materials
Article
Research on the Impact Mechanical Properties of Real-Time
High-Temperature Granite and a Coupled Thermal–Mechanical
Constitutive Model
Yubai Li 1 , Yue Zhai 1,2, *, Yifan Xie 1 and Fandong Meng 1
Citation: Li, Y.; Zhai, Y.; Xie, Y.;
Meng, F. Research on the Impact
Mechanical Properties of Real-Time
High-Temperature Granite and a
Coupled Thermal–Mechanical
Constitutive Model. Materials 2023,
16, 2773. https://doi.org/10.3390/
ma16072773
Academic Editors: Carlos
Garcia-Mateo and Changho Lee
Received: 24 February 2023
Revised: 22 March 2023
Accepted: 27 March 2023
Published: 30 March 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Materials 2023, 16, 2773. https://doi.org/10.3390/ma16072773
1 School of Geological Engineering and Geomatics, Chang’an University, Xi’an 710064, China;
liyubai@chd.edu.cn (Y.L.)
2 Key Laboratory of Mine Geological Hazard Mechanism and Control, Shaanxi Institute of Geological Survey,
Xi’an 710054, China
* Correspondence: zy@chd.edu.cn
Abstract: Studying the mechanical behavior of rocks under real-time high-temperature conditions
is of great significance for the development of energy caverns, nuclear waste disposal projects, and
tunneling engineering. In this study, a real-time high-temperature impact compression test was
conducted on Sejila Mountain granite to explore the effects of temperature and external load on its
mechanical properties. Based on the concepts of damage mechanics and statistics, a coupled thermal–
mechanical (T-M) damage constitutive model was established, which considers the temperature effect
and uses the double-shear unified strength as the yield criterion. The parameter expressions were
clarified, and the accuracy and applicability of the model were verified by experimental data. The
research results indicated that high temperatures had an obvious damaging and deteriorating effect
on the strength of the granite, while an increase in impact velocity had an enhancing effect on the
strength of the granite. The established constitutive model theoretical curve and test curve showed
a high degree of agreement, indicating that the coupled T-M model can objectively represent the
evolution process of damage in rocks and the physical meaning of its parameters is clear.
Keywords: granite; thermal–mechanical coupling; SHPB; unified strength theory; constitutive model
1. Introduction
The development and utilization of geothermal resources, the deep burial disposal of
nuclear waste, and tunneling engineering involve real-time high-temperature rocks. Com-
pared with room-temperature conditions, the stress status of rocks in a high-temperature
environment usually changes, which affects the stability of a rock mass [1,2]. In addition,
underground rock engineering often involves explosion-proof structural engineering, exca-
vation and blasting, etc., and it is necessary to predict and control the damage degree of
rocks under impact load. Therefore, it is of great theoretical significance and practical value
to study the law governing the impact mechanical properties of rocks under real-time high
temperatures for underground rock engineering.
The physical and mechanical properties of rocks are greatly affected by temperature
and impact load [3,4]. With the increase in the strain rate, the energy absorption rate of
a rock increases, the damage degree increases, and the peak stress is affected differently
by the rock particle size [5–7]. Kumari et al. studied the effect of temperature on the
mechanical behavior of rock specimens through triaxial tests under high temperature and
high pressure and found that the type of rock instability is affected by the temperature and
mainly characterized by a brittle–plastic transformation of the rock [8]. Because the rock will
acquire brittleness after being subjected to a high temperature, the mechanical properties
of the rock after cooling, following a high-temperature treatment, are significantly different
from those during the high-temperature treatment. At present, real-time high-temperature
https://www.mdpi.com/journal/materials
Materials 2023, 16, 2773 2 of 15

impact mechanical testing is mainly carried out by three test methods. The first method
consists in heating the specimen and part of the pressure bar simultaneously [9]. Although
the change of the wave impedance of steel elastic pressure bars caused by high temperature
is corrected during data processing, the experimental error caused by the effects of the
high temperature on the yield strength and elastic modulus of the bar is still unavoidable.
The second method consists in placing temperature-insensitive spacers at both ends of
the specimen [10]. Only the spacer block is heated during the test. This method can
effectively improve the drawbacks of the first method but it has higher requirements
for the material of the spacer block, the test equipment is more complicated, and the
operation is cumbersome. The third test method consists in heating only the specimens
in a furnace [11]. The specimens reaching the prefabrication temperature are transferred
to the split Hopkinson Pressure Bar (SHPB) device, and the dynamic mechanical test is
completed. This method must be completed in a short time to avoid heat transfer after the
bar comes in contact with the specimen. If the heat loss is too large during the transfer
process, temperature compensation should be considered. Due to the limitation of the test
technology, the results of real-time high-temperature impact mechanical tests have large
errors, and the analysis and understanding of the impact mechanism of rock damage are
controversial [12–14].
Various scholars have conducted a deep study on the failure mechanism of thermally
damaged rocks by applying new mechanical theories, in order to define an appropriate
method to describe the stress–strain relationship in response to rock damage [15]. Since
Dougill [16] introduced the concept of damage mechanics into the field of rock mechanics,
damage mechanics’ principles have become important for studying rock mass damage.
For example, Cao et al. [17] divided the rock into two parts, damaged and undamaged,
and proposed a damage constitutive equation considering rock hardening and softening.
Mubarak et al. [18] analyzed the damage constitutive model of a rock under cyclic loading
conditions. Deng et al. [19] proposed a damage constitutive model based on continuous
damage mechanics and the maximum entropy distribution of micro-element strength.
However, the above models cannot accurately describe the mechanical behavior of rocks at
high temperature. Therefore, it is necessary to build a constitutive model of rocks subjected
to high temperature considering the effects of temperature. Liu and Xu [20] carried out a
study on the change of the main mechanical parameters of granite (20~600 ◦ C), obtained
the fitting equation of uniaxial compressive strength and the Poisson’s ratio of granite
depending on the temperature, and used them to define thermal damage. Rong et al. [21]
defined thermal damage from the perspective of acoustic emission and established a
damage coupling equation considering the crack closure effect. Zhang et al. [22] were
the first to propose a thermal–mechanical (T-M) factor and established a one-dimensional
nonlinear coupling model. At present, constitutive models considering the temperature
effects are mainly divided into three types. The first type is the Thermo–Elasto–Plastic
damage model based on the damage theory and Thermo-Elasticity theory [23,24]. The
second type is a rheological model based on creep tests and the Thermo-Viscoplastic
theory [25,26]. The third type is the thermal damage constitutive model of rocks based
on statistical distribution [27,28]. The statistical damage constitutive model can directly
and accurately describe the evolution process of rock damage, so as to better describe the
mechanism of rock damage. Therefore, this model has obvious advantages over the other
two models. Due to the randomness in the evolution of the internal defect distribution
in a rock, it is generally considered that it satisfies the Weibull distribution [29] or the
normal distribution [30], and the damage discriminating variable is usually expressed by
the strength yield criterion. The strength theories include the maximum normal stress
strength theory, the maximum shear stress theory, the maximum normal strain theory, the
octahedral shear stress theory [31], the MC strength theory [32], the DP strength theory [33],
the Griffith and modified Griffith theory [34], the HB empirical strength theory [35], etc.
In addition, Xie et al. [36] used the method of continuous damage mechanics to study the
thermal damage of rocks when the temperature changed and established a rock thermal
 Materials 2023, 16, x FOR PEER REVIEW 3 of 16

[33], the Griffith and modified Griffith theory [34], the HB empirical strength theory [35],
Materials 2023, 16, 2773 etc. In addition, Xie et al. [36] used the method of continuous damage mechanics to 3study of 15
the thermal damage of rocks when the temperature changed and established a rock ther-
mal damage failure criterion based on the energy theory. Yu and He [37] put forward the
double-shear unified strength theory of materials on the basis of the double-shear stress
damage failure criterion based on the energy theory. Yu and He [37] put forward the
idea, which has strong applicability and has been applied in some practical projects [38–
double-shear unified strength theory of materials on the basis of the double-shear stress
40].
idea, which has strong applicability and has been applied in some practical projects [38–40].
To sum up, the existing studies mostly focused on the mechanical properties of rocks
To sum up, the existing studies mostly focused on the mechanical properties of rocks
under the action of a single field, and research on the mechanical properties of rocks under
under the action of a single field, and research on the mechanical properties of rocks under
T-Mcoupling
T-M coupling isis less
less developed.
developed. How Howtotoreduce
reducethetheexperimental
experimental error
errorand make
and make thethe
ex-
perimental results more convincing, and how to rationally analyze
experimental results more convincing, and how to rationally analyze the experimental the experimental re-
sults through
results through theory
theoryareare
stillstill
in the exploratory
in the exploratorystage. Therefore,
stage. in thisinstudy,
Therefore, Sejila Moun-
this study, Sejila
tain granite was used as the test material, and an impact compression
Mountain granite was used as the test material, and an impact compression test test under real-time
under
high temperature
real-time was carried
high temperature wasout through
carried outthe improved
through SHPB system.
the improved SHPBThe changes
system. The in
granite mechanical properties at different temperatures and impact
changes in granite mechanical properties at different temperatures and impact velocities velocities are dis-
cussed,
are and a coupled
discussed, T-M constitutive
and a coupled model ofmodel
T-M constitutive graniteofbased onbased
granite the double-shear
on the double-uni-
fied strength
shear unified yield criterion
strength yieldiscriterion
established to provide theoretical
is established to providesupport for high-temper-
theoretical support for
ature rock engineering-related calculations and numerical simulations.
high-temperature rock engineering-related calculations and numerical simulations.

2.2.Materials
Materialsand
andMethods
Methods
2.1. Specimen Preparation
2.1. Specimen Preparation
The
The granite
granite specimens
specimens were
were collected
collected from
from the
the tunnel
tunnel engineering
engineering sitesite near
near Sejila
Sejila
Mountain,
Mountain, China. The surface was gray-white with black particles, the particlesdense,
China. The surface was gray-white with black particles, the particles were were
and theand
dense, specimens were well
the specimens were cemented and free
well cemented andof free
cracks. According
of cracks. to International
According to Interna-
Society for Rockfor
tional Society Mechanics (ISRM), an
Rock Mechanics impactancompression
(ISRM), test specimen
impact compression test is processed
specimen is into
pro-
acessed
cylinder of Φ48 mm × 25 mm, and a static compression test specimen
into a cylinder of Φ48 mm × 25 mm, and a static compression test specimen is pro-is processed to
obtain a sample of Φ48 mm × 100 mm. A double-sided grinding machine
cessed to obtain a sample of Φ48 mm × 100 mm. A double-sided grinding machine was was used for
fine
usedprocessing to ensure to
for fine processing that the parallelism
ensure error was error
that the parallelism less than
was 0.05
less mm,
than and0.05 the
mm,error
and
of the height and diameter of the specimen was not more than 0.3 mm.
the error of the height and diameter of the specimen was not more than 0.3 mm. The max- The maximum
deviation of the end
imum deviation face
of the perpendicular
end to the axis
face perpendicular of axis
to the the specimen did notdid
of the specimen exceed 0.25◦
not exceed
(Figure 1) [41].1) [41].
0.25° (Figure

Figure1.1.Granite
Figure Granitespecimens.
specimens.

2.2.
2.2.Test
TestEquipment
EquipmentandandTestTestMethods
Methods
The
The test adopted the SHPB system
test adopted the SHPB system with
with aa pressure
pressure barbardiameter
diameter of of 50
50mmmmand andanan
intelligent real-time high-temperature furnace. The schematic diagram is
intelligent real-time high-temperature furnace. The schematic diagram is shown in Figureshown in Figure 2.
The elastic modulus of the pressure bars was 210210
GPa, thethe
density was 7850 kg/m 3
2. The elastic modulus of the pressure bars was GPa, density was 7850 kg/m,3,thethe
propagation velocity of the stress wave was 5170 m/s, the length of the
propagation velocity of the stress wave was 5170 m/s, the length of the impact bar was 400 impact bar was
400
mm, mm,
the the length
length of the
of the incident
incident barbarwaswas44004400
mm, mm,
andand
thethe length
length of the
of the transmission
transmission bar
bar was 3000 mm. In this
was 3000 mm. In ◦this experiment, experiment,
five five
workingworking conditions
conditions were were
chosen, chosen,
i.e., roomi.e.,temper-
room
temperature, ◦ C, 600 ◦ C, and 800 ◦ C. The heating rate was 10 ◦ C/min during
ature, 200 °C,200400 C,
°C,400
600 °C, and 800 °C. The heating rate was 10 °C/min during the heat-
the
ingheating
process.process. After reaching
After reaching the targetthe target temperature,
temperature, the temperature
the temperature was kept
was kept constant for
constant for 2 h to ensure that specimens were heated evenly. After the specimens were
heated, the control system pushed the incident bar and the transmission bar to both ends
of the specimen synchronously, and the impact test was completed in the real-time high-
temperature furnace. This method effectively avoids heating the pressure bar and thus
Materials 2023, 16, x FOR PEER REVIEW 4 of 16

2 h to ensure that specimens were heated evenly. After the specimens were heated, the
Materials 2023, 16, 2773 control system pushed the incident bar and the transmission bar to both ends of the spec- 4 of 15
imen synchronously, and the impact test was completed in the real-time high-temperature
furnace. This method effectively avoids heating the pressure bar and thus affecting its
mechanical
affecting itsparameters
mechanicaland also avoids
parameters andthe temperature
also avoids theloss caused byloss
temperature the transfer
caused byof the
the
specimen. Thespecimen.
transfer of the control pressure in this
The control test was
pressure 0.2test
in this MPa,was 0.4
0.2MPa,
MPa,and 0.6 MPa,
0.4 MPa, andMPa,
and 0.6 the
corresponding initial impact
and the corresponding initial velocities were 5.4
impact velocities m/s,5.4
were 8.8m/s,
m/s,8.8and
m/s,11.3and
m/s. Three
11.3 m/s.sets
Threeof
specimens were prepared
sets of specimens for allfor
were prepared working conditions,
all working and the
conditions, andmechanical
the mechanicalteststests
werewerere-
repeated
peated three
three times.
times. TheThe median
median of of
thethe test
test data
data was
was selected
selected forfor
thethe analysis.
analysis.

(a)

(b)
Figure
Figure 2.
2. SHPB
SHPB test
test system
system with an intelligent
with an intelligent real-time
real-time high-temperature
high-temperature furnace.
furnace. (a)
(a) Abridged
Abridged
general view, (b) real-time high-temperature furnace.
general view, (b) real-time high-temperature furnace.

Based
Based on
on the
theone-dimensional
one-dimensional stress
stress wave
waveassumption
assumptionand andthethehomogeneity
homogeneityassump-
assump-
tion, the three-wave method
tion, the three-wave method [42,43] was
was used to calculate the raw data:
 . 𝜇1 −µ 𝜇−2µ 𝐶0 C
𝜀̇ 
= ε−= − 1 ls = 2
= (−𝜀 0
(−ε+ + 𝜀 ε+r +𝜀𝑡 )ε t )
 𝑙𝑠 C 𝑙𝑠 l s 𝑖 i 𝑟
ε 𝐶= ls0 (−ε i + ε r + ε t )dτ (1)
 𝜀 = 0 (−𝜀EA

+ 𝜀 + 𝜀 )𝑑𝜏 (1)
𝑙 = 2A𝑖 s (ε i𝑟+ ε r 𝑡+ ε t )
σ
𝑠
. 𝐸𝐴
where ε is the strain rate, ε i is the 𝜎 =incident
(𝜀 +wave
𝜀𝑟 + 𝜀strain,
𝑡) ε r is the reflected wave strain,
{ 2𝐴𝑠 𝑖
ε t is the transmitted wave strain, C0 is the one-dimensional elastic wave velocity in the
where
pressure 𝜀̇ is thels strain
bar, rate, 𝜀𝑖ofisthe
is the length theimpact
incident wave
bar, E is strain, 𝜀𝑟 ismodulus,
the elastic the reflected wave
As and strain,
A are the
𝜀cross-sectional
𝑡 is the transmitted wave strain, 𝐶 is the one-dimensional elastic
area of the specimen0and the pressure bar, respectively. wave velocity in the
pressure bar, 𝑙𝑠 is the length of the impact bar, 𝐸 is the elastic modulus, 𝐴𝑠 and
𝐴2.3.
areResults
the cross-sectional area of the specimen and the pressure bar, respectively.
The stress–strain curves of the granite specimens at different temperatures and impact
velocities were determined, as shown in Figure 3. At 800 ◦ C, the granite specimen had
lost its bearing capacity. Since this occurrence did not meet the assumptions of the SHPB
test, the impact mechanical test of the specimen under this working condition was not
carried out.
 2.3. Results
The stress–strain curves of the granite specimens at different temperatures and im-
pact velocities were determined, as shown in Figure 3. At 800 °C, the granite specimen had
lost its bearing capacity. Since this occurrence did not meet the assumptions of the SHPB
Materials 2023, 16, 2773
test, the impact mechanical test of the specimen under this working condition was 5not of 15

carried out.

(a) (b)

(c) (d)
Figure 3. Stress–strain curves at different temperatures with specimens treated at 800 °C. (a) Stress–◦
Figure 3. Stress–strain curves at different temperatures with specimens treated at 800 C.
strain curves at different temperatures (5.4 m/s), (b) stress–strain curves at different temperatures
(a) Stress–strain curves at different temperatures (5.4 m/s), (b) stress–strain curves at different tem-
(8.8 m/s), (c) stress–strain curves at different temperatures (11.3 m/s), (d) granite specimens treated
at peratures
800 °C. (8.8 m/s), (c) stress–strain curves at different temperatures (11.3 m/s), (d) granite specimens
treated at 800 ◦ C.
It can be seen in Figure 3 that the test curves have no obvious compaction section,
It can be seen in Figure 3 that the test curves have no obvious compaction section,
and the slope of some curves rises sharply at the initial stage. Although molybdenum-
and the slope of some curves rises sharply at the initial stage. Although molybdenum-
based grease was applied to both sides of the specimen before the test, the data oscillation
based grease was applied to both sides of the specimen before the test, the data oscillation
phenomenon caused by the uneven contact surface would still occur at the moment when
phenomenon caused by the uneven contact surface would still occur at the moment when
the incident bar hit the specimen, which directly led to a sharp increase in the deformation
the incident bar hit the specimen, which directly led to a sharp increase in the deformation
modulus. For such phenomena, it should be considered that the dynamic deformation
modulus. For such phenomena, it should be considered that the dynamic deformation
modulus between the zero point and 50% of the peak stress was always less than or equal
modulus between the zero point and 50% of the peak stress was always less than or equal
toto
thethe
dynamic
dynamicdeformation
deformationmodulus
modulusat 50%
at 50%of the
of peak stress
the peak pointpoint
stress in theinanalysis of theof
the analysis
test data. Differently from what observed at 8.8 m/s and 11.3 m/s, the test curve
the test data. Differently from what observed at 8.8 m/s and 11.3 m/s, the test curve at 5.4 m/s
impact
at 5.4 velocity showed
m/s impact an obvious
velocity showedapproximate
an obvious plateau beforeplateau
approximate reaching the peak
before stress,
reaching the
which is because when the impact velocity is low, the shaping effect on the wave
peak stress, which is because when the impact velocity is low, the shaping effect on is far less
the
wave is far less than that of a high-velocity impact. The incident wave cannot be well
approximated to a sine wave, but the compressive strength value and strain rate of the
corresponding specimen will be improved. At the same time, it was also shown that even
if the external load reached the maximum impact resistance of the specimen, due to the
extremely short action time, there was a short adaptation stagnation period after the rock
was broken, resulting in resistance to the peak load for a short time. In addition, this
also showed that the granite underwent a certain degree of brittle–plastic transformation
Materials 2023, 16, 2773 6 of 15

at high temperature. At the same impact velocity, the peak strain at high temperature
increased compared with that at room temperature, which once again indicated that the
brittleness of the rock began to change to plasticity under the action of high temperature.
Under the same impact velocity, the peak stress at 200 ◦ C was lower than that at room
temperature, while the peak stress at 400 ◦ C was significantly higher than that at 200 ◦ C,
the peak stress at 600 ◦ C was significantly reduced, and the bearing capacity was lost
at 800 ◦ C. Before 400 ◦ C, due to the complete evaporation of the water absorbed in the
minerals inside the specimen and the release of water from the crystal lattice, the effective
stress increased. The strength of granite increased with the increase of the effective stress.
However, the loss of water from the specimen also reduced its strength to some extent.
Therefore, before reaching 400 ◦ C, the peak strength of the specimen fluctuated due to the
interaction of multiple mechanisms [44]. At 600 ◦ C, quartz undergoes a phase transition,
and the volume of the crystal structure suddenly increases, resulting in a large, unbalanced
force around the cracks, which manifests as a rapid decline in the strength of the specimen
macroscopically [45]. A temperature higher than 600 ◦ C commonly leads to mineral
intragranular displacement and intergranular dislocation [46], and the rapid development
of fractures leads to a sharp drop in the peak stress of granite and to the rapid deterioration
of its mechanical properties.
At the same temperature, compared with 5.4 m/s, the peak stress of the specimens in-
creased significantly when the impact velocity was 8.8 m/s and 11.3 m/s and as the impact
velocity increased. The influence of the initial damage caused by the high temperature on
the curve was significantly reduced, and the impact velocity replaced the influence of high
temperature on the curve, becoming dominant. It can be seen that the dynamic impact
mechanical properties of granite were the result of the coupling between high temperature
and impact velocity, and as the impact velocity increased, the impact of the initial damage
caused by the high temperature on mechanical properties such as peak stress decreased.

3. Constitutive Model
3.1. Thermal Damage Variable
When studying mechanical properties such as the deformation, strength, and damage
of rocks at high temperature, in addition to exploring their regularity from an experimental
point of view, it is important to establish a coupled T-M damage model from a theoretical
point of view. Similar to other rocks, granite is generally considered to be an aggregate
of minerals. When the temperature increases, thermal stress makes the mineral particles
inside granite expand and squeeze, causing fracture expansion and connection, resulting in
thermal damage. A large number of experiments have proved that a high temperature has
a softening effect on granite, which is shown by the decrease of the elastic modulus with the
increase in temperature [47–49]. Therefore, the elastic modulus is a function of temperature
and s can be chosen to reflect the thermal damage variable ∅T , which is defined as

∅T = 1 − ET /E0 (2)

where ET is the elastic modulus at temperature T; E0 is the initial elastic modulus of

the rock.

3.2. Mechanical Damage Variable

Granite has initial defects during its natural formation, and when it is loaded, mi-
crofractures in its internal structure continue to develop. The change of the microstructure
leads to the formation of macroscopic cracks, which continue to expand, and finally macro-
scopic damage occurs. Therefore, the formation and development of microcracks are the
root cause of the damage. A large number of studies proved that the evolution of the
defect distribution in a rock satisfies a statistical law, and the strength of micro-elements
Materials 2023, 16, 2773 7 of 15

approximately obeys the Weibull distribution [50]; the distribution density function h(w) is:

m w m −1 w m
h(w) = ( ) exp[−( ) ] (3)
w0 w0 w0
where w is the discriminant variable of micro-element damage; m and w0 are the parameters
in the Weibull distribution.
The mechanical damage variable ∅ M can be written as:
Z w
w m
∅M = h(w) dw = 1 − exp[−( ) ] (4)
0 w0

3.3. Coupled T-M Damage Constitutive Model

According to the above analysis, the total T-M damage ∅ is:

∅ = ∅T + ∅ M − ∅T ∅ M (5)

Based on Lemaitre’s principle and Hooke’s law, the constitutive relation is:
w m
σ = E0 ε(1 − ∅) = E0 ε(1 − ∅T )(1 − ∅ M ) = ET εexp[−( ) ] (6)
w0

where ε is the axial strain, σ is the axial stress of the rock.

In this study, a strength yield criterion was considered to represent w. Yu and He [37]
proposed a unified strength theory based on the double-shear yield criterion. Starting from
a unified physical model, the theory takes into account all stress components and their
different effects on material failure and can be applied to various geotechnical materials.
Both the M-C strength theory and the double-shear strength theory are special cases and
contain new calculation criteria that are more reasonable than the D-P criterion [33]. The
unified strength theory based on the double-shear stress yield criterion is expressed as:
1− a 1− b √ +a √
2√ ◦
(
3a I1 + 1+b J2 sinθ + a 3 J2 cosθ, 0 ≤ θ ≤ θb
w= 1− a a+ ab+b √ √ ◦ (7)
3a I1 + a(1+b) J2 sinθ + 2√+ a+ ab−b J2 cosθ, θb ≤ θ ≤ 60
3( a+ ab)

where I1 is the first invariant of the stress tensor, J2 is the second invariant of the stress
tensor, θ is the shear stress angle, a and b are the parameters of the corresponding function.
Among them, I1 and J2 can be expressed as:
(
I1 = σ10 + σ20 + σ30

2 2
2 (8)
J2 = 61 [ σ10 − σ20 + (σ20 − σ30 ) + σ30 − σ10 ]

and
σ1 Eε 1
σ10 = σ1 −µσ2 −µσ3
σ2 Eε 1
σ20 = σ1 −µσ2 −µσ3 (9)
σ3 Eε 1
σ30 = σ1 −µσ2 −µσ3

a and b can be expressed as:

1−sinθ
(
a= 1+sinθ
(1+ a)τ0 −σt (10)
b= σt −τ0

In the above equation, τ0 is the shear strength, σt is the tensile strength, µ is the Poisson’s
ratio, E is the elastic modulus.
By combining Equations (7)–(10), the unified strength theory expressed in the form of
principal stress can be obtained. Under uniaxial compression test conditions (σ20 = σ30 = 0),
Materials 2023, 16, 2773 8 of 15

the unified strength theory based on the double-shear stress yield criterion can be ex-
pressed as:
Eε 1 (σ1 /a − σ3 ) Eε
w= = 1 (11)
σ1 − 2µσ3 a
Substituting Equation (11) into Equation (6), the damage evolution model of the
specimen can be obtained, which is expressed as:

(1 − sinθ ) E0 ε m
σ = ET εexp[−( ) ] (12)
(1 + sinθ )w0

3.4. Constitutive Model Parameters Determination

The mechanical parameters of the high-temperature rock statistical damage model
proposed in this study include ET , E0 , and ϕ and can be measured by routine rock tests.
Therefore, choosing method to determine the distribution parameters w0 and m was the
key problem to solve the application of the statistical damage constitutive model. At
present, the methods for determining distribution parameters are mainly the inversion
analysis method [51], the linear fitting method [52], and the peak point method [53]. The
peak point method focuses on the highest point of the test curve due to its clear physical
meaning and simple processing. The influence of other test points on a rock stress–strain
curve can be weakened, and the problem that the statistical damage constitutive equation
has no mathematical meaning within a certain range can be avoided. The stress–strain
curve obtained by theoretical calculation completely depends on the properties of the
statistical damage constitutive equation itself. It requires the model to satisfy the following
governing equations:

 σ(ε) = σc
ε=ε c (13)
 δσ =0
δε σ =σc ,ε=ε c

where σc and ε c are the peak stress and peak strain of the stress–strain curve, respectively.
Combining Equations (6) and (13):

m = 1/[ln( ET ε c ) − lnσc ]

(14)
w0 = E0aε c m1/m

It can be seen from Equation (14) that the model parameters m and w0 can be obtained
from the experimental parameters E0 , σc and ε c , which shows that the mechanical parame-
ters of the high-temperature rock statistical damage constitutive model proposed in this
study are easy to obtain, and their physical meaning is clear.

4. Model Validation
In order to analyze the rationality and applicability of the above-described statistical
damage constitutive model, the test results were fitted with the model curve. In this study,
the RMT-150 C rock mechanical test system was used to conduct static compression tests
at room temperature and obtain the initial mechanical properties of the granite specimen.
The initial elastic modulus was 42.873 GPa, the Poisson’s ratio was 0.160, and the peak
stress was 186.351 MPa. At room temperature, the longitudinal wave velocity of the Sejila
Mountain granite was 2770 m/s, and the rock internal friction angle was 45◦ . According
to the above test results and the parameter determination method, the parameters of the
model built are shown in Table 1. Among them, the dynamic elastic modulus was calculated
by taking two points corresponding to 40% and 60% of the peak stress. The theoretical
curve of the rock stress–strain relationship at different temperatures could be obtained
from Equation (12), as shown in Figure 4.
Materials 2023, 16, 2773 9 of 15

Table 1. Values of the model parameters.

Temperature/◦ C ET /MPa w0 m σc εc
20 13,979.2 3.161 9.215 124.66 0.00994
200 45,206.58 1.393 0.693 101.09 0.00947
5.4 m/s
400 11,665.95 5.743 1.141 103.43 0.018
600 13,504.7 4.651 1.036 92.52 0.01799
20 17,581.72 2.818 21.762 164.40 0.00979
200 28,706.13 2.275 3.942 143.23 0.00643
8.8 m/s
400 17,289.46 3.087 11.700 158.89 0.01001
600 11,940.04 5.794 1.719 112.92 0.01692
20 28,106.52 3.952 1.948 188.91 0.01123
200 12,836.02 6.139 5.196 189.44 0.01789
Materials 2023, 16, x FOR PEER REVIEW 11.3 m/s 10 of 16
400 21,098.09 5.850 2.198 218.99 0.01636
600 14,346.63 4.542 3.210 132.80 0.01264

(a)

(b)

Figure 4. Cont.
Materials2023,
Materials 2023,16,
16, 2773
x FOR PEER REVIEW 11 10
of of
1615

(c)
Figure
Figure 4.
4. Test
Test curve
curve and
and theoretical
theoretical curve
curvefitting.
fitting.(a)
(a)Stress–strain
Stress–straincurves
curvesatatdifferent
differenttemperatures
temperatures
(5.4 m/s), (b) stress–strain curves at different temperatures (8.8 m/s), (c) stress–strain curves at dif-
(5.4 m/s), (b) stress–strain curves at different temperatures (8.8 m/s), (c) stress–strain curves at
ferent temperatures (11.3 m/s).
different temperatures (11.3 m/s).

As
As shown
shown in in Figure
Figure 4, 4, the
the theoretical
theoretical curvecurve obtained
obtainedaccording
accordingto tothe
thestatistical
statisticaldamage
dam-
age constitutive model showed the same general trend
constitutive model showed the same general trend as the experimental curve, which as the experimental curve, which
reflects
reflects the
the changes
changes of of granite
granitestrength
strengthwith withtemperature.
temperature.ItItshows showsthat thatthe thecoupled
coupled T-MT-M
damage constitutive model based on the unified strength theory
damage constitutive model based on the unified strength theory proposed in this study can proposed in this study
can reflect
reflect the deformation
the deformation characteristics
characteristics of granite
of granite underunder
different different temperatures
temperatures and
and uniaxial
uniaxial compression tests. It can also well characterize the
compression tests. It can also well characterize the stress–strain curve under different stress–strain curve under dif-
ferent temperatures. However, there is also a certain deviation,
temperatures. However, there is also a certain deviation, which is within a reasonable which is within a reason-
able
range.range.
The The reasonreason is that
is that the the viscous
viscous characteristics
characteristics of the
of the rockrock in the
in the impact
impact test
test were
were not
not considered
considered in thein process
the process of establishing
of establishing the equation
the equation andand would would improve
improve the the dy-
dynamic
namic
strength strength
of the rock of the rock
[54]. The[54].
curve The wascurve was generally
generally lower than lower than the experimental
the experimental curve.
curve.Compared with the results at 5.4 m/s, we found that the theoretical curve fitting effect
Compared
was better whenwith the the
impactresults at 5.4 m/s,
velocity was we 8.8 found
m/s and that11.3
the m/s.
theoretical
This is curve fitting
because theeffect
wave
was better when the impact velocity was 8.8 m/s and 11.3
shaper improved the smoothness of the stress wave at the expense of partial strain rate m/s. This is because the waveand
shaper improved
compressive the smoothness
strength. At 5.4 m/s,of in the
orderstress wavetruly
to more at the expense
reflect of partial strain
the mechanical rate
properties
and compressive
of the specimen itself, strength. At 5.4 m/s,
the selected wave in shaper
order tocouldmorenot truly reflect
shape well thethemechanical
rectangular prop-
wave
erties of thebyspecimen
generated the impact itself,
intothean selected
approximate wavesine shaper could
wave, not shape
resulting in a well
longthe rectangular
plateau segment
wave
in thegenerated
test curve.by the impact into an approximate sine wave, resulting in a long plateau
segment in thetotest
In order curve.
further illustrate the rationality and universality of the theoretical model
In order to further
based on the double-shear illustrate
unified thestrength
rationality and universality
theory, the uniaxialof the theoretical
impact compression model test
based
results of marble and coal under real-time high temperature from Yin et al., Yang test
on the double-shear unified strength theory, the uniaxial impact compression et al.,
results
Louis etofal.,marble
and Liu andandcoalXv.under
werereal-time
compared highwithtemperature
ours [55–58]. from TheYin et al.,obtained
results Yang et al.,after
Louis et al., and Liu and Xv. were compared with ours [55–58].
the model was fitted are shown in Figure 5. It can be seen that the theoretical results The results obtained after
the modelinwas
obtained thisfitted are shown
research showed in Figure
the same 5. Itgeneral
can be trend
seen that the experimental
as the theoretical results resultsob-of
tained in this research showed the same general trend as the
other scholars. They could better reflect the variations of the rock pre-peak intensity atexperimental results of other
scholars. They could better
different temperatures, andreflect
the errorthewasvariations
withinof the rock pre-peak
a reasonable range. Theintensity
error atwasdifferent
because
temperatures, and the error was within a reasonable range.
in the process of compressive deformation of the rock, there was sometimes a compaction The error was because in the
process
section before entering the elastic section. At this stage, the pores of the rock itselfsec-
of compressive deformation of the rock, there was sometimes a compaction were
tion before entering
compacted, causingthe theelastic
initialsection.
shape of At the
this curve
stage, to
thesag,
pores of theisrock
which itself were
inconsistent com-
with the
pacted,
theoreticalcausing
curve. the initial shape of the curve to sag, which is inconsistent with the theo-
retical curve.
Materials 2023, 16, x FOR PEER REVIEW 12 of 16

Materials 2023, 16, 2773 11 of 15

Figure5.5.Comparison
Figure Comparisonof
ofexperimental
experimentalresults
resultsin
inliterature
literatureand
andtheoretical
theoreticalcurves
curves[55–58].
[55–58].

5. Discussion
5. Discussion
M coupling is a process in which two physical fields, the stress field and the temper-
M coupling is a process in which two physical fields, the stress field and the temper-
ature field, influence each other, that is, the temperature has an influence on the stress
ature field, influence each other, that is, the temperature has an influence on the stress
deformation, and the stress deformation also affects the temperature. In other words, T-M
deformation, and the stress deformation also affects the temperature. In other words, T-
coupling requires the coexistence of a stress field and a temperature field and refers to the
M coupling requires the coexistence of a stress field and a temperature field and refers to
study of a rock mechanical properties under real-time high temperature, rather than to
the study of a rock mechanical properties under real-time high temperature, rather than
the study of a rock mechanical properties after a high-temperature treatment. There is an
to the study of a rock mechanical properties after a high-temperature treatment. There is
essential difference between the two. Studies have shown that the brittleness of granite
an essential difference between the two. Studies have shown that the brittleness of granite
is restored after applying a high temperature, and the mechanical properties of a rock
is restored after applying a high temperature, and the mechanical properties of a rock
specimen change suddenly only when the brittle–plastic transition occurs in specimen.
specimen the
Therefore, change suddenlyofonly
deterioration when the brittle–plastic
the mechanical properties of rock transition
specimensoccursunderin the
specimen.
action
Therefore, the deterioration of the mechanical properties of rock
of real-time high temperature proceeds continuously, while the mechanical behavior of rock specimens under the ac-
tion of real-time high temperature proceeds continuously, while
specimens after cooling following a high-temperature treatment shows a sudden change, the mechanical behavior
of rockisspecimens
which closely related aftertocooling
the phasefollowing a high-temperature
transition in the rock structure. treatment shows a sudden
change, which is closely related to the phase transition
It can be seen in this study that the coupling effect was nonlinear in the rock structure.
and changed contin-
It can be seen in this study that the coupling effect
uously throughout the loading process. The effect of temperature on granite was nonlinear and changed
damage was con-
tinuouslyAtthroughout
obvious. the loading
the same impact velocity,process. The effectdamage
the mechanical of temperature
increasedonwith granite damage
the increase
was
of theobvious.
temperature.At the same
This isimpact
because velocity, the mechanical
the thermal expansion damage increased
coefficients of the with the in-
minerals
crease of the temperature. This is because the thermal expansion
contained in granite are different, and the physical properties such as the thermal elasticity coefficients of the min-
erals contained in granite are different, and the physical properties
of minerals behave differently in different directions, that is, anisotropy. This can lead to such as the thermal
elasticity
an of minerals
uncoordinated behave
thermal differently
expansion in different
across directions, that
particle boundaries. is, anisotropy.
Another reason is that This
can
as the lead to an uncoordinated
temperature thermalmotion
rises, the thermal expansion across
of the particleincreases,
molecules boundaries. Another
resulting in rea-
the
son is that as
weakening temperature
of the force betweenrises, the thermal
molecules, causingmotion of theto
the lattice molecules
dislocateincreases,
or crack. result-
ing in the weakening
According of the force model
to the constitutive between molecules,above,
established causingthethe lattice to dislocate
determination of w0 , m,or
Ecrack.
T , and E 0 was the key issue in building the model. In order to clarify the influence of
According on
each parameter to the
the constitutive
stress–strainmodelcurve,established
the influence above,
of thethe determination
model parameters 𝑤0the
of on , m,
𝐸𝑇 , and 𝐸0curve
theoretical was is the keyfurther
here issue in buildingbythe
discussed model.
taking theIn order toatclarify
specimen ◦
200 Cthe underinfluence
8.8 m/s of
each
as parameter
an example andonanalyzing
the stress–strain
the resultscurve,
shown the in
influence
Figure 6. of the model parameters on the
theoretical
It can be curve
seenisinhere
Figurefurther
6 thatdiscussed
with the by takingofthe
decrease thespecimen at 200 °C under
model distribution parameters
8.8 m/s
was0 or
an m, the peak
example and stress and peak
analyzing strain ofshown
the results the specimen
in Figureboth
6. decreased significantly, but
there was no difference in the initial deformation stage of the specimen. Therefore, the
model parameters w0 and m mainly characterize the strength characteristics of the specimen
without affecting its lithological characteristics such as the elastic modulus. The parameter
ET mainly affected the slope and peak stress of the loading section of the stress–strain curve.
The increase of ET correspondingly increased the slope and peak stress of the loading
Materials 2023, 16, 2773 12 of 15

section of the stress–strain curve, but the peak strain was not affected. The effect of changes
in the
Materials 2023, 16, x FOR PEER REVIEW parameter E0 on the theoretical curve was exactly opposite to that of changes 13 of in
16
distribution parameter. With the decrease of E0 , the peak stress and peak strain of the
stress–strain curve both increased.

(a) (b)

(c) (d)
Figure 6. Influence of the model parameters on the stress–strain curves. (a) Effect of the parameter
Figure 6. Influence of the model parameters on the stress–strain curves. (a) Effect of the parameter
w_0 on the theoretical curve, (b) effect of the parameter m on the theoretical curve, (c) effect of the
w0 on the theoretical curve, (b) effect of the parameter m on the theoretical curve, (c) effect of the
parameter 𝐸𝑇 on the theoretical curve (d) effect of the parameter 𝐸0 on the theoretical curve.
parameter ET on the theoretical curve (d) effect of the parameter E0 on the theoretical curve.
It can
For beconstitutive
this seen in Figure 6 thatthis
model, with the decrease
study used the of the model
extreme value distribution
method toparameters
determine
𝑤 or m, the peak stress and peak strain of the specimen both
0 model parameters according to the rock yield criterion. This determination
the decreased significantly,
methodbutis
there was no difference in the initial deformation stage of the specimen.
suitable for the solution of loading tests at different temperatures, allows the easy deter- Therefore, the
model parameters
mination 𝑤0 and m mainly
of a rock mechanical characterize
parameters, the strengthfor
and is convenient characteristics of the speci-
engineering applications.
men without affecting its lithological characteristics such as the elastic
In the process of establishing the rock statistical thermal damage model, the mechanical modulus. The pa-
rameter 𝐸
parameters𝑇 and statistical distribution parameters used in the model have nothingstress–
mainly affected the slope and peak stress of the loading section of the to do
strainlithology,
with curve. The increase
so this model ofis𝐸also
𝑇 correspondingly
applicable to other increased the slope
rocks with and lithologies.
different peak stress of
the loading
It is worthsection
notingofthat
the the
stress–strain curve,
statistical T-M but the
coupling peak strain
constitutive was of
model not affected.
rocks The
proposed
effect
in thisof changes
study in the
ignores theparameter
influence of 𝐸0 primary
on the theoretical
defects of curve
the rockwas andexactly opposite
the partial to that
closure of
of changes inafter
micro-cracks distribution
heating.parameter.
Therefore, theWith the decrease
initial compaction of 𝐸segment
0 , the peak stress
of the and peak
stress–strain
strain cannot
curve of the stress–strain
be represented.curveInboth increased.
addition, the value of each parameter of this model has
nothing to do with the post-peak point, andused
For this constitutive model, this study therethe extreme
is no value
constraint onmethod to determine
the post-peak point;
the model parameters according to the rock yield criterion. This
therefore, the model cannot describe the mechanical behavior of a specimen after determination method
failure.is
suitable for the solution of loading tests at different temperatures, allows the easy deter-
mination of a rock mechanical parameters, and is convenient for engineering applications.
In the process of establishing the rock statistical thermal damage model, the mechanical
parameters and statistical distribution parameters used in the model have nothing to do
Materials 2023, 16, 2773 13 of 15

6. Conclusions
In this study, a real-time high-temperature impact compression test was conducted
on Sejila Mountain granite using the SHPB system and an intelligent synchronous high-
temperature furnace. The variation laws of peak stress, peak strain, and elastic modulus
were revealed, and a coupled T-M damage constitutive model based on the unified strength
theory was established. The main conclusions are as follows:
The ultimate temperature bearing capacity of Sejila Mountain granite was found to be
less than 800 ◦ C. When the temperature exceeded 400 ◦ C, the stress decreased significantly
with the increase of the temperature. The peak strain at high temperature was higher than
that at room temperature. At the same temperature, an increase in impact velocity had a
significant enhancing effect on the strength of the specimens.
The coupled T-M damage constitutive model established based on the unified strength
theory can fully reflect the strength characteristics of a rock before failure, can well describe
the stress–strain curves at different temperatures, and provides curves with a high degree
of fitting with the experimental curves.
The model parameter ET has a significant effect on the peak stress and slope of the
theoretical curve, but not on the peak strain. The distribution parameters of the model
mainly characterize the strength characteristics of a specimen. With a decrease in the
distribution parameters, the peak strength and peak strain of the theoretical curve decrease
significantly. The effect of the parameter E0 on the theoretical curve is exactly the opposite
of that of the distribution parameter.

Author Contributions: Conceptualization, Y.L. and Y.Z.; methodology, Y.L. and F.M.; formal analysis,
Y.L. and Y.X.; writing—original draft preparation, Y.X.; writing—review and editing, Y.L.; funding
acquisition, Y.L. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by National Natural Science Foundation of China (grant num-
ber 211026220437).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data that support the findings of this study are available on request
from the corresponding author.
Acknowledgments: This research was carried out in the School of Geological Engineering and
Geomatics at Chang’an University.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Kumari, W.G.P.; Ranjith, P.G.; Perera, M.S.A.; Li, X.; Li, L.H.; Chen, B.K.; Avanthi Isaka, B.L.; De Silva, V.R.S. Hydraulic fracturing
under high temperature and pressure conditions with micro CT applications: Geothermal energy from hot dry rocks. Fuel 2018,
230, 138–154. [CrossRef]
2. Haido, J.H.; Tayeh, B.A.; Majeed, S.S.; Karpuzcu, M. Effect of high temperature on the mechanical properties of basalt fibre
self-compacting concrete as an overlay material. Constr. Build. Mater. 2020, 268, 121725. [CrossRef]
3. Klewiah, I.; Berawala, D.S.; Walker, H.C.A.; Andersen, P.; Nadeau, P.H. Review of experimental sorption studies of CO2 and CH4
in shales. J. Nat. Gas Sci. Eng. 2019, 73, 103045. [CrossRef]
4. Brantut, N.; Heap, M.; Meredith, P.; Baud, P. Time-dependent cracking and brittle creep in crustal rocks: A review. J. Struct. Geol.
2013, 52, 17–43. [CrossRef]
5. Yavuz, H.; Tufekci, K.; Kayacan, R.; Cevizci, H. Predicting the Dynamic Compressive Strength of Carbonate Rocks from
Quasi-Static Properties. Exp. Mech. 2013, 53, 367–376. [CrossRef]
6. Wasantha, P.L.P.; Ranjith, P.G.; Zhao, J.; Shao, S.S.; Permata, G. Strain Rate Effect on the Mechanical Behaviour of Sandstones with
Different Grain Sizes. Rock Mech. Rock Eng. 2015, 48, 1883–1895. [CrossRef]
7. Mishra, S.; Meena, H.; Parashar, V.; Khetwal, A.; Chakraborty, T.; Matsagar, V.; Chandel, P.; Singh, M. High Strain Rate Response
of Rocks Under Dynamic Loading Using Split Hopkinson Pressure Bar. Geotech. Geol. Eng. 2018, 36, 531–549. [CrossRef]
Materials 2023, 16, 2773 14 of 15

8. Kumari, W.; Ranjith, P.; Perera, M.; Shao, S.; Chen, B.; Lashin, A.; Arifi, N.; Rathnaweera, T. Mechanical behaviour of Australian
Strathbogie granite under in-situ stress and temperature conditions: An application to geothermal energy extraction. Geothermics
2017, 65, 44–59. [CrossRef]
9. Yin, T.; Wang, C.; Wu, Y.; Wu, B. A waveform modification method for testing dynamic properties of rock under high temperature.
J. Rock Mech. Geotech. Eng. 2021, 13, 833–844. [CrossRef]
10. Gilat, A.; Wu, X. Elevated temperature testing with the torsional split hopkinson bar. Exp. Mech. 1994, 34, 166–170. [CrossRef]
11. Liu, S.; Xu, J.; Wang, H. Numerical Simulation Analysis of Real-Time High-Temperature Impact Test Technique of Rock Materials.
Geotech. Test. J. 2022, 45, 212–224. [CrossRef]
12. Liu, S.; Xu, J. Analysis on damage mechanical characteristics of marble exposed to high temperature. Int. J. Damage Mech. 2015,
24, 1180–1193. [CrossRef]
13. Xu, X.L.; Gao, F.; Zhang, Z.Z.; Lin, C. Experimental study of the effect of loading rates on mechanical properties of granite at
real-time high temperature. Rock Soil Mech. 2015, 36, 2184–2192.
14. Meng, L.; Li, T.; Liao, A.; Zeng, P. Anisotropic Mechanical Properties of Sandstone Under Unloading Confining Pressure at High
Temperatures. Arab. J. Sci. Eng. 2018, 43, 5283–5294. [CrossRef]
15. Finnie, I.; Cooper, G.A.; Berlie, J. Fracture propagation in rock by transient cooling. Int. J. Rock Mech. Min. Sci. Geomech. Abstr.
1979, 16, 11–21. [CrossRef]
16. Dougill, J.W. On stable progressively fracturing solids. Z. Angew. Math. Phys. (ZAMP) 1976, 27, 423–437. [CrossRef]
17. Cao, W.; Tan, X.; Zhang, C.; He, M. Constitutive model to simulate full deformation and failure process for rocks considering
initial compression and residual strength behaviors. Can. Geotech. J. 2019, 56, 649–661. [CrossRef]
18. Mubarak, H.; Saji, R.P.; Sousa, R.; Sassi, M.; Abu Al-Rub, R.K. Quantification of Plasticity and Damage in Berea Sandstone through
Monotonic and Cyclic Triaxial Loading under High-Confinement Pressures. J. Mater. Civ. Eng. 2022, 34, 04022116. [CrossRef]
19. Deng, J.A.; Cu, D.S. On a statistical damage constitutive model for rock materials. Comput. Geosci. 2011, 37, 122–128. [CrossRef]
20. Liu, Q.; Xu, X. Damage analysis of brittle rock at high temperature. Chin. J. Rock Mech. Eng. 2000, 19, 408–411.
21. Rong, G.; Yao, M.; Peng, J.; Sha, S.; Tan, J. Influence of initial thermal cracking on physical and mechanical behaviour of a coarse
marble: Insights from uniaxial compression tests with acoustic emission monitoring. Geophys. J. Int. 2018, 214, 1886–1900.
22. Zhang, Z.Z.; Gao, F.; Xu, X.L. Experimental study of temperature effect of mechanical properties of granite. Rock Soil Mech. 2011,
32, 2346–2352.
23. Dittmann, M.; Aldakheel, F.; Schulte, J.; Schmidt, F.; Krüger, M.; Wriggers, P.; Hesch, C. Phase-field modeling of porous-ductile
fracture in non-linear thermo-elasto-plastic solids. Comput. Methods Appl. Mech. Eng. 2020, 361, 112730. [CrossRef]
24. Ganczarski, A.; Skrzypek, J. A Study on Coupled Thermo-Elasto-Plastic-Damage Dissipative Phenomena: Models and Application
to Some Innovative Materials. J. Therm. Stress. 2009, 32, 698–751. [CrossRef]
25. Lukic, D.C.; Zlatanovic, E.M.; Jokanovic, I.M. Tunnel lining load with consideration of the rheological properties of rock mass and
concrete. Geomech. Eng. 2020, 21, 53–62.
26. Singh, A.; Kumar, C.; Kannan, L.G.; Rao, K.S.; Ayothiraman, R. Rheological Behaviour of Rock Salt Under Uniaxial Compression.
Procedia Eng. 2017, 173, 639–646. [CrossRef]
27. Molladavoodi, H.; RahimiRezaei, Y. Heterogeneous Rock Simulation Using DIP-Micromechanics-Statistical Methods. Adv. Civ.
Eng. 2018, 2018, 7010817. [CrossRef]
28. Dubé, M.; Doquet, V.; Constantinescu, A.; George, D.; Rémond, Y.; Ahzi, S. Modeling of thermal shock-induced damage in a
borosilicate glass. Mech. Mater. 2010, 42, 863–872. [CrossRef]
29. Kolo, I.; Abu Al-Rub, R.K.; Sousa, R.L. Computational Modelling of Fracture Propagation in Rocks Using a Coupled Elastic-
Plasticity-Damage Model. Math. Probl. Eng. 2016, 2016, 3231092. [CrossRef]
30. Girard, L.; Weiss, J.; Amitrano, D. Damage-Cluster Distributions and Size Effect on Strength in Compressive Failure. Phys. Rev.
Lett. 2012, 108, 225502. [CrossRef]
31. Saurav, R.; Kumar, S.N. Strength behaviour of sandstone subjected to polyaxial state of stress. Int. J. Min. Sci. Technol. 2017, 27,
889–897.
32. Lelovic, S.; Vasovic, D. Determination of Mohr-Coulomb parameters for modelling of concrete. Crystals 2020, 10, 808. [CrossRef]
33. Grange, S.; Mencacci, S.; Hild, F. Modelling dynamic fragmentation of rocks. J. Phys. IV Fr. 2006, 134, 125–131. [CrossRef]
34. Rahimi, R.; Nygaard, R. Effect of rock strength variation on the estimated borehole breakout using shear failure criteria. Géoméch.
Geophys. Geo-Energy Geo-Resour. 2018, 4, 369–382. [CrossRef]
35. Melkoumian, N.; Priest, S.D.; Hunt, S.P. Further Development of the Three-Dimensional Hoek–Brown Yield Criterion. Rock Mech.
Rock Eng. 2009, 42, 835–847. [CrossRef]
36. Xie, W.H.; Shun, C.L.I.; Gao, F. Energy fracture criterion of thermal damage-mechanical coupling. J. Xi’an Univ. Sci. Technol. 2007,
27, 341–346.
37. Yu, M.H.; He, L.N. A new model and theory on yield and failure of materials under the complex stress state. Mech. Behav. Mater.
VI 1992, 841–846. [CrossRef]
38. Guo, Z.Y.; Wang, H.N.; Jiang, M.J. Elastoplastic analytical investigation of wellbore stability for drilling in methane hydrate-
bearing sediments. J. Nat. Gas Sci. Eng. 2020, 79, 103344. [CrossRef]
39. Li, Y.; Liu, Q.L.; Cai, Q.C. Unification of Hoek-Brown Criterion on the Basis of United Strength Theory. Appl. Mech. Mater. 2014,
501–504, 415–418. [CrossRef]
Materials 2023, 16, 2773 15 of 15

40. Yang, Q.; Zan, Y.; Xie, L.G. Comparative analysis of the nonlinear unified strength criterion for rocks and other three-dimensional
Hoek–Brown strength criteria. Géoméch. Geophys. Geo-Energy Geo-Resour. 2018, 4, 29–37. [CrossRef]
41. Del Greco, O.; Ferrero, A.M.; Oggeri, C. Experimental and analytical interpretation of the behaviour of laboratory tests on
composite specimens. Int. J. Rock Mech. Min. Sci. Géoméch. Abstr. 1993, 30, 1539–1543. [CrossRef]
42. Kully, R.M. Development of new method for direct measurement of high strain rate testing parameters. Exp. Appl. Mech. 2016, 4,
139–147.
43. Othman, R.; Bissac, M.N.; Collet, P.; Gary, G. Testing with SHPB from quasi-static to dynamic strain rates. J. Phys. IV Fr. 2003, 110,
397–404. [CrossRef]
44. Zuo, J.-P.; Wang, J.-T.; Sun, Y.-J.; Chen, Y.; Jiang, G.-H.; Li, Y.-H. Effects of thermal treatment on fracture characteristics of granite
from Beishan, a possible high-level radioactive waste disposal site in China. Eng. Fract. Mech. 2017, 182, 425–437. [CrossRef]
45. Chen, S.; Yang, C.; Wang, G. Evolution of thermal damage and permeability of Beishan granite. Appl. Therm. Eng. 2017, 110,
1533–1542. [CrossRef]
46. Chen, Y.L.; Wang, S.R.; Ni, J.; Azzam, R.; Fernandez-Steeger, T.M. An experimental study of the mechanical properties of granite
after high temperature exposure based on mineral characteristics. Eng. Geol. 2017, 220, 234–242. [CrossRef]
47. Wingquist, C.F. Elastic Moduli of Rock at Elevated Temperatures; US Department of the Interior, Bureau of Mines: Washington, DC,
USA, 1970.
48. Kuforiji, H.I.; Olurin, O.T.; Akinyemi, O.D.; Idowu, O.A. Wave velocity variation with temperature: Influential properties of
temperature coefficient (partial derivative V/partial derivative T) of selected rocks. Environ. Earth Sci. 2021, 80, 638. [CrossRef]
49. Ahmed, H.M.; Hefni, M.A.; Ahmed, H.A.; Adewuyi, S.O.; Hassani, F.; Sasmito, A.P.; Saleem, H.A.; Moustafa, E.B.; Hassan, G.S.A.
Effect of elevated temperature on rhyolitic rocks’ properties. Materials 2022, 15, 3204. [CrossRef]
50. Grange, S.; Forquin, P.; Mencacci, S.; Hild, F. On the dynamic fragmentation of two limestones using edge-on impact tests. Int. J.
Impact Eng. 2008, 35, 977–991. [CrossRef]
51. Choi, Y.; Chung, W.K.; Ha, J.; Shin, S.R. Formation estimation of shaly sandstone reservoir using joint inversion from well logging
data. Geophys. Geophys. Explor. 2019, 22, 1–11. [CrossRef]
52. Liu, Y.-S.; Qiu, Z.-Z.; Zhan, X.-C.; Liu, H.-N.; Gong, H.-N. Study of statistical damage constitutive model of layered composite
rock under triaxial compression. Appl. Math. Nonlinear Sci. 2021, 6, 299–308. [CrossRef]
53. Zhu, Z.; Tian, H.; Wang, R.; Jiang, G.; Dou, B.; Mei, G. Statistical thermal damage constitutive model of rocks based on Weibull
distribution. Arab. J. Geosci. 2021, 14, 495. [CrossRef]
54. Yang, L.; Li, Z.-D. Nonlinear Variation Parameters Creep Model of Rock and Parametric Inversion. Geotech. Geol. Eng. 2018, 36,
2985–2993. [CrossRef]
55. Yin, T.B.; Wang, P.; Li, X.B.; Shu, R.H.; Ye, Z.Y. Effects of thermal treatment on physical mechanical characteristics of coal rock. J.
Cent. South Univ. 2016, 23, 2336–2345. [CrossRef]
56. Yang, R.; Fang, S.; Li, W.; Wei, G.; Li, Q.; Liang, S. Temperature effects on dynamic compressive behavior of siliceous sandstone.
Arab. J. Geosci. 2020, 13, 382. [CrossRef]
57. Wong, L.N.Y.; Li, Z.; Kang, H.M.; Teh, C.I. Dynamic Loading of Carrara Marble in a Heated State. Rock Mech. Rock Eng. 2017, 50,
1487–1505. [CrossRef]
58. Liu, S.; Xu, J. Study on dynamic characteristics of marble under impact loading and high temperature. Int. J. Rock Mech. Min. Sci.
2013, 62, 51–58. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.