H13I-01 13:40h
Experimental and Theoretical Investigations of Reaction-Coupled Flow and Transport in Porous Media
In some systems, it is possible to observe complex patterns of coupling between fluid and flow and mass transport when reactions involving a solid phase are operative. For example, dissolution and precipitation reactions can change a porous medium's physical properties such as porosity and permeability. These changes influence fluid flow, which affect the concentration of dissolved solids, the composition of solid phases, and the rate and direction of advective transport. Both experimental and modeling studies were conducted to investigate the coupling between flow and transport due to effects of fluid density, dissolution/precipitation reactions, and heterogeneity in medium properties. The complex chemical system is created by pumping a dilute Fe(ClO4)3 solution through a medium created by mixing glass beads and crushed calcite. Fe3+ rapidly hydrolyzes to produce hydroxo complexes and H+. As pH increases through reaction with calcite, a poorly crystallized solid, ferric oxyhydroxide precipitates. Two-dimensional flow tank studies are use to verify a novel modeling approach. In the model, there is full coupling of flow and transport due to permeability changes from dissolution/precipitation reactions. Further, TOUGHREACT is used to study reaction-front dynamics, and how the aqueous phase concentrations depend upon this pattern of evolution. Both the experimental and theoretical results highlight the complexity of coupling in systems with heterogeneous reactions. The important implication of this study is that details of interactions between pore fluid and the porous medium need to be well characterized in order to predict the changing aqueous concentrations.
H13I-02 13:55h
Pore Size Controls on the Temporal Evolution of Porosity in Reacting Basalts
Basaltic lavas exhibit considerable, yet heterogeneous, porosity (typically 10-25 % in vesiculated zones, but sometimes up to 85 % in scoraceous flowtops) and concomitantly large permeabilities (0.001 to 10 darcies). The heterogeneous distribution of porosity occurs not only on the scale of a lava flow (e.g., between vesiculated flowtops and massive centers of thick aa flows) but also at smaller scales in terms of the abundance and size distribution of pores. Low temperature chemical alteration of basaltic lavas (i.e., weathering and zeolite facies metamorphism) results in dissolution of primary basalt phases (glass, anorthite-rich plagioclase, olivine, pyroxenes) and precipitation of a repeatable sequence of mineral parageneses. Secondary minerals formed during this process, such as zeolites and mafic phyllosilicates, have open crystal structures (low densities) and thus large molar volumes. The net effect is an increase in mineral volume. Reaction path modeling of typical zeolite facies mineral parageneses in basaltic lavas indicates that the net increase in mineral volume (the volume of new minerals formed minus that of the minerals dissolved) is up to 27 %. The new volume of mineral is accommodated by replacement (often pseudomorphic) of the primary phases and growth of minerals within pore spaces. Superposition relations within vesicles indicate that mafic phyllosilicates form first, followed by one or more generations of zeolites and other silicates. Observations of the distribution of various stages in this paragenesis between individual pores (which range in size from sub-mm to over one cm) indicate that pore size and shape largely control the extent of reaction progress recorded at the scale of an individual pore. For instance, within a given sample, mafic phyllosilicate linings on pores are observed to be of uniform thickness. If the thickness of the phyllosilicate lining exceeds the radius of the pore (or one half the minimum thickness of an elongate pore), the pore becomes completely filled during this stage and is not part of the storage capacity of the lava. Similarly, late stage mineral generations and residual porosity, if present, are typically only developed in the largest pores. These relations, along with recently discovered evidence of restricted fluid flow networks around and between vesicles in zeolite facies lavas, suggest that porosity modification in these systems is largely controlled by local (mm-scale) transport of material from the lava matrix into the pores and can be used to generate geometric predictions of the fate of pore space during reactive fluid flow. The fact that pore size and shape influences the preservation of porosity (and thus to some extent permeability) during reactive fluid flow in basaltic lavas suggests that semi-empirical models of reaction-porosity-permeability couplings that are typically used in other media do not apply well to vesicular materials.
H13I-03 14:10h
Use of Bioluminescence to Study Reactive Solute Transport and Biofilm Growth and Activity in Porous Media
Using a meso-scale porous media flat plate reactor we utilized a naturally bioluminescent biofilm (V. fischeri) and tracer studies to obtain information on the interactions between biofilms and reactive flow in porous media. The growth and development of the V. fischeri biofilm in a porous media geometry was studied using digital time lapse images of the bioluminescent signal given off by the developing biofilm. The effect of biofilm development on porous media hydrodynamics was examined using dye tracer studies and image analysis. The natural bioluminescence of the V. fischeri allowed real-time, in-situ study of biofilm development in porous media, without destruction of the biofilm. Dye studies and image analysis enabled the study of effects of biofilm accumulation on porous media hydraulics, with comparisons to plug flow and completely mixed systems with varying degrees of biofilm accumulation. The hydraulic conductivity of the porous media/biofilm system was continuously monitored showing a 1 to 4 order of magnitude decrease in hydraulic conductivity as a function of biofilm thickness and accumulation. The real-time nature of the study permitted us to visualize dynamic flow channel formation within the biofilm/porous media system. In addition, the sensitivity of the V. fischeri biofilm to dissolved oxygen allowed us to capture real-time images of reactive transport within the system. Using bioluminescent imaging, the location of active biomass, as well as the relative degree of biological activity, could be visualized and monitored over time. This work is the first meso-scale visualization of the interactions between biofilm and flow in porous media.
H13I-04 14:25h
Influence of Cemented Layers on Contaminant Transport in Mine Tailings
Exposure of sulfide-mine tailings to atmospheric oxygen leads to the initiation of a series of reactions, including sulfide oxidation, acid neutralization and metal attenuation reactions. As oxygen ingresses into the tailings, the oxidation front moves downward and inward from the edges of the tailings surface. At or near the acid neutralization front, secondary phases can accumulate, leading to the formation of hardpan layers. Field studies were conducted at three mine sites to evaluate the role of cemented layers in influencing contaminant transport from oxidized tailings. Detailed field measurements were made, including collection of water and gas samples from the vadose and groundwater zones. Cores were collected for mineralogical and chemical analyses to evaluate the extent of sulfide mineral oxidation and accumulation of secondary phases. Calculations of mineral saturation indices were made using ion-pair and ion-interaction models that were modified to account for the very high solute concentrations observed in the tailings pore waters. At a site that has been oxidizing for 25 years, a massive Fe(III)-bearing hardpan, containing gypsum, goethite and jarosite, has formed over the last 15 years. At a site that has been oxidizing for 35 years, an Fe(III)-bearing hardpan is also present. At a site that has been oxidizing for 70 years, a massive Fe(II)-bearing hardpan containing melanterite and gypsum is present below the zone of active oxidation. Above this zone, there are discontinuous Fe(III)-bearing cemented layers that are likely oxidized remnants of the original Fe(II) hardpan. Calculated mineral saturation indices are consistent with the observed accumulations of secondary phases. Transient perched water table conditions have developed above the massive Fe(II) hardpan, leading to the lateral transport of sulfide oxidation products along the hardpan and the formation of seepage zones above the permanent water table. Chemical extractions and mineralogical analyses show that trace metals are concentrated in the hardpan layers. Simulations using a reactive solute transport model show close agreement between the predicted and observed accumulations of secondary phases.
H13I-05 14:40h
Mathematical Modelling of Hydraulic Permeability Evolution in the Damage Zone Surrounding Geological Faults
Geological faults are planar structures, oriented in three-dimensional space, on which shear displacement has occurred. As the rock shears, the material within and around the shear plane is damaged causing a huge variability in fluid flow properties. Faults can be barriers to flow, conduits, or combinations of the two, and their hydraulic properties vary considerably over both space and time. It is critical for the prediction of both future and historical fluid (or gas) migration through fault zones to be able to assess their spatial and temporal hydraulic evolution. This is particularly relevant, due to the large timescales involved, when modelling historic migration in oil fields, or when simulating the transport of radionuclides following the deep burial of radioactive waste. Fluid flow and structural deformation are fully coupled processes within fault zones: Small scale discontinuities, such as fractures and deformation bands of reduced permeability, are formed in the damage zone surrounding faults and have a strong influence on flow properties. Fluid flow in the subsurface is traditionally modelled using Darcy's law, and structural deformation using Navier's law. The main aim of this research is to investigate and validate our understanding of these coupled processes in the damage zone surrounding faults. In the research presented here, fault zone evolution is modelled using a finite element approach. The coupled hydro-mechanical model has been developed and validated using standard benchmark data. The model has been used to simulate deformation and fluid flow in damage zone structures mapped at the Big Hole fault, Utah. Results demonstrate propagation of the existing slip surfaces leading to increased flow. The model is now being applied to explore temporal and spatial fault evolution in the Sierra Nevada based on data sets collected by B\"{u}rgmann and Pollard. Results will validate and enhance scientific understanding of the physical processes inherent in fault development. Ultimately, the model will be extended to include chemical processes such as mineralisation, and will provide an invaluable tool for predicting the heterogeneous hydrogeological properties of faults.
H13I-06 14:55h
Coupling diffusion and high-pH precipitation/dissolution in the near field of a HLW repository in clay by means of reactive solute transport models
The reference concept for a HLW repository in clay in Spain includes a 75 cm thick bentonite buffer which surrounds canisters. A concrete sustainment 20 cm thick is foreseen between the bentonite buffer and the clay formation. The long term geochemical evolution of the near field is affected by a high-pH hyperalkaline plume induced by concrete. Numerical models of multicomponent reactive transport have been developped in order to quantify the evolution of the system over 1 Ma. Water flow is negligible once the bentonite buffer is saturated after about 20 years. Therefore, solute transport occurs mainly by diffusion. Models account for aqueous complexation, acid-base and redox reactions, cation exchange, and mineral dissolution precipitation in the bentonite, the concrete and the clay formation. Numerical results obtained witth CORE2D indicate that the high-pH plume causes significant changes in porewater chemistry both in the bentonite buffer and the clay formation. Porosity changes caused by mineral dissolution/precipitation are extremely important. Therefore, coupled modes of diffusion and reactive transport accounting for changes in porosity caused by mineral precipitation are required in order to obtain realistic predictions.
H13I-07 INVITED 15:10h
In Situ Characterization of Porosity and Permeability Changes at High Pressure: Application to Geological Sequestration
The global energy system is dominated by fossil fuels, which are abundant and relatively inexpensive. Carbon dioxide emissions resulting from the use of fossil fuels are responsible for most of the projected human influence on climate. As a society, if we wish to manage the risks of climate change, finding methods and developing new technologies so that fossil fuels become net zero-carbon emitting is a critical part of an overall climate change response strategy. An important technology receiving increasing attention is capturing CO$_{2}$ from large stationary power sources and recycling the carbon back into the ground where it may be used for additional resource recovery (oil or natural gas) or simple sequestration. Successful implementation of carbon capture and sequestration requires a fundamental understanding of the chemical reactions of CO$_{2}$ within the host formation and impacts on porosity and permeability. In this paper, we will discuss the experimental challenges associated with measurements of porosity and permeability changes under high-pressure conditions and attributing observed changes to specific dissolution-precipitation reactions or dissociation-formation reactions in the case of natural gas hydrates. Application of new techniques, such as pulsed field gradient NMR and scanning laser Raman LIDAR will be described that hold promise for in-situ measurements. Measurement of gas permeability in gas hydrate-bearing sediments will also be discussed. Such data are extremely rare, principally because of the difficulties involved in stabilizing the gas hydrate under high pressure. Gas flow rate data collected over gas hydrate saturations between 10% and 70% in Accusand show poor correlation with classical models such as Brooks-Corey.