Search

Geological carbon storage in deep aquifers involves displacement of resident brine by supercritical CO2 (scCO2), which is an unstable drainage process caused by the invasion of less viscous scCO2. The unstable drainage is greatly complicated by aquifer heterogeneity and anisotropy and regarded as one of the key factors accounting for the uncertainty in storage capacity estimates. The impacts of pore-scale characteristics on the unstable drainage remain poorly understood. In this study, scCO2 drainage experiments were conducted at 40 °C and 9 MPa using a homogeneous elliptical micromodel with low or high anisotropy, a homogeneous/isotropic hexagonal micromodel, and a heterogeneous sandstone-analog micromodel. Each initially water-saturated micromodel was invaded by scCO2 at different rates with logCa (the capillary number)ranging from −7.6 to −4.4, and scCO2/water images were obtained. The measured CO2 saturations in these centimeter-scale micromodels vary considerably from 0.08 to 0.93 depending on the pore-scale characteristics and capillary number. It was also observed that scCO2 drainage follows the classic flow-regime transition from capillary fingering through crossover to viscous fingering for either of the low-anisotropy elliptical and heterogeneous micromodels, but with disparate crossover zones. The crossover zones of scCO2 saturation were then unified with the minimum scCO2 saturation occurring at logCa*=-4.0 using the complete capillary number (Ca*)that considers pore characteristics. For the hexagonal and the high-anisotropy elliptical micromodels, a monotonic increase in scCO2 saturation with increasing Ca* (without crossover)was observed. It appears that the complete capillary number is more appropriate than the classic capillary number when characterizing flow regimes and CO2 saturation in different pore networks.

Dissolution trapping is one of the most important mechanisms for geological carbon storage (GCS). Recent laboratory and field experiments have shown non-equilibrium dissolution of supercritical CO2 (scCO2) and coupled scCO2 dissolution and water flow, i.e., scCO2 dissolution at local pores/pore throats creating new water-flow paths, which in turn enhance dissolution by increased advection and interfacial area. However, the impacts of pore-scale characteristics on these coupled processes have not been investigated. In this study, imbibition and dissolution experiments were conducted under 40 °C and 9 MPa using a homogeneous/isotropic hexagonal micromodel, two homogeneous elliptical micromodels with low or high anisotropy, and a heterogeneous sandstone-analog micromodel. The four micromodels, initially saturated with deionized (DI)-water, were drained by injecting scCO2 to establish a stable scCO2 saturation. DI water was then injected at different rates with logCa (the capillary number) ranging from −6.56 to −4.34. Results show that bypass of scCO2 by displacing water is the dominant mechanism contributing to the residual CO2 trapping, triggered by heterogeneity in pore characteristics or pore-scale scCO2-water distribution. Bypass can be enhanced by pore heterogeneity or reduced by increasing transverse permeability, resulting in relatively low (

Recently, both core- and pore-scale imbibition experiments have shown non-equilibrium dissolution of supercritical CO2 (scCO2) and a prolonged depletion of residual scCO2. In this study, pore-scale scCO2 dissolution and mass transfer under drainage conditions were investigated using a two-dimensional heterogeneous micromodel and a novel fluorescent water dye with a sensitive pH range between 3.7 and 6.5. Drainage experiments were conducted at 9 MPa and 40 °C by injecting scCO2 into the sandstone-analogue pore network initially saturated by water without dissolved CO2 (dsCO2). During the experiments, time-lapse images of dye intensity, reflecting water pH, were obtained. These images show non-uniform pH in individual pores and pore clusters, with average pH levels gradually decreasing with time. Further analysis on selected pores and pore clusters shows that (1) rate-limited mass transfer prevails with slowly decreasing pH over time when the scCO2-water interface area is low with respect to the volume of water-filled pores and pore clusters, (2) fast scCO2 dissolution and phase equilibrium occurs when scCO2 bubbles invade into water-filled pores, significantly enhancing the area-to-volume ratio, and (3) a transition from rate-limited to diffusion-limited mass transfer occurs in a single pore when a medium area-to-volume ratio is prevalent. The analysis also shows that two fundamental processes – scCO2 dissolution at phase interfaces and diffusion of dsCO2 at the pore scale (10–100 µm) observed after scCO2 bubble invasion into water-filled pores without pore throat constraints – are relatively fast. The overall slow dissolution of scCO2 in the millimeter-scale micromodel can be attributed to the small area-to-volume ratios that represent pore-throat configurations and characteristics of phase interfaces. This finding is applicable for the behavior of dissolution at pore, core, and field scales when water-filled pores and pore clusters of varying size are surrounded by scCO2 at narrow pore throats.

In modeling of geological carbon storage, dissolution of supercritical CO2 (scCO2) is often assumed to be instantaneous with equilibrium phase partitioning. In contrast, recent core-scale imbibition experiments have shown a prolonged depletion of residual scCO2 by dissolution, implying a non-equilibrium mechanism. In this study, eight pore-scale scCO2 dissolution experiments in a 2D heterogeneous, sandstone-analog micromodel were conducted at supercritical conditions (9MPa and 40°C). The micromodel was first saturated with deionized (DI) water and drained by injecting scCO2 to establish a stable scCO2 saturation. DI water was then injected at constant flow rates after scCO2 drainage was completed. High resolution time-lapse images of scCO2 and water distributions were obtained during imbibition and dissolution, aided by a scCO2-soluble fluorescent dye introduced with scCO2 during drainage. These images were used to estimate scCO2 saturations and scCO2 depletion rates. Experimental results show that (1) a time-independent, varying number of water-flow channels are created during imbibition and later dominant dissolution by the random nature of water flow at the micromodel inlet, and (2) a time-dependent number of water-flow channels are created by coupled imbibition and dissolution following completion of dominant imbibition. The number of water-flow paths, constant or transient in nature, greatly affects the overall depletion rate of scCO2 by dissolution. The average mass fraction of dissolved CO2 (dsCO2) in water effluent varies from 0.38% to 2.72% of CO2 solubility, indicating non-equilibrium scCO2 dissolution in the millimeter-scale pore network. In general, the transient depletion rate decreases as trapped, discontinuous scCO2 bubbles and clusters within water-flow paths dissolve, then remains low with dissolution of large bypassed scCO2 clusters at their interfaces with longitudinal water flow, and finally increases with coupled transverse water flow and enhanced dissolution of large scCO2 clusters. The three stages of scCO2 depletion, common to experiments with time-independent water-flow paths, are revealed by zoom-in image analysis of individual scCO2 bubbles and clusters. The measured relative permeability of water, affected by scCO2 dissolution and bi-modal permeability, shows a non-monotonic dependence on saturation. The results for experiments with different injection rates imply that the non-equilibrium nature of scCO2 dissolution becomes less important when water flow is relatively low and the time scale for dissolution is large, and more pronounced when heterogeneity is strong.

Acquisition of porous media saturated hydraulic conductivity data in the laboratory is usually time consuming and costly because of the manual labor associated with the currently available techniques. Lately, there has been increased interest in automating hydraulic conductivity laboratory techniques to reduce analysis time and improve data consistency. A new apparatus was developed that is able to measure hydraulic conductivity values with the constant-flux, constant-head, and falling-head methods in a fully automated fashion. The apparatus can be used for both packed columns and undisturbed field cores. The column design is such that water is forced to flow in a nominally one-dimensional manner through the porous medium. An analysis is initiated with a 'smart search,' yielding an estimate of the saturated hydraulic conductivity. The operator can use this estimated value to obtain conductivity values using one, two, or all three methods. Besides installing and removing the columns, no manual efforts are required. Hydraulic conductivity data for standard laboratory sands showed that application of the three methods resulted in similar results, indicating that external flow resistance was small. A comparison with literature data for the same sands, obtained with a constant-flux method, showed differences Abbreviations: IHCA, integrated hydraulic conductivity apparatus.

A testing methodology is presented that supports the development of a field design for in situ gaseous treatment of sediments with diluted hydrogen sulfide. This approach involves the collection of column breakthrough test results at various flow rates, allowing a relationship to be developed between pore velocity of the carrier gas and velocity of the hydrogen sulfide reaction front that permits sizing to the field scale. A regression lit of a set of laboratory column breakthrough test data collected in this study is utilized to illustrate the development of a field design based oil a two-dimensional radial flow analytical model. Information regarding treatment time and hydrogen sulfide consumption characteristics associated with in situ gaseous treatment can then be obtained from this model and used as a basis for estimation of treatment schedule and costs. The regression relationship can also be utilized in numerical models in more complex geometries to support the field design of in situ gaseous treatment operations. CE Database subject headings: Laboratory tests: Remedial action: Subsurface environment: Vadose zone: Gas: Sediment: Columns.

Flow of nonvolatile nonaqueous phase liquid (NAPL) and aqueous phases that account for mobile, entrapped, and residual NAPL in variably saturated water-wet porous media is modeled and compared against results from detailed laboratory experiments. Residual saturation formation in the vadose zone is a process that is often ignored in multifluid flow simulators, which might cause an overestimation of the volume of NAPL that reaches the ground water. Mobile NAPL is defined as being continuous in the pore space and flows under a pressure gradient or gravitational body force. Entrapped NAPL is defined as being occluded by the aqueous phase, occurring as immobile ganglia surrounded by aqueous phase in the pore space and formed when NAPL is replaced by the aqueous phase. Residual NAPL is defined as immobile, nonwater entrapped NAPL that does not drain from the pore spaces and is conceptualized as being either continuous or discontinuous. Free NAPL comprises mobile and residual NAPL. The numerical model is formulated on mass conservation equations for oil and water, transported via NAPL and aqueous phases through variably saturated porous media. To account for phase transitions, a primary variable switching scheme is implemented for the oil-mass conservation equation over three phase conditions: (1) aqueous or aqueous-gas with dissolved oil, (2) aqueous or aqueous-gas with entrapped NAPL, and (3) aqueous or aqueous gas with free NAPL. Two laboratory-scale column experiments are modeled to verify the numerical model. Comparisons between the numerical simulations and experiments demonstrate the necessity to include the residual NAPL formation process in multifluid flow simulators., Introduction The relationships between the various forms of non-aqueous phase liquid (NAPL) discussed in this paper--free, entrapped, residual, mobile, and immobile--are given in Equation 1, where the subscripts n, f, [...]

Numerical simulations with the multifluid flow simulator STOMP are compared with quantitative results of a detailed lighter-than-water nonaqueous-phase liquid (LNAPL) infiltration and redistribution experiment in a 1.67-m-long, 1-m-high, and 0.05-m-wide flow cell. The experiment was performed to test the ability of commonly used nonhysteretic and hysteretic constitutive relations between relative permeability (k), fluid saturation (S), and capillary pressure head (h) to describe multifluid flow in two dimensions. The fluid and sand parameters necessary to apply the constitutive relations were obtained independently. The flow cell was filled with a homogeneous sand mixture under saturated conditions. After partial drainage of the sand, the LNAPL was slowly injected for 12 h from a small source area located at the surface. A dual-energy gamma radiation system was used to determine LNAPL and water saturations at 255 locations during infiltration and redistribution. The results show a reasonable match between the experimental and numerical data, indicating that the constitutive relations used are adequate to describe relatively slow LNAPL infiltration and redistribution. The differences between the nonhysteretic and hysteretic simulations are small. This implies that hysteresis, a result of nonwetting fluid entrapment and pore geometry, was not important in this experiment.

Introduction In most natural aquifers, the heterogeneity of the permeability is primarily responsible for the scale dependency of the dispersivity. Guven et al. (1986) among others noted that one reason [...]

Nonintrusive, in situ salt concentration measurement in saturated sand columns using a gamma radiation technique is discussed. A seriesof vertical displacement experiments with deionized water and a NaI solution demonstrated the technique. Longitudinal dispersivity values were derived and compared between resident concentration breakthrough curves and conventional effluent curves. It was found that apparent dispersivities are affected by the density configuration of the liquids.