energies

Article
Fracture Processes of Rock-Like Specimens
Containing Nonpersistent Fissures under
Uniaxial Compression
Miao Chen 1,2 , Shengqi Yang 1, * , Ranjith Pathegama Gamage 1,2 , Wendong Yang 3 ,
Pengfei Yin 1 , Yuanchao Zhang 1 and Qiangyong Zhang 4
1 State Key Laboratory for Geomechanics and Deep Underground Engineering, School of Mechanics and Civil
Engineering, China University of Mining and Technology, Xuzhou 221116, China;
miaochen@cumt.edu.cn (M.C.); ranjith.pg@monash.edu (R.P.G.); yinpengfei@cumt.edu.cn (P.Y.);
yczhang@cumt.edu.cn (Y.Z.)
2 Deep Earth Energy Laboratory, Department of Civil Engineering, Building 60, Monash University,
Melbourne, VC 3800, Australia
3 College of Pipeline and Civil Engineering, China University of Petroleum, Qingdao 266580, China;
yangwd@upc.edu.cn
4 Research Center of Geotechnical and Structural Engineering, Shangdong University, Jinan 250100, China;
qiangyongz@sdu.edu.cn
* Correspondence: yangsqi@cumt.edu.cn




Received: 25 November 2018; Accepted: 25 December 2018; Published: 28 December 2018


Abstract: Many rock engineering accidents have proven that the coalescence of discontinuities
in surrounding rock can have a major impact on the security and stable operation of energy
infrastructure. To give an insight into the understanding of the crack propagation and coalescence
in fissured rock masses, a series of uniaxial compression experiments were conducted on rock-like
specimens containing nonpersistent fissures. The digital speckle correlation method (DSCM) and the
acoustic emission (AE) monitoring system were adopted to capture the real-time strain field on the
specimens’ surfaces and microfracturing events within specimens, respectively. The experimental
results indicated that the strength and deformation modulus of specimens were significantly affected
by fissure inclination. The damage process showed obvious progressive stain localization failure
characteristics. The clear and intuitive full-field strain field development was successfully monitored
by the DSCM technique. The real-time strain accumulation, crack initiation, propagation, and
coalescence were also analyzed. Each time, the saltation of the strain field was usually accompanied
by the fluctuation of the stress curve and obvious AE events. Crack coalescence modes between
fissures changed from tension coalescence mode to mixed tension-shear coalescence mode, then
to shear coalescence mode with an increase in fissure inclination. Five basic failure modes were
identified from the experimental results: Tensile failure across the fissure planes, rotation failure of
newly generated blocks, mixed failure mode, shear failure, and splitting failure. An investigation of
the fracture processes of rock-like specimens containing nonpersistent fissures using these methods
can enhance understanding of the fracture behavior of jointed rocks.

Keywords: stability; nonpersistent fissures; strain evolution and fracture behavior; digital speckle
correlation method

1. Introduction
The stability of surrounding rock is crucial for the security and stable operation of rock
engineering, such as underground space development, reservoir engineering, slope projects, and

Energies 2019, 12, 79; doi:10.3390/en12010079 www.mdpi.com/journal/energies

Energies 2019, 12, x FOR PEER REVIEW 2 of 24
Energies 2019, 12, 79 2 of 24
The stability of surrounding rock is crucial for the security and stable operation of rock
engineering, such as underground space development, reservoir engineering, slope projects, and
engineering. As
mining engineering. As aa common
common complicated
complicated geological
geological mass, many discontinuities with variant
scales distributed in the rock mass are a result of
distributed in the rock mass are a result of geological geological structure or engineering
structure influence,
or engineering which
influence,
causes causes
which the mechanical behavior
the mechanical of a rock
behavior ofmass
a rock tomass
showto asshow
discontinuous, heterogeneous,
as discontinuous, and strongly
heterogeneous, and
anisotropic.
strongly When a disturbed
anisotropic. load (causing
When a disturbed load by crustal by
(causing stress, blasting,
crustal stress,orblasting,
excavationor unloading)
excavation
is applied [1,2],
unloading) new cracks
is applied are readily
[1,2], new cracks initiated
are readily and propagated,
initiated sometimes sometimes
and propagated, coalescing with other
coalescing
cracks.
with Thecracks.
other coalescence between fissures
The coalescence betweenwillfissures
lead to will
damage
lead or
to even
damagefailure of thefailure
or even rock mass.
of theMany
rock
rock mass engineering accidents have derived from the coalescence of discontinuities
mass. Many rock mass engineering accidents have derived from the coalescence of discontinuities in in surrounding
rock [3,4]. Figure
surrounding rock1a[3,4].
shows a pillar
Figure 1afailure
showsin aanpillar
underground
failure in minean caused by coalescence
underground along two
mine caused by
pre-existing along
coalescence fissures.
twoFigure 1b shows
pre-existing a step Figure
fissures. failure observed
1b showsnear the failure
a step headscarp of the near
observed Aishihik
the
River Landslide,
headscarp of theand Figure 1c
Aishihik shows
River rock slopeand
Landslide, stability
Figure that
1cisshows
discontinuity-controlled.
rock slope stability Therefore,
that is
extensive experimental studies
discontinuity-controlled. have been
Therefore, conducted
extensive on brittlestudies
experimental rock orhaverock-like
been specimens
conductedcontaining
on brittle
pre-existing
rock fissures
or rock-like [5–9] to investigate
specimens containingcrack initiation,
pre-existing propagation,
fissures [5–9] to and coalescence.
investigate Theinitiation,
crack common
crack patterns [10] observed in previous investigations of rock and rock-like
propagation, and coalescence. The common crack patterns [10] observed in previous investigations specimens containing
onerock
of singleand
pre-existing
rock-like fissure undercontaining
specimens uniaxial compression
one single are summarized
pre-existing as follows:
fissure under (1) Wing
uniaxial
cracks initiate from the tips of fissures and propagate toward the direction
compression are summarized as follows: (1) Wing cracks initiate from the tips of fissures and of maximum compressive
stress withtoward
propagate increasingthe load; (2) secondary
direction of maximum cracks also initiate
compressive from
stress the increasing
with tips of fissures
load;and propagate
(2) secondary
coplanarly
cracks also or nearlyfrom
initiate coplanarly
the tipstoofthe fissureand
fissures or with an angle
propagate similar toorthe
coplanarly wingcoplanarly
nearly cracks but to in the
opposite direction.
fissure or with an angle similar to the wing cracks but in the opposite direction.

Figure 1.
Figure 1. Jointed
Jointed rock
rockmass.
mass.(a)
(a)Two
Twopre-existing
pre-existingfissures in in
fissures a pillar [10];
a pillar (b) (b)
[10]; stepstep
failure on aon
failure slope [11];
a slope
and (c) multiple closely spaced discontinuities [12].
[11]; and (c) multiple closely spaced discontinuities [12].

In addition
In addition to to the
the study
study ofof crack
crack initiation
initiation andand propagation
propagation of of aa single
single fissure,
fissure, the
the coalescence
coalescence
behavior of cracks when specimens contain more than one fissure has also
behavior of cracks when specimens contain more than one fissure has also been extensively studied, been extensively studied,
mainly by
mainly by compression
compression experiments
experiments or or numerical
numerical simulations
simulations on on specimens
specimens with with two
two oror three
three
pre-existing fissures [13–21]. Previous studies have shown that the coalescence
pre-existing fissures [13–21]. Previous studies have shown that the coalescence often occurs by the often occurs by the
linkage of two fissures through tensile, shear, or mixed tensile-shear cracks,
linkage of two fissures through tensile, shear, or mixed tensile-shear cracks, and the patterns of crackand the patterns of crack
coalesce are
coalesce are found
found to to be
be strongly
strongly dependent
dependent on on the
the geometry
geometry of of the
the fissures.
fissures.
Although significant improvements have
Although significant improvements have been obtained for thebeen obtained for the understanding
understanding of of the
the crack
crack
coalescence pattern
coalescence pattern of of specimens
specimens containing
containing fewfew fissures,
fissures, the
the fissures
fissures contained
contained in in engineering
engineering rockrock
masses often appear in groups with similar orientation and characteristics.
masses often appear in groups with similar orientation and characteristics. The crack behavior of The crack behavior of aa
specimen containing multiple fissures becomes more complex as the
specimen containing multiple fissures becomes more complex as the high number of fissures high number of fissures provides
more possible
provides morecoalescences. Sagong and
possible coalescences. Bobetand
Sagong [22]Bobet
performed a compression
[22] performed on gypsumon
a compression specimens
gypsum
with 3 and 16 fissures, and the comparison results showed that the cracking
specimens with 3 and 16 fissures, and the comparison results showed that the cracking pattern in pattern in specimens
with multiple
specimens withfissures
multiple wasfissures
similarwasto the specimen
similar to thewith two fissures.
specimen with two Prudencio
fissures.and Van Sintand
Prudencio Jan Van
[23]
observed
Sint threeobserved
Jan [23] main failure threemodes
maininfailure
biaxial modes
tests with in specimens
biaxial tests containing nonpersistent
with specimens joints
containing
and indicated the failure modes and maximum strengths were dependent
nonpersistent joints and indicated the failure modes and maximum strengths were dependent on on joint systems geometry.
Basedsystems
joint on the experimental
geometry. Based results [23],experimental
on the Bahaaddini results
et al. [24] carried
[23], out a numerical
Bahaaddini et al. [24] simulation
carried outto a
study the effects of joint geometric parameters on the failure mechanism
numerical simulation to study the effects of joint geometric parameters on the failure mechanism and mechanical properties
of rock
and mass containing
mechanical properties nonpersistent
of rock mass joints. Their numerical
containing resultsjoints.
nonpersistent showed thatnumerical
Their the failureresults
mode
was controlled
showed that theby fissures
failure modeinclination and step
was controlled byangle andinclination
fissures the angle and between respectand
step angle fissures and
the angle
Energies 2019, 12, 79 3 of 24

principal stress direction. Chen [25] experimentally investigated the deformation behavior of jointed
rock masses for gypsum specimens and claimed that deformation behavior is related to the closing of
fissures, the fracture process, and shear rupture along the shear plane. Using PFC3D software (4.0,
Itasca Consulting Group, Minnesota, MN, USA), Fan et al. [26] investigated the effect of nonpersistent
joints on macromechanical behavior of a jointed rock block and demonstrated that the joint dip angle
and joint continuity factor played an important role on strength and the deformability modulus of
jointed rock blocks. Gui et al. [27] performed numerical simulations to study the influence of openings
on rock mechanical behavior: The effect of size, alignment, orientation, shape, and opening ratio of
openings on the mechanical behavior were investigated. Meanwhile, the impact of joint geometry
and arrangement on the strength and deformability of jointed rock masses were also symmetrically
investigated [28–31].
Previous work has promoted understanding of the mechanical behavior and crack behaviors
of rock masses containing nonpersistent fissures. However, most experimental studies have mainly
estimated the fracture process, including the statistical description of cracks from photographs on
a macro scale or using strain gauges to measuring small changes in specimen deformation near the
crack. The strain field development of the specimens, which is crucial in revealing the mechanism
of fracture behavior and the failure process under compression, cannot be easily obtained from
traditional monitoring methods. Digital image correlation (DIC) is an optical noncontact deformation
measurement technique that can be used to monitor full-field strain evolution on a specimen surface
during the deformation process [32]. Compared to the traditional contact measurement technique
(such as a displacement extensometer, photoelastic coating, and a strain gauge), it has many advantages:
It is convenient and avoids local stress disturbance in direct contact measurements. Additionally, DIC
can capture the propagation of a secondary crack, which is difficult to observe by traditional methods
because the secondary crack remains closed after initiation [22]. Thus, DIC has been widely used in
fields such as materials testing, fracture mechanics, and project monitoring [33–36]. High-resolution
DIC was applied by Carroll et al. [37] to obtain the strain fields of fatigue crack growth in an alloy at the
grain level. Zhao et al. [38,39] combined 3D-DIC with numerical simulation to investigate the fracture
process of artificial rock-like specimens with an open hole or double flaws under uniaxial compression.
Li et al. [40] employed DIC to investigate the influence of pre-existing holes with different geometries
on the strength and fracture behavior of specimens containing a hole during a uniaxial compression
test. However, despite numerous previous studies using DIC to investigate the fracture processes in
rocks, due to the complexity of fissure distribution, research about the use of DIC to detect real-time
full-field strain development of specimens containing a set of pre-existing parallel fissures has been
rarely reported.
This study adopted the digital speckle correlation method (DSCM) and acoustic emission (AE)
techniques to detect the fracture processes of specimens containing a set of parallel fissures with
different fissure inclinations. The characteristics of the full-field strain field development, acoustic
emission, mechanical characteristics, and crack coalescence patterns during the loading process were
analyzed. The structure of this paper is as follows. The preparation of the uniaxial compression
experiment on rock-like specimens containing nonpersistent fissures is introduced in Section 2.
In Section 3.1, the effect of fissure inclination on specimens’ mechanical behavior and deformation
characteristics are investigated. In Section 3.2, the crack coalescence patterns and failure modes of
specimens with various fissure inclinations are analyzed and categorized. In Section 3.3, the strain
field development and AE event characters under uniaxial loading associated with different failure
modes are analyzed to enhance understanding of the basics of the fracture processes. Section 4
summarizes and discusses the characteristics of nonpersistent fissures on AE characteristics and strain
field. The main conclusions of this study are presented at the end of the paper.
Energies 2019, 12, 79 4 of 24

Energies 2019, 12, x FOR PEER REVIEW 4 of 24

2. Experimental Testing
2. Experimental Testing
2.1. Rock-Like Specimens with Nonpersistent Fissures for the Experiment
2.1. Rock-Like Specimens with Nonpersistent Fissures for the Experiment
It is difficult to investigate the rock fracture process at an engineering scale. Therefore, real
rock andIt is difficult
rock-liketomaterials
investigate
arethe rock fracture
usually adoptedprocess
in the at an engineering
investigation scale.fracture
of rock Therefore, real rock
behavior in
aand rock-likeGypsum
laboratory. materialshasare usually
proven adopted
to be a goodinrock-like
the investigation
material for ofrock
rockmechanics
fracture behavior
experimentsin a
laboratory.
and Gypsum has
for investigating proven
crack to be
behavior a good rock-like
[13,14,22,28]. In ourmaterial for rock
experiments, mechanics
gypsum experiments
was used and
to simulate
for investigating
rock behavior. The crack behavior
mixture used[13,14,22,28]. In our
as the rock-like experiments,
material in this gypsum
study was was used to simulate
composed of gypsum,rock
behavior.
aqueous The mixture
solution used
with 0.5% as the rock-like
antifoaming material
agent, and in this
silica sand at astudy
weightwas
ratiocomposed of gypsum,
of 10:6:1. The physics
aqueous
and solution
mechanical with 0.5%
properties of antifoaming
the rock-likeagent, andand
material silica sand sandstone
natural at a weight areratio
shownof 10:6:1.
in TableThe1.
physics
The and mechanical
comparison indicatedproperties of thematerial
that the similar rock-likehasmaterial
similarand natural properties
mechanical sandstone to aresandstone,
shown in
Table is
which 1. suitable
The comparison indicated
in simulating that the similar material has similar mechanical properties to
real rock.
sandstone, which is suitable in simulating real rock.
Table 1. Physics and mechanical properties of rock-like material.
Table 1. Physics and mechanical properties of rock-like material.
Uniaxial
Tensile Elastic Poisson’s Cohesion Friction Density
Material Type Compressive
Uniaxial Strength
Tensile
(MPa) Elastic (GPa)
Modulus Ratio (MPa) (◦ ) (kg/m3 )
Strength (MPa) Poisson’s Cohesion Friction Density
Material Type Compressive Strength Modulus
Rock-like Ratio (MPa) (°) (kg/m3)
Strength (MPa)
35.3 (MPa)
3.0 (GPa)
8.3 0.16 12.6 18.9 1546
material
Rock-like material 35.3 3.0 8.3 0.16 12.6 18.9 1546
Sandstone 20–170 4–25 3–35 0.02–0.2 4–40 25–60 2200–2710
Sandstone 20–170 4–25 3–35 0.02–0.2 4–40 25–60 2200–2710

The
The dimensions
dimensions ofof the cuboid specimens
specimens werewere 160160××8080× ×
50 50
mm mm (high
(high × wide
× wide × thickness).
× thickness). This
This study mainly focused on the effects of the fissure inclination on crack behavior,
study mainly focused on the effects of the fissure inclination on crack behavior, mechanical properties, mechanical
properties,
and failureand failure
modes. modes.
In the In the
fissured fissured
region, theregion,
fissure the fissure inclination
inclination had a range hadof a0°range 0◦ toother
of and
to 90°, 90◦ ,
and other
joints joints parameters
parameters were fixed.wereThefixed. The geometrical
geometrical parameters
parameters and fissures
and fissures arrangement
arrangement of the of the
test
test specimens are shown in Figure 2. Pre-existing fissures were created by inserting
specimens are shown in Figure 2. Pre-existing fissures were created by inserting 0.6-mm-thick alloy 0.6-mm-thick
alloy bars pre-cut
bars into into pre-cut
slotsslots
in theinmold
the mold (Figure
(Figure 3c) and
3c) and pulling
pulling themthem
out out after
after the the solidification
solidification started.
started.
12
m
m

m
m
10
160mm

10
m
m

Figure2.2. Geometrical
Figure Geometricalparameters
parametersand
andfissure
fissurearrangement
arrangementof
ofthe
thespecimens.
specimens.

The
The specimen
specimen preparation
preparation procedures
procedures were
were carefully
carefully controlled
controlled to toobtain
obtainreliable
reliableexperiment
experiment
results. The weighed modeling materials were first mixed in a blender (Figure
results. The weighed modeling materials were first mixed in a blender (Figure 3a,b). Next, 3a,b). Next, the mixturethe
was poured
mixture wasinto a mold
poured anda placed
into mold andon aplaced
vibrationon table (Figure table
a vibration 3d) until there3d)
(Figure were no there
until obvious signs
were no
of bubbles.
obvious After
signs of the mixture
bubbles. became
After solidified,
the mixture the strips
became were the
solidified, pulled outwere
strips slowly, andout
pulled theslowly,
specimens
and
were removed from
the specimens werethe mold. from
removed Finally,
thethe specimens
mold. Finally,were cured a minimum
the specimens of 20a days
were cured beforeofthe
minimum 20
test
daysat before
room temperature.
the test at roomAt least 3 specimens
temperature. At with
least the same inclination
3 specimens with thewere prepared
same to getwere
inclination the
stable data.
prepared toIfget
there was andata.
the stable obvious difference
If there in the results
was an obvious of two
difference inidentical tests,
the results a third
of two specimen
identical tests,
was tested. The finished specimens are shown in Figure 3d.
a third specimen was tested. The finished specimens are shown in Figure 3d.
Energies
Energies 2019,
2019, 12,
12, x79FOR PEER REVIEW 55of
of24
24
Energies 2019, 12, x FOR PEER REVIEW 5 of 24

(a) Weigh (b) Stir

aluminum alloy frame nickel alloy strips

vibrostand
Slots

(d)Vibration (e) finished specimens

(c) experimental mold

Figure 3. Preparation of rock-like rock specimens containing a set of pre-existing parallel fissures
Preparation of
Figure 3. Preparation of rock-like
rock-like rock specimens
specimens containing
containing a set of pre-existing parallel fissures
inclined in various directions. (a) weigh; (b) stir; (c) experimental mold; (d) vibration and (e) finished
inclined ininvarious
variousdirections.
directions. (a) weigh; (b) stir; (c) experimental
(a) weigh; (b) stir; (c) experimental mold; (d)mold; (d)and
vibration vibration and
(e) finished
specimens.
(e) finished
specimens. specimens.

2.2.
2.2. Testing
Testing System
System
2.2. Testing System
The
The uniaxial
uniaxial compressive
compressive experiment
experiment of of the
the specimens
specimens containing
containing nonpersistent
nonpersistent fissures
fissures was
was
The uniaxial compressive experiment of the specimens containing nonpersistent fissures was
performed
performed atatthethe testing
testing system
system in theinState
theKeyState Key Laboratory
Laboratory for Geomechanics
for Geomechanics and Deep
and Deep Underground
performed at the testing system in the State Key Laboratory for Geomechanics and Deep
Underground Engineering,
Engineering, China ChinaofUniversity
University Mining and of Mining and Technology,
Technology, as shown inasFigure
shown4.in AFigure 4. A
mechanic
Underground Engineering, China University of Mining and Technology, as shown in Figure 4. A
mechanic servo-controlled loading system (CMT5305, MTS Industrial Systems
servo-controlled loading system (CMT5305, MTS Industrial Systems Company, Shenzhen, China)Company, Shenzhen,
mechanic servo-controlled loading system (CMT5305, MTS Industrial Systems Company, Shenzhen,
China) with a maximum
with a maximum loadingofcapacity
loading capacity of 300
300 kN was usedkN
for was used for
the uniaxial the uniaxial
compressive compressive
experiments, the
China) with a maximum loading capacity of 300 kN was used for the uniaxial compressive
experiments,
loading modethe wasloading mode control
displacement was displacement
loading, andcontrol loading,
the loading rate and the loading
was 0.06 mm/min, rate wascould
which 0.06
experiments, the loading mode was displacement control loading, and the loading rate was 0.06
mm/min, which could
ensure a quasistatic ensurecondition.
loading a quasistatic loading condition.
mm/min, which could ensure a quasistatic loading condition.

Servo loading
system

Servo loading AE Digital speckle

system sensor correlation method
LED Light

Camera
AE monitoring
system

Figure 4.
Figure Testingsystem
4. Testing system for
for the
the specimens
specimens containing
containing nonpersistent
nonpersistent fissures
fissures in
in the
the uniaxial
uniaxial
Figure 4. Testing
compressive system for the specimens containing nonpersistent fissures in the uniaxial
experiment.
compressive experiment.
compressive experiment.
To obtain
To obtainthethereal-time fracture
real-time evolution
fracture process,process,
evolution an opticalan
noncontact
optical deformation
noncontact measurement
deformation
To (XTOP
system obtain3D the real-time Company,
Technology fracture evolution
Shenzhen, process,
China), an optical
“XTDIC”, was noncontact
adoptedwastodeformation
capture
measurement system (XTOP 3D Technology Company, Shenzhen, China), “XTDIC”, adoptedthe to
measurement
strain field system
over the (XTOPof3D
course theTechnology
loading Company,
process. This Shenzhen,
system wasChina),
a DSCM“XTDIC”, was of
consisting adopted to
a digital
capture the strain field over the course of the loading process. This system was a DSCM consisting of
capture the strain field over the course of the loading process. This system was a DSCM consisting of
acamera
digitaland a data
camera andprocessing system. The
a data processing high-precision
system. digital camera
The high-precision was
digital used to
camera collect
was usedtheto speckle
collect
aimages
digitalofcamera
the and a data
specimens processing
during the system. The
experiment. By high-precision
analyzing the digital images
speckle cameraof was used to collect
specimens before
the speckle images of the specimens during the experiment. By analyzing the speckle images of
the
and speckle
after images
surface of the
deformation specimens
by image during
matching,the experiment.
the By
displacement analyzing
field was the speckle
obtained by images
tracking of
the
specimens before and after surface deformation by image matching, the displacement field was
specimens
movement before and after surface deformation by image matching, the displacement field was
obtained byof geometric
tracking the points
movementon the ofspecimen
geometricsurface. To the
points on improve the image
specimen recognition
surface. To improve degree,
the
obtained by tracking the movement of geometric points on the specimen surface. To improve the
image recognition degree, black paint was randomly sprayed on a clean surface of the gypsum
image recognition degree, black paint was randomly sprayed on a clean surface of the gypsum
Energies 2019, 12, 79 6 of 24

black paint was randomly sprayed on a clean surface of Servo loading specimen to produce an artificial
the gypsum
system
speckle pattern. Note that the speckle particles were to be randomly distributed and uniform in size.
During DSCM application, a region of interest (ROI) was selected and then discretized into small
virtual subsets. Image correlation was to match the small square subsets of the underformed image
with the position in the deformed surface image, as shown in Figure 5. Point p and q in the reference
Servo loading AE Digital speckle
system (x, y) were mapped to point p* and
image with coordinates q* in the image after deformation with
sensor correlation method
coordinates (x*, y*), respectively. LED Light
The relation function between the two coordinates [41] could be performed as

∂u ∂u
x∗ = x + u + ∆x + ∆y, (1)
∂x ∂y
Camera
AE monitoring
∗ ∂v ∂v
systemy = y+v+ ∆x + ∆y, (2)
∂x ∂y

where u, v, ∂u ∂u ∂v ∂v
∂x , ∂y , ∂x , ∂y , are six deformation parameters; and u and v are the displacement of initial
tip position relative to the (x, y) coordinate system.
Fig. 4 Testing system for the specimens containing non-persistent fissures in the uniaxial compressive experiment
y*

q*
ROI

deformation p* x*
y Deformed subset

q v
Δy
Image p
analysis grid
x u
Δx
Y
Undeformed subset

X
Fig. 5 Basic principles of the digital speckle correlation method
Figure 5. Basic principles of the digital speckle correlation method.

At the same time, the AE events were monitored by using an AE monitoring system (PCI-II,
Physical Acoustic Corporation, Princeton, NJ, USA) to identify microfracturing of specimens during
uniaxial compression. In this experiment, an acoustic emission probe (Physical Acoustic Corporation,
Princeton, NJ, USA) was attached to the back of the specimens, and the frequency of the AE system
was set to 3 MHz.

3. Analysis of Experimental Results

3.1. Effect of Fissure Inclinations on Mechanical Behavior

Figure 6 shows the complete axial stress–strain curves for intact specimens and specimens with
different fissure inclinations under uniaxial compression. It can be seen that the characteristics of the
stress–strain curves varied due to the influence of the fissure inclination, and the mechanical behavior
of the specimen with α = 90◦ was similar to the intact specimen. Based on a characteristic analysis
of the curves, except for the specimen with α = 90◦ , the remaining curves existed a stress fluctuation
before peak strength. This was mainly due to microcrack initiation in fissure tips during the loading
process: The new crack’s initiation resulted in local stress adjustment and caused the stress–strain
curve to fall. However, the effect of this phenomenon on the integral forced status of the specimen was
limited. Thus, the stress curve increased again, and then the stress–strain curve fluctuated before peak
 Energies 2019, 12, 79 7 of 24

strength. AE counts began to appear stable after this point. Combined with AE count characteristics
and 2019,
Energies strain
12,field
x FORdevelopment,
PEER REVIEW this point was identified as the crack initiation point: The stress 7 of 24 at
this point was considered to be the crack initiation stress point. Crack initiation represents the stress
reveal
level the
wherefailure mechanismbegins,
microfracturing of jointed
and itrock to reduce
is especially engineering
important disasters
to reveal related
the failure to rock of
mechanism
fracturing [42,43].
jointed rock Many engineering
to reduce methods have been used
disasters to determine
related the crack[42,43].
to rock fracturing initiation
Manystress of brittle
methods have
rocks
beenunder
used touniaxial
determinecompression, such as stress
the crack initiation the volumetric strain-based
of brittle rocks method
under uniaxial [44], the crack
compression, such as
volumetric strainstrain-based
the volumetric method [3], and [44],
method the volumetric stiffness strain
the crack volumetric method [45]. [3],
method In and
ourthestudy, the
volumetric
combination of the DSCM technique and the AE monitoring system successfully
stiffness method [45]. In our study, the combination of the DSCM technique and the AE monitoring identified crack
initiation
system in jointed rock
successfully from aspects
identified crack of AE events,
initiation the strain
in jointed rockfield,
fromand the stress-strain
aspects of AE events,curve. In
the strain
addition to the
field, and thespecimen of α =curve.
stress-strain 90°, the
In overall
addition stages of specimen
to the the stress–strain 90◦ , the
of α = curves during
overallthestages
complete
of the
failure process curves
stress–strain could during
be divided into the failure
the complete following fourcould
process stages
bebased
dividedoninto
curvethe sharpness
following fourand stages
AE
characteristics:
based on curvenamely,
sharpnesstheandstages of initial compaction,
AE characteristics: namely, thelinear
stages elastic
of initialdeformation, nonlinear
compaction, linear elastic
deformation,
deformation, and post-peak
nonlinear failure, as shown
deformation, in Figurefailure,
and post-peak 7. as shown in Figure 7.

Fig. 6 stress-strain
Figure curves
6. Stress–strain of specimens
curves of specimens.
Figure 6. Stress–strain curves of specimens.
10 Axial stress Ea=7.06GPa
100
Accumulated σp
Accumulated AE counts (×1000)

AE counts
8 Ea=7.49GPa
80
Axial stress/ MPa

σci
6 Stress fluctuation
60
Es=7.51GPa

4 40

Stage I: Stage II: Stage III: Stage IV:

2 initial linear elastic nonlinear post-peak
20
compaction deformation deformation failure
0 0
0 1 2 3 4 5 6
-3
Axial strain/ 10
Figure 7. Characteristics
Fig. 7 characteristics of theofstress-strain
the stress–strain curve
curve and AE and accumulated
accumulated acoustic
count of emission
the specimen with(AE) counts of
α=15°
the specimen with α = of ◦
15the. stress–strain curve and accumulated acoustic emission (AE) counts of
Figure 7. Characteristics
the specimen with α = 15°.

Figure 8 displays the strength variation of specimens containing different inclinations of

nonpersistent fissures. It can be seen that the fissure inclinations had a significant effect on the
strength behavior of the specimens. The peak strength initially decreased, and then increased as
Energies 2019, 12, 79 8 of 24

Figure 8 displays the strength variation of specimens containing different inclinations of

Energies 2019, 12, x fissures.
nonpersistent FOR PEER REVIEW
It can be
seen that the fissure inclinations had a significant effect on 8 ofthe
24
strength behavior of the specimens. The peak strength initially decreased, and then increased as fissure
fissure inclination
inclination increased
increased in thefrom
in the range range
0◦ from
to 90◦0° to 90°:
: 9.05 MPa 9.05
wasMPa
the was the minimum
minimum strengthstrength
at α = 30◦at, and
α=
30°,
28.2and
MPa28.2
wasMPa was the maximum
the maximum strength strength
at α = 90◦at. α = 90°.

1
0
0
3
0

Experiment values of crack initiation stress

Average value of crack initiation stress

8
7
.
4
2
5

8
0
Experiment values of peak stress
Average value of peak stress
Experiment values of λ
2
0

6
4
.
7

6
0
a Average value of λ
6
2
.
6
)

P
M
1
5

5
2
.
7

λ )
s
4
4
.
7

4
0
s
(

(
r
t
5

1
0

S
3
3
.
6

0
0

1
5

3 i
0 s

4 c
5 l

6
0

7
5

9
0
F
s
u
r
e
s
i
n
i
n
a
t
i
o
n

()
°

Figure 8. Influence
Figure 8. Influenceofofthethe fissure
fissure inclination
inclination on crack
on crack initiation
initiation stress
stress and and
peak peak
stress stress
of the of the
specimens
specimens
containingcontaining nonpersistent
nonpersistent fissures. fissures.

Figure 8 also
Figure alsoshows
showsthe thechange
change rule of crack
rule initiation
of crack stress
initiation magnitude
stress with the
magnitude variety
with of fissure
the variety of
inclinations. As with ◦
fissure inclinations. Asthe intact
with the specimen, with the
intact specimen, withspecimen with αwith
the specimen = 90α =split
90° directly at theatpeak
split directly the
point,point,
peak the crack initiation
the crack phenomenon
initiation did notdid
phenomenon appear before the
not appear peakthe
before strength, and therefore
peak strength, the crack
and therefore
initiation stress of the specimen with = 90 ◦ was not considered. It can be seen from the curve that
the crack initiation stress of the specimen with α = 90° was not considered. It can be seen from the
α
the crack
curve thatinitiation
the crackstress of thestress
initiation specimens
of the first decreased
specimens firstand then increased
decreased and thenasincreased
fissure inclinations
as fissure
increased from 0 ◦ to 90◦ . However, unlike the peak strength, the minimum crack initiation stress was
inclinations increased from 0° to 90°. However, unlike the peak strength, the minimum crack
3.8 MPa atstress
α = 45 ◦.
initiation was 3.8 MPa at α = 45°.
Here, the
Here, the crack
crack initiation
initiation stress
stress coefficient
coefficient λλ is
is introduced,
introduced, namely
namely the
the ratio
ratio of
of crack
crack initiation
initiation
stress to the peak strength,
stress to the peak strength,
σ
λ = ci , (3)
𝜆 = σp, (3)
where σσciciisisthe
where thecrack
crackinitiation
initiationstress
stress (MPa)
(MPa) andand σσppisisthe
thepeak
peakstrength
strength(MPa).
(MPa).
The trends
The trends inin Figure
Figure 88show showthat
thatthethecrack
crackinitiation
initiationstress coefficient
stress decreased
coefficient decreasedby by
increasing the
increasing
fissure inclinations from 0 ◦ to 45◦ , which indicated that the specimen containing fissures with α = 45◦
the fissure inclinations from 0° to 45°, which indicated that the specimen containing fissures with α =
waswas
45° thetheeasiest forfor
easiest crack
crackinitiation.
initiation.InInaddition,
addition,thisthiscrack
crackinitiation
initiation stress
stress coefficient increased by
coefficient increased by
increasing the fissure inclinations from 45 ◦ to 90◦ . A comparative analysis of the crack initiation stress
increasing the fissure inclinations from 45°to 90°. A comparative analysis of the crack initiation stress
coefficient showed
coefficient showed thatthat there
there was
was aa certain
certain relationship
relationship between
between peak
peak strength
strength and
and crack
crack initiation
initiation
stress, that
stress, that the
the higher
higher thethe crack
crack initiation
initiationstress,
stress,thethehigher
higherthe
thepeak
peakstrength,
strength,as asshown
shownin inFigure
Figure8.8.
This demonstrates that by suppressing the microcrack initiation before peak
This demonstrates that by suppressing the microcrack initiation before peak strength, the bearing strength, the bearing
capacity of the specimen can be improved, and gives guidance in improving
capacity of the specimen can be improved, and gives guidance in improving the bearing capacity of the bearing capacity of
jointed surrounding
jointed surrounding rock rock during
during engineering.
engineering.
The pre-existing fissures
The pre-existing fissures also also hadhadan impact on theon
an impact deformational characteristics
the deformational of the specimens.
characteristics of the
Figure 9 shows the influence of fissure inclinations on elastic modulus under
specimens. Figure 9 shows the influence of fissure inclinations on elastic modulus under uniaxial uniaxial compression.
compression. The elastic modulus Es was defined as the slope of the linear segment before the crack
initiation in the stress–strain curve. The trend of Es was basically the same as the peak strength, i.e.,
Es first decreased from 0° to 30° and then increased, with a minimum of 6.7 GPa at 30° and a
maximum of 8.0 GPa at 90°. It should be noted that the strain–stress curves of specimens may
experience multiple stress fluctuations when entering the nonlinear deformation stage after the
Energies 2019, 12, 79 9 of 24

The elastic modulus Es was defined as the slope of the linear segment before the crack initiation in the
stress–strain curve. The trend of Es was basically the same as the peak strength, i.e., Es first decreased
Energies
from ◦2019,
02019,
to 30 12,◦ x FOR then
PEER REVIEW
increased, with a minimum of 6.7 GPa at 30◦ and a maximum of 8.099GPa of 24
Energies 12, x and
FOR PEER REVIEW of 24
◦
at 90 . It should be noted that the strain–stress curves of specimens may experience multiple stress
crack initiation, as shown in Figure 6. Each stress fluctuation in the stress-strain curve will lead to a
fluctuations
crack initiation,when entering
as shown in the nonlinear
Figure 6. Eachdeformation stage in
stress fluctuation after
thethe crack initiation,
stress-strain as shown
curve will lead toina
stress redistribution within the specimen. The influence of initial microcrack initiation and
Figure redistribution
stress 6. Each stress fluctuation
within theinspecimen.
the stress-strain curve willof
The influence lead to a stress
initial redistribution
microcrack within
initiation and
propagation has limited effects on the overall load-bearing structure of specimens, but will lead to
the specimen.has
propagation The influence
limited effectsof on
initial
the microcrack initiation structure
overall load-bearing and propagation has limited
of specimens, but willeffects
lead on
to
structural damage of the specimen. As seen from Figure 7, the average modulus was slightly less
the overalldamage
structural load-bearingof thestructure
specimen. of As
specimens,
seen frombutFigure
will lead to structural
7, the damagewas
average modulus of the specimen.
slightly less
than the previous period.
As
thanseen
thefrompreviousFigure 7, the average modulus was slightly less than the previous period.
period.

Figure 9. Effects of the fissure inclinations on the elastic modulus of specimens containing
Figure 9.
Figure Effects ofof the
9. Effects the fissure
fissure inclinations
inclinations on
on the
the elastic
elastic modulus
modulus of
of specimens
specimens containing
containing
nonpersistentfissures.
nonpersistent fissures.
nonpersistent fissures.

3.2. Effect
3.2. Effect of
of Fissure
Fissure Inclinations
Inclinations on
on Crack
CrackCoalescence
CoalescencePatterns
Patternsand
andFailure
FailureModes
Modes
3.2. Effect of Fissure Inclinations on Crack Coalescence Patterns and Failure Modes
3.2.1.
3.2.1. Crack
Crack Coalescence
Coalescence Patterns
Patterns
3.2.1. Crack Coalescence Patterns
For
For specimens
specimens containing
containing more
more than
than one
one fissure,
fissure, coalescence
coalescence occurs
occurs when
when two
two initial
initial fissures
fissures
are For specimens containing more than one fissure, coalescence occurs when two initial fissures
are connected
connected through
through aanewly
newlyformed
formedcrack.
crack. In
In specimens
specimens withwith multiple
multiple fissures,
fissures, the
thecoalescences
coalescences
are connected complex.
become through a newly10 formed crack. In specimens with multiple fissures, the coalescences
become more more complex. Figure
Figure 10 shows
shows aa sketch
sketch ofof different
different crack
crack types
types that
that were
were identified
identified based
based
become
on more complex. Figure 10 shows a sketch of different crack types that were identified based
on their
their morphology,
morphology,failure
failuremodes,
modes,andandpropagation
propagationmechanism
mechanismin ingranite
granitespecimens
specimens containing
containing aa
on their morphology,
single failure modes, and propagation mechanism inpatterns
granite specimens containing a
singlefissure
fissure[46] andand
[46] werewereused for analyzing
used the crack
for analyzing thecoalescence
crack coalescence ofpatterns
specimens of containing
specimens
single
two fissure
fissures [46]
[47]. and
Here, were
these used
crack for
patternsanalyzing
could be the
used crack
as a coalescence
reference for patterns
analyzing theof specimens
propagation
containing two fissures [47]. Here, these crack patterns could be used as a reference for analyzing the
containing
and two fissures
coalescence [47]. Here, these crack patterns could be used as a reference for analyzing the
propagation andpattern of specimens
coalescence in specimens
pattern of this experiment.
in this experiment.
propagation and coalescence pattern of specimens in this experiment.

Figure 10. A
Figure 10. A sketch
sketch of
of the
theinitiated
initiatedcrack
crackclassifications
classifications of
of specimens
specimens containing
containing fissures
fissures (modified
(modified
Figure 10. A sketch of the initiated crack classifications of specimens containing fissures (modified
from
fromYang
YangandandJing
Jing[46]
[46]and
andYang
Yangetetal.
al.[47]).
[47]).
from Yang and Jing [46] and Yang et al. [47]).

Table 2 summarizes the crack coalescence patterns on the back of the rock-like specimens
Table 2 summarizes the crack coalescence patterns on the back of the rock-like specimens
containing nonpersistent fissures under uniaxial compression based on the classification. It could be
containing nonpersistent fissures under uniaxial compression based on the classification. It could be
concluded that the final macrofracture patterns of specimens containing nonpersistent fissures were
concluded that the final macrofracture patterns of specimens containing nonpersistent fissures were
all a mixture of various crack types, which were much more complicated than those in specimens
 Energies 2019, 12, 79 10 of 24

Table 2 summarizes the crack coalescence patterns on the back of the rock-like specimens
containing nonpersistent fissures under uniaxial compression based on the classification. It could be
Energies
Energies 2019,2019,
Energies
2019,
Energies 12, xx FOR
12,
2019, FOR
12,
12, PEER
xx PEER
FOR
FOR REVIEW
PEER
REVIEW
PEER REVIEW
REVIEW 10 of
10 of 10
24
24
10 of
of 24
24
Energies 2019,
Energies 12, x 12,
2019,
concluded FOR PEERPEER
x that
FOR theREVIEW REVIEW
final macrofracture patterns of specimens containing nonpersistent10fissures of10
24of 24
were
containing all
containing a mixture
few fewfissures. of various
fissures.Tensile crack
wingwing
Tensile types,
crack which
was was
crack were
always always much
the first
first
the more
crack complicated
to initiate
initiate than
in our
our those
specimens, in specimens
containing
containing
containing few fissures.
few
few fewfissures. Tensile
fissures.
Tensile wing
Tensile
wing crack
wing
crack was
crack
was always
was
always alwaysthe
the the first
first crack
first
crack
crack
to
crack to
to initiate
tocrack in
initiate
initiate
in
in our
our specimens,
specimens,
specimens,
and
containing
it
and containing
often
it initiated
often
fissures.
fewfrom
initiated fissures.
from
Tensile
the tips
the of
wing
Tensile
tips
crack
wing
pre-existing
of
was
crack
pre-existing
always
was
fissures always
fissuresthat
the first
the
were
that
crack
first
near
were the
near
to to in
initiate
specimen
the
our
in
initiate
specimen
specimens,
our
end,
specimens,
in end,
our
exceptspecimens,
except
and it
and
andand often
it
it often
it initiated
often
initiated
often from
initiated
initiatedfrom the
from
fromthe tips
the
tips
the of
tips
of
tipspre-existing
of pre-existing
pre-existing
of pre-existingfissures
fissures
fissuresthat
fissuresthat were
that
were
that near
were
near
were the
near
the
near specimen
the specimen
specimen
the specimen end,
end, except
end,
except
end, except
except
for and
for specimens
specimens it
specimens often
withwith initiated
α == α 0° and
==andfrom
0° the
90°. 90°.
and tips
NoteNote of pre-existing
that that
mostmost of the
the
of fissures
cracks
the that
outside
cracks were near
the fissured
outside the
fissured
the specimen
region
fissured region end,
werewere except
for for specimens
for specimens
for specimens with
withwithα
with 0°
α =withα
α0°=and 0° 90°.
and
0°=and90°. Note
90°.
Note
90°.
◦ and that
Note
◦that
Note most
that
most
that of
most
of most
most of
the cracks
of thethe
cracks outside
cracks
cracksoutside the
outside
outsidethe thethe
fissured region
fissured
region
fissured were
region
region were were
were
tensile
tensilefor specimens
cracks, and and
cracks, the the crack α 0
types
crack 90
within
types . Note
the the
within that
fissured
fissured of
regionthe
region cracks
werewere outside
closely the
related
closely fissured
related region
to fissure
fissure
to were
tensile
tensile cracks,
tensile
cracks,
tensile and
cracks,
cracks,and the
and
andthe crack
the
crack
the types
crack
types
crack within
types
within
types the
within
the
within fissured
the fissured
fissured
the fissured region
region region
region were
were closely
were
closely
were related
closely
closely related
related
related to
to to fissure
fissure
fissure
to fissure
tensileThe
inclinations.
inclinations. cracks,main
The and
main the crack
crack crack types
coalescence within
coalescence the
pattern fissured
pattern changed region
changed from were
from closely
tensile related
crack
tensile toto aafissure
crack to mixtureinclinations.
aa mixture
inclinations.
inclinations.
inclinations.
inclinations. The
TheThe main
Themain main
main crack
crack coalescence
crack
crack coalescence
coalescence
coalescence pattern
pattern
pattern
pattern changed
changed changed
changed from
from from
fromtensile
tensile crack
tensile
tensilecrack to
crack
to ato
crack tomixture
mixturemixture
a mixture
The
coalescence,
coalescence, main
and and crack
thenthen coalescence
to shear
shear
to crack pattern
coalescencechanged from
withwith tensile
the increase
increase crack to
in fissure
fissure a mixture
inclination. coalescence, and then to
coalescence,
coalescence,
coalescence, and then
and to
then to shearcrack
shear crack
crack coalescence
coalescence
coalescencewith the
with the
the increase in
increase in
in fissure inclination.
inclination.
fissure inclination.
shearand
coalescence, crackthen
and to shear
then
coalescence to crack
shearwith coalescence
crack increasewith
the coalescence the increase
with
in fissure the increase
inclination. in fissure
in inclination.
fissure inclination.
Table
Table 2. Crack
Table
2. Crack
Table 2. coalescence
2. Crack
coalescence
Crack patterns
coalescence
patterns
coalescence on the
patterns
on the
patterns on back
the
onback
the of the
back
of
backtheofspecimens’
the backs.
specimens’
ofspecimens’
the backs.
specimens’ backs.
backs.
Table 2. Crack
Table 2.
Table coalescence
Crack
2. Crack patterns
coalescence
coalescence on the
patterns on
patternsback
the of the
onback
the of
backspecimens’
the backs.
ofspecimens’ backs.
the specimens’ backs.
Crack
Crack Crack
Crack
Failure
Failure mode
Failure
mode
Failure
Failure andand
mode
and
mode
Mode overall
and
overall crackCrack
overall
crack
andOverall
overall crack
crack Crack
Crack
Crack coalescence
Crack coalescence
coalescence
Crack coalescence
Coalescence mode
mode
Mode in the
mode
in
mode the
in
in the Crack
the coalescenc Crack
α
α α
αα Failure mode
Failure and
mode overall
and crack
overall crack Crack coalescence
Crack coalescence mode in
mode the
in theCrack coalescenc
Coalescence
coalescenc
coalescenc Type
α α distribution
distribution
Distribution
distribution
distribution fissured
Fissured
fissured region
fissured
Regionregion
region
fissured region coalescenc
coalescenc
distribution
distribution fissured region
fissured region eee type
typeee type
type
type
e type

0° 0°
0° 0°0
◦ Tww;; T
T TTT
s;wS ;;;m
TTTss;s; ;SSSmmm
Tw; TTss;;wwwS;S
m
0° 0° Tm s; Sm

Tww;T
T ;TssTT;;;wwwT
T;T
;T ; Taa;;
aa;;ss;; T
15° 15° TTww;T;TT T;T
15° 15°
15◦
15° 15° SsTT
as;; S
TTas ;Ta;; asSS;;mmTT;;aas
m;as
sm ;
; ; Sm ;
as; S Tmas;; Sm;
as

Tww;; T
T TTT ;;;;S
STT ;;;;;; S m;
SSmm;;
asw mas
30° 30°
30° 30° Tw; TT as
w;; S
w Tmmasas
30° 30°
30◦ TSSwaa;;; asL
LT cS
S aa;;; SLLm cc ; Sa ; Lc
Sa; LSc a; Lc
cas

Tww;; T
T TTT ;wT
ss;w
T
;; as
T
T
as ;;ss;; TTasas;;
45° 45°
45° 45° Tw; STT s;w;TT ass;; Tas;
45° 45° m S
SSm S m m
m Sm
 Energies 2019, 12, 79 Tw; Tasw; T
Smas; Sm; 11 of 24
30° 30°
Sa; LSca; Lc

Table 2. Cont.

Failure Mode and Overall Crack Crack Coalescence Mode in the

α Crack Coalescence Type
Distribution Fissured Region

Tw; Tsw;;TTass; Tas;

45° 45°45◦ TSwm; TSs ; Tas ; Sm
m

Energies
Energies
Energies 2019,2019,
2019,
Energies 12, 12,
12, xx FOR
FOR xxx PEER
FOR PEER REVIEW
REVIEW
PEERPEER
REVIEW 11 of 11
11 of 24 of
24 24
Energies 2019,2019,
Energies 12, x 12,
2019, 12,
FOR FOR
FOR
PEERPEER REVIEW
REVIEW
REVIEW 11 of 11
24 of
11 of 24
24

T
as;s;s;;S
TTss;; TTas Tmmas
ST ; ; Smm;;
TT Tmasas;;;;; SS
as
60° 60°
60° 60◦ Ts; TS T
ass;s;S T Sm;; Sa
60° 60°
60° Saa S Saa m
Sa Sa

75° 75°
75° TT;s;s;;SS;T
TTss;; TTas TmmTas
;; ;; S
S ;;; ;
75° 75° Ts; TT
as
75°75
◦ as
T
as;ss;ST mas;as; SSm
;mmm

90° 90°
90° TTss Tss
90° 90°
90°90◦ Ts T
TTs s

3.2.2.
3.2.2.
3.2.2. Failure
Failure
Failure
3.2.2. Modes
Modes
Failure Modes
Modes
3.2.2.
3.2.2. Failure
Failure Modes
3.2.2.Modes
Failure Modes
As As mentioned
mentioned
As mentioned
mentioned
As previously,
previously,
previously,
previously, pre-existing
pre-existing
pre-existing
pre-existing fissure
fissure fissure inclinations
inclinations
inclinations
fissure inclinationsplay play
playplay an an important
an important important
important
an role role
in
rolerolein thein failure
the
in the failure
failure
the failure
As As mentioned
mentioned previously,
previously, pre-existing
pre-existing fissure fissure inclinations
inclinations play play
an an important
important role role
in thein failure
the failure
behavior
behavior
behavior of
of As
of mentioned
specimens:
specimens:
specimens: The
The previously,
The
failure
failurefailure
modes
modes pre-existing
modes
of
of the
theof thefissure
specimens
specimens inclinations
specimens are
are are
shown
shown play
shown in
in an
in
Table
Tableimportant
Table
22 Five
Five2 role
Five
dominant
dominantin the failure
dominant
behavior
behavior
behavior of
of specimens:
specimens:
of specimens: The TheThe
failurefailure
failure
modes modes
modesof the of
of the
the specimens
specimens
specimens are
are shown
are shown shown in
in Table
in Table Table
2 Five22 Five
Five dominant
dominant
failure
failure behavior
failure
modes
modes modescouldofcould
could specimens:
be
be classified
classified The
be classified
classified failure
according
according modes
accordingto
to to of
crack
crack crackthe specimens
coalescence
coalescence
coalescence modesare
modes shown
modeswith
with with
the
the 2 Fivedominant
in pre-existing
Table
the pre-existing
pre-existing
pre-existing dominant
fissures
fissures failure
fissures
failure
failure
failure modes modes
modes
could could
couldbe be
be classified
classified according
according
according to
totocrack crack
to crack coalescence
coalescence
coalescence modes modes
modeswith with
with
the the
the pre-existing
pre-existing fissures
fissures
fissures
and and
the modes
the failure
the failure
and and failure could
failure
process
process be
processof classified
of specimens,
specimens, according
of specimens,
specimens, as as shown
as shown
shown showncrack
in coalescence
in Figure
in Figure
Figure Figure
11. 11.
11. 11. modes with the pre-existing fissures and the
and and
the the
the failure
failure process
process of
of specimens, as
as shown in
in Figure 11.
failure process
process of of specimens,
specimens, as as shown
shown in in Figure
Figure 11.
11.
1.
1. 1.1. Failure
Failure
Failure mode
mode
Failure mode
mode III (tensile
II (tensile
(tensile(tensile
failure
failurefailure
across
across
failure across
the
across the fissure
the fissure
fissure
the fissure planes):
planes):
planes): This This
failure
ThisThis
planes): failure failure
mode
mode
failure mode
only
mode only
onlyonlyappearedappeared
appeared
appeared in
in thein the
the
in the
1. 1.Failure
Failure
mode modeI (tensile(tensile
failurefailure
across across
the the fissure
fissure planes):
planes): This This
failure failure
mode mode
only only appeared
appeared in in the
the
specimen
specimen
1.specimenwith
Failure with
α =
mode α
0°. = I0°. Mechanically,
Mechanically,
(tensile the
failure the formation
formation
across the of of
this
fissure this
type type
planes): of of
failure
This failure
mode
failure mode
was
mode was
due
only due
to theto the
appeared
specimen
specimen
specimen with
withwith α
with =
α = 0°. 0°.
α Mechanically,
α ==Mechanically,
0°.
0°. Mechanically,
Mechanically, the formation
the
the formation
the formation
◦ .within
formation of this
of type
of this
of this typetype
this of
type failure
of
of failure
of failure mode
failure
mode mode
modewas
was wasdue
due due
was to
due the
to the to the in
to the
production
production the of of
tensile
specimen tensile
stresses
with stresses
=within
0 the the
central
Mechanically, central surface
surface
the of
formation of open-type
open-type of this fissures
fissures
type of under under
failure loading
mode loading
production
production
production
production of tensile
of tensile
of
of tensile stresses
tensile
stressesstresses
α
stresseswithin
within the the
within
within
the central
the
centralcentralsurface
central surface
surface
surface of open-type
of open-type
of
of open-type
open-type fissures under
fissures
fissures
fissures under under
underloading
loadingloading due
was
loading
conditions
conditions
conditions and
to the
conditions and
andproduction subsequent
subsequent
subsequent
and subsequent initiation
ofinitiation
tensile
initiation of
stresses
initiation of within
of tensile
tensile
of tensile
crackscracks
the
cracks near
central near
nearnear the
the the
central
surface
centralcentral
of fissure.fissure.
open-type
fissure. With
With With
the
fissures the
under
the the
conditions
conditions and and subsequent
subsequent initiation
initiation of of tensile
tensile tensile
crackscracks
cracks
near near
the the
the
centralcentral
central fissure.
fissure.
fissure. With With
Withthe the
propagation
propagation
propagation
propagation of
of wing of
wing
wing
of wing
cracks
cracks cracks
to
to the to
the
the the
upper
upper upperand
and andand
lower
lower lower
end end
end endface, face, macrotensile
macrotensile
face,face,
macrotensile cracks
cracks cracks
went went
wentwent through
throughthrough
propagation
propagation of of wing
wing
cracks cracks
cracks
to to the
the upper
to upper upperand and
lower lower
lower
end end
face, face, macrotensile
macrotensile
macrotensile cracks cracks
cracks
went went
throughthrough
through
the
the the entire
entire
entire
the entire specimen.
specimen.
specimen.
specimen. The
The The increasing
increasing
increasing
The increasing tensile
tensile tensile
crack
crack
tensile crack apertures
apertures
apertures
crack apertures led
led ledthe
to
to
led to bending
the
to the bending
bending
the bending failure
failurefailure
of
of
failure of rock
rock
rock
of rock
the entire
the entire specimen.
specimen. The The increasing
increasing tensile tensile
crack crack apertures
apertures led to ledthe to bending
the bending failurefailure
of rockof rock
near near
the
nearnear the free
the free
free
the free
surface surface
surface
surface (Figure
(Figure (Figure
11a). 11a).
11a).11a).
(Figure This This
ThisThisfailure
failure failure
mode
failuremode mode
was
waswas
mode was categorized
categorized
categorized
categorized as as tensile
as tensile
tensile
as tensile
failure
failurefailure
across
across
failure across
across
near near
the the free
free surfacesurface (Figure (Figure
11a). 11a).
This This
failure failuremode mode
was was categorized
categorized as as tensile
tensile failurefailure
across across
the
the the
fissurefissure
fissure
the planes
planes
fissure planes
[28].
planes[28]. [28].
[28].
the fissure
the fissure planes planes[28].[28].
Energies 2019, 12, 79 12 of 24

loading conditions and subsequent initiation of tensile cracks near the central fissure. With the
propagation of wing cracks to the upper and lower end face, macrotensile cracks went through
the entire specimen. The increasing tensile crack apertures led to the bending failure of rock
near the free surface (Figure 11a). This failure mode was categorized as tensile failure across the
fissure planes [28].
2. Failure mode II (rotation failure of newly generated blocks): For specimens with α = 15◦ and
30◦ , as axial compression stress increased, wing cracks developed from pre-existing fissure tips.
Energies 2019, 12, x FOR PEER REVIEW 12 of 24
The wing cracks of parallel fissures coalesced and cut out a newly generated block that could
rotate. The strength
2. Failure mode IIof this failure
(rotation failuremodel wasgenerated
of newly controlled by the
blocks): Forrotation
specimensof the
withblock (Figure
α = 15° and 11b,c)
and was smaller than the strength of the other failure modes. Note that, although the tips.
30°, as axial compression stress increased, wing cracks developed from pre-existing fissure final crack
The wing cracks of parallel fissures coalesced and cut out a newly◦ generated block that could
coalescence in a horizontal rock bridge for specimens with α = 30 was similar to “stepped failure
rotate. The strength of this failure model was controlled by the rotation of the block (Figure
modes”, as reported by Prudencio and Van Sint Jan [23], the final failure of the core bearing
11b,c) and was smaller than the strength of the other failure modes. Note that, although the
structurefinalwas
crackstill categorized
coalescence in a as a rotation
horizontal rockfailure
bridgeof fornewly generated
specimens with α blocks.
= 30° was similar to
3. Failure“stepped
mode III (mix
failure failureasmode):
modes”, reported When the fissure
by Prudencio and Van inclinations
Sint Jan [23], approached
the final failure45of◦the
, the load
core bearing structure was still categorized as a rotation failure of newly
increased, and tensile wing cracks were propagated to link the parallel fissures tips. Some shear generated blocks.
3. Failure mode III (mix failure mode): When the fissure inclinations approached 45°, the load
cracks were propagated along the inclined direction to interact with the coplanar fissures
increased, and tensile wing cracks were propagated to link the parallel fissures tips. Some
(Figureshear
11d,e). Thus the macro failure pattern was a mixture of shear failure and tension failure,
cracks were propagated along the inclined direction to interact with the coplanar fissures
and this failure
(Figure phenomenon
11d,e). Thus the macrocould be classified
failure pattern wasasa a mixedoffailure.
mixture shear failure and tension failure,
4. failure phenomenon could be ◦
Failure mode IV (shear failure): At α = 60 and 75 , the failure.
and this classified as a ◦mixed shear crack initiated from the fissure
tip4.andFailure mode IV through
propagated (shear failure): At α = 60° and rock
the intervening 75°, the shear crack
bridge untilinitiated from the
it coalesced fissure
with thetipadjacent
and propagated through the intervening rock bridge until it coalesced
coplanar crack tip, finally forming a macro shear plane along the fissure plane. This failure mode with the adjacent
coplanar crack tip, finally forming a macro shear plane along the fissure plane. This failure
was classified as shear failure (Figure 11e,f). Even though there was wing crack coalescence in
mode was classified as shear failure (Figure 11e,f). Even though there was wing crack
specimenscoalescence 60◦ during
with αin= specimens withloading,
α = 60°shear
duringcracks
loading,played
shear acracks
key role in the
played final
a key rolefailure
in the mode.
5. Failurefinal
mode V (splitting
failure mode. ◦
failure): When fissure inclination reached 90 , the specimen finally
5. due
failed Failure mode V (splitting
to splitting failure, failure):
as shown When fissure inclination
in Figure 11g. Notereached 90°, the specimen
that, although a macro finally
failure plane
crossedfailed
throughdue pre-existing
to splitting failure,
fissures,as shown in Figure
the failure process11g. showed
Note that, although
that a macrooffailure
the influence pre-existing
plane crossed through
◦ pre-existing fissures, the failure process showed that the influence of
fissures with α = 90 on the failure process could be neglected. Therefore, this failure mode was
pre-existing fissures with α = 90°on the failure process could be neglected. Therefore, this
classified as amode
failure splitting failure.as a splitting failure.
was classified

T T
T T T
Rotation
T
Rotation

T T
T
T T T
T

Figure
Figure 11. Five11.failure
Five failure
modes. modes. (a) Tensile
(a) Tensile failure
failure across
across thethe fissure
fissure planes;(b,c)
planes; (b,c)rotation
rotation failure
failure of
of newly
newly generated blocks; (d) mix failure mode; (e,f) shear failure; (g) splitting
generated blocks; (d) mix failure mode; (e,f) shear failure; (g) splitting failure. failure

3.3. Fracture Evolution Process

Energies 2019, 12, 79 13 of 24

Energies 2019, 12, x FOR PEER REVIEW 13 of 24

3.3. Fracture Evolution Process
The experimental results proved that the final failure modes could be divided into five patterns.
The experimental results proved that the final failure modes could be divided into five patterns.
Using 0°, 30°, 60°, 75°, and 90° fissure inclinations, the fracture evolution process was analyzed
Using 0◦ , 30◦ , 60◦ , 75◦ , and 90◦ fissure inclinations, the fracture evolution process was analyzed based
based on the strain field of specimens combined with curves of stress, AE counts, and accumulated
on the strain field of specimens combined with curves of stress, AE counts, and accumulated AE counts
AE counts over time. Note that the black areas of the strain fields were due to the failure of subsets
over time. Note that the black areas of the strain fields were due to the failure of subsets caused by
caused by macrocrack or surface spalling on specimen surfaces during crack propagation. For the
macrocrack or surface spalling on specimen surfaces during crack propagation. For the convenience of
convenience of comparative analysis, the fissures are marked ①~⑨.
comparative analysis, the fissures are marked ~ 1 .9

3.3.1. Tensile
3.3.1. TensileFailure
Failureacross
acrossthe
theFissure
FissurePlanes
Planes
The curves
The curves of ofstress
stressandandacoustic
acousticemission
emissioncharacteristics
characteristics during
during the the loading
loading period
period for
forαα==00° ◦
are shown
are shown in in Figure
Figure 12.12. The
The strain
strain field
field of of point
point (a)(a) was
was used
used asas the
the reference
reference image
image toto which
which the the
subsequent photographs (points b–h) were correlated. The stress curve
subsequent photographs (points b–h) were correlated. The stress curve of the specimen initially showed of the specimen initially
showed nonlinear
nonlinear deformation, deformation,
which usuallywhich usually
resulted resulted
from from the
the closure closure
of open of open microcracks.
microcracks. The stress curveThe
stress curve was smooth, and acoustic emission characteristics were not
was smooth, and acoustic emission characteristics were not obvious. Along with the load increases, the obvious. Along with the
load increases,
strain concentration the strain concentration
zone began to form near zone thebegan to form
center of thenear
rightthe centeras
fissures, ofshown
the right fissures,
in Figure as
13b.
shown in Figure 13b. This was due to loading, which could easily lead to
This was due to loading, which could easily lead to the appearance of tensile stress in the center of the appearance of tensile
stressfissures.
open in the center of open
The strain fissures. The zone
concentration straininitially
concentration zone initially
only appeared in theonly appeared
right in the right
side, possibly due
side,
to thepossibly due to the slight
slight prominence of theprominence
top and bottom of theoftoptheand bottom ofonthe
specimens thespecimens
right side, oncausing
the rightstress
side,
causing stresson
concentration concentration
the right side onduring
the right theside during
initial stage.the initial stage.
When the specimen was loaded to 64.7% σ p, the stress curve had a little fluctuation, and evident
When the specimen was loaded to 64.7% σp , the stress curve had a little fluctuation, and evident
AE events occurred at
AE events occurred at the
the same
same time.
time. It It could
could be be observed
observed fromfrom thethe strain
strain field
field that
that the
the strain
strain
concentration zone in the right joints coalesced, and that the cracks initiated
concentration zone in the right joints coalesced, and that the cracks initiated from the center of the from the center of the
rightfissures
right fissuresand andpropagated
propagatedto tothe
thetoptopand andbottom
bottomedge edgeof ofthe
thespecimen.
specimen.The Thetime
timebetween
betweenpoints
points
(b) and (c) was approximately 0.3 s, indicating that the new crack growth
(b) and (c) was approximately 0.3 s, indicating that the new crack growth was rapid. The AE counts was rapid. The AE counts
started to
started to appear
appear more
morefrequently
frequently than
than before,
before,which
which meant
meant the
the crack
crack began
began to topropagate
propagate from
fromthis
this
point,and
point, andtherefore
thereforethisthispoint
pointcould
couldbe beconsidered
consideredtotobe bethe
thecrack
crackinitiation
initiationpoint.
point.

15 100 150
Axial stress
AE counts
Accumulated AE counts (×1000)

12 Accumulated AE counts 80 120

Axial stress (MPa)

AE counts (×100)

9 60 90

6 40 60

3 20 30

0 0 0
0 100 200 300 400 500 600
Time (s)
◦ ).
Figure 12.Characteristics
Figure12. Characteristicsof
ofstress
stressand
andacoustic
acousticemissions
emissionsduring
duringthe
theloading
loadingprocess
process(α(α==00°).

Then, the load continued to increase, the crack of the right fissures continued to propagate, and
Then, the load continued to increase, the crack of the right fissures continued to propagate, and
the strain concentration zone appeared on the left side of the right fissures, as shown in Figure 13d.
the strain concentration zone appeared on the left side of the right fissures, as shown in Figure 13d.
When the axial stress reached the peak stress point (e), the crack initiated from the center of fissure 2
When the axial stress reached the peak stress point (e), the crack initiated from the center of fissure
and propagated toward the top edge of the specimen; the strain concentration zone between fissure
② and propagated toward the top edge of the specimen; the strain concentration zone between
2 and fissure 9 coalesced, and macrocracks appeared; this led to a slight drop in the stress curve.
fissure ② and fissure ⑨ coalesced, and macrocracks appeared; this led to a slight drop in the stress
The strain concentration zone appeared in the center of fissure 7 and propagated toward the bottom
curve. The strain concentration zone appeared in the center of fissure ⑦ and propagated toward the
bottom edge of the specimen, and the strain concentration zone appeared between the right tip of
fissure ⑦ and the left tip of fissure ③. At the point (f) just after the peak stress, the stress curve had
Energies 2019, 12, 79 14 of 24

edge of the specimen, and the strain concentration zone appeared between the right tip of fissure 7
Energies 2019, 12, x FOR PEER REVIEW 14 of 24
and the left tip of fissure . 3 At the point (f) just after the peak stress, the stress curve had a large drop

aand associated
large drop and with obviouswith
associated AE counts:
obviousIt AE
cancounts:
be seenItfrom
can the strain
be seen fieldthe
from that the crack
strain initiated
field that from
the crack
the center of fissure 7 experienced rapid crack propagation. In the stage of post-peak stress softening,
initiated from the center of fissure ⑦ experienced rapid crack propagation. In the stage of post-peak
the tension
stress cracks
softening, intension
the the upper and in
cracks lower fissuresand
the upper continued to extend
lower fissures toward the
continued end surface
to extend toward of the
the
specimen;
end surfacethe
of upper right side
the specimen; theofupper
the specimen
right sidebegan
of thetospecimen
spall; and the specimen
began gradually
to spall; and lost its
the specimen
bearing capacity, as shown in Figure 13g,h.
gradually lost its bearing capacity, as shown in Figure 13g,h.

Figure 13.
Figure Evolution of
13. Evolution of strain
strain fields
fields for
for the
the specimen
specimen with 0◦ corresponding
with αα == 0° corresponding to
to points
points (a–h)
(a–h) in
in
Figure 12.
Figure 12.
3.3.2. Rotation Failure of Newly Generated Blocks
3.3.2. Rotation Failure of Newly Generated Blocks
The curves of stress and acoustic emission characteristics during the loading period for α = 30◦
The curves of stress and acoustic emission characteristics during the loading period for α = 30°
are shown in Figure 14. Point (a) in the strain field (Figure 15a) is the reference image. As shown in
are shown in Figure 14. Point (a) in the strain field (Figure 15a) is the reference image. As shown in
Figure 15b, the strain concentration zone began to form near the fissure tips with the increasing load,
Figure 15b, the strain concentration zone began to form near the fissure tips with the increasing load,
but the stress concentration at the fissure tips did not exceed the material strength, and therefore there
but the stress concentration at the fissure tips did not exceed the material strength, and therefore
was no crack initiation. When the load came to 3.8 MPa (38.8% σp ), the obvious wing crack (Figure 15c)
there was no crack initiation. When the load came to 3.8 MPa (38.8% σp), the obvious wing crack
rapidly initiated along the direction of axial loading from the left fissure tips of fissures
1 2 and the
(Figure 15c) rapidly initiated along the direction of axial loading from the left fissure tips of fissures
right tips of fissures
8 ,
9 which caused a drop in the stress curve, and the strain concentration degree
①② and the right tips of fissures ⑧⑨, which caused a drop in the stress curve, and the strain
of the tips of fissures 3
4
5 6 grew after the first stress drop. After point (c), with the axial stress
concentration degree of the tips of fissures ③④⑤⑥ grew after the first stress drop. After point (c),
continuing to increase, the strain concentration zones at the left tip of fissure 3 and the right tip of
with the axial stress continuing to increase, the strain concentration zones at the left tip of fissure ③
fissure 7 developed into wing crack and propagated gradually to the edge of the specimen, as shown
and the right tip of fissure ⑦ developed into wing crack and propagated gradually to the edge of
in Figure 15d–g. When the specimen was loaded to peak stress, the newly separated blocks between
the specimen, as shown in Figure 15d–g. When the specimen was loaded to peak stress, the newly
adjacent noncollinear fissures were cutting out, with the wing crack between fissures coalesced, and
separated blocks between adjacent noncollinear fissures were cutting out, with the wing crack
the specimen gradually lost its bearing capacity with the wing crack initiated from the upper and
between fissures coalesced, and the specimen gradually lost its bearing capacity with the wing crack
lower fissure tips propagating to the end face, as shown in Figure 15h. The increasing load led to the
initiated from the upper and lower fissure tips propagating to the end face, as shown in Figure 15h.
rotation of newly generated blocks, which finally resulted in the failure of the specimen, as shown in
The increasing load led to the rotation of newly generated blocks, which finally resulted in the
Figure 11c. Due to the failure of the image analysis grid caused by the block rotation, the strain-field
failure of the specimen, as shown in Figure 11c. Due to the failure of the image analysis grid caused
developments after the peak strength were difficult to obtain in the final failure.
by the block rotation, the strain-field developments after the peak strength were difficult to obtain in
the final failure.
Energies 2019,12,
Energies 2019, 12,79
x FOR PEER REVIEW 15
15 of
of 24
24
Energies 2019, 12, x FOR PEER REVIEW 15 of 24

10 25
10 Axial stress g 25
40
Axial stress g 40
AE counts
AE counts AE counts
Accumulated h

Accumulated AE counts (×1000)

8 Accumulated AE counts h 20 32

Accumulated AE counts (×1000)

8 20 32

Axial stress (MPa)

AE counts (×100)
Axial stress (MPa)

AE counts (×100)
6 e 15 24
6 e 15 24
Crack
Crack cc
initiation
f
f
4 initiation 10 16
4 38.8% p 10 16
38.8% p
b dd
2 aa
2 5
b 5
8
8

0 0 0
0
0 50 100 150 200 250 300 350 0 0
0 50 100 150
Time (s)200 250 300 350
Time (s)
Figure 14.
14. Characteristics of
of stress and
and acoustic
acoustic emissions
emissions during
during the
the loading
loadingprocess
process(α
(α== 30
30°).
◦ ).
Figure 14.Characteristics
Figure Characteristics ofstress
stress and acoustic emissions during the loading process (α = 30°).

Figure 15. Evolution of strain fields for the specimen with α = 30◦ corresponding to points (a–h) in
Figure 15. Evolution of strain fields for the specimen with α = 30° corresponding to points (a-h) in
Figure 13.15. Evolution of strain fields for the specimen with α = 30° corresponding to points (a-h) in
Figure
Figure 13.
Figure 13.
3.3.3. Mix Failure Mode
3.3.3. Mix Failure Mode
3.3.3. Mix 16
Figure Failure
showsMode
the characteristics of the stress and acoustic emissions for the specimen containing
Figure 16fissures
nonpersistent shows with
the characteristics
= 45◦ duringof the the stressprocess,
loading and acoustic
and the emissions for thestrain
corresponding specimen
filed
Figure 16 shows the αcharacteristics of the stress and acoustic emissions for the specimen
containing
atcontaining
points (a–h)nonpersistent
are shown in fissures with
Figurewith α =
17. When 45° during
the the
pressure loading process, and the corresponding
nonpersistent fissures α = 45° during the was loaded
loading to point
process, and(b) the(38.5% σp ), the
corresponding
strain concentration
strain filed at points zone
(a–h) appeared
are shownininthe Figure
upper 17.tip
When
of the pressure
fissure
1 and thewaslower
loadedtiptoofpoint (b)
fissure (38.5%
and
strain filed at points (a–h) are shown in Figure 17. When the pressure was loaded to point (b)9(38.5%
σ p), the strain concentration zone appeared in the upper tip of fissure ① and the lower tip of fissure
propagated along
σp), the strain the direction
concentration of loading,
zone appeared accompanied
in the upperby tipa of
first stress①
fissure dropandand
the certain
lower tipAEofcounts.
fissure
⑨ loading
As and propagated
increased,along
the the
straindirection of loading,
localization zone accompanied
was concentrated by a first
at thestress drop andfissure
pre-existing certaintips.
AE
⑨ and propagated along the direction of loading, accompanied by a first stress drop and certain AE
counts.
In As loading
this period increased,
(points b–d), thethe strain
stress localization
curve was on the zone
rise,was concentrated
accompanied by at the pre-existing
certain AE counts, fissure
which
counts. As loading increased, the strain localization zone was concentrated at the pre-existing fissure
tips. In this
indicated period
that (points b–d),
the microcracks weretheinstress curve was on the rise, accompanied by certain AEpoint
counts,
tips. In this period (points b–d), the stable
stress propagation.
curve was onWhen the
the rise, axial stress was
accompanied byloaded
certaintoAE (e),
counts,
which
a which indicated
tensile indicated that
wing crackthat the microcracks
initiated from the were
upper in
tip stable
of propagation.
fissure When the axial stress
3 and propagated rapidly along the loading was loaded
the microcracks were in stable propagation. When the axial stress was loaded
to point (e),
direction. a tensile wing crack initiated from thepeak
upper tip (point
of fissure ③ wing
and propagated rapidly
to point When the axial
(e), a tensile stress
wing was
crack increased
initiated frominto the upperstress f), new
tip of fissure ③ and cracks appeared
propagated at
rapidly
along the loading direction. When the axial stress was increased into peak stress (point f), new wing
along the loading direction. When the axial stress was increased into peak stress (point f), new wing
 Energies
Energies 12,12,
2019,
2019, 79 PEER REVIEW
x FOR 16 16
of of
24 24
Energies 2019, 12, x FOR PEER REVIEW 16 of 24

cracks appeared at the right tip of fissures ⑧⑨. Meanwhile, tensile cracks [22,48] coalesced
cracks appeared at the right
8 .9 tip of fissures ⑧⑨. Meanwhile, tensile cracks [22,48] coalesced
the right
between tip of fissures
fissures ②⑤ and Meanwhile,
fissures ④⑦. The tensile cracks
crack [22,48]and
initiation coalesced between
propagation resulted in
fissures 2 5 and
energy
between
fissures fissures
4 .
7 The②⑤ crack fissures ④⑦.
andinitiation and The crack initiation
propagation resulted and
in propagation
energy release. resulted
Subsequent in energy
stresses
release. Subsequent stresses were redistributed in the specimen, immediately leading to a high strain
release. Subsequent stresses
were redistributed were redistributed in the specimen, immediately leading to zone
a highinstrain
concentration zone ininthe theleft
specimen, immediately
tip of fissure ⑦, as shownleading
into a high
Figure strain
17g. Then,concentration
the strain concentration the left
concentration
tip of fissure zone
,
7 as inshown
the leftintip of fissure
Figure 17g. ⑦,
Then,as shown
the in Figure
strain 17g. Then,
concentration zone the
ofstrain
the concentration
right tip of fissure
zone of the right tip of fissure ⑦ propagated to a wing crack, and the rock bridge between fissures
zone
of the right tip
7 propagated to aofwing
fissure ⑦ propagated
crack, and the rock tobridge
a wingbetween
crack, and the rock
fissures 1 2 bridge
between
3 coalesced fissures
with a shear
①②③ coalesced with a shear crack. The propagation of wing crack and coalescence of shear cracks
①②③ coalesced
crack. The with
propagation a shear
of wingcrack. The propagation of wing crack and coalescence of shear cracks
resulted in integral instability, andcrack and coalescence
the stress curve dropped of shear cracks
to 51.0% σp. resulted in integral instability,
resulted
and theinstress
integral instability,
curve droppedand the stress
to 51.0% σp . curve dropped to 51.0% σp.
12 40 80
Axial stress
12
Axial stress g 40 80
AE counts g
10 AE counts

(×1000)
Accumulated AE counts
10

(×1000)
Accumulated AE counts
f 30 60
ef 30 60
e
(MPa)

(×100)
8

counts
(MPa)

(×100)
8

counts
d
stress

counts
6 20 40
d

AEAE
stress

Crack b

counts
6 20 40
Crack

Accumulated
initiation
Axial

Accumulated
initiation c

AEAE
h
Axial

4 38.5% p
4 38.5% p c h 10 20
10 20
2
2 a
a
0 0 0
00 100 200 300 400 0
500 0
0 100 200Time (s)300 400 500
Time (s)
Figure 16.16.
Figure Characteristics of of
Characteristics stress and
stress acoustic
and emissions
acoustic during
emissions the
during loading
the process
loading (α(α= =
process 45◦ ).
45°).
Figure 16. Characteristics of stress and acoustic emissions during the loading process (α = 45°).

Figure
Figure17.17. Evolution
Evolution of of strain
strain fields
fields forfor the
the specimen
specimen with
with α α= =
45° ◦ corresponding
45corresponding to to points
points (a–h)
(a–h) inin
Figure 17. Evolution of strain fields for the specimen with α = 45° corresponding to points (a–h) in
Figure
Figure 16.16.
Figure 16.
3.3.4.
3.3.4. Shear
Shear Failure
Failure along
along the
the Fissure
Fissure Plane
Plane
3.3.4. Shear Failure along the Fissure Plane
The
The stress
stress and
and AEAEcount
countcurves
curvesofofthe
thespecimen
specimencontaining
containing nonpersistent
nonpersistent fissures
fissures with
with 75◦
α α= =75°
The stress
subjected and AEcompressive
count curvesstress
of the arespecimen incontaining nonpersistent fissures with αfields
= 75°
subjected totouniaxial
uniaxialcompressive stress are shownin
shown Figure 18,
Figure 18, and
and the
the corresponding
corresponding strain
strain fieldsof
subjected to uniaxial compressive stress are shown in Figure 18, and the corresponding strain fields
ofmark
markpoints
points(a–h) areare
(a–h) shown
shownin Figure
in Figure19. Taking pointpoint
19. Taking (a) in(a)
theininitial loadingloading
the initial period period
as a reference
as a
of mark points (a–h) are shown in Figure 19. Taking point (a) in the initial loading period as a
reference point, the small strain concentration only appeared in the fissure tip before the crack
reference point, the small strain concentration only appeared in the fissure tip before the crack
Energies 2019, 12, 79 17 of 24

point, the 2019,

Energies small 12,strain concentration
x FOR PEER REVIEW only appeared in the fissure tip before the crack initiation, 17 of 24as
shown in Figure 19b. When the axial stress was loaded up to point (c) (74.3% σp ), a strain localized
Energies 2019,
as12, x FOR PEER REVIEW19b. When the axial stress was loaded up to point (c) (74.3% 17 of 24 p), a
zoneinitiation,
appeared shown
in in Figure
the upper tip of fissure 1 and the lower tip of fissure 9 and propagated σalong
strain localized
the direction zone appeared in same
the upper tip of fissure ① and the lower of fissure ⑨
tipobvious and
initiation,ofasaxial
shown loading.
in Figure At 19b.
the Whentime, stress
the axial fluctuation
stress was loaded occurred with
up to point (c) (74.3%AE σp),counts,
a
propagated along the direction of axial loading. At the same time, stress fluctuation occurred with
which implied
strain that zone
localized the specimen
appeared entered
in the upperinto tip
theofcrack
fissure ① and period.
initiation the lower The of fissure ⑨
tip specimen wasandloaded
obvious AE counts, which implied that the specimen entered into the crack initiation period. The
propagated
to point (e) withalong the direction
increasing of axial loading.AtAtthis
axial deformation. the moment,
same time,coalescence
stress fluctuation occurred
of strain with
concentration
specimen AE wascounts,
loadedwhichto point (e) with increasing axial deformation.
into theAt thisinitiation
moment, coalescence of
zonesobvious
occurred between implied
the collinear that the specimen
fissures, and the strainentered crack
concentration along the period.
direction Theof the
strain
specimenconcentration
was loaded zones occurred
to higher,
point (e)as with between
increasing the collinear
axial19e. fissures,
deformation. and the strain concentration
fissure inclination became shown in Figure As with At thethis moment,
prior coalescence
specimens, thereofwas a
along
strainthe direction ofzones
concentration the fissure
occurred inclination
between became
the collinearhigher, as shown
fissures, and thein Figure 19e. As with the
strain concentration
strong relationship between the coalescence of the strain concentration zone and the development of
prior
alongspecimens,
the direction there
of was a strong
the fissure relationship
inclination between
became higher,theascoalescence of the 19e.
shown in Figure strainAsconcentration
with the
the zone
stress–strain
and the curve (i.e., the
development of coalescence
the stress–strainof high
curve strain
(i.e., concentration
the coalescence zones
of high always
strain led to a drop
concentration
prior specimens, there was a strong relationship between the coalescence of the strain concentration
in stress,
zones which
zone andalways subsequently
led to a drop
the development
decreased
in stress,
of the
the bearing
which
stress–strain
capacity decreased
subsequently
curve
of the specimen).
(i.e., the coalescence of the
highbearing
When the specimen
capacity
strain concentration of the
wasspecimen).
loaded to
zones always point
When (f),
led the which
to aspecimen was just before
was loaded
drop in stress, peak strength,
whichtosubsequently
point (f), which shear crack
was just
decreased coalescence
the before
bearingpeak occurred
strength,
capacity of the in the
shear
rockcrack
bridge
specimen). along
coalescence the fissure
When occurred
the specimenplane,
in thewas leading
rock to
bridge
loaded the initiation
along(f),
to point thewhich of
fissure an
was anti-tensile
plane, leading
just before secondary
peakto the crack
shearoffrom
initiation
strength, an
the anti-tensile
lower tip ofsecondary
crack coalescencefissure .
occurred Afterward,
3 crack infrom the the
the rock sliptip
bridge
lower ofofthe
along theshear
fissure ③.plane
fissure ledleading
plane,
Afterward,to the
thetofailure of of
thethe
the initiation
slip specimen
of
shear an
plane
anti-tensile
(Figure
led 19g,h)
to secondary
and
the failure a of thecrack
rapid dropfrom
specimenin the thestress–strain
lower
(Figure tip ofand
19g,h) fissure
curve. ③. Afterward,
a rapid drop in thethe slip of the curve.
stress–strain shear plane
led to the failure of the specimen (Figure 19g,h) and a rapid drop in the stress–strain curve.

Figure
Figure 18. Characteristics
18. 18.
Figure Characteristics
Characteristics ofof
ofstress
stress
stressand
andandacoustic
acousticemissions
acoustic emissions
emissionsduring
duringthe
during loading
the
the process
loading
loading (α = (α
process 75°). 75◦ ).
==75°).

[％][％]
4.750
4.750
①
④①
⑦④
⑦
⑤
⑧ ②
② 4.218
4.218
⑧⑤ ③
⑥
⑨ ③
⑨⑥ 3.685
3.685
3.152
(a)σ1=0MPa, (b)σ1=13.97MPa (c)σ1=15.12MPa (d)σ1=17.13MPa
3.152
(a)σ 1=0MPa,
ε1=0×10 -3 (b)σ 1=13.97MPa
ε1=2.54×10-3 (c)σ 1=15.12MPa
ε1=2.73×10 ε(d)σ 1=17.13MPa 2.619
1=3.01×10
-3 -3
ε1=0×10 -3
ε1=2.54×10-3 ε1=2.73×10-3 ε1=3.01×10-3 2.619
2.086
2.086
1.553
1.553
1.021
1.021
0.488
0.488
(e)σ1=17.32MPa(f)σ1=19.45MPa (g)σ1=10.05MPa (h)σ1=6.88MPa -0.045
ε1=3.16×10
(e)σ
-3 ε1=3.64×10
=17.32MPa(f)σ
-3
1=19.45MPa (g)σ =10.05MPaε(h)σ
ε1=3.77×10 -3
1=4.03×10
-3
1=6.88MPa
1 1 -0.045
ε1=3.16×10 -3 εfields
1=3.64×10
-3
ε1=3.77×10 -3 ε◦1=4.03×10-3 to points (a–h) in
Figure 19. Evolution
Figure of strain
19. Evolution fields
of strain for
forthe
thespecimen
specimen withαα==75°
with 75corresponding
corresponding to points (a–h) in
Figure
Figure 18. 18.
Figure 19. Evolution of strain fields for the specimen with α = 75° corresponding to points (a–h) in
Figure 18.
 Energies 2019, 12, x FOR PEER REVIEW 18 of 24
Energies 2019,
Energies 12,12,
2019, x FOR
79 PEER REVIEW 18 18
of 24
of 24

3.3.5. Splitting Failure

3.3.5. Splitting Failure
3.3.5.When
Splitting
theFailure
fissure inclination approached 90°, the failure process was different from other
When the fissure inclination approached 90°, the failure process was different from other
inclinations.
When FiguresFigures
the fissure 20 inclination
and 21 show the stress 90 and
◦ , AE count characteristic curves of afromspecimen
inclinations. 20 and 21 showapproached
the stress and AE the failure
count process was
characteristic different
curves of a specimen other
containing
inclinations. nonpersistent
Figures 20 fissures
and 21 with
show α
the = 90°
stress during
and AE the uniaxial
count compression
characteristic curvesprocess
of a and the
specimen
containing nonpersistent fissures with α = 90° during the uniaxial compression process and the
corresponding strain fieldsfissures
containing nonpersistent of markwithpoints=(a–c), respectively.
90◦respectively.
during the As As seencompression
uniaxial from the strain field of point
corresponding strain fields of mark pointsα(a–c), seen from the strainprocess field of andpointthe
(b) (Figure 21b), only small strain concentrations appeared between fissure ② and fissure ③. When
(b) (Figure 21b), only small strain concentrations appeared between fissure ② and fissure ③. When(b)
corresponding strain fields of mark points (a–c), respectively. As seen from the strain field of point
the specimen was small
loadedstrain
to peak stress, cracks rapidly between
propagated from 2 the top end to 3 the bottom
(Figure
the 21b),was
specimen only loaded to peakconcentrations
stress, cracks appeared
rapidly propagatedfissure from the and end to .
top fissure the When
bottom the
end via
specimen fissures
was loaded ③⑤⑦ (Figure
to peak stress,21c). The specimen failed, and the axial supporting capacity
end via fissures ③⑤⑦ (Figure 21c).cracks
The rapidly
specimen propagated
failed, and fromthethe top supporting
axial end to the bottomcapacityend
dropped,
via fissuresaccompanied
3 5 7 by a21c).
(Figure largeTheamount of AEfailed,
specimen events. andThetheAE events
axial stayed atcapacity
supporting a very low level
dropped,
dropped, accompanied by a large amount of AE events. The AE events stayed at a very low level
before the peak stress,
accompanied and dense of AEAE events only appeared after reaching a peak stress. Compared to
before the peak by a large
stress, andamount
dense AE events events.
only The AE events
appeared stayed
after reaching atpeak very low level
stress. Comparedbeforeto the
other fissure
peakfissure
stress, and inclinations,
dense AEthe the
events strain concentration
onlyconcentration
appeared after of specimen surface was not obvious before
other inclinations, strain ofreaching
specimen peak stress.was
surface Compared to other
not obvious fissure
before
reaching
inclinations,peak the stress,
strain demonstratingofthat
concentration the pre-existing
specimen surface was fissures
not had before
obvious little effect
reachingon peak
the failure
stress,
reaching peak stress, demonstrating that the pre-existing fissures had little effect on the failure
behavior of
demonstrating the jointed
that the rock mass, and the specimen with α = 90° did not experience a crack initiation
behavior of the jointed rockpre-existing
mass, and the fissures had with
specimen little αeffect
= 90°ondidthe
notfailure
experiencebehavior
a crackof initiation
the jointed
period
rock similar
mass, and totheother specimens.
specimen with α = 90 ◦ did not experience a crack initiation period similar to
period similar to other specimens.
other specimens.
30 5 3
30 Axial stress 5 3
Axial stress
AE counts
AE counts

Accumulated AE counts (×1000)

25 Accumulated AE counts 4

Accumulated AE counts (×1000)

25 Accumulated AE counts 4
Axial stress (MPa)

AE counts (×100)
20 2
Axial stress (MPa)

AE counts (×100)
20 3 2
3
15
15
2
2
10 1
10 1
1
5 1
5

0 0 0
0 0 100 200 300 400 500 600 0
700 0
0 100 200 300 Time400
(s) 500 600 700
Time (s)
Figure 20. Characteristics
Characteristics of stress and acoustic emissions during the loading process (α
(α =
= 90°).
90◦ ).
Figure
Figure 20.20.
Characteristics of of stress
stress andand acoustic
acoustic emissions
emissions during
during thethe loading
loading process
process (α = 90°).

Figure 21.
Figure Evolution of
21. Evolution of strain
strain fields
fields for
for the
the specimen
specimen with
with α
α= 90◦ corresponding
= 90° corresponding to
to points
points (a–c)
(a–c) in
in
Figure 21.
Figure Evolution
19. of strain fields for the specimen with α = 90° corresponding to points (a–c) in
Figure 19.
Figure 19.
4. Discussion
4. Discussion
4. Discussion
4.1. AE Count Characteristics
4.1. AE Count Characteristics
4.1. AEInCount Characteristics
Section 3, we investigated the mechanical behavior and fracture process of specimens containing
In Section
nonpersistent 3, we investigated
fissures the mechanical
at various the
fissure behavior
inclinations. The andand fracture
failure behaviorprocess of specimens
of specimens is the
In Section 3, we investigated mechanical behavior fracture process of specimens
containing nonpersistent fissures at various fissure inclinations. The failure behavior of specimens is
containing nonpersistent fissures at various fissure inclinations. The failure behavior of specimens is
the macroscopic representation of microscopic deformation and fracture, and AE events have
the macroscopic representation of microscopic deformation and fracture, and AE events have
Energies 2019, 12, 79 19 of 24

macroscopic representation of microscopic deformation and fracture, and AE events have proven to be
closely associated with the microcracking process of the brittle material under compression [46].
Based on the general characteristics of AE data collected in experiments, the AE activity could
be divided into three periods: A quiet period, slow rise period, and active period. A specimen of
α = 45◦ is shown in Figure 22 as an example. In the stage of initial compaction, only the unevenness
30 5 3
of the top and bottom of the Axial samplestress
was gradually compacted, and the micropores or microdefects
in the sample were closed under AE counts
the load. Thus there were few AE events. As the load continued to
b

Accumulated AE counts (×1000)

25 Accumulated AE counts
increase, the strain concentration zone started to appear around the fissures, 4but there was no initiation
of a new crack, so the number of AE events remained low. Due to the stress concentration around
Axial stress (MPa)

AE counts (×100)
the fissures, some 20 new microcracks were initiated when the stress exceeded material 2 strength and
3
there was an obvious AE event. Then, when the AE activity entered the slow rise period, the AE
event became more15active, with the propagation of microcracks under load, and the accumulated AE
counts gradually increased as time increased. The transition from the c quiet 2 period to the slow rise
10
period of the AE events indicated a progressive accumulation of damage in the specimen. 1 Of note, the
accumulated AE counts of specimens may have experienced several steep slopes during this period.
1
As shown in Figure 519 of the specimen with α = 75◦ , the first steep slope corresponded to the beginning
a
of tension crack initiation from fissures 1 ;9 and the second slope was caused by the shear crack
coalescence along the 0 direction of fissures 1
2 3 and 7 8 .
9 As the stress 0continued 0 to load, the AE
0 100 200 300 400 500 600
events became more active, and the accumulation of AE counts increased sharply at the peak stress.700
Time (s)
The dramatic increase in the accumulated AE events was mainly caused by a large number of local
fractures inFig.20 Characteristics
the specimen and of stress and acoustic
forecasted emissions during
the development of athe loading process
macroscopic (α=90°)
crack. The appearance of
[％]
penetrating cracks in the specimen caused serious damage to the bearing structure, and the bearing
4.610
capacity of the specimen after the peak strength was mainly realized by 4.076
the resistance of uneven
surfaces on penetrating cracks. In this period, cracks continued to propagate, 3.543and the accumulated AE
counts were much greater⑦than ④ ①
in other periods. 3.009
The progressive increase ⑧⑤ ② process of AE events also revealed that the2.476 failure process of rock-like
specimens containing nonpersistent
⑨⑥ ③ fissures was a cumulative process. 1.942 Unlike the intact specimens
or specimens containing one single fissure, the AE events may have experienced 1.409 several intense
0.875
activities before peak stress. This was mainly due to the fact that a greater number of fissures provided
additional opportunities for coalescence: Every crack initiation and coalescence 0.342 resulted in energy
(a)σ1=0MPa, (b)σ1=28.96MPa, (c)σ1=4.99MPa, -0.192
release that was accompanied ε1=0×10 -3 by AE ε =3.56×
events.
1 10 -3
However, ε1=3.66×10
local stress
-3 adjustment had limited effects
on the overall load-bearing
Fig.21 Evolution of strainstructure of specimen
fields for the the specimens. Therefore,
with α=90° the to
corresponding characteristics of19
point (a-c) in Fig. AE events of
specimens containing nonpersistent fissures was more complex.

12 40 60
Axial stress
AE counts
Accumulated AE counts (×1000)

10 Accumulated AE counts 50
30
Axial stress (MPa)

AE counts (×100)

8 40

6 20 30
Stress
fluctuation
Active
4 Crack Initiation 20
period
10
2 10
Slow rise period
Quiet period
0 0 0
0.0 0.5 1.0 1.5 2.0
Axial strain(10-3)
Fig.22
Figure AE event
22. AE event characteristics
characteristicsduring
duringloading process process
the loading (α=45°) (α = 45◦ ).
Energies 2019, 12, 79 20 of 24

4.2. Evolution Process of the Strain Field

In this study, DSCM was proven to be a quick and effective tool to reveal clear and intuitive
strain accumulation processes and fracture behaviors of the specimen surface. The process analysis
of a strain field showed that fissure inclination affected the characteristics of stress distribution in
specimens and thus affected the fracture process. In the early loading period, the specimen was in
the uniform compression process, and the strain field was distributed uniformly on the specimen
surface. Along with the stress increase, the strain concentration zone gradually appeared around the
pre-existing fissures. The strain concentration zone of a specimen appeared first in the fissure tip,
except for the specimen of α = 0◦ , for which it appeared in the center of the fissure.
Initial cracks were primarily propagated from the fissure tips, especially the fissure tips near
the ends of the specimen. Although the strain value in the initial strain localization stage was small,
the early effect of pre-existing fissures on the overall damage evolution of the specimens could be
predicted. With loading increased, the strain concentration zone propagated and coalesced with other
strain concentration zones in the rock bridge area. The coalescence of high strain concentration zones
will always lead to a stress drop and obvious AE counts. The failure process showed that the main
reason for the overall failure was due to the coalescence of new cracks between fissures, resulting
in the failure of the core bearing structure, except for specimens with α = 90◦ , which presented with
intact material failure. This was because the pre-existing fissures in this experiment were mainly
concentrated in the center of the specimen. After the crack coalescence in the core bearing structure,
the cracks experienced rapid coalescence with other cracks and propagation toward the edge of the
specimen, which eventually led to the final failure.
Figure 23 gives the horizontal and vertical displacement nephogram of specimens with α = 30◦ ,
60◦ , and 75◦ at the specified mark points during the experiment. Overall, the horizontal and vertical
displacement had antisymmetric distribution. The horizontal displacement was positive for right
displacement and negative for left displacement: As the upper end was the loading end and the lower
end was the fixed end, the vertical displacement was mainly focused on the top of the specimens.
As seen from the displacement nephogram, there were some dividing lines between the adjacent
considerably different displacement zones. These dividing lines reflected the paths of crack initiation,
propagation, and coalescence. Note that the dividing lines were mainly concentrated along the fissures,
especially in the fissure tips, which also illustrates the significant effect of pre-existing fissures on
the fracture evolution process. Compared to conventional studies that have described the fracture
process mainly from statistical photographs on a macro scale, this study provided full-field strain
analysis during the overall process of failure. This work indicates that monitoring of the failure process
of specimens using the DSCM technique is an effective tool to investigate fracturing processes and
fracture mechanisms in jointed rock.
 Energies 2019, 12, 79 21 of 24
Energies 2019, 12, x FOR PEER REVIEW 21 of 24

Figure 23. Horizontal and vertical displacement nephogram of specimens with α = 30◦ , 60◦ , and 75◦ at
Figure 23. Horizontal and vertical displacement nephogram of specimens with α = 30°, 60°, and 75°
specified mark points.
at specified mark points.
5. Conclusions
5. Conclusions
Research on the mechanical properties and fracture behavior of specimens containing
Research onfissures
nonpersistent the mechanical
helps in better properties and fracture
understanding the failure behavior
mechanisms of specimens
of jointedcontaining
rock. In this
nonpersistent
study, rock-likefissures helps in
specimens better understanding
containing nine pre-existing the nonpersistent
failure mechanisms fissuresofwere
jointed
used rock. In this
to investigate
study, rock-like specimens
the mechanical properties and containing nine pre-existing
failure behavior of jointed rock nonpersistent
by a uniaxial fissures were experiment.
compression used to
investigate
The DSCM the technique
mechanicaland properties and failure
AE monitoring behavior
system wereofadopted
jointed rock in the byexperiments
a uniaxial compression
to reveal the
experiment. The DSCM
real-time strain technique
field and and AE monitoring
AE characteristics system loading.
during uniaxial were adopted in the experiments
The following conclusionstocan
reveal the real-time strain field and AE characteristics during uniaxial loading. The following
be drawn.
conclusions can be drawn.
1. Fissure inclination has a significant impact on the mechanical behavior and crack coalescence
1. Fissure inclination
patterns has a significant
of specimens containing impact on the mechanical
nonpersistent fissures. behavior
The peakand crack coalescence
strength of specimens
patterns of specimens containing nonpersistent fissures.
containing nonpersistent fissures initially decreased and then increased as fissure The peak strength of specimens
inclination
containing
increased nonpersistent
in the rangefissures ◦
from 0 initially ◦
to 90 , withdecreased and then
a minimum ◦
increased
at 30 as fissure
. The crack inclination
initiation stress of
increased
specimens in the range
first from 0°
decreased andto then
90°, with a minimum
increased as fissure at inclination
30°. The crack initiation
increased from 0◦ toof75◦ .
stress
specimens
Unlike the firstpeak decreased andthe
strength, then increased
minimum as fissure
crack initiation inclination
stress was increased
at α = 45 ◦ . The
from 0° totrend
75°. of
Unlike themodulus
elastic peak strength, the minimum
was basically the same as crack initiation
the peak strength,stress was decreased
which at α = 45°. The0◦trend
from to 30◦ofand
elastic
thenmodulus
increased. was basically the same as the peak strength, which decreased from 0° to 30°
2. andCrack
then increased.
coalescence between fissures plays an important role in the final failure modes. As fissure
2. Crack coalescence
inclination between
increased, crackfissures
coalescenceplaysmodes
an important role inchanged
between fissures the finalfromfailure
tensilemodes. As
coalescence
fissure inclination increased, crack coalescence modes between
mode to mixed tension-shear coalescence mode, then to shear coalescence mode. Five basic fissures changed from tensile
coalescence
failure modesmodewere to mixed tension-shear
classified coalescence mode,
from the experimental results:then to shear
1 tensile coalescence
failure across the mode.
fissure
Fiveplanes;
basic failure
modes were classified from
2 rotation failure of newly generated blocks; the experimental results: ①
3 mix failure mode; tensile failure across
4 shear failure; 5
thesplitting failure.② rotation failure of newly generated blocks; ③ mix failure mode; ④ shear
fissure planes;
The ⑤
3. failure; splitting
clear failure. damage evolution and fracturing process of specimens with five
and intuitive
3. Thefailure
clear and intuitive
modes during damage evolution and
the experiment werefracturing
observedprocessusing the of specimens
DSCM. The with five failure
damage process
modes during the
of specimens experiment
containing were observed
nonpersistent fissuresusing
showed theobvious
DSCM.progressive
The damage strain process of
localization
specimens containing nonpersistent fissures showed obvious progressive
failure characteristics. In the initial loading stage, the strain concentration zone was concentrated strain localization
failure
around characteristics.
the pre-existing In fissures:
the initial loading
Although stage, value
the strain the strain
was small, concentration zone was
the early influence of the
concentrated
pre-existing around the
fissures onpre-existing fissures: process
the overall damage Although the strain value
of specimens could was small, theThen,
be predicted. earlythe
influence
primaryofstrainthe pre-existing
concentration fissures on the overall
zone developed into a damage
high strain process of specimens
localization could bethe
band, indicating
predicted. Then, the primary strain concentration zone developed into a high strain
localization band, indicating the developmental path of macrocracks. The saltation of strain
Energies 2019, 12, 79 22 of 24

developmental path of macrocracks. The saltation of strain fields was usually accompanied by a
fluctuation of the stress curve and obvious AE events.

Author Contributions: Methodology—writing—original draft preparation, M.C.; conceptualization—writing

and review, S.Y.; review and editing, R.P.G.; editing, W.Y. and Q.Z.; Experiment, P.Y. and Y.Z.
Funding: This research was funded by the Independent Innovation Project for Double First-level Construction
(China University of Mining and Technology), grant number 2018ZZCX04).
Acknowledgments: The authors would like to express their sincere gratitude to the editor and two anonymous
reviewers for their valuable comments, which have greatly improved this paper. The first author also thanks the
Chinese Scholarship Council for its financial support.
Conflicts of Interest: The authors declare no conflict of interest.

Nomenclature
,
1 ,
2 ,
3 ,
4 ,
5 ,
6 ,
7 ,
8
9 numbers of fissures in the specimen
AE acoustic emission
DIC digital image correlation
DSCM digital speckle correlation method
Ea reloading average modulus
Es elastic modulus, the slope of approximate linear part in the stress–strain curve
F far-field crack
Lc lateral crack
ROI a region of interest
Sa anti-shear crack
Sm main shear crack
Sp surface spalling
Ss secondary shear crack
Ta anti-tensile wing crack
Ts secondary tensile crack
Tw tensile wing crack
λ the ratio of crack initiation stress to the compressive strength
σci the crack initiation stress
σp the peak strength

References
1. Yang, S.Q.; Tian, W.L.; Ranjith, P.G. Experimental Investigation on Deformation Failure Characteristics of
Crystalline Marble Under Triaxial Cyclic Loading. Rock Mech. Rock Eng. 2017, 50, 2871–2889. [CrossRef]
2. Yang, S.Q.; Chen, M.; Fang, G.; Wang, Y.C.; Meng, B.; Li, Y.H.; Jing, H.W. Physical experiment and numerical
modelling of tunnel excavation in slanted upper-soft and lower-hard strata. Tunn. Undergr. Space Technol.
2018, 82, 248–264. [CrossRef]
3. Lajtai, E.Z.; Lajtai, V.N. The evolution of brittle fracture in rocks. J. Geol. Soc. 1974, 130, 1–16. [CrossRef]
4. Yang, S.Q.; Chen, M.; Jing, H.W.; Chen, K.F.; Meng, B. A case study on large deformation failure mechanism
of deep soft rock roadway in Xin’An coal mine, China. Eng. Geol. 2017, 217, 89–101. [CrossRef]
5. Cotterell, B. Brittle Fracture in Compression. Int. J. Fract. Mech. 1972, 8, 195–208. [CrossRef]
6. Horii, H.; Nemat-Nasser, S. Compression-induced microcrack growth in brittle solids: Axial splitting and
shear failure. J. Geophys. Res. 1985, 90, 3105–3125. [CrossRef]
7. Wong, R.; Chau, K.; Tang, C.; Lin, P. Analysis of crack coalescence in rock-like materials containing three
flaws—Part I: Experimental approach. Int. J. Rock Mech. Min. Sci. 2001, 38, 909–924. [CrossRef]
8. Lee, H.; Jeon, S. An experimental and numerical study of fracture coalescence in pre-cracked specimens
under uniaxial compression. Int. J. Solids Struct. 2011, 48, 979–999. [CrossRef]
9. Bobet, A. The initiation of secondary cracks in compression. Eng. Fract. Mech. 2000, 66, 187–219. [CrossRef]
10. Esterhuizen, G.S.; Dolinar, D.R.; Ellenberger, J.L. Pillar strength in underground stone mines in the United
States. Int. J. Rock Mech. Min. Sci. 2011, 48, 42–50. [CrossRef]
Energies 2019, 12, 79 23 of 24

11. Brideau, M.-A.; Yan, M.; Stead, D. The role of tectonic damage and brittle rock fracture in the development
of large rock slope failures. Geomorphology 2009, 103, 30–49. [CrossRef]
12. Hearn, G.J. C4 Rock slope stabilization. Geol. Soc. Lond. Eng. Geol. Spec. Publ. 2011, 24, 189–208. [CrossRef]
13. Bobet, A.; Einstein, H.H. Fracture coalescence in rock-type materials under uniaxial and biaxial compression.
Int. J. Rock Mech. Min. Sci. 1998, 35, 26. [CrossRef]
14. Wong, L.N.Y.; Einstein, H.H. Crack Coalescence in Molded Gypsum and Carrara Marble: Part 1. Macroscopic
Observations and Interpretation. Rock Mech. Rock Eng. 2009, 42, 475–511. [CrossRef]
15. Yang, S.Q.; Liu, X.R.; Jing, H.W. Experimental investigation on fracture coalescence behavior of red sandstone
containing two unparallel fissures under uniaxial compression. Int. J. Rock Mech. Min. Sci. 2013, 63, 82–92.
[CrossRef]
16. Yang, S.Q.; Yang, D.S.; Jing, H.W.; Li, Y.H.; Wang, S.Y. An Experimental Study of the Fracture Coalescence
Behaviour of Brittle Sandstone Specimens Containing Three Fissures. Rock Mech. Rock Eng. 2012, 45, 563–582.
[CrossRef]
17. Yang, S.Q.; Tian, W.L.; Huang, Y.H. Failure mechanical behavior of pre-holed granite specimens after elevated
temperature treatment by particle flow code. Geothermics 2018, 72, 124–137. [CrossRef]
18. Yang, S.Q.; Huang, Y.H.; Ranjith, P.G. Failure mechanical and acoustic behavior of brine saturated-sandstone
containing two pre-existing flaws under different confining pressures. Eng. Fract. Mech. 2018, 193, 108–121.
[CrossRef]
19. Zhou, X.P.; Bian, J.; Qian, Q.H. Numerical simulation of crack growth and coalescence in rock-like materials
containing multiple pre-existing flaws. Rock Mech. Rock Eng. 2015, 48, 1097–1114. [CrossRef]
20. Zhou, X.P.; Zhang, Y.X.; Ha, Q.L.; Zhu, K.S. Micromechanical modeling of the complete stress-strain
relationship for crack weakened rock subjected to compressive loading. Rock Mech. Rock Eng. 2008, 41,
747–769. [CrossRef]
21. Zhou, X.P.; Cheng, H.; Feng, Y.F. An Experimental Study of Crack Coalescence Behaviour in Rock-Like
Materials Containing Multiple Flaws Under Uniaxial Compression. Rock Mech. Rock Eng. 2014, 47, 1961–1986.
[CrossRef]
22. Sagong, M.; Bobet, A. Coalescence of multiple flaws in a rock-model material in uniaxial compression. Int. J.
Rock Mech. Min. Sci. 2002, 39, 229–241. [CrossRef]
23. Prudencio, M.; Van Sint Jan, M. Strength and failure modes of rock mass models with non-persistent joints.
Int. J. Rock Mech. Min. Sci. 2007, 44, 890–902. [CrossRef]
24. Bahaaddini, M.; Sharrock, G.; Hebblewhite, B.K. Numerical investigation of the effect of joint geometrical
parameters on the mechanical properties of a non-persistent jointed rock mass under uniaxial compression.
Comput. Geotech. 2013, 49, 206–225. [CrossRef]
25. Chen, X.; Liao, Z.; Peng, X. Deformability characteristics of jointed rock masses under uniaxial compression.
Int. J. Min. Sci. Technol. 2012, 22, 213–221. [CrossRef]
26. Fan, X.; Kulatilake, P.H.S.W.; Chen, X. Mechanical behavior of rock-like jointed blocks with
multi-non-persistent joints under uniaxial loading: A particle mechanics approach. Eng. Geol. 2015,
190, 17–32. [CrossRef]
27. Gui, Y.L.; Zhao, Z.Y.; Zhang, C.; Ma, S.Q. Numerical investigation of the opening effect on the
mechanical behaviours in rocks under uniaxial loading using hybrid continuum-discrete element method.
Comput. Geotech. 2017, 90, 55–72. [CrossRef]
28. Yang, X.; Kulatilake, P.H.S.W.; Jing, H.; Yang, S. Numerical simulation of a jointed rock block mechanical
behavior adjacent to an underground excavation and comparison with physical model test results.
Tunn. Undergr. Space Technol. 2015, 50, 129–142. [CrossRef]
29. Cao, R.H.; Cao, P.; Lin, H.; Pu, C.Z.; Ou, K. Mechanical behavior of brittle rock-like specimens with
pre-existing fissures under uniaxial loading: Experimental studies and particle mechanics approach.
Rock Mech. Rock Eng. 2016, 49, 763–783. [CrossRef]
30. Yang, X.X.; Jing, H.W.; Tang, C.A.; Yang, S.Q. Effect of parallel joint interaction on mechanical behavior of
jointed rock mass models. Int. J. Rock Mech. Min. Sci. 2017, 92, 40–53. [CrossRef]
31. Cheng, Y.; Wong, L.N.Y. Microscopic characterization of tensile and shear fracturing in progressive failure in
marble. J. Geophys. Res. Solid Earth 2018, 123, 204–225. [CrossRef]
32. Yates, J.R.; Zanganeh, M.; Tai, Y.H. Quantifying crack tip displacement fields with DIC. Eng. Fract. Mech.
2010, 77, 2063–2076. [CrossRef]
Energies 2019, 12, 79 24 of 24

33. Abanto-Bueno, J.; Lambros, J. Investigation of crack growth in functionally graded materials using digital
image correlation. Eng. Fract. Mech. 2002, 69, 1695–1711. [CrossRef]
34. Hedan, S.; Fauchille, A.-L.; Valle, V.; Cabrera, J.; Cosenza, P. One-year monitoring of desiccation cracks in
Tournemire argillite using digital image correlation. Int. J. Rock Mech. Min. Sci. 2014, 68, 22–35. [CrossRef]
35. Yin, P.F.; Yang, S.Q. Experimental investigation of the strength and failure behavior of layered sandstone
under uniaxial compression and Brazilian testing. Acta Geophys. 2018, 66, 585–605. [CrossRef]
36. Zhou, X.P.; Lian, Y.J.; Wong, L.N.Y.; Berto, F. Understanding the fracture behavior of brittle and ductile
multi-flawed rocks by uniaxial loading by digital image correlation. Eng. Fract. Mech. 2018, 199, 438–460.
[CrossRef]
37. Carroll, J.D.; Abuzaid, W.; Lambros, J.; Sehitoglu, H. High resolution digital image correlation measurements
of strain accumulation in fatigue crack growth. Int. J. Fatigue 2013, 57, 140–150. [CrossRef]
38. Zhao, C.; Matsuda, H.; Morita, C.; Shen, M.R. Study on Failure Characteristic of Rock-Like Materials with an
Open-Hole Under Uniaxial Compression. Strain 2011, 47, 405–413. [CrossRef]
39. Zhao, C.; Bao, C.; Tian, J.; Matsuda, H.; Morita, C. Experimental study of coalescence mode of cracks and
strength of rock with double flaws based on strain localization. Chin. J. Rock Mech. Eng. 2015, 34, 10.
40. Li, D.; Zhu, Q.; Zhou, Z.; Li, X.; Ranjith, P.G. Fracture analysis of marble specimens with a hole under
uniaxial compression by digital image correlation. Eng. Fract. Mech. 2017, 183, 109–124. [CrossRef]
41. Bruck, H.; McNeill, S.; Sutton, M.A.; Peters, W. Digital image correlation using Newton-Raphson method of
partial differential correction. Exp. Mech. 1989, 29, 261–267. [CrossRef]
42. Wasantha, P.; Ranjith, P.; Viete, D.; Haque, A.; Bouazza, A. Crack initiation stress for saturated sandstone
in triaxial compression. In Proceedings of the 46th US Rock Mechanics/Geomechanics Symposium,
Chicago, IL, USA, 24–27 June 2012.
43. Kong, R.; Feng, X.-T.; Zhang, X.; Yang, C. Study on crack initiation and damage stress in sandstone under
true triaxial compression. Int. J. Rock Mech. Min. Sci. 2018, 106, 117–123. [CrossRef]
44. Brace, W.; Paulding, B., Jr.; Scholz, C. Dilatancy in the fracture of crystalline rocks. J. Geophys. Res. 1966, 71,
3939–3953. [CrossRef]
45. Eberhardt, E.; Stead, D.; Stimpson, B.; Read, R. Identifying crack initiation and propagation thresholds in
brittle rock. Can. Geotech. J. 1998, 35, 222–233. [CrossRef]
46. Yang, S.Q.; Jing, H.-W. Strength failure and crack coalescence behavior of brittle sandstone samples containing
a single fissure under uniaxial compression. Int. J. Fract. 2011, 168, 227–250. [CrossRef]
47. Yang, S.Q. Crack coalescence behavior of brittle sandstone samples containing two coplanar fissures in the
process of deformation failure. Eng. Fract. Mech. 2011, 78, 3059–3081. [CrossRef]
48. Park, C.H.; Bobet, A. Crack initiation, propagation and coalescence from frictional flaws in uniaxial
compression. Eng. Fract. Mech. 2010, 77, 2727–2748. [CrossRef]

© 2018 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 (http://creativecommons.org/licenses/by/4.0/).