P09 .

Downloaded 05/24/22 to 133.3.201.92. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/page/policies/terms

Effect of confining pressure during thermal treatment processes on rock properties

Ziyang Zhou*, Hitoshi Mikada, Junichi Takekawa, Shibo Xu, (Kyoto University)

Summary investigated for the efficient development of geothermal

reservoirs.
In the present study, we investigate the thermal effect on
physical and mechanical properties of rock mass using In order to study the influence of this difference and better
numerical experiments based on the discrete element reproduce the state of thermal storage rock formation, we
method (DEM) to understand the influence of the thermal build a thermal-mechanical coupling model to conduct a
treatment process in geothermal reservoir rocks. We series of numerical simulation tests based on the discrete
propose a new thermal-mechanical coupling model for element method (DEM) (Cundall and Strack, 1979) to
DEM to focus on the cooling effect under various analyze the influence of different thermal treatment
confining pressure. After verifying the proposed method process on rock properties. We verify the proposed method
by comparing numerical results with those in laboratory by comparing the results with laboratory experiment
experiments under various temperatures, uniaxial results. Then, thermal treatment with different confining
compression tests with different thermal treatment pressure on rock specimens' physical and mechanical
processes are conducted. We apply a confining pressure to properties is investigated.
rock specimens for representing the realistic environment
of geothermal reservoirs. Our numerical results show that Methodology
the confining pressure during the thermal treatment process
has a great influence on the physical and mechanical A rock mass is regarded as the assembly of small particles
properties of rock. If the confining pressure is not applied in DEM, and therefore the deformation behavior (large
DOI:10.1190/iceg2021-075.1

to the specimen during the cooling process, its strength deformation behavior in particular) of rock could be
degraded significantly. Our results indicate that it is crucial simulated in a simple manner. As one of the most
to consider the coupling effect of thermal and confining commonly used numerical models in DEM, the bonded-
pressure conditions. particle model (BPM) (Potyondy and Cundall, 2004)
directly mimics the cemented granular material in which
Introduction the cement is deformable and may break, and thus
correspond very well with real rock. In BPM, neighboring
As the global population continues to grow, the electricity particles are bonded together, and this inter-particle
demand is increasing sharply. On this behalf, geothermal interaction is simplified as normal, shear, and rotational
energy has become one of the most promising renewable springs. Consequently, the increment of force and moment
energy resources because of its huge, practically could be calculated through a linear relationship based on
inexhaustible reserves. The temperature of geothermal Hooke's law. For the commercial production of electricity,
fields for the development rises over time to where the the temperature of the rock mass is very high. Under such
ductile region of rocks could be seen. The application of an environment, the mechanical behaviors of rock will
enhanced geothermal systems (EGS) that uses hydraulic change significantly compared with the normal
fracturing would become necessary (Reinsch et al., 2017). temperature state. Based on the experimental results
It becomes an inevitable demand to understand the effect obtained by Yang et al. (2017), these differences are
of high temperatures on the physical and mechanical mainly caused by different thermal-expansion behaviors of
properties of the reservoir rocks. For this purpose, many various grains such as quartz, feldspar, biotite, and many
laboratory experiments have been carried out (Yang et al., other rock-forming minerals. In addition, the mechanical
(2017; Rossi et al., 2018; Shen et al., 2020). However, the properties of the mineral grains themselves will also
heating mechanism of the thermal storage rock formation change proportional to temperature accordingly.
is complicated, which may include radioactive heat
generation, tectonic activity, volcanic activity, and other A numerical model considering the thermally-induced
ways (Lu et al., 2017). In these cases, the deformation of cracks is proposed to reproduce the above mechanism in
thermal storage rock formation is constrained by the the DEM simulation, different from the bond damage
surrounding rock mass while being heated and always determination algorithm in the traditional DEM. A damage
keeps its thermal energy before being developed, rather variable D is introduced to specify the damage of bonds
than being cooled to room temperature in laboratory with the mechanism that when D equals zero, the bond is
experiments. These characteristics are different from the intact; alternatively, when D equals one, the bond is
commonly used thermal treatment process of rock completely damaged. An exponential form is applied to
specimens in laboratory experiments. The above difference calculate the damage variable D to ensure a smooth
in rock's physical and mechanical properties should be transition between the intact and broken bonds.

ICEG2021, 25-28 October 2021, Al Ain, UAE

Sixth International Conference on Engineering Geophysics (ICEG)

Virtual, 25–28 October 2021 10.1190/iceg2021-075.1 Page 294
 Effect of confining pressure during thermal treatment processes on rock properties
Downloaded 05/24/22 to 133.3.201.92. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/page/policies/terms

D 1 e , (1)
Validation of the proposed method
where ε , ε ε and ε are the total and ultimate strains in
the normal direction, and and the total strain and ultimate Based on the above theories, numerical simulations of
strains in the shear direction, respectively. Then, the uniaxial compression tests after different temperature
strength and stiffness of the bond will be updated by treatments are carried out to validate the proposed scheme.
multiplying the damage variable so that the bonds are Firstly, the temperature of numerical specimens is
gradually damaged with the increase in strain. Considering increased to their target temperature (200, 300, 400, 500,
the different thermal expansion behavior of mineral grains 600, 700, and 800 ) at a slow rate of 0.45 /min to
in real rock, different thermal expansion coefficients α are reduce the influence of the heating rate itself. In this
assigned to DEM particles so that the radius of particles process, the radius of the particles expands as the
will change for the temperature T as follows. temperature increases. After the model reaches a steady
state, the specimens are cooled to room temperature (25 )
∆ , (2) at the same rate. Subsequently, the particle radius will also
shrink as the temperature decreases. The different thermal
where and are the radius of the i-th particle for the expansion and contraction behavior of DEM particles
temperature T and room temperature, and the thermal caused by different thermal expansion coefficients will
expansion coefficient of the particle, and ∆ the cause thermal stress and generate thermally-induced cracks.
temperature change, respectively. With the expansion or
shrink of particles, the corresponding force will also The numerical model after the above process is used for
change to cause thermal stress in the DEM model. uniaxial compression test simulations. Figure 1 shows the
evolution of static elastic modulus and uniaxial
DOI:10.1190/iceg2021-075.1

By considering the effect of high temperature on the compressive strength (UCS) with increasing heat treatment
mineral grains, such as chemical changes and transgranular temperature for the numerical specimens. The uniaxial
cracks, the elastic modulus of the particles is negatively compressive strength and static elastic modulus of granite
correlated with the temperature, which is described as specimens are significantly decreased in high temperatures,
follows. and the changing trends of both are consistent with the
experimental results illustrated in Figure 2.
1 1 300 , (3)
Numerical simulations with different constraints and
where , , and, are the contact stiffness between thermal treatment process
particles at temperature T and room temperature, and a
coefficient. It is worth mentioning that this equation is only To better reproduce the environment of thermal storage
applied when the temperature is higher than 300 . When rock formations constrained by surrounding rock masses,
the temperature is lower than 300 , the contact stiffness the numerical specimens after different heat treatment
does not change. processes are applied for uniaxial compression simulation,
and the specific processing flow could be summarized as:
In addition, when the temperature is low (< 300 ), the
thermal stress is not enough to induce micro-cracks in the Apply the confining pressure to numerical specimens to a
granite, so the increasing mutual attraction and bonding predetermined value.
strength between mineral grains caused by mutual distance
reduction dominate. The increase in bond strength in the Increase the temperature of specimens to the target
numerical simulation is expressed as follows. temperature while keeping the confining pressure constant;
After the model stabilizes, the temperature of the
1 , (4) specimens in group A is lowered to room temperature, as a
comparison, the temperature of specimens in group B
remains unchanged, the confining pressure of both groups
where is a coefficient, which is set to be 0.0005, and remained unchanged;
and are the bond strength at temperature T and room
temperature, respectively. In these equations, when T is
greater than the threshold temperature (300 in this
paper), T takes the threshold temperature.

ICEG2021, 25-28 October 2021, Al Ain, UAE

Sixth International Conference on Engineering Geophysics (ICEG)

Virtual, 25–28 October 2021 10.1190/iceg2021-075.1 Page 295
 Effect
E of conffining pressu
ure during th
hermal treatm
ment processses on rock p
properties
Downloaded 05/24/22 to 133.3.201.92. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/page/policies/terms

(a)

Figgure 3: Crack patteerns of uniaxial coompression simulattion

(b)
DOI:10.1190/iceg2021-075.1

afteer thermal treatm

ments at 600 annd confining presssure
Figure 1:: Influence of tem
mperature on (a) Static elastic durring thermal treatmment: (a) 0MPa, ((b) 2MPa, (c) 4M MPa,
modulus; (b) Uniaxial com mpressive strength
h obtained by (d) 6MPa, (e) 8MP Pa and (f) Room temperature conntrol
numericall simulation. grooup, the color of cracks here repreesents the normaliized
craack width.

Afterr the above treatm

ment, numericall specimens werre used
for unniaxial compresssion simulation.

Numeerical simulationn results of uniaaxial compressioon tests

after different thermmal treatment prrocesses are shoown in
Figurre 3. From Figu ure 3, it could be observed thaat after
therm
mal treatment aat 600 , the failure mode of the
uniaxxial compressioon specimen sshows typical high-
tempeerature failure characteristics; due to the thhermal
(a) treatm
ment, initial defe
fects caused by tthermal stress eexist in
numeerical specimen bbefore the uniaxxial compressionn test is
perforrmed. Thereforee, the crack patttern of the speecimen
after the uniaxial coompression test is more complicated
than that of room teemperature (Figgure 3, (f)), whhich is
maniffested as multipple axial cracks; and many smalll-scale
crackks develop and ccoalesce to form m a complex cracck net.
In adddition, with thee increase of thhe confining prressure
during
ng the thermal treatment (Figu ure 3, (a)-(e)),, these
charaacteristics do nott change significcantly.

Figurres 4 and 5 m may show the foollowing charactteristic

behavvior of rock specimens under sttress when the hheating
(b) treatm
ment process iinvolves both heating and ccooling
Figure 2: Influence of tem mperature on (a) Dynamic and proceedures. If the heaating treatment pprocess is heatinng only,
static elasstic modulus; (b) Uniaxial compreessive strength the iincrease in connfining pressuree during the hheating
and crackk damage thresholdd (Yang et al., 2017). ment may increaase the uniaxiall compressive sttrength
treatm
and sstatic modulus oof the processed specimen. Due to the

ICEG2021, 25-28 Octobber 2021, Al Ainn, UAE

Sixth International Conference on Engineering Geophysics (ICEG)

Virtual, 25–28 October 2021 10.1190/iceg2021-075.1 Page 296
 Effect
E of conffining pressu
ure during th
hermal treatm
ment processses on rock p
properties
Downloaded 05/24/22 to 133.3.201.92. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/page/policies/terms

numeerical specimen. When the connfining pressure is not

applieed during the thermal treatm ment, the heatinng and
coolinng stages will generate therm mal stress insidde the
numeerical specimen. Therefore, the sspecimen after hheating
and ccooling treatmennt contains more initial defects thhan the
specimmen that has onlly undergone heeating treatment, which
is man
anifested by the llower strength.

Whenn confining prressure is appplied during thhermal

treatm
ment, during thee heating stage, the particles unndergo
Figure 4: Uniaxial comprressive strength of numerical therm
mal expansion; duue to the confiniing pressure appplied to
specimen
n after different thermal treatmeent processes the sspecimen, greateer thermal stress may be gennerated
insidee the specimen, which may causse more initial ddefects.
Thereefore, when therre is no coolingg stage, the strenngth of
the sppecimen may bee reduced. Howeever, when the thhermal
treatm
ment process inccludes the cooling stage, the paarticles
will sshrink during tthe cooling stagge, and the connfining
pressuure applied to the specimenn helps to makke the
specim men piece denser, the initial thhermal stress craacks in
the innterior of the sppecimen will allso be closed, aand the
inter--particle interlocck will be strenggthened. Therefoore, the
specim men strength coould be improvedd, and within a ccertain
Figure 5:: Static elastic modulus of th
he numerical rangee, the higher thhe confining prressure, the moore the
specimen
n after different thermal treatmen
nt processes strenggth increases. R Regardless of whhether the specim men is
DOI:10.1190/iceg2021-075.1

cooleed or not, the hhigher confiningg pressure durinng the

mal treatment w
therm will always hellp to form a denser
increase in confining pressu ure during heatin
ng treatment, thee
specim men, and the sttatic elastic moddulus characterizzes the
static mod dulus of the specimen durring processing g
behavvior of the speecimen before failure, so wiithin a
continues too increase. The uniaxial
u compressive strength off
speciffic range, a highgher confining ppressure always results
the specimmen appears to increase,
i but th
he trend for thiss
in a higher modullus of elasticityy. The static elastic
change is not
n clear, and itt may not alway ys be concluded d
moduulus of the speciimen after cooliing is higher thaan that
that there is a simple positive correllation with thee
of thee specimen withoout cooling for tthe same reason.
confining pressure.
p Due tot the pressure change in thee
heating treaatment process, the
t confining preessure increases,
Concclusion
but the uniiaxial compressiive strength may y decrease. Thee
comparison n results of the tw
wo numerical simulation groupss
The Thermal-mechaanical coupling model is appllied to
show that thhe static moduluus of the specimeen cooled duringg
investtigate the mecchanism of thhermal expansiion of
the thermall treatment proccess is higher than
t that of thee
particcles by DEM. T The simulated reesults are verifieed with
specimen without
w thermal treatment. Wheether or not thee
laboraatory experimennt results of uniaaxial compressioon tests
specimens are
a cooled in thee heat has a moree complex effectt
underr various temperratures, then tessted on the speccimens
on the strenngth of the test piece. The streength of the testt
after different therm mal treatment processes. Appplying
piece not placed under confining presssure during thee
confinning pressure too the numericall specimen during the
thermal treeatment process is significantly y lower than thee
thermmal treatment prrocess helps inccrease the speccimen's
strength of the uncooled test piece. The strrength of the testt
static elastic moduluss because it couuld help form a denser
piece increeases with the confining presssure during thee
structture. When thee confining preessure is not aapplied
thermal treatment, and thee strength gap between
b the two
o
during
ng the thermal trreatment, coolingg the heated speecimen
numerical test
t piece group ps with and with hout the thermall
will rreduce the streength of the sppecimen due too more
treatment process
p gradually
y narrows. Wheen the confining g
substaantial thermal sttress. As the coonfining pressuree rises,
pressure ex xceeds 4 MPaa, the strength of the cooled d
the sttrength of the sspecimen after ccooling treatmennt will
specimen iss higher than thaat of the uncooledd specimen.
graduually increase annd exceed that off the specimen thhat has
not beeen cooled undeer the same condditions.
Discussion

The effect of the cooling phase

p on the speecimen observedd
in the numerical
n simu
ulation is com mplicated. Thee
mechanism m of this phenom menon should be b related to thee
initial defeccts generated affter the thermal treatment of thee

ICEG2021, 25-28 Octobber 2021, Al Ainn, UAE

Sixth International Conference on Engineering Geophysics (ICEG)

Virtual, 25–28 October 2021 10.1190/iceg2021-075.1 Page 297
 Effect of confining pressure during thermal treatment processes on rock properties
Downloaded 05/24/22 to 133.3.201.92. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/page/policies/terms

References
1) Anna, S., Mirosława, B., Tomasz, J.: High temperature
versus geomechanical parameters of selected rocks –
the present state of research. Journal of Sustainable
Mining, 12, pp. 45-51, 2013.
2) Barbier, E.: Geothermal energy technology and current
status: an overview, Renewable and Sustainable
Energy Reviews, 6 (1)(2), pp.3-65, 2002. doi:
10.1016/S1364-0321(02)00002-3.
3) Cundall, P.A. and Strack, O.D.L.: A discrete numerical
model for granular assemblies. Geotechnique, 29 (1),
pp.47-65, 1979.
4) Cundall, P.A. and Strack, O.D.L.: A bonded-particle
model for rock. International Journal of Rock
Mechanics and Mining Sciences, 41 (8), pp.1329-1364,
2004. doi: 10.1016/j.ijrmms.2004.09.011
5) Lu, C., Lin, W., Gan, H., Liu, F., W, G.: Occurrence
types and genesis models of hot dry rock resources in
China. Environmental Earth Sciences, 76 (646), 2017.
doi: 10.1007/s12665-017-6947-4
6) Reinsch, T., Dobson, P., Asanuma, H., Huenges, E.,
Poletto, F., Sanjuan, B.: Utilizing supercritical
geothermal systems: a review of past ventures and
DOI:10.1190/iceg2021-075.1

ongoing research activities. Geothermal Energy, 5 (1),

art. no. 16, 2017. doi: 10.1186/s40517-017-0075-y
7) Rossi, E., Kant, M.A., Madonna, C. Saar, M., Rudolf
von Rohr, P.: The Effects of High Heating Rate and
High Temperature on the Rock Strength: Feasibility
Study of a Thermally Assisted Drilling Method. Rock
Mechanics and Rock Engineering, 51, pp. 2957–2964,
2018. doi: 10.1007/s00603-018-1507-0
8) Shen, Y., Hou, X., Yuan, J., Xu, Z., Hao, J., Gu, L. and
Liu, Z.: Thermal deterioration of high-temperature
granite after cooling shock: multiple-identification and
damage mechanism. Bulletin of Engineering Geology
and the Environment, 79, pp. 5385-5398, 2020.
9) Shu, R., Yin, T., Li, X.: Effect of Heating Rate on the
Dynamic Compressive Properties of Granite.
Geofluids, 2019. doi: 10.1155/2019/8292065
10) Tester, J. W., Anderson, B. J., Batchelor, A. S.,
Blackwell, D. D., DiPippo, R., Drake, E. M., ... &
Veatch, R. J. The future of geothermal energy.
Massachusetts Institute of Technology, 358, 2006.
11) Yang, S., Ranjith, P., Jing, H., Tian, W. and Ju, Y.: An
experimental investigation on thermal damage and
failure mechanical behavior of granite after exposure
to different high temperature treatments. Geothermics,
65, pp. 180-197, 2017.

ICEG2021, 25-28 October 2021, Al Ain, UAE

Sixth International Conference on Engineering Geophysics (ICEG)

Virtual, 25–28 October 2021 10.1190/iceg2021-075.1 Page 298