You are on page 1of 8

Healing and Shear Stress Reduction of Rock Salt Single Fracture

under Slide-Hold-Slide Direct Shear Experiments

Kiyoshi Kishidaa*, Takao Yanob and Hideaki Yasuharac


a
Associate Professor, Kyoto University, Kyoto, Japan
b
Technitian, Kyoto UNiversity, Kyoto, Japan
c
Associate Professor, Ehime University, Matsuyama, Japan
*
Corresponding Author’s E-mail: kishida.kiyoshi.3r@kyoto-u.ac.jp

ABSTRACT
The evolution of the long-term mechanical, hydraulic, and transport characteristics of rock
fractures must be predicted in advance by considering the issue of the underground deposits of the
energy by products of high-level radioactive waste. This paper presents slide-hold-slide direct shear
experiments conducted for rock salt (halite) specimens with single fractures so as to investigate the
effects of load holding on the mechanical properties of rock joints. From the experimental results, it is
confirmed that the shear stress reduction can be observed during load holding. It is also confirmed that
the shear strength recovery can be observed after load holding and it is found to increase with the
increment of load holding period. This is because the dominant factor in the shear strength recovery
during holding may be attributed to a purely mechanical process, like creep deformation at the
contacting asperities.

Keywords: Healing, Halite, Rock Fracture, Slide-hold-slide

1. INTRODUCTION
In the nuclear waste disposal and CO2 geo-sequestration projects, it is of significant importance to
discuss and evaluate the long-term changes in the mechanical and the hydro-mechanical properties of
jointed rock masses. While these projects are being conducted, the hydraulic conductivity of the rock
masses may change due to deformation resulting from changes in the stress distribution induced by the
excavation of the underground cavern and the underground deposits of waste matter in the short-term
range.
The chemo-mechanical fluid-rock interactions, which result in a reduction in permeability and the
recovery of fracture strength under hydrothermal and confining pressure conditions, have been
reported for sandstone (Tenthorey, et al., 2003). A significant strength recovery (i.e., 75% of the initial
rock strength) of the fractured sandstone is measured in a short pressure-holding period of 6 hrs. It is
also reported that during the holding, the permeability of the rock fractures decreases monotonically
with time. The applied confining stress of 50 MPa and the temperature of 927°C are both significantly
high. This is because the deep underground conditions are being simulated during interseismic periods.
Yasuhara et al. (2005) reported such a strength recovery using a quartz aggregate under relatively low
confining stress (i.e., 5 MPa) and temperature (i.e., 65°C) conditions, and explained that this evolution
of strength may be attributed to the chemo-mechanical process of the pressure solution. This chemo-
mechanical process may contribute to strength recovery even under relatively low stress and
temperature conditions (Yasuhara, et al., 2005; Niemeijer, et al., 2008; Kishida, et al., 2011). One of
the pioneer research works in the recovery of shear strength was conducted by Dieterich (1972, 1974,
1994). In the cyclic process of slide and hold in friction table experiments using particle quartz,
Dieterich (1972) confirmed that the increments in strength recovery were apparent in the case of the
longer holding period. The reasons for this phenomenon were considered to be that the contacted area
on the rock surface was practically very small due to the roughness and that the contacted asperities

594
were applied to a high level of stress. Consequently, creep deformation occurred at the contacted
asperities, due to the high stress condition, and the contacted area was expanded as time-dependent
(Dieterich , 1994).
In order to clarify the influence of the holding state on the shear strength in the shear process of a
single rock fracture, slide-hold-slide (SHS) direct shear-flow coupling tests were carried out on single
rock fractures at several confining stresses and under saturated/unsaturated conditions (Kishida, et al.,
2011). In this research work, two types of specimens were employed, namely, granite and artificial
(mortar) replica specimen including natural single rock joint. Consequently, the mortar specimen
could be confirmed as the significant shear strength recovery on the SHS process. On the other hand,
the clearly shear strength recovery could not be observed on the SHS direct shear test of granite rock
specimen. In this research, the SHS direct shear tests are carried out on the halite (rock salt) which is
intentionally introduced tensile single rock fracture. The uniaxial compressive strength of the halite is
almost equivalent. In addition, since the halite consists of mainly single mineral of NaCl, it is thought
that a conceptual model of pressure solution presented by Yasuhara, et al. (2011) can applied to
discuss the phenomena of SHS process. These are the reason why the halite is introduced to the SHS
direct shear tests.

2. SLIDE-HOLD-SLIDE (SHS) DIRECT SHEAR EXPERIMENTS


2.1 Speciemen
As mentioned above, the halite which is a kind of the sedimentary rock is applied to this research
work. A single tensile fracture is artificially created by cutting away at a halite block. Then, a
rectangular solid is formed with a cross section of 120 mm by 80 mm (80 mm is the shear direction)
and a height of 120 mm under dry condition. The uniaxial compressive strength of the halite is 49.6
MPa and is also equivalent as the mortar specimens which were employed by Kishida, et al. (2011).
Photo 1 shows the specimen including the tensile facture.

Photo 1. Rock specimen including tensile rock fracture

Measuring the joint surface roughness and evaluating the contact points between two surfaces are
very challenging and yet important tasks necessary to discussions on the mechanical and the hydro-
mechanical behavior of rock fractures. The 3-D roughness profiling system (Sato and Sawada, 2011)
is adopted to profile the joint surface roughness before and after performing the tests. This system
consists of an X–Y positioning table, and a laser scan micrometer, with a measurement allowance spot
dimension of 70 μm. The rough joint surface is measured at 0.25 mm intervals. The profiling data are
accumulated and obtained into the computer as the digital format. Contour maps of the joint surface
roughness on both sides of the halite fracture profiled by a laser scan micrometer are shown in Figure 1.

2.2 Experimental procedure and conditions


The shear tester is employed (Ohnishi, et al., 2000). The shear unit of the apparatus is controlled by
an electric and oil pressure servo mechanism which includes a feedback system. Four items, namely,
shear (horizontal) load, shear displacement, normal (vertical) load, and normal displacement,

595
(a) Bottom side (b) Upper side
Figure 1. Contour maps of joint surface roughness through Halite tensile fracture (Before performing
the SHS direct shear experiments)

are measured in the shear unit, and the obtained data are automatically recorded in the computer. The
loads are measured by two types of load cells for the vertical load and for the horizontal load. In the
experiments, the area sheared along a single fracture decreases as the shearing process advances. In
order to maintain a constant normal confining stress, the normal load is automatically adjusted by
considering the reduction in shear area. Electric gap sensors are employed to measure both vertical and
horizontal displacements.
In this paper, the halite specimen including single rock joint is employed for the SHS direct shear
experiment under various normal confining stress conditions and constant room temperature condition
at 20 °C. The tests were performed at every predetermined shear displacement up to 3 mm and the
constant normal confining stresses at 0.5, 1.0, and 3.0 MPa. The measured parameters in these
experiments are the shear displacement, the shear stress, the normal displacement, the normal stress,
the discharge under a constant shear displacement velocity of 0.1 mm/min and various holding periods.
In order to clarify that the re-shearing behavior of rock joint after various periods holding is
affected as the contact asperities on rock joint under the constant normal confining stress condition,
SHS direct shear experiments are carried out. In the SHS direct shear experiment, the shear are
conducted to some shear displacement. And then, the shear displacement is hold for some periods
under the same normal confining stress condition. After that, the shear is re-started. The concept of the
SHS direct shear experiment is shown in Figure 2.

σn
τ
Shear stress [MPa]

Slide

Slide

Holding

Shear displacement [mm]

Figure 2. The conceptual image of slide-hold-slide process on the direct shear test of rock joint

In this research, 3 kinds of normal confining stress conditions are employed on the one rock salt
specimen which includes tensile fracture. The SHS direct shear tests are conducted under room
temperature (20 °C) and constant shear velocity of 0.1 mm/min. They are also conducted until the
shear displacement at 3.0 mm and 4 cases, namely, Case I of 0.5 MPa of the constant normal confining

596
stress condition, Case II of 1.0 MPa, Case III of 0.5 MPa and Case IV of 3.0 MPa. The SHS processes
are conducted on the residual stress state and various holding periods from 60 seconds to 7 days are
employed. Table 1 shows the experimental cases and the holding periods in this research work.

Table 1. Experiment’s cases, normal confining stress and holding period


Normal confinig
Case Holding period
stress, σn [MPa]
I 0.5 MPa 60 sec., 180 sec., 300 sec., 420 sec., 900 sec.
60 sec., 180 sec., 300 sec., 420 sec., 900 sec., 1800 sec., 10800 sec.,
II 1.0 MPa
5 days, 7 days
60 sec., 180 sec., 300 sec., 420 sec., 900 sec., 1800 sec., 10800 sec.,
III 0.5 MPa
2 days, 4 days
IV 3.0 MPa 60 sec., 180 sec., 300 sec., 420 sec., 900 sec., 1800 sec., 10800 sec.

3. EXPERIMENTAL RESULTS AND DISCUSSIONS


3.1 Results
Figure 3 shows the shear stress - shear displacement and the dilation - shear displacement relations
of the SHS direct shear experiments on the rock salt. From all cases, the shear stress increases at the
initial phase of the experiments, and then, the shear stress reaches at the peak shear strength. After that,
the shear stress slightly decreases such as strain softening. Finally, the shear stress reaches to the
residual stress state. This is a typical mechanical behavior of rock joint under relative low normal
confining condition. Figure 4 shows the peak shear strength and normal confining stress and the
residual state stress and normal confining stress relations. With cyclic using the same specimen
including the joint surface roughness, the Coulomb’s failure criterion is applied to the peak shear
strength and the strength parameters, such as cohesion, c, and internal friction angle, φ, are evaluated.
Consequently, c is 0.305 MPa and φ is 33.4 °. It is observed that, in the constant normal confining
stress of 3.0 MPa, the maximum shear strength is smaller than the Coulomb’s failure criterion. It is
thought that the joint surface roughness becomes smooth and the asperities become weak with the
cyclic shear processes.

Case I, σ n = 0.5 MPa Case I, σ = 0.5 MPa


n

Case II, σ = 1.0 MPa Case II, σn = 1.0 MPa


n

Case III, σn = 0.5 MPa Case III, σn = 0.5 MPa


Case IV, σ dn = 3.0 MPa Case IV, σn = 3.0 MPa

2.5 0.8

2.0 0.6
Shear stress [MPa]

Dilation [mm]

1.5 0.4

1.0 0.2

0.5 0.0

0.0 -0.2
0 0.5 1 1.5 2 2.5 3 0.0 0.50 1.0 1.5 2.0 2.5 3.0
Shear displacement [mm] Shear displacement [mm]
(a) The shear stress and shear displacement relation (b) The dilation and shear displacement relation
Figure 3. The SHS direct shear on the tensile single joint of rock salt

597
3.5

Rssidual shear stress [MPa]


3.0 Maximum shear strength [MPa]

2.5

Shear stress [MPa]


2.0

1.5 φ = 33.4 o
c = 0.31 [MPa]

1.0

0.5

0.0
0 0.5 1 1.5 2 2.5 3 3.5
Normal stress[MPa]

Figure 4. The maximum shear strength and the residual shear stress in each case are plotted against
the constant normal confining stress

Figure 5 shows the enlarged views of SHS process at the residual stress state. In every SHS
processes, the shear stress is reducing in various hold period. And then, the shear stress is increasing in
the process of re-sliding. The shear stress in the process of re-sliding takes over the value at the start
time of the holding process. The shear stress reaches at the peak, and then, it reaches the residual stress
state. In all cases, as the holding period becomes longer, it is confirmed that the decrement of the shear
stress in the holding process is increasing and the increment of the shear stress at the re-sliding process
is increasing. Therefore, it is confirmed that the time dependence of shear strength recovery can be
observed.

3.2 Discussion of the holding time dependency on the shear strength recovery
In this section, the time dependency of the shear strength recovery will be discussed. For the
experimental results, the holding time dependency on the shear strength recovery can be confirmed. In
this section, we will discuss the holding time dependency on the shear strength recovery. According to
previous research works (Dieterich, 1972 & 1994), it is thought that the evolution of the shear strength
recovery occurred due to the change in the coefficient of friction on the rock fracture surface since the
contact area within the rock fracture was increased by the plastic and/or the visco-plastic deformation
of the rocks and the consolidation of the gouge. Then, it was also proposed as the relationship between
the holding time and the shear strength recovery, as follows:
τ s = τ 0 + A log10 th (1)
where τs, τ0, th, and A are the peak shear strength at the post-holding, the shear stress at the pre-holding,
the holding time, and the constant value, respectively. A is the key parameter to estimate the shear
strengthening in considering the effects of both normal stress and material strength of intact part. In
this research work, the shear strength recovery is defined as the difference between τs and τ0. Figure 6
shows the relationship between the shear strength recovery and the holding time. In each case, it is
confirmed that the shear stress recovery increases log-linearly with the holding time. Equation (1)
replicates well the shear strength recovery and the holding time relation. Consequently, a good
correlation can be confirmed in all cases between Equation (1) and the experimental plots. Kishida, et
al. (2011) conducted the SHS direct shear test through the mortal replica sample. They observed the
log-linear correlation between the shear strength recovery and the holding period in the range of less
than 104 sec holding time. On the other hand, they reported that Equation (1) underestimated the
actual measurements of the shear strength recovery for the holding time of more than 104 sec. In this
research work, we cannot observe the above mentioned phenomena.

598
Case I, σ n = 0.5 MPa
Case III, σn = 0.5 MPa Case II, σn = 1.0 MPa

0.55 1.20

0.50 1.10
Shear stress [MPa]

Shear stress [MPa]


1.00
0.45
0.90
0.40
0.80

0.35 0.70

0.30 0.60
1.2 1.6 2 2.4 2.8 1.2 1.6 2 2.4 2.8
Shear displacement [mm] Shear displacement [mm]

(a) Case I and III (b) Case III

Case IV, σ dn = 3.0 MPa

2.4

2.3
Shear stress [MPa]

2.2

2.1

2.0

1.9

1.8
1.2 1.6 2 2.4 2.8
Shear displacement [mm]

(c) Case IV
Figure 5. Enlarged view of the shear stress – shear displacement curves around the shear
displacements from 1.2 to 3.0 mm.

Case I : A = 0.0048
0.30 Case II : A = 0.0085
Shear strength recovery [MPa]

Case III : A = 0.0145


0.25 Case IV : A = 0.0546

0.20

0.15

0.10

0.05

0.00
1 2 3 4 5 6
10 10 10 10 10 10
Hold time, t [sec]
h

Figure 6. Shear strength recovery, τs – τ0, plotted against the log hold time. The log linear regression
lines are evaluated for the shear strength recovery.

599
In Figure 7, Dieterich’s A constant value for the regression lines of Equation (1) is plotted against
the normal confining stress. From this figure, the dependency of the normal confining stress can be
confirmed and constant value A for the wet fracture condition is larger than that for the dry fracture
condition. And, constant value A for the rough fracture condition is also larger than that of the smooth
fracture condition. In Figure 7, Dieterich’s A constant values from Kishida et al.(2011), and tuff
specimen are also plotted.
The specimens in Figure 7 are difference the joint surface roughness. However, the influence of
the joint surface roughness cannot be confirmed in Figure 7. On the other hand, Dieterich’s A constant
values increase with the increment of the normal constant confining stress. The uniaxial compressive
strengths in Figure 7 are almost same. Dieterich’s A constant values are small since the holding period
in this research work is relatively short. Based on the these conditions, it is thought that the strength
recovery are affected by the pressure on the contact area.

Halite (Rock Salt)


Mortal specimen, M-2, Wet (Kishida, et la., 2011)
Mortal specimen, M-3, Dry (Kishida, et la., 2011)
Mortal specimen, M-3, Wet (Kishida, et la., 2011)
Limestone
0.10

0.08
Dieterich's A value

0.06

0.04

0.02

0.00
0 1 2 3 4 5 6 7 8
Normal confining Stress, σn [MPa]

Figure 7. A constant value for Dieterich’s law plotted against the normal confining stress. Confining
stress dependency is observed.

4. CONCLUSIONS
Experimental measurements of SHS direct shear tests on tensial joint of salt rock have allowed the
evolution of the shear strength recovery. The shear strength recovery increases with the increment of
the holding period and is followed by Dieterich’s law.
With discussed our previous results for SHS direct shear test on rock joint, it is confirmed that
Dieterich’s A constant value is depended on the normal confining stress. On the other hand, the
influence of the joint surface roughness is not confirmed in Dieterich’s A constant value.

ACKNOWLEDGEMENT
This work has been supported partially by JSPS Grant-in-Aid for Scientific Research (C) Number
24560602. And, the masurement of joint surface roughness has been strongly supported by Mr.
Sawada and Mr. Sato at JAEA. These supports are gratefully acknowledged.

600
REFERENCES
Dieterich, J.H., 1972, Time-dependent friction in rocks, J of Geophysical Research, 77, pp.3690-3697.
Dieterich, J.H., 1994, Direct observation of frictional contacts; New insights for state-dependent
properties, Pure Applied Geophysics, 243, pp.283-302.
Kishida, K., Kawaguchi, Y., Nakashima, S., Yasuhara, H., 2011, Estimation of shear strength recovery
and permeability of single rock fractures in shear-hold-shear type direct shear tests, IJRMMS, 48;
782 – 793.
Niemeijer A, Marone C, Elsworth D., 2008, Healing of simulation fault gouges aided by pressure
solution: Results from rock analogue experiments. J of Geophysical Research, 113:B04204.
Ohnishi, Y., Yano, T., Iwamoto, H., Nishikawa, N., 2000, The history of the research and
development of the apparatuses measuring both the shear strength and the roughness profile of
rock joints. J Geotech Eng JSCE, 645/III-50:307–20. [in Japanese].
Sato, H. and Sawada, A., 2011, An evaluation of fracture characteristic for replica specimen casted by
granite artificial tensile fracture, JAEA-Research 2011-052, Japan Atomic Energy Agency,. (in
Japanese)
Tenthorey, E., Cox, S.F., Todd, H.F., 2003, Evolution of strength recovery and permeability during
fluid-rock reaction in experimental fault zones. Earth Plannet Sci Letters, 206:161-172.
Yasuhara, H., Marone, C., Elsworth D., 2005, Fault zone restrengthening and frictional healing: The
role of pressure solution. J of Geophysical Research. 110:B6 B06310.

601

You might also like