- Email: [email protected]
Temperature Influence on the Mechanical Behaviour of a Compacted Bentonite
305 TEMPERATURE INFLUENCE ON THE MECHANICAL BEHAVIOUR OF A COMPACTED BENTONITE Maria Victoria Villar \ Antonio Lloret 2 ^) Centro de Investigaciones Energeticas, Medioambientales y Tecnologicas (CIEMAT), Madrid, Spain 2) Universitat Politecnica de Catalunya (UPC), Barcelona, Spain Abstract: The design of high level radioactive waste (HLW) repositories in deep geological media -in which bentonite is proposed as sealing material- leads to the necessity of deepening the study of the behaviour of clays when subjected to hydraulic and thermal changes. The paper presents the results of an experimental study on the effects of temperature on the volumetric behaviour of a compacted bentonite used in FEBEX, which is a project for the study of the near field for a HLW repository according to the Spanish concept. The experimental programme includes swelling under load tests, swelling pressure tests and oedometric compression tests with and without control of suction. The results indicate that temperature reduces the swelling capacity and the swelling pressure and increases the volumetric strains induced by vertical loads due to a reduction of the yielding stress.
1. INTRODUCTION The coupling between thermal, hydraulic and mechanical processes in the bentonite barrier of HLW repositories is recognised as a crucial aspect to evaluate their behaviour. The heating induced by the radioactive waste and the hydration with water supplied by the surrounding rock affect important properties of the compacted bentonite. In particular, the temperature changes affect the mechanical response of the clay in several aspects: swelling pressure, swelling and collapse, thermal dilatation and contraction, compressibility, yielding and effects on time dependent behaviour. In relation to radioactive waste disposal, during last years a number of laboratory results referring to thermal effects on saturated soils have been presented (Baldi et al. 1988, Towhata et al. 1993, Tanaka et al. 1997, Sultan et al. 2002, Burghignoli et al. 2000). However, results focused on the thermal influence on volume change behaviour of unsaturated soils are still limited (Wiebe et al. 1998, Romero et al. 2003). In particular, information concerning the temperature effects on the mechanical behaviour of highly expansive clays in unsaturated conditions is still scarce (Lingnau et al. 1996, Romero et al. 2001). The work presented here is being performed in the framework of FEBEX (Full-scale Engineered Barriers Experiment in Crystalline Host Rock), which is a project for the study of the near field for a HLW repository in crystalline rock according to the Spanish concept: the waste canisters are surrounded by a clay barrier constructed from
highly-compacted bentonite blocks (ENRESA 2000). In this paper, experimental results obtained on compacted specimens subjected to temperature, suction and stress are presented, as well as an attempt to interpret the cause of the temperature effects observed.
2. BENTONITE PROPERTIES The tests have been performed with the FEBEX bentonite, which is the clay used for the FEBEX Project in the in situ (Grimsel, Switzerland) and the mock-up (Madrid, Spain) tests (ENRESA 2000). The FEBEX bentonite has a content of montmorillonite higher than 90 percent. The cation exchange capacity (CEC) is of 102±4 meq/lOOg, and the major exchangeable cations are: Ca (42 %), Mg (33 %), Na (23 %) and K (2 %). The liquid limit of the bentonite is 102±4 percent. The hygroscopic water content in equilibrium with the laboratory atmosphere (relative humidity 50±10 %, temperature 21 ±3 °C, suction 120 MPa) is 13.7±1.3 percent. The pore size distribution is bi-modal, as it is characteristic of this type of materials. The volume of intra-aggregate pores (smaller than 0.006 pm) is very similar for samples compacted at different dry densities, and represents 73-78 percent of total pore volume when the bentonite is compacted at a dry density of 1.7 g/cm\ Retention curves for the FEBEX bentonite at constant and free volume conditions for various dry densities and temperatures have been reported by Villar (2002) and Villar & Lloret (2002). Figure 1
306 shows the water retention curves for the bentonite compacted at dry density 1.65 g/cm^ in wetting paths performed at 20, 40 and 60 °C (Villar & Lloret 2002). The retention capacity for the same dry density is slightly lower the higher the temperature. The saturated permeability of the bentonite compacted at a dry density of 1.7 g/cm^ is of the order of 10"^"^ m/s. Other termo-hydro-mechanical and geochemical characteristics of the FEBEX bentonite are detailed in ENRESA (2000) and Villar (2002). 1000
oedometric cell located inside a thermostatic bath (Figure 2). The two inlets of the cell were closed during the tests to avoid any water exchange with the environment, consequently the samples remained unsaturated.
Load cell
1 Temperature!
- a - 20 °C 1 o 40 °C N A 60°C 11
/^
100
5
10
^ V 10
y
15 20 25 Water content (%)
Figure 2. Schematic layout of the oedometric cell inside the thermostatic bath.
30
Figure 1. Water retention curves at constant volume of FEBEX bentonite compacted at dry density of 1.65 g/cm^.
3. EXPERIMENTAL PROCEDURES Five types of tests have been performed in oedometric conditions at various temperatures following different stress paths and suction changes.
3.1 Compression tests with liygroscopic water content The influence of temperature on the compressibility of unsaturated clay has been studied by compression tests in oedometric conditions. The initial dry density of the samples has been 1.70 g/cm^, and the clay was compacted uniaxially with its hygroscopic water content inside the oedometer ring (diameter of 5.0 cm, length of the specimen of 1.2 cm). A vertical pressure between 22 and 26 MPa was applied to manufacture the specimens. Three different samples were heated up to 25, 60 and 80 °C under a low vertical load (0.4 MPa). Afterwards, the samples were progressively loaded by steps up to 20 MPa. The tests were performed in an
3.2 Soalcing under vertical load tests The influence of temperature on the swelling capacity of the clay was checked by soaking under load tests. They have been performed in an oedometer whose cell is placed in a thermostatic bath. Specimens of dry density 1.60 g/cm^ were obtained by uniaxial compaction of the FEBEX clay with its hygroscopic water content. Vertical stresses of 15±1 MPa were applied for the manufacturing of the specimens, whose final height was 1.2 cm and diameter 5.0 cm. The swelling strain experienced by the clay upon saturation with distilled water has been determined for temperatures between 30 and 80 °C. The tests have been performed under vertical loads of 0.5, 1.5 and 3.0 MPa, after having reached the stabilisation of the target temperature.
3.3 Swelling pressure tests The determination of the swelling pressure as a function of temperature in the interval between 25 and 80 °C has been performed in the oedometers shown in Figure 2. The clay has been uniaxially compacted with its hygroscopic water content at initial dry densities of 1.50 and 1.60 g/cm^ Vertical stresses of 11 and 16 MPa, respectively, were applied to obtain specimens of 5.0 diameter and 1.2 cm height.
307 Once the temperature has stabilised, the sample is hydrated at constant volume through the bottom face with deionised water injected at a pressure of 0.6 MPa, while the upper outlet remains open to atmosphere. At the same time, the swelling pressure exerted by the clay is measured by a load cell installed in the loading frame and the vertical deformation of the specimen is measured by two LVDT's. The values of load, strain and water exchange are automatically recorded. The final density may differ slightly from the nominal one due to the small displacement allowed by the equipment.
reaching a null suction state. Finally the sample was unloaded. Loading ram
Water deposit
3.4 Compression tests after saturation
B B Specimen [ m Porous stone |
In this type of tests the combined effect of the temperature and the hydration on the behaviour of the clay was investigated. The tests were performed at 25 and 80 °C in the oedometer shown in Figure 2, and the procedure to manufacture the samples was the same described in section 3.1. After the initial heating under a low vertical stress, the samples were saturated with distilled water under a vertical load of 1 MPa. Afterwards, the vertical load was progressively increased up to 20 MPa.
Figures. Schematic layout of controlled oedometer cell thermostatic bath.
^^Oedometerring—-Membranej
the suction inside the
3.5 Wetting and compression tests under controiled suction In these tests, bentonite was hydrated and loaded controlling the suction applied to the samples by using the axis translation technique. Figure 3 shows the layout of the oedometer cell, which was immersed inside a thermostatic bath (Romero et al. 2001). Suction is applied by changing the pressure of the gas phase in the pores of the sample by injecting nitrogen in the cell to the desired pressure. The bottom of the sample is in contact with water at atmospheric pressure through a cellulose membrane permeable to water but not to air. Three tests on bentonite compacted at an initial dry density of 1.7 g/cm^ with its hygroscopic water content have been performed at different temperatures (20, 40 and 60 °C), following the suction and stress paths indicated in Figure 4. In the first step, under a vertical load of 0.1 MPa, the suction of the sample was equalised to 14 MPa (due to mechanical limitations of the cell, this is the maximum suction that can be controlled). Afterwards, keeping this suction constant, the vertical load was increased up to 5 MPa. Under this vertical load, suction was decreased by steps until
0.01
0.1
1
Vertical pressure (MPa) Figure 4. Stress path followed in wetting and loading tests under controlled suction at different temperatures.
4. RESULTS 4.7 Compression tests witti tjygroscopic water content Figure 5 shows the volume change measured in the samples loaded at different temperatures keeping their water content constant. As it has been observed in saturated materials, the temperature increases the compressibility of the bentonite. A certain reduction in the size of the elastic domain with the temperature can be observed (Hueckel & Borsetto 1990), despite the fact that the vertical stresses applied are smaller than the compaction load.
308 4.3 Swelling pressure tests
0.60
***-*s ^.
0.58 o S 0.56
I
">. ||-^-25°C
I ^
0.54 -H 11 • 60°C 0.52 r] ^ 80 °c
4
0.50 0.1
1 10 100 Vertical stress (MPa)
Figure 5. Volume change measured during loading at different temperatures and constant water content (initial dry density 1.70 g/cm^).
4.2 Soaking under vertical load tests The final strains reached in soaking tests performed at different temperatures are plotted as a function of the temperature and overload of the test in Figure 6 (Villar & Loret 2002). The swelling capacity decreases with temperature, although the influence of temperature is less evident when the overload is high. This behaviour can be explained if we consider that at high temperatures the amount of adsorbed water in the microstructure of the clay is smaller that at low temperatures (Ma & Hueckel 1992), what reduces the interlamellar swelling capacity, which is the prevailing mechanism in the swelling of a Cabentonite. -25 1
20
40
60
80
100
Temperature (°C) Figure 6. Final strain of samples of initial dry density 1.60 g/crn^ saturated under different vertical loads and temperatures.
The results of swelling tests are shown in Figure 7, in which the dispersion of data can be mostly attributed to the variations in dry density caused by the small displacement allowed by the equipment. A clear decrease of swelling pressure as a function of temperature is observed. This behaviour is wellmatched with the observed increase in the compressibility of the bentonite in the compression tests and the reduction of swelling strains in the soaking tests. Lingnau et al. (1996) and Romero et al. (2003) found also a reduction in swelling pressure with temperature for a sand/bentonite mixture and for a moderately expansive clay, respectively. 1
().U
:
Dry density (g/cm )| 1? 5.0 ^ ^
8
o 1.60
An " 4.0
1 3.0 S 2.0 1 c u 1.0 ^ nn20
D 1.50
- - i
o -[
60
80
100
Temperature (°C) Figure 7. Effect of temperature on swelling pressure for two different dry densities.
4.4 Compression tests after saturation Figure 8 shows the evolution of the swelling during saturation under a vertical load of 1 MPa at two different temperatures for samples of initial dry density 1.70 g/cm^ The increase of water permeability and the decrease of swelling capacity of the bentonite with temperature can be observed. After saturation the samples were loaded under oedometric conditions. The volume changes measured during loading are plotted in Figure 9. Hydration induces a significant increase of pore volume in both samples, but this swelling is greater in the sample tested at 25 °C (Figure 8), following the same trend of the soaking under load tests. This increase of volume is associated to irreversible changes in the macrostructure of the material and reduces the apparent preconsolidation stress of the bentonite (Gens & Alonso 1992, Alonso et al. 1999). These structural changes prevail over the increase of the compressibility of the bentonite with temperature, and the cold sample presents the
309 smallest preconsolidation pressure and the biggest volume change under load application.
r
1.40 1.20
^
80 ''C
-^25°C
il
II
w> 0.80
I
0.60
00
the yield point at 20 °C is obtained around a vertical stress of 1.5 MPa, whereas the samples tested at 40 and 60 °C present this point at higher stresses. In the subsequent wetting from 14 to 0 MPa, again the higher swelling strains are observed at laboratory temperature (Figure 10). In the final unloading steps the strain changes are more relevant in the samples tested at high temperatures.
/ 0.40 0.20 -\—tt,A ^tf-r^Ti
0.00
0.0001
0.01
/ ^/ 1 Time (h)
100
10000
suction decrease from 120 to 14 MPa
-J
Figure 8. Swelling during saturation ofbentonite under a vertical stress of 1 MPa.
initial state (vertical load = 0.1 MPa)
0.80 -&-25°C A 80 °C
0.75 o •f 0.70
I 0.65 j 0.60
1
10 100 Suction+ P«m (MPa)
Figure 10. Results of isothermal oedometric tests with control of suction and temperature.
AM S
0.55 0.1
1 10 Vertical stress (MPa)
100
Figure 9. Volume change due to loading after saturation under a vertical stress of 1 MPa.
4.5 Wetting and compression tests under controiied suction The results of the oedometric tests performed with control of suction are summarised in Figure 10 (Romero et al. 2001, Villar 2002). The swelling that occurs initially when suction decreases from the 120 MPa corresponding to the hygroscopic conditions of the clay to the 14 MPa experienced in the oedometer is greater the lower the temperature. Afterwards, the stiffness of the bentonite on loading from 0.1 to 5 MPa apparently increases with temperature (the results of this part of the path are detailed in Figure 11). It must be taken into account that after the volume changes due to the first suction reduction, the structure of the three samples is not the same and, therefore, their mechanical behaviour is strongly affected: the sample tested at 20 °C presents the higher void ratio and the more compressible structure. In fact,
0.1 1 Vertical jM-essure (MPa)
Figure 11. Evolution of void ratio during loading under suction 14 MPa {detail of Figure 10).
5. CONCLUSIONS Different experimental techniques and equipments to study the influence of the temperature on the mechanical behaviour of the FEBEX bentonite under saturated and partially saturated states have been presented. Under unsaturated conditions the compressibility of the bentonite increases with the temperature of the test. This result is coherent with the thermally induced reduction of the size of the yield surface proposed by several authors.
310 A decrease of the final swelling strain on isothermal suction reduction has been observed at elevated temperatures in both suction controlled and soaking tests. This effect can be explained by transfer induced by temperature between intraaggregate adsorbed water and inter-aggregate free water. Swelling pressure also decreases with temperature. This decrease may be explained by the reduction of the swelling capacity and of the size of the elastic domain at high temperatures. In tests with suction reduction, the structure changes due to hydration are more relevant in the subsequent mechanical behaviour of the bentonite than the effects of temperature.
6. ACKNOWLEDGEMENTS Work co-funded by ENRESA and the European Commission and performed as part of the Fifth EURATOM Framework Programme, key action Nuclear Fission (1998-2002), Project FEBEX II (EC Contract FIKW-CT-2000-00016). The laboratory work has been performed at CIEMAT by R. Campos and J. Aroz. The helpful discussions with P. Rivas and P.L. Martin (CIEMAT) and E. Romero and A. Gens (UPC) are greatly acknowledged.
REFERENCES Alonso, E.E., Vaunat, J. & Gens, A. 1999. Modelling the mechanical behaviour of expansive clays. Eng. Geol. 54: pp. 173-183. Baldi, G., Hueckel, T., & Pellegrini, R. 1988. Thermal volume changes of mineral-water system in low-porosity clay soils. Can. Geotech. J. 25(4): pp. 807-825. Burghignoli, A., Desideri, A. & Miliziano, S. 2000. A laboratory study on the thermomechanical behaviour of clayey soils. Can. Geotech. J. 37: pp. 764-780. ENRESA. 2000. FEBEX Project. Full-scale engineered barriers experiment for a deep geological repository for high level radioactive waste in crystalline host rock. Final Report. Pub. Tec. 1/2000. 354 pp. Gens, A. & Alonso, E.E. 1992. A framework for the behaviour of unsaturated expansive clays. Can. Geotech. J. 29: pp. 1013-1032. Hueckel, T. & Borsetto, M. 1990. Thermoplasticity of saturated soils and shales: constitutive equations. J. Geotech. Eng. ASCE 116(12): pp. 1765-1777. Lingnau, B.E., Graham, J., Yarechewski, D., Tanaka, N. & Gray, M.N. 1996. Effects of
temperature on strength and compressibility of sand-bentonite buffer. Eng. Geol. 41(1-4): pp. 103-115. Ma, C. & Hueckel, T. 1992. Stress and pore pressure in saturated clay subjected to heat from radioactive waste: a numerical simulation. Can. Geotech. J. 29: pp. 1087-1094. Romero, E., Villar, M.V. & Lloret, A., 2001. Thermo-hydro-mechanical behaviour of two heavily overconsolidated clays. Proc. 6* Int. Workshop Key Issues in Waste Isolation Research. ENPC, Paris. Romero, E., Gens, A. & Lloret, A. 2003. Suction effects on a compacted clay under nonisothermal conditions. Geotechnique 53 (1): pp. 65-81. Sultan, N., Delage, P. & Cui, Y.J. 2002. Temperature effects on the volume change behaviour of Boom clay. Eng. Geol. 64: pp. 135-145. Tanaka N., Graham, J. & Crilly, T. 1997. Stressstrain behaviour of reconstituted illitic clay at different temperatures. Eng. Geol. 47: pp. 339350. Towhata, I., Kuntiwattanakul, P., Seko, I. & Ohishi, K. 1993. Volume change of clays induced by heating as observed in consolidation tests. Soils and Foundations 33(4): pp. 170-183. Villar, M.V. 2002. Thermo-hydro-mechanical characterisation of a bentonite from Cabo de Gata. Pub. Tec. ENRESA 01/2002. 258 pp. Madrid. Villar, M.V. & Lloret, A. 2002. Temperature influence on the hydro-mechanical behaviour of a compacted bentonite. Proc. Int. Meeting Clays in Natural and Engineered Barriers for Radioactive Waste Confinement. ANDRA, Reims. Wiebe, B., Graham, J., Tang, G. X. & Dixon, D. 1998. Influence of pressure, saturation and temperature on the behaviour of unsaturated sand-bentonite. Can. Geotech. J. 35: pp. 194205.
1. INTRODUCTION The coupling between thermal, hydraulic and mechanical processes in the bentonite barrier of HLW repositories is recognised as a crucial aspect to evaluate their behaviour. The heating induced by the radioactive waste and the hydration with water supplied by the surrounding rock affect important properties of the compacted bentonite. In particular, the temperature changes affect the mechanical response of the clay in several aspects: swelling pressure, swelling and collapse, thermal dilatation and contraction, compressibility, yielding and effects on time dependent behaviour. In relation to radioactive waste disposal, during last years a number of laboratory results referring to thermal effects on saturated soils have been presented (Baldi et al. 1988, Towhata et al. 1993, Tanaka et al. 1997, Sultan et al. 2002, Burghignoli et al. 2000). However, results focused on the thermal influence on volume change behaviour of unsaturated soils are still limited (Wiebe et al. 1998, Romero et al. 2003). In particular, information concerning the temperature effects on the mechanical behaviour of highly expansive clays in unsaturated conditions is still scarce (Lingnau et al. 1996, Romero et al. 2001). The work presented here is being performed in the framework of FEBEX (Full-scale Engineered Barriers Experiment in Crystalline Host Rock), which is a project for the study of the near field for a HLW repository in crystalline rock according to the Spanish concept: the waste canisters are surrounded by a clay barrier constructed from
highly-compacted bentonite blocks (ENRESA 2000). In this paper, experimental results obtained on compacted specimens subjected to temperature, suction and stress are presented, as well as an attempt to interpret the cause of the temperature effects observed.
2. BENTONITE PROPERTIES The tests have been performed with the FEBEX bentonite, which is the clay used for the FEBEX Project in the in situ (Grimsel, Switzerland) and the mock-up (Madrid, Spain) tests (ENRESA 2000). The FEBEX bentonite has a content of montmorillonite higher than 90 percent. The cation exchange capacity (CEC) is of 102±4 meq/lOOg, and the major exchangeable cations are: Ca (42 %), Mg (33 %), Na (23 %) and K (2 %). The liquid limit of the bentonite is 102±4 percent. The hygroscopic water content in equilibrium with the laboratory atmosphere (relative humidity 50±10 %, temperature 21 ±3 °C, suction 120 MPa) is 13.7±1.3 percent. The pore size distribution is bi-modal, as it is characteristic of this type of materials. The volume of intra-aggregate pores (smaller than 0.006 pm) is very similar for samples compacted at different dry densities, and represents 73-78 percent of total pore volume when the bentonite is compacted at a dry density of 1.7 g/cm\ Retention curves for the FEBEX bentonite at constant and free volume conditions for various dry densities and temperatures have been reported by Villar (2002) and Villar & Lloret (2002). Figure 1
306 shows the water retention curves for the bentonite compacted at dry density 1.65 g/cm^ in wetting paths performed at 20, 40 and 60 °C (Villar & Lloret 2002). The retention capacity for the same dry density is slightly lower the higher the temperature. The saturated permeability of the bentonite compacted at a dry density of 1.7 g/cm^ is of the order of 10"^"^ m/s. Other termo-hydro-mechanical and geochemical characteristics of the FEBEX bentonite are detailed in ENRESA (2000) and Villar (2002). 1000
oedometric cell located inside a thermostatic bath (Figure 2). The two inlets of the cell were closed during the tests to avoid any water exchange with the environment, consequently the samples remained unsaturated.
Load cell
1 Temperature!
- a - 20 °C 1 o 40 °C N A 60°C 11
/^
100
5
10
^ V 10
y
15 20 25 Water content (%)
Figure 2. Schematic layout of the oedometric cell inside the thermostatic bath.
30
Figure 1. Water retention curves at constant volume of FEBEX bentonite compacted at dry density of 1.65 g/cm^.
3. EXPERIMENTAL PROCEDURES Five types of tests have been performed in oedometric conditions at various temperatures following different stress paths and suction changes.
3.1 Compression tests with liygroscopic water content The influence of temperature on the compressibility of unsaturated clay has been studied by compression tests in oedometric conditions. The initial dry density of the samples has been 1.70 g/cm^, and the clay was compacted uniaxially with its hygroscopic water content inside the oedometer ring (diameter of 5.0 cm, length of the specimen of 1.2 cm). A vertical pressure between 22 and 26 MPa was applied to manufacture the specimens. Three different samples were heated up to 25, 60 and 80 °C under a low vertical load (0.4 MPa). Afterwards, the samples were progressively loaded by steps up to 20 MPa. The tests were performed in an
3.2 Soalcing under vertical load tests The influence of temperature on the swelling capacity of the clay was checked by soaking under load tests. They have been performed in an oedometer whose cell is placed in a thermostatic bath. Specimens of dry density 1.60 g/cm^ were obtained by uniaxial compaction of the FEBEX clay with its hygroscopic water content. Vertical stresses of 15±1 MPa were applied for the manufacturing of the specimens, whose final height was 1.2 cm and diameter 5.0 cm. The swelling strain experienced by the clay upon saturation with distilled water has been determined for temperatures between 30 and 80 °C. The tests have been performed under vertical loads of 0.5, 1.5 and 3.0 MPa, after having reached the stabilisation of the target temperature.
3.3 Swelling pressure tests The determination of the swelling pressure as a function of temperature in the interval between 25 and 80 °C has been performed in the oedometers shown in Figure 2. The clay has been uniaxially compacted with its hygroscopic water content at initial dry densities of 1.50 and 1.60 g/cm^ Vertical stresses of 11 and 16 MPa, respectively, were applied to obtain specimens of 5.0 diameter and 1.2 cm height.
307 Once the temperature has stabilised, the sample is hydrated at constant volume through the bottom face with deionised water injected at a pressure of 0.6 MPa, while the upper outlet remains open to atmosphere. At the same time, the swelling pressure exerted by the clay is measured by a load cell installed in the loading frame and the vertical deformation of the specimen is measured by two LVDT's. The values of load, strain and water exchange are automatically recorded. The final density may differ slightly from the nominal one due to the small displacement allowed by the equipment.
reaching a null suction state. Finally the sample was unloaded. Loading ram
Water deposit
3.4 Compression tests after saturation
B B Specimen [ m Porous stone |
In this type of tests the combined effect of the temperature and the hydration on the behaviour of the clay was investigated. The tests were performed at 25 and 80 °C in the oedometer shown in Figure 2, and the procedure to manufacture the samples was the same described in section 3.1. After the initial heating under a low vertical stress, the samples were saturated with distilled water under a vertical load of 1 MPa. Afterwards, the vertical load was progressively increased up to 20 MPa.
Figures. Schematic layout of controlled oedometer cell thermostatic bath.
^^Oedometerring—-Membranej
the suction inside the
3.5 Wetting and compression tests under controiled suction In these tests, bentonite was hydrated and loaded controlling the suction applied to the samples by using the axis translation technique. Figure 3 shows the layout of the oedometer cell, which was immersed inside a thermostatic bath (Romero et al. 2001). Suction is applied by changing the pressure of the gas phase in the pores of the sample by injecting nitrogen in the cell to the desired pressure. The bottom of the sample is in contact with water at atmospheric pressure through a cellulose membrane permeable to water but not to air. Three tests on bentonite compacted at an initial dry density of 1.7 g/cm^ with its hygroscopic water content have been performed at different temperatures (20, 40 and 60 °C), following the suction and stress paths indicated in Figure 4. In the first step, under a vertical load of 0.1 MPa, the suction of the sample was equalised to 14 MPa (due to mechanical limitations of the cell, this is the maximum suction that can be controlled). Afterwards, keeping this suction constant, the vertical load was increased up to 5 MPa. Under this vertical load, suction was decreased by steps until
0.01
0.1
1
Vertical pressure (MPa) Figure 4. Stress path followed in wetting and loading tests under controlled suction at different temperatures.
4. RESULTS 4.7 Compression tests witti tjygroscopic water content Figure 5 shows the volume change measured in the samples loaded at different temperatures keeping their water content constant. As it has been observed in saturated materials, the temperature increases the compressibility of the bentonite. A certain reduction in the size of the elastic domain with the temperature can be observed (Hueckel & Borsetto 1990), despite the fact that the vertical stresses applied are smaller than the compaction load.
308 4.3 Swelling pressure tests
0.60
***-*s ^.
0.58 o S 0.56
I
">. ||-^-25°C
I ^
0.54 -H 11 • 60°C 0.52 r] ^ 80 °c
4
0.50 0.1
1 10 100 Vertical stress (MPa)
Figure 5. Volume change measured during loading at different temperatures and constant water content (initial dry density 1.70 g/cm^).
4.2 Soaking under vertical load tests The final strains reached in soaking tests performed at different temperatures are plotted as a function of the temperature and overload of the test in Figure 6 (Villar & Loret 2002). The swelling capacity decreases with temperature, although the influence of temperature is less evident when the overload is high. This behaviour can be explained if we consider that at high temperatures the amount of adsorbed water in the microstructure of the clay is smaller that at low temperatures (Ma & Hueckel 1992), what reduces the interlamellar swelling capacity, which is the prevailing mechanism in the swelling of a Cabentonite. -25 1
20
40
60
80
100
Temperature (°C) Figure 6. Final strain of samples of initial dry density 1.60 g/crn^ saturated under different vertical loads and temperatures.
The results of swelling tests are shown in Figure 7, in which the dispersion of data can be mostly attributed to the variations in dry density caused by the small displacement allowed by the equipment. A clear decrease of swelling pressure as a function of temperature is observed. This behaviour is wellmatched with the observed increase in the compressibility of the bentonite in the compression tests and the reduction of swelling strains in the soaking tests. Lingnau et al. (1996) and Romero et al. (2003) found also a reduction in swelling pressure with temperature for a sand/bentonite mixture and for a moderately expansive clay, respectively. 1
().U
:
Dry density (g/cm )| 1? 5.0 ^ ^
8
o 1.60
An " 4.0
1 3.0 S 2.0 1 c u 1.0 ^ nn20
D 1.50
- - i
o -[
60
80
100
Temperature (°C) Figure 7. Effect of temperature on swelling pressure for two different dry densities.
4.4 Compression tests after saturation Figure 8 shows the evolution of the swelling during saturation under a vertical load of 1 MPa at two different temperatures for samples of initial dry density 1.70 g/cm^ The increase of water permeability and the decrease of swelling capacity of the bentonite with temperature can be observed. After saturation the samples were loaded under oedometric conditions. The volume changes measured during loading are plotted in Figure 9. Hydration induces a significant increase of pore volume in both samples, but this swelling is greater in the sample tested at 25 °C (Figure 8), following the same trend of the soaking under load tests. This increase of volume is associated to irreversible changes in the macrostructure of the material and reduces the apparent preconsolidation stress of the bentonite (Gens & Alonso 1992, Alonso et al. 1999). These structural changes prevail over the increase of the compressibility of the bentonite with temperature, and the cold sample presents the
309 smallest preconsolidation pressure and the biggest volume change under load application.
r
1.40 1.20
^
80 ''C
-^25°C
il
II
w> 0.80
I
0.60
00
the yield point at 20 °C is obtained around a vertical stress of 1.5 MPa, whereas the samples tested at 40 and 60 °C present this point at higher stresses. In the subsequent wetting from 14 to 0 MPa, again the higher swelling strains are observed at laboratory temperature (Figure 10). In the final unloading steps the strain changes are more relevant in the samples tested at high temperatures.
/ 0.40 0.20 -\—tt,A ^tf-r^Ti
0.00
0.0001
0.01
/ ^/ 1 Time (h)
100
10000
suction decrease from 120 to 14 MPa
-J
Figure 8. Swelling during saturation ofbentonite under a vertical stress of 1 MPa.
initial state (vertical load = 0.1 MPa)
0.80 -&-25°C A 80 °C
0.75 o •f 0.70
I 0.65 j 0.60
1
10 100 Suction+ P«m (MPa)
Figure 10. Results of isothermal oedometric tests with control of suction and temperature.
AM S
0.55 0.1
1 10 Vertical stress (MPa)
100
Figure 9. Volume change due to loading after saturation under a vertical stress of 1 MPa.
4.5 Wetting and compression tests under controiied suction The results of the oedometric tests performed with control of suction are summarised in Figure 10 (Romero et al. 2001, Villar 2002). The swelling that occurs initially when suction decreases from the 120 MPa corresponding to the hygroscopic conditions of the clay to the 14 MPa experienced in the oedometer is greater the lower the temperature. Afterwards, the stiffness of the bentonite on loading from 0.1 to 5 MPa apparently increases with temperature (the results of this part of the path are detailed in Figure 11). It must be taken into account that after the volume changes due to the first suction reduction, the structure of the three samples is not the same and, therefore, their mechanical behaviour is strongly affected: the sample tested at 20 °C presents the higher void ratio and the more compressible structure. In fact,
0.1 1 Vertical jM-essure (MPa)
Figure 11. Evolution of void ratio during loading under suction 14 MPa {detail of Figure 10).
5. CONCLUSIONS Different experimental techniques and equipments to study the influence of the temperature on the mechanical behaviour of the FEBEX bentonite under saturated and partially saturated states have been presented. Under unsaturated conditions the compressibility of the bentonite increases with the temperature of the test. This result is coherent with the thermally induced reduction of the size of the yield surface proposed by several authors.
310 A decrease of the final swelling strain on isothermal suction reduction has been observed at elevated temperatures in both suction controlled and soaking tests. This effect can be explained by transfer induced by temperature between intraaggregate adsorbed water and inter-aggregate free water. Swelling pressure also decreases with temperature. This decrease may be explained by the reduction of the swelling capacity and of the size of the elastic domain at high temperatures. In tests with suction reduction, the structure changes due to hydration are more relevant in the subsequent mechanical behaviour of the bentonite than the effects of temperature.
6. ACKNOWLEDGEMENTS Work co-funded by ENRESA and the European Commission and performed as part of the Fifth EURATOM Framework Programme, key action Nuclear Fission (1998-2002), Project FEBEX II (EC Contract FIKW-CT-2000-00016). The laboratory work has been performed at CIEMAT by R. Campos and J. Aroz. The helpful discussions with P. Rivas and P.L. Martin (CIEMAT) and E. Romero and A. Gens (UPC) are greatly acknowledged.
REFERENCES Alonso, E.E., Vaunat, J. & Gens, A. 1999. Modelling the mechanical behaviour of expansive clays. Eng. Geol. 54: pp. 173-183. Baldi, G., Hueckel, T., & Pellegrini, R. 1988. Thermal volume changes of mineral-water system in low-porosity clay soils. Can. Geotech. J. 25(4): pp. 807-825. Burghignoli, A., Desideri, A. & Miliziano, S. 2000. A laboratory study on the thermomechanical behaviour of clayey soils. Can. Geotech. J. 37: pp. 764-780. ENRESA. 2000. FEBEX Project. Full-scale engineered barriers experiment for a deep geological repository for high level radioactive waste in crystalline host rock. Final Report. Pub. Tec. 1/2000. 354 pp. Gens, A. & Alonso, E.E. 1992. A framework for the behaviour of unsaturated expansive clays. Can. Geotech. J. 29: pp. 1013-1032. Hueckel, T. & Borsetto, M. 1990. Thermoplasticity of saturated soils and shales: constitutive equations. J. Geotech. Eng. ASCE 116(12): pp. 1765-1777. Lingnau, B.E., Graham, J., Yarechewski, D., Tanaka, N. & Gray, M.N. 1996. Effects of
temperature on strength and compressibility of sand-bentonite buffer. Eng. Geol. 41(1-4): pp. 103-115. Ma, C. & Hueckel, T. 1992. Stress and pore pressure in saturated clay subjected to heat from radioactive waste: a numerical simulation. Can. Geotech. J. 29: pp. 1087-1094. Romero, E., Villar, M.V. & Lloret, A., 2001. Thermo-hydro-mechanical behaviour of two heavily overconsolidated clays. Proc. 6* Int. Workshop Key Issues in Waste Isolation Research. ENPC, Paris. Romero, E., Gens, A. & Lloret, A. 2003. Suction effects on a compacted clay under nonisothermal conditions. Geotechnique 53 (1): pp. 65-81. Sultan, N., Delage, P. & Cui, Y.J. 2002. Temperature effects on the volume change behaviour of Boom clay. Eng. Geol. 64: pp. 135-145. Tanaka N., Graham, J. & Crilly, T. 1997. Stressstrain behaviour of reconstituted illitic clay at different temperatures. Eng. Geol. 47: pp. 339350. Towhata, I., Kuntiwattanakul, P., Seko, I. & Ohishi, K. 1993. Volume change of clays induced by heating as observed in consolidation tests. Soils and Foundations 33(4): pp. 170-183. Villar, M.V. 2002. Thermo-hydro-mechanical characterisation of a bentonite from Cabo de Gata. Pub. Tec. ENRESA 01/2002. 258 pp. Madrid. Villar, M.V. & Lloret, A. 2002. Temperature influence on the hydro-mechanical behaviour of a compacted bentonite. Proc. Int. Meeting Clays in Natural and Engineered Barriers for Radioactive Waste Confinement. ANDRA, Reims. Wiebe, B., Graham, J., Tang, G. X. & Dixon, D. 1998. Influence of pressure, saturation and temperature on the behaviour of unsaturated sand-bentonite. Can. Geotech. J. 35: pp. 194205.