geosciences
Article
Evaluating the Effect of Different Stress Path Regimes on
Borehole Deformation Using Convergence Measuring Device
Jun Hyuk Heo 1, *, Noune Melkoumian 1 and Sam S. Hashemi 2
Citation: Heo, J.H.; Melkoumian, N.;
Hashemi, S.S. Evaluating the Effect of
Different Stress Path Regimes on
Borehole Deformation Using
Convergence Measuring Device.
Geosciences 2022, 12, 317. https://
doi.org/10.3390/geosciences12090317
Academic Editors: Mohamed Shahin
and Jesús Martínez-Frías
Received: 16 August 2022
Accepted: 22 August 2022
Published: 26 August 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Geosciences 2022, 12, 317. https://doi.org/10.3390/geosciences12090317
1 School of Civil, Environmental, and Mining Engineering, University of Adelaide, Adelaide, SA 5000, Australia
2 Department of Energy Resources Engineering, Stanford University, Stanford, CA 94305, USA
* Correspondence: jun.heo@adelaide.edu.au
Abstract: A laboratory study was conducted to investigate the borehole deformation of poorly
cemented sandstone rocks with Uniaxial Compressive Strength (UCS) less than 10 MPa under
different stress path regimes by using a convergence measuring device (CMD). Synthetic thick-walled
hollow cylinders (TWHCs) comprised of sand grains, Portland cement and water were prepared
for this study. A series of mechanical tests including uniaxial and triaxial compression tests were
performed to examine the physical properties of the artificial sandstones. A vertical displacement
loading rate of 0.07 mm/min and confining pressure at a rate of 0.2 MPa/min were applied for
the experiments. The CMD was deployed inside the TWHC specimen to measure the borehole
deformation. Five different stress paths were applied to the specimens to investigate the effect of
stress paths, and three different cement agent contents (10%, 12% and 14%) were considered to study
the effect of cement content on the borehole failure. The effect of the cement content on the borehole
failure was found to be more significant than the effect of change in stress path regimes.
Keywords: borehole deformation; convergence measuring device; thick-walled hollow cylinder;
borehole monitoring
1. Introduction
Drilling boreholes and creating other cavities in the ground disturbs the local in situ
stress field due to the stress concentration around the cavity. Under the stress concentration,
weak bonding between the particles in poorly cemented formation may break when there
is an inadequate cement agent present between the interfaces of sand grains, leading
to inelastic damaged zone and borehole failure. Understanding the responses of poorly
cemented formation is important for the prediction of failure modes and mechanical
characteristics. Borehole convergence is a fundamental parameter in borehole stability
analysis involving deformational calculations. In order to ensure long term stability of
modern drilling operations, the acquisition of geotechnical information about encountered
formation is critical for early detection of instability issues and minimizing borehole failure.
While various convergence measuring devices such as caliper log, borehole extensometer,
and convergence monitor are being used in the field, relatively few efforts have been
made to accurately measure the borehole convergence in laboratory experiments for weak
rocks under different stress path regimes. The cell of Bonnechere, the USBM gage and
the CSIRO HI cell have been proposed for laboratory-based tests to measure the borehole
deformation [1]. However, their application requires an extensive installation procedure
and is limited to a specific borehole diameter size matching the size of the monitoring
device, i.e., the diameter must be more than 30 mm. In the current study, a cost-effective,
versatile and reliable convergence measuring device with a quick and easy installation
procedure was developed based on the concept of the LDT proposed by Goto et al. [2] to
evaluate the borehole deformation in thick-walled hollow cylinder specimens. These new
CMDs have been calibrated and tested in laboratory conditions in order to verify their
performance under different stress paths for poorly cemented sands.
https://www.mdpi.com/journal/geosciences
Geosciences 2022, 12, 317 2 of 14

Laboratory experiments involving the measurement of the diametric deformation

of a borehole by placing a two-arm caliper inside a hollow cylinder were carried out by
Tronvoll [3]. However, the effect of different stress paths was not investigated.
Igarashi et al. [4] developed a borehole tangential deformation gage to measure the tan-
gential deformation of the borehole wall. Papamichos et al. [5] performed laboratory
experiments on hollow cylinder specimens of Red Wildmore sandstone where they used
cantilever strain gage pairs to study the effect of the stress anisotropy on the borehole
deformation. Bujok et al. [6] conducted experimental studies on the effects of the well
convergence on the deformational shape of the borehole. Wu et al. [7] conducted polyaxial
tests with a four-arm cantilever borehole gage placed inside the borehole within a block
specimen to measure the borehole deformation. These tests involved a relatively stronger
sandstone with a uniaxial compressive strength (UCS) greater than 10 MPa.
Laboratory experiments are found to be one of the most effective methods for studying
the borehole failure mechanism. An experimental study on the effect of borehole inclination
was conducted by Rawlings et al. [8] using both synthetic and natural weak sandstone
specimens. Haimson and Song [9] observed a deep and narrow ‘slot’ shaped breakout in
poorly cemented sandstone from a TWHC test. Papamichos et al. [5] studied the effect of
anisotropic stress paths on weak natural sandstone. Younessi et al. [10] conducted labora-
tory tests on TWHC as well as true triaxial tests on synthetic sandstone specimens represent-
ing poorly cemented formations. Bujok et al. [6] studied the effect of borehole convergence
using a specially made pressure vessel by applying confining pressure onto a wellbore
model consisting of different layers of rock materials. Hashemi et al. [11] carried out TWHC
tests on poorly cemented sands focusing on the effects of cement content, anisotropic stress
paths and borehole sizes. Khormali et al. [12,13] conducted experimental studies on reser-
voir rocks. Li et al. [14] demonstrated that the effect of confining pressure in the range of 2
MPa ~ 13 MPa using hollow cylinder specimens on two different sandstones was nearly the
same. Yan et al. [15] conducted a comparative study between unconsolidated and weakly
consolidated sandstones and concluded that weakly consolidated sandstones tended to
develop localised failure, whereas unconsolidated sandstones failed uniformly. Further-
more, Nouri et al. [16] analysed failure modes of sandstones, and Fu et al. [17] recently
investigated the breakage process and the crack propagation paths of cubic specimens.
However, limited laboratory-based experimental studies have been conducted to investi-
gate the effect of borehole deformation using borehole convergence measuring sensors. This
indicates a needs for the development of sensors that are capable of accurately measuring
borehole convergence.
This study aims to investigate the application of the CMD on monitoring borehole
deformation under different stress path regimes. A series of laboratory-based TWHC tests
were designed and conducted under various stress paths. Both far-field and an element
on the borehole wall were considered for applying different stress paths. The borehole
behaviour was monitored by a video camera to the record borehole in order to determine
the borehole failure along the borehole for the specimens. The results present a realistic
understanding of the failure behaviour of poorly cemented sand formations under different
stress paths.

2. Experimental Study
The Uniaxial Compressive Strength (UCS), triaxial and TWHC tests are the most
widely used and versatile rock mechanics tests for determining the parameters required
for borehole stability analysis and understanding the behaviour of a granular formation
under different stress conditions. In this study, the borehole stability experiments were
performed on TWHC specimens in a modified Hoek cell. During each test, the CMDs were
deployed inside the borehole of the TWHC specimen for measuring the borehole diameter
convergence, and a micro camera with the resolution of 225 pixels per inch (PPI) was placed
inside the hollow platen of the Hoek cell for recording the borehole wall deformation
during the testing and for its real-time visual monitoring. The TWHC specimen was
Geosciences 2022, 12, 317 3 of 14

subjected to an axial stress along the borehole axis, and the confining stress was provided
by a servo-controlled hydraulic device.
A number of laboratory facilities were used for conducting the tests. For applying
vertical stress to the specimen, a servo-controlled axial loading system of 300 kN capacity
with 0.1 N accuracy was used. For applying and maintaining the external confining
pressure at a very low level, the Hoek cell was connected to an automatic hydraulic machine
which had a relief valve and a pressure gage with an accuracy of 0.01 MPa. A 60-channel
data acquisition system was connected to computers to monitor and record axial force,
axial displacement, axial strain, lateral strain, convergence measurement and time into a
storage device.

3. Lab Testing Program

The TWHC specimens were prepared in the laboratory under controlled conditions
to manufacture poorly cemented sandstone specimens that resemble those of the natural
rocks. Synthetic rock specimens are frequently used as an alternative to natural rocks for
conducting laboratory-based borehole stability experiments [10,11]. The preparation of
synthetic rocks involves mixing sand grains, a cementing agent and water. The mechanical
properties of the prepared synthetic rock specimens depend on the individual components
used in the mixture. A small variation in the initial components has a significant influence
on the mechanical properties of the final specimens [10]. This indicates the importance
of the careful selection of the components for the mixture. It has been observed that
the mechanical properties of the synthetic rocks are primarily controlled by the cemen-
tation as the sand grains are bonded together by a cementing agent [18]. Liévano and
Kanji [19] tested synthetic rock specimens made under controlled conditions and found
that the mechanical properties of the synthetic specimens closely resemble those of the
natural rocks.
Synthetic sandstone specimens which are close to the weak formations in terms of
mechanical parameters such as the UCS are less than 10 MPa. For further experimental
study of borehole convergence behaviour of this type of sandstone, other parameters such
as high porosity and permeability were replicated. The synthetic sandstone specimens
used for this study were prepared according to the mixture preparation procedure for
poorly cemented formation proposed by Hashemi et al. [11] to achieve homogeneous
properties. The mixture was composed of natural silica sands with two different grain size
ranges, namely, coarse grain sands with grain size between 0.425 mm and 1.4 mm, and
fine grain sands with a grain size between 0.15 mm and 0.355 mm, Portland cement type
II (specific gravity, Gs = 3.15 g/cm3 ) and water. The components were thoroughly mixed
together to achieve a homogeneous mixture for the specimens, and the time spent between
pouring water into the dry mixture and compacting it into metal moulds was maintained
to be within 30 min to avoid initial setting of the cement. Each specimen was prepared by
compacting the mixture into three equal layers of 42 mm thickness. The bottom surface
between the layers was scratched before the subsequent layer was compacted on top to
ensure thorough interlocking between the successive layers. The mixture was not strong
enough to bond sand grains in the early curing stage. Thus, the specimens were left in the
mould for five days, and then removed from the mould and cured for another three days
wrapped in a plastic film at a room temperature (18–23 ◦ C) before testing. The opposite
ends of the specimen are made perfectly parallel. Further details for specimen preparation
can be found in Hashemi et al. [11].

4. Design of the CMD

The convergence measuring device (CMD) was designed based on the local deforma-
tion transducer (LDT) developed at the Institute of Industrial Science, University of Tokyo,
Tokyo, Japan [2]. The LD is a simple and low-cost device that is capable of accurately
measuring axial strains to 10–4 %. The working principle of the CMD is such that when the
borehole of the TWHC specimen is deformed under triaxial stress conditions, the move-
 4. Design of the CMD
The convergence measuring device (CMD) was designed based on the local defor-
mation transducer (LDT) developed at the Institute of Industrial Science, University of
Tokyo, Tokyo, Japan [2]. The LD is a simple and low-cost device that is capable of accu-
Geosciences 2022, 12, 317 4 of 14
rately measuring axial strains to 10–4%. The working principle of the CMD is such that
when the borehole of the TWHC specimen is deformed under triaxial stress conditions,
the movement of the CMD legs triggers either increase or decrease in the gage strain de-
ment ofon
pending thethe
CMD legs stress
applied triggers either
paths. Theincrease or decrease
measurements in from
taken the gage strain
the two depending
strain gages
on the applied stress paths. The measurements taken from the two strain
are then averaged and converted to displacement measurement through a calibration pro- gages are then
averaged and converted to displacement measurement through a calibration
cess by co-relating the CMD leg displacement and the output from the strain gages. The process by
co-relating the CMD leg displacement and the output from the strain gages.
strain gage currently in use has a resistance of 120 ohms with a grid dimension of 2.5 mm The strain
gageand
wide currently
10 mm in long
use has
(TMLa resistance of 120
FLA-10-11). Theohms
two with
straina gages
grid dimension
form a half of 2.5 mm wide
Wheatstone
bridge to enhance measurement accuracy and minimize the temperature effect. bridge to
and 10 mm long (TML FLA-10-11). The two strain gages form a half Wheatstone
enhance
Figuremeasurement
1 shows theaccuracy
final CMDandmanufactured
minimize the temperature
for testing. Iteffect.
consists of strain gage
Figure 1 shows the final CMD manufactured for testing. It consists of strain gage
attached to either side of a V-shaped metal frame which are positioned right opposite each
attached to either side of a V-shaped metal frame which are positioned right opposite each
other as shown below.
other as shown below.

Figure 1. Convergence Measuring Device.

Figure 1. Convergence Measuring Device.
Each strain gage is placed near the closed end of the metal frame where it exhibits the
Each strain gage is placed near the closed end of the metal frame where it exhibits
strain in compression or tension when the open ends of the CMD legs move. To allow a
the straincontact
smooth in compression or tensionwall,
with the borehole when the openplastic
a circular ends of thewas
bead CMD legs move.
attached To allow
to each side of
athe
smooth contact with the borehole wall, a circular
outer surface of the CMD legs using a strong glue. plastic bead was attached to each side
of the outer surface of the CMD legs using a strong glue.
5. Calibration of the CMD
5. Calibration of the CMD
The calibration of the CMD was carried out by comparing and modifying the output
fromThe
thecalibration
strain gageofversus
the CMD was carried out
the displacement by comparing
output and modifying
from the digital calibrator.the output
The initial
from the strain
distance betweengage versus
the open the
endsdisplacement output
of the CMD was set from
at 15 the
mm,digital
whichcalibrator.
matches theTheborehole
initial
distance
diameter between the open
of the TWHC ends of the
specimens. CMD
Both thewas
CMD setand
at 15the
mm, which
digital matchesreadings
calibrator the borehole
were
diameter
zeroed for of the
theinitial
TWHC specimens. Both
measurement, the CMD
and then and themeasurements
subsequent digital calibrator
werereadings
taken atwere
every
zeroed
0.1 mmfor the initial measurement,
displacement as the distance and thenthe
between subsequent
CMD legsmeasurements
was contractedwereby 0.1taken
mm at ata
every 0.1its
step to mmfulldisplacement
range capacity. as the distance
A digital between the
micrometre CMD legs
(Mitutoyo was contracted
digimatic micrometreby 164)
0.1
mm
withataaresolution
step to itsoffull
0.01range capacity.
mm was used toA digital micrometre
get reference (Mitutoyo
measurements fordigimatic micro-of
the calibration
metre 164) with
the CMD. a resolution
The calibration wasof carried
0.01 mm was
out by used to get the
comparing reference measurements
measurements from theforCMD
the
and the digital
calibration of themicrometre
CMD. The output versus
calibration wastime.
carried out by comparing the measurements
from the CMD and the digital micrometre output versus time.
6. Experimental Devices and Calibration
6. Experimental Devices
A triaxial stress and Calibration
system (MTS) was used for the test. This system is capable of sup-
plying, maintaining,
A triaxial monitoring
stress system (MTS) and
wascontrolling
used for theconfining pressure
test. This system asiswell as measuring
capable of sup-
accurate radial and axial displacements. Hoek cell was used to maintain
plying, maintaining, monitoring and controlling confining pressure as well as measuring pre-defined confin-
ing pressure on the specimen. Hydrostatic loading was applied for each
accurate radial and axial displacements. Hoek cell was used to maintain pre-defined con- of the stress paths
tested. In order to measure accurate volumetric strain measurement,
fining pressure on the specimen. Hydrostatic loading was applied for each of the stress two axial variable
displacement
paths tested. Intransformers (LVDT)
order to measure and strain
accurate gaugesstrain
volumetric were used. For the UCS
measurement, twotesting, the
axial var-
specimens were prepared in three different cement contents to study the
iable displacement transformers (LVDT) and strain gauges were used. For the UCS test- effect of cementing
agent.
ing, UCS tests were
the specimens wereconducted
prepared inonthreethe specimens of 65.3contents
different cement mm in diameter
to study theandeffect
127 mmof
in height.
A hollow cylinder test is used to apply an axial load and an external pressure along
the curved surface of the cylinder specimen that provides a ready method for studying
the strength. The modified Hoek cell designed and manufactured by Hashemi and Melk-
oumian [20] consisting of two sets of cylindrical platens was used for conducting the TWHC
tests. A micro camera was fitted inside the test cell to monitor and record the borehole wall
in real time. The Hoek cell was synchronised with a pressure maintainer specifically tuned
for applying a low confining pressure without leakage. The CMDs were placed inside the
 the curved surface of the cylinder specimen that provides a ready method for studying
the strength. The modified Hoek cell designed and manufactured by Hashemi and
Melkoumian [11] consisting of two sets of cylindrical platens was used for conducting the
TWHC tests. A micro camera was fitted inside the test cell to monitor and record the bore-
Geosciences 2022, 12, 317 hole wall in real time. The Hoek cell was synchronised with a pressure maintainer specif- 5 of 14
ically tuned for applying a low confining pressure without leakage. The CMDs were
placed inside the TWHC specimen as shown in Figure 2a. A hollow cylindrical metal
spacer made from hardened steel matching the diameter of the TWHC specimen was
TWHC specimen as shown in Figure 2a. A hollow cylindrical metal spacer made from
placed in between the specimen and the hollow platen to maintain the CMD in the vertical
hardened steel matching the diameter of the TWHC specimen was placed in between the
position
specimenduring
and thetesting.
hollowAplaten
total of
to two CMDs
maintain thewere
CMD used forvertical
in the each test to improve
position thetesting.
during accu-
racy of the convergence measurement. The top CMD was placed alongside
A total of two CMDs were used for each test to improve the accuracy of the convergence the camera
and approximately
measurement. 42 mm
The top CMD from
wasthe top end
placed surface the
alongside of the specimen,
camera and the bottom42CMD
and approximately mm
was placed on the direct opposite location of the top CMD.
from the top end surface of the specimen, and the bottom CMD was placed on the direct
opposite location of the top CMD.

(a) (b)
Figure
Figure 2.
2. (a)
(a) The
The CMDs
CMDs placed
placed inside
inside the
the Hoek
Hoek cell
cell and
and (b)
(b) the
the view
view from the micro
from the micro camera.
camera.

The end surfaces of each specimenspecimen were prepared to be flat and parallel to each other other
by applying
applying a thin layer of dental paste. LubricationLubrication was was applied
applied on on the end surfaces in
order to minimise
minimise thethefriction
frictionbetween
betweenthe theplaten
platenand andthethe specimen.
specimen. Furthermore,
Furthermore, spher-
spherical
ical platens
platens werewere placed
placed on of
on top topthe
of hollow
the hollow
platenplaten to ensure
to ensure that that a uniform
a uniform vertical
vertical stress
stress was
applied
was onto onto
applied the TWHC
the TWHC specimen during
specimen testing
during and toand
testing minimise the bedding
to minimise error. A
the bedding pair
error.
of pair
A axialofand lateral
axial and strain
lateralgages
strainwere
gagesattached directly directly
were attached to the specimen surface tosurface
to the specimen measureto
local deformations.
measure A small load
local deformations. of approximately
A small 5 N was applied
load of approximately 5 N was by applied
the loading machine
by the load-
as the
ing top ram
machine aswas
the lowered
top ram down to the surface
was lowered down of tothe
theplaten
surface prior to the
of the commencement
platen prior to the
of the test to ensure
commencement fulltest
of the contact between
to ensure fullthe top ram
contact and the
between platen.
the top ram Theand
imagethe captured
platen. Theby
the micro camera as shown in Figure 2b was checked for the clear focal
image captured by the micro camera as shown in Figure 2b was checked for the clear focal point of the camera
and toofensure
point that the
the camera CMDs
and were positioned
to ensure that the CMDs vertically. The experiments
were positioned wereThe
vertically. performed
experi-
with the
ments werefollowing two with
performed loadingthe stages:
following first,
two simultaneous
loading stages: vertical
first,and confining stresses
simultaneous vertical
wereconfining
and applied up to a pre-defined
stresses were applied stress
up level to simulate stress
to a pre-defined the hydrostatic stress condition
level to simulate the hydro-in
the first stage of loading. The second stage was followed by increasing the vertical load
at a constant displacement rate of 0.07 mm/min. Confining pressure was varied in each
test to apply desired confining stress to the TWHC specimen. Figure 3 illustrates borehole
convergence observed by the camera placed above the specimen during a TWHC test under
the first stress path without the CMD placed inside the borehole. The camera captured
three different stages of borehole convergence; Figure 3 shows sand particles beginning to
dislodge followed by visibly deformed borehole and then the fully converged borehole.
 creasing the vertical load4at a constant displacement rate of 0.07 mm/min.
6 MPa Confining pres-

Devia
sureREVIEW
Geosciences 2022, 12, x FOR PEER was varied in each test to apply desired confining stress to the TWHC specimen.13 of 15
4 MPa
2
Figure 3 illustrates borehole convergence observed by the camera placed above the spec-
R REVIEW imen during a TWHC test under the first stress 13 of 15
path without the CMD2 MPaplaced inside the
0
borehole. The camera captured three different stages of borehole convergence; Figure 3
Geosciences 2022, 12, 317 0
shows sand particles beginning 5
to dislodge 10
followed 15 20deformed
by visibly 25 borehole6 and
of 14

then the fully converged borehole. Radial strain (%)

(b)
sciences 2022, 12, x FOR PEER REVIEW 12 13 of 15
(b)
12 PEER REVIEW
Geosciences 2022, 12, x FOR 10 13 of 15

Deviatoric Stress (MPa)

(c)
10 8
Deviatoric Stress (MPa)

12

8 10 6 (b)

Deviatoric Stress (MPa)

12
6 8 4 (b) 6 MPa
10 12 4 MPa
Deviatoric Stress (MPa)
4 6 2 6 MPa
8 2 MPa
10 4 MPa
2 Deviatoric Stress (MPa) 4 0 6 MPa
6 8 02 MPa 5 10 15 20 25
4 MPa
0 2 Radial strain (%)
0 5 10 4 15 20 25 2 MPa
Figure 3. Borehole convergence
6 stages observed by the camera. 6 MPa
Radial strain (%) 0
7. Stress Paths 4 MPa
2
7. Stress Paths 4 0 5 10 6 MPa 15
The effect
effect of 2 MPa
The of stress
stress paths
paths on
on the
the behaviour
Radialof
behaviour TWHC
strain
of TWHC
(c) (%) was
was4investigated
investigated
MPa
by
by applying
applying
five 0 stress 2paths. Two approaches were used
different for analysing the stress paths: (1)
(1)
five different stress paths. Two approaches were used for analysing the stress paths:
considering 0 stresses
the 5 on12the 10boundary 15 of the20 specimens 25 and
2 MPa
(2) considering certain
considering (c) the stresses on the boundary of the specimens and (2) considering certain
0 radial Radial
stresses
stresses at
Figureat the
8. the borehole
Stress versus
borehole wall. In thestrain
wall.strain:
In the first
(a) (%)
approach
10%approach
first (the first
cement content,
(the first and
(b) and second
12% cement
second stress
stress paths),
content, (c) 14% the
paths), ce-
the
12 stress ( ) and0 the 10
ment content.
vertical 5
confining 10
pressure ( 15 ) were20considered 25 and the induced
Deviatoric Stress (MPa)

vertical stress (𝜎𝑧) z and the confining pressure (𝜎𝑐𝑜𝑛𝑓) conf were considered and the induced
stresses at the borehole wall were Radial subsequently
calculated strain (%) based onon thetheapplied stresses on
10 stresses at the borehole wall8were calculated subsequently based applied stresses
8.5. Fifth Stress Path
Deviatoric Stress (MPa)

the boundary of the TWHC specimens. In the

on the boundary of the TWHC specimens. In the second approach (the third, fourth and second approach (the third, fourth and fifth
stress paths), the
fifthprincipal stresses
8 fifth In the
stress paths), stress
the path,
principal (c)(σ𝜃, and
both
6 stresses
r , σr
(𝜎𝜃, ) for
z𝜎𝑟, were
𝜎𝑧) anincreased
element on
for an element the borehole wall
simultaneously
on the borehole at the were
same
wall
considered,
rate until and
the the
failurestresses
at thewere applied
borehole wallto the was boundaries
observed of the
by the specimen
camera. to induce
While the the
yield
were considered, 12 and the stresses were applied to the boundaries of the specimen to in-
desired
stresses magnitude
increase of tangential
consistently 4of with
and vertical
the increase stresses on
in the stressesthe borehole
cement content, wall
6the and
specimenson bothwith
6 duce the desired magnitude tangential (c)
and vertical on the borehole MPa wall and
ends12% ofcement
the specimen,content respectively.
behaved differently. For these specimens, the measured radial strain
on both ends 10 of the12specimen, respectively. 4 MPa
Deviatoric Stress (MPa)

was In the first stress paths (Equation

2 a 6lower
MPa stress (1)) both state,the vertical stress and the confining pres-
4 In significantly
the first stress higher
pathsat (Equation (1a)) both thecompared
vertical stressto theand specimens
the
2
with 10%
confining
MPa
pres- and
sure14% wereof increased
cement simultaneously
content. This is likely at the
to same
be caused rate byup to
one the
4 MPa at the same rate up to the pre-defined value, whichof pre-defined
the two CMDs value,
being which
placed
sure were 8increased 10 simultaneously
Deviatoric Stress (MPa)

2 simulated
at a location a hydrostatic
in the borehole condition
0 where actingthe on on the boundary
borehole converged of the specimens.
significantly moreThen, thanin thethe
simulated a hydrostatic condition acting
2subjected
MPa the boundary of the specimens. Then, in at
the
second
other step,
CMD.
6 the specimen
Therefore, was
the 0
averaging of tothea vertical
5
two CMD loading
readings increment
10 resulted (normal
in 15
such faulting)
a different
8
second step, the specimen was subjected to a vertical loading increment (normal faulting)
0 corresponding
reading compared to a constant displacement
to the results obtained rate ofthe
forRadial 0.07other
mm/min.
strain (%) During
cement During
contents. the test,yield
the level
0 corresponding to a constant displacement rate of 0.07 mm/min. the Thetest, the level stress
the5specimen
of for confining
4 pressure
with 6
10
10% applied
cement to 15
the
content external
was surface
significantly 6 of
MPathe specimens
less than the was kept
specimen constant
with 14%
of the confining pressure applied to the external surface of the specimens was kept con-
bycement
an Radial
automatic strainpressure
content. (%)
The TWHC gauge system. In
specimens with the10% second and4 stress
14%
MPa path when
cement content thebehaved
stress level simi-
stant
reached by the an automatic pressure gauge system. Inpressure the second stress path when the stress
larly,
level as 2ishydrostatic
Figure
reached
8. Stress4versus
indicated
the
condition,
by
hydrostatic
radial
the stress
condition,
the
strain: confining
(a) 10%
versus the radialcement
confining strain 2
(reverse
content, (b)612%
behaviour.
pressure
MPa
MPafaulting)
When
(reverse
was
cementcompared
faulting)
increased,
content, (c) 14%
wasto in-
ce-
that
corresponding
ment content. to a constant pressure rate of 0.2 MPa/min. The4 lateral displacement
MPa The lateral displace- rate
creased,
Figure 8. Stress versus radial strain: (a)was
10%measured
cement corresponding
0 content, by four to amm
30 constant
lateral pressure
strain rate
gauges ce-of which
0.2 MPa/min.
were connected to the data
2(b) 12% cement content, (c) 14%
ment content. ment rate
acquisition8.5. was 0 measured
and
Fifth linkedPath
Stress toby thefour 30 mmpump
5hydraulic lateral strain gauges
10 systems. 15 which 2 MPawere connected to the
data In acquisition
the In third and0 linked to the paths,hydraulicand pump were systems.
theand fifthfourth
stressstress
Radial
path, strainboth σ𝜃 (%)and σr z wereincreased
increasedsimultaneously
simultaneouslyup attothe a same
8.5. Fifth Stress Path In
certainrate the
value. thirdThen, and asfourth stress
0per the experiment paths,
5 𝜎𝜃 and
plan, 𝜎𝑧
the were
10
specimen increased
was simultaneously
15
subjected either to up or to
until the failure at the borehole wall was observed by the camera. While the yield z
a certain
increment. value. In Then,
orderasto per the experiment
achieve thewithsame plan,
the the
tangential (%) specimen
and was subjected either to 𝜎𝑧
In the fifth stress path, both σ𝜃 andstresses
σr were increased
increase simultaneously
consistently Radial
theat strain
increase same in thevertical
cementstressescontent, atthethe borehole
specimens with
or 𝜎𝜃
wall inincrement.
the initial In
step order
ofby to
the achieve
test, the
thecontent, same
confining tangential
pressure and vertical stresses at the of bore-
rate until the failure atFigure 8. Stress
the borehole versus
wall 12% radial
was strain:
observed
cement (a)
content 10% thecement
camera.
behaved While
differently. (b) the
12%
For thesevalue
cement
yield content,was(c)derived
specimens, 14%
thece- in terms
measured the strain
radial
stresses increase consistently with hole
ment content. wall instress,
tangential
the increase the
in the
was significantly
initial
andcement steprate
thehigher
of the
of test,
thethe
atincreasing
content, a lower
confining
specimensthe vertical
stress with
state,
pressure
stress value
compared was wasconstant
to kept
derived and
the specimens
in terms
withwas 10% and
12% cement content behaved differently. of
Figure the
the same tangential
Forasof
8. Stress the rate
versus
these stress,
radial
specimens, and
of increase the
strain:the rate
in
(a)measured
10% of increasing
thecement
tangential content, the vertical
stress
(b) 12%at the stress
borehole
cement was kept
content, wall. constant
(c) 14% and
It isce-worth
14% cement content. This is likelyradial to be straincaused by one of the two CMDs being placed
was significantly higher 8.5.
at Fifth Stress
a lowerment Pathstate,that
content.
mentioning
stress comparedin an unsupported borehole, 10% r at the andborehole wall is zero.
at a location in thetoborehole
the specimens wherewith the borehole converged significantly more than at the
In the fifth
14% of cement content. This is likely to be In
stressthe fifth
path,
caused path,
both
byTherefore,σ𝜃 and
one of thethe σr
two z were increased
increased
CMDs being simultaneously
simultaneously until
at the the
same failure at the
other CMD. averaging of theplacedtwo CMD readings resulted in such a different
rate until 8.5.
the Fifth
borehole
failure
at a location in the borehole where the borehole Stress
wall
at Path
the was observed
borehole
converged wall by was
significantly the real-time
observed
more camera
by
than the
at recording.
camera.
the While It should
the yieldbe mentioned
reading compared to the results obtained for the other cement contents. The yield stress
other CMD. Therefore, stresses that
increase
the averaging In for each
fifth
consistently
of the TWHC
stress with specimen,
path,the both
increase σ𝜃 ininand the z were
σrcement
were calculated
increased
content, separately
simultaneously
the specimens basedwith on the
atspecimen
the stress
same
fortwo CMD
specimen readings
with 10% resulted
cement such
content awas
different
significantly less than the with 14%
application
rate untilcement
the area
failure and the
at cement
the borehole size. Equations (1)–(5) show the principal stress status
yield at simi-
reading compared to the 12% cement
results content
obtained behaved
for the other
content. Theborehole
differently. TWHC For these
contents. wallThe
specimens wasyield
specimens, observed
with the by
10%measured
stress and the14% camera.
radialWhile
cement strain
content the behaved
the boundary
stresses increase ofaspecimens and borehole wallin for each stress path. The difference between
for specimen with 10%was significantly
cement content higher
waslarly, asatconsistently
significantly lowerless
is indicated stresswith
than
by the the
state,
the increase
compared
specimen
stress versus the
toradial
with the cement
14% straincontent,
specimens with
behaviour. the10%specimens
When and compared with to that
14% of 12%
cementthe first
cement
content. and fourth
content
This is stress
behaved
likely
cement content. The TWHC specimens with 10% and 14% cement content behaved simi- paths
to be is
differently. in
caused the byinitial
For these
one step
of of the
specimens,
the two experiment.
CMDs the measured
being In the
placed first
radial path,
strain the
larly, as is indicated byatthe
a locationwasmagnitude
significantly
in theradial
stress versus boreholeof z
strain is
higher
whereconsiderably
at a lower
the borehole
behaviour. lower
stress
Whenconverged than
state, that
compared
compared significantly for
to that the tofourth
the path
specimens
more than at theuntil the
with end
10% of
and the
other CMD. 14% of cementthe
Therefore, content.
averaging Thisofisthe likelytwoto CMDbe caused
readings by resulted
one of the two CMDs
in such a differentbeing placed
at a location
reading compared to thein the borehole
results obtained where forthe theborehole
other cement converged significantly
contents. The yieldmore stress than at the
for specimen otherwithCMD. 10% Therefore,
cement content the averaging of the two less
was significantly CMD readings
than resultedwith
the specimen in such14%a different
readingThe
cement content. compared
TWHC to the results
specimens withobtained
10% and for14% the cement
other cement content contents.
behavedThe simi-yield stress
larly, as is indicated by the stress versus radial strain behaviour. When compared to thatwith 14%
for specimen with 10% cement content was significantly less than the specimen
cement content. The TWHC specimens with 10% and 14% cement content behaved simi-
 Geosciences 2022, 12, 317 7 of 14

first stage of the experiment. The same relationship was applied to the second and third
stress paths.

σ1 = σ2 = σ3 = σconf = σz (first step)
first stress path (1)
σ1 = σz > σ2 = σ3 = σconf (second step)

σ1 = σ2 = σ3 = σconf = σz (first step)
second stress path (2)
σ1 = σ2 = σconf > σ3 = σz (second step)

σ1 = σ2 = σθ = σz > σ3 = σr = 0 (first step)
third stress path (3)
σ1 = σθ = σ2 = σz > σr = 0 (second step)

σ1 = σ2 = σθ = σz > σ3 = σr = 0 (first step)
fourth stress path (4)
σ1 = σz > σ2 = σθ > σr = 0 (second step)

σ1 = σθ = σ2 = σz > σr = 0 ( first step)
fifth stress path (5)
σ1 = σθ = σ2 = σz > σr = 0 ( second step)

8. Results
To minimise the influence of rock heterogeneity on the test results, repetitive triaxial
tests were performed on three specimens for each cement content under various confining
pressures. Different values of confining pressure of 2 MPa, 4 MPa and 6 MPa.

8.1. First Stress Path

Geosciences 2022, 12, x FOR PEER REVIEW
In the first stress path, both the axial stress and the confining pressure were increased
simultaneously at the same rate up to the pre-defined level, in order to simulate hydrostatic
condition on the boundary of the specimens. Then, in the second step, the specimen was
subjected to an axial loading increment corresponding to a constant displacement rate of
0.07 mm/min. During the test, the level of the confining pressure applied to the external
surface of the specimens was kept constant by an automatic pressure gauge system. The
results showed the relatively linear function of the deviatoric stress followed by a constant
increase in the radial strain as the borehole failure initiated. As shown in Figure 4a–c,
this behaviour was more profound at a lower level of cement content of 10% compared
to specimens with a higher cement content of 12% and 14%. Prior8 to of the
15 initiation of the
borehole failure, the radial strain of the specimens with a higher amount of cement content
increased less dramatically compared to that for the specimens with 10% cement content,
indicating that the load-bearing capacity increases when the cement content is increased.
(a)
11
10
9
Deviatoric Stress (MPa)

8
7
6
5
4
6MPa
3
2 4MPa

1 2MPa
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Radial Strain (%)

Figure 4. Cont.

12
(b)
11
 2 4MPa

1 2MPa
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Radial Strain (%)
Geosciences 2022, 12, 317 8 of 14

12
(b)
11
10
9

Deviatoric Stress (MPa) 8

7
6
5
4
3 6MPa

2 4MPa
Geosciences 2022, 12, x FOR PEER REVIEW 1 2MPa 9 of 15

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
Radial Strain (%)
(c)
18

16

14
Deviatoric Stress (MPa)

12

10

4
6MPa
2 4MPa
2MPa
0
0 0.05 0.1 0.15 0.2 0.25 0.3
Radial Strain (%)

Figure 4. Deviatoric stress (MPa) versus radial strain (%): (a) 10% cement content, (b) 12% cement
Figure 4. content, (c)stress
Deviatoric 14%(MPa)
cement content.
versus radial strain (%): (a) 10% cement content, (b) 12% cement
content, (c) 14% cement content.
8.2. Second Stress Path
8.2. Second Stress
ThePath
increase in the cement content had a minimal effect on the stiffness of the speci-
Themens
increase in the
tested cement
under thecontent
secondhad a minimal
stress effect[21]
path. Mogi on the stiffnessthat
suggested of the
thespeci-
stiffness depends on
mens tested under the second stress path. Mogi [20] suggested that the stiffness depends
the rock material and, as the same sand grain sizes were used for the specimen preparation,
on the rock
the material
specimen and, as the same
stiffness sand grain
remained sizes were
unchanged used
in the for the
tests. The specimen
level of prepa-
the confining pressure
ration, the
required to initiate a borehole convergence increased with thethe
specimen stiffness remained unchanged in the tests. The level of confining
increase in the cement con-
pressure required to initiate a borehole convergence increased with the increase in the
tent. This is due to the strengthening effect of the cement. Hashemi et al. [11] have also
cement content. This is due to the strengthening effect of the cement. Hashemi et al. [11]
observed that the poorly cemented specimens with a higher cement content demonstrate a
have also observed that the poorly cemented specimens with a higher cement content
higher
demonstrate peak peak
a higher strength duedue
strength to the strengthening
to the strengtheningeffect
effect of thecementing
of the cementing agent. Generally, for
agent.
Generally, for the second path with an increase in the cement content, the radial strain increased for a
the second path with an increase in the cement content, the radial strain
increasedcertain confining
for a certain stress.stress.
confining WithWithincreasing
increasingthe the
cement
cement content,
content, the
thestiffness
stiffnessof the specimens
in the vertical
of the specimens direction
in the vertical obtained
direction higherhigher
obtained valuesvalues
than than
in theinlateral orientation
the lateral orien- for the second
tation forpath; however,
the second path;the rate of the
however, increase
rate ofwas lower
increase wasfor the second
lower path. path.
for the second
Figure 5 shows that the effect of cement content has a higher influence on stabilizing
a borehole than that of the confining stress applied for the second stress path. The level of
confining pressure required to initiate a borehole convergence increased with the increase
in the cement content. Such observation agrees with [11], where the poorly cemented spec-
imens with a higher cement content demonstrated a higher peak strength due to the
strengthening effect of the cementing agent. Figure 6 below illustrates the borehole failure
Geosciences 2022, 12, 317 9 of 14

Figure 5 shows that the effect of cement content has a higher influence on stabilizing a
borehole than that of the confining stress applied for the second stress path. The level of
confining pressure required to initiate a borehole convergence increased with the increase
in the cement content. Such observation agrees with [11], where the poorly cemented
specimens with a higher cement content demonstrated a higher peak strength due to
the strengthening effect of the cementing agent. Figure 6 below illustrates the borehole
failure process observed from the camera installed inside the borehole. Moreover, the
specimens showed a more ductile behaviour when the equal amount of vertical stress and
the maximum principal stress were applied to them. The borehole failure was observed
at lower strains for the second path compared to the first path, which suggests that the
Geosciences 2022, 12, x FOR PEER REVIEW 10 of 15
tangential stress has the important influence on the borehole failure in poorly cemented
Geosciences 2022, 12, x FOR PEER REVIEW 10 of 15
specimens. Results displayed that the shear bands were formed, and the produced sand
mainly consisted of uniform loose grains.

Figure 5. Confining pressure

Confining pressure (MPa) versus
pressure (MPa)
(MPa) versus radial
radial strain
strain (%).
Figure 5.
Figure 5. Confining versus radial strain (%).
(%).

Figure 6. Borehole failure observations.

Figure 6. Borehole failure observations.
8.3. Third
8.3. Third Stress
Stress Path
Path
8.3. Third Stress
Figure 7a–c Path
7a–c present
Figure present the
the results
results on
on applying
applying the
the third
third stress
stress path
path on
on the
the TWHC
TWHC speci-
spec-
mens Figure
with 7a–c12%
10%, present
and the results
14% cement on applying
contents the 2third
under MPa,stress
4 MPapath
and on
6 the TWHC
MPa, spec-
respectively.
imens with 10%, 12% and 14% cement contents under 2 MPa, 4 MPa and 6 MPa, respec-
imens with 10%,
The confining 12%
and and 14%
vertical cement
stresses werecontents underfor
considered 2 MPa, 4 MPaatand
an element the6borehole
MPa, respec-
wall
tively. The confining and vertical stresses were considered for an element at the borehole
tively.
based onThethe
confining andaxial
lateral and vertical stresses
stresses wereon
applied considered for anofelement
the boundary the TWHC at the borehole
specimens.
wall based on the lateral and axial stresses applied on the boundary of the TWHC speci-
wall based on
The stresses theapplied
were lateral and axial
to the stressesatapplied
specimen the sameonrate
theuntil
boundary
a givenofvalue
the TWHC speci-
was reached.
mens. The stresses were applied to the specimen at the same rate until a given value was
mens.
Then theTheconfining
stresses were
stressapplied to the specimen
was increased until an at the sameat
instability rate
theuntil a given
borehole value
wall waswas
ob-
reached. Then the confining stress was increased until an instability at the borehole wall
reached. Then the confining stress was increased until an instability at the borehole wall
was observed, while the vertical stress was kept unchanged. The main difference between
was observed, while the vertical stress was kept unchanged. The main difference between
the second and third stress paths was in the initial stage of the test, in which the axial
the second and third stress paths was in the initial stage of the test, in which the axial
stress applied to the specimen was considerably higher in the third path for the corre-
stress applied to the specimen was considerably higher in the third path for the corre-
sponding borehole size and specimen outer diameter. According to Figure 7a, with an
 Geosciences 2022, 12, 317 10 of 14

served, while the vertical stress was kept unchanged. The main difference between the
second and third stress paths was in the initial stage of the test, in which the axial stress
applied to the specimen was considerably higher in the third path for the corresponding
borehole size and specimen outer diameter. According to Figure 7a, with an increase in the
vertical stress, the strength of the specimen reached prior to the initiation of the borehole
failure was noticeably higher on the specimen with 14% cement under 6 MPa than the
es 2022, 12, x FOR PEER REVIEW 11 of 15
specimen that was under 4 MPa and 12% of cement content. It illustrates that the effect
of vertical stress as a supporting pressure in increasing the ductility for a given cement
content and the specimens exhibited a higher ductile behaviour when a greater vertical
Unlike the specimens
stress wasunder the Furthermore,
applied. first stress path, theincreasing
with trends of graphs
cement from the third
content, the borehole failure at
stress path in general show
the wall wasthat the vertical
observed stressradial
at higher kept strains.
the specimen in the contraction
mode and therefore, the dilation
Unlike was observed
the specimens underat higher
the firstradial
stressstrains. Thetrends
path, the resultsofalso
graphs from the third
showed that withstress
an increase
path iningeneral
cementshow
content,
thatthe
theradial strain
vertical would
stress keptbethe
increased
specimen for in the contraction
certain vertical stress in this stress path, similar to that in the second stress path. This con-
mode and therefore, the dilation was observed at higher radial strains. The results also
firms that with an increase
showed thatinwith
the cement content,
an increase the lateral
in cement strain
content, theincreases in allwould
radial strain the be increased for
stress paths. However, the effect of cement content had less influence than the
certain vertical stress in this stress path, similar to that in the second change in stress path. This
vertical stress. Increasing cement content from 10% to 12% did not result in any significant
confirms that with an increase in the cement content, the lateral strain increases in all the
variation in radial strain.
stress paths. However, the effect of cement content had less influence than the change in
vertical stress. Increasing cement content from 10% to 12% did not result in any significant
variation in radial strain.

(a)
14

12
Deviatoric Stress (MPa)

10

8
6 MPa
6
4 MPa
4
2 MPa
2

0
0 5 10 15 20
Radial Strain (%)

(b)
14

12
Deviatoric Stress (MPa)

10

6
6 MPa
4
4 MPa
2
2 MPa
0
0 5 10 15
Radial Strain (%)
Figure 7. Cont.
es 2022, 12, x FOR PEER REVIEW 12 of 15
(c)
Geosciences 2022, 12, 317 11 of 14
14

(MPa)Stress (MPa)
12
10 (c)
8
14
6
12

Deviatoric
6 MPa
10
4
4 MPa
Deviatoric Stress 28 2 MPa
06
6 MPa
4 0 5 10 15
4 MPa
2
Radial Strain (%)
2 MPa
0
Figure 7. Deviatoric stress 0versus radial strain:
5 (a) 10% cement 10 content, (b) 12%
15 cement content, (c)
14% cement content. Radial Strain (%)
Figure 7. Deviatoric stress versus radial strain: (a) 10% cement content, (b) 12% cement content,
8.4. Fourth Stress Path
(c)stress
Figure 7. Deviatoric 14% cement content.
versus radial strain: (a) 10% cement content, (b) 12% cement content, (c)
Figure 8a–c illustrate the stress−strain behaviour for the fourth stress path for 10%,
14% cement content.
12% and 14% cement 8.4. content
Fourth Stress Path
respectively. In this path, the confining stress at the borehole
wall
8.4. was increased
Fourth along
Stress Path with
Figure theillustrate
8a–c axial stresstheat the same rate
stress–strain up to 2 for
behaviour MPa.theThen,
fourthvertical
stress path for 10%, 12%
stress was increased and 14%while
cementconfining
contentstress was kept
respectively. unchanged.
In this The trendstress
path, the confining of the
at the borehole wall
Figure 8a–c illustrate the stress−strain behaviour for the fourth stress path for 10%,
was increased
stress−strain relationships in thealong
fourth with
paththewasaxial stresstoatwhat
similar the same rate up toin
was observed 2 MPa. Then, vertical stress
the first
12% and 14% cement content respectively. In this path, the confining stress at the borehole
path where the axial wasstress
increased
was muchwhilelower
confining stress
than that forwas
thekept
fourthunchanged.
path in theThe trend
initial stageof the stress–strain
wall was increased along with the axial stress at the same rate up to 2 MPa. Then, vertical
ofstress
the test. Figure relationships
8a–c show in the
that thefourth pathinstability
borehole was similar to what
occurs at was
a observed
higher stress inand
the first path where
was increased while confining stress was kept unchanged. The trend of the
lower radial strainthe
foraxial
the stress
fourth was
pathmuch
in lower than
different that contents.
cement for the fourth path in the axial
initial stage of the test.
stress−strain relationships in the fourth path was similar to what was Since the in
observed initial
the first
stress was higher Figure 8a–c
forstress
the fourth show that
path,lowerthe borehole
the specimens instability
behaved occurs
in path at
a brittle a higher
manner before and lower radial
stress
path where the axial was much than that for the fourth in the initial stage
the borehole strain for Such
convergence. the fourth path is
behaviour in likely
different
to becement
due tocontents.
the Sincethethe initial axial stress was
of the test. Figure 8a–c show that the borehole instability occurs at aeffect
higherof stressloading
and
rate on the higher
strength of fortested
the the fourth path, the specimens behaved in a brittle manner before the borehole
specimens.
lower radial strain for the fourth path in different cement contents. Since the initial axial
convergence. Such behaviour is likely to be due to the effect of the loading rate on the
stress was higher for the fourth path, the specimens behaved in a brittle manner before
strength of the tested specimens.
the borehole convergence. Such behaviour is likely to be due to the effect of the loading
rate on the strength of the tested specimens.
(a)
7

6 (a)
Stress (MPa)

57

46
Stress (MPa)

6 MPa
35
Deviatoric

4 MPa
24
62MPa
MPa
13
Deviatoric

4 MPa
2 2 MPa
0
1 0 5 10 15
Radial strain (%)
0
Figure 08. Cont. 5 10 15
Radial strain (%)
2022, 12, x FOR PEER REVIEW 13 of 15
(b)
Geosciences 2022, 12, 317 12 of 14
12 (b)
12
10

Deviatoric Stress (MPa)

10 (b)

Stress (MPa)
8
12
8
6
10

Stress (MPa)
6
4 6 MPa
DeviatoricDeviatoric
8
4 6 MPa 4 MPa
2
6 4 MPa 2 MPa
2
0
4 2 MPa
0 5 10 15
6 MPa 20 25
0
Radial
20 strain
4 MPa(%)
2 0 5 10 15 25
Radial strain (%) 2 MPa
0
0 5 10 15 20 25
Radial strain (%) (c)
12 (c)
12
10
Deviatoric Stress (MPa)

10 (c)
Stress (MPa)

8
12
8
6
10
Stress (MPa)

6 6 MPa
4
DeviatoricDeviatoric

8 4 MPa
4 6 MPa
2
6 4 MPa 2 MPa
2
0 2 MPa
4 0 5 6 MPa 10 15
0
4 MPa
2 0 5 10 Radial strain (%)
15
Radial strain (%) 2 MPa
0
Figure
Figure 8.
8. Stress
Stressversus
versusradial strain:
radial (a) 10%
strain: cement
(a) 10% content,
cement (b) 12%
content, (b) cement content,
12% cement (c) 14%(c)ce-14%
content,
ment 0
content. 5 10 15
cement
Figure 8. Stress versus content.
radial strain: (a) 10% cement content,
Radial strain (%)(b) 12% cement content, (c) 14% ce-
ment content. 8.5. Fifth Stress Path
8.5. Fifth Stress Path
8.5. Fifth Stressversus In the fifth stress path, both σ𝜃 and σrr were increased simultaneously at the same
Path radial
Figure 8. Stress
rate until strain: (a) 10% cement content, (b) 12% cement content, (c) 14% ce-
the failure at the borehole wall was observed observed by the the camera.
camera. While the yield
ment Incontent.
the fifth stress path, both σ𝜃 and σr were increased simultaneously at the same
stresses increase consistently with the increase in
stresses increase consistently with the increase in the cement content, the the cement content, the specimens
specimens with
with
rate until the failure 12% at the borehole
cement content wall wasdifferently.
behaved observed by For the camera.
these While the
specimens, themeasured
yield radial strain
8.5. Fifth increase
Stress Path12% cement content behaved differently. For these specimens, the measured radial strain
stresses consistently
was significantly with higher
the increase in thestress
at aa lower
lower cement content,
state, comparedthe specimens
to the with with 10% and
the specimens
specimens
was significantly higher at stress state, compared to with 10% and
12% In cement content
the fifth stress
14% behaved
path, both differently.
σ𝜃 andThis For
σr werethese specimens,
increased the measured
simultaneously of radial
at
thethe strain
same
14% ofof cement
cement content.
content. This is likely
is likely to be
to be caused
caused by one
by one of the two
two CMDs being
CMDs being placed
placed
was until
rate significantly higher
the failure
at a at at aborehole
the
location lower
in thestress
wallstate,
boreholewas wherecompared
observed the to the specimens
by
borehole camera. withsignificantly
While
converged 10%yield
the and more than at
at a location in the borehole where the borehole converged significantly more than at the
14% of cement
stresses increasecontent.
the other This
consistently is
CMD. likely
with theto be caused
increase
Therefore, the bycement
in averaging
the one ofof the two
content,
the twoCMDs
the being
specimens
CMD placed
readings with
other CMD. Therefore, the averaging of the two CMD readings resultedresulted
in such ain such a
different
at a location
12% in the
cement content borehole
behaved
different where
reading the
differently.
comparedborehole
For tothese
the converged
specimens,
results significantly
obtainedthe measured
for the more than
radial
other cementat the
straincontents. The yield
reading compared to the results obtained for the other cement contents. The yield stress
othersignificantly
was CMD. Therefore, the
higherfor averaging
atspecimen
a lower stress of the two
state, CMD readings
compared to the resulted
specimens in such
with a 10%
different
and
for specimen with 10% cement content was significantly less than the specimen with with
stress with 10% cement content was significantly less than the specimen 14%
reading
14% compared
of cement 14%to cement
content. the results
This obtained
iscontent.
likely be for
toThe caused
TWHC the other
by onecement
specimens of the contents.
two
with 10%CMDs Thebeing
and 14%yield stresscontent behaved
placed
cement
cement content. The TWHC specimens with 10% and 14% cement content behaved simi-
fora specimen
at location inwiththe 10% cement
borehole
similarly, iscontent
where wasby
the borehole significantly
theconverged lesssignificantly
than the specimen withat14%
morebehaviour.
than theWhen compared
larly, as is as indicated
indicated by the stress stress
versusversus
radialradial
strain strain
behaviour. When compared to that
cement
other content.
CMD. The
Therefore, TWHC
the specimens
averaging of with
the two10% CMD and 14%
readingscement content
resulted in behaved
such
to that of the first and second stress paths where the confining pressure and axial stress a simi-
different
larly, as compared
reading is indicated toby
were thethe stressat
results
increased versus
obtained radial
a constant for rate
thestrain
other
in the behaviour.
cement When
path,compared
contents.
fifth stress The yieldto
the specimens that
stress tolerated a higher
for specimen withlevel 10%ofcement
stresses content was significantly
surpassing 25 MPa, as can lessbe than
seentheinspecimen
Figure 9. with 14%
It is important to mention
cement content. The thatTWHC
the trend specimens
of the graphwith 10%remained and 14% cement content
unchanged behaved
for different levelssimi-
of stresses applied,
larly, as is indicated by the stressthe
demonstrating versus radialcontraction
consistent strain behaviour. mode under Whenthe compared
fifth stressto that
path.

Geosciences 2022, 12, 317
of the first and second stress paths where the confining pressure and axial stress were
increased at a constant rate in the fifth stress path, the specimens tolerated a higher level
of stresses surpassing 25 MPa, as can be seen in Figure 9. It is important to mention that
the trend of the graph remained unchanged for different levels of stresses applied, demon-13 of 14
strating the consistent contraction mode under the fifth stress path.

35

30

25

Stress (MPa) 20
6MPa
15
4MPa
10 2MPa

0
0 2 4 6 8
Radial strain (%)

Figure 9. Deviatoric stress versus radial strain.

Figure 9. Deviatoric stress versus radial strain.
9. Conclusions
9. Conclusions
The effects of five different stress path regimes and three different cement agent con-
tents
Theon the borehole
effects failure stress
of five different stresswere
pathinvestigated
regimes andexperimentally
three differentusing
cementtheagent
TWHC con-tests.
Five
tents ondifferent
the boreholestress failure
regimesstress
werewereapplied to the synthetic
investigated TWHC poorly
experimentally usingcemented
the TWHC sand
specimens
tests. with the
Five different CMDs
stress deployed
regimes wereinside the to
applied borehole. In conjunction
the synthetic TWHC poorly with the CMDs, the
cemented
borehole was monitored by the camera. The five different
sand specimens with the CMDs deployed inside the borehole. In conjunction with stress paths included applying
the
normal
CMDs, theand strike–slip
borehole faulting stress
was monitored by theregimes
camera.toThe synthetic TWHCstress
five different poorly cemented
paths included sand
specimens.
applying normalTheand specimens under
strike−slip the first
faulting stress
stress path showed
regimes that increase
to synthetic TWHC in the cement
poorly ce-
content
mented sand increased
specimens. the The
radial strain. The
specimens increase
under the firstin stress
the cement content that
path showed had increase
a minimal
effect on the stiffness of the specimens tested under the second
in the cement content increased the radial strain. The increase in the cement content stress path. The specimens
had
under third stress path showed that the radial strain increased
a minimal effect on the stiffness of the specimens tested under the second stress path.with the increase in cement
The
content. under
specimens The effect of stress
third cement content
path had that
showed less influence
the radialon the borehole
strain increasedfailure
with under
the in-the
fourth
crease stress path
in cement than The
content. the change
effect ofincement
vertical stress.had
content Theless
specimens
influence under theborehole
on the fifth stress
pathunder
failure tolerated a higher
the fourth level
stress of than
path stresses. Finally,inthe
the change effectstress.
vertical of theThe
cement content
specimens on the
under
borehole failure was found to be more significant than the effect of confining
the fifth stress path tolerated a higher level of stresses. Finally, the effect of the cement pressure in all
stress paths.
content on the borehole failure was found to be more significant than the effect of confin-
ing pressure in all stress paths.
Author Contributions: Conceptualization, J.H.H.; validation, J.H.H.; formal analysis, J.H.H.; investi-
gation, J.H.H.; writing—original draft preparation, J.H.H.; writing—review and editing, J.H.H.; super-
Author Contributions: Conceptualization, J.H.H.; validation, J.H.H.; formal analysis, J.H.H.; inves-
vision, N.M. and S.S.H. All authors have read and agreed to the published version of
tigation, J.H.H.; writing—original draft preparation, J.H.H.; writing—review and editing, J.H.H.;
the manuscript.
supervision, N.M. and S.S.H. All authors have read and agreed to the published version of the man-
uscript.
Funding: This research received no external funding.
Funding: This
Conflicts ofresearch
Interest:received no external
The authors declare funding.
no conflict of interest.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Amadei, B.; Stephansson, O. Rock Stress and Its Measurement, 1st ed.; Chapman & Hall: London, UK, 1997.
References
2. Goto, S.; Tatsuoka, F.; Shibuya, S.; Kim, Y.; Sato, T. A simple gauge for local small strain measurements in the laboratory. Soils
1. Amadei, B.; Stephansson, O. Rock Stress and Its Measurement, 1st ed.; Chapman & Hall: London, UK, 1997.
Found. 1991, 31, 169–180. [CrossRef]
2. 3. Goto, S.; Tatsuoka,
Tronvoll, F.; Shibuya,
J. Experimental S.; Kim, Y.; Sato,
Investigation T. A simple
of Perforation gauge
Cavity for localIn
Stability. small strain measurements
Proceedings in the
of the 33rd U.S. laboratory.on
Symposium Soils
Rock
Found. 1991, 31, 169–180.
Mechanics, Santa Fe, NM, USA, 3–5 June 1992; Tillerson, J.R., Wawersik, W.R., Eds.; Balkema: Rotterdam, The Netherlands, 1992;
pp. 365–373.
4. Igarashi, A.; Ito, T.; Sekine, K.; Hayashi, K. Development of borehole tangential deformation gage and its application for
deter-mining the stress in rock. In Proceedings of the 42nd US Rock Mechanics Symposium and the 2nd U.S.—Canada Rock
Mechan-ics Symposium, San Francisco, CA, USA, 29 June–2 July 2008.
5. Papamichos, E.; Tronvoll, J.; Skjaerstein, A.; Unander, T.E. Hole stability of red wildmoor sandstone under anisotropic stresses
and sand production criterion. J. Pet. Sci. Eng. 2010, 72, 78–92. [CrossRef]
Geosciences 2022, 12, 317 14 of 14

6. Bujok, P.; Porzer, M.; Labus, K.; Klempa, M.; Pavluš, J. Experimental modeling of abandoned shallow oil wells convergence. Eng.
Geol. 2013, 157, 1–7. [CrossRef]
7. Wu, B.; Chen, Z.; Zhang, X. Stability of borehole with Breakouts—An Experimental and Numerical Modelling Study. In
Proceedings of the 50th US Rock Mechanics/Geomechanics Symposium, Houston, TX, USA, 26–29 June 2016.
8. Rawlings, C.G.; Barton, N.R.; Bandis, S.C.; Addis, M.A.; Gutierrez, M.S. Laboratory and numerical discontinuum modeling of
wellbore stability. J. Pet. Technol. 1993, 45, 1086–1092. [CrossRef]
9. Haimson, B.; Song, I. Laboratory study of borehole breakouts in Cordova Cream: A case of shear failure mechanism. Int. J. Rock
Mech. Min. Sci. 1993, 30, 1047–1056. [CrossRef]
10. Younessi, A.; Rasouli, V.; Wu, B. Sand production simulation under true-triaxial stress conditions. Int. J. Rock Mech. Min. Sci. 2013,
61, 130–140. [CrossRef]
11. Hashemi, S.S.; Melkoumian, N.; Taheri, A.; Jaksa, M. The failure behavior of poorly cemented sands at a borehole wall using
laboratory tests. Int. J. Rock Mech. Min. Sci. 2015, 77, 348–357. [CrossRef]
12. Khormali, A.; Sharifov, A.R.; Torba, D. Experimental and modeling analysis of asphaltene precipitation in the near wellbore
region of oil wells. Pet. Sci. Technol. 2018, 36, 1030–1036. [CrossRef]
13. Khormali, A.; Sharifov, A.R.; Torba, D.I. Experimental and modeling study of asphaltene adsorption onto the reservoir rocks. Pet.
Sci. Technol. 2018, 36, 1482–1489. [CrossRef]
14. Li, J.; Zhang, G.; Liu, M. Experimental investigation on the effect of confining pressure on the tensile strength of sandstone using
hollow cylinder tensile test method. Arab. J. Geosci. 2019, 12, 768. [CrossRef]
15. Yan, M.; Deng, J.; Yu, B.; Li, M.; Zhang, B.; Xiao, Q.; Tian, D. Comparative study on sanding characteristics between weakly
consolidated sandstones and unconsolidated sandstones. J. Nat. Gas Sci. Eng. 2020, 76, 103183. [CrossRef]
16. Nouri, M.; Khanlari, G.; Sarfarazi, V.; Rafiei, B. Failure Modes Analysis of Sandstones (Case Study: Markazi Province Sand-stones).
Sci. Q. J. Iran. Assoc. Eng. Geol. 2020, 13, 61–86.
17. Fu, J.; Haeri, H.; Yavari, M.D.; Sarfarazi, V.; Marji, M.F. Effects of the Measured Noise on the Failure Mechanism of Pre-Cracked
Concrete Specimens Under the Loading Modes I, II, III, and IV. Strength Mater. 2021, 53, 938–949. [CrossRef]
18. Saidi, F.; Bernabé, Y.; Reuschlé, T. The mechanical behaviour of synthetic, poorly consolidated granular rock under uniaxial
compression. Tectonophysics 2003, 370, 105–120. [CrossRef]
19. Liévano, V.R.; Kanji, M.A. Simulation of arenaceous weak rocks by means of cemented sands. In Proceedings of the 12th ISRM
Congress, Beijing, China, 16–21 October 2011; OnePetro: Richardson, TX, USA, 2011.
20. Hashemi, S.S.; Melkoumian, N. Effect of different stress path regimes on borehole instability in poorly cemented formations.
J. Pet. Sci. Eng. 2016, 146, 30–49.
21. Mogi, K. Experimental Rock Mechanics; CRC Press: Boca Raton, FL, USA, 2006.