Citronelle Dome CO2 EOR Pilot Test and Opportunities For CO2 Storage
Citronelle Dome CO2 EOR Pilot Test and Opportunities For CO2 Storage
Citronelle Dome CO2 EOR Pilot Test and Opportunities For CO2 Storage
Peter M. Walsh, Konstantinos Theodorou, and P. Corey Shum University of Alabama at Birmingham, Birmingham, AL Ermson Z. Nyakatawa, Xiongwen Chen, Kathleen A. Roberts, and Latasha Lyte Alabama Agricultural and Mechanical University, Normal, AL Gary N. Dittmar, Keith Murphy, Steve Walker, Thomas Boelens, Pete Guerra, Tommy Miller, Tommy Henderson, Michael Sullivan, Danny Beasley, Steven Brewer, and Franklin Everett Denbury Resources, Inc., Plano, TX, and Citronelle, AL Jack C. Pashin, Denise J. Hills, and David C. Kopaska-Merkel Geological Survey of Alabama, Tuscaloosa, AL Richard A. Esposito and Kirk M. Ellison Southern Company, Birmingham, AL Eric S. Carlson, Peter E. Clark, Akand W. Islam, Csar A. Turmero, and Francis Dumkwu University of Alabama, Tuscaloosa, AL Shen-En Chen, Wenya Qi, Yangguang Liu, and Peng Wang University of North Carolina at Charlotte, Charlotte, NC Research Experience in Carbon Sequestration 2012 Birmingham, Alabama June 3-13, 2012
Discovered: Depth: Oil gravity: Available wells: Formation: Original oil in place: Cumulative oil production: Cumulative gas production: Secondary recovery (waterflood): CO2-EOR pilot test:
1955
~11,000
ft
42-46o API 413 Donovan Sands (Lower Cretaceous) 537 million barrels 170 million barrels 15 billion cubic feet since 1961 1981-1984
Carbon Sequestration Atlas of the United States and Canada, U.S. Department off Energy, Office of Fossil Energy, National Energy Technology Laboratory, 2007, p. 51.
1.0 E+07
8.0 E+06
2.0 E+06
1965
1975
1985 Year
1995
2005
2015
Oil production is approaching the ultimate recovery possible using conventional secondary recovery technology.
Denbury specializes in CO2 EOR Natural CO2 Supply at Jackson Dome, MS Proved plus probable tertiary oil reserves as of 12/31/11
Denbury Resources Inc., EOR: The Economically Viable CCS Solution, Presentation at the Indiana Center for Coal Technology Research, June 5, 2008. http://www.purdue.edu/discoverypark/energy/assets/pdfs/cctr/presentations/Denbury-CCTR-June08.pdf
The pipeline serves Eucutta, Soso, Martinville, and Heidelberg Fields An early plan was to extend the pipeline to Citronelle
http://www.denbury.com/interest-owner-relations/contact-owner-relations/default.aspx
http://www.industcards.com/st-coal-usa-al.htm
Plant Barry
2,657 megawatts (electric) 5 generating units firing coal 2 units firing natural gas ~10 million tons/year of CO2 from coal-fired generation
Southeast Regional Carbon Sequestration Partnership (SECARB) Phase III Anthropogenic CO2 Injection Field Test http://www.secarbon.org/files/anthropogenic-test.pdf
http://en.wikipedia.org/wiki/Pumpjack
J. C. Pashin and R. A. Esposito, "Citronelle Dome: A Giant Opportunity for Multi-Zone Carbon Storage and EOR in the Mississippi Interior Salt Basin of Alabama," Annual Convention of the Gulf Coast Association of Geological Societies and the Gulf Coast Section of the Society for Sedimentary Geology, Corpus Christi, TX, October 21-23, 2007
Rodessa Formation
Detailed description of the geology assists in the interpretation of field data and provides input to large-scale, fine grid reservoir simulations.
Sand 14-1
The inverted five-spot well pattern is shown shaded, with the injector in magenta at the center.
Sand 16-2
Production history since 2002 and simulations of oil yields from different CO2 and water injection plans.
Yellow: Red: 7,500 tons CO2, followed by water 7,500 tons CO2, followed by water, another 7,500 tons CO2, and water
CO2 Storage Tank and Pump CO2 supplied by Denbury from Jackson Dome
6000
4500
3000
1500
Replace tubing in injector
0 Dec 1 Jan 1 Feb 1 Mar 1 Apr 1 May 1 Jun 1 Jul 1 Aug 1 Sep 1 Oct 1
Continuous injection of CO2 began in January 2010 and continued without interruption at an average rate of 31 tons/day to September 2010, with a total of 8036 tons injected. The injection history was in good agreement with the prediction from the reservoir simulation by Eric Carlson using SENSOR (Coats Engineering, Inc.).
70
60
50
40
On injecting CO2, production at Tank Battery B-19-11, which receives oil from only one producer in the test pattern (Well B-19-11), continued for four months on the trajectory that it had been following for the previous 10 months. In May 2010, production at the battery abruptly declined, by approximately the typical production from Well B-19-11 (8 to 9 bbl/day), then continued a steady decline, with no significant response to the termination of CO2 injection and return to water injection in September 2010.
30
20
CO2
Water
10
0 1/09 4/09 7/09 10/09 1/10 4/10 7/10 10/10 1/11 4/11 7/11 10/11 1/12 4/12
Time (month/year)
The path from the injector to Well B - 1 9 - 11 , w h e r e e x c e s s i v e breakthrough occurred, coincides approximately with the direction of maximum horizontal compressive stress in the Southeast.
Slope of on the log-log plot is evidence for a vertical fracture intersecting the wellbore. Behavior at long times (approximately radial flow) gives permeability (0.4 millidarcy). Permeability and behavior at short times give fracture wall area and length (600 to 1000 ft).
70
60 50 40 30 20
CO2
10 0 1/09 25
Water
20
15
10
0 1/09 12,000
4/09
4/10
4/11
4/12
7/12
10,000
8,000
6,000
4,000
CO2
Water
2,000
0 1/09
4/09
4/10
4/11
4/12
7/12
Time (month/year)
500
Water
Shut-in
CO2
Water
400
300
200
100
5/1/2008
5/1/2009
5/1/2010
5/1/2011
5/1/2012
Time (month/day/year)
Water injection rate following CO2 injection is less than half of that before CO2 injection. Ratio of water to CO2 viscosity at reservoir conditions ~ 5. Treatments with surfactant and acid are planned, to restore injectivity.
Plant Barry produces ~ 10 million short tons CO2/year from coal-fired generation (~ 9 billion kg CO2/year). 10-year-long commercial CO2-EOR project would require ~ 9% of Plant Barrys CO2 production. Injection of CO2 into saline formations at Citronelle, or distribution by pipeline to other CO2 sinks, would be needed to sequester larger fractions of the CO2 produced at Plant Barry, and for longer times.
J. C. Pashin and R. A. Esposito, "Citronelle Dome: A Giant Opportunity for Multi-Zone Carbon Storage and EOR in the Mississippi Interior Salt Basin of Alabama," Annual Convention of the Gulf Coast Association of Geological Societies and the Gulf Coast Section of the Society for Sedimentary Geology, Corpus Christi, TX, October 21-23, 2007
At the rate of 10 million tons CO2/year, these formations could provide storage for 50 years.
Middle Donovan Sands: 24 - 100 million tons Donovan oil-bearing Sands: 115 - 460 million tons Total: 500 - 2000 million tons
Conclusions
The structure of Citronelle Dome and absence of faults make it an attractive site for CO2enhanced oil recovery and CO2 sequestration. Satisfactory response of three of the producing wells in the test pattern to injection of 8000 tons of CO2 reinforces the expectation of profitable CO2-EOR at Citronelle. Three unexpected problems were associated with the CO2 injection, requiring consideration in the design of a commercial CO2 flood: 1. The conventional power oil pumps in service at Citronelle were subject to excessive erosion/corrosion by particles and scale mobilized by CO2/water. The problem was resolved by installation of more robust pumps. 2. A hydraulic fracture intersecting the injector was evidently opened by water or CO2 injection during the test. Bypassing of CO2, probably through the fracture, caused a large deficit in incremental recovery from one of the producers. Active control of downhole pressures is proposed to minimize the opening of fractures. 3. Injectivity to water was low, following the injection of CO2. Treatments with surfactant and acid are planned, to determine the cause. The result may have significant bearing on the choice between continuous CO2 injection or water-alternating-gas recovery. The depleted oil reservoirs, and saline formations above the oil reservoirs, are expected to be capable of storage for 500 million to 2 billion tons of CO2, sufficient to accept CO2 produced at Alabama Power Plant Barry for 50 years.
Acknowledgments
This work is supported by: U.S. Department of Energy/National Energy Technology Laboratory Under Cooperative Agreement No. DE-FC26-06NT43029 Denbury Resources, Inc. Plano, TX, and Citronelle, AL Southern Company Services, Inc. Birmingham, AL Geological Survey of Alabama Tuscaloosa, AL Special thanks to: Richard A. Esposito and Mark S. Berry, Southern Company Services, Inc. Melinda M. Lalor, UAB Ronald T. Evans Jack C. Pashin, Geological Survey of Alabama Chandra Nautiyal, DOE/NETL
Minimum Miscibility Pressure for Citronelle oil is sufficient to guarantee a miscible flood.
Laboratory at the University of Alabama; viscometer is in the fume hood at center
CO2 in air measured monthly at 100 locations in and around the test site
Average and standard deviation of volume fractions of CO2 in ambient air near the ground across Citronelle
CO2 Breakthrough
On the morning of May 25, 2010, four months after the start of continuous CO2 injection, high pressure was detected in the liquid/gas separator at Tank Battery B-19-11, where oil from Well B-19-11 is gathered. Gas in the head space of the power oil tank was tested using a Draeger Tube, which indicated a high level of CO2. Produced gas samples, for detailed analysis, were collected on the same day.
__________________________________________________________________________________________________ C1 C2 C3 O2 N2 CO2 13CO2 vol% vol% vol% vol% vol% vol% per 1000 __________________________________________________________________________________________________ Ave. of two samples before CO2 injecHon 43.0 6.9 7.7 0.18* 29.7 2.7 -12.8 Ave. of two samples, May 25, 2010 1.2 0.8 1.9 0.06 0.4 90.4 -3.0 CO2 from Jackson Dome 0.005 0.005 0.02 0.05 0.2 99.5 -2.9 __________________________________________________________________________________________________ *O2 + Ar 13CO2 = [(13C/12C)sample/(13C/12C)reference - 1] x 1000
Analyses of Injected CO2 and Samples of Produced Gas from Well B-19-11, before and during CO2 InjecHon.
Conclusion: CO2 in produced gas at B-19-11 is breakthrough from injecHon at B-19-10 #2.
120
100
80
STB/D 60
40
20
CO2 injection
Water injecHon
CO2 injecHon
0 10160
10260
10360
10460
10560
10660
10760
10860
10960
CumulaHve-Aher-Federal-Income-Tax-Cash Flow behaviors versus Hme, for CO2 injecHon-slug sizes from 7500 to 90000 tons.
Phase
III:
Perform
economic
and
market
analysis
for
WAG
recovery
scenarios,
with
reservoir
simulaHon
validated
by
the
performance
of
the
Phase
II
-
Phase
III
WAG.
Lithology
Core from the CO2 injection well at an injection point (Sand 16-2)
Graphic representation of rock types, with thin-section photomicrographs. Width and color of core log indicate rock type: Brown: mudstone Yellow: sandstone Orange: conglomerate
Petrology - Sandstone
David Kopaska-Merkel, Geological Survey of Alabama
Porous sandstone containing poorly rounded and poorly sorted grains. Quartz grains are white, potassium feldspar grains are stained speckled yellow. Porosity is filled with blue epoxy and consists of a combination of primary interparticle pores and feldspar molds.
Gary Dittmar, Tommy Miller, Tommy Henderson, and Michael Sullivan, Denbury Resources
Gary Dittmar, Tommy Miller, Tommy Henderson, and Michael Sullivan, Denbury Resources
CO2 Injector: B-19-10 #2 Producers: B-19-7 B-19-8 B-19-9 Produced fluids to B-19-8 Tank Battery Producer: B-19-11 Produced fluids to B-19-11 Tank Battery
Denbury Resources
Vegetation Monitoring
Xiongwen Chen and Kathleen Roberts, Alabama A&M University
Does escape or seepage of CO2 affect plant growth? 10 m x 10 m test plot chosen near each well and tank battery, and the golf course (controls). Trees and shrubs tagged and their circumferences recorded. Sites are visited yearly to monitor species distribution and plant growth (450 specimens).
Kathleen Roberts and Xiongwen Chen recording plant species and stem circumferences in a test plot.
The figure shows the yearly average fractional increase in plant stem and trunk cross section area. Elevated CO2 in air or soil is expected to increase the growth rate. In all cases except one, the growth rate has declined since CO2 injection began in late 2009. The exception is site VP1, near the CO2 injector.
Soil Monitoring
Ermson Nyakatawa and Latasha Lyte, Alabama A&M University
Gas sample extracted through probe at depths up to 3 ft. Soil moisture measured at depths up to 3 ft. Gas flow at surface collected in canister for analysis. Soil thermometer. CO2 evolution from microbial and plant respiration observed year round.
Well Number
Rodessa Formation
Screen shot of a dynamic visualization of CO2 (green), oil (yellow), and water (blue) showing the evolution of their saturations during WAG recovery from the Donovan 14-1 Sand . The vertical lines are the injector (red) and the four producers (black). An aerial photo of the area surrounding the test pattern is shown at the top.
Caprock Properties
Breakthrough Pressure Critical pressure at onset of capillary flow Permeability Controls rate of flow after breakthrough Chemical Reactions May increase or decrease permeability Fracture Gradient May limit rate of CO2 injection or height of CO2 column
Figure by Jack C. Pashin, Geological Survey of Alabama, 2009.
Assurance of seal layer integrity is a key element in the process of gaining the confidence of plant managers, regulators, investors, and the public in geologic sequestration.
70 60 50 40 30 20 10 0
CO2
Water
70 60 50 40 30 20 10 0
CO2
Water
12000
10000 8000 6000 4000 2000 0 -2000 -4000 -6000 -8000 -10000 -12000 10/09 1/10 4/10
CO2
Water
7/10
10/10
1/11
4/11
7/11
10/11
1/12
4/12
7/12
Time (month/year)
http://www.industcards.com/st-coal-usa-al.htm
Plant Barry
2,657 megawatts (electric) 5 generating units firing coal 2 units firing natural gas ~10 million tons/year of CO2 from coal-fired generation
http://www.southerncompany.com/aboutus/about.aspx
http://www.industcards.com/st-coal-usa-al.htm
http://www.secarbon.org/wp-content/uploads/2011/05/Dittmar.pdf
Proposal for coping with fractures: Conduct a step-rate test to determine the fracture opening stress Install a sensor to monitor downhole pressure and adjust the CO2 injection pressure to prevent the fracture from opening
The path from the injector to Well B - 1 9 - 11 , w h e r e e x c e s s i v e breakthrough occurred, coincides approximately with the direction of maximum horizontal compressive stress in the Southeast.
Linear flow into (or out from) a high conductivity fracture Solution to the differential equation gives the pressure decay at the wellbore ~ t
Slope of on the log-log plot is evidence for a vertical fracture intersecting the wellbore. Behavior at long times (approximately radial flow) gives permeability (0.4 millidarcy). Permeability and behavior at short times give fracture wall area and length (600 to 1000 ft).