Kentucky Geological Survey Report of Investigations

Abstract

The Kentucky Geological Survey’s Marvin Blan No. 1 well was drilled in east-central Hancock County, Ky., about 4 mi southwest of the Ohio River, to demonstrate CO2 injection in the Western Kentucky Coal Field, following the mandate and partial funding from Kentucky’s House Bill 1, August 2007. Installation of a groundwater monitoring well was required as a condition of obtaining a U.S. Environmental Protection Agency Underground Injection Control Class V Permit prior to drilling the Blan well; however, no groundwater was encountered under the Blan well site. The groundwater monitoring well was immediately plugged and abandoned in accordance with State regulations, and the UIC permit was amended to require monitoring of two domestic water wells and two developed springs within approximately 2 mi of the Blan well site. Drilling of the Blan well commenced in April 2009 and was completed in June 2009.

Testing CO2 injection and storage was completed in two phases during 2009 and 2010. The Blan well penetrated an unfaulted Early Pennsylvanian through Neoproterozoic stratigraphic section characteristic of western Kentucky north of the Rough Creek Graben. Minor hydrocarbon shows were encountered during drilling. Whole-diameter 4-in. cores were recovered from the Late Devonian New Albany Shale, Late Ordovician Maquoketa Shale and Black River Group, Middle Cambrian–Lower Ordovician Knox Group (Beekmantown Dolomite, Gunter Sandstone, and Copper Ridge Dolomite), and Precambrian Middle Run Sandstone. Electric logs recorded in the Marvin Blan No. 1 can serve as type logs for western Kentucky. Structural dip in the well was found to be homoclinal, dipping approximately 0.5° west above the Knox unconformity, 1° west in the Knox Group and Eau Claire Formation, and about 3.5° north in the Middle Run.

The Knox Group, the target interval of the well, has a complex lithology including fabric-preserving primary dolomite and fabric-destructive secondary dolomite, vugfilling saddle dolomite, vug-lining chert, chert nodules and fracture fills, and nodular to disseminated pyrite in the Beekmantown, Gunter, and Copper Ridge dolomite facies, and fine-grained quartz sand with dolomite cement in the sandstone facies of the Gunter. CO2 storage capacity of the Knox was evidenced by reservoir properties of porosity and permeability and the injection testing programs.

Reservoir seals were evaluated in the Knox and overlying strata. Within the Knox, permeabilities measured in vertical core plugs from the Beekmantown and Copper Ridge Dolomites suggest that intraformational seals may be problematic. Three stratigraphic intervals overlying the Knox in the Marvin Blan No. 1 well may provide seals for potential CO2 storage reservoirs in western Kentucky: the Wells Creek Formation, Black River Group, and Maquoketa Shale. The Wells Creek and Black River had permeabilities suggesting that these intervals may act as secondary sealing strata. The primary reservoir seal for the Knox, however, is the Maquoketa. The Maquoketa is a dark gray, calcareous, silty, fissile shale. Maximum seal capacity calculated from permeabilities measured in vertical core plugs from the Maquoketa exceeded the net reservoir height in the Knox by about two orders of magnitude. Rock strength measured in core plugs from the Maquoketa suggests that any CO2 migrating from the Knox would likely have sufficient pressure to fracture the Maquoketa.

Phase 1 injection testing used 18,454 bbl of synthetic brine and 323 tons of CO2 (equivalent to 1,765 bbl of fluid or 5,646 mcf of gaseous CO2), and phase 2 injection testing used a total of 4,265 bbl of synthetic brine and 367 tons of CO2 (2,000 bbl of liquid or 6,415 mcf of gaseous CO2). Calculating the reservoir volume required to store a volume of supercritical CO2 used data provided by wireline electric logs, analysis of whole and sidewall cores, wireline temperature and pressure surveys, and analysis of formation waters collected prior to injection tests. The most likely storage capacities calculated in the Knox in the Marvin Blan No. 1 ranges from 450 tons per surface acre in the phase 2 Gunter interval to 3,190 tons per surface acre for the entire Knox section. At the completion of testing, the injection zone in the Marvin Blan No. 1 well was permanently abandoned with cement plugs in accordance with Kentucky and U.S. Environmental Protection Agency regulations.

Regional extrapolation of CO2 storage potential based on the results of a single well test can be problematic unless corroborating evidence can be demonstrated. Core analysis from the Knox is not available from wells in the region surrounding the Marvin Blan No. 1 well, although indirect evidence of porosity and permeability can be demonstrated in the form of active saltwater-disposal and gas-storage wells injecting into the Knox. This preliminary regional evaluation suggests that the Knox reservoir may be found throughout much of western Kentucky. The western Kentucky region suitable for CO2 storage in the Knox is limited updip, to the east and south, by the depth at which the base of the Maquoketa lies above the depth required to ensure storage of CO2 storage in its supercritical state and the deepest a commercial well might be drilled for CO2 storage. The resulting prospective region has an area of approximately 6,000 mi2, beyond which it is unlikely that suitable Knox reservoirs may be developed. Faults in the subsurface, which serve as conduits for CO2 migration and compromise sealing strata, may mitigate the area with Knox reservoirs suitable for CO2 storage.

The data from the Marvin Blan No. 1 well make an important contribution to understanding the subsurface strata in western Kentucky, and clarify relationships between electric-log responses, lithology, and rock properties, and effectively demonstrate the CO2 storage potential of the Knox and sealing capacity of the Maquoketa. The results of the injection tests in the Blan well, however, provide a basis for evaluating supercritical CO2 storage in Cambrian-Ordovician carbonate reservoirs throughout the Midcontinent.

Publication Date

2014

Series

Series XII

Report Number

Report of Investigations 25

Digital Object Identifier (DOI)

https://doi.org/10.13023/kgs.ri25.12

Funding Information

Project funding for drilling and testing the Marvin Blan No. 1 well in 2009 was provided by the Commonwealth of Kentucky, the Western Kentucky Carbon Storage Foundation, the Illinois Office of Coal Development, and the U.S. Department of Energy, National Energy Technology Laboratory. Additional testing of the Marvin Blan No. 1 well in 2010 was made possible by a U.S. Department of Energy grant from the American Recovery and Reinvestment Act to the University of Illinois and several partners, including the Kentucky Geological Survey, to further evaluate the CO2 storage potential of deep saline reservoirs underlying much of the Midwest.

Included in

Geology Commons

Share

COinS