© 2015 University of North Dakota Energy & Environmental Research Center.
Enhanced Oil Recovery (EOR) in Tight Oil: Lessons Learned from Pilot Tests
in the Bakken
Tight Oil Optimization Workshop Calgary, Alberta, Canada
March 12, 2015
James Sorensen John Hamling
EERC Research Program Partners
Additional Support Sponsoring Partners
2
www.bakkendispatch.com
Today’s Objectives
• Review of publicly available records – Five North Dakota Bakken injection tests – Elm Coulee Bakken injection test
• Lesson’s learned
3
Injection Tests
Five North Dakota Bakken injection tests are known: 1. #9660: Water, tested March–April 1994 (50 days)
2. #16713: CO2, tested September–October 2008 (29 days)
3. #17170: Water, tested April–May 2012 (?? days)
4. #16986: Waterflood followed by field gas injection
– Waterflood, tested April 2012 – February 2014 (672 days)
– Field gas, tested June–August 2014 (54 days?)
5. #24779: Vertical CO2, test began February 11, 2014 (duration unknown)
• All Class II wells. Details of tests are limited.
4
Injection Test Locations
5
#9660 – Water Injection Test
• Meridian Oil Company. • Converted existing horizontal well. • Freshwater injection into Upper Bakken Shale. • Injection began March 8, 1994: Shut in April 27, 1994 (50 days).
– “shut in for approximately 1–2 months to evaluate its performance.”
• Request to put back on pump July 19, 1994. – “test was found to be unsuccessful.”
Injected Water Volumes • March 1994: 7616 bbl (avg. 1389 psi). • April 1994: 5644 bbl (avg. 1096 psi).
6
#9660 – Water Injection Test
7
0
2000
4000
6000
8000
10000
12000
Mar-89 Aug-90 Dec-91 May-93 Sep-94 Jan-96 Jun-97
Oil,
bbl
Date
Monthly Production, #9660
Monthly Oil
Injection Test Duration
#16713 – CO2 Injection Test • EOG. • Fractured (April 2008) with sand and gel, no report of multistages; however,
well diagram shows six packers in production zone. • Permit includes a detailed injection plan.
– Planned 60-day soak time with return to production; later altered to 30 days.
– Food-grade CO2 from Praxair. • Injection began September 15, 2008: CO2 injection completed on
October 14, 2008 (29 days). • After 11-day injection, breakthrough occurred 1 mile away in an offset well.
Injected CO2 volumes:
– “September 2008: 5010 bbl.” – “October 2008: 4862 bbl.”
• No posttest results; no records of any kind after March 2010.
8
#16713 – CO2 Injection Test
0
5000
10000
15000
20000
25000
30000
35000
Aug-07 Feb-08 Sep-08 Mar-09 Oct-09 May-10 Nov-10 Jun-11 Dec-11 Jul-12 Jan-13 Aug-13 Mar-14 Sep-14
Oil,
bbl
Date
Monthly Oil Production, #16713
Monthly Oil
Injection Test Duration
On Pump
Well Returned to Sales
9
#16713 CO2 Injection Test Offset Wells
“After injecting CO2 for 11 days into the Austin 1-02H (#16713), we have begun to see breakthrough from Austin 1-02H (#16713) to the Austin 2-03H(#16768), over a mile away. The other offset wells we are monitoring, the Austin 9-11H (#17075) and the Bruhn 1-12H (#17475), have yet to show an increase in CO2 concentrations.
The concentration observed in the Austin 1-02H (#16713) has increased from a background reading of 5000 ppm the week before injection began and during the first days of the injection to approximately 25,000 ppm. Based on our calculations this translates to approximately
4 Mcfd of the approximately 1000 Mcfd we are injecting into #16713.”
10
#16713 CO2 Injection Test: Offset Wells
11
#16713 CO2 Injection Test: Gas Production in Offset Well #16768
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
Aug-07 Apr-08 Dec-08 Aug-09 May-10 Jan-11 Sep-11 May-12 Jan-13 Oct-13 Jun-14
Gas
, Mcf
Date
Monthly Gas Production in Offset Well #16768
Monthly Gas
Injection Test Duration
Additional production from injection test?
12
#17170 – Water Injection Test • EOG. • Fractured (August 2008) with sand and gel, no report of multistage; however,
seven packers are illustrated in the well diagram. • Taken off production April 22, 2012. • Produced water injection test.
– “Huff ‘n’ puff.” • Injection test began May 3, 2012: No available notes on completion of test.
– Contradictory injection dates listed on state Web site. • Planned 30-day injection with 10-day soak.
– Cycle to repeat until deemed uneconomical; returned to production. • August 20, 2012, additional reserve pits were installed to collect fracture sand • Requested “low-pressure injection through artificial lift” on October 12, 2012
(sundry notice), i.e., artificial lift was initiated. • No newer records.
Injected Produced Water Volumes • April 2012: 10,380 bbl. • May 2012: 28,797 bbl.
13
0
5000
10000
15000
20000
25000
Jun-08 Dec-08 Jul-09 Jan-10 Aug-10 Feb-11 Sep-11 Apr-12 Oct-12 May-13 Nov-13 Jun-14
Oil,
bbl
Date
Monthly Oil Production, #17170
Monthly Oil
"Low Pressure Injection"
Injection Start
#17170 – Water Injection Test
14
#16986 – Water and Field Gas Injection Tests
• EOG – Middle Bakken horizontal. – Well currently listed as “Inactive gas injector.”
• Time line: – Spudded January 28, 2008. – Began producing in April 2008. – Fractured June 2008 (sand and gel, no note of multistage OR presence of production
packers). – On pump late July 2008. – September 2008 applied for permit for recompletion and injection of food-grade CO2.
♦ Permit approved late October 2008. ♦ Permit rescinded October 2009. ♦ No evidence to suggest conversion occurred.
– December 2011, request for conversion to EOR injection well (produced water injection, “waterflood pilot”); approved February 2012.
– Water injection began April 16, 2012. ♦ Periodic injection until February 2014. ♦ No additional details in well file.
– Returned to production in March 2014.
Cumulative Injected Produced Water Volumes (North Dakota Industrial Commission [NDIC])
♦ 438,969 bbl
15
#16986 – Field Gas Injection Test • Time line, continued:
– June 2014 requested change to gas injection. – Test consisted of injection of field gas with some produced water injection. – Water used to “manage effects of gas mobility in the fracture system” or, if needed, “build
system pressure with less gas volume.” – Goal “evaluate and test the technical feasibility and production performance results of injecting
produced gas into the Bakken Formation for the purposes of secondary recovery.”
– Injection began June 27, 2014. – Appeared to have communication with the production well. – Injection data provided through August 20, 2014: Injection end date unknown (ongoing?). – Injected field gas volumes:
♦ June 2014: 4598 Mcf ♦ July 2014: 50,871 Mcf ♦ August 1–20, 2014: 33,260 Mcf ♦ Cumulative total: 88,729 Mcf
– No posttest production data available.
16
0
5000
10000
15000
20000
25000
30000
Apr-08 Oct-08 May-09 Dec-09 Jun-10 Jan-11 Jul-11 Feb-12 Aug-12 Mar-13 Oct-13 Apr-14
Volu
me,
bbl
Date
Monthly Production, #16986 Montly Oil
Monthly Water
Injection Test Duration
CO2 Injection Permit
Approved
CO2 Permit Rescinded
Request Waterflood
Injection Permit;
Approved Feb. 2012
Field Gas
Injection Start
#16986 – Water and Field Gas Injection Tests
0
10,000
20,000
30,000
40,000
50,000
60,000
Apr-12 Jun-12 Aug-12 Nov-12 Jan-13 Apr-13 Jun-13 Sep-13 Nov-13 Feb-14
Wat
er In
ject
ed, b
bl
Date
Waterflood, Monthly Injection Volumes #16986
Water Injected
Waterflood Injection Start
Resume Production
17
Monthly Oil Monthly Water
#16986 – Water and Field Gas Injection Tests: Offset Wells
18
#16986 – Waterflood: Production from Offset Well #16461
19
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
0
5000
10000
15000
20000
25000
Jan-07 Sep-07 Jun-08 Feb-09 Oct-09 Jun-10 Feb-11 Nov-11 Jul-12 Mar-13 Nov-13 Jul-14
Volu
me,
bbl
Date
Offset Well, #16461, Monthly Production
Monthly Oil
Monthly Water
Monthly Gas
Gas
Pro
duce
d, M
cf
Waterflood Injection
0
10
20
30
40
50
60
Apr-12 Oct-12 May-13 Nov-13
Wat
er In
ject
ed,
Mbb
l
Date
Water Injected #16986
WaterInjected
#16986 – Water and Field Gas Injection Tests: Offset Wells
20
0
1000
2000
3000
4000
5000
6000
7000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
Aug-06 Apr-07 Jan-08 Sep-08 May-09 Jan-10 Sep-10 Jun-11 Feb-12 Oct-12 Jun-13 Mar-14
Volu
me,
bbl
Date
Offset Well, #16346, Monthly Production
Monthly OilMonthly WaterMonthly Gas
Gas
Pro
duce
d, M
cf
#16986 – Waterflood: Production from Offset Well #16346
21
Waterflood Injection
0
10
20
30
40
50
60
Apr-12 Oct-12 May-13 Nov-13
Wat
er In
ject
ed,
Mbb
l
Date
Water Injected #16986 WaterInjected
0
5000
10000
15000
20000
25000
30000
Apr-08 Oct-08 May-09 Dec-09 Jun-10 Jan-11 Jul-11 Feb-12 Aug-12 Mar-13 Oct-13 Apr-14
Volu
me,
bbl
Date
Monthly Production, #16986 Montly Oil
Monthly Water
Injection Test Duration
CO2 Injection Permit
Approved
CO2 Permit Rescinded
Request Waterflood
Injection Permit;
Approved Feb. 2012
#16986 – Water and Field Gas Injection Tests
0
10,000
20,000
30,000
40,000
50,000
60,000
Apr-12 Jun-12 Aug-12 Nov-12 Jan-13 Apr-13 Jun-13 Sep-13 Nov-13 Feb-14
Wat
er In
ject
ed, b
bl
Date
Waterflood, Monthly Injection Volumes #16986
Water Injected
Waterflood Injection Start
Resume Production
Field Gas
Injection Start
22
Monthly Oil Monthly Water
#16986 – Water and Field Gas Injection Tests
0
500
1000
1500
2000
2500
3000
3500
4000
0
200
400
600
800
1000
1200
1400
1600
1800
2000
6/25/2014 7/2/2014 7/9/2014 7/16/2014 7/23/2014 7/30/2014 8/6/2014 8/13/2014 8/20/2014
Gas
Inje
cted
, Mcf
Date
#16986 Daily Field Gas Injection Volume and Pressure
Daily Volume
Daily Pressure
Pres
sure
, psi
g
Uncertainty in Reported
Pressure Values
23
#16986 – Water and Field Gas Injection Tests
0
10,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
90,000
100,000
6/25/2014 7/2/2014 7/9/2014 7/16/2014 7/23/2014 7/30/2014 8/6/2014 8/13/2014 8/20/2014
Gas
Inje
cted
, Mcf
Date
#16986 Cumulative Field Gas Injection Volume
24
#16986 – Water and Field Gas Injection Tests
“Gas injection operations began on the Parshall 20-03H (#16986) on June 27, 2014, which
represented the first day where we had consistent gas injection rate. On July 2, 2014, the
Patten 1-02H (#16461), which is one of three wells on the 1280 EOR pilot area, had gas
production of 177 Mcf and oil production of 33 bbl. Preinjection GOR for this well was
approximately 400 scf/bbl; therefore, we would estimate that of the 177 Mcf produced on
this day, 164 Mcf was incremental as a result of gas injection operations. To mitigate the
volume of gas channeling through to the Patten 1-02H (#16461), our first operational course of
action was to reduce the VFD speed of the pump to help build bottom hole pressure in this well.
On 7 /3 we continued to observe instantaneous gas rates on the Patten 1-02H (#16461)
and these rates were showing an escalation from the previous day. We decided to stop
the pump on the Patten 1-02H (#16461) and operate this well on an as needed basis. We
feel this will help mitigate the volume of gas that it being cycled from injection to surface and
help build BHP in the injection well.”
25
#16986 – Water and Field Gas Injection Tests: Offset Wells
26
0
50
100
150
200
250
300
0
20
40
60
80
100
120
6/18 6/28 7/8 7/18 7/28 8/7 8/17 8/27
Oil,
bbl
Date
Monthly Production, #16461
Daily OilDaily Gas
Gas
, Mcf
#16986 – Field Gas Production from Offset Well #16461
0
2
4
6
8
10
12
6/18/14 7/8/14 7/28/14 8/17/14
Wat
er, b
bl
Date
#16461 Daily Water Production
16461 Water
27
No notable production response was observed in
other offset wells over this time period.
#24779 – CO2 Injection Test • Whiting. • Class II, vertical well, inactive. • Drilled as a stratigraphic test well.
– Collected 366’ of core. – No production.
• Test designed to see if the formation can accept CO2 gas. • Planned for 20-day test.
– Cemented production casing. – Planned to use packers to isolate the Middle Bakken zone. – Planned injection of 10 Mscf of CO2. – Short soak period (days). – Produce well, collect samples, and reinject all fluids except samples.
• Injection reported to start on February 11, 2014. • Results unavailable/unknown.
28
Elm Coulee Test in Montana
DFN From NW McGregor (Mission Canyon)
From NW McGregor (Mission Canyon)
• Conventional huff ‘n’ puffs respond within days or weeks.
• Response at Elm Coulee appears to have taken months.
0
200
400
600
800
1000
1200
1400
BB
LS O
il/M
onth
Production Month
Burning Tree – State 36-2H Oil Production
Oil Production
CO₂ Huff 'n’ Puff Test Period
Period of Possible Incremental Oil
Recovery?
Put back on pump
• Could the delayed response be a reflection of the dominance of diffusion as a mechanism for CO2 movement in the Bakken?
• Future field tests need robust baseline characterization, injection, and monitoring data to determine fate and effect of CO2 in the reservoir.
29
What Did We Learn?... • Water and gas injectivity into various lithofacies of the Bakken petroleum
system has been demonstrated.
• Production responses to injection were observed, which indicates that fluid mobilization can be influenced.
• Laboratory results suggest potential for high mobilization under the right conditions.
– A more complete understanding of these conditions can be gained from field tests.
Engineered tests within a well-characterized geologic setting will
help inform successful injection programs by providing fundamental knowledge needed to dovetail lab studies, geologic
models, and reservoir simulation. This will allow evaluation of EOR scenarios that can guide more successful pilot and field EOR
development. Greg Latza Photography
30
Recap
Andrew Burton / Getty Images
Too few data exist for the six injection tests performed in the Bakken to perform thorough engineering and geologic analysis,
nor are the designs or test objectives fully understood.
Unless we are able to gain insight into these previous tests, we will need to develop a basic understanding of injectivity and reservoir performance in order to better engineer and adapt
successful EOR programs in unconventional reservoirs.
Unconventional reservoirs will require an unconventional approach
31
Contact Information
Energy & Environmental Research Center University of North Dakota 15 North 23rd Street, Stop 9018 Grand Forks, ND 58202-9018 World Wide Web: www.undeerc.org Telephone No. (701) 777-5287 Fax No. (701) 777-5181 James Sorensen, Senior Research Manager [email protected] John Hamling, Senior Research Manager [email protected]
32
Acknowledgment This material is based upon work supported by the U.S. Department of Energy
National Energy Technology Laboratory under Award No. DE-FC26-08NT43291.
Disclaimer This presentation was prepared as an account of work sponsored by an agency of the
United States Government. Neither the United States Government, nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes
any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed or represents that its use would not
infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors
expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.
33
