. Because the micro-models are transparent, full visualization and quantification of the fluid distributions is possible. Initially the micro-models are saturated with decane (wetting phase). Nitrogen (non-wetting phase) is invaded into the system by the application of pressure in increments. At each increment, the system is allowed to equilibrate and then digital images of fluid distributions within the pore structure are acquired. The images are analyzed to determined fluid saturations, interfacial areas per volume and curvature of the interfaces. The curvatures of the interfaces are calculated using level set methods Pressure measurements are also made with pressure transducers during the experiment. From the data, we have established that the interfacial area per volume between non-wetting and wetting fluids lifts the ambiguity associated with the hysteretic relationship between capillary pressure and saturation in porous media. The interface between the non-wetting and wetting phases is composed of two subsets: one with a unique curvature determined by the capillary pressure, and the other with a distribution of curvatures dominated by disjoining pressure. This work provides experimental support for theoretical predictions that the capillary-dominated subset plays a role analogous to a state variable. Any comprehensive description of multiphase flow properties must include this interfacial area with the traditional variables of pressure and fluid saturation. Research is continuing to examine the role of pore structure on the relationship among capillary pressure, saturation and interfacial area per volume. Acknowledgments: Geosciences Research Program, Office of Basic Energy Sciences US Department of Energy. LJPN and DDN wish to acknowledge University Faculty Scholar program at Purdue University
H32A-03 10:50h
3D Flow of Water Through the Undisturbed Samples of Coarse Sandy Loam Visualized by MRI
The filling of pores of undisturbed soil samples (each 9 cm of diameter and 8.5 cm height) of coarse sandy loam (Korkusova Hu_, CR) was monitored under changing flow conditions using 3D nuclear magnetic resonance imaging (MRI). The soil under study represents a heterogeneous porous medium of random structure. The flow in the particular samples was imaged during recurrent ponded infiltration-outflow and tension infiltration-outflow experiments. The boundary condition at the top of the samples was maintained by means of the disc infiltrometer. Free outflow was used as the bottom boundary condition. Pressure head was measured by tensiometer installed in the body of each sample. Values of hydraulic pressure heads, sample weights and cumulative flow data were recorded during the experiments. At the saturation and near-saturation stages, significant preferential flow was observed together with the instability of steady state flow rates. From the MRI videos of water propagation throughout the samples it is evident, that the flow instability has been caused due to air phase redistribution. Questionable is the role of several order differences of local pore velocity values and its impact on the filling of the sample within the time scale of the experiment.
H32A-04 11:05h
Magnetic Resonance Imaging and Theoretical Analysis of Particle Deposition in Porous Media
Measurements of particle deposition and mobilization in porous columns were performed using nuclear magnetic resonance imaging (MRI). Use of MRI enabled acquisition of detailed measurements that quantify spatial and temporal evolution of particle transport patterns and porosity changes due to particle deposition and mobilization. In contrast to previous studies, measurements indicate that at least for the considered particle sizes, particle deposition tends to increase with distance into the column. A theoretical model of particle deposition and porosity reduction was developed to accompany the measurements. Because of difficulties associated with identifying unique parameterizations for processes of particle straining, attachment and exclusion, a simple phenomenological, physically-motivated model is suggested. The model requires a minimum number of fitting parameters, and matches the essential features of the experimental measurements on spatial and temporal flow and deposition patterns. Moreover, resulting fluctuations in the model deposition coefficient induce a memory effect in the deposition which can be treated by means of a convolution in time.
H32A-05 11:20h
Calculation of the Darcy-scale effective diffusion and dispersion tensors for porous media: Volume averaging with 3-dimensional closure solutions
The method of volume averaging has been used successfully for decades to develop the appropriate macroscopic form of transport equations in multiphase systems. However, until recently, the solution of the closure problem that allows one to predict the effective parameters that appear in the macroscopic transport equations has been difficult to solve in three dimensions. We will report on the solution to the closure problem in simple but three dimensional unit cells for the problems of diffusion and dispersion in porous media. Comparisons with laboratory data and prospects for computations in realistically structured media will also be discussed.
H32A-06 11:35h
Simulation of coupled flow, transport, and reaction in porous media by lattice Boltzmann method
We develop lattice-Boltzmann models for simulating pore-scale fluid flow, solute transport, and chemical dissolution/deposition in porous media. The processes of advection, diffusion, and surface reactions, together with temporal geometrical changes in the pore space, are taken into account. The numerical results show that at high Peclet and Peclet-Damkohler numbers, wormholes are formed and permeability increases greatly due to the dissolution process. At low Peclet and high Peclet-Damkohler numbers, reactions mainly occur at the inlet boundary, resulting in the face dissolution and the slowest increase of the permeability in the dissolution process. At moderate Peclet and Peclet-Damkohler numbers, reactions are generally nonuniform, with more in the upstream and less in the downstream. At very small Peclet-Damkohler number, dissolution or precipitation is highly uniform, and these two processes can be approximately reversed by each other. We also develop methods for upscaling the transport and reaction systems at a pore-scale to a macroscopic level and investigate the conditions under which such upscaling can be done.
H32A-07 11:50h
Lattice Boltzmann Methods and Their Boundary Conditions for Solute Transport
Lattice Boltzmann methods (LBM) are proving exceptionally versatile for modeling flow and solute transport in porous media. A number of authors have demonstrated solute transport capabilities, but little appears to have been done to evaluate these solutions in the context of well-known analytical solutions of the convection-dispersion equation, boundary conditions, and concentration detection modes. We consider LBM simulations of Taylor dispersion and dispersion in porous media in light of four 'classical' analytical solutions and their boundary conditions (BCs). 'First type' BCs are simple constant concentration boundaries while 'third type' BCs specify solute flux and allow for simultaneous convective and dispersive flux across the boundary. We also impose a zero dispersive/diffusive flux BC. We use resident and flux-averaged concentration detection modes. We explore different LBM boundary formulations.
H32A-08 12:05h
Pore-Scale Simulation of Multiphase Flow in Porous and Fractured Media Using Smoothed Particle Hydrodynamics.
A two-dimensional numerical model based on smoothed particle hydrodynamics (SPH) was used to simulate unsaturated and multiphase flow on a pore scale in porous and fractured media. A combination of standard SPH equations with pair-wise fluid-fluid and fluid-solid particle-particle interactions allowed surface tension and realistic three-phase contact dynamics to be simulated. The accuracy of the model was verified by calculating the surface tension from simulations of small-amplitude fluid drop oscillations, capillary rise and drop movement between two parallel plates. All three experiments led to a consistent value for the surface tension. The dependency of receding and advancing contact angles on droplet velocity was studied. Incorporation of surface tension and fluid-solid interactions allowed unsaturated and multiphase flow in porous and fractured media to be realistically simulated. The simulation results compare well with laboratory experiments and analytical solutions.