- Email: [email protected]
The influence of an excavation damaged zone on the thermal-mechanical and hydro-mechanical behaviors of an underground excavation
Engineering Geology 101 (2008) 110–123
Contents lists available at ScienceDirect
Engineering Geology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e n g g e o
The influence of an excavation damaged zone on the thermal-mechanical and hydro-mechanical behaviors of an underground excavation S. Kwon ⁎, W.J. Cho HLW disposal research center, Korea Atomic Energy Research Institute, Daejeon, Republic of Korea
A R T I C L E
I N F O
Article history: Received 22 November 2007 Received in revised form 25 March 2008 Accepted 13 April 2008 Available online 24 April 2008 Keywords: Excavation damaged zone Underground research tunnel Thermo-mechanical coupling Hydro-mechanical coupling Tunnelling Radioactive waste disposal
A B S T R A C T In Korea, a reference disposal system, KRS, was proposed in 2006 after 10 years of research and development. In the KRS, the high-level radioactive waste repository is considered to be located in a crystalline rock likes granite. For a validation of the feasibility, safety, and stability of the KRS, an underground research tunnel, KURT was constructed in Nov. 2006. During the construction of KURT by a controlled blasting, the size and characteristics of an excavation damaged zone(EDZ) were investigated by in situ as well as laboratory tests. The possible influences of an EDZ around a tunnel on the thermal, hydraulic and mechanical behaviors of the near field were investigated by using hydro-mechanical and thermo-mechanical coupling analyses. From this study, it was found that the existence of an EDZ can influence the thermal, hydraulic, and mechanical behaviors of the near field and it was recommended that an EDZ should be considered as an important parameter during the design of underground repositories. © 2008 Elsevier B.V. All rights reserved.
1. Introduction A blasting impact and stress redistribution after an excavation induce an excavation damaged or disturbed zone(EDZ) around an excavation. An investigation into the size and characteristics of this zone is important from safety and stability points of view especially when the construction and operation time of an underground facility is long and its design criteria is rigorous. In an underground radioactive waste repository, which requires a long construction and operation time as well as an extremely long monitoring time after its closure, the mechanism of an EDZ development is one of the most important research topics. In the case of an underground high-level radioactive waste(HLW) repository, which is typically assumed to be located at several hundred meters deep in a rock mass where the in situ stress is high, the development of an EDZ, which changes the thermal, mechanical, hydraulic, and chemical behaviors in the near field, is inevitable. In the international cooperation research project,
Abbreviations: BHTV, BoreHole TeleViewer; BIPS, Borehole Image Processing System; CANDU, Canadian Deuterium Uranium; DECOVALEX, DEvelopment of COupled models and their VALidation against EXperiments; EDZ, Excavation damaged zone; HLW, high-level radioactive waste; KAERI, Korea Atomic Energy Research Institute; KRS, Korea reference disposal system; KURT, KAERI Underground Research Tunnel; LILW, low and intermediate level waste; PWR, Pressurized Water Reactors; THMC, ThermoHydro-Mechanical-Chemical; URL, Underground Research Laboratory; Watt/tHM, Watt per tons of heavy metal. ⁎ Corresponding author. E-mail address: [email protected] (S. Kwon). 0013-7952/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.enggeo.2008.04.004
DECOVALEX, which was started in 1992, an EDZ was considered as an important parameter for the Thermo-Hydro-Mechanical-Chemical (THMC) coupling analysis. In the last phase of the project, DECOVALEX-THMC, three of the following five tasks were related to an EDZ (Hudson and Jing, 2007). Task 1: Influence of a near-field THM and PA Task 2: THMC studies of an EDZ Task 3: EDZ in argillaceous rocks Task 4: Property change in an EDZ and the near field due to THC and THM processes for volcanic and crystalline rocks Task 5: Long-term climate change The development of an EDZ is dependent on many parameters such as the excavation method, tunnel geometry, blasting technique, and the rock and the in situ stress conditions. To investigate the size of an EDZ and the characteristics of it, in situ tests have been carried out in many URLs including the Aspo Hard Rock Laboratory and Stripa mine in Sweden, the Kamaish mines in Japan, the URL in Canada, the Mol in Belgium, the Tournemire in France, the WIPP in USA, the Olkiluoto research tunnel in Finland, and the Mont Terri in Switzerland (Dale and Hurtado, 1996; Backblom and Martin, 1999; Sato et al., 2000; NEA/RWM, 2002; Bossart et al., 2002; Cai and Kaiser, 2005). According to Backblom and Martin (1999), an EDZ might not be crucial for the overall safety of a repository due to the effectiveness of engineered barriers and the self-sealing and self-healing capabilities of an EDZ in claystone or salt rock; however, the results are very dependent on which nuclide is studied, what assumptions are made,
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
and what type of models are used for the engineered barrier and the rock. They also concluded that excluding an EDZ may result in nonconservative maximum radiation dose estimations and thus the relevant performance assessments should be more explicit and transparent with a consideration of an EDZ. In Korea, a Korean reference disposal system(KRS) was developed for the permanent disposal of spent fuels in a deep geological formation(Lee et al., 2006). In order to validate the feasibility, safety, stability, and reliability of this system, an underground research tunnel, KAERI Underground Research Tunnel(KURT), was constructed as a generic URL in 2006. The characteristics of an EDZ around a tunnel were investigated with various laboratory as well as in situ tests. In this study, the influence of an EDZ on the mechanical, hydraulic, and thermal behaviors around the KURT and the conceptual repository design adapting the KRS was investigated based on three-dimensional computer simulations and the rock properties measured from the laboratory and in situ tests.
111
2. Introduction to the KRS and the KURT 2.1. Korean reference disposal system In Korea, it is an urgent issue to develop a reasonable management system for the spent fuels being generated from the 20 nuclear power plants currently operating at 4 reactor sites. The amount of the spent fuels accumulated and stored at the reactor sites was more than 8600 tons in 2007. It will be increased to 36 000 tons in 2040, if the currently operating plants and those under construction and planned for are operated for their expected life times, which ranges from 30 to 60 years. In order to develop the KRS with a consideration of the characteristics of the spent fuels and the geological conditions in the Korean peninsula, a long-term R&D program has been carried out since 1997. According to the KRS, an underground HLW repository is supposed to be located in a crystalline rock mass like granite at about 500 m below the surface. The canisters containing Pressurized Water
Fig. 1. Disposal concept being considered for the Korean reference disposal system.
112
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Table 1 Major components for the Korea reference disposal system Components Spent fuel
Canister
Buffer
Backfill Rock
Disposal tunnel
Deposition hole
Characteristics Type Amount Disposal rate Heat generation Cooling time Diameter Length Number
PWR 20 000 tU 380 canister/yr 1540 W 40 years 1.02 m 4.83 m 11375
CANDU 16 000 tU 146 canisters/yr 760 W 30 years 1.02 m 4.83 m 2835
Compacted bentonite Dry density: 1800 kg/m3 Thickness: 0.5 m Maximum temperature b 100 °C Crushed rock 70% + bentonite 30% Grain size b 22 mm Crystalline rock (granite) Depth 500 m Thermal gradient : 30 °C/km Drill and blasting Tunnel spacing : 40 m Underground area : 8 km2 Tunnel number : 700 tunnels Width 5 m, height 6.15 m, length 251 m Vertical emplacement Total number : 11944 holes Spacing : PWR 6 m, CANDU 4 m Diameter :2.02 m
Reactors (PWR) and Canadian Deuterium Uranium(CANDU) spent fuels are assumed to be emplaced in vertical boreholes drilled with a spacing of 6 m and 4 m, respectively, in the floor. Because of the different characteristics of these two spent fuels, it is recommended to dispose of them in separate areas. The migration of radionuclides from the disposed high-level waste will be delayed by natural barriers as well as engineered barriers including a waste form, canister, buffer, and backfill. Compacted bentonite blocks will be installed around a
canister in a deposition hole as a buffer. A mixture of bentonite and crushed rock is being considered as the backfilling material. The dimensions and material type of the canisters for the two types of spent fuels are designed to be exactly identical to make their encapsulation and handling processes in the repository simple. The principal design concept and major parts of the KRS were reported in Lee et al. (2006). Fig. 1 shows the underground repository concept being considered for the KRS. Table 1 is a brief description of the important parts of the KRS. Cooling time in Table 1 means the time for a temporary storage in the pools or dry storage at the reactor sites. 2.2. KAERI Underground Research Tunnel (KURT) In order to validate the suggested disposal system, the KRS, a small scale underground research tunnel, KURT, was constructed in Daejeon, which is located about 150-km south of Seoul, in 2006. The KURT portal is located on a hill-side about 110 m above sea level (35°25′ N, 127°22′ E) . Fig. 2 shows the location and layout of the KURT. At the end of the access tunnel, the overburden is about 90 m. For the first 40 m of the access tunnel, 0.3 m thick concrete lining was installed to support the tunnel in the weak rock which had suffered from a weathering. The other areas were supported mainly by shotcrete and rock bolts. The geological conditions of the site had been investigated in 2003. The main rock type in the area is granite, which is the host rock type for the low and intermediate level waste(LILW) repository in the Gyung-Ju area and a candidate host rock type for an HLW repository. During the site characterization, various rock and rock mass properties were determined from laboratory and in situ tests. Two boreholes were drilled at the tunnel portal area: (a) one is a 160 m long vertical borehole; and (b) the other is a 250 m long declined borehole with the same direction and slope as the access tunnel. In the boreholes, in situ tests including hydraulic fracturing tests, a hydraulic tests, BHTV/BIPS, borehole shear tests, and Goodman jack tests were carried out. Hydraulic fracturing and impression packer tests were implemented at 7 locations ranging from 50 m to 155 in the vertical borehole. Total
Fig. 2. Schematic drawing of the KURT.
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
113
Fig. 3. Variation of the hydraulic conductivities along the 10% declined borehole.
of 36 hydraulic tests were carried out with a 10 m interval using a double packer method to determine the hydraulic conductivities along the boreholes. At first an injection test was carried out at a pressure of 4 kgf/cm2 for 20 min. After the injection test, a fall-off test was followed to check on the pressure drop under the condition of noinjection for 20 min. Fig. 3 shows the results from the hydraulic pressure tests in the declined borehole. Hydraulic conductivities were usually lower than 10− 6 m/s except for 180 m, where it was 10− 5 m/s (Kwon et al., 2004). There is a tendency for decreasing hydraulic conductivities with an increasing depth. This can be explained by the improvement of the rock quality, which was confirmed by the increase of the RQD and RMR with the depth (Kwon et al., 2006). Table 2 lists the properties from the laboratory and in situ borehole tests. At a shallow depth, the rock strength and modulus are lower than those at a deeper location due to a weathering at a shallow depth. In the cases of friction angle and cohesion, however, it was not possible to derive any depth effect, because the triaxial compression tests were only carried out for the rock cores from deep locations (Kwon et al., 2004). The construction of the KURT was started in May, 2005. The total tunnel length of the KURT with an access tunnel and two research modules is 255 m. A 10% downward slope and a tunnel direction toward the top of the mountain were chosen to effectively achieve the tunnel depth with a short access tunnel and to locate the research modules, in which the major in situ experiments will be carried out, in relatively good geological conditions (Kwon et al., 2006). A controlled blasting technique was applied with Emulite and Kinex, which was used for the production and perimeter holes, respectively. About 90 blast holes with a 38 mm diameter each were drilled by a two-boom jumbo for each blast round. Every day one blasting round was completed and a 1 m–3 m advance could be achieved depending on the rock condition and the blasting design. During the construction of the KURT, a fracture mapping was carried out after a blasting. In the fracture mapping, the dip and dip direction, the fracture roughness, fracture filling materials, and the water inflow of the fractures longer than 2 m were recorded. In several locations, fractures zones were found around faults or dykes at 45 m, 67 m, 112 m, 115 m, 120 m, and 138 m from the tunnel entrance (Kim et al., 2006). Fig. 4 shows the total inflow recorded during and after the construction of the tunnel. After the construction of research module 2, which crosses water conducting fractures, and the drilling of the 20–50 m long boreholes in the floor and wall for research purposes, a significant amount of inflow was recorded. In contrast, no detectable water inflow was recorded during the construction of research module 1, which was constructed in a dry conditioned rock without
any water conducting fractures. The inflow was quickly stabilized with the installation of packers at the boreholes. During the construction of the remaining parts of research module 2 and research module 1, the amount of inflow was maintained at around 20–25 tons/day. The water is collected at a sump pit installed at the end of the access tunnel and pumping out to the surface automatically. Table 3 lists the measured rock properties in the investigated ranges divided by the major fracture zones and fault zones(Kwon et al., 2006). The rock mass properties were determined from RMR and Hoek–Brown's empirical equations (Hoek et al., 2002). Fig. 5 shows the geological conditions along the tunnel and the EDZ study area. Discontinuities with dip directions similar to the tunnel direction could not be included. Only the major discontinuities with a dip direction more or less perpendicular to the tunnel direction were plotted in Fig. 5. 3. EDZ size and properties In situ tests to investigate the characteristics of an EDZ at the KURT were carried out after the construction of a 8 m long turning shelter at about 60 m from the tunnel entrance. Seven boreholes were drilled into the tunnel wall for the borehole tests before the excavation of the access tunnel as shown in Fig. 6. The construction of the access tunnel
Table 2 Rock properties from the laboratory and in situ tests using the declined borehole Properties
Unit
Number of tests Average
Depth effect
Young's modulus Deformation modulus Poisson's ratio Density UCS Tensile strength Friction angle of rock Cohesion of rock Dynamic elastic modulus Max. horizontal stress Min. horizontal stress In situ stress ratio K Hydraulic conductivity RQD RMR
GPa GPa
12 3 12 12 60 14 6 6 12
26.5 5.04 0.23 2623 61.65 9.5 49 17.6 52
Ed = 0.1332Z + 0.38 – – UCS = 135.02 − 779.6/Z0.5 St = 0.0283Z + 5.8218 – – –
MPa MPa
7
m/s
22 252 m 252 m
4.6 5.6 1.98 4.26e− 7 64.5 59
0.0201Z⁎ + 3.505 0.0174Z⁎ + 2.7455 K = 22.607Z⁎− 0.5882 ln(k) = −14.59 − 0.017Z RQD = 94.984 − 280.3/Z0.5 RMR = 17.21lnZ − 20.18
kg/m3 MPa MPa MPa GPa
Z is the distance along the declined borehole. Z* is the depth in the vertical borehole.
(continued on next page)
114
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Fig. 4. Inflow change with the tunnel excavation.
was restarted and the response of the rock mass was measured. Two more boreholes were drilled after the construction of the access tunnel to measure the property changes due to a blasting impact. The rock cores obtained from the drilling before and after a blasting were used for the rock core observation and laboratory tests. From the in situ and laboratory tests, the following conclusions could be drawn: 1) According to the RQDs of the rock cores from the EDZ study area, the average RQD in the 0–2 m range, where the blasting impact is significant, was 67, while it was 81 at a deeper range. This means the RQD was decreased by about 17% by the blasting(Kwon and Cho, 2007). 2) From the Goodman jack test before and after a blasting, it was possible to observe that the deformation modulus after a blasting was lower than that before a blasting. The influence of a blasting could even be found at 1.5 m from the tunnel wall. Fig. 7 shows the result from the Goodman jack test in borehole 8, which was drilled after the excavation of the access tunnel. The horizontal as well as vertical deformation moduli were measured at 0.3 m intervals. The distance from the wall, Z, was calculated with a consideration for the direction of a borehole. It was possible to observe that the deformation modulus was more or less consistent to 1.5 m from the wall and then it was increased with the depth. The average vertical deformation modulus, 16 GPa, was about 25% higher than the horizontal deformation modulus, 12 GPa . This might be related to the characteristics of the fracture network in the study area. With more measurements and further studies on the fracture network, we will be able to explain the difference. 3) From the laboratory tests, the EDZ size could be estimated to be around 1.1–1.5 m. Fig. 8(a) shows the general trend for the variation of the rock properties with the distance from the tunnel wall. As shown in Fig. 8 (b), Young's modulus before and after an excavation change with the distance from the wall which is similar to the general trend. The results from the laboratory tests are summarized in Table 4. The elastic modulus and rock strength in the EDZ were decreased by about 50% and 15%, respectively(Kwon and Cho., 2007) . 4. Hydro-mechanical analysis with the EDZ 4.1. Model mesh To investigate the effect of an EDZ around the KURT, a hydromechanical coupling analysis was carried out using a three-dimen-
sional FDM code, FLAC3D. Because a groundwater mainly flows through joints, continuum models like FLAC3D have disadvantages for describing the hydraulic processes on a small scale. On a large scale, however, continuum models are useful for investigating the general hydraulic processes and for assessing different scenarios from a sensitivity analysis point of view. Similarly, a rock could be assumed to have isotropic and homogeneous mechanical properties. In this study, the strengths and deformation modulus determined by using the empirical equations suggested by Hoek et al. (2002) were used. The model mesh shown in Fig. 9 was developed in order to consider various factors, which can influence the mechanical and hydraulic behaviors of the KURT. The width, length, and height of the initial model mesh were 100 m, 250 m, and 300 m, respectively. Total number of zones in the model mesh was 27250. Factors considered in the modeling were: − Surface topography: The actual surface topography above the tunnel was implemented in the model as shown in Fig. 9. − Rock property change along the tunnel: The mechanical properties at different ranges divided by the major fracture zones were also inputted in the model as shown in Fig. 10(a). Since the major fracture zones were more or less vertical, 10 m thick vertical layers were inputted as shown in Fig. 10(b) for the hydraulic conductivities, which were measured at 10 m intervals in the declined borehole. − In situ stress: According to the hydraulic fracturing tests carried out in a vertical borehole near the tunnel portal, the maximum horizontal stress was in a direction of 96.5° from the north and the
Table 3 Rock mass properties in the investigated ranges along the declined borehole Parameters
Interval (m) Dip of the range UCS (MPa) Em (GPa) Bulk modulus (GPa) Shear modulus (GPa) Cohesion (MPa) Friction angle Tensile strength (MPa)
Range(m) 30–82
82–125
125–177
177–192
192–252
52.9 88 30.8 3.2 2.13 1.28 1.7 35 0.014
42.6 68–88 58.4 12.1 8.07 4.84 4.2 40 0.102
51 68–90 71.6 37.9 25.3 15.16 6.7 46 0.49
15.2 90 77.6 6.21 4.14 2.48 4.6 36 0.047
60.1 90 82.7 20.4 13.6 8.16 6.4 42 0.23
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
115
Fig. 5. KURT layout and the geological conditions around the EDZ study area.
in situ stress ratio was decreased with the depth as listed in Table 2. In order to simulate the actual in situ stress variation with the depth, the model meshes above the current surface topography were removed step by step as shown in Fig. 9. Different initial stress ratios (K) in the initial flat mesh were used to simulate the measured in situ stress condition and finally K = 1 was chosen for the 300 m thick model mesh. The displacements developed during each step were initialized, but the stress and the plastic zone developed during this phase were stored for the following excavation phase. With this approach, it was possible to realize a measured stress ratio distribution, in which the horizontal stress was higher than the vertical stress (Kwon et al., 2006). − EDZ : It was assumed that a 1–2 m thick EDZ was developed immediately after an excavation. An EDZ in a floor is normally larger than that in a roof or a wall because of a heavy charge in a floor (Autio, 2003). From Fig. 11, which shows the RQD variation with the depth for the wall and the floor, it is possible to see that the damaged zone in the floor is larger than that in the wall. The
average RQD values were calculated from the rock cores from 25 boreholes in the KURT. In this study, therefore, the size of an EDZ in the floor was modeled to be larger than those in the roof and the wall. − Tunnel geometry: The horse shoe shape tunnel size of 6m × 6 m was implemented in the model. − Tunnel slope: In the model, the tunnel had a −10% slope, the same as the access tunnel in the KURT. − Sequential excavation : In the model, the tunnel was excavated with a 10 m advance in order to consider the influence of a sequential excavation. 4.2. Initial and boundary conditions In the model, zero displacement and no-flow boundaries were applied at the bottom and the four sides. The mechanical properties of an EDZ were adjusted based on the laboratory rock core tests. The bulk modulus and shear modulus were reduced by a half of the original rock properties as listed in Table 5. The cohesion and tensile strength in an EDZ were
Fig. 6. Boreholes drilled for the in situ EDZ study at KURT.
116
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Fig. 7. Deformation modulus change with the distance from the wall.
adjusted to 80% of the original values. The hydraulic conductivity of an EDZ was assumed to be one tenth of the original value(Table 6). For the hydraulic analysis, it was assumed that the rock was fully saturated at about 20 m above the tunnel entrance and the pore pressures at the boundaries below the water table were fixed to simulate the horizontally maintained water table. The initial saturation above the water table was assumed to be 0.5 in the calculation. 4.3. Hydro-mechanical calculations Fig. 12 shows the pore pressure contour and the flow vector after the excavation of the access tunnel. It is possible to observe that the distribution of the pore pressure and the flow vectors in the zones are different along the tunnel length. A relatively higher pore pressure and lower flow were calculated in the lower permeable zones. Because of the assumption that the rock above the water table is partially saturated, the flow vectors could reach the surface. Fig. 13 shows the calculated total inflow across the entire tunnel with a tunnel advance for the cases with and without an EDZ. After an excavation, the inflow suddenly increases and then it decreases smoothly to a stable value with time. The peak inflow at each excavation step depends on the hydraulic conductivity of the excavated range. As expected, the excavation of a higher permeable zone results a greater inflow. It is possible to see that the inflow increases steadily with the tunnel advances. Without an EDZ, the daily inflow, after a complete excavation of the access tunnel, was calculated to be about 6 tons. The inflow with the consideration of an EDZ is about 20% larger than that from the case without an EDZ. Even with a consideration of an EDZ, the inflow is much lower than the measured inflow of 25 tons/day. The lower inflow rate from the calculation might be due to the tunnel length difference and the assumptions of a hydraulic conductivity in an EDZ. In this study, the hydraulic conductivity of an EDZ is one order of a magnitude higher than the original value. Because the hydraulic conductivity of an EDZ can be increased by more than 2 orders of a magnitude(Sato et al., 2000), the difference with regards to the measured inflow rate can be
decreased with a higher conductivity in an EDZ. With an accurate measurement of hydraulic properties in an EDZ and a sensitivity analysis with varying mechanical and hydraulic properties in the future study, a clearer explanation of an EDZ effect will be possible. Transient hydraulic test is also required to investigate a rock diffusivity, which effects the stabilization of a flow with time. As shown in Fig. 14, the elastic and plastic displacements around the tunnel after the excavation of the access tunnel were increased by about 50% with the consideration of an EDZ in the model. The maximum displacement of about 8 mm was recorded in the wall around the tunnel portal. From Fig. 14, it is possible to observe that a displacement along the tunnel is largely dependent on a variation of the deformation modulus for the rock mass. When the bulk modulus is over 10 GPa, the displacements around the tunnel were less than 2 mm, even with an EDZ. Such a small deformation, in the order of 0.1%, for the 6 m tunnel diameter is related to the adjustment of the rock properties in an EDZ based on a rock core test. For a more realistic modeling, therefore, it is required to carry various in situ tests to measure the mechanical properties. 5. Thermo-mechanical analysis for the reference disposal system 5.1. Model mesh, initial, and boundary conditions For the conceptual design of the KRS, a thermo-mechanical coupling analysis was carried out with the model mesh in Fig.15. An underground repository with a deposition hole spacing of 6 m and tunnel spacing of 40 m could be modeled by a quarter model with two vertical symmetry planes on the X and Y axes. For a detailed analysis, the disposal tunnel, fuel part, outshell, backfill, buffer, EDZ, and rock were included in the model. The properties of the fuel part, which represents the inside of the outshell, were determined with the assumption that the fuel and cast insert were uniformly mixed. Important buffer and backfill properties could be derived from the laboratory tests as well as the literature review undertaken by Cho et al. (2001). The material properties of them are summarized in Tables 7 and 8.
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
117
Fig. 8. General trend of a rock property change before and after a blasting.
The rock properties were determined using a 500 m deep borehole in the Goseong area in the north-eastern part of the Korean peninsula (Park et al., 2001). Hydraulic fracturing tests were carried out at 10 locations with 50 m intervals. At a shallow depth from the surface to 200 m, the Table 4 Representative results of the laboratory tests Distance (m)
Uniaxial compressive strength P wave velocity S wave velocity Young's modulus Triaxial compressive strength Brazilian tensile strength Porosity
Property change in Zone 1 (%)
D1
D2
0.7 0.7 0.9 0.7 0.9 0.7 More than 1.5
0.9 1.5 1.5 1.5 1.1 1.1 –
−14 −17 −12 −57 −5 to −11 −13 108
horizontal stresses are much higher than the vertical stress. At the locations deeper than 200 m, the average in situ stress ratio was almost 1.0 (Park et al., 2002). The rock properties for an EDZ were adjusted to be similar to the previous hydro-mechanical analysis. The mechanical and thermal properties including the bulk modulus, shear modulus, cohesion, tensile strength, and thermal expansion coefficient for the EDZ were reduced by 50% from those of the intact rock. It was assumed that the thermal conductivity of an EDZ was increased by 100%, because a higher permeability in an EDZ can allow for a greater groundwater movement, which results in a greater heat transfer from a canister to a rock mass. In the cases of the friction angle and the specific heat, the same values as the intact rock were used. The in situ stresses at the repository depth, 500 m, were assumed to be hydrostatic based on the hydraulic fracturing test in the Goseong area. The surface temperature was assumed to be 20 °C and a geothermal gradient of 3 °C/100 m was used. For the PWR spent fuel generated from the Korean
118
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Fig. 9. Model mesh for the mechanical-hydraulic coupling analysis for the KURT.
nuclear power plants, its decay heat can be calculated using the following equation; 1
P ðt Þ ¼ 852:34e 0:2642þ0:130889t ðWatt=tHMÞ 1btb30 years P ðt Þ ¼ 14548 t 0:76204 ðWatt=tHMÞ tN30 years where, t is the elapsed time (year) after extracting the spent fuel from a reactor. Fig. 16 shows the decay heat change with time after its discharge from the reactors. Subsequent heating of a rock mass by a heat-generating waste will increase the stresses in a buffer, canister, and rock mass because of a thermal expansion. Because of the higher thermal expansion of water than that of rock, the heat from the canister will also increase the pore pressure in the rock. This will lead to higher hydraulic gradients around the deposition hole and influence on the hydraulic behavior in the near field. 5.2. Influence of an EDZ on the thermo-mechanical results Two cases with and without an EDZ were modeled for the disposal tunnel and the deposition hole containing PWR spent fuel for 50 years.
It is normal to simulate much longer computation span in radioactive waste disposal projects to investigate the long-term complex THMC coupling behavior. In this study, however, 50 years of computation span was enough to check the peak temperature, which was used for the evaluation of the influence of EDZ. The decay heat from the spent fuel was calculated with a consideration of a 40 years cooling after the discharge from the reactors. Without a consideration of an EDZ, the peak temperature at the contact surface between the canister and the buffer was 92 °C as shown in Fig. 17. It was decreased to 89 °C with the consideration of an EDZ around the tunnel and borehole. When bentonite is used for the buffer material, the maximum temperature of the buffer is recommended to be lower than 100 °C in order to avoid a non-boiling condition and a mineralogical alternation of the bentonite, which can be occurred with the hydrothermal reaction between the bentonite and groundwater at higher temperature. Since it is usual to design a deposition hole and a tunnel spacing with the thermal criteria for a buffer, the meaning of the 3 °C difference for the maximum buffer temperatures with and without an EDZ is significant from a repository design point of view.
Fig. 10. Assignment of the mechanical and hydraulic properties.
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
119
Fig. 11. RQD variation with the depth for the wall and the floor.
Kwon et al. (2003) carried out a sensitivity analysis of the near field during the development of a disposal system. In the sensitivity analysis, seven parameters including the tunnel size, buffer and backfill thickness, deposition hole and tunnel spacing were considered. The buffer temperature for 50 years after an emplacement was compared for these cases and the main effect of each parameter could be determined as shown in Table 9. Negative main effect of a parameter means the buffer temperature decreases with an increase of this parameter. In the case of the tunnel spacing, the main effect was −4.3. This means the buffer temperature can be decreased by only 4.3 °C when the tunnel spacing is increased from 36 m to 44 m. Compared to the influence of the parameters on the buffer temperature listed in Table 9, a temperature change of 3 °C with the consideration of an EDZ is significant. Because of the possible influence of an EDZ on the temperature distribution around an underground disposal area, it is highly recommended to carry out an in situ long-term heater test to evaluate the thermal properties in an EDZ and to reduce the uncertainty of a modelling. The development of an EDZ after an excavation reduced the stress concentration around the excavation as shown in Fig. 18. Because of the lower modulus and strengths of the EDZ than those of the intact rock, the maximum principal stresses are calculated to be a half of those from the case without the consideration of an EDZ. Such a stress reduction with the consideration of an EDZ can affect the mechanical and hydraulic behavior of the rock mass around an underground repository by changing the confinement pressure on the discontinuities. From the thermal and mechanical points of view, therefore, it is highly recommended to include the characteristics of an EDZ as an important parameter for an HLW repository design and safety
1) The groundwater inflow into the KURT was increased by about 20% with the consideration of an EDZ, in which the hydraulic conductivity was assumed to be one order of a magnitude higher than the original rock mass. 2) The excavation of a higher permeable zone resulted in a greater inflow into the KURT. The inflow immediately after the excavation increased sharply and decreased with time to a stable value.
Table 5 Mechanical properties in the EDZ
Table 6 Hydraulic properties in the EDZ
Range (m)
Bulk modulus (GPa)
Shear modulus (GPa)
Cohesion (MPa)
Friction angle (deg.)
Tensile strength (MPa)
40–80 70–130 130–180
1.07 4.04 12.65
0.64 2.42 7.58
1.36 3.36 5.36
35 40 46
0.011 0.08 0.098
analysis. Also, various in situ and laboratory tests should be undertaken for a better characterization of an EDZ in different geological and blasting conditions. 6. Conclusions For the safe management of an underground facility, an understanding of an EDZ effects is important especially when the construction and operation time is long as an underground radioactive waste repository. In Korea, a reference disposal system, KRS, was proposed in 2006 and a small scale underground research tunnel, KURT, for a validation of the KRS was constructed in Nov., 2006. During the construction of the KURT, the characteristics of an EDZ were investigated from in situ as well as laboratory tests. Based on the in situ and laboratory tests, the size of an EDZ around the KURT could be estimated to be 1.5 m from the tunnel wall. The elastic modulus and deformation modulus in an EDZ was decreased by about 56% and 40%, respectively, when compared to the original rock properties. In order to evaluate the influence of an EDZ on the thermal, hydraulic and mechanical behaviors, hydro-mechanical and thermomechanical coupling analyses were carried out using FLAC3D. From this study, the following conclusions could be drawn:
Range (m) Hydraulic conductivity Range (m) Hydraulic conductivity
40–50 7.15e-6
50–60 7.15e-6
60–70 1.75e-6
70–80 3.42e-6
80–90 3.06e-6
90–100 3e-7
100–110 1.05e-7
110–120 120–130 130–140 140–150 150–160 160–170 170–180 2.44e-6 5.19e-6 8.54e-9 2e-7 8.18e-9 1.86e-7 9.24e-8
120
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Fig. 12. Pore pressure contour and flow direction after the excavation.
3) After the completion of the excavation for the KURT, a relatively higher pore pressure and lower flow rate were observed in the lower permeable zones. 4) When the properties of an EDZ determined from laboratory tests were applied, the displacement around the KURT was increased by about 50% when compared to the case without the consideration of an EDZ. 5) The peak temperature at the canister and buffer contact in the KRS was increased by 3 °C with the consideration of an EDZ, in
which the thermal conductivity was assumed to be 1 order of a magnitude higher than that of the intact rock. Since such an influence is significant when compared to the influence of the repository design parameters considered in a sensitivity analysis, it is highly recommended to carry out in situ tests to evaluate the thermal properties in an EDZ, and to reduce the uncertainty of a modeling. 6) When the mechanical properties including the bulk modulus, shear modulus, cohesion, tensile strength, and thermal expansion
Fig. 13. Variation of total inflow with a sequential excavation.
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
121
Fig. 14. Variation of the displacements in the roof, floor, and wall after the excavation.
coefficient of an EDZ were reduced by 50% from those of the intact rock, the maximum principal stresses in the roof and wall were significantly decreased. From the study, it was possible to observe the influence of an EDZ on the thermal, hydraulic, and mechanical behaviors of underground excavations. In a future study, therefore, it is highly recommended to measure the thermal, hydraulic and mechanical properties of an EDZ carefully for a better understanding of the EDZ effects. With more in situ tests likes an in situ heater test, transient hydraulic tests, and Goodman jack tests, the influence of an EDZ on the thermal, hydraulic, and mechanical behaviors of underground facilities will be evaluated more precisely.
References Autio, J., 2003. The effect of the excavation damaged zone on the migration of radionuclides. The 2002 International EDZ workshop, Toronto, p. 7. paper. Backblom, G., Martin, C.D., 1999. Recent experiments in hard rocks to study the excavation response: implications for the performance of a nuclear waste geological repository. Tunnelling and Underground Space Technology 14, 377–394. Bossart, P., Peter, M.M., Moeri, A., Trick, T., Mayer, J., 2002. Geological and hydraulic characterization of the excavation disturbed zone in the Opalinus Clay of the Mont Terri Rock Laboratory. Engineering Geology 66, 19–38. Cai, M., Kaiser, P.K., 2005. Assessment of excavation damaged zone using a micromechanics model. Tunnelling and Underground Space Technology 20, 301–310. Cho, W.J., Lee, J.W., Kang, C.H., 2001. A Compilation and Evaluation of Thermal and Mechanical Properties of Compacted Bentonite for the Performance Assessment of Engineered Barriers in the High-level Waste Repository, KAERI/TR-1826/2001, KAERI. Table 7 Material properties of the fuel part and the outshell Material type
Fuel + cast steel
Model type Young's modulus Poisson's ratio Density Thermal conductivity Specific heat Thermal expansion
GPa kg/m3 W/m °K J/kg °K 1/°K
Stainless steel
Elastic model
Elastic model
190 0.3 6500 43 424 1.2e−5
200 0.3 8000 15.2 504 8.2e−6
Table 8 Material properties of the rock, buffer and backfill Material type
Modulus
Fig. 15. Model mesh for a Thermal-Mechanical coupling analysis for the Korean reference disposal concept.
Unit
GPa
Rock
Buffer
Backfill
Granite
Compacted bentonite
Bentonite 70% + Sand 30%
Bulk = 37.04 Shear = 21.65 2650
Bulk = 0.038 Shear = 0.029 Dry 1800 Wet 2100 2.04
Density
kg/m
Thermal conductivity Specific heat Thermal expansion UCS Cohesion Friction angle
W/m oK
2.523
Bulk = 0.345 Shear = 0.258 Dry 1800 Wet 2100 1.47
J/kg oK 1/oK
1 576 1.92e−5
888 3.1e−4
900 3.1e−4
MPa MPa Degree
22.5 61
7.66 1.1 50
0.93 1.1 17
3
122
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Fig. 16. Decay heat change after the discharge of PWR spent fuels.
Dale, T., Hurtado, L.D.,1996. WIPP Air-Intake Shaft Disturbed Rock Zone Study. SAND96-1327C. Hoek, E., Carranza-Torres, C., Corkum, B., 2002. Hoek–Brown failure criterion-2002 edition. Proceedings of the Fifth North American Rock Mechanics Symposium, Toronto, Canada, pp. 267–273. Hudson, J.A., Jing, L., 2007. DECOVALEX-THMC, Task B, Understanding and characterizing the excavation disturbed zone(EDZ). SKI report 2007, p. 08. Kim, K.S., Cho, S.I., Lee, J.H., Lim, W.M., Ryu, S.W., 2006. Fracture network model of KURT rock mass. Abstracts of Proceedings of the Korean Radioactive waste Society, Jeju, Korea, pp. 230–231. Kwon, S., Park, J.H., Choi, J.W., Kang, C.H., 2003. Thermo-mechanical sensitivity analysis of repository design parameters using Korean geological conditions. KAERI report, KAERI/TR-2360/2003. Kwon, S., Park, J.H., Cho, W.J., 2004. Geotechnical characteristics of the site for an underground research tunnel in KAERI, KAERI report. KAERI/TR-2805/2004. Kwon, S., Cho, W.J., Han, P.S., 2006. Concept development of an underground research tunnel for validating the Korean reference HLW disposal system. Tunnelling and Underground Space Technology 21, 203–217.
Kwon, S., Cho, W.J., 2007. Investigation of excavation disturbed zone around a tunnel by blasting. Explosives & Blasting 25, 15–29. Lee, J.Y., Cho, D.K., Kim, S.G., Choi, H.J., Choi, J.W., Hahn, P.S., 2006. Development of the Korean reference vertical disposal system concept for spent fuels. Waste Management'06, Tucson, AZ. Park, B.Y., Bae, D.S., Kim, C.S., Kim, K.S., Ko, Y.K., 2001. Evaluation of the basic mechanical and thermal properties of deep crystalline rocks. KAERI report. KAERI/TR-1828/ 2001. Park, B.Y., Bae, D.S., Kim, C.S., Kim, K.S., Ko, Y.K., Won, K.S., 2002. Measuring the initial earth pressure of granite using hydraulic fracturing test: Goseong and Yuseong areas. KAERI report. KAERI/TR-2038/2002. NEA, 2002. Characterization and Representation of the Excavation Disturbed or Damaged Zone(EDZ). NEA/RWM(2002). Sato, T., Kikuchi, T., Sugihara, K., 2000. In-situ experiments on an excavation disturbed zone induced by mechanical excavation in Neogene sedimentary rock at Tono mine, central Japan. Engineering Geology 56, 97–108.
Fig. 17. Influence of an EDZ on the near-field temperatures.
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123 Table 9 Main effects of the parameters on the buffer temperature for 50 years after an emplacement
Max. value Min. value Average temp. with max. value Average temp. with min. value Main effect
Tunnel width
Tunnel height
Tunnel spacing
Hole spacing
Sidebuffer
Upper buffer
Backfill
6.6 m 5.4 m 79.5 °C
7.7 m 6.3 m 79.5 °C
44 m 36 m 77.3 °C
6.6 m 5.4 m 76.2 °C
0.55 m 1.65 m 0.45 m 1.35 m 79.8 °C 79.4 °C
1.1 m 0.9 m 79.5 °C
79.4 °C
79.4 °C
81.6 °C
82.7 °C
79.1 °C
79.5 °C
79.5 °C
0.06
0.06
−4.32
−6.52
0.68
− 0.1
−0.02
Fig. 18. Influence of an EDZ on the near field maximum principal stresses.
123
Contents lists available at ScienceDirect
Engineering Geology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e n g g e o
The influence of an excavation damaged zone on the thermal-mechanical and hydro-mechanical behaviors of an underground excavation S. Kwon ⁎, W.J. Cho HLW disposal research center, Korea Atomic Energy Research Institute, Daejeon, Republic of Korea
A R T I C L E
I N F O
Article history: Received 22 November 2007 Received in revised form 25 March 2008 Accepted 13 April 2008 Available online 24 April 2008 Keywords: Excavation damaged zone Underground research tunnel Thermo-mechanical coupling Hydro-mechanical coupling Tunnelling Radioactive waste disposal
A B S T R A C T In Korea, a reference disposal system, KRS, was proposed in 2006 after 10 years of research and development. In the KRS, the high-level radioactive waste repository is considered to be located in a crystalline rock likes granite. For a validation of the feasibility, safety, and stability of the KRS, an underground research tunnel, KURT was constructed in Nov. 2006. During the construction of KURT by a controlled blasting, the size and characteristics of an excavation damaged zone(EDZ) were investigated by in situ as well as laboratory tests. The possible influences of an EDZ around a tunnel on the thermal, hydraulic and mechanical behaviors of the near field were investigated by using hydro-mechanical and thermo-mechanical coupling analyses. From this study, it was found that the existence of an EDZ can influence the thermal, hydraulic, and mechanical behaviors of the near field and it was recommended that an EDZ should be considered as an important parameter during the design of underground repositories. © 2008 Elsevier B.V. All rights reserved.
1. Introduction A blasting impact and stress redistribution after an excavation induce an excavation damaged or disturbed zone(EDZ) around an excavation. An investigation into the size and characteristics of this zone is important from safety and stability points of view especially when the construction and operation time of an underground facility is long and its design criteria is rigorous. In an underground radioactive waste repository, which requires a long construction and operation time as well as an extremely long monitoring time after its closure, the mechanism of an EDZ development is one of the most important research topics. In the case of an underground high-level radioactive waste(HLW) repository, which is typically assumed to be located at several hundred meters deep in a rock mass where the in situ stress is high, the development of an EDZ, which changes the thermal, mechanical, hydraulic, and chemical behaviors in the near field, is inevitable. In the international cooperation research project,
Abbreviations: BHTV, BoreHole TeleViewer; BIPS, Borehole Image Processing System; CANDU, Canadian Deuterium Uranium; DECOVALEX, DEvelopment of COupled models and their VALidation against EXperiments; EDZ, Excavation damaged zone; HLW, high-level radioactive waste; KAERI, Korea Atomic Energy Research Institute; KRS, Korea reference disposal system; KURT, KAERI Underground Research Tunnel; LILW, low and intermediate level waste; PWR, Pressurized Water Reactors; THMC, ThermoHydro-Mechanical-Chemical; URL, Underground Research Laboratory; Watt/tHM, Watt per tons of heavy metal. ⁎ Corresponding author. E-mail address: [email protected] (S. Kwon). 0013-7952/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.enggeo.2008.04.004
DECOVALEX, which was started in 1992, an EDZ was considered as an important parameter for the Thermo-Hydro-Mechanical-Chemical (THMC) coupling analysis. In the last phase of the project, DECOVALEX-THMC, three of the following five tasks were related to an EDZ (Hudson and Jing, 2007). Task 1: Influence of a near-field THM and PA Task 2: THMC studies of an EDZ Task 3: EDZ in argillaceous rocks Task 4: Property change in an EDZ and the near field due to THC and THM processes for volcanic and crystalline rocks Task 5: Long-term climate change The development of an EDZ is dependent on many parameters such as the excavation method, tunnel geometry, blasting technique, and the rock and the in situ stress conditions. To investigate the size of an EDZ and the characteristics of it, in situ tests have been carried out in many URLs including the Aspo Hard Rock Laboratory and Stripa mine in Sweden, the Kamaish mines in Japan, the URL in Canada, the Mol in Belgium, the Tournemire in France, the WIPP in USA, the Olkiluoto research tunnel in Finland, and the Mont Terri in Switzerland (Dale and Hurtado, 1996; Backblom and Martin, 1999; Sato et al., 2000; NEA/RWM, 2002; Bossart et al., 2002; Cai and Kaiser, 2005). According to Backblom and Martin (1999), an EDZ might not be crucial for the overall safety of a repository due to the effectiveness of engineered barriers and the self-sealing and self-healing capabilities of an EDZ in claystone or salt rock; however, the results are very dependent on which nuclide is studied, what assumptions are made,
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
and what type of models are used for the engineered barrier and the rock. They also concluded that excluding an EDZ may result in nonconservative maximum radiation dose estimations and thus the relevant performance assessments should be more explicit and transparent with a consideration of an EDZ. In Korea, a Korean reference disposal system(KRS) was developed for the permanent disposal of spent fuels in a deep geological formation(Lee et al., 2006). In order to validate the feasibility, safety, stability, and reliability of this system, an underground research tunnel, KAERI Underground Research Tunnel(KURT), was constructed as a generic URL in 2006. The characteristics of an EDZ around a tunnel were investigated with various laboratory as well as in situ tests. In this study, the influence of an EDZ on the mechanical, hydraulic, and thermal behaviors around the KURT and the conceptual repository design adapting the KRS was investigated based on three-dimensional computer simulations and the rock properties measured from the laboratory and in situ tests.
111
2. Introduction to the KRS and the KURT 2.1. Korean reference disposal system In Korea, it is an urgent issue to develop a reasonable management system for the spent fuels being generated from the 20 nuclear power plants currently operating at 4 reactor sites. The amount of the spent fuels accumulated and stored at the reactor sites was more than 8600 tons in 2007. It will be increased to 36 000 tons in 2040, if the currently operating plants and those under construction and planned for are operated for their expected life times, which ranges from 30 to 60 years. In order to develop the KRS with a consideration of the characteristics of the spent fuels and the geological conditions in the Korean peninsula, a long-term R&D program has been carried out since 1997. According to the KRS, an underground HLW repository is supposed to be located in a crystalline rock mass like granite at about 500 m below the surface. The canisters containing Pressurized Water
Fig. 1. Disposal concept being considered for the Korean reference disposal system.
112
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Table 1 Major components for the Korea reference disposal system Components Spent fuel
Canister
Buffer
Backfill Rock
Disposal tunnel
Deposition hole
Characteristics Type Amount Disposal rate Heat generation Cooling time Diameter Length Number
PWR 20 000 tU 380 canister/yr 1540 W 40 years 1.02 m 4.83 m 11375
CANDU 16 000 tU 146 canisters/yr 760 W 30 years 1.02 m 4.83 m 2835
Compacted bentonite Dry density: 1800 kg/m3 Thickness: 0.5 m Maximum temperature b 100 °C Crushed rock 70% + bentonite 30% Grain size b 22 mm Crystalline rock (granite) Depth 500 m Thermal gradient : 30 °C/km Drill and blasting Tunnel spacing : 40 m Underground area : 8 km2 Tunnel number : 700 tunnels Width 5 m, height 6.15 m, length 251 m Vertical emplacement Total number : 11944 holes Spacing : PWR 6 m, CANDU 4 m Diameter :2.02 m
Reactors (PWR) and Canadian Deuterium Uranium(CANDU) spent fuels are assumed to be emplaced in vertical boreholes drilled with a spacing of 6 m and 4 m, respectively, in the floor. Because of the different characteristics of these two spent fuels, it is recommended to dispose of them in separate areas. The migration of radionuclides from the disposed high-level waste will be delayed by natural barriers as well as engineered barriers including a waste form, canister, buffer, and backfill. Compacted bentonite blocks will be installed around a
canister in a deposition hole as a buffer. A mixture of bentonite and crushed rock is being considered as the backfilling material. The dimensions and material type of the canisters for the two types of spent fuels are designed to be exactly identical to make their encapsulation and handling processes in the repository simple. The principal design concept and major parts of the KRS were reported in Lee et al. (2006). Fig. 1 shows the underground repository concept being considered for the KRS. Table 1 is a brief description of the important parts of the KRS. Cooling time in Table 1 means the time for a temporary storage in the pools or dry storage at the reactor sites. 2.2. KAERI Underground Research Tunnel (KURT) In order to validate the suggested disposal system, the KRS, a small scale underground research tunnel, KURT, was constructed in Daejeon, which is located about 150-km south of Seoul, in 2006. The KURT portal is located on a hill-side about 110 m above sea level (35°25′ N, 127°22′ E) . Fig. 2 shows the location and layout of the KURT. At the end of the access tunnel, the overburden is about 90 m. For the first 40 m of the access tunnel, 0.3 m thick concrete lining was installed to support the tunnel in the weak rock which had suffered from a weathering. The other areas were supported mainly by shotcrete and rock bolts. The geological conditions of the site had been investigated in 2003. The main rock type in the area is granite, which is the host rock type for the low and intermediate level waste(LILW) repository in the Gyung-Ju area and a candidate host rock type for an HLW repository. During the site characterization, various rock and rock mass properties were determined from laboratory and in situ tests. Two boreholes were drilled at the tunnel portal area: (a) one is a 160 m long vertical borehole; and (b) the other is a 250 m long declined borehole with the same direction and slope as the access tunnel. In the boreholes, in situ tests including hydraulic fracturing tests, a hydraulic tests, BHTV/BIPS, borehole shear tests, and Goodman jack tests were carried out. Hydraulic fracturing and impression packer tests were implemented at 7 locations ranging from 50 m to 155 in the vertical borehole. Total
Fig. 2. Schematic drawing of the KURT.
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
113
Fig. 3. Variation of the hydraulic conductivities along the 10% declined borehole.
of 36 hydraulic tests were carried out with a 10 m interval using a double packer method to determine the hydraulic conductivities along the boreholes. At first an injection test was carried out at a pressure of 4 kgf/cm2 for 20 min. After the injection test, a fall-off test was followed to check on the pressure drop under the condition of noinjection for 20 min. Fig. 3 shows the results from the hydraulic pressure tests in the declined borehole. Hydraulic conductivities were usually lower than 10− 6 m/s except for 180 m, where it was 10− 5 m/s (Kwon et al., 2004). There is a tendency for decreasing hydraulic conductivities with an increasing depth. This can be explained by the improvement of the rock quality, which was confirmed by the increase of the RQD and RMR with the depth (Kwon et al., 2006). Table 2 lists the properties from the laboratory and in situ borehole tests. At a shallow depth, the rock strength and modulus are lower than those at a deeper location due to a weathering at a shallow depth. In the cases of friction angle and cohesion, however, it was not possible to derive any depth effect, because the triaxial compression tests were only carried out for the rock cores from deep locations (Kwon et al., 2004). The construction of the KURT was started in May, 2005. The total tunnel length of the KURT with an access tunnel and two research modules is 255 m. A 10% downward slope and a tunnel direction toward the top of the mountain were chosen to effectively achieve the tunnel depth with a short access tunnel and to locate the research modules, in which the major in situ experiments will be carried out, in relatively good geological conditions (Kwon et al., 2006). A controlled blasting technique was applied with Emulite and Kinex, which was used for the production and perimeter holes, respectively. About 90 blast holes with a 38 mm diameter each were drilled by a two-boom jumbo for each blast round. Every day one blasting round was completed and a 1 m–3 m advance could be achieved depending on the rock condition and the blasting design. During the construction of the KURT, a fracture mapping was carried out after a blasting. In the fracture mapping, the dip and dip direction, the fracture roughness, fracture filling materials, and the water inflow of the fractures longer than 2 m were recorded. In several locations, fractures zones were found around faults or dykes at 45 m, 67 m, 112 m, 115 m, 120 m, and 138 m from the tunnel entrance (Kim et al., 2006). Fig. 4 shows the total inflow recorded during and after the construction of the tunnel. After the construction of research module 2, which crosses water conducting fractures, and the drilling of the 20–50 m long boreholes in the floor and wall for research purposes, a significant amount of inflow was recorded. In contrast, no detectable water inflow was recorded during the construction of research module 1, which was constructed in a dry conditioned rock without
any water conducting fractures. The inflow was quickly stabilized with the installation of packers at the boreholes. During the construction of the remaining parts of research module 2 and research module 1, the amount of inflow was maintained at around 20–25 tons/day. The water is collected at a sump pit installed at the end of the access tunnel and pumping out to the surface automatically. Table 3 lists the measured rock properties in the investigated ranges divided by the major fracture zones and fault zones(Kwon et al., 2006). The rock mass properties were determined from RMR and Hoek–Brown's empirical equations (Hoek et al., 2002). Fig. 5 shows the geological conditions along the tunnel and the EDZ study area. Discontinuities with dip directions similar to the tunnel direction could not be included. Only the major discontinuities with a dip direction more or less perpendicular to the tunnel direction were plotted in Fig. 5. 3. EDZ size and properties In situ tests to investigate the characteristics of an EDZ at the KURT were carried out after the construction of a 8 m long turning shelter at about 60 m from the tunnel entrance. Seven boreholes were drilled into the tunnel wall for the borehole tests before the excavation of the access tunnel as shown in Fig. 6. The construction of the access tunnel
Table 2 Rock properties from the laboratory and in situ tests using the declined borehole Properties
Unit
Number of tests Average
Depth effect
Young's modulus Deformation modulus Poisson's ratio Density UCS Tensile strength Friction angle of rock Cohesion of rock Dynamic elastic modulus Max. horizontal stress Min. horizontal stress In situ stress ratio K Hydraulic conductivity RQD RMR
GPa GPa
12 3 12 12 60 14 6 6 12
26.5 5.04 0.23 2623 61.65 9.5 49 17.6 52
Ed = 0.1332Z + 0.38 – – UCS = 135.02 − 779.6/Z0.5 St = 0.0283Z + 5.8218 – – –
MPa MPa
7
m/s
22 252 m 252 m
4.6 5.6 1.98 4.26e− 7 64.5 59
0.0201Z⁎ + 3.505 0.0174Z⁎ + 2.7455 K = 22.607Z⁎− 0.5882 ln(k) = −14.59 − 0.017Z RQD = 94.984 − 280.3/Z0.5 RMR = 17.21lnZ − 20.18
kg/m3 MPa MPa MPa GPa
Z is the distance along the declined borehole. Z* is the depth in the vertical borehole.
(continued on next page)
114
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Fig. 4. Inflow change with the tunnel excavation.
was restarted and the response of the rock mass was measured. Two more boreholes were drilled after the construction of the access tunnel to measure the property changes due to a blasting impact. The rock cores obtained from the drilling before and after a blasting were used for the rock core observation and laboratory tests. From the in situ and laboratory tests, the following conclusions could be drawn: 1) According to the RQDs of the rock cores from the EDZ study area, the average RQD in the 0–2 m range, where the blasting impact is significant, was 67, while it was 81 at a deeper range. This means the RQD was decreased by about 17% by the blasting(Kwon and Cho, 2007). 2) From the Goodman jack test before and after a blasting, it was possible to observe that the deformation modulus after a blasting was lower than that before a blasting. The influence of a blasting could even be found at 1.5 m from the tunnel wall. Fig. 7 shows the result from the Goodman jack test in borehole 8, which was drilled after the excavation of the access tunnel. The horizontal as well as vertical deformation moduli were measured at 0.3 m intervals. The distance from the wall, Z, was calculated with a consideration for the direction of a borehole. It was possible to observe that the deformation modulus was more or less consistent to 1.5 m from the wall and then it was increased with the depth. The average vertical deformation modulus, 16 GPa, was about 25% higher than the horizontal deformation modulus, 12 GPa . This might be related to the characteristics of the fracture network in the study area. With more measurements and further studies on the fracture network, we will be able to explain the difference. 3) From the laboratory tests, the EDZ size could be estimated to be around 1.1–1.5 m. Fig. 8(a) shows the general trend for the variation of the rock properties with the distance from the tunnel wall. As shown in Fig. 8 (b), Young's modulus before and after an excavation change with the distance from the wall which is similar to the general trend. The results from the laboratory tests are summarized in Table 4. The elastic modulus and rock strength in the EDZ were decreased by about 50% and 15%, respectively(Kwon and Cho., 2007) . 4. Hydro-mechanical analysis with the EDZ 4.1. Model mesh To investigate the effect of an EDZ around the KURT, a hydromechanical coupling analysis was carried out using a three-dimen-
sional FDM code, FLAC3D. Because a groundwater mainly flows through joints, continuum models like FLAC3D have disadvantages for describing the hydraulic processes on a small scale. On a large scale, however, continuum models are useful for investigating the general hydraulic processes and for assessing different scenarios from a sensitivity analysis point of view. Similarly, a rock could be assumed to have isotropic and homogeneous mechanical properties. In this study, the strengths and deformation modulus determined by using the empirical equations suggested by Hoek et al. (2002) were used. The model mesh shown in Fig. 9 was developed in order to consider various factors, which can influence the mechanical and hydraulic behaviors of the KURT. The width, length, and height of the initial model mesh were 100 m, 250 m, and 300 m, respectively. Total number of zones in the model mesh was 27250. Factors considered in the modeling were: − Surface topography: The actual surface topography above the tunnel was implemented in the model as shown in Fig. 9. − Rock property change along the tunnel: The mechanical properties at different ranges divided by the major fracture zones were also inputted in the model as shown in Fig. 10(a). Since the major fracture zones were more or less vertical, 10 m thick vertical layers were inputted as shown in Fig. 10(b) for the hydraulic conductivities, which were measured at 10 m intervals in the declined borehole. − In situ stress: According to the hydraulic fracturing tests carried out in a vertical borehole near the tunnel portal, the maximum horizontal stress was in a direction of 96.5° from the north and the
Table 3 Rock mass properties in the investigated ranges along the declined borehole Parameters
Interval (m) Dip of the range UCS (MPa) Em (GPa) Bulk modulus (GPa) Shear modulus (GPa) Cohesion (MPa) Friction angle Tensile strength (MPa)
Range(m) 30–82
82–125
125–177
177–192
192–252
52.9 88 30.8 3.2 2.13 1.28 1.7 35 0.014
42.6 68–88 58.4 12.1 8.07 4.84 4.2 40 0.102
51 68–90 71.6 37.9 25.3 15.16 6.7 46 0.49
15.2 90 77.6 6.21 4.14 2.48 4.6 36 0.047
60.1 90 82.7 20.4 13.6 8.16 6.4 42 0.23
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
115
Fig. 5. KURT layout and the geological conditions around the EDZ study area.
in situ stress ratio was decreased with the depth as listed in Table 2. In order to simulate the actual in situ stress variation with the depth, the model meshes above the current surface topography were removed step by step as shown in Fig. 9. Different initial stress ratios (K) in the initial flat mesh were used to simulate the measured in situ stress condition and finally K = 1 was chosen for the 300 m thick model mesh. The displacements developed during each step were initialized, but the stress and the plastic zone developed during this phase were stored for the following excavation phase. With this approach, it was possible to realize a measured stress ratio distribution, in which the horizontal stress was higher than the vertical stress (Kwon et al., 2006). − EDZ : It was assumed that a 1–2 m thick EDZ was developed immediately after an excavation. An EDZ in a floor is normally larger than that in a roof or a wall because of a heavy charge in a floor (Autio, 2003). From Fig. 11, which shows the RQD variation with the depth for the wall and the floor, it is possible to see that the damaged zone in the floor is larger than that in the wall. The
average RQD values were calculated from the rock cores from 25 boreholes in the KURT. In this study, therefore, the size of an EDZ in the floor was modeled to be larger than those in the roof and the wall. − Tunnel geometry: The horse shoe shape tunnel size of 6m × 6 m was implemented in the model. − Tunnel slope: In the model, the tunnel had a −10% slope, the same as the access tunnel in the KURT. − Sequential excavation : In the model, the tunnel was excavated with a 10 m advance in order to consider the influence of a sequential excavation. 4.2. Initial and boundary conditions In the model, zero displacement and no-flow boundaries were applied at the bottom and the four sides. The mechanical properties of an EDZ were adjusted based on the laboratory rock core tests. The bulk modulus and shear modulus were reduced by a half of the original rock properties as listed in Table 5. The cohesion and tensile strength in an EDZ were
Fig. 6. Boreholes drilled for the in situ EDZ study at KURT.
116
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Fig. 7. Deformation modulus change with the distance from the wall.
adjusted to 80% of the original values. The hydraulic conductivity of an EDZ was assumed to be one tenth of the original value(Table 6). For the hydraulic analysis, it was assumed that the rock was fully saturated at about 20 m above the tunnel entrance and the pore pressures at the boundaries below the water table were fixed to simulate the horizontally maintained water table. The initial saturation above the water table was assumed to be 0.5 in the calculation. 4.3. Hydro-mechanical calculations Fig. 12 shows the pore pressure contour and the flow vector after the excavation of the access tunnel. It is possible to observe that the distribution of the pore pressure and the flow vectors in the zones are different along the tunnel length. A relatively higher pore pressure and lower flow were calculated in the lower permeable zones. Because of the assumption that the rock above the water table is partially saturated, the flow vectors could reach the surface. Fig. 13 shows the calculated total inflow across the entire tunnel with a tunnel advance for the cases with and without an EDZ. After an excavation, the inflow suddenly increases and then it decreases smoothly to a stable value with time. The peak inflow at each excavation step depends on the hydraulic conductivity of the excavated range. As expected, the excavation of a higher permeable zone results a greater inflow. It is possible to see that the inflow increases steadily with the tunnel advances. Without an EDZ, the daily inflow, after a complete excavation of the access tunnel, was calculated to be about 6 tons. The inflow with the consideration of an EDZ is about 20% larger than that from the case without an EDZ. Even with a consideration of an EDZ, the inflow is much lower than the measured inflow of 25 tons/day. The lower inflow rate from the calculation might be due to the tunnel length difference and the assumptions of a hydraulic conductivity in an EDZ. In this study, the hydraulic conductivity of an EDZ is one order of a magnitude higher than the original value. Because the hydraulic conductivity of an EDZ can be increased by more than 2 orders of a magnitude(Sato et al., 2000), the difference with regards to the measured inflow rate can be
decreased with a higher conductivity in an EDZ. With an accurate measurement of hydraulic properties in an EDZ and a sensitivity analysis with varying mechanical and hydraulic properties in the future study, a clearer explanation of an EDZ effect will be possible. Transient hydraulic test is also required to investigate a rock diffusivity, which effects the stabilization of a flow with time. As shown in Fig. 14, the elastic and plastic displacements around the tunnel after the excavation of the access tunnel were increased by about 50% with the consideration of an EDZ in the model. The maximum displacement of about 8 mm was recorded in the wall around the tunnel portal. From Fig. 14, it is possible to observe that a displacement along the tunnel is largely dependent on a variation of the deformation modulus for the rock mass. When the bulk modulus is over 10 GPa, the displacements around the tunnel were less than 2 mm, even with an EDZ. Such a small deformation, in the order of 0.1%, for the 6 m tunnel diameter is related to the adjustment of the rock properties in an EDZ based on a rock core test. For a more realistic modeling, therefore, it is required to carry various in situ tests to measure the mechanical properties. 5. Thermo-mechanical analysis for the reference disposal system 5.1. Model mesh, initial, and boundary conditions For the conceptual design of the KRS, a thermo-mechanical coupling analysis was carried out with the model mesh in Fig.15. An underground repository with a deposition hole spacing of 6 m and tunnel spacing of 40 m could be modeled by a quarter model with two vertical symmetry planes on the X and Y axes. For a detailed analysis, the disposal tunnel, fuel part, outshell, backfill, buffer, EDZ, and rock were included in the model. The properties of the fuel part, which represents the inside of the outshell, were determined with the assumption that the fuel and cast insert were uniformly mixed. Important buffer and backfill properties could be derived from the laboratory tests as well as the literature review undertaken by Cho et al. (2001). The material properties of them are summarized in Tables 7 and 8.
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
117
Fig. 8. General trend of a rock property change before and after a blasting.
The rock properties were determined using a 500 m deep borehole in the Goseong area in the north-eastern part of the Korean peninsula (Park et al., 2001). Hydraulic fracturing tests were carried out at 10 locations with 50 m intervals. At a shallow depth from the surface to 200 m, the Table 4 Representative results of the laboratory tests Distance (m)
Uniaxial compressive strength P wave velocity S wave velocity Young's modulus Triaxial compressive strength Brazilian tensile strength Porosity
Property change in Zone 1 (%)
D1
D2
0.7 0.7 0.9 0.7 0.9 0.7 More than 1.5
0.9 1.5 1.5 1.5 1.1 1.1 –
−14 −17 −12 −57 −5 to −11 −13 108
horizontal stresses are much higher than the vertical stress. At the locations deeper than 200 m, the average in situ stress ratio was almost 1.0 (Park et al., 2002). The rock properties for an EDZ were adjusted to be similar to the previous hydro-mechanical analysis. The mechanical and thermal properties including the bulk modulus, shear modulus, cohesion, tensile strength, and thermal expansion coefficient for the EDZ were reduced by 50% from those of the intact rock. It was assumed that the thermal conductivity of an EDZ was increased by 100%, because a higher permeability in an EDZ can allow for a greater groundwater movement, which results in a greater heat transfer from a canister to a rock mass. In the cases of the friction angle and the specific heat, the same values as the intact rock were used. The in situ stresses at the repository depth, 500 m, were assumed to be hydrostatic based on the hydraulic fracturing test in the Goseong area. The surface temperature was assumed to be 20 °C and a geothermal gradient of 3 °C/100 m was used. For the PWR spent fuel generated from the Korean
118
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Fig. 9. Model mesh for the mechanical-hydraulic coupling analysis for the KURT.
nuclear power plants, its decay heat can be calculated using the following equation; 1
P ðt Þ ¼ 852:34e 0:2642þ0:130889t ðWatt=tHMÞ 1btb30 years P ðt Þ ¼ 14548 t 0:76204 ðWatt=tHMÞ tN30 years where, t is the elapsed time (year) after extracting the spent fuel from a reactor. Fig. 16 shows the decay heat change with time after its discharge from the reactors. Subsequent heating of a rock mass by a heat-generating waste will increase the stresses in a buffer, canister, and rock mass because of a thermal expansion. Because of the higher thermal expansion of water than that of rock, the heat from the canister will also increase the pore pressure in the rock. This will lead to higher hydraulic gradients around the deposition hole and influence on the hydraulic behavior in the near field. 5.2. Influence of an EDZ on the thermo-mechanical results Two cases with and without an EDZ were modeled for the disposal tunnel and the deposition hole containing PWR spent fuel for 50 years.
It is normal to simulate much longer computation span in radioactive waste disposal projects to investigate the long-term complex THMC coupling behavior. In this study, however, 50 years of computation span was enough to check the peak temperature, which was used for the evaluation of the influence of EDZ. The decay heat from the spent fuel was calculated with a consideration of a 40 years cooling after the discharge from the reactors. Without a consideration of an EDZ, the peak temperature at the contact surface between the canister and the buffer was 92 °C as shown in Fig. 17. It was decreased to 89 °C with the consideration of an EDZ around the tunnel and borehole. When bentonite is used for the buffer material, the maximum temperature of the buffer is recommended to be lower than 100 °C in order to avoid a non-boiling condition and a mineralogical alternation of the bentonite, which can be occurred with the hydrothermal reaction between the bentonite and groundwater at higher temperature. Since it is usual to design a deposition hole and a tunnel spacing with the thermal criteria for a buffer, the meaning of the 3 °C difference for the maximum buffer temperatures with and without an EDZ is significant from a repository design point of view.
Fig. 10. Assignment of the mechanical and hydraulic properties.
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
119
Fig. 11. RQD variation with the depth for the wall and the floor.
Kwon et al. (2003) carried out a sensitivity analysis of the near field during the development of a disposal system. In the sensitivity analysis, seven parameters including the tunnel size, buffer and backfill thickness, deposition hole and tunnel spacing were considered. The buffer temperature for 50 years after an emplacement was compared for these cases and the main effect of each parameter could be determined as shown in Table 9. Negative main effect of a parameter means the buffer temperature decreases with an increase of this parameter. In the case of the tunnel spacing, the main effect was −4.3. This means the buffer temperature can be decreased by only 4.3 °C when the tunnel spacing is increased from 36 m to 44 m. Compared to the influence of the parameters on the buffer temperature listed in Table 9, a temperature change of 3 °C with the consideration of an EDZ is significant. Because of the possible influence of an EDZ on the temperature distribution around an underground disposal area, it is highly recommended to carry out an in situ long-term heater test to evaluate the thermal properties in an EDZ and to reduce the uncertainty of a modelling. The development of an EDZ after an excavation reduced the stress concentration around the excavation as shown in Fig. 18. Because of the lower modulus and strengths of the EDZ than those of the intact rock, the maximum principal stresses are calculated to be a half of those from the case without the consideration of an EDZ. Such a stress reduction with the consideration of an EDZ can affect the mechanical and hydraulic behavior of the rock mass around an underground repository by changing the confinement pressure on the discontinuities. From the thermal and mechanical points of view, therefore, it is highly recommended to include the characteristics of an EDZ as an important parameter for an HLW repository design and safety
1) The groundwater inflow into the KURT was increased by about 20% with the consideration of an EDZ, in which the hydraulic conductivity was assumed to be one order of a magnitude higher than the original rock mass. 2) The excavation of a higher permeable zone resulted in a greater inflow into the KURT. The inflow immediately after the excavation increased sharply and decreased with time to a stable value.
Table 5 Mechanical properties in the EDZ
Table 6 Hydraulic properties in the EDZ
Range (m)
Bulk modulus (GPa)
Shear modulus (GPa)
Cohesion (MPa)
Friction angle (deg.)
Tensile strength (MPa)
40–80 70–130 130–180
1.07 4.04 12.65
0.64 2.42 7.58
1.36 3.36 5.36
35 40 46
0.011 0.08 0.098
analysis. Also, various in situ and laboratory tests should be undertaken for a better characterization of an EDZ in different geological and blasting conditions. 6. Conclusions For the safe management of an underground facility, an understanding of an EDZ effects is important especially when the construction and operation time is long as an underground radioactive waste repository. In Korea, a reference disposal system, KRS, was proposed in 2006 and a small scale underground research tunnel, KURT, for a validation of the KRS was constructed in Nov., 2006. During the construction of the KURT, the characteristics of an EDZ were investigated from in situ as well as laboratory tests. Based on the in situ and laboratory tests, the size of an EDZ around the KURT could be estimated to be 1.5 m from the tunnel wall. The elastic modulus and deformation modulus in an EDZ was decreased by about 56% and 40%, respectively, when compared to the original rock properties. In order to evaluate the influence of an EDZ on the thermal, hydraulic and mechanical behaviors, hydro-mechanical and thermomechanical coupling analyses were carried out using FLAC3D. From this study, the following conclusions could be drawn:
Range (m) Hydraulic conductivity Range (m) Hydraulic conductivity
40–50 7.15e-6
50–60 7.15e-6
60–70 1.75e-6
70–80 3.42e-6
80–90 3.06e-6
90–100 3e-7
100–110 1.05e-7
110–120 120–130 130–140 140–150 150–160 160–170 170–180 2.44e-6 5.19e-6 8.54e-9 2e-7 8.18e-9 1.86e-7 9.24e-8
120
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Fig. 12. Pore pressure contour and flow direction after the excavation.
3) After the completion of the excavation for the KURT, a relatively higher pore pressure and lower flow rate were observed in the lower permeable zones. 4) When the properties of an EDZ determined from laboratory tests were applied, the displacement around the KURT was increased by about 50% when compared to the case without the consideration of an EDZ. 5) The peak temperature at the canister and buffer contact in the KRS was increased by 3 °C with the consideration of an EDZ, in
which the thermal conductivity was assumed to be 1 order of a magnitude higher than that of the intact rock. Since such an influence is significant when compared to the influence of the repository design parameters considered in a sensitivity analysis, it is highly recommended to carry out in situ tests to evaluate the thermal properties in an EDZ, and to reduce the uncertainty of a modeling. 6) When the mechanical properties including the bulk modulus, shear modulus, cohesion, tensile strength, and thermal expansion
Fig. 13. Variation of total inflow with a sequential excavation.
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
121
Fig. 14. Variation of the displacements in the roof, floor, and wall after the excavation.
coefficient of an EDZ were reduced by 50% from those of the intact rock, the maximum principal stresses in the roof and wall were significantly decreased. From the study, it was possible to observe the influence of an EDZ on the thermal, hydraulic, and mechanical behaviors of underground excavations. In a future study, therefore, it is highly recommended to measure the thermal, hydraulic and mechanical properties of an EDZ carefully for a better understanding of the EDZ effects. With more in situ tests likes an in situ heater test, transient hydraulic tests, and Goodman jack tests, the influence of an EDZ on the thermal, hydraulic, and mechanical behaviors of underground facilities will be evaluated more precisely.
References Autio, J., 2003. The effect of the excavation damaged zone on the migration of radionuclides. The 2002 International EDZ workshop, Toronto, p. 7. paper. Backblom, G., Martin, C.D., 1999. Recent experiments in hard rocks to study the excavation response: implications for the performance of a nuclear waste geological repository. Tunnelling and Underground Space Technology 14, 377–394. Bossart, P., Peter, M.M., Moeri, A., Trick, T., Mayer, J., 2002. Geological and hydraulic characterization of the excavation disturbed zone in the Opalinus Clay of the Mont Terri Rock Laboratory. Engineering Geology 66, 19–38. Cai, M., Kaiser, P.K., 2005. Assessment of excavation damaged zone using a micromechanics model. Tunnelling and Underground Space Technology 20, 301–310. Cho, W.J., Lee, J.W., Kang, C.H., 2001. A Compilation and Evaluation of Thermal and Mechanical Properties of Compacted Bentonite for the Performance Assessment of Engineered Barriers in the High-level Waste Repository, KAERI/TR-1826/2001, KAERI. Table 7 Material properties of the fuel part and the outshell Material type
Fuel + cast steel
Model type Young's modulus Poisson's ratio Density Thermal conductivity Specific heat Thermal expansion
GPa kg/m3 W/m °K J/kg °K 1/°K
Stainless steel
Elastic model
Elastic model
190 0.3 6500 43 424 1.2e−5
200 0.3 8000 15.2 504 8.2e−6
Table 8 Material properties of the rock, buffer and backfill Material type
Modulus
Fig. 15. Model mesh for a Thermal-Mechanical coupling analysis for the Korean reference disposal concept.
Unit
GPa
Rock
Buffer
Backfill
Granite
Compacted bentonite
Bentonite 70% + Sand 30%
Bulk = 37.04 Shear = 21.65 2650
Bulk = 0.038 Shear = 0.029 Dry 1800 Wet 2100 2.04
Density
kg/m
Thermal conductivity Specific heat Thermal expansion UCS Cohesion Friction angle
W/m oK
2.523
Bulk = 0.345 Shear = 0.258 Dry 1800 Wet 2100 1.47
J/kg oK 1/oK
1 576 1.92e−5
888 3.1e−4
900 3.1e−4
MPa MPa Degree
22.5 61
7.66 1.1 50
0.93 1.1 17
3
122
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123
Fig. 16. Decay heat change after the discharge of PWR spent fuels.
Dale, T., Hurtado, L.D.,1996. WIPP Air-Intake Shaft Disturbed Rock Zone Study. SAND96-1327C. Hoek, E., Carranza-Torres, C., Corkum, B., 2002. Hoek–Brown failure criterion-2002 edition. Proceedings of the Fifth North American Rock Mechanics Symposium, Toronto, Canada, pp. 267–273. Hudson, J.A., Jing, L., 2007. DECOVALEX-THMC, Task B, Understanding and characterizing the excavation disturbed zone(EDZ). SKI report 2007, p. 08. Kim, K.S., Cho, S.I., Lee, J.H., Lim, W.M., Ryu, S.W., 2006. Fracture network model of KURT rock mass. Abstracts of Proceedings of the Korean Radioactive waste Society, Jeju, Korea, pp. 230–231. Kwon, S., Park, J.H., Choi, J.W., Kang, C.H., 2003. Thermo-mechanical sensitivity analysis of repository design parameters using Korean geological conditions. KAERI report, KAERI/TR-2360/2003. Kwon, S., Park, J.H., Cho, W.J., 2004. Geotechnical characteristics of the site for an underground research tunnel in KAERI, KAERI report. KAERI/TR-2805/2004. Kwon, S., Cho, W.J., Han, P.S., 2006. Concept development of an underground research tunnel for validating the Korean reference HLW disposal system. Tunnelling and Underground Space Technology 21, 203–217.
Kwon, S., Cho, W.J., 2007. Investigation of excavation disturbed zone around a tunnel by blasting. Explosives & Blasting 25, 15–29. Lee, J.Y., Cho, D.K., Kim, S.G., Choi, H.J., Choi, J.W., Hahn, P.S., 2006. Development of the Korean reference vertical disposal system concept for spent fuels. Waste Management'06, Tucson, AZ. Park, B.Y., Bae, D.S., Kim, C.S., Kim, K.S., Ko, Y.K., 2001. Evaluation of the basic mechanical and thermal properties of deep crystalline rocks. KAERI report. KAERI/TR-1828/ 2001. Park, B.Y., Bae, D.S., Kim, C.S., Kim, K.S., Ko, Y.K., Won, K.S., 2002. Measuring the initial earth pressure of granite using hydraulic fracturing test: Goseong and Yuseong areas. KAERI report. KAERI/TR-2038/2002. NEA, 2002. Characterization and Representation of the Excavation Disturbed or Damaged Zone(EDZ). NEA/RWM(2002). Sato, T., Kikuchi, T., Sugihara, K., 2000. In-situ experiments on an excavation disturbed zone induced by mechanical excavation in Neogene sedimentary rock at Tono mine, central Japan. Engineering Geology 56, 97–108.
Fig. 17. Influence of an EDZ on the near-field temperatures.
S. Kwon, W.J. Cho / Engineering Geology 101 (2008) 110–123 Table 9 Main effects of the parameters on the buffer temperature for 50 years after an emplacement
Max. value Min. value Average temp. with max. value Average temp. with min. value Main effect
Tunnel width
Tunnel height
Tunnel spacing
Hole spacing
Sidebuffer
Upper buffer
Backfill
6.6 m 5.4 m 79.5 °C
7.7 m 6.3 m 79.5 °C
44 m 36 m 77.3 °C
6.6 m 5.4 m 76.2 °C
0.55 m 1.65 m 0.45 m 1.35 m 79.8 °C 79.4 °C
1.1 m 0.9 m 79.5 °C
79.4 °C
79.4 °C
81.6 °C
82.7 °C
79.1 °C
79.5 °C
79.5 °C
0.06
0.06
−4.32
−6.52
0.68
− 0.1
−0.02
Fig. 18. Influence of an EDZ on the near field maximum principal stresses.
123