Assessing geochemical reactions during CO2 injection into an oil-bearing reef in the Northern Michigan basin. (January 2019)

• Record Type: Journal Article
• Authors: Welch, Susan A.; Sheets, Julie M.; Place, Matthew C.; Saltzman, Matthew R.; Edwards, Cole T.; Gupta, Neeraj; Cole, David R.
• Is Part Of: Applied geochemistry. Volume 100(2019)
• Page Start: 380
• Page End: 392
• Publication Date: 2019-01
• Subjects: Niagaran reef -- Isotope geochemistry -- Carbon sequestration -- Geochemical monitoring; Environmental geochemistry -- Periodicals; Water chemistry -- Periodicals; Geochemistry -- Social aspects -- Periodicals; Geochemistry -- Periodicals; 551.9
• DOI: 10.1016/j.apgeochem.2018.12.008
• Languages: English
• ISSNs: 0883-2927

Abstract

Abstract: The Dover 33 Reef, part of the Niagaran Reef Complex in Northern Michigan (USA), has been the focus of an enhanced oil recovery/carbon capture utilization and storage (EOR/CCUS) project as part of the Phase III-Midwest Regional Carbon Sequestration Partnership (Gupta et al., 2013a, b). The Dover 33 structure has experienced significant CO2 flooding in the past two decades, and over the course of the current injection study (between February 2013 and July 2016) has received approximately 100 to 1000 tonnes/day into the central injection well (L-M) 1–33). As part of the geochemical monitoring effort of the study, gas and fluid samples were collected from Dover 33 reef, and several other nearby reef structures, to assess the impact of CO2 injection on the geochemical processes occurring within in the reef. The injected gas is composed of approximately 95% CO2, with a δ 13 CCO2 of ∼20.5‰, which is consistent with previously published compositions of Antrim shale gas, the source of the CO2 . The concentrations and isotopic compositions of higher pressure gas collected from the L-M 5–33 monitoring well were similar to those measured in the injection well, but did exhibit a small but systematic shift in isotopic composition towards lower values over the course of the study, suggesting mixing and dilution between the gas in the reservoir and the injected gas. In contrast the δ 13 CCO2 of gas samples from the monitoring well with the lower surface pressure, L-M 2–33, are consistently lower throughout the study, ∼18.5‰, indicating that reactions with the injected CO2 are occurring within the reef or with the well casing. Fluid samples were collected to assess the extent of interaction among the injected gas, the reservoir rock, and the brine. The brine samples are acidic (pH ∼ 4.1 to 4.9) with a total salt content of nearly 400 g/L. Analysis of the isotopic composition of dissolved inorganic carbon (DIC) in the Dover 33 brine shows that δ 13 C is higher than the injected gas (27–33‰) suggesting that the gas is not in equilibrium with DIC in the brine, and that there has been little isotopic exchange with carbonate minerals in the reef. The water isotope composition of the brine, δ 18 O and δD, plot below the meteoric water line, indicating that the water is not of recent meteoric origin and has undergone isotopic exchange with both gas and minerals within the reef structure. The 87 Sr/ 86 Sr ratios of the brine samples range from 0.70865 to 0.70869, consistent with Silurian seawater composition. Geochemical modelling of the brine composition shows that the predicted CO2 solubility as DIC is much greater than the measured DIC, and that the brines are supersaturated with respect to carbonate minerals, suggesting the potential for significant trapping of CO2 in both dissolved and mineral form. Highlights: The Dover 33 Reef has been the focus of an EOR/CCUS project as part of the MRCSP. The composition of gas within the reef structure closely reflects the injected gas. δ 18 O reflects isotopic exchange between the injected gas and the brine. Geochemical modelling suggest the potential for CO2 trapping in dissolved and mineral form.