sand mixtures

Soils and Foundations ]]]];](]):]]]–]]]

HOSTED BY

The Japanese Geotechnical Society

Soils and Foundations www.sciencedirect.com journal homepage: www.elsevier.com/locate/sandf

Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures Jiang-Feng Liua,b,n, Frédéric Skoczylasa, Jian Liuc b

a Laboratoire de Mécanique de Lille (LML), Ecole Centrale de Lille, BP 48, F-59651 Villeneuve d'Ascq Cedex, France State Key Laboratory for GeoMechanics and Deep Underground Engineering, China University of Mining & Technology, Xuzhou 221116, China c CNNC Beijing Research Institute of Uranium Geology (BRIUG), Beijing 100029, China

Received 3 December 2013; received in revised form 19 June 2014; accepted 6 July 2014

Abstract Highly compacted bentonite-based materials are often considered as buffer or sealing materials for deep high-level radioactive waste repositories. In situ, the initial state of bentonite-based materials is only partially saturated, which has a very high suction that will promote water absorption from the host rock. In addition, a gradient of water saturation will be formed between the external part and the central part of the compacted bentonite blocks. In this paper, water retention tests, under both constant-volume and free-swelling conditions, were performed to investigate the suction behavior of a compacted bentonite/sand mixture. In order to investigate the sealing ability of the partially saturated bentonite/sand mixture, gas permeability tests were also carried out under the in situ confining stress. It was found that the confining conditions have a limited effect on the water retention capacity of the compacted bentonite/sand mixture at lower levels of relative humidity (RH), while this influence is significant at higher RH levels. The results of gas permeability tests show that gas permeability is very sensitive to the water content and the confining pressure. When the sample (stable at RH ¼98%) was subjected to a in situ confining pressure (7–8 MPa), the gas permeability was very low (1.83×10–14 m/s) which indicates that gas tightness can be obtained even though the sample is not fully saturated. & 2014 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved.

Keywords: Radioactive waste repositories; Bentonite/sand mixture; Water retention capacity; Gas permeability

1. Introduction High-level radioactive waste (HLRW) repositories are usually constructed in bedrock (e.g., Callovo-Oxfordian argillite) several hundred meters below the ground surface. Such deep geological repositories are usually composed of a natural geological barrier (host rock), an engineered barrier made of a metallic canister, and bentonite-based materials. In this context, compacted bentoniten Corresponding author at: Laboratoire de Mécanique de Lille (LML), Ecole Centrale de Lille, BP 48,F-59651 Villeneuve d'Ascq Cedex, France. E-mail address: jeafl[email protected] (J.-F. Liu). Peer review under responsibility of The Japanese Geotechnical Society.

based materials are usually used to seal tunnels and galleries, or as buffer materials around the waste containers, the purpose of which is to create a “low permeable zone” around them (Alonso et al., 2006). To seal a repository gallery, the clay barrier (buffer) is usually formed by blocks of compacted bentonite-based materials arranged in vertical slices, which are put in place with initial construction gaps (Villar and Lloret, 2007). In addition, these gaps also exist between the host rock and the compacted bentonite blocks. The gaps account for 6.6% (FEBEX mock-up tests), 9% (French concept, according to Andra (the French radioactive waste management agency)), and 14% (SEALEX in situ tests) of the volume of the gallery (Martin et al., 2006; Andra, 2005a; Barnichon and Deleruyelle, 2009).

http://dx.doi.org/10.1016/j.sandf.2014.09.011 0038-0806/& 2014 The Japanese Geotechnical Society. Production and hosting by Elsevier B.V. All rights reserved.

Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]]

2

The initial state of the compacted bentonite-based materials is usually partially saturated. Once placed in the galleries, they will be progressively hydrated due to the underground water infiltrating from the host rock. As indicated by some researchers, the amount of water that infiltrates into the bentonite-based materials depends largely on their swelling properties (Pierre, 2006; Siemens and Blatz, 2009; Cui et al., 2008; Ye et al., 2009a). The swelling properties of expansive materials, with the same initial conditions, clearly differ depending on the confinement conditions (constantvolume or free swelling) (Villar, 2007; Wang et al., 2012, 2013; Ye et al., 2009a; Cui et al., 2008; Tang et al., 2013), the confining stress applied to the materials (Mollins et al., 1996; Ng and Pang, 2000a, 2000b; Lloret and Villar, 2007; Hoffmann et al., 2007; Villar and Gómez-Espina, 2007; Cui et al., 2011), the chemical components of the water (Studds et al., 1998; Abdullah et al., 1999; Zhu et al., 2013; Wang et al., 2014), and/or the temperature (Ye et al., 2009b; Cui et al., 2011; Ye et al., 2013). In addition, the initial physical properties (e.g., dry density, water content, and claymineral content) and the initial state (e.g., loose or compacted) also have a great influence on the swelling properties of the materials (Komine and Ogata., 1994, 1999; Villar and Lloret, 2008; Siemens and Blatz, 2009; Tang and Cui, 2010; Agus et al., 2010). In situ, the infiltration of water into the bentonite-based barrier is a very slow process. During this process, a gradient of water saturation will be presented between the external and the central parts of the clay barrier, as shown in Fig. 1. The external part of the barrier, that is hydrated first, will swell and will be confined in an extremely stiff host rock. As a consequence, it will compress the internal part not yet hydratedand apply a confining pressure. To assess the sealing ability of the barrier, especially for the central unsaturated part, it is essential to measure its permeability under in situ confining stress. In terms of the permeability of compacted bentonite-based materials, similar to their swelling

COx claystone

properties, it also depends on the initial state (water content, porosity, and dry density) and the boundary conditions (confining stress, gas pressure, and temperature) (Villar et al., 2012; Vangpaisal and Bouazza, 2004; Cho et al., 1999; Sällfors and Öberg-Högsta, 2002; Lloret and Villar, 2007; Villar and Lloret, 2004; Didier et al., 2000). In this study, firstly, the suction behaviors of four compacted bentonite/sand mixtures at both constant-volume and freeswelling conditions, were investigated. Then, the gas permeabilities of five partially saturated bentonite/sand mixtures, under in situ confining pressure, were measured to evaluate their sealing abilities.

2. Theoretical model 2.1. Equation of Kelvin–Laplace The Kelvin–Laplace equation describes the relationship between the capillary pressure, Pcap , and the relative humidity, RH. The RH of the air above the meniscus in a capillary pore is given by Kelvin's equation (Thomson, 1871) and cited by Galvin (2005) as lnðRH Þ ¼ 

υm 2γ cos θ r RT

ð1Þ

where υm is the molar volume, R is the universal gas constant, T is the temperature, γ is the surface tension, r is the radius of a droplet, and θ is the contact angle. Indeed, for a porous medium, it is assumed that this equation describes the relationship between the inside RH and the maximum radius of the pores which are filled with water under this RH. With the equation of

Water saturation gradient

CO x formation Cell head ~ 8.0 m

Usable part ~ 30.0 m

Water

H2

H2

H2

H2

Disposal package

Permanent sleeve

Metal plug

Concrete plug

Compacted bentonite/sand mixture Scale

CO x formation Fig. 1. Schematic diagram of the in situ saturation process: example of the access drift and storage gallery for type C waste, at a depth of -500m, after Andra (2005b). Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]]

Young–Laplace (Pellicer and Garcia-Morales, 2000), 2γ cos θ ð2Þ r According to Eqs. (1) and (2), capillary pressure Pcap can be expressed as the following equation: RT Pcap ¼  lnðRH Þ ð3Þ υm

Pcap ¼

2.2. Water retention model The sorption isotherm (relationship between the degree of saturation Sr and relative humidity RH) can be determined by the water retention model and the Kelvin–Laplace equation. One of the most popular retention models is the model of van Genuchten (the VG model) (van Genuchten, 1980), namely,    m Pcap n Sr ¼ 1 þ ð4Þ Pr in which m ¼ 1 1/n and Pr are two parameters which are related to the size distribution of the pores in the porous medium. However, this model cannot properly simulate the experimental results when the degree of water saturation is close to 1 (Feyen et al., 1998; Liu, 2012). Therefore, Gerke and Genuchten (1993) proposed a dual-porosity theory to improve the fitting effect. It is assumed that in a porous medium, there are two pore systems: a macrospore or the fracture system and a less permeable matrix pore system. Durner (1994) presented the following retention model which contains the two pores systems:         pcap n1  m1 pcap n2  m2 S r ¼ ω1 1 þ þ ω2 1 þ ð5Þ p1 p2 where ω1 is the volume fraction of the first pore system and ω2 ¼ 1  ω1 is the volume fraction of the second pore system. m1 ¼ 1 ð1=n1 Þ, p1 , m2 ¼ 1  ð1=n2 Þ, and p2 are related to the size distribution of the first and the second pore systems. Therefore, there are 5 independent parameters in this model, namely, ω1 , m1 , p1 ; m2 , and p2 . In order to facilitate the numerical simulation, Liu (2012) proposed a dual-porosity model (the DP model) that contains the following three independent parameters:     Sr ¼ ω1 1  eðp1 =pcap Þ þ ω2 1  eðp2 =pcap Þ ð6Þ

3

thermal conductivity. As required by Andra, the target swelling pressure is between 7–8 MPa. To obtain this swelling pressure, the mixture should be compacted to yield a dry density of 1.77 g/cm3 and a water content of 15.2%. To obtain such a water content, the bentonite/sand powder mixture was put in a climatic chamber with a constant RH (85%) and temperature (22 1C) and then compacted in a steel tube at an axial pressure of 12 MPa. The resulting samples are 12.5 or 25 mm in height and 37 mm in diameter. Two tests were performed under constant-volume condition (SO1 and SO2), as shown in Photo 1. In the tests, the sample was made to swell in the tube, while the radial deformations were obstructed by the inner surface of the tube and the axial strain was blocked by the use of two porous plates. Two other tests were performed under freeswelling condition (SF1 and SF2). Five samples (SF3-1–SF3-5) were used to perform the gas permeability tests. More information can be found in Table 1.

3.2. Experimental methodology 3.2.1. Method to obtain different saturation levels under constant-volume/free-swelling conditions In order to obtain uniformly partially water-saturated bentonite/ sand samples, each sample was placed in a hermetic chamber at a given RH of 75%, 85%, 92%, or 98%. These RH were provided by various salt solutions (Greenspan, 1977). Full water saturation was achieved through mass stabilization in a hermetic chamber at an RH of 100% (over pure distilled water), while for the constantvolume condition the sample was put in a triaxial cell where water could be injected directly to accelerate the saturation process. Within each hermetic chamber, the sample was put above the water surface, where the actual RH was close to the required value.

Constant-volume condition

Free-swelling condition

In fact, this dual retention model is based on the supposition that the sample has a bimodal distribution of pore size (Maekawa et al., 1999). 3. Materials and experimental methods 3.1. Materials and sample preparation The materials used in our study consist of a mixture of 70% bentonite (MX80-type Wyoming sodic montmorillonite) and 30% sand. Sand was chosen as the inert additive in order to limit the swelling ability of the bentonite and to improve its

Photo 1. Compacted bentonite/sand mixtures used for both free-swelling and constant-volume conditions.

Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]]

4

Table 1 Nomenclature of the samples and the confinement conditions of the tests. Samples

Number

Confinement conditions

Notes

SO1 SO2 SF1 SF2 SF3-1–S3-5

1 1 1 1 5

Constant-volume condition

RH: 75%. 85%. 92%. 98% and 100%.

(H¼ 25 mm, D¼ 42.5 mm) (H ¼ 25 mm, D¼ 42.5 mm) (H ¼ 12.5 mm, D¼ 42.5 mm) (H¼ 25 mm, D ¼ 42.5 mm) (H ¼ 25 mm, D¼ 37.6 mm)

Free-swelling condition Free-swelling condition (water retention tests)

3.2.2. Experimental procedure Except for samples drying, all tests were performed in an air-conditioned room with a mean temperature of 22 1C. The whole experimental procedure is as follows: Water retention tests (1). Samples preparation as described above. (2). Weighing of the different samples and pieces of experimental mounting (tube, plates, etc.). (3). Equilibrium at RH ¼ 75%, 85%, 92%, and 98% (until mass stabilization). (4). Full water saturation were obtained in a hermetic chamber with RH ¼ 100%, or in a triaxial cell. (5). Finally, these samples were dried in an oven t.

Gas permeability tests were conducted in the triaxial cell under confined conditions after water retention tests.

times, namely, Sr ¼

VPw mt  mdry ¼  100% V P  total msat  mdry

ð9Þ

where msat is the sample mass measured at the fully saturated state, V P  w is the pore volume when filled with water, and V P  total is the total pore volume. In addition, under free-swelling condition, the sample volume (V RH ) was measured at the end of each RH step. The relative volume variation (V r ) is chosen to express the change in volume during the hydration process. Vr ¼

V RH  V ini  100% V ini

ð10Þ

where V ini is the initial volume of the sample, namely, just after compaction. Gas permeability tests (1). Repeat the 1st–3rd steps of the water retention tests (under the free-swelling condition). (2). Gas permeability tests were performed in a triaxial cell with argon gas (under confined conditions, and with confining pressures of Pc: 3 MPa, 5 MPa, 8 MPa, 5 MPa, and 3 MPa). (3). Repeat the 5th step of the water retention tests, and then these samples were put in a hermetic chamber with RH ¼ 100%.

3.2.3. Calculation of parameters In order to avoid the initial difference between the different samples (e.g., initial mass), we use the relative mass variation (mr) to express the mass change of the sample during the hydration process, mt  mini  100% ð7Þ mr ¼ mini where mt is the sample mass at time t (day) and mini is the initial mass, i.e., just after compaction. Similarly, water content w is calculated as follows: mt  mdry  100% ð8Þ w¼ mdry where mdry is the dry mass of the sample. Also, the degree of saturation can be derived from the sample mass at different

3.2.4. Method to measure gas permeability The experimental set-up used for the gas permeability tests is shown in Fig. 2. The experimental device includes a confining cell, an oil pump with a digital manometer used to record the applied confining pressure, a buffer tank (0.4 L) supplied with argon gas by a big gas tank, and two manometers that can record the gas pressure in the buffer tank and in the gas tank. Prior to being placed inside the triaxial cell, the sample was jacketed inside a protective Viton™ membrane. Confining pressure (Pc) was applied and controlled by the oil pump. The maximum Pc was limited to 8 MPa, corresponding to the in situ swelling pressure (Liu, 2013). In steady-state conditions, Darcy's law can be used directly to determine the gas permeability. As shown in Fig. 2, one side of the circular cylindrical sample was subjected to a given gas pressure ðP1 Þ, while the other side was subjected to a constant atmospheric level ðP0 Þ. The average gas volume flow rate ðQg Þ was evaluated by measuring the time ðΔtÞ required to obtain a small decrease in pressure ðΔPÞ. By assuming a quasi-static flow and the ideal gas state equation, Qg can be calculated as follows: Qg ¼

V 0 ΔP Pmean Δt

ð11Þ

where V 0 is the volume of the buffer reservoir, and Pmean is the average upstream gas pressure, namely, Pmean
¼ P1  ΔP=2. Therefore, effective gas permeability K ef f m=s can be computed with the following equation: K ef f ¼ K int K rg ¼

ρgQg 2hPmean A ðP2mean  P20 Þ

ð12Þ

Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]]

5

Fig. 2. Experimental setup used for steady state gas permeability tests, consisting of triaxial cell (to apply confinement) and gas system to impose pressure gradient and to measure flow rate.

There are three basic definitions in this equation, namely, K ef f ; K int , and K rg . When applying a gas pressure gradient to a material partially saturated with water, gas flows through the pores filled with air. The measured gas permeability is defined as K ef f . Theoretically, when considering the single phase flow (gas or water) in a porous medium, the permeability ðK int Þ is called the intrinsic permeability, which is independent of the fluid. The relative permeability ðK rg Þ is the ratio of K ef f to K int . ρ is the density of the gas, g is the acceleration due to gravity, h is the height of the sample, and A is the cross-sectional area of the sample. In fact, the density of gas ρ will change with the gas pressure, so the value of ρ should be corrected according to the injected gas pressure or back pressure, depending on what flow is used for the calculation (see Eqs. (14) and (15)). m PV ¼ nRT ¼ RT ð13Þ M m PM ρ¼ ¼ ð14Þ V RT where P is the gas pressure, V is the gas volume, n is the amount of substance of gas, T is the temperature of the gas (295.15 K in our tests), m is the gas mass, and M is the molar mass (39.9 g/mole for argon). 4. Results and analysis 4.1. Water retention properties 4.1.1. Water retention tests under constant-volume condition: SO1 and SO2 Fig. 3 shows the variation in relative mass of samples SO1 and SO2 versus time under different RH, while samples SO1 and SO2 were saturated under constant-volume condition. The two samples were prepared under the same conditions. The first and main observation is that quite a long period is needed for the whole process: 123 days (SO1) vs. 228 days (SO2). For a pre-compacted bentonite/sand mixture, the initial water potential or total suction is not uniform due to incomplete hydration. The hydration process is in fact a process of the redistribution of pore water until an internal equilibrium state is achieved. The total time needed to reach this equilibrium state lies in the pore size (or the coefficient of

12

Relative mass variation (%) SO1_75% SO1_85% SO1_92% SO1_98% SO1_100% SO2_75% SO2_85% SO2_92% SO2_98% SO2_100%

10 8 6

RH 100%: full saturation in the triaxial cell

4 2 0

0

50

100 150 Time (days)

200

250

Fig. 3. Relative mass variation of samples SO1 and SO2 vs. time and RH.

permeability) and the degree of interaction between the pore water and the clay unit layer (Agus et al., 2010). The latter is related to the chemistry of the pore water and the physicochemical characteristics of the clay. Since the water permeability of the compacted bentonite/sand mixture (the same as the materials used in this study) is extremely low, within the range of 10–20– 10–21 m2 (Liu, 2013), and the interaction between the clay aggregate and the water molecules is very active, it is not strange that such a long time is needed for the whole hydration process. In addition, samples SO1 and SO2 were prepared under the same conditions, although their total hydration time was not the same. This difference is related to the different plates used for the two tests: plastic plates (SO2) and steel plates (SO1). In fact, two steel plates were used for the first test, but the test lasted a very long time. Thus, it was decided to use plastic porous plates with large holes to accelerate the hydration process of the second test. In terms of the increase in relative mass at each RH step, their values are found to be quite close at lower and higher RH, e.g., 0.65% (SO1-RH¼ 75%) vs. 0.52% (SO2-RH¼ 75%), 5.86% (SO1-RH¼ 98%) vs. 5.80% (SO2-RH¼ 98%), 11.34% (SO1-RH¼ 100%) vs.11.24% (SO2-RH¼ 100%), while a little difference exists in the intermediate RH, e.g., 2.45% (SO1RH¼ 85%) vs. 1.91% (SO2-RH¼ 85%), 3.74% (SO1-RH¼

Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]]

6

1

Sr

0.9

SO1-VG model

Table 2 Parameters used for the VG model and the DP model.

SO1-DP model

Model no. VG model

DP model

SO2-VG model

0.8 0.7

n

SO2-DP model SO1-Experimental results

ω1

R2

P1

R2

P2

1.14 4.9  105 0.9533 0.3782 1.26  106 6.90  107 0.9968 1.14 5.3  105 0.9870 0.3990 1.52  106 8.20  107 0.9942

SO1 SO2

SO2-Experimental results

Pr

0.6 0.5

Suction (MPa) 1000.0 SO1-Constant volume-1.77

0.4

SO2-Constant volume-1.77 MX-80 Bentonite/sand mixture (70/30)--Constant volume-1.67 (Wang, 2012)

0.3 75%

Pure FEBEX bentonite-Constant volume-1.70 (Villar, 2007)

80%

85%

90%

95%

100%

RH

Pure MX-80 bentonite-Constant volume-1.70 (Marcial, 2003)

100.0

Fig. 4. Sorption isotherm of samples SO1 and SO2: the VG model and the DP model are used to fit the experimental data.

92%) vs. 3.21% (SO2-RH¼ 92%). The water retention capacity (total soil water potential ψ w ) of the soil mainly consists of two components, i.e., a matric component (ψ m ) and an osmotic component (ψ o ). The matric potential results from the combined effects of the capillary and the adsorptive forces within the soil matrix, which is associated with the pore structure of the sample. The osmotic potential is determined by the presence of solutes in the soil water. In fact, small changes in the waiting time before compaction (when the bentonite powder matures at RH¼ 85%, before compaction) or in the compaction process itself can lead to small changes in the distribution of the radius of the pores. And these changes will lead to the disparity of the matric potential (ψ m ) between the two samples, which may contribute to the minor difference in their increase in relative mass at RH¼ 85% and RH¼ 92%. Fig. 4 shows the relationship between the relative humidity and the degree of saturation. It is clear that this relationship (RH vs. Sr) is similar for both samples. The models of van Genuchten (VG) and Dual-Porosity (DP) were utilized to fit the experimental data, and the related parameters were determined by the least squares method (see Table 2). It is found that both models can simulate the experimental results very well. When comparing their correlation coefficients R2 (Table 2), it is found that the fitting effect of the DP model is better than that of the VG model, since the DP model adds a parameter. Feyen et al. (1998) and Liu (2012) indicated that the DP model cannot properly simulate the experimental results when Sr is close to 1; however, this has not been found in the present study. Perhaps it is related to the limited number of fitting points, namely, only four points are available for the curve fitting. Fig. 5 presents the water retention curves (WRCs: suction vs. w) of the two samples. As shown in the figure, suction gradually decreases with the increase of water content. Furthermore, it is observed that the WRCs of sample SO2 are similar to the WRCs of the bentonite/sand mixture measured by Wang et al. (2013), while their composition is the same (70% MX80 bentoniteþ 30% sand), but with a slight difference in dry density (1.78 g/cm3 vs. 1.67 g/cm3). When compared with pure bentonite, both FEBEX bentonite (ρd ¼ 1.70 g/cm3) and MX80

10.0

1.0 10

14

18

22

26

30

34

38

w (%)

Fig. 5. Water retention curves (Suction vs. w) obtained under constant-volume condition: SO1 and SO2 (ρd ¼1.77 g/cm3), MX-80 bentonite/sand mixtureWang (ρd ¼1.67 g/cm3), pure MX80 bentonite (ρd ¼ 1.70 g/cm3) and Pure FEBEX bentonite (ρd ¼ 1.70 g/cm3).

bentonite (ρd ¼ 1.70 g/cm3), it is clear that the water retention capacity of the mixture is smaller than that of pure bentonite if we overlook their minor differences of dry density. In fact, water is mainly absorbed by bentonite, rather than sand, and the water retention capacity will increase with the increase in bentonite content (Agus et al., 2010). The disparity attenuates as the water content decreases (or suction increases), which indicates that there is an insignificant contribution of the bentonite content to the water retention capacity of the sample at lower water contents. In addition, it is noted that at the same suction value, the amount of water absorbed by pure FEBEX bentonite is larger than that absorbed by pure MX80 bentonite. This means that the water retention capacity of the FEBEX bentonite is higher than that of the MX80 bentonite under constant-volume condition. A similar phenomenon was also observed by Villar and Lloret (2007). 4.1.2. Water retention tests under free swelling-condition: SF1 and SF2 The results of water retention tests (mr vs. time) of samples SF1 and SF2, obtained under free-swelling condition, are presented in Fig. 6. Similar to samples SO1 and SO2, the total swelling time of samples SF1 and SF2 is also quite long: 458 days (SF1) and 554 days (SF2). Considering that the size of the samples is only 12.5 (25) mm  42.5 mm, it is foreseen that the in situ saturation process will be an extremely long

Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]] Suction (MPa )

Relative mass variation (%) 30

1000.0

SF1_75% SF1_85% SF1_92% SF1_98% SF1_100% SF2_75% SF2_85% SF2_92% SF2_100% SF2_98%

25 20 15

7

SF1-unconfined-1.77 SF2-unconfined-1.77 MX80 bentonite/sand mixture (70/30)-unconfined-1.67(Wang, 2012) Pure MX80 bentonite-unconfined-1.70 (Marcial et al., 2003) Pure FEBEX bentonite-unconfined-1.76 (Lloret and Villar, 2007) Pure FEBEX bentonite-unconfined-1.67(Lloret and Villar, 2007)

100.0

10.0

1.0

10 5

0.1 10

0

20

30

40

50

60

70

w (%)

0

100

200

300 400 Time (days)

500

600

Fig. 6. Relative mass variation of samples SF1 and SF2 vs. time and RH.

period. The differences in swelling kinetics between the two samples may be attributed to the scaling effect or to small changes occurring during the preparation. It is particularly noticeable that at RH ¼ 100%, both the swelling time (TRH ¼ 100%) and the increase in relative mass (mr) are surprising. For example, TRH ¼ 100% accounts for 65.1% -SF1 (74.7%SF2) of the total swelling time, and mr at RH ¼ 100% is about 17.5%-SF1 (25.1%-SF2), which is 1.86-SF1 (2.82-SF2) times the mr at RH ¼ 98%. The difference in mr at RH ¼ 100% (17.5% vs. 25.1%) between the two samples may be attributed to the different swelling times at RH ¼ 100% (298 days vs. 413 days). When calculating the Sr , according to Eq. (9), the total pore volume (V P  total ) is assumed to be constant during the hydration process. This assumption does not stand under free-swelling condition since the sample volume changes a lot during the hydration process. This means that the Sr calculated with Eq. (9) is not suitable to express the water retention capacity of the samples which swell under freeswelling condition. Therefore, we do not present the RH-Sr curves for samples SF1 and SF2. Fig. 7 shows the water retention curves (suction-w) for the bentonite or bentonite/sand mixture measured under freeswelling condition. Once again, it is found that the behaviors of the WRCs obtained for samples SF1 and SF2 are similar to that of the bentonite/sand mixture (70% MX80 bentoniteþ 30% sand, ρd ¼ 1.67 g/cm3) measured by Wang et al. (2013). When comparing with the pure MX80 bentonite (ρd ¼ 1.70 g/cm3) and the FEBEX bentonite (ρd ¼ 1.76 g/cm3 or 1.67 g/cm3), the water retention capacity of the bentonite/sand mixture is smaller than that of pure bentonite. In addition, different with the WRCs measured under constant-volume condition, here, at lower water contents (w¼ 10–21%), the retention capacity of the pure FEBEX bentonite is similar to that of the pure MX80 bentonite, while at higher water contents (w421%), the retention capacity of pure FEBEX is higher than that of the pure MX80 bentonite (see Fig. 7). However, more results will be needed to verify

Fig. 7. Water retention curves (Suction vs. w) obtained under free-swelling condition: SF1 and SF2 (ρd ¼ 1.77 g/cm3), bentonite/sand mixture-Wang (ρd ¼1.67 g/cm3) (Wang, 2012), pure MX80 bentonite (ρd ¼1.70 g/cm3) and Pure FEBEX bentonite (ρd ¼ 1.67 or 1.76 g/cm3). Comment: the dry density is the initial dry density.

Relative volume variation

80% 70% 60%

Relative volume evolution with RH

50% 40%

30% 20%

10% 0% 75%

80%

85%

90%

95%

100%

RH Fig. 8. Relative volume variation of sample SF2 vs. RH (measured at the end of each RH step).

these observations. In addition, we find that under free-swelling condition, the initial dry density has little influence on the retention capacity of the FEBEX bentonite. For sample SF2, volume VRH was measured at the end of each adsorption stage, i.e., RH¼ 75%, 85%, 92%, 98%, and 100%. As can be seen from Fig. 8, the compacted bentonite/sand mixture displays a very good swelling ability under moist environment. In addition, this swelling capacity depends largely on the value of RH. For example, at lower RH, e.g., RH¼ 75%, the volume change is very small, only 0.69%, while at higher RH, the volume change is significant, e.g., at RH¼ 98% and 100%, the volume changes are 27.15% and 72.48%, respectively. This means that the swelling capacity of the compacted bentonite/sand mixture, under higher RH, is amazing. Under constant-volume condition, this moistureinduced deformation will change to a “swelling” pressure. In fact, at the initial state, the swelling of the compacted bentonite/sand

Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]]

8

Bentonite/sand mixture (70:30)

Pure bentonite

cracks

Photo 2. Pure bentonite and bentonite/sand mixture (70:30) (in dry state).

mixture will fill the initial gap between the seals and the host rock and may penetrate through the open fractures (e.g., in EDZ) into the surrounding host rock, which is favorable for the sealing of the disposal pits. However, once sealing is obatined, the swelling pressures of these compacted bentonite/sand mixtures will increase gradually with the time until becoming stable. If the swelling pressure is too high, it may lead to rock damage rather than having a positive supporting effect on the sidewall of the host rock. This is why sand is added to the bentonite powder, namely, to limit its swelling pressure to a proper value. In addition, cracking caused by free swelling or shrinkage is one of the major reasons of inaccuracy when measuring the volume change of the sample (Péron et al., 2007). Cracking phenomena are especially obvious for pure bentonite (see Photo 2). It is found that more cracks will be produced during the drying process for pure bentonite than for the bentonite/sand mixture used in this

Fig. 9. Schematic diagram of unsaturated bentonite–sand mixture (Agus et al., 2010). Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]]

study. Indeed, as shown in Fig. 9, a continuous sand matrix exists in the bentonite/sand mixture and leads to the shrinkage or the swelling of the clay aggregate partially within the voids of the sand matrix. Therefore, the compacted bentonite/sand mixture is less sensitive to drying or wetting when comparing with pure bentonite. Furthermore, as proposed by Péron et al. (2007), an improved ASTM Standard (ASTM D 2325, test method for capillarymoisture relationships for coarse- and medium-textured soils by porous-plate apparatus) can be used to avoid cracking when measuring the sample volume. 4.1.3. Comparison of water retention properties under constant-volume and free-swelling conditions Fig. 10 shows a comparison of the water saturation kinetics of the samples under both constant-volume and free-swelling conditions. Due to the different methods used under RH¼ 100%, for example, samples SO1 and SO2 were put in a triaxial cell and water was injected directly into the samples, while samples SF1 and SF2 were put in a hermetic chamber with RH¼ 100%, we did not plot the saturation time at RH¼ 100%. It can be observed that the confinement conditions have little effect on the saturation kinetics when RHr85%. With the increase of RH (RH485%), the difference in swelling kinetics becomes more and more pronounced. In particular, at RH¼ 100%, the difference in terms of the relative mass is noticeable, e.g., 17.50% (SF1) and 25.12% (SF2) vs. 11.34% (SO1) and 11.24% (SO2). As shown in Fig. 9, at lower water contents, the water will first fill the intra-aggregate pores and lead to the hydration and swelling of the clay aggregate, creating a soft gel (Kröhn, 2005; Cui et al., 2008; Ye et al., 2009a; Wang et al., 2013). This soft gel will fill the inter-aggregate pores, where enough space will be left in the inter-aggregate pores. In this phase, the effect of the intra-aggregate governing suction is predominant when comparing with the inter-aggregate capillary phenomenon (Romero et al., 1999). Before the inter-aggregate pores are fully filled with the swelling clay aggregate, the water absorption is insensitive to the confinement conditions. Then, with the increase of RH, under free-swelling condition, the volume of the inter-aggregate pores will increase due to the swelling of the intra-aggregate. As a result, more water will be absorbed. This is

Relative mass variation (%) 30 25 20 15

SO1_75%

SF1_75%

SO1_85%

SF1_85%

SO1_92%

SF1_92%

SO1_98%

SF1_98%

SO1_100%

SF1_100%

SO2_75%

SF2_75%

SO2_85%

SF2_85%

SO2_92%

SF2_92%

SO2_98%

SF2_98%

SO2_100%

SF2_100%

10 5 0

0

50

100

150

200

250

Time (days)

Fig. 10. Comparison of saturation kinetics obtained under both free-swelling and constant-volume conditions.

9

Suction (MPa ) 50 SO1-Constant-volume SO2-Constant-volume SF1-Free swelling SF2-Free swelling

40

30

20

10

0 10

18

26

34 w (%)

42

Fig. 11. Comparison of WRCs (Suction vs. w) obtained under both freeswelling and constant-volume conditions.

not the case when the swelling is confined. After the available inter-aggregate pores are fully filled with the expanding intraaggregate, no more water can be absorbed. In Fig. 11, a comparison of the suction-w relationship is presented. It is found that the suction-w relationship is independent of the confinement conditions when w is between 10% and 18%. With the increase of w, however, the influence of the confinement conditions becomes more and more obvious. Similar phenomena have also been observed by Delage et al. (1998), which the WRCs of compacted bentonite at higher suction levels are insensitive to the confinement conditions. As explained before, this is due to the intraaggregate governing suction at lower water contents. However, with the increase of water content, this difference becomes more and more significant, because at higher water contents, the inter-aggregate pores play a dominant role in the suction, while a large increase in the volume of the inter-aggregate pores is experienced by the compacted bentonite/sand mixture during the hydration process under free-swelling condition (Lloret and Villar, 2007). 4.2. Gas permeability tests on partially saturated bentonite/ sand mixture under confinement condition In the PGZ in situ experiment, the expected swelling pressure of the fully saturated bentonite/sand mixture is between 7–8 MPa (Liu, 2013). As indicated before, in order to investigate the gas tightness of the central unsaturated clay barrier under in situ confining stress, another series of samples (SF 3-1–SF 3-5) were used to perform gas permeability tests. As shown in Fig. 12, gas permeability is correlated to the water volume ratio (ew) defined as the ratio of water volume (Vw) to bentonite volume (Vb) (Toll, 1995; Romero et al., 2011; Wang et al., 2013): ew ¼

V w wGb ¼ Sr e ¼ Vb B

ð15Þ

Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]]

10

Gas permeability (m/s)

Gas permeability (m/s)

1.E-07

1.E-06

1.E-08

1.E-07

1.E-09

1.E-08

1.E-10

1.E-09 Pc=3 MPa (loading)

1.E-11

y = 4E-13e14.217x R² = 0.9762

1.E-10

Pc=5 MPa (loading)

1.E-11

Pc=8 MPa (loading)

1.E-12

Pc=5 MPa (unloading)

1.E-13 1.E-14

1.E-12

Pc=3 MPa (unloading)

0

0.2

0.4 0.6 Water volume ratio, ew

0.8

1

0

0.2

0.4

0.6

0.8

1

(1-e w)

Fig. 12. Gas permeability versus confining pressure and water volume ratio.

where w is the water content of the mixture, Gb is the specific gravity of bentonite (Gb ¼ 2:77), B is the bentonite content (in dry mass, B ¼ 0:7) in the mixture, and e is the void ratio of the mixture. It can be observed that gas permeability is very sensitive to the water volume ratio ew (or water content) and the confining pressure Pc (porosity). For example, at Pc ¼ 3 MPa, the gas permeability decreased about five orders of magnitude, while ew increased from 0.07 to 0.95. In addition, at the same ew (e.g., ew ¼ 0:95), the gas permeability decreased about two orders of magnitude, while the Pc increased from 3 MPa to 8 MPa. During the unloading phase, the hysteresis effect can be found. For example, the gas permeability measured at Pc ¼ 5 MPa, (ew ¼ 0:68, loading phase) is about 2.96  10–12 m/s, while the corresponding value is about 1.43  10–12 m/s at the unloading phase. This means that some inter-aggregate pores were still closed even though Pc had returned to the initial value. In addition, at Pc ¼ 8 MPa and ew ¼ 0:95, it is found that the gas permeability is extremely low, about 1.83  10–14 m/s. This means that tightness to gas can be achieved under in situ confining pressure (7–8 MPa) even though the sample is not fully saturated. In order to calculate the intrinsic permeability, the gas permeability was correlated to 1  ew . Here, only the gas permeability at Pc ¼ 3 MPa (loading phase) was correlated, because higher levels of confining pressure will clearly alter the porosity of the sample. As presented in Fig. 13, the gas permeability (Keff m/s) was correlated to 1 ew through the following equation: K ef f ¼ K int K rg ¼ 4  10  13 e14:27ð1  ew Þ

1.E-13

ð16Þ

According to this equation, intrinsic permeability K int is about
6.30 
10  7 m/s (or 8.70  10–14 m2, according to k m2 ¼ k m=s ðρg=μÞ, where μ is the dynamic viscosity of argon) if we set the value of ew to 0 in this equation (K rg will be 1 for the dry sample). In fact, we have tried to measure the intrinsic permeability of the dry bentonite/sand sample, and the order of magnitude is about 10–14 m2 (Liu, 2013). When comparing with the K int derived from the saturated water permeability measurements (Liu, 2013), the values for K int measured with

Fig. 13. Gas permeability measured at Pc ¼3 MPa (loading phase) as function of water volume ratio ð1 ew Þ.

gas (dry sample, 10–14 m2) are extremely higher than those measured with water (water saturated sample, 10–20–10–21 m2). Similar phenomena were also observed by Lloret and Villar (2007). They attributed this difference to the different accessible pore sizes to water and gas (0:3 μm vs:20 μm). In fact, for swelling clay, such as bentonite, it is difficult to measure the intrinsic permeability, since its pores microstructure will be changed significantly due to hydration (swelling) or drying (shrinkage).

5. Conclusion The aim of this study was to investigate the water retention properties of a compacted bentonite/sand mixture under both constant-volume and free-swelling conditions, and the gas permeability of a partially saturated bentonite/sand mixture under in situ confining stress. The results show that the confinement conditions have a significant influence on the saturation kinetics of the samples (at higher RH). More precisely, under freeswelling condition, it was found that more water was absorbed and the swelling period was much longer. In regard to WRCs, it was observed that the suction-w relationship is independent of the confinement conditions when the water content is within 10–18%, while this influence is obvious at higher water contents. This phenomenon can be attributed to the different swelling mechanisms under lower water contents (intra-aggregate governing suction) and higher water contents (inter-aggregate governing suction). In addition, under free-swelling condition, the volume change was amazing at higher RH. For example, at RH¼ 100%, the volume increased 72.48% when comparing with the initial volume (just after compaction). This is due to a large increase in the volume of the inter-aggregate pores experienced by the compacted bentonite/sand mixture during the hydration process. For gas permeability tests, it is clear that the gas permeability is very sensitive to the water volume ratio (or water content) and the confining pressure (or porosity). The gas permeability decreased gradually with the increase of

Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]]

the water volume ratio (or increase in confining pressure). When the sample (at ew ¼ 0:95) was subjected to in situ confining pressure (7–8 MPa), the gas permeability was very low (1.83  10–14 m/s); this indicates that tightness to gas can be obtained even though the sample is not fully saturated. In addition, the gas permeability at Pc ¼ 3 MPa (loading phase) was correlated with 1  ew . This relationship can be expressed by an empirical equation, K ef f ¼ 4  10  13 e14:27ð1  ew Þ , with a very high fitting accuracy (R2 ¼ 0.9762). According to this equation, the intrinsic gas permeability can be obtained when the value of ew is set to 0. A significant difference between the values of intrinsic permeability derived from the water and gas flow measurements hints that the microstructure of the pores will be changed greatly due to hydration (swelling) or drying (shrinkage). Acknowledgments The research leading to these results has received funding from the European Atomic Energy Community's Seventh Framework Programme (FP7/2007-2011) under Grant Agreement no. 230357, the FORGE project. The support from the State Key Laboratory for GeoMechanics and Deep Underground Engineering, China University of Mining & Technology (SKLGDUEK1201) is also greatly acknowledged. The authors are also very grateful to the editor and the anonymous reviewers for their elaborate reviews and valuable suggestions. References Abdullah, W.S., Alshibli, K.A., Al-Zou'bi, M.S., 1999. Influence of pore water chemistry on the swelling behavior of compacted clays. Appl. Clay Sci. 15 (5), 447–462. Agus, S.S., Schanz, T., Fredlund, D.G., 2010. Measurements of suction versus water content for bentonite-sand mixtures. Can. Geotechn. J. 47 (5), 583–594. Alonso, E.E., Olivella, S., Arnedo, D., 2006. Mechanisms of gas transport in clay barriers. J. Iber. Geol. 32 (2), 175–196. Andra, D., 2005a. Référentiel des matériaux d'un stockage de déchets à haute activité et à vie longue. Tome 4: Les matériaux à base d'argilites excavé et remaniées. Rapport Andra no. CRPASCM040015B. Andra Dossier, 2005b. Dossier 2005 Tome Argile Tome Architecture and Management of a Geological Repository. Technical Report. Barnichon, J.D., Deleruyelle, F., 2009. Sealing Experiments at the Tournemire URL. EUROSAFE. Cho, W.J., Lee, J.O., Chun, K.S., 1999. The temperature effects on hydraulic conductivity of compacted bentonite. Appl. Clay Sci. 14 (1), 47–58. Cui, Y.J., Tang, A.M., Loiseau, C., Delage, P., 2008. Determining the unsaturated hydraulic conductivity of a compacted sand-bentonite mixture under constant-volume and free-swell conditions. Phys. Chem. Earth, Parts A/B/C 33 (Suppl. 1), S462–S471. Cui, Y.J., Tang, A.M., Qian, L.X., Ye, W.M., Chen, B., 2011. Thermal– mechanical behavior of compacted GMZ bentonite. Soils Found. 51 (6), 1065–1074. Delage, P., Cui, Y.J., Yahia A.ï. M., De Laure, E., 1998. On the saturated hydraulic conductivity of a dense compacted bentonite. In: Proceedings of the 2nd International Conference on Unsaturated Soils, vol. 1. International Academic Publishers. Beijing, pp. 344–349. Didier, G., Bouazza, A., Cazaux, D., 2000. Gas permeability of geosynthetic clay liners. Geotext. Geomembr. 18 (2), 235–250. Durner, W., 1994. Hydraulic conductivity estimation for soils with heterogeneous pore structure. Water Resour. Res. 30 (2), 211–223.

11

Feyen, J., Jacques, D., Timmerman, A., Vanderborght, J., 1998. Modelling water flow and solute transport in heterogeneous soils: a review of recent approaches. J. Agric. Eng. Res. 70 (3), 231–256. Galvin, K.P., 2005. A conceptually simple derivation of Kelvin equation. Chem. Eng. Sci. 60, 4659–4660. Gerke, H.H., Genuchten, M.V., 1993. A dual-porosity model for simulating the preferential movement of water and solutes in structured porous media. Water Resour. Res. 29 (2), 305–319. Greenspan, L., 1977. Humidity fixed points of binary saturated aqueous solutions. J. Res. Natl. Bur. Stand. A. Phys. Chem. 81A, 89–96. Hoffmann, C., Alonso, E.E., Romero, E., 2007. Hydro-mechanical behavior of bentonite pellet mixtures. Phys. Chem. Earth, Parts A/B/C 32 (8), 832–849. Komine, H., Ogata, N., 1994. Experimental study on swelling characteristics of compacted bentonite. Can. Geotechn. J. 31 (4), 478–490. Komine, H., Ogata, N., 1999. Experimental study on swelling characteristics of sand–bentonite mixture for nuclear waste disposal. Soils Found. 39 (2), 83–97. Kröhn, K.P., 2005. Results and interpretation of bentonite resaturation experiments with liquid water and water vapour, Unsaturated Soils: Experimental Studies. Springer, Berlin Heidelberg, 257–269. Liu, J., 2012. Etude expérimentale de la perméabilité relative des matériaux cimentaires et simulation numérique du transfert d’eau dans le béton (Ph.D. thesis). Ecole central de Lille, France (in French). Liu, J.F., 2013. Sealing Efficiency of an Argillite-Bentonite Plug Subjected to Gas Pressure, in the Context of Deep Underground Nuclear Waste Storage (Ph.D. thesis). Ecole central de Lille, France (in French). Lloret, A., Villar, M.V., 2007. Advances on the knowledge of the thermohydro-mechanical behavior of heavily compacted “FEBEX” bentonite. Phys. Chem. Earth, Parts A/B/C 32 (8), 701–715. Maekawa, K., Chaube, R., Kishi, T., 1999. Modeling of Concrete Performance: Hydration, Microstructure Formation and Mass Transport. Routledge, Taylor & Francis. Martin, P.L., Barcala, J.M., Huertas, F., 2006. Large-scale and long-term coupled thermo-hydro-mechanic experiments with bentonite: the febex mock-up test. Journal of Iberian Geology 32 (2), 259–282. Mollins, L.H., Stewart, D.I., Cousens, T.W., 1996. Predicting the properties of bentonite–sand mixtures. Clay Miner. 31 (2), 243–252. Ng, C.W., Pang, Y.W., 2000a. Influence of stress state on soil–water characteristics and slope stability. J. Geotechn. Geoenviron. Eng. 126 (2), 157–166. Ng, C.W., Pang, Y.W., 2000b. Experimental investigations of the soil–water characteristics of a volcanic soil. Can. Geotechn. J. 37 (6), 1252–1264. Pellicer, J., Garcia-Morales, V., Hermandez, et M.J., 2000. On the demonstration of the Young–Laplace equation in introductory physics courses. Trans. Am. Geophys. Union 38, 126–129. Péron, H., Hueckel, T., Laloui, L., 2007. An improved volume measurement for determining soil water retention curves. Geotechn. Test. J. 30 (1), 1. Pierre, D., 2006. Some microstructure effects on the behavior of compacted swelling clays used for engineered barriers. Chin. J. Rock Mech. Eng. 25, 4. Romero, E., Della Vecchia, G., Jommi, C., 2011. An insight into the water retention properties of compacted clayey soils. Géotechnique 61 (4), 313–328. Romero, E., Gens, A., Lloret, A., 1999. Water permeability, water retention and microstructure of unsaturated compacted Boom clay. Eng. Geol. 54 (1), 117–127. Sällfors, G., Öberg-Högsta, A.L., 2002. Determination of hydraulic conductivity of sand-bentonite mixtures for engineering purposes. Geotechn. Geol. Eng. 20 (1), 65–80. Siemens, G., Blatz, J.A., 2009. Evaluation of the influence of boundary confinement on the behavior of unsaturated swelling clay soils. Can. Geotechn. J. 46 (3), 339–356. Studds, P.G., Stewart, D.I., Cousens, T.W., 1998. The effects of salt solutions on the properties of bentonite–sand mixtures. Clay Miner. 33 (4), 651–660. Tang, A.M., Cui, Y.J., 2010. Effects of mineralogy on thermo-hydromechanical parameters of MX80 bentonite. J. Rock Mech. Geotechn. Eng. 2 (1), 91–96. Tang, C.S., Huang, L.M., Wang, J., Liu, Y.M., 2013. Influence of boundary condition on the swelling behavior of GMZ buffer/backfilling material in HLW repository, Rock Characterisation, Modelling and Engineering Design Methods, 199.

Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011

12

J.-F. Liu et al. / Soils and Foundations ] (]]]]) ]]]–]]]

Thomson, W., 1871. On the equilibrium of a vapour at a curved surface of liquid. Phil. Mag. 42, 448–452. Toll, D.G., 1995. A conceptual model for the drying and wetting of soil. In: Proceedings of the First International Conference on Unsaturated Soils. Paris, France, vol. 2, pp. 805–810. van Genuchten, M.T., 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44 (5), 892–898. Vangpaisal, T., Bouazza, A., 2004. Gas permeability of partially hydrated geosynthetic clay liners. J. Geotechn. Geoenviron. Eng. 130 (1), 93–102. Villar, M.V., 2007. Water retention of two natural compacted bentonites. Clays Clay Miner. 55 (3), 311–322. Villar, M.V., Gómez-Espina, R., 2007. Retention curves of two bentonites at high temperature, Experimental Unsaturated Soil Mechanics. Springer Proceedings in Physics Volume 112, Berlin, Heidelberg, 267–274. Villar, M.V., Lloret, A., 2004. Influence of temperature on the hydromechanical behavior of a compacted bentonite. Appl. Clay Sci. 26 (1), 337–350. Villar, M.V., Lloret, A., 2007. Dismantling of the first section of the FEBEX in situ test: THM laboratory tests on the bentonite blocks retrieved. Phys. Chem. Earth, Parts A/B/C 32 (8), 716–729. Villar, M., Lloret, A., 2008. Influence of dry density and water content on the swelling of a compacted bentonite. Appl. Clay Sci. 39 (1), 38–49. Villar, M.V., Martín, P.L., Romero, F.J., Barcala, J.M., 2012. Results of the Tests on Bentonite (Part 1). FORGE Report D3.15.

Wang, Q., 2012. Hydro-Mechanical Behavior of Bentonite-Based Materials used for High-Level Radioactive Waste Disposal (Ph.D. thesis). Université Paris-Est, France. Wang, Q., Tang, A.M., Cui, Y., Delage, P., 2012. Experimental study on the swelling behavior of bentonite/claystone mixture. Eng. Geol. 124, 59–66. Wang, Q., Tang, A.M., Cui, Y.J., Delage, P., Barnichon, J.D., Ye, W.M., 2013. The effects of technological voids on the hydro-mechanical behavior of compacted bentonite–sand mixture. Soils Found. 53 (2), 232–245. Wang, Q., Tang, A.M., Cui, Y., Delage, P., Gatmiri, B., 2014. Long-term effect of water chemistry on the swelling pressure of a bentonite-based material. Appl. Clay Sci. 87, 157–162. Ye, W.M., Cui, Y.J., Qian, L.X., Chen, B., 2009a. An experimental study of the water transfer through confined compacted GMZ bentonite. Eng. Geol. 108 (3), 169–176. Ye, W.M., Wan, M., Chen, B., Chen, Y.G., Cui, Y.J., Wang, J., 2009b. Effect of temperature on soil–water characteristics and hysteresis of compacted Gaomiaozi bentonite. J. Central South Univ. Technol. 16, 821–826. Ye, W.M., Wan, M., Chen, B., Chen, Y.G., Cui, Y.J., Wang, J., 2013. Temperature effects on the swelling pressure and saturated hydraulic conductivity of the compacted GMZ01 bentonite. Environ. Earth Sci. 68 (1), 281–288. Zhu, C.M., Ye, W.M., Chen, Y.G., Chen, B., Cui, Y.J., 2013. Influence of salt solutions on the swelling pressure and hydraulic conductivity of compacted GMZ01 bentonite. Eng. Geol. 166, 74–80.

Please cite this article as: Liu, J.-F., et al., Experimental research on water retention and gas permeability of compacted bentonite/sand mixtures. Soils and Foundations (2014), http://dx.doi.org/10.1016/j.sandf.2014.09.011