GEOCHEMICAL CONSIDERATIONS FOR SUPERCRITICAL CO2 INJECTION AT CITRONELLE OIL FIELD, MOBILE COUNTRY, ALABAMA

Geological sequestration has been proposed as a promising method for reduction of net CO₂ emissions from coal-fired power plants. The petroleum industry has used CO₂-flood for decades as a method for enhanced oil recovery (EOR) in aging petroleum reservoirs. Injection of CO₂ at existing oil fields therefore has the potential to return a dual benefit to society. However, present understanding of storage of CO₂ in sedimentary basins does not permit confident prediction of the permanence of CO₂ sequestration or of the geochemical reactions likely to be induced between the injected supercritical CO₂, the formation fluids and the reservoir rock materials. The goal of this study is to produce and test a predictive model of fluid-rock interactions induced in a sandstone reservoir by the injection of supercritical CO₂.

The Citronelle Oil Field, located in northern Mobile Country, Alabama, was previously selected as the site for a pilot study of supercritical CO₂ injection as a means of EOR and carbon sequestration led by Peter Walsh at the University of Alabama at Birmingham. The current study dovetails with this other DOE/NETL-sponsored project, which began injection of CO₂ at the end of November 2009. Approximately 170 MMbbl of oil have been produced from fluvial sands of the Lower Cretaceous Rodessa Formation since 1955 by primary and secondary methods. The Rodessa Formation is bounded above by a thick sequence of anhydrite and shale, and is therefore considered to be an excellent site for safe CO₂ storage.

Aqueous fluids collected in late July from three producing wells situated closest to the injection well are presumed to represent pre-injection samples, based on flow modeling by Walsh et al. which predicts that only one third to one half of the first 7,500 ton slug of CO₂ had been injected by late July 2010. The aqueous formation fluids are saline, with Na concentrations ranging from 4,000-30,000 ppm and Ca concentrations ranging from 1,800-16,000 ppm. A second set of samples will be collected in late August 2010. Preliminary geochemical models are being run to calculate mineral saturation indices and establish the pre-injection chemical characteristics of the produced aqueous fluids. It is expected that it may take up to one year to see breakthrough of the first CO₂ slug in the producing wells nearest the injection well.