A binational effort between the United States and Canada is under way to characterize the lowermost aquifer system in the Williston and Alberta Basins of the northern Great Plains prairie region of North America in the United States and Canada. This 3-year project, begun in 2011, is being conducted with the goal of determining the potential for geologic storage of carbon dioxide (CO2) in rock formations of the 1.34 million-km2 Cambro-Ordovician saline system (COSS). The focus of this report is to evaluate and discuss geochemical modeling and laboratory studies performed by the Energy & Environmental Research Center to determine potential chemical reactions between CO2, brine, and rock on the portion of the COSS that occurs in North Dakota, Montana, and South Dakota. Although the modeling and laboratory activities were conducted independently, the results of the two different activities were compared to each other to establish a greater understanding of the validity and applicability of the modeling and laboratory approaches. The geochemical modeling study was performed using publicly available PHREEQC software and databases. Rock samples, mineralogy, and water analysis data for both the sandstone injection target and the shale cap rock were also obtained from publicly available sources. The laboratory-based exposure tests entailed exposure of various COSS rock samples to CO2 for 28 days at formation pressure and temperature. The results of the geochemical modeling were consistent with existing literature, and suggested that because most of the COSS comprises quartz-rich sandstone, much of the rock matrix will be nonreactive. Reactions can, however, occur with secondary components (clays, carbonates, micas, K-feldspar) that can be contained within the sandstone and the heterogeneous mixed lithology zones between the primary sand layers. The geochemical modeling study predicted that a geochemical effect from the interaction of CO2 with the COSS minerals and formation water was the dissolution of calcite and concurrent formation of dolomite. The source of Mg2+ for this reaction was either from Mg2+ contained within secondary formation minerals, such as illite, phlogopite, celadonite, and clinochlore, or from Mg2+ in the formation water. The modeling calculations also indicated a potential reaction of the CO2 with illite and K-feldspar in the formation. The K-feldspar was predicted to decompose into quartz, clay, and carbonates, thus trapping the CO2 in a mineral form. vi The results of the laboratory experiments generally compared favorably with the modeling portion of the study. The analytical data generated from the exposed samples show a variable mix of concentrations of K-feldspar and a general trend of decreases in illite. Illite and Kfeldspar behaviors were generally in agreement with the geochemical modeling results. The most significant reactions occurred between CO2 and dolomite/calcite and glauconite in the sandstone. Glauconite contained within samples completely dissolved and decomposed during the exposure experiments. The decomposition of glauconite will form siderite and quartz as well as some ions that will remain in solution. This phenomenon in glauconite-rich areas may increase local permeability as well as provide a mineral-trapping mechanism for CO2. Samples of shale that were exposed to CO2 showed no change in morphology or chemistry, with the exception of halite precipitation. The formation of halite that was seen is most likely an artifact of the samples being dried (not rinsed) after exposure to CO2 and brine solutions. The modeling calculations and laboratory experiments both suggest that CO2 interactions with the COSS mineral phases (reservoir and cap rock) are not detrimental to CO2 storage. Large areal changes in porosity and permeability are not anticipated from the interactions of CO2 with the COSS. Minerals within the COSS that did react with CO2 are typically found in lower concentrations in the quartz-dominated sandstone or within the low-porosity cap rock. Any reactions with the cap rock are not likely to penetrate past the CO2–cap rock interface because of low porosity/permeability. Variations in formation water chemistry, mineral content, and porosity in the COSS can result in large variations in the amount of CO2 that can be trapped. These variations occur both geographically between different areas of the COSS and vertically at each location. Additional focused efforts are needed on both the modeling and rock–CO2 exposure fronts to better understand the potential effects of CO2 storage in the COSS. With respect to future modeling efforts, additional data are necessary for more robust calculations to address the effects of pressure, kinetics, and concentrated brines in the COSS. Laboratory-based CO2 exposure experiments could be improved by implementing more advanced sampling methodologies for highly heterogeneous rocks to ensure that observed differences in chemistry are accurate. Heterogeneity in the rock samples provides for challenging interpretation of results when minute changes in chemistry are observed. Improvements to the CO2 exposure methodology to allow for better detection of minute mineralogical changes within the rock fabric will greatly aid in the refinement of this experimental process.
- PDFWDP_D89_BasalCambrianGeochem_Final_Mar13.pdf