Brittleness and Rock Strength of the Bakken Formation,

Williston Basin, North Dakota

Jingqi Xu, Steve Sonnenberg

Bakken Consortium
Department of Geology and Geological Engineering
Colorado School of Mines
Outline

 Introduction

 Motivation

 Brittleness Index

 Rock Hardness

 Key Takeaways
Introduction
Williston Basin Map Stratigraphy

Sonnenberg and Pramudito, 2009 Johnson, 2014

Bakken Formation Geomechanical Features

1 inch 1 inch
Shear Fractures Natural Fractures Deformation Expulsion fractures
Motivation
Generally, a brittle shale play has high
probability to develop a large complex
fracture networks during hydraulic
fracturing.
• To understand the
brittleness/ductility of the Bakken
Formation
• To predict prospective hydraulic
fracturing candidates within the
Bakken Formation

R2=0.75
Tinnin, et al., 2014

• To develop an empirical relationship for a rapid,

accurate, and cost-efficient estimate of the rock
strength of the Bakken Formation via rebound
hammer
Zahm and Enderlin, 2010
Introduction
Two principle elastic parameters for evaluation of rock geomechanical
properties
3𝑉𝑉𝑉𝑉2−4𝑉𝑉𝑉𝑉2 2
1 𝑉𝑉𝑉𝑉 −2𝑉𝑉𝑉𝑉
2
Edynamic= ρ𝑉𝑉𝑠𝑠2 ( ) ν dynamic= ( 2 )
𝑉𝑉𝑉𝑉2−𝑉𝑉𝑉𝑉2 2 𝑉𝑉𝑉𝑉 −𝑉𝑉𝑉𝑉2
Well A Well A, Hognose, J Horst
Depth (ft) 10 30 50 70 0.15 0.25 0.35
10570 70
Upper Bakken Member
Upper
10580 Bakken ν Middle Bakken Member

60 Lower Bakken Member

10590 Pronghorn Member

Dynamic Young's Modulus (GPa)

10600 50

Middle
10610
Bakken 40
10620

30
10630

10640
Lower 20
Bakken
10650

Pronghorn 10
10660 0.1 0.2 0.3 0.4
Young's Modulus (Gpa) Poisson's Ratio Dynamic Possion's Ratio
Mineral Composition
Mass %
0 20 40 60 80 100
Lodgepole
10573
10576
Upper
10579
Bakken
10582
10588 Quartz

10591 K-feldspar
10594
Plagioclase
10597 Upper
10600 Calcite Bakken
10603
Dolomite
10606
10609 Ankerite/Fe-dolomite
Depth (ft)

10612
10615
Siderite Middle
10618 Pyrite Bakken Middle
10621 Marcasite Bakken
10624
Chlorite
10627
10630 Kaolinite
10633
Illite/Mica
10636
10639 Illite/Smectite (I/S)
10642
Lower
10645
Lower Bakken
10648 Bakken
10651
10654 Pronghorn
10657 Pronghorn
XRD (X-ray diffraction) results of Well A Three Forks ECS (elemental capture spectroscopy) log
Brittleness
Brittleness

If the rock has a large region of elastic

behavior but only a small region of
ductile behavior the rock is considered
brittle. In contrast, If the material has a
small region of elastic behavior and a
large region of ductile behavior,
absorbing much energy before failure,
it is considered ductile (opposite of
brittle).

Brittleness Index http://homepage.smc.edu/grippo_alessandro/structural.html

• Mineralogy-based Brittleness Index

• Elastic Log-based Brittleness Index

Higher the magnitude of the brittleness index, the more brittle the rock is
Brittleness Index
• Mineralogy-based Brittleness Index
𝑄𝑄𝑧𝑧
𝐵𝐵𝐵𝐵(Jarvie, et al. , 2007) =
𝑄𝑄𝑄𝑄 + 𝐶𝐶𝐶𝐶 + 𝐶𝐶𝐶𝐶𝐶𝐶

𝑄𝑄𝑄𝑄 + 𝐷𝐷𝐷𝐷𝑙𝑙
𝐵𝐵𝐵𝐵(𝑊𝑊𝑊𝑊𝑊𝑊g and Gale, 2009) =
𝑄𝑄𝑄𝑄 + 𝐶𝐶𝐶𝐶 + 𝐷𝐷𝐷𝐷𝐷𝐷 + 𝐶𝐶𝐶𝐶𝐶𝐶 + 𝑇𝑇𝑇𝑇𝑇𝑇

𝑊𝑊𝑄𝑄𝑄𝑄𝑄𝑄 + 𝑊𝑊𝑊𝑊𝑎𝑎 + 𝑊𝑊𝑊𝑊𝑜𝑜𝑙𝑙

𝐵𝐵𝐵𝐵(J𝑖𝑖n et al. , 2014) =
𝑊𝑊𝑇𝑇𝑇𝑇𝑇𝑇𝑎𝑎𝑎𝑎

𝑄𝑄𝑄𝑄 + 𝐶𝐶𝐶𝐶 + 𝐷𝐷𝐷𝐷𝑙𝑙 + 𝐹𝐹𝐹𝐹𝐹𝐹 + 𝑃𝑃𝑃𝑃

𝐵𝐵𝐵𝐵(this study) =
𝑄𝑄𝑄𝑄 + 𝐶𝐶𝐶𝐶 + 𝐷𝐷𝐷𝐷𝐷𝐷 + 𝐹𝐹𝐹𝐹𝐹𝐹 + 𝑃𝑃𝑃𝑃 + 𝐶𝐶𝐶𝐶𝐶𝐶 + 𝑇𝑇𝑇𝑇𝑇𝑇

TOC Determination
• Schmoker Equation (1983)
Brittleness Index
Geologic Depth Mineral Brittleness Index
Section (ft) 0 20 40 60 80 100
10570

Upper Bakken
10580
Shale

10590

10600

Middle Bakken Jarvie, 2007

10610
Member
Wang, 2009

10620 Jin, 2014

this study
10630

10640
Lower Bakken
Shale
10650

Pronghorn Member
10660
Brittleness Index
• Elastic Log-based Brittleness Index
1 𝐸𝐸−𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝑣𝑣𝑚𝑚𝑚𝑚𝑚𝑚−𝑣𝑣
𝐵𝐵𝐵𝐵 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝑟𝑟 𝑎𝑎𝑎𝑎𝑎𝑎 𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵, 2007 = [
2 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸−𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸
+ 𝑣𝑣 −𝑣𝑣𝑚𝑚𝑚𝑚𝑚𝑚
]
𝑚𝑚𝑚𝑚𝑚𝑚

Depth (ft) Young's Modulus (Gpa) Poisson's Ratio Mineral Brittleness Index Elastic Brittleness Index
10 30 50 70 0.15 0.25 0.35 0 20 40 60 80 100 0 20 40 60 80 100
10570

Upper
10580 Bakken
Shale

10590

10600

Jarvie,
Middle
10610 2007 Bakken
Wang, Member
2009
10620 Jin,
2014 Grieser,
This 2007
10630 study

10640 Lower
Bakken
Shale
10650

Pronghorn
10660
Brittleness Index

Rock brittleness is a function of various complex

70 parameters rather than simply elastic properties
and mineral compositions
60 • Rock anisotropy, temperature, fluid type and
diagenesis
50
• Rock deformability varies as a function of
Elastic BI Greiser

confining pressure
40
• Pore pressure in reservoir influence elastic
properties, high pore pressure makes the rock
30
less consolidated
20
• It is hard to agree with Young’s modulus and
y = 1.2668x - 42.969
R² = 0.751 Poisson’s ratio having the same share of
10 influence on rock brittleness as imposed by
20 40 60 80 100 arithmetic averaging in the equation
Mineral BI this study
Rock Strength
Laboratory Measurements
• Leeb rebound hardness measurement
• Uniaxial compressive measurement
• Triaxial compressive measurement
• Acoustic velocity measurement
Hardness Measurement
Leeb rebound hardness as
rock strength proxy parameter?

Equotip Bambino 2

Induced Fractures
Gillian, 2012

Unconfined Compression Test

• Frequently used to characterize the mechanical
behavior of rock
1 inch
• Primary index test for strength and deformability
of intact rock
• Most useful for rock comparison and classification
Hardness Measurement
Leeb Rebound Hardness (HLD)
500.00 600.00 700.00 800.00 900.00
10525.00 J. HORST 1-11H
Middle Bakken Member

10535.00

HLD vs. UCS

55.0
10545.00

Leeb 50.0
Depth (ft)

Rebound
10555.00 Hardness
45.0
UCS UCS (Mpa)

10565.00 40.0

35.0
10575.00 UCS= 0.0754*Leeb Rebound Hardness - 4.9794
R² = 0.21169
30.0
500 550 600 650 700 750
Leeb Rebound Hardness (HLD)
10585.00
0 20 40 60
UCS (Mpa)
Hardness Measurement
Leeb Rebound Hardness (HLD)
500.00 600.00 700.00 800.00 900.00
10525.00 J. HORST 1-11H
Middle Bakken Member

10535.00

HLD vs. Elastic BI

10545.00 75

Leeb
Depth (ft)

70
Rebound
10555.00 Hardness
Elastic BI Elastic BI 65

10565.00
60

55
10575.00
Elastic BI= 0.0448*Leeb Rebound Hardness + 35.793
R² = 0.22878
50
500 550 600 650 700 750
10585.00 Leeb Rebound Hardness (HLD)
40 50 60 70 80
Elastic BI
Hardness Measurement
Low correlation between variables:
• Reduced rock strength as a result of previous
sampling and geomechanical tests
• Errors exist during hardness measurements (e.g.,
edge effect, fractures, weathered samples )
• Errors derived from the UCS measurement
• Possible depth mismatch between UCS and Leeb
hardness measurements

Seeking more wells with UCS

data and intact core samples
Triaxial Measurement
Well: Danks 17-44H (no sonic log)
Interval: Lower Bakken Shale and Pronghorn Member
Confining pressure: 1200 psi
Estatic vs. HLD 𝑣𝑣static vs. HLD
50 0.30
Static Young's Modulus (GPa)

0.29
0.28

Static Poisson's ratio

40
y = -0.0213x + 31.342 0.27 y = -6E-05x + 0.2768
R² = 0.1028 0.26 R² = 0.4236
30 0.25
0.24
0.23
20
0.22
0.21
10 0.20
400 500 600 700 400 500 600 700
Leeb Rebound Hardness (HLD) Leeb Rebound Hardness (HLD)

CS vs. HLD
160
Compressive Strength (Mpa)

140

120

100

80
y = 0.2831x - 51.708
R² = 0.7362
60
400 500 600 700
Leeb Rebound Hardness (HLD)
Acoustic Measurement
Well: Danks 17-44H (no sonic log)
Confining pressure: 1200 psi, Interval: Lower Bakken Shale and Pronghorn Member
Edynamic vs. HLD Kdynamic vs. HLD
50 35
Dynamic Young's Modulus

y = -0.0113x + 23.269
y = -0.0092x + 30.823

Dynamic Bulk Modulus

R² = 0.0235
R² = 0.0076 30
40
25
(GPa)

(Gpa)
30
20
20
15

10 10
400 500 600 700 400 500 600 700
Leeb Rebound Hardness (HLD) Leeb Rebound Hardness (HLD)

𝑣𝑣dynamic vs. HLD Gdynamic vs. HLD

0.30 20
y = -8E-05x + 0.2907 y = -0.0031x + 12.042
Dynamic Shear Modulus
0.29
Dynamic Poisson's ratio

R² = 0.089 R² = 0.0054
0.28 15
0.27
0.26
(Gpa)

0.25 10
0.24
0.23 5
0.22
0.21
0.20 0
400 500 600 700 400 500 600 700
Leeb Rebound Hardness (HLD) Leeb Rebound Hardness (HLD)
Key Takeways

• In the case of Bakken Formation, quartz, calcite, dolomite, feldspar,

and pyrite are the main contributors to the brittleness, whereas
clay and organic matter control the ductility
• The new mineralogical brittleness model demonstrates a
remarkable positive correlation with the elastic log-based
brittleness model
• The rock hardness result suggests a fair correlation with the rock
compressive strength (confined/unconfined) and elastic brittleness
index
• The rebound hammer hardness test has a restricted application due
to poor core condition and inevitable measurement errors
Acknowledgement

• All Colorado School of Mines Bakken Research Consortium

members
• North Dakota Geological Survey for access to the cores
• Dipanwita Nandy for her guidance and suggestions
• Yulia Faulkner for her assistance in ECS log analyses
• Special thanks to Enerplus Corporation and Marathon Oil
Corporation for providing data for this research