Numerical modeling of injection and mineral trapping of CO2 with H2S and SO2 in a sandstone formation
Introduction
Concern over the consequences of global warming due to increasing levels of anthropogenic carbon dioxide (CO2) in the atmosphere has led to a variety of proposals to curtail, if not prevent, further increases. One such approach is to inject CO2 from stationary generators (such as fossil-fuel power plants) into reasonably accessible structural reservoirs in deep permeable geologic formations. Candidates include aquifers in sedimentary formations, structural traps in depleted oil and gas fields, and deep unmineable coal seams. Deep aquifers are relatively abundant in certain regions of the United States, and therefore are logical targets for the eventual disposal of CO2. Such aquifers commonly contain brackish or saline water, and those with salinities exceeding 10,000 mg/L total dissolved solids are excluded by the U.S. Environmental Protection Agency as underground sources of drinking water or USDWs as defined in 40 CFR § 144.3 (U.S. EPA, 2001).
Previous investigations of CO2 disposal in aquifers include geochemical (Gunter et al., 2000, Perkins et al., 2002, Xu et al., 2004, Zerai et al., 2006) and hydrologic modeling (Pruess et al., 2003), reactive transport modeling (Johnson et al., 2001, McPherson and Lichtner, 2001, Knauss et al., 2005, White et al., 2005, Xu et al., 2005), laboratory experiments (Pearce et al., 1996, Rochelle et al., 1996, Gunter et al., 1997, Kaszuba et al., 2003, Soong et al., 2004, Carroll and Knauss, 2005, Kaszuba et al., 2005, Palandri and Kharaka, 2005, Pokrovsky et al., 2005, Rosenbauer et al., 2005), and field studies (Lohuis, 1993, Gunter et al., 1993, Bachu et al., 1994, Perkins and Gunter, 1995, Gunter et al., 1996, Gunter et al., 1997, Kharaka et al., 2006). Large-scale aquifer disposal of CO2 is currently practiced in the Norwegian sector of the North Sea (Korbol and Kaddour, 1995).
Conventional coal-burning power plants are one of the primary contributors of excess CO2 to the atmospheric inventory. However, the energy penalty for separation and compression of CO2 from conventional coal combustion is steep and could lead to a 30–40% reduction in useable power output (Nsakala et al., 2001). More technologically advanced integrated gas combined cycle (IGCC) plants, in which coal is gasified with restricted oxygen, are thermodynamically more efficient (i.e., produce less CO2 for a given power output) and are more suitable for CO2 capture. Therefore, if CO2 capture and deep subsurface disposal were to be considered seriously, the preferred approach would be to build replacement IGCC plants with integrated CO2 capture, rather than retrofit existing conventional plants.
The fate of minor quantities of sulfur and nitrogen compounds during combustion or gasification is of considerable interest, as their release into the atmosphere leads to the formation of urban ozone and acid rain, the destruction of stratospheric ozone, and global warming. Coal also contains many trace elements that are potentially hazardous to human health and the environment, such as mercury and arsenic, and their release into the atmosphere is restricted under the Clean Air Act Amendments (CAAA) of 1990. During CO2 separation and capture, these constituents could inadvertently contaminate the separated CO2 and be co-injected. The concentrations and speciation of the co-injected contaminants would differ markedly, depending on whether CO2 is captured during the operation of a conventional power plant or an IGCC plant. The specific nature of the plant design and CO2 separation technology could also afford an opportunity to deliberately capture environmental pollutants in the gaseous state and co-inject them with the CO2 to mitigate problems associated with co-disposal with ash in surface impoundments. With the deliberate and efficient separation and capture of volatile pollutants, their concentrations would be roughly equivalent to their concentrations in the coal feed.
In this study, we evaluate the co-injection of either sulfur dioxide (SO2) or hydrogen sulfide (H2S) with CO2. SO2 would have to be deliberately removed from the flue gas exhaust using flue gas desulfurization (FGD) technology prior to separation of CO2 using methyl ethanolamine (MEA) absorption/stripping process in conventional plants to prevent poisoning of the MEA. The recovery of SO2 in the FGD circuit would, however, require some modifications in conventional practice. In contrast, H2S is very conveniently separated together with CO2 using MEA in IGCC plants (Treviño Coca, 2003).
A precedent for the co-injection of H2S with CO2 already exists, as these acid gases were separated during natural gas production in western Canada and re-injected into deep hydrocarbon reservoirs and saline aquifers (Bachu et al., 2005). By the end of 2002, approximately 40 of such acid-gas injection facilities were active having disposed of close to 1.5 Mt CO2 and 1 Mt H2S. Approximately 16 acid-gas injection facilities are also operating in the United States (Bachu et al., 2005). The relatively trouble-free operation of these facilities suggests that acid-gas injection is a mature and safe technology that can be applied under similar conditions elsewhere. Current acid-gas injection operations would be somewhat analogous to future large-scale co-disposal of H2S with CO2 from IGCC plants. However, whereas acid gases from natural gas treatment range in composition from 2 to 84 vol.% H2S, most coals contain no more than 5 wt.% sulfur, and therefore the concentration of H2S in CO2 from IGCC plants is unlikely to exceed 1.5 vol.%.
Knauss et al. (2005) have already presented the results of reactive transport simulations to investigate the impact of mixtures of dissolved CO2, H2S, or SO2 on a geological formation. The results suggest that relatively high concentrations of H2S in CO2 injection would not adversely impact injectivity compared to the injection of CO2 alone, while co-injection of SO2 may produce anhydrite in quantities sufficient to decrease formation porosity and permeability. A significant conclusion is that, if co-injection can be accomplished, the costs associated with separation and disposal of the gases would be minimized. Knauss et al. (2005) used an analytical expression to calculate a 1-D radial Darcy flow that approximates the flow field. A single aqueous flow phase was used in which CO2 and other acid gases were dissolved. The simulations extended to only 100 years, whereas the alteration of primary minerals and precipitation of secondary carbonates are expected to take place over much longer time frames. Gunter et al. (2000) used a batch geochemical model to simulate the interaction of industrial waste streams comprising CO2 and H2S with the minerals in typical carbonate and sandstone aquifers from the Alberta Basin, Canada. The results show that these acid gases can be neutralized with formation of secondary minerals, such as calcite, siderite, anhydrite/gypsum and pyrrhotite. As expected, siliciclastic aquifers demonstrate better "mineral trapping" characteristics for CO2 than carbonate aquifers, because high-pressure CO2 enhances carbonate dissolution, and the availability of co-sequestering hydroxide components such as Mg(OH)2, Ca(OH)2 and Fe(OH)2 are less abundant in carbonates.
Mineral alteration rates due to injection of CO2 in deep aquifers are very slow, especially with respect to aluminosilicates, and extensive alteration is not expected during experimental studies in the laboratory. Experiments simulating the injection horizon environment have been conducted for up to several months in duration, with only minimal alteration and undetectable concentrations of secondary alteration phases. The problem can be circumvented in part through experimental studies at substantially elevated operating temperatures and pressures. Although such studies can provide useful insights, their value is constrained by dissimilarities in system rates, mechanisms, and reaction products. Experimental studies must therefore be supplemented by computer simulations with due consideration to model calibration using, where possible, both laboratory experimental data, and/or field data reflecting evolving or evolved natural systems.
In earlier papers, we performed a series of modeling studies in which various geochemical aspects of CO2 injection were investigated. In Xu et al. (2004), we presented a comparative analysis of the chemical interaction of aqueous solutions under high CO2 partial pressures with three different rock types; (1) glauconitic sandstone from the Alberta Sedimentary Basin, (2) an arenaceous sediment from the Gulf Coast of the United States, and (3) a dunite. Two subsequent studies focused on reactive transport simulations involving only a quartzose lithic arkose representative of the Frio Formation of the Texas Gulf Coast. The Frio formation is the shallowest of three reservoir quality sandstones and is found at depths ranging between 5000 and 20,000 ft., sufficient to ensure adequate CO2 densities for effective disposal. The first study involved CO2 injection in a 1-D radial region surrounding the injection well to analyze CO2 immobilization through carbonate precipitation (Xu et al., 2003); the second presented simulation results on mass transfer, mineral alteration, and consequent CO2 sequestration in the same arkose, when confined on either side by a reactive shale (Xu et al., 2005).
In this study, we present numerical simulations of the injection of CO2–H2S and CO2–SO2 mixtures into an arkose with similar hydrogeologic properties and mineral composition to that used in the preceding two studies. Our objectives are (1) to analyze changes in aqueous chemical composition, mineral alteration, acid-gas immobilization through precipitation, and changes in porosity induced by the injection, and (2) to compare modeling results with prior investigations and with limited field observations of analogous natural systems as a basis for validation. We use a fully coupled model of multiphase CO2 fluid flow into a saline aqueous (H2O + NaCl) phase, transport of aqueous species, and geochemical reactions. Reactive geochemical transport simulations are performed over a period of 10,000 years. The sensitivity of the model to differing dissolution rates and kinetic schemes on the evolution of the chemical system and on CO2 sequestration are also addressed.
Section snippets
Simulation method
The present simulations employed the nonisothermal reactive geochemical transport code TOUGHREACT (Xu and Pruess, 2001, Xu et al., 2006). This code introduces reactive chemistry into the multiphase fluid and heat flow code TOUGH2 (Pruess, 2004). More information on the TOUGHREACT can be found at the website (http://www-esd.lbl.gov/TOUGHREACT/). A new fluid property module, ECO2N, based on work by Spycher and Pruess (2005) was used, which provides an accurate description of the thermophysical
Fluid flow conditions
A single-layer uniform sandstone formation with a thickness of 10 m is considered in the present model (Fig. 1). Hydrological parameter specifications of the formation are chosen to be representative of those of the Texas Gulf Coast at a depth of about 2 km (Table 2). The formation is assumed to extend infinitely in the horizontal direction. A radial grid is used with spacing that increases away from the well.
This fluid flow simplification does not consider non-uniform sweep that may occur as a
Acid-gas simulations
The output of the simulations conducted for this study is very extensive, essentially consisting of three categories of information: (1) the aqueous phase composition, (2) the distribution of primary and secondary minerals, and (3) the physical properties of the system (such as porosity). For convenience, we present selected information in graphical form as a function of radial distance from the well bore, and at discrete time intervals of 10, 100, 1000, and 10,000 years. Although the objective
Model limitations
The simulations presented in this paper provide valuable insights regarding the chemical consequences of co-injecting H2S and SO2 with CO2 in the subsurface environment. The results are, however, constrained by the limitations of current reactive chemical transport simulators to replicate all of the complex physical, hydrological and geochemical processes that are expected to occur, but have not been incorporated in the underlying models. Some omissions can be justified, because they would not
Earlier modeling
The present model results are generally consistent with those by Knauss et al. (2005). The co-injection of H2S, compared to injection CO2 alone, does not significantly affect pH distribution and the mineral alteration pattern, whereas the co-injection of SO2 results in a substantially different pH distribution and mineral alteration pattern. The formation of sulfate-bearing minerals such as anhydrite will change the reservoir porosity and permeability. Formation of dawsonite and silica phases
Findings and recommendations
We have developed a conceptual model for injection of CO2 with H2S or SO2 in a sandstone formation, using hydrogeologic properties and mineral compositions commonly encountered in U.S. Gulf Coast sediments. We have performed six simulations of acid-gas injection into a 1-D radial region of an arkosic formation surrounding an injection well, at 200 bar pressure and 75 °C. Major findings are as follows:
- (1)
The co-injection of H2S, compared to injection CO2 alone, does not significantly affect pH
Acknowledgements
We thank Pat Dobson and Nicolas Spycher for reviews of the manuscript. We appreciate John Kaszuba and the anonymous reviewer for their comments during the review process, which greatly improve the quality of the paper. This work was partly supported by the Director, Office of Science, Office of Basic Energy Sciences, and partly by the Zero Emission Research and Technology project (ZERT), of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231 with Lawrence Berkeley National
References (103)
- et al.
Aquifer disposal of CO2: hydrodynamic and mineral trapping
Energy Convers. Manag.
(1994) - et al.
Deep injection of acid gas in Western Canada, Chapter 48 in Underground Injection Science and Technology
- et al.
Dependence of labradorite dissolution kinetics on CO2(aq), Al(aq), and temperature
Chem. Geol.
(2005) - et al.
Kinetics and mechanism of polythionate oxidation to sulfate at low pH by O2 and Fe3+
Geochim. Cosmochim. Acta
(2003) - et al.
Aquifer disposal of CO2-rich gases: reaction design for added capacity
Energy Convers. Manag.
(1993) - et al.
Technical and economic feasibility of CO2 disposal in aquifers within the Alberta Sedimentary Basin, Canada
Energy Convers. Manag.
(1996) - et al.
Aquifer disposal of acid gases: modelling of water–rock reactions for trapping of acid wastes
Appl. Geochem.
(2000) - et al.
SUPCRT92: a software package for calculating the standard molal thermodynamic properties of minerals, gases, aqueous species, and reactions from 1 to 5000 bars and 0 to 1000 degrees C
Comput. Geosci.
(1992) - et al.
Diagenesis of siliceous oozes — I. Chemical controls on the rate of opal-A to opal-CT transformation — an experimental study
Geochim. Cosmochim. Acta
(1977) - et al.
Carbon dioxide reaction processes in a model brine aquifer at 200 °C and 200 bars: implications for geologic sequestration of carbon
Appl. Geochem.
(2003)
Experiment evaluation of mixed fluid reactions between supercritical carbon dioxide and NaCl brine: relevance to the integrity of a geologic carbon repository
Chem. Geol.
Gas–water–rock interactions in sedimentary basins: CO2 sequestration in the Frio Formation, Texas, USA
J. Geochem. Explor.
Evaluation of the impact of CO2, co-contaminant gas, aqueous fluid, and reservoir rock interactions on the geologic sequestration of CO2
Chem. Geol.
Sleipner vest CO2 disposal — injection of removed CO2 into the Utsira Formation
Energy Convers. Manag.
Sulfur isotopic effects in the disproportionation reaction of sulfur dioxide in hydrothermal fluids: implications for the δ34S variations of dissolved bisulfate and elemental sulfur from active crater lakes
J. Volcanol. Geotherm. Res.
Chemical weathering rate laws and global geochemical cycles
Geochim. Cosmochim. Acta
Mineralogical and geochemical consequences of the long-term presence of CO2 in natural reservoirs: an example from the Springerville-St. Johns Field, Arizona, and New Mexico, U.S.A
Chem. Geol.
The effect of sulfate on aluminum concentrations in natural waters: some stability relations in the system Al2O3–SO3–H2O at 298 K
Geochim. Cosmochim. Acta
Ferric iron-bearing sediments as a mineral trap for CO2 sequestration: iron reduction using sulfur-bearing waste gas
Chem. Geol.
Natural occurrences as analogues for the geological disposal of carbon dioxide
Energy Convers. Manag.
Dissolution kinetics of calcite, dolomite, and magnesite at 15 °C and 0 to 50 atm pCO2
Chem. Geol.
Calculation of multicomponent chemical equilibria and reaction processes in systems involving minerals, gases and aqueous phase
Geochim. Cosmochim. Acta
Quartz solubility at low temperatures
Geochim. Cosmochim. Acta
Experimental investigation of CO2–brine–rock interactions at elevated temperature and pressure: implications for CO2 sequestration in deep-saline aquifers
Fuel Process. Technol.
Experimental investigation of aluminum–silica aqueous complexing at 300 °C
Chem. Geol.
Experimental and simulation studies on mineral trapping of CO2 with brine
Energy Convers. Manag.
CO2–H2O mixtures in the geological sequestration of CO2: II. Partitioning in chloride brines at 12–100 °C and up to 600 bar
Geochim. Cosmochim. Acta
A new kinetic approach to modeling water–rock interaction: the role of nucleation, precursors and Ostwald ripening
Geochim. Cosmochim. Acta
Silicate complexes of aluminum(III) in aqueous systems
Coord. Chem. Rev.
Biogenic silica dissolution in sediments of the Southern Ocean. I Solubility
Deep-Sea Res.
The Ladbroke Grove–Katnook carbon dioxide natural laboratory: a recent CO2 accumulation in a lithic sandstone reservoir
Energy
Simulation of reactive transport of injected CO2 on the Colorado Plateau, Utah, USA
Chem. Geol.
Numerical simulation to study mineral trapping for CO2 disposal in deep aquifers
Appl. Geochem.
Mineral sequestration of carbon dioxide in a sandstone–shale system
Chem. Geol.
TOURGHREACT: a simulation program for non-isothermal multiphase reactive geochemical transport in variably saturated geologic media
Comput. Geosci.
Silica activity and the smectite–illite reaction
Geology
A review of hazardous chemical species associated with CO2 capture from coal-fired power plants and their potential fate during CO2 geologic storage
Lawrence Berkeley National Laboratory Report LBNL-59731
Assessment of feldspar solubility constants in water in the range 0° to 350 °C at vapor saturation pressures.
Am. J. Sci.
Diagenesis and reservoir quality of the Aldebaran Sandstone, Denison Trough, east-central Queensland, Australia
Sedimentology
Postdepositional history of the Permian Sequence in the Denison Trough, Eastern Australia
Am. Assoc. Pet. Geol. Bull.
Continental-scale magmatic carbon dioxide seepage recorded by dawsonite in the Bowen–Gunnedah–Sydney Basin system, Eastern Australia
J. Sediment. Res., Sect. A Sediment. Pet. Proc.
The basic aluminum sulphates
J. Chem. Soc.
Aluminosilicate phases during initial clay formation: hydrogen ion–aluminum–oxalic acid–silicic acid–sodium system
J. Soil Sci.
Scarbroite, Al5(OH)13CO3·5H2O, compared with gibbsite and hydrotalcite
Min. Mag.
Synthesis, characterization, and energetics of solid solution along the dolomite–ankerite join, and implications for the stability of ordered CaFe(CO3)2
Am. Mineral.
Studies on alkaline earth sulfites — VIII. Thermal behaviour of Ca3(SO3)2(SO4)·12H2O and its solid solution in calcium sulfite tetrahydrate and crystallographic data on dehydrated salts
Rev. Chim. Miner.
The interrelation between gas and oil relative permeabilities
Prod. Mon.
Role of convective mixing of the long-term storage of carbon dioxide in deep saline formations
Aquifer disposal of CO2-rich greenhouse gases: extension of the time scale of experiment for CO2-sequestering reactions by geochemical modeling
Mineral. Petrol.
Equilibrium and structural studies of silicon(IV) and aluminum(III) in aqueous solution. 15. A potentiometric study of speciation and equilibria in the aluminum(3+)–carbon dioxide(g)–hydroxide system
Acta Chem. Scand., Ser. A, Phys. & Inorg. Chem.
Cited by (355)
Two-dimensional modeling of CO<inf>2</inf> storage in high heat flow areas: Insights from the Las Tres Virgenes geothermal field, Mexico
2024, International Journal of Greenhouse Gas ControlCarbon capture and sequestration technology for environmental remediation: A CO<inf>2</inf> utilization approach through EOR
2024, Geoenergy Science and EngineeringThe impact of secondary silicate mineral precipitation kinetics on CO<inf>2</inf> mineral storage
2024, International Journal of Greenhouse Gas Control