1
Shale Fracturing: The Geology And Technology That Sustained The Boom
Welcome!The webinar will begin promptly at Noon (CST)
Dr. Jon OlsonProfessor ChairmanPetroleum and GeosystemsEngineering DepartmentThe University of Texas at Austin
Dr. Mukul SharmaProfessor "Tex" Moncrief ChairDepartment of Petroleum and Geosystems EngineeringThe University of Texas at Austin
Dr. Zoya HeidariAssistant ProfessorDepartment of Petroleum and Geosystems EngineeringThe University of Texas at Austin
2
Shale Fracturing: the geology and technology that led to the boom
Dr. Jon E. OlsonChair and Professor
Petroleum & Geosystems Engineering
3
Outline• why is fracturing important• what is a hydraulic fracture• lessons from geology• fracture mechanics tests in the lab• small-‐scale fluid injection experiments to demonstrated hydraulic fracture complexity
4
2011 – horizontal well, 15 fracs
1996 – vertical well, single frac
Barnett Shale: Technology Matters
without fracturing
Dr. Jon Olson
5
US Natural Gas Productionfrom US Energy Information Agency, through July 2015
Dr. Jon Olson
6
What is a hydraulic fracture?
7
Rock Failure Modes: dispelling myths
unconfined compressive test
hydraulic fracturing in acrylic
Dr. Jon Olson
8 from slb.com
SHmax,maximum horizontal
stress
Shmin,minimum horizontal
stress
Gross Fracture geometry is systematic & predictable
Dr. Jon Olson
9
Hydraulic Fracture Geometry
Horizontal well with multiple transverse fractures
Horizontal well with longitudinal fracture
Vertical well with vertical fracture
10
Lessons on fracture propagation and interaction
from geology
11
Muddy Gap, WY
S∝H
thick beds 11Jon OlsonUT-‐Austin, FRAC
Systematic natural fractures
1212
Impact of lithology on fracture growth
• natural fracture spacing in shales often closer than other lithologies
• depends strongly on mineral make-‐up of rock
• ductile clay layers can be fracture arrestors within more brittle shale interbeds
Dr. Jon Olson
1313
Impact of lithology on fracture growth
• natural fracture spacing in shales often closer than other lithologies
• depends strongly on mineral make-‐up of rock
• ductile clay layers can be fracture arrestors within more brittle shale interbeds
Dr. Jon Olson
1414
Impact of lithology on fracture growth
ductile claybarrier
Huron Shale, Ohio
fracture arrest
15
Fracture Interaction with Bedding Planes
• siliceous mudstone, Miocene Monterrey Fm.
• vertical quartz-filled fracture selectively propagates across gray beds (marly), along white beds (phosphatic)
1 cm
from Portuguese Bend, Palos Verdes, CADr. Jon Olson
16
1 cm
from Portuguese Bend, Palos Verdes, CA
• siliceous mudstone, Miocene Monterrey Fm.
• vertical quartz-filled fracture selectively propagates across gray beds (marly), along white beds (phosphatic)
Fracture Interaction with Bedding Planes
Dr. Jon Olson
17
Natural-‐natural fracture interaction
Jon Olson, UT-Austin, FRAC
Miocene Monterey Formation, Palos Verdes, CA
Dr. Jon Olson
18
Diverting along the interface of thicker fracture
Natural-‐natural fracture interaction
Jon Olson, UT-Austin, FRAC
Miocene Monterey Formation, Palos Verdes, CA
Dr. Jon Olson
19
Fracture crossing of thinner frac
Natural-‐natural fracture interaction
Jon Olson, UT-Austin, FRAC
20
Natural-‐natural fracture interaction
bedding plane
Jon Olson, UT-Austin, FRAC
Miocene Monterey Formation, Palos Verdes, CA
Dr. Jon Olson
21
Natural-‐natural fracture interaction
Miocene Monterey Formation, Palos Verdes, CA
bedding plane
bedding-‐parallel vein
Jon Olson, UT-Austin, FRAC
Dr. Jon Olson
22
Natural-‐natural fracture interactionMiocene Monterey Formation,
Palos Verdes, CA
bedding plane
bedding-‐parallel vein
cross-‐bed vein #1
Jon Olson, UT-Austin, FRAC
Dr. Jon Olson
23
Natural-‐natural fracture interactionMiocene Monterey Formation,
Palos Verdes, CA
bedding plane
bedding-‐parallel vein
cross-‐bed vein #1
cross-‐bed vein #2a
cross-‐bed vein #2b
Jon Olson, UT-Austin, FRAC
Dr. Jon Olson
24Jon Olson, UT-Austin, FRAC
Natural-‐natural fracture interaction
Miocene Monterey Formation, Palos Verdes, CA
bedding plane
bedding-‐parallel vein
cross-‐bed vein #1
cross-‐bed vein #2a
cross-‐bed vein #2b
Diverting along the interface of thicker fracture
Dr. Jon Olson
25
Natural-‐natural fracture interactionMiocene Monterey Formation,
Palos Verdes, CA
bedding plane
bedding-‐parallel vein
cross-‐bed vein #1
cross-‐bed vein #2a
cross-‐bed vein #2b
Fracture crossing of thinner frac
Jon Olson, UT-Austin, FRAC
26
Natural Fracture Summary
• natural veins fluid-‐driven fractures occurring at depth (i.e., good analogy for hy-‐frac)
• bedding planes and pre-‐existing veins can divert fracture propagation (frictional interfaces would do same)
• vein thickness increases chance of diverting propagation
Dr. Jon Olson
27
Fracture Mechanics Testing of Cores –
Marcellus Shale
28
(a)
Marcellus Core Testing• test vein strength,
fracture interaction with Marcellus Core
• saw and grind samples rather than plug to reduce damage/breakage
• propagate induced fractures using Semi-‐Circular Bending test
fromLee, Olson, Holder, Gale and Myers, 2015, JGRdoi:10.1002/2014JB011358
29
(a) (b)
(c)
(d)
• test vein strength, fracture interaction with Marcellus Core
• saw and grind samples rather than plug to reduce damage/breakage
• propagate induced fractures using Semi-‐Circular Bending test
fromLee, Olson, Holder, Gale and Myers, 2015, JGRdoi:10.1002/2014JB011358
Marcellus Core Testing
30
(a) (b)
(c)
(d)
(e)
• test vein strength, fracture interaction with Marcellus Core
• saw and grind samples rather than plug to reduce damage/breakage
• propagate induced fractures using Semi-‐Circular Bending test
fromLee, Olson, Holder, Gale and Myers, 2015, JGRdoi:10.1002/2014JB011358
Marcellus Core Testing
31
(a) (b)
(c)
(d)
(e)
• test vein strength, fracture interaction with Marcellus Core
• saw and grind samples rather than plug to reduce damage/breakage
• propagate induced fractures using Semi-‐Circular Bending test
fromLee, Olson, Holder, Gale and Myers, 2015, JGRdoi:10.1002/2014JB011358
Marcellus Core Testing
32
(a) (b)
(c)
(d)
(e)
Marcellus Core Testing
fromLee, Olson, Holder, Gale and Myers, 2015, JGRdoi:10.1002/2014JB011358
• test vein strength, fracture interaction with Marcellus Core
• saw and grind samples rather than plug to reduce damage/breakage
• propagate induced fractures using Semi-‐Circular Bending test
33
• sample diameter = 2-‐4 in• vein thickness = 0.01-‐0.075 in• failure along flaws in calcite
vein-‐fill (fluid inclusion trails and cleavage)
Failure Occurs Within the Cement
failed SCB sample
thin section of calcite vein fill
plane polarized light crossed nicolsDr. Jon Olson
34
Multiple Saw Cuts & GrindingDiameter = 2.5 inThickness of the vein= 0.009 in
(a) (b)
(c) (d)
SCB test Resultscrossing preferred at more orthogonal approach angle
35
0.075 in 0.05 in
0.035 in 0.01 in
(a) (b)
(c) (d)
Impact of Vein Thicknesscrossing preferred for thinner veins
36
Testing summary
• demonstrated that well-‐cemented veins can provide planes of weakness
• tests at varying approach angles can quantify vein strength• critical energy release rate of veins ~ ¼ shale matrix
• vein toughness, KIc, was estimated to be higher than shale
• failure depends on strength and stiffness of veinsKIC(vein)~0.8 MPa-‐m1/2 KIC(shale)~0.5 MPa-‐m1/2
Dr. Jon Olson
37
Small-‐scale Laboratory Hydraulic Fracture
Experiments
38
Hydraulic fracturing in fractured reservoirs
Fisher et al 2004 Warpinski & Teufel, 1987
39
Physical Experiments: Interaction between cemented flaws and fluid driven cracks
Hydraulic Injection Tube (Wellbore)
hydrostone = gypsum-based cement39
Dr. Jon Olson
40
Natural Fractures (glass slides)
Sh,max
Pour Hydrostone Blocks with Embedded Discontinuities
Sh,max Natural Fractures (glass slides)
Dr. Jon Olson
41
Load Frame – apply Svert, SHmax, Shmin
Flatjacks
Hydraulic Injection Tube (wellbore)
Flatjack Pressure Lines
Dr. Jon Olson
42
Uncovering Hydraulic Fractures
42Dr. Jon Olson
43
Complex Interaction – Oblique Case
wellbore
natural fracture
43Dr. Jon Olson
44
Complex Fracture propagation/interaction
Bahorich et al., SPE 190197
45
Complex Fracture propagation/interaction
Bahorich et al., SPE 190197
46
Complex Fracture propagation/interaction
diverts along natural fracture
mixed mode I-‐II non-‐planar curving
Bahorich et al., SPE 190197
47
Complex Fracture propagation/interaction
part of fracture propagates under natural fracture
diverts along natural fracture
mixed mode I-‐II non-‐planar curving
Bahorich et al., SPE 190197
48
Summary• shale gas has made energy more affordable and secure in the United States
• much of hydraulic fracture complexity has an analogy in natural fracture examples
• one key to complexity is the interaction of hydraulic fractures with natural fractures
• laboratory testing can be used to quantify pre-‐existing fracture strengths and to run small-‐scale hydraulic fracture tests to illustrate potential geometries
Dr. Jon Olson
49
Acknowledgment: FRAC Consortium Sponsors
http://www.beg.utexas.edu/frac
50
51
Well Spacing, Fracture Spacing, Sequencing and Fluid Management in Pad
Fractured Horizontal Wells
Mukul M. SharmaUniversity of Texas at Austin
Dr. Mukul Sharma
52
JIP FACT SHEETSponsors for 2014-‐15
•RPSEA / DOE: 2 projects for $2,400,000 for 2013-2016•The State of Texas pays faculty salaries.•Your funds are leveraged about 50:1
Other potential members that have expressed interest:• PTT• Petrobras•ONGC• Chesapeake
1. Air Liquide2. Anadarko Petroleum3. Aramco4. Baker Hughes5. BHPBilliton6. BP7. Chief Oil & Gas8. Chevron9. ConocoPhillips10. Devon Energy11. Eni12. Ferus13. FSTI Inc.*14. Hess15. Linde16. Lubrizol*17. MeadWestVaco (MWV)
18. Nalco (Ecolabs Inc.)*19. Nexen Energy ULC20. Noble Energy*21. Oxy22. Pioneer23. PEMEX24. Praxair25. Range Resources26. Sanchez Oil & Gas*27. Schlumberger28. Shell29. Southwestern Energy30. Statoil31. Talisman32. Unimin33. Weatherford34. Wintershall
53
The People Involved
Murtadha Al Tammar Haotian WangPrateek Bhardwaj Chu-‐Hsiang WuMichael Brothers Weiwei WuChris Blyton Mingyuan YangEric C. Bryant Shiting YiMichael Carey Jason YorkDeepen Gala Peng ZhangChang Min Jung Junhao ZhouHojung Jung Saud Alquwizani*Emmanouil Karantinos Samarth Agarwal*Ashish Kumar Saptaswa Basu*Dongkeun Lee Stephen A. Bryant*Hisanao Ouchi Jameson P. Gips*Javid Shiriyev Jongsoo Hwang*Igor Shovkun Eric R. Lehman*Kaustubh Shrivastava Lionel Ribeiro*Jeffrey Stewart Roman Shor*Sanjay Surya Do Shin*
Research Staff:§ Yaniv Brick§ Philip Cardiff§ Ajeetha Kamilla§ Jin Lee§ Ripudaman Manchanda§ Anand S. Nagoo§ Rodney T. Russell
Graduate Students:
54
Top Five Ideas Worth Trying
1. Treat fracture design as a multi-frac, multi-well problem. This will provide the optimum,– Well spacing, fracture spacing, fracture sequencing– Fracture design
2. Liquids lifting through entire life of well– Wellbore trajectory– Artificial lift design
3. Better zipper frac sequencing4. Improve proppant placement5. Refracturing
– New methods to divert fluids during refracturing– Better candidate selection
Dr. Mukul Sharma
55
Pad Fracturing: The Big PictureLearning from Experience
• There is no general consensus on most recommendations regarding well spacing, proppants, pumping rates, fluids, flowback…….– Too many variables– Too few wells and too little data– Expensive learning
• There is a real and significant financial benefit to accelerating the learning process (capital efficiency).
• How do we accomplish this?– Learn from existing wells (data analysis)– Physics based models– Combine the two (iterate).
Dr. Mukul Sharma
56
Multiple Non-‐Planar Fracture PropagationFracture Stage with 4 Perforation Clusters
Q
q1
q2
q3
q4
57
Signatures of Fracture Complexity / Interference
Bakken
Barnett
Eagle Ford
Fracture Trajectory vs. Distance from Stage 1 (ft) Net Closure Pressure vs. Stage Number
Ref: Roussell and Sharma, 2012, ARMA 12-‐633
58
Complex Fracture Networks
Ref: Warpinski, N.R. et al, SPE 114173, 2008
In addition to stress shadowing, fracture complexity also arises due to:q Complex rock fabricq Natural fracturesq Bedding planesq Heterogeneity
q Stress anisotropyq Shear failure
Stress interference, natural fractures, heterogeneities, pore pressure depletion, rock fabric can all lead to fracture complexity. Dr. Mukul Sharma
59
Bossier TGS, Anadarko (Sharma et al. 2004)
Barnett Shale, Devon
(Fisher et al. 2005)
Fracture Complexity is a Strong Function of Rock Fabric and Local Stress Contrast
Dr. Mukul Sharma
60
Effect of Rock Fabric and Stress Anisotropy
Ref: SPE 173374-‐ Arbitrary Fracture Propagation in Heterogeneous Poroelastic Formations Using a Finite Volume-‐Based Cohesive Zone Model • Eric C. Bryant, M. M. Sharma, 2015.
61
We Can Control Fracture Complexity, To Some Degree
Wellbore
Fracture 1
Stress Reversal Region
Reoriented σhmaxdirection
Region of Low Stress Contrast
σhmax – σhmin(psi)
Wellbore
Stress Reversal Region
Decreasing σhmin
away from the fracture
σhmin(psi)
Fracture 1
σhmin is increased close to the propped-‐open fracture and exceeds the original σhmax value causing reorientation of the σhmax direction.
Fractures that propagate in regions of low stress contrast are likely to show more fracture complexity.
SPE 159899
62
Pad Fracturing Design Workflow
3-‐D Fracture Model: Estimate fracture dimensions /
complexity
3-‐D Fracture Interference Model: Estimate number of fractures per
stage
3-‐D Reservoir Model:Simulate production, reservoir drainage
and NPV.
Parametric Study: Well spacing, fracture spacing and fracture dimensions.
Fracture Design Recommendations: Sand volume, fracture sequencing,
fluids, proppant schedule.
63
Estimating Optimum Well Spacing and Fracture Spacing (Lf = 160 ft)
Dr. Mukul Sharma
64
• There are many reasons to treat the fracture design problem on a pad scale (a multi-‐well, multi-‐fracproblem):– Interference between fractures – Fracture complexity / reservoir heterogeneity– Some ability to control fracture complexity
• Computing pore pressure, stress and failure maps is important for:– Establishing well spacing – Selecting locations for infill or step-‐out wells – Selecting fracture spacing in the new well– Fracture designs in new wells– The feasibility of an infill well
Summary
65
Top Five Ideas Worth Trying
1. Treat fracture design as a multi-frac, multi-well problem. This will provide the optimum,– Well spacing, fracture spacing, fracture sequencing– Fracture design
2. Liquids lifting through entire life of well– Wellbore trajectory– Artificial lift design
3. Better zipper frac sequencing4. Improve proppant placement5. Refracturing
– New methods to divert fluids during refracturing– Better candidate selection
Dr. Mukul Sharma
66
Wellbore Liquids Management
q Well productivity is a strong function of wellbore liquids loading and wellbore trajectory
q Every unconventional well will be on artificial lift for 90% of its life.
q To properly manage wellbore fluids we must:
q Unload liquids from fractures and the wellbore
q Unload liquids from the reservoir matrix
q Integrate wellbore models with reservoir inflow
q Properly design and manage artificial lift.
q Obtain good estimates of BHP from THP
67
How Good Are Our Wellbore Models?
Cousins, Denton and Hewitt (1965) – Exp. # 49, data also in Table 12.1 of Wallis’s One Dimensional Flow textbook
Dr. Mukul Sharma
68
Run numbers 7.02 to 7.09 in the reference: Hewitt, G. F., King, I., Lovegrove, P. C.: Holdup and pressure drop measurements in the two-‐phase annular flow of air-‐water mixtures, UK AERE Report R3764, June (1961)
Run numbers FHOPI-‐46 to FHOPI-‐49 in the reference: Crowley, C. J., Sam, R. G., Rothe, P. H.: Investigation of two-‐phase flow in horizontal and inclined pipes at large pipe size and high gas density, Project PR-‐172-‐507, Pipeline Research Committee, AGA, February (1986)
How Good Are Our Wellbore Models?
Dr. Mukul Sharma
69
Owen, D. G.: An experimental and theoretical analysis of equilibrium annular flows, Ph.D. Dissertation, Dept. of Chem. Eng., U. of Birmingham (1986)
Also available from Theofanous, T. G., Amarasooriya, W. H.: Dataset no. 1 -‐pressure drop and entrained fraction in fully developed flow, Multiphase Sci. and Tech., v. 6, part 1, pp. 5-‐13 (1992)
How Good Are Our Wellbore Models?
70
Kumar, N.: Improvements for flow correlations for gas wells experiencing liquid loading, Soc. of Petroleum Engineers, Paper No. 92049 (2005)
Chierici, G. L., Ciucci, G. M., Sclocchi, G.: Two-‐phase vertical flow in oil wells -‐prediction of pressure drop, Soc. of Petroleum Engineers, Paper No. 4316, August (1974)
How Good Are Our Wellbore Models?
Dr. Mukul Sharma
71
Slightly-‐Inclined Flow: Horizontal Wells
Ø In long laterals, segmentation is important because of the need to Ø Capture the local multiphase
flow developmentØ Different wellbore regions and
reservoir zonesØDifferent well inflow from multiple entry points with productivity indices (PIs)
ØDifferent lithologies, rock qualities, stress gradients
Ajayi et al.: Stimulation Design for Unconventional Resources, Schlumberger Oilfield Review, Summer (2013) Dr. Mukul Sharma
72
Toe-‐Up WellborePFF Simulator
Dr. Mukul Sharma
73
Correct trend is captured for both pressure and gas
holdup
Toe-‐Up Wellbore: PFF SimulatorComparison with Well Data
74
Undulating WellborePFF Simulator
Dr. Mukul Sharma
75
Undulating WellborePFF Simulator
76
Liquids Unloading
q Obtaining good estimates of BHP from THP
q Effect of wellbore trajectory on well productivityq Unloading of liquids from fractures
q Integrating wellbore models with reservoir inflow
q Unloading of liquids from the reservoir matrixq Design of artificial lift.
Dr. Mukul Sharma
77
• Competing forces : drawdown vs. gravity vs. capillary
• Capillary forces inhibiting flow of water
• Gravity pulling liquid to the bottom
• Turner’s critical velocity for vertical gas wells:
• For σ = 60 dynes/cm, ρL=58 lb/ft3, ρg = 3 lb/ft3,
Ucrit = 8.4 ft/s !
Water flowing downward, assisted by gravity and
drawdown
Water flowing upward at high gas rates and downward at
low gas rates
Loading in fracture?? High Sw, Low gas rel perm
Liquid Loading Inside Fracture
Dr. Mukul Sharma
78
Can the Gas Lift the Liquid in the Fracture?
Dr. Mukul Sharma
79
Integrating Wellbore Flow with Reservoir Inflow
Ø To model the impact of wellbore trajectory on hydraulically-‐fractured horizontal well productivity it is important to account for the effect of water unloading in the wellbore the fracture and the rock matrix
80
Integrating Wellbore Flow with Reservoir Inflow
Dr. Mukul Sharma
81
Liquids Removal Takes Time
Property Value
Matrix Permeability 1 µD
Fracture Permeability 2 D
Drawdown 2000 psi
Variation of water saturation inside Fracture Variation of water saturation in matrix near fracture
82
Water and Condensate BlockingØ Water /condensate blocking can cause a severe reduction in gas and condensate relative perms.
Ø Chemical treatment using a non-ionic fluorinated surfactant increases gas and condensate relative permeability by a factor of 2.
Ø Proppant can be treated as well to improve proppant-pack conductivity.
Ø 6 field trials conducted. More underway.
83
References• “Flow-‐Through Drying of Porous Media”, AIChE Journal, July, 2006, J.
Mahadevan, M. M. Sharma and Y.C. Yortsos.• “Water Removal from Porous Media by Gas Injection: Experiments and
Simulation” Transport in Porous Media Journal, 2007, J. Mahadevan, M. M. Sharma, and Y. Yortsos.
• “Evaporative Clean-‐up of Water-‐Blocks in Gas Wells”, Journal of Petroleum Technology, pp 46-‐68, October 2005, J. Mahadevan, M. M. Sharma and Y.C. Yortsos.
• “Factors Affecting Clean-‐up of Water Blocks: A Laboratory Investigation,” SPE Journal, September 2005, J. Mahadevan, M.M. Sharma.
• “Evaporative Clean-‐up of Water-‐Blocks in Gas Wells”, SPE 94215, presented at the 2005 SPE Production and Operations Symposium held in Oklahoma City, OK, April 17-‐19, 2005, J. Mahadevan and M.M. Sharma
• “Cleanup of Water Blocks in Depleted Low-‐Permeability Reservoirs”, SPE 89837 presented at the SPE Annual Technical Conference and Exhibition, Houston, Texas, 26–29 September 2004, B. Parekh, and M.M. Sharma.
84
Top Five Ideas Worth Trying
1. Treat fracture design as a multi-frac, multi-well problem. This will provide the optimum,– Well spacing, fracture spacing, fracture sequencing– Fracture design
2. Liquids lifting through entire life of well– Wellbore trajectory– Artificial lift design
3. Better zipper frac sequencing4. Improve proppant placement5. Refracturing
– New methods to divert fluids during refracturing– Better candidate selection
Dr. Mukul Sharma
85
Shale Fracturing: The Geology And Technology That Sustained The Boom
Q&APlease enter your questions in the chat box on the left.
Dr. Jon OlsonProfessor ChairmanPetroleum and GeosystemsEngineering DepartmentThe University of Texas at Austin
Dr. Mukul SharmaProfessor "Tex" Moncrief ChairDepartment of Petroleum and Geosystems EngineeringThe University of Texas at Austin
Dr. Zoya HeidariAssistant ProfessorDepartment of Petroleum and Geosystems EngineeringThe University of Texas at Austin
86
Thank you!To receive CEU verification, please visit the link below and complete a short survey. You MUST click on this link and fill out the information now to be eligible for CEU verification (only takes 30 seconds).
https://www.surveymonkey.com/r/6FFFRJYIf you did not already pay the $25 fee for CEU verification, please call 512-‐232-‐5199 or 512-‐471-‐3506 to pay the fee within 24 hours
Dr. Jon Olson: [email protected]
Dr. Mukul Sharma: [email protected]
Dr. Zoya Heidari: [email protected]
Comments? [email protected]