Jens T. Birkholzer

Modeling Basin- and Plume-Scale Processes of CO2 Storage for Full-Scale Deployment

Integrated modeling of basin- and plume-scale processes induced by full-scale deployment of CO(2) storage was applied to the Mt. Simon Aquifer in the Illinois Basin. A three-dimensional mesh was generated with local refinement around 20 injection sites, with approximately 30 km spacing. A total annual injection rate of 100 Mt CO(2) over 50 years was used. The CO(2)-brine flow at the plume scale and the single-phase flow at the basin scale were simulated. Simulation results show the overall shape of a CO(2) plume consisting of a typical gravity-override subplume in the bottom injection zone of high injectivity and a pyramid-shaped subplume in the overlying multilayered Mt. Simon, indicating the important role of a secondary seal with relatively low-permeability and high-entry capillary pressure. The secondary-seal effect is manifested by retarded upward CO(2) migration as a result of multiple secondary seals, coupled with lateral preferential CO(2) viscous fingering through high-permeability layers. The plume width varies from 9.0 to 13.5 km at 200 years, indicating the slow CO(2) migration and no plume interference between storage sites. On the basin scale, pressure perturbations propagate quickly away from injection centers, interfere after less than 1 year, and eventually reach basin margins. The simulated pressure buildup of 35 bar in the injection area is not expected to affect caprock geomechanical integrity. Moderate pressure buildup is observed in Mt. Simon in northern Illinois. However, its impact on groundwater resources is less than the hydraulic drawdown induced by long-term extensive pumping from overlying freshwater aquifers.

Effect of dissolved CO2 on a shallow groundwater system: a controlled release field experiment.

Capturing carbon dioxide (CO(2)) emissions from industrial sources and injecting the emissions deep underground in geologic formations is one method being considered to control CO(2) concentrations in the atmosphere. Sequestering CO(2) underground has its own set of environmental risks, including the potential migration of CO(2) out of the storage reservoir and resulting acidification and release of trace constituents in shallow groundwater. A field study involving the controlled release of groundwater containing dissolved CO(2) was initiated to investigate potential groundwater impacts. Dissolution of CO(2) in the groundwater resulted in a sustained and easily detected decrease of ~3 pH units. Several trace constituents, including As and Pb, remained below their respective detections limits and/or at background levels. Other constituents (Ba, Ca, Cr, Sr, Mg, Mn, and Fe) displayed a pulse response, consisting of an initial increase in concentration followed by either a return to background levels or slightly greater than background. This suggests a fast-release mechanism (desorption, exchange, and/or fast dissolution of small finite amounts of metals) concomitant in some cases with a slower release potentially involving different solid phases or mechanisms. Inorganic constituents regulated by the U.S. Environmental Protection Agency remained below their respective maximum contaminant levels throughout the experiment.

Modeling studies and analysis of seepage into drifts at Yucca mountain

Abstract An important issue for the long-term performance of underground nuclear waste repositories is the rate of water seepage into the waste emplacement drifts. A prediction of the seepage rate is particularly complicated for the potential repository site at Yucca Mountain, NV, which is located in a thick sequence of unsaturated, fractured tuffs. Underground openings in unsaturated media might act as capillary barriers, diverting water around them. In the present work, we study the potential rates of seepage into drifts as a function of predicted percolation flux at Yucca Mountain, based on a stochastic model of the fractured rock mass in the drift vicinity. A variety of flow scenarios are considered, assuming estimated present-day and predicted future climate conditions. We show that the heterogeneity in the flow domain is a key factor controlling seepage rates, since it causes channelized flow and local ponding in the unsaturated flow field. The rates of seepage are related in a complex non-linear manner to the rock properties, the size and shape of the drift, the degree of heterogeneity, and the assumed percolation scenario.

Solute channeling in unsaturated heterogeneous porous media

Numerical simulations have been performed to study flow and solute transport phenomena in strongly heterogeneous, variably saturated porous media. Different saturation scenarios were applied varying from fully saturated to highly unsaturated conditions, corresponding to different infiltration rates into the soil. It was found that the solute travels along preferred flow paths, which may be called channels. The degree of channeling, the location of channels, and the hydraulic properties along channels are a function of the mean saturation in the flow domain. Strong channeling effects were obtained in both fully saturated and in low-saturation cases. At intermediate saturation values, channeling effects are less significant, and the system exhibits a more homogeneous flow pattern. The dispersion of solutes as shown in the calculated tracer breakthrough curves essentially reflects the degree of channeling and thus is saturation dependent; in cases when flow channeling is less evident we observe much smaller dispersion than in cases with strong channeling. The hydraulic properties of the channels appeared to be an invariant of the actual location and geometry, indicating that they may be an intrinsic characteristic of the soil heterogeneity and the water saturation.

Field tests and model analyses of seepage into drift

Abstract This paper focuses on field test results and model analyses of the first set of five niche seepage tests conducted in the Exploratory Studies Facility at Yucca Mountain. The test results suggest that (1) a niche opening (short drift excavated for this study) acts as a capillary barrier; (2) a seepage threshold exists; and (3) the seepage is a fraction of the liquid released above the ceiling. Before seepage quantification, air injection and liquid release tests at two niche locations were conducted to characterize the fracture flow paths. Nearly two-order-of-magnitude changes in air permeability values were measured before and after niche excavation. The dyed liquid flow paths, together with a localized wet feature potentially associated with an ambient flow path, were mapped during dry excavation operations. After niche excavation, the seepage is quantified by the ratio of the water mass dripped into a niche to the mass released above the opening at selected borehole intervals. For the first set of five tests conducted at Niche 3650 site, the ratios range from 0% (no dripping for two tests) to 27.2%. Changes in flow path distributions and water accumulation near seepage threshold were observed on the niche ceiling. The seepage test results compare reasonably well with model results without parameter adjustments, using capillary barrier boundary condition in the niche and two-dimensional and three-dimensional conceptualizations to represent discrete fracture and fracture network for the flow paths. Model analyses of the niche tests indicate that the seepage is very sensitive to the niche boundary condition and is moderately sensitive to the heterogeneity of the fracture flow paths and to the strengths of matrix imbibition. Strong capillary strength and large storage capacity of the fracture flow paths limit the seepage. High permeability value also enhances diversion and reduces seepage for low liquid release rate.

Predictions and observations of the thermal–hydrological conditions in the Single Heater Test

Abstract The Single Heater Test (SHT) is one of two in-situ thermal tests included in the site characterization program for the potential underground nuclear waste repository at Yucca Mountain. Coupled thermal–hydrological–mechanical–chemical processes in the fractured rock mass around the heater were monitored by numerous sensors emplaced among 30 boreholes. Periodic active testing of cross-hole radar tomography, neutron logging, electrical resistivity tomography, and interference air permeability tests probed the change of moisture content in the rock mass. Thermal–hydrological processes in the SHT have been simulated using a 3-D numerical model and compared to the monitored data. The good agreement between the temperature data and simulated results indicates that the thermal–hydrological responses of the SHT in the 9 months of heating are well-represented by the coupled thermal–hydrological numerical model. The dominant heat transfer process is by conduction and the signature of vapor and liquid counter flow is subtle in the temperature data. The simulated result of a dry-out zone of about 1 m (at the end of the heating phase) around the heater hole, and a condensation zone of increased liquid saturation outside of the dry-out zone, is consistent with the radar tomography and air permeability data. Tomography data and post-test laboratory measurements indicate that the moisture content is larger below than above the heater horizon, suggesting gravity drainage of condensate in the fractures. Model studies show that gravity drainage occurs in simulations using the dual permeability conceptual model, but is absent in the effective-continuum model, where matrix and fractures are required to be in thermodynamic equilibrium at all times.

Pressure Buildup and Brine Migration During CO2 Storage in Multilayered Aquifers

Carbon dioxide injection into deep saline formations may induce large-scale pressure increases and migration of native fluid. Local high-conductivity features, such as improperly abandoned wells or conductive faults, could act as conduits for focused leakage of brine into shallow groundwater resources. Pressurized brine can also be pushed into overlying/underlying formations because of diffuse leakage through low-permeability aquitards, which occur over large areas and may allow for effective pressure bleed-off in the storage reservoirs. This study presents the application of a recently developed analytical solution for pressure buildup and leakage rates in a multilayered aquifer-aquitard system with focused and diffuse brine leakage. The accuracy of this single-phase analytical solution for estimating far-field flow processes is verified by comparison with a numerical simulation study that considers the details of two-phase flow. We then present several example applications for a hypothetical CO2 injection scenario (without consideration of two-phase flow) to demonstrate that the new solution is an efficient tool for analyzing regional pressure buildup in a multilayered system, as well as for gaining insights into the leakage processes of flow through aquitards, leaky wells, and/or leaky faults. This solution may be particularly useful when a large number of calculations needs to be performed, that is, for uncertainty quantification, for parameter estimation, or for the optimization of pressure-management schemes.

Results from an International Simulation Study on Coupled Thermal, Hydrological, and Mechanical Processes near Geological Nuclear Waste Repositories

As part of the ongoing international code comparison project DECOVALEX, four research teams used five different models to simulate coupled thermal, hydrological, and mechanical (THM) processes near underground waste emplacement drifts. The simulations were conducted for two generic repository types with open or back-filled repository drifts under higher and lower post-closure temperature, respectively. In the completed first model inception phase of the project, a good agreement was achieved between the research teams in calculating THM responses for both repository types, although some disagreement in hydrological responses are currently being resolved. Good agreement in the basic thermal-mechanical responses was achieved for both repository types, even with some teams using relatively simplified thermal-elastic heat-conduction models that neglect complex near-field thermal-hydrological processes. The good agreement between the complex and simplified (and well-known) process models indicates that the basic thermal-mechanical responses can be predicted with a relatively high confidence level. The research teams have now moved on to the second phase of the project, the analysis of THM-induced permanent (irreversible) changes and the impact of those changes on the fluid flow field near an emplacement drift.

Monitoring CO2 Intrusion and Associated Geochemical Transformations in a Shallow Groundwater System Using Complex Electrical Methods

The risk of CO(2) leakage from a properly permitted deep geologic storage facility is expected to be very low. However, if leakage occurs it could potentially impact potable groundwater quality. Dissolved CO(2) in groundwater decreases pH, which can mobilize naturally occurring trace metals commonly contained in aquifer sediments. Observing such processes requires adequate monitoring strategies. Here, we use laboratory and field experiments to explore the sensitivity of time-lapse complex resistivity responses for remotely monitoring dissolved CO(2) distribution and geochemical transformations that may impact groundwater quality. Results show that electrical resistivity and phase responses correlate well with dissolved CO(2) injection processes. Specifically, resistivity initially decreases due to increase of bicarbonate and dissolved species. As pH continues to decrease, the resistivity rebounds toward initial conditions due to the transition of bicarbonate into nondissociated carbonic acid, which reduces the total concentration of dissociated species and thus the water conductivity. An electrical phase decrease is also observed, which is interpreted to be driven by the decrease of surface charge density as well as potential mineral dissolution and ion exchange. Both laboratory and field experiments demonstrate the potential of field complex resistivity method for remotely monitoring changes in groundwater quality due to CO(2) leakage.

Experimental study on effects of geologic heterogeneity in enhancing dissolution trapping of supercritical CO2

Dissolution trapping is one of the primary mechanisms that enhance the storage security of supercritical carbon dioxide (scCO2) in saline geologic formations. When scCO2 dissolves in formation brine produces an aqueous solution that is denser than formation brine, which leads to convective mixing driven by gravitational instabilities. Convective mixing can enhance the dissolution of CO2 and thus it can contribute to stable trapping of dissolved CO2. However, in the presence of geologic heterogeneities, diffusive mixing may also contribute to dissolution trapping. The effects of heterogeneity on mixing and its contribution to stable trapping are not well understood. The goal of this experimental study is to investigate the effects of geologic heterogeneity on mixing and stable trapping of dissolved CO2. Homogeneous and heterogeneous media experiments were conducted in a two-dimensional test tank with various packing configurations using surrogates for scCO2 (water) and brine (propylene glycol) under ambient pressure and temperature conditions. The results show that the density-driven flow in heterogeneous formations may not always cause significant convective mixing especially in layered systems containing low-permeability zones. In homogeneous formations, density-driven fingering enhances both storage in the deeper parts of the formation and contact between the host rock and dissolved CO2 for the potential mineralization. On the other hand, for layered systems, dissolved CO2 becomes immobilized in low-permeability zones with low-diffusion rates, which reduces the risk of leakage through any fault or fracture. Both cases contribute to the permanence of the dissolved plume in the formation.