Feasibility of CO<Subscript>2</Subscript> migration detection using pressure and CO<Subscript>2</Subscript> saturation monitoring above an imperfect primary seal of a geologic CO<Subscript>2</Subscript> storage formation: a numerical investigation

A numerical model was developed to investigate the potential to detect fluid migration in a (homogeneous, isotropic, with constant pressure lateral boundaries) porous and permeable interval overlying an imperfect primary seal of a geologic CO₂ storage formation. The seal imperfection was modeled as a single higher-permeability zone in an otherwise low-permeability seal, with the center of that zone offset from the CO₂ injection well by 1400 m. Pressure response resulting from fluid migration through the high-permeability zone was detectable up to 1650 m from the centroid of that zone at the base of the monitored interval after 30 years of CO₂ injection (detection limit = 0.1 MPa pressure increase); no pressure response was detectable at the top of the monitored interval at the same point in time. CO₂ saturation response could be up to 774 m from the center of the high-permeability zone at the bottom of the monitored interval, and 1103 m at the top (saturation detection limit = 0.01). More than 6% of the injected CO₂, by mass, migrated out of primary containment after 130 years of site performance (including 30 years of active injection) in the case where the zone of seal imperfection had a moderately high permeability (10^??17 m² or 0.01 mD). Free-phase CO₂ saturation monitoring at the top of the overlying interval provides favorable spatial coverage for detecting fluid migration across the primary seal. Improved sensitivity of detection for pressure perturbation will benefit time of detection above an imperfect seal.     

Time-lapse crosswell seismic and VSP monitoring of injected CO<Subscript>2</Subscript> in a brine aquifer

Seismic surveys successfully imaged a small scale CO₂ injection (1,600 ton) conducted in a brine aquifer of the Frio Formation near Houston, Texas. These time-lapse borehole seismic surveys, crosswell and vertical seismic profile (VSP), were acquired to monitor the CO₂ distribution using two boreholes (the new injection well and a pre-existing well used for monitoring) which are 30 m apart at a depth of 1,500 m. The crosswell survey provided a high-resolution image of the CO₂ distribution between the wells via tomographic imaging of the P-wave velocity decrease (up to 500 m/s). The simultaneously acquired S-wave tomography showed little change in S-wave velocity, as expected for fluid substitution. A rock physics model was used to estimate CO₂ saturations of 10–20% from the P-wave velocity change. The VSP survey resolved a large (∼70%) change in reflection amplitude for the Frio horizon. This CO₂ induced reflection amplitude change allowed estimation of the CO₂ extent beyond the monitor well and on three azimuths. The VSP result is compared with numerical modeling of CO₂ saturations and is seismically modeled using the velocity change estimated in the crosswell survey.     

Analytical model for screening potential CO<Subscript>2</Subscript> repositories

Assessing potential repositories for geologic sequestration of carbon dioxide using numerical models can be complicated, costly, and time-consuming, especially when faced with the challenge of selecting a repository from a multitude of potential repositories. This paper presents a set of simple analytical equations (model), based on the work of previous researchers, that could be used to evaluate the suitability of candidate repositories for subsurface sequestration of carbon dioxide. We considered the injection of carbon dioxide at a constant rate into a confined saline aquifer via a fully perforated vertical injection well. The validity of the analytical model was assessed via comparison with the TOUGH2 numerical model. The metrics used in comparing the two models include (1) spatial variations in formation pressure and (2) vertically integrated brine saturation profile. The analytical model and TOUGH2 show excellent agreement in their results when similar input conditions and assumptions are applied in both. The analytical model neglects capillary pressure and the pressure dependence of fluid properties. However, simulations in TOUGH2 indicate that little error is introduced by these simplifications. Sensitivity studies indicate that the agreement between the analytical model and TOUGH2 depends strongly on (1) the residual brine saturation, (2) the difference in density between carbon dioxide and resident brine (buoyancy), and (3) the relationship between relative permeability and brine saturation. The results achieved suggest that the analytical model is valid when the relationship between relative permeability and brine saturation is linear or quasi-linear and when the irreducible saturation of brine is zero or very small.     

CO<Subscript>2</Subscript> fluid inclusions in Jack Hills zircons

The discovery of Hadean to Paleoarchean zircons in a metaconglomerate from Jack Hills, Western Australia, has catalyzed intensive study of these zircons and their mineral inclusions, as they represent unique geochemical archives that can be used to unravel the geological evolution of early Earth. Here, we report the occurrence and physical properties of previously undetected CO₂ inclusions that were identified in 3.36–3.47 Ga and 3.80–4.13 Ga zircon grains by confocal micro-Raman spectroscopy. Minimum P–T conditions of zircon formation were determined from the highest density of the inclusions, determined from the density-dependence of the Fermi diad splitting in the Raman spectrum and Ti-in-zircon thermometry. For both age periods, the CO₂ densities and Ti-in-zircon temperatures correspond to high-grade metamorphic conditions (≥5 to ≥7 kbar/~670 to 770 °C) that are typical of mid-crustal regional metamorphism throughout Earth’s history. In addition, fully enclosed, highly disordered graphitic carbon inclusions were identified in two zircon grains from the older population that also contained CO₂ inclusions. Transmission electron microscopy on one of these inclusions revealed that carbon forms a thin amorphous film on the inclusion wall, whereas the rest of the volume was probably occupied by CO₂ prior to analysis. This indicates a close relationship between CO₂ and the reduced carbon inclusions and, in particular that the carbon precipitated from a CO₂-rich fluid, which is inconsistent with the recently proposed biogenic origin of carbon inclusions found in Hadean zircons from Jack Hills.     

Sealing efficiency analysis for shallow-layer caprocks in CO<Subscript>2</Subscript> geological storage

The CO₂ migrated from deeper to shallower layers may change its phase state from supercritical state to gaseous state (called phase transition). This phase transition makes both viscosity and density of CO₂ experience a sharp variation, which may induce the CO₂ further penetration into shallow layers. This is a critical and dangerous situation for the security of CO₂ geological storage. However, the assessment of caprock sealing efficiency with a fully coupled multi-physical model is still missing on this phase transition effect. This study extends our previous fully coupled multi-physical model to include this phase transition effect. The dramatic changes of CO₂ viscosity and density are incorporated into the model. The impacts of temperature and pressure on caprock sealing efficiency (expressed by CO₂ penetration depth) are then numerically investigated for a caprock layer at the depth of 800 m. The changes of CO₂ physical properties with gas partial pressure and formation temperature in the phase transition zone are explored. It is observed that phase transition revises the linear relationship of CO₂ penetration depth and time square root as well as penetration depth. The real physical properties of CO₂ in the phase transition zone are critical to the safety of CO₂ sequestration. Pressure and temperature have different impact mechanisms on the security of CO₂ geological storage.     

On CO<Subscript>2</Subscript> fluid flow and heat transfer behavior in the subsurface,following leakage from a geologic storage reservoir

Geologic storage of CO₂ is expected to produce plumes of large areal extent, and some leakage may occur along fractures, fault zones, or improperly plugged pre-existing wellbores. A review of physical and chemical processes accompanying leakage suggests a potential for self-enhancement. The numerical simulations presented here confirm this expectation, but reveal self-limiting features as well. It seems unlikely that CO₂ leakage could trigger a high-energy run-away discharge, a so-called “pneumatic eruption,” but present understanding is insufficient to rule out this possibility. The most promising avenue for increasing understanding of CO₂ leakage behavior is the study of natural analogues.     

Dynamics of CO<Subscript>2</Subscript> fluxes and concentrations during a shallow subsurface CO<Subscript>2</Subscript> release

A field facility located in Bozeman, Montana provides the opportunity to test methods to detect, locate, and quantify potential CO₂ leakage from geologic storage sites. From 9 July to 7 August 2008, 0.3 t CO₂ day⁻¹ were injected from a 100-m long, ~2.5-m deep horizontal well. Repeated measurements of soil CO₂ fluxes on a grid characterized the spatio-temporal evolution of the surface leakage signal and quantified the surface leakage rate. Infrared CO₂ concentration sensors installed in the soil at 30 cm depth at 0–10 m from the well and at 4 cm above the ground at 0 and 5 m from the well recorded surface breakthrough of CO₂ leakage and migration of CO₂ leakage through the soil. Temporal variations in CO₂ concentrations were correlated with atmospheric and soil temperature, wind speed, atmospheric pressure, rainfall, and CO₂ injection rate.     

Depth of volcanic basalt degassing forecasted from CO2 fluid inclusions

Fluid inclusions have recorded the history of degassing in basalt. Some fluid inclusions in olivine and pyroxene phenocrysts of basalt were analyzed by micro-thermometry and Raman spectroscopy in this paper. The experimental results showed that many inclusions are present almost in a pure CO2 system. The densities of some CO2 inclusions were computed in terms of Raman spectroscopic characteristics of CO2 Fermi resonance at room temperature. Their densities change over a wide range, but mainly between 0.044 g/cm3 and 0.289 g/cm3. Their micro-thermometric measurements showed that the CO2 inclusions examined reached homogenization between 1145.5℃ and 1265℃ . The mean value of homogenization temperatures of CO2 inclusions in basalts is near 1210℃. The trap pressures (depths) of inclusions were computed with the equation of state and computer program. Distribution of the trap depths makes it know that the degassing of magma can happen over a wide pressure (depth) range, but mainly at the depth of 0.48 km to 3.85 km. This implicates that basalt magma experienced intensive degassing and the CO2 gas reservoir from the basalt magma also may be formed in this range of depths. The results of this study showed that the depth of basalt magma degassing can be forecasted from CO2 fluid inclusions, and it is meaningful for understanding the process of magma degassing and constraining the inorganogenic CO2 gas reservoir.     

Near-surface soil carbon detection for monitoring CO<Subscript>2</Subscript> seepage from a geological reservoir

The promise of the Brookhaven National Laboratory (BNL) Inelastic Neutron Scattering (INS) System was evaluated for use as a long-term, in-field monitor to detect cumulative changes in belowground carbon resulting from the leakage of CO₂ stored in deep geological reservoirs. This system underwent tests at a facility constructed specifically for testing, under controlled conditions, various detection systems for monitoring near-surface transport and accumulations of CO₂ fluxes emanating from a shallow buried, slotted horizontal well. The INS System was assessed by comparing the results from placing it above the horizontal well at a spot with a known high CO₂ leak identified and quantified the previous years, with those obtained from background readings adjacent to the well. At two different “Hot Spots”, a suppression of about 14% in 2008 and about 7% in 2009 in carbon content above the well in comparison to the background signal was observed. An overview of these results is presented.     

A stochastic inversion workflow for monitoring the distribution of CO<Subscript>2</Subscript> injected into deep saline aquifers

We present a two-step stochastic inversion approach for monitoring the distribution of CO₂ injected into deep saline aquifers for the typical scenario of one single injection well and a database comprising a common suite of well logs as well as time-lapse vertical seismic profiling (VSP) data. In the first step, we compute several sets of stochastic models of the elastic properties using conventional sequential Gaussian co-simulations (SGCS) representing the considered reservoir before CO₂ injection. All realizations within a set of models are then iteratively combined using a modified gradual deformation algorithm aiming at reducing the mismatch between the observed and simulated VSP data. In the second step, these optimal static models then serve as input for a history matching approach using the same modified gradual deformation algorithm for minimizing the mismatch between the observed and simulated VSP data following the injection of CO₂. At each gradual deformation step, the injection and migration of CO₂ is simulated and the corresponding seismic traces are computed and compared with the observed ones. The proposed stochastic inversion approach has been tested for a realistic, and arguably particularly challenging, synthetic case study mimicking the geological environment of a potential CO₂ injection site in the Cambrian-Ordivician sedimentary sequence of the St. Lawrence platform in Southern Québec. The results demonstrate that the proposed two-step reservoir characterization approach is capable of adequately resolving and monitoring the distribution of the injected CO₂. This finds its expression in optimized models of P- and S-wave velocities, density, and porosity, which, compared to conventional stochastic reservoir models, exhibit a significantly improved structural similarity with regard to the corresponding reference models. The proposed approach is therefore expected to allow for an optimal injection forecast by using a quantitative assimilation of all available data from the appraisal stage of a CO₂ injection site.     

Capturing and storage of CO<Subscript>2</Subscript> by micron-nano minerals: Evidence from the nature

The increase of CO2 in atmosphere is a main factor leading to "greenhouse effect", which causes more and more serious global environmental problems. The reduction of CO2 is a challenge for the survival of human beings, and it is also a big technical problem. CO2 fluid-rock interaction is a key scientific problem involved in geo-logical storage. The CO2 fluid-rock interaction has a variety of multi-scale changes. Due to great differences in the quantity of surface atoms and surface energy between micron-nano-sized minerals, and ions and crystals, the speed and efficiency of CO2 fluid-rock interaction on a micron-nano scale are much higher than those on other scales. As is known from the natural world, the micron-nano structures of pores and the surface chemical modification of natural porous minerals (zeolite, diatomite, sepiolite, palygorskite, halloysite, etc.) should be further investigated, which can be used as the micron-nano -mineral porous materials with high capacity and high efficiency for capturing CO2. Through simulating the adsorption capacity and process of CO2 by minerals in the natural world, the micron-nano technology is applied to calcium- and magnesium-based minerals (olivine, pyroxene, feldspar, clay, etc.) so as to improve the activity of calcium and magnesium and enlarge the reaction contact area. In this way, the efficiency of capturing and storage of CO2 by calcium- and magnesium-based minerals can be greatly improved. These minerals can also be used as the micron-nano-mineral materials with large capacity and high efficiency for capturing and storing CO2.     

Preparation of carbon molecular sieves and its impregnation with Co and Ni for CO<Subscript>2</Subscript>/N<Subscript>2</Subscript> separation

The carbon molecular sieves (CMSs) prepared by carbonaceous materials as precursors are effective in CO₂/N₂ separation. However, selectivity of these materials is too low, since hydrocarbon cracking for developing the desired microporosity in carbonaceous materials has not been done effectively. Hence, in this study, cobalt and nickel impregnation on the precursor was conducted to introduce catalysts for hydrocarbon cracking. Cobalt and nickel impregnation, carbonization under N₂ atmosphere, and chemical vapor deposition (CVD) by benzene were conducted on the extruded mixtures of activated carbon and coal tar pitch under different conditions to prepare CMSs. The best CMS prepared by carbon deposition on the cobalt-impregnated samples exhibited CO₂ adsorption capacity of 54.79 mg/g and uptake ratio of 28.9 at 0 °C and 1 bar. In terms of CO₂ adsorption capacity and uptake ratio, CMSs prepared by carbon deposition on non-impregnated and cobalt-impregnated samples presented the best results, respectively. As benzene concentration and CVD time increased, equilibrium adsorption capacity of CO₂ decreased, and uptake ratio increased. Cobalt was found to be the best catalyst for benzene cracking in the CVD process.     

The Role of CaCO<Subscript>3</Subscript> Reactions in the Contemporary Oceanic CO<Subscript>2</Subscript> Cycle

The present analysis adjusts previous estimates of global ocean CaCO₃ production rates substantially upward, to 133 × 10¹² mol yr^?1 plankton production and 42 × 10¹² mol yr^?1 shelf benthos production. The plankton adjustment is consistent with recent satellite-based estimates; the benthos adjustment includes primarily an upward adjustment of CaCO₃ production on so-called carbonate-poor sedimentary shelves and secondarily pays greater attention to high CaCO₃ mass (calcimass) and turnover of shelf communities on temperate and polar shelves. Estimated CaCO₃ sediment accumulation rates remain about the same as they have been for some years: ~20 × 10¹² mol yr^?1 on shelves and 11 × 10¹² mol yr^?1 in the deep ocean. The differences between production and accumulation of calcareous materials call for dissolution of ~22 × 10¹² mol yr^?1 (~50 %) of shelf benthonic carbonate production and 122 × 10¹² mol yr^?1 (>90 %) of planktonic production. Most CaCO₃ production, whether planktonic or benthonic, is assumed to take place in water depths of <100 m, while most dissolution is assumed to occur below this depth. The molar ratio of CO₂ release to CaCO₃ precipitation (CO₂↑/CaCO₃↓) is <1.0 and varies with depth. This ratio, Ψ, is presently about 0.66 in surface seawater and 0.85 in ocean waters deeper than about 1000 m. The net flux of CO₂ associated with CaCO₃ reactions in the global ocean in late preindustrial time is estimated to be an apparent influx from the atmosphere to the ocean, of +7 × 10¹² mol C yr^?1, at a time scale of 10²–10³ years. The CaCO₃-mediated influx of CO₂ is approximately offset by CO₂ release from organic C oxidation in the water column. Continuing ocean acidification will have effects on CaCO₃ and organic C metabolic responses to the oceanic inorganic C cycle, although those responses remain poorly quantified.     

Using hyperspectral plant signatures for CO<Subscript>2</Subscript> leak detection during the 2008 ZERT CO<Subscript>2</Subscript> sequestration field experiment in Bozeman,Montana

Hyperspectral plant signatures can be used as a short-term, as well as long-term (100-year timescale) monitoring technique to verify that CO₂ sequestration fields have not been compromised. An influx of CO₂ gas into the soil can stress vegetation, which causes changes in the visible to near-infrared reflectance spectral signature of the vegetation. For 29 days, beginning on July 9, 2008, pure carbon dioxide gas was released through a 100-m long horizontal injection well, at a flow rate of 300 kg day⁻¹. Spectral signatures were recorded almost daily from an unmown patch of plants over the injection with a “FieldSpec Pro” spectrometer by Analytical Spectral Devices, Inc. Measurements were taken both inside and outside of the CO₂ leak zone to normalize observations for other environmental factors affecting the plants. Four to five days after the injection began, stress was observed in the spectral signatures of plants within 1 m of the well. After approximately 10 days, moderate to high amounts of stress were measured out to 2.5 m from the well. This spatial distribution corresponded to areas of high CO₂ flux from the injection. Airborne hyperspectral imagery, acquired by Resonon, Inc. of Bozeman, MT using their hyperspectral camera, also showed the same pattern of plant stress. Spectral signatures of the plants were also compared to the CO₂ concentrations in the soil, which indicated that the lower limit of soil CO₂ needed to stress vegetation is between 4 and 8% by volume.     

An integrative approach to robust design and probabilistic risk assessment for CO<Subscript>2</Subscript> storage in geological formations

CO₂ storage in geological formations is currently being discussed intensively as a technology with a high potential for mitigating CO₂ emissions. However, any large-scale application requires a thorough analysis of the potential risks. Current numerical simulation models are too expensive for probabilistic risk analysis or stochastic approaches based on a brute-force approach of repeated simulation. Even single deterministic simulations may require parallel high-performance computing. The multiphase flow processes involved are too non-linear for quasi-linear error propagation and other simplified stochastic tools. As an alternative approach, we propose a massive stochastic model reduction based on the probabilistic collocation method. The model response is projected onto a higher-order orthogonal basis of polynomials to approximate dependence on uncertain parameters (porosity, permeability, etc.) and design parameters (injection rate, depth, etc.). This allows for a non-linear propagation of model uncertainty affecting the predicted risk, ensures fast computation, and provides a powerful tool for combining design variables and uncertain variables into one approach based on an integrative response surface. Thus, the design task of finding optimal injection regimes explicitly includes uncertainty, which leads to robust designs with a minimum failure probability. We validate our proposed stochastic approach by Monte Carlo simulation using a common 3D benchmark problem (Class et al., Comput Geosci 13:451–467, 2009). A reasonable compromise between computational efforts and precision was reached already with second-order polynomials. In our case study, the proposed approach yields a significant computational speed-up by a factor of 100 compared with the Monte Carlo evaluation. We demonstrate that, due to the non-linearity of the flow and transport processes during CO₂ injection, including uncertainty in the analysis leads to a systematic and significant shift of the predicted leakage rates toward higher values compared with deterministic simulations, affecting both risk estimates and the design of injection scenarios.