Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Impact of pore fluid chemistry on mechanical and thermal behavior of clay-based materials Tabiatnejad, Bardia 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2017_february_tabiatnejad_bardia.pdf [ 5.5MB ]
Metadata
JSON: 24-1.0340683.json
JSON-LD: 24-1.0340683-ld.json
RDF/XML (Pretty): 24-1.0340683-rdf.xml
RDF/JSON: 24-1.0340683-rdf.json
Turtle: 24-1.0340683-turtle.txt
N-Triples: 24-1.0340683-rdf-ntriples.txt
Original Record: 24-1.0340683-source.json
Full Text
24-1.0340683-fulltext.txt
Citation
24-1.0340683.ris

Full Text

IMPACT OF PORE FLUID CHEMISTRY ON MECHANICAL AND THERMAL BEHAVIOR OF CLAY-BASED MATERIALS  by  Bardia Tabiatnejad    A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE COLLEGE OF GRADUATE STUDIES  (Civil Engineering)    THE UNIVERSITY OF BRITISH COLUMBIA  (Okanagan)  January 2017   © Bardia Tabiatnejad, 2017  iThe undersigned certify that they have read, and recommend to the College of Graduate Studies for acceptance, a thesis entitled:     submitted by                                Bardia Tabiatnejad  in partial fulfilment of the requirements of the degree of              Doctor of Philosophy  .    Dr. Sumi Siddiqua, School of Engineering  Supervisor   Dr. Greg Siemens, Department of Civil Engineering, Royal Military College    Co‐Supervisor  Dr. Kasun Hewage, School of Engineering   Supervisory Committee Member  Dr. Liwei Wang, School of Engineering   University Examiner  Dr. Jinyuan Liu, Department of Civil Engineering, Ryerson University  External Examiner, Professor (please print name and university in the line above)   October 18, 2016  (Date Submitted to Grad Studies)   iiAbstract  The impact of pore fluid salinity on thermal and mechanical properties of clay-based materials are examined in this thesis. The proposed concept for deep geological repository in Canada consists of a series of clay-based material to support the used-fuel bundles in place and minimize the local groundwater infiltration, which is highly saline. The coupled process of local groundwater infiltration and thermal desiccation in the engineered barrier causes the material to experience unsaturated conditions with increased salt content levels. In the previous studies, the behavior of the material has been investigated in the distilled water conditions, which was necessary to consider the change in the pore fluid salinity considering the infiltration of the groundwater. This research investigates the impact of the pore fluid salinity on the mechanical behavior and thermal conductivity of the clay-based materials which are designed to be use in the repository in future. To accomplish this, three experimental testing programs have been carried out.  The first experimental program examined the shrinkage and stress-strain behavior of bentonite-sand buffer specimens prepared with high and low salt concentrations. The results were compared to the available data from the literature for the specimens prepared with distilled water. Specimens were dried out in desiccators and installed in triaxial cells. The soil water characteristic, compression and shear strength of the specimens were presented and impact of the pore fluid salinity was discussed.   iiiThe second experimental program evaluated the effect salinity of the pore fluid on the behavior of the light backfill material in unsaturated conditions. Specimens prepared with distilled water and saline solution were tested and their compressibility and shear strength were discussed.   The third experimental program investigated the impact of pore fluid salinity on the thermal conductivity of mixtures of bentonite and silica sand at low and high dry densities. In continue, a number of thermal conductivity prediction models were selected to calculate the thermal conductivity values of the material with distilled water and saline pore fluids. The calculated values were compared to the obtained experimental data and a suitable model for predicting thermal conductivity in the saline pore fluid condition was presented.    ivPreface  A version of chapter three has been published in the journal of Environmental Earth Sciences.  Tabiatnejad, B., Siddiqua, S. and Siemens, G. (2016), "Impact of pore fluid salinity on the mechanical behavior of unsaturated bentonite-sand mixture", Environmental Earth Sciences, vol. 75, no. 22, pp. 1434. I wrote the manuscript, which was further edited by Dr. Sumi Siddiqua and Dr. Greg Siemens.  A version of chapter five has been published in the conference proceedings of 69th Canadian Geotechnical Conference, GeoVancouver 2016, Vancouver, Canada, 2 -5 October 2016; Tabiatnejad, B., Siddiqua, S., and Siemens, G. “Thermal properties of an engineered barrier in the presence of saline pore fluid”. I prepared the specimens, measured the experimental data based on the appropriate procedures, and analyzed the laboratory data. My supervisors guided me through the process and reviewed and finalized the conference paper.   vTable of Contents  Abstract .................................................................................................................................... ii Preface ..................................................................................................................................... iv Table of Contents .................................................................................................................... v List of Tables ........................................................................................................................ viii List of Figures ......................................................................................................................... ix List of Symbols, Abbreviations and Mathematical Notations .......................................... xv Acknowledgements ............................................................................................................ xviii Dedication ............................................................................................................................. xix Chapter 1: Introduction ....................................................................................................... 20 1.1  Background ......................................................................................................................... 20 1.2  Research Question............................................................................................................... 24 1.3  Research Objectives ............................................................................................................ 26 1.4  Thesis Structure and Organization ...................................................................................... 27 Chapter 2: Shrinkage and Mechanical Behavior of Bentonite-sand Buffer ................... 29 2.1  Overview ............................................................................................................................. 29 2.2  Materials and Sample Preparation ...................................................................................... 29 2.3  Experimental Program ........................................................................................................ 34 2.3.1  Suction Measurements .................................................................................................... 37 2.3.2  High Pressure triaxial Equipment ................................................................................... 39 2.4  Results and Discussions ...................................................................................................... 41  vi2.4.1  Constant Suction Drying ................................................................................................. 41 2.4.2  Isotropic Compression .................................................................................................... 46 2.4.3  Triaxial Shearing ............................................................................................................ 55 2.5  Experimental Limitations .................................................................................................... 58 2.6  Summary ............................................................................................................................. 61 Chapter 3: Mechanical behavior of Light Backfill ............................................................ 62 3.1  Overview ............................................................................................................................. 62 3.2  Materials and Methods ........................................................................................................ 62 3.2.1  Light Backfill Material ................................................................................................... 62 3.2.2  Solution Preparation ....................................................................................................... 64 3.2.3  Specimen Preparation ..................................................................................................... 64 3.2.4  Specimen Compaction .................................................................................................... 64 3.2.5  Triaxial Equipment ......................................................................................................... 65 3.3  Experimental Program ........................................................................................................ 66 3.3.1  Results and Discussions .................................................................................................. 68 3.3.1.1  Soil Water Characteristic Curve ............................................................................ 68 3.3.1.2  Isotropic Compression ........................................................................................... 71 3.3.1.3  Triaxial Shear Strength .......................................................................................... 76 3.4  Summary ............................................................................................................................. 81 Chapter 4: Effect of pore fluid salinity on thermal conductivity of bentonite-sand mixtures and evaluation of prediction models ................................................................... 83 4.1  Overview ............................................................................................................................. 83 4.2  Materials ............................................................................................................................. 83 4.3  Thermal Conductivity Measurement ................................................................................... 85 4.4  Results and Discussions ...................................................................................................... 91  vii4.5  Calculated Thermal Conductivity Values using Prediction Models ................................... 94 4.6  Summary ........................................................................................................................... 108 Chapter 5: Conclusions and Recommendations .............................................................. 110 5.1  Summary and Conclusions ................................................................................................ 110 5.2  Originality and Contribution ............................................................................................. 111 5.3  Applications ...................................................................................................................... 111 5.4  Limitations and Recommendations ................................................................................... 112 References ............................................................................................................................ 114   viiiList of Tables  Table 2.1 Summary of Wyoming bentonite material properties (Blatz et al. 2007) ............... 30 Table 2.2 Proportioning specification used for Frac sand component .................................... 30 Table 2.3 Shear strength parameters for BSB specimens prepared with 100 g/L CaCl2 ....... 57 Table 2.4 Shear strength parameters for BSB specimens prepared with 250 g/L CaCl2 ....... 57 Table 3.1 Properties of the bentonite component of Light backfill (after Graham et al. 1997) ............................................................................................................................... 63 Table 4.1 Test Series for Thermal Conductivity Measurements ............................................ 84 Table 4.2 Mean value of R (μ) and root mean square error (δ) values for the selected thermal conductivity prediction models for the specimens prepared at target dry density of 1.67 Mg/m3 ...................................................................................................... 100 Table 4.3 Mean value of R (μ) and root mean square error (δ) values for the selected thermal conductivity prediction models for the specimens prepared at target dry density of 1.24 Mg/m3 ...................................................................................................... 107 Table 5.1 Limitations and recommendations in this study ................................................... 113  ixList of Figures  Figure 1.1 Illustration of Deep Geological Repository Concept (NWMO 2015) ................... 22 Figure 1.2 Conceptual schematic of a deep geological repository showing anticipated evolution of thermal energy, degree of saturation and pore water saline concentration distributions with time (from Tabiatnejad et al. 2016) ................... 25 Figure 2.1 Frac sand size ranges provided by SIL industrial minerals ................................... 31 Figure 2.2 Static compaction unit in the geotechnical lab at the Royal Military College ...... 32 Figure 2.3 Compacted Bentonite-sand buffer specimen prepared with distilled water .......... 33 Figure 2.4 Schematic view of the setup for desiccating the BSB specimens ......................... 35 Figure 2.5 Constant suction drying in the desiccators (a) immediately after compaction (b) after the suction equilibrium .................................................................................. 36 Figure 2.6 Desiccated BSB specimen before installing in the triaxial cell ............................. 37 Figure 2.7 Suction tip, RH sensor and signal amplifier for measuring internal total suction levels in the specimens (Tabiatnejad et al. 2016) .................................................. 38 Figure 2.8 A BSB sample installed in the high-pressure triaxial apparatus ........................... 40 Figure 2.9 Total suction versus bulk water content for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2.................................................... 41 Figure 2.10 Total suction versus degree of saturation for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2.................................................... 42 Figure 2.11 Volumetric strain versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2.................................................... 44  xFigure 2.12 Dry density versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2 .............................................................. 45 Figure 2.13 Dry density versus degree of saturation for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2.................................................... 46 Figure 2.14 Volumetric strain versus mean stress for BSB specimens with distilled water, 100 g/L and 250 g/ L CaCl2 .................................................................................. 47 Figure 2.15 EMDD versus total suction for BSB specimens with distilled water, 100 g/L and 250 g/L CaCl2 ................................................................................................. 48 Figure 2.16 Loading modulus versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2.................................................... 49 Figure 2.17 Loading modulus versus degree of saturation for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2 .......................................... 50 Figure 2.18 Virgin compression modulus versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2 ............................ 51 Figure 2.19 Virgin compression modulus versus degree of saturation for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2 .................... 52 Figure 2.20 Unloading modulus versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2.................................................... 53 Figure 2.21 Unloading modulus versus degree of saturation for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2 ............................ 54 Figure 2.22 Preconsolidation stress versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2 .......................................... 55  xiFigure 2.23 Deviator stress versus axial strain for BSB specimens with distilled water and 250 g/L CaCl2 ........................................................................................................ 56 Figure 2.24. Leaked oil into (a) an as-compacted specimen (b) a desiccated specimen ........ 58 Figure 2.25. Fixing and improving the sealing of different parts of the setup such as (a) sharp edges of the pedestal cap (b) wrapped sharp edges with geotextile (c) smoothing sharp edges on the membrane stretcher (d) using new plugs on the top cap .......................................................................................................................... 60 Figure 3.1 Close view of the LBF material as provided by AECL......................................... 63 Figure 3.2 Typical compacted LBF specimen for (a) Triaxial experiments (b) suction measurements ......................................................................................................... 65 Figure 3.3 Installed specimen in the GDS Triaxial cell with on specimen Hall Effect transducer (a) front view (b) side view .................................................................. 66 Figure 3.4 Suction measurement setup for LBF specimens ................................................... 67 Figure 3.5 Soil water characteristic curves for LBF specimens prepared with distilled water and 250 g/L solution versus degree of saturation .................................................. 68 Figure 3.6 Soil water characteristic curves for LBF specimens prepared with distilled water and 250 g/L solution versus bulk water content .................................................... 69 Figure 3.7 Isotropic compression curves for LBF specimens with distilled water and 250 g/L CaCl2 solution ................................................................................................. 72 Figure 3.8 Loading modulus (κi) for LBF with distilled water and 250 g/L CaCl2 solution at varying degrees of saturation ............................................................................. 73 Figure 3.9 Compression Loading modulus (λ) for LBF with distilled water and 250 g/L CaCl2 solution at varying degrees of saturation .................................................... 74  xiiFigure 3.10 Unload modulus (κu) for LBF with distilled water and 250 g/L CaCl2 solution at varying degrees of saturation ............................................................................. 75 Figure 3.11 Pre-consolidation pressure values for LBF with distilled water and 250 g/L CaCl2 solution at varying degrees of saturation .................................................... 76 Figure 3.12 Stress-strain curves for LBF specimens with distilled water at different degrees of saturation ........................................................................................................... 77 Figure 3.13 Stress-strain curves for LBF specimens with 250 g/L CaCl2 at different degrees of saturation ........................................................................................................... 77 Figure 3.14 Typical failure mode of sheared specimens ........................................................ 78 Figure 3.15 E1% for the specimens with distilled water and 250 g/L CaCl2 solution ........... 79 Figure 3.16 E50 for the specimens with distilled water and 250 g/L CaCl2 solution ............ 80 Figure 3.17 qeot for the specimens with distilled water and 250 g/L CaCl2 solution ............ 81 Figure 4.1 KD2 Pro thermal meter and TR-1 Sensor (from Rozanski and Sobotka 2015) .... 85 Figure 4.2 Typical temperature measurements by KD2 Pro Decagon thermal meter for both heating and cooling cycles ..................................................................................... 88 Figure 4.3 Linear portion of the temperature measurement plot for (a) heating and (b) cooling phases ........................................................................................................ 89 Figure 4.4 Schematic illustration of the thermal needle probe inserted in the cylindrical specimen ................................................................................................................ 90 Figure 4.5 Thermal Conductivity measurement of the bentonite-sand specimen prepared at (a) 1.67 Mg/m3 dry density with (a) 250 g/L CaCl2 solution at ~ 40% degree of saturation and (b) completely dry specimen .......................................................... 91  xiiiFigure 4.6 Thermal conductivity values for the specimens with dry density of 1.67 Mg/m3 prepared with distilled water and saline solutions ................................................. 92 Figure 4.7 Thermal conductivity values for the specimens with target dry density of 1.24 Mg/m3 prepared with distilled water and saline solutions .................................... 93 Figure 4.8 Evaluation of. Johansen (1977) model for the BSB specimens prepared at 1.67 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions ................... 95 Figure 4.9 Evaluation of. Cote and Konrad (2005) model for the BSB specimens prepared at 1.67 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions ....... 96 Figure 4.10 Evaluation of. Lu et al. (2007) model for the BSB specimens prepared at 1.67 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions ................... 97 Figure 4.11 Evaluation of. Chen (2008) model for the BSB specimens prepared at 1.67 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions ................... 98 Figure 4.12 Evaluation of. Bruggeman (1935) model for the BSB specimens prepared at 1.67 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions ........... 99 Figure 4.13 Evaluation of. Johansen (1977) model for the BSB specimens prepared at 1.24 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions ................. 102 Figure 4.14 Evaluation of. Cote and Konrad (2005) model for the BSB specimens prepared at 1.24 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions ..... 103 Figure 4.15 Evaluation of. Lu et al. (2007) model for the BSB specimens prepared at 1.24 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions ................. 104 Figure 4.16 Evaluation of. Chen (2008) model for the BSB specimens prepared at 1.24 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions ................. 105  xivFigure 4.17 Evaluation of. Bruggeman (1935) model for the BSB specimens prepared at 1.24 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions ......... 106  xvList of Symbols, Abbreviations and Mathematical Notations NWMO Nuclear Waste Management Organization BSB Bentonite-Sand Buffer LBF Light-Backfill DBF Dense Backfill NaCl Sodium Chloride CaCl2 Calcium Chloride Ψ Soil suction or total suction R Universal (molar) gas constant T Absolute temperature ߭௪଴ Specific volume of water or the inverse of water ߱௩ Molecular mass of water vapor ݑത௩ Partial pressure of pore water vapor  ݑത௩଴ Saturation pressure of water vapor RH Relative Humidity HAE High Air Entry TCS Thermal Conductivity Sensor σᇱ Effective Normal Stress  σ Normal Stress u୵ Pore Water Pressure τ୤ Shear strength on the failure plane at failure cᇱ Effective cohesion    xviሺσ୤ െ u୵ሻ୤	 Effective normal stress on the failure plane at failure σ୤୤ Total normal stress on the failure plane at failure u୵୤ Pore water pressure at failure ∅ᇱ Effective angle of internal friction ݁ Current void ratio ݁଴ Initial void ratio ߣ Isotropic logarithmic compression index ݌ᇱ Current mean effective stress  ݁௬ Void ratio at yielding point ݇ Swelling Index ݌௬ᇱ  Effective stress at yielding point ܯ Critical stress ratio ݍ Deviator Stress κ Thermal Conductivity  ܣ Cross Section Area ݈ Length of the specimen ܦ Thermal Diffusivity th Heating time ∆ܶ Partial pressure of pore water vapor  Q Heat input per unit length of heater λ Thermal conductivity ݎ Distance from the heated needle ܭ௖ Thermal Conductivity Sensor  xviiܭ௠,௣ Predicted Thermal Conductivity ௗܲ Volume fraction of the spherical particles ܭ௘ Kersten number ܵ௥ Degree of Saturation ܭ௦௔௧ Thermal conductivity at saturated condition ܭௗ௥௬ Thermal Conductivity at dry condition ߩௗ Dry density ߩ௦ Density of solid phase ݊ Porosity 	ܭ௜ Thermal conductivity of ice ܭௌ௢௟௜ௗ Thermal conductivity of solid phase ܭௐ௔௧௘௥ Thermal conductivity of water χ Particle shape effect  a Empirical parameter in Lu et al. (2007) b Empirical parameter in Lu et al. (2007) α Soil type effect in Lu et al. (2007) Mw  Molecular weight of water ρw Density of water ܵ Suction  w Bulk water content μ Mean value of R  δ Root mean square error    xviiiAcknowledgements  Sincerest gratitude to my advisors Dr. Sumi Siddiqua and Dr. Greg Siemens for their guidance and mentorship during this research program. They always provided their time and knowledge to help me develop both academically and professionally.  Thank you to Dr. Kasun Hewage for serving as my Ph.D. committee member, and Dr. Ray Taheri for serving as external examiner in my Ph.D. proposal defense.   Thanks to Kristine Mattson, Dexter Gaskin, Steve VanVolkinburgh and Kim Nordstrom, lab technicians at Royal Military College of Canada and UBC Okanagan for their support, friendship, and help during my experiments.  Special thanks to my dear friend Dr. Yazan Qasrawi, for his friendship in the difficulties I faced during my Ph.D. program. Thanks to my fellow graduate students at the geotechnical research group at UBC Okanagan, especially Mr. Amin Bigdeli.  Financial support is acknowledged from the Nuclear Waste Management Organization and the National Science and Engineering Research Council of Canada.  My deepest thanks go to my family. I do not have suitable words to fully describe their endless moral, emotional, and financial support throughout my years of education.   xixDedication  To my parents and my brother!   20Chapter 1: Introduction  1.1 Background  Clays are formed by chemical and mechanical decomposition of small rock particles (Mitchell 1993). Clay is a raw material with more than hundred applications in different industries such as agriculture, engineering, and construction. Their properties are greatly dependent on their mineral structure and composition. Numerous different clay minerals form various types of clays. Regarding the various properties of each clay type, their industrial applications are different. Bentonite is one of the most popular clay types with numerous applications in industry (Murray H.H. 2006).  One type of clay, bentonite, is mainly composed of the smectite group, which their predominate minerals are calcium and sodium montmorillonite minerals. Calcium montmorillonite can be found in many regions of the world, but sodium bentonite is rare. The most popular places for sodium bentonite are in the states of Montana and Wyoming in the United States. The smectite mineral group has unique physical and chemical properties such as high expandable layers, high surface area, high absorption capacity, high swelling capacity, and high viscosity. Considering these properties, they are used in numerous industrial applications (e.g. Drilling muds, clay barriers, ceramics and refractions, cosmetics, desiccants, medicines, etc.) (Murray H.H. 2006).  One industry that will exploit the physical and chemical properties for protection of the environments is long-term isolation of used nuclear fuel. Used nuclear fuel remains highly radioactive for a long time following removal form the reactors and needs proper shielding and  21protection to avoid environmental risks (NWMO 2005). One of the options for disposing the nuclear used fuel, which many countries are working on to develop their regional concept, is the deep geological repository (DGR) (Sarkar and Siddiqua 2016). In this method, used fuel bundles will be isolated from the environment through a multi-barrier sealing system (NWMO 2015). Owing to the high swelling capability and low permeability of sodium bentonite, it is one of the main material options to be used in the design of the engineered barriers in the Canadian concept of DGR. The clay-based engineered barriers facilitate sealing the nuclear used fuel containers, inhibits the groundwater infiltration into the layers, and cuts off the microbial activity (Gates et al. 2009, Zheng et al. 2015). They will be placed around the containers and fill the gap with the host rock. Figure 1.1 illustrates a schematic view of the Canadian concept of DGR.  The engineered barriers are designed to support the containers in place and function under different nontypical environmental stresses for a lengthy period of hundreds to thousands of years (Siddiqua et al. 2014). The engineered barriers, host rock and other components of the deep geological repository would undergo coupled hydraulic-mechanical, thermal and biological processes to reach an equilibrium situation after a long-time. During their operation time, they will be subjected to moisture gain and loss and stress changes (NWMO 2005, Garcia-Bengochea et al. 1979; Hoffman et al. 2006).  The swelling, permeability and strength of the barriers are dependent on their initial dry density and composition. Different barriers are designed based on their specific application in the repository, which are highly compacted 100% bentonite (HCB), bentonite-sand buffer  22(BSB), light backfill (LBF), low to medium dense gap fill (GF) and dense backfill (DBF) (Siddiqua et al. 2011a&b).    Figure 1.1 Illustration of Deep Geological Repository Concept (NWMO 2015)  During the operation life of the repository, local groundwater permeates into the disposal vault from the host rock to the used fuel containers which results in hydrating the clay-based engineered barriers. The local ground water of the potential locations proposed for  23construction of the geological repository is high in salt concentration which changes the pore fluid chemistry of the layers through hydrating them. Since the physical properties of clay are highly sensitive to the salinity of the constituent pore fluid (Siddiqua et al. 2014) infiltration of the saline groundwater can change the behavior of the engineered barriers within the repository. According to a review by Nuclear Waste Management Organization (NWMO 2005) crystalline rock and sedimentary rock are introduced as potential host media for a geological repository in Canada. Data collected by Gascoyne et al. (1987) and Mazurek (2004) shows the presence of a great amount of dissolved salt in the local groundwater at proposed depth for the repository. In crystalline rock of the Canadian shield total dissolved solids (TDS) varies from 8 to more than 100 g/L at the mentioned depth under the ground., and in sedimentary rock TDS is greater than 200 g/L. Salt speciation is typically Na‐Ca‐Cl at shallow depths, changing to Ca‐Na‐Cl at greater depths (Priyanto et al. 2008).  On the other hand, the used fuel containers are at elevated temperatures (up to 100°C) at the installation time in the emplacement rooms in the repository vault. Due to the radioactive degradation process of the used fuel, they keep generating heat for an extended period which develops a thermal gradient flowing from the containers towards the host rock. This thermal gradient repels the water molecules from the innermost layers in the system towards the surrounding host rock (NWMO 2005; Kjartanson et al. 2003). The heat dissipation rate of the designed barriers is dependent on their thermal properties and the moisture change related to thermal gradients. Having an elevated thermal conductivity over a wide range of moisture contents is considered in the design of the engineered barriers. Also, as mentioned, saline groundwater penetration into the barriers changes the pore fluid chemistry in the soil-water system. This change could influence the  24thermal conduction capability of the bentonite materials. Previous studies have investigated the impact of salt concentrations on the thermal conductivity of soil materials (Noborio and McInnes 1993, Abu-Hamdeh and Reeder 2000, Casas et al. 2013). The pore water chemistry influence on the thermal conductivity of the designed barriers in the Canadian DGR concept should be investigated and reported. Also, as mentioned above, the barriers would dry out due to the elevated temperatures; therefore, it is reasonable to evaluate the thermal properties of the salted barriers under a wide range of moisture contents.  1.2 Research Question   At the depth of the repository, the engineered barriers will be subjected to coupled processes of hydration and desiccation. The local groundwater infiltrates into the barrier and saturates the clay-based material from the surrounding host rock towards the used fuel container. Meanwhile, the thermal energy dissipates from the container towards the in-situ rock and dries the material out (NWMO 2005, Kjartanson et al. 2003, and Steefel et al. 2010). These coupled processes are schematically shown in Figure 1.2.  Infiltration of the saline groundwater in to the barrier increases the salt concentration of the pore fluid. When the existing thermal gradient dries the material out, salt remains in the material while experiencing a wide range of degrees of saturation.  Both unsaturated conditions and increased salt concentrations in the pore fluid affect the mechanical behavior and thermal conductivity of the clay-based materials. The impact of the pore fluid salinity on the hydro-mechanical behavior of the clay-based materials have been studied in previous works (Siddiqua et al. 2011b, Siddiqua et al. 2014). Also, the behavior of  25the materials has been investigated in unsaturated conditions within a range of different saturation degrees, in which the specimens were prepared with distilled water (Blatz et al. 2002, Blatz et al. 2007). The situation of the increased salt content within different degrees of saturations is a condition which occurs for the clay-based barrier between the distance of surrounding host rock and the used fuel container. This situation has not been studied previously on the materials designed to be used in the repository facility.    Figure 1.2 Conceptual schematic of a deep geological repository showing anticipated evolution of thermal energy, degree of saturation and pore water saline concentration distributions with time (from Tabiatnejad et al. 2016)  In this research program two of the materials designed for the multi-barrier sealing system, one at high density and the other at low density were selected to study the impact of  26pore fluid salinity on their mechanical behavior and thermal conduction ability.  For this purpose, a series of triaxial experiments are performed to study the stress-strain behavior of clay-based materials in different salt concentrations and degrees of saturations. The impact of the pore fluid salinity on the compressibility and shear strength of the materials are discussed based on the obtained results. In the second phase of the experimental program, thermal conductivity experiments are reported. Based on the experimental data the impact of salt concentrations on the thermal conductivity values is discussed.  1.3 Research Objectives  The goal of this research is to investigate the impact of the salinity of the constituent pore fluid on the behavior of the clay-based materials compacted at high and low dry densities in unsaturated conditions. This study focuses on the effect of increased salt contents in the material on the shrinkage, compressibility and shear strength of the clay-based barriers in two dry densities. Moreover, it studies the thermal conductivity of the clay-based material in the saline pore fluid situations and compares them with the thermal conductivity of the material in distilled water pore fluid conditions. The findings from this study elaborate the changes in the behavior of the clay-based materials around the used fuel containers when the saline groundwater infiltrates and the barriers dry out at the same time. The results and calculated parameters will be used in future behavior modeling of the barriers. Moreover, to be able to predict the thermal conductivity values for the saline pore fluid situations several thermal conductivity prediction models are selected and evaluated. Specific goals of this research can be listed as:  27  Objective 1: Study and quantify the impact of pore fluid salinity on the shrinkage, compressibility and shear strength of the BSB material in unsaturated conditions. Also, obtain the water retention curve of the material in the presence of salt in the constituent pore fluid.  Objective 2: Study and quantify the effect of pore fluid salinity on compressibility and shear strength of the LBF material in unsaturated conditions and obtain the soil water characteristic curves.    Objective 3: Investigate the impact of salt concentrations on the thermal conductivity of the mixtures of bentonite and silica sand in two low and high dry densities.  ‐ Objective 4: Evaluate the existing thermal conductivity prediction models for calculation of thermal conductivity values for mixtures of bentonite clay and silica sand in both pure water and saline solution situations.  1.4  Thesis Structure and Organization  This thesis consists of five chapters to achieve the mentioned objectives. Chapter 1 presents an introduction to different elements of this study and elaborates the research question and objectives and the organization of the thesis.  Chapter 2 presents the experiments and the obtained results on the BSB material. The chapter begins with a description of the material and the experimental methodology and continues with the results from the soil water characteristic curves, isotropic compression, and shear strength. In the end, the impact of the salinity on the hydraulic and mechanical behavior of the material is discussed. Chapter 3 starts with  28describing the light backfill material along with the specimen preparation procedure and methods for performing the triaxial tests. The obtained results on the stress-strain behavior and total suction measurements are presented with their interpretation. The focus of chapter 4 is on the thermal conductivity subject. The procedures for both experimental part and prediction models results are presented, measured thermal conductivity values for the mixtures with different pore fluid situations are illustrated in figures. The calculated thermal conductivity values by the prediction models are plotted versus the obtained experimental values, and their performance in prediction is evaluated. Chapter 5 presents the summary of the research work, the conclusions, and recommends future research studies.      29Chapter 2: Shrinkage and Mechanical Behavior of Bentonite-sand Buffer1  2.1 Overview  This chapter first presents a description of the bentonite-sand buffer (BSB), and the procedures followed to prepare the specimens. Experimental methodology is presented in detail for shrinkage behavior tests and the triaxial experiments. Obtained results on the shrinkage behavior and stress-strain behavior of the specimens are presented along with the soil water characteristic curves. In continue, the results are discussed, and the conclusions based on them are presented.  2.2 Materials and Sample Preparation   Bentonite-sand buffer (BSB) specimens were prepared by mixing equal dry mass of Wyoming bentonite clay and well-graded silica sand with the CaCl2 salt solution. The Wyoming bentonite clay is purchased from the Bentonite Performance Minerals LLC, and its commercial name is NATIONAL® Standard mesh 200. The material properties of the used clay are summarized in Table 2.1                                                  1 A version of chapter three has been published in the journal of Environmental Earth Sciences.  has been accepted for publication in the Journal of Environmental Earth Sciences Tabiatnejad, B., Siddiqua, S. and Siemens, G. (2016), "Impact of pore fluid salinity on the mechanical behavior of unsaturated bentonite-sand mixture", Environmental Earth Sciences, vol. 75, no. 22, pp. 1434.  30Table 2.1 Summary of Wyoming bentonite material properties (Blatz et al. 2007) Property Values Liquid limit, wL (%) 625 Plastic limit, wP (%) 45 Plasticity index, Ip (%) 580 Montmorillonite content (%) 90 Predominant cations Sodium (Na+)  The sand component of BSB is crushed, medium, sub-angular silica sand. Frac sand in six different size ranges was purchased from SIL industrial minerals in Alberta Canada. Based on the particle size distribution of each size range, a proportioning recipe was prepared to mix them to meet the specifications documented in Dixon et al. (1994). Table 2.2  shows the recipe used for mixing the different size ranges of the Frac sand to prepare the sand component of the material. Also, pictures of the different sizes ranges are shown in Figure 2.1. Sand and clay were oven-dried for 48 hours.   Table 2.2 Proportioning specification used for Frac sand component Frac Sand Sizes Percentage (%) 8 x 12 4.40 12 x 20 9.90 16 x 30 10.80 20 x 40 23.00 40 x 70 18.80 50 x 100 19.20 100 x 140 13.90   31          a) 8 x 12        b) 12 x 20            c) 16 x 30              d) 20 x 40           e) 40 x 70    f) 50 x 140 Figure 2.1 Frac sand size ranges provided by SIL industrial minerals  Bentonite and sand mixtures were prepared based on the timed procedure provided by Dixon et al. (1994). The entire process takes 15 minutes and includes six steps of light mixing, grinding, tamping, scraping, a second grinding step and placing in the sealed bags at the end. The soil mixture was sealed in two plastic bags and was kept in a cold place for 48 hours for  32the completion of clay hydration and homogeneous distribution of moisture throughout the mixture. Static compaction device in the geotechnical lab at the Royal Military College (RMC) was used to compact the specimens. The device (shown in Figure 2.2) consisted of a bottle jack with the load capacity of 8000 kg and a load frame, purchased from Humboldt MFG. CO. The triaxial specimens were compacted in five 20 mm lifts to a target diameter and height of 50 and 100 mm, respectively, to achieve a target density of 1.67 Mg/m3 and a degree of saturation of 85%. A compacted BSB specimen prepared with distilled water is illustrated in Figure 2.3.   Figure 2.2 Static compaction unit in the geotechnical lab at the Royal Military College   33  Figure 2.3 Compacted Bentonite-sand buffer specimen prepared with distilled water  As presented in Siddiqua et al. (2011 & 2014) ground water at proposed geological location of the Canadian nuclear waste repository contains a high concentration of Na, Ca and Cl ions. Therefore, for this study, CaCl2 salt was selected to prepare synthetic solutions in the laboratory. Since two different maximum concentrations, >100 and >200, were observed in the Canadian Shield and the Ordovician-age sediments, respectively, 100 g/L and 250 g/L concentrations of CaCl2 solution were considered in the experimental program.  The existence of salt in the pore fluid adds a component to the soil system i.e. solid, gas, solvent and solute. Considering this change, volume-mass relationships were corrected (followed by Siddiqua et al. 2011) to accommodate for the presence of salts in the solution.   342.3 Experimental Program  Experiments were designed to accommodate two groups of specimens which can be categorized as (i) as-compacted and (ii) desiccated.  Three as-compacted specimens were prepared with three different pore fluids: distilled water, 100 g/L and 250 g/L CaCl2 solutions. They were tested using high-pressure triaxial apparatus with a capacity of 10 MPa.  Before installing specimens on the triaxial cell, suction measurements were recorded. Following the installation of a sample on the triaxial cell, pressure was applied in increments. The maximum applied pressure was as high as 6 MPa and then unloaded to a pressure of 3 MPa.  The specimens were sheared at a constant strain rate of 0.014 mm/min (20 mm over 24 hr) along the stress path of deviator stress (q)/mean stress (p) ratio of 3:1, keeping the cell pressure constant at 3 MPa. The q and p stresses are defined as q = σ1-σ3 and p = (σ1+2σ3)/3, where σ1 and σ3 are major and minor principal stresses, respectively. The specimens were sheared at a rate of 0.014 mm/s to ensure enough time for suction equilibrium throughout the shearing (Blatz and Graham, 2003). Following the shearing phase, the specimens were removed from the cell for post-test measurements of mass, volume change, and gravimetric water content.    Desiccated specimens were prepared by imposing three different target total suction levels (80, 160 and 300 MPa) to the specimens using the vapor equilibrium technique (Tessier 1984; Delage et al. 1998; Romero 1999; Delage and Cui 2000; Villar 2000; Tang and Cui 2005). Saturated binary salt solutions (LiCl, CaCl2, and MgNO3) were used to create the desired relative humidity environments in three different glass desiccators. The solutions were over-saturated with the salt so that excess undissolved salt remained at the bottom of the  35container. Figure 2.4 illustrates a schematic view of a soil specimen in a desiccator for drying the samples out.    Figure 2.4 Schematic view of the setup for desiccating the BSB specimens  The soil mixtures and specimens were prepared as described in the material section to maintain the same properties as they would if they were placed in the repository environment.  Relative humidity (RH) values for the reference binary salts were obtained from Greenspan (1977). In sealed desiccators at different relative humidities, the specimens lost moisture so that the total suction of specimens increased until an equilibrium level was established. Water molecules were drawn out of the BSB samples and absorbed by the oversaturated salt solution. Figure 2.5a shows the specimens in the desiccator immediately after compaction prior to sealing with the glass lid, and Figure 2.5b shows the desiccated specimen in the desiccator after the 30-day equilibrium period.    36 (a)  (b) Figure 2.5 Constant suction drying in the desiccators (a) immediately after compaction (b) after the suction equilibrium   The desiccators were placed in a Styrofoam box in a temperature controlled room to minimize any possible temperature gradients. Tang et al. (1998) performed similar tests and  37observed equilibrium for these materials after 30 days of exposure to each constant RH environment. The specimens were weighed daily, and if a sample’s weight did not change more than 0.2% in three consecutive days, it was then considered equilibrated. The specimens’ dimensions were precisely measured before and after the desiccation process; volume strain and changes in dry density were calculated based on the measurements. Figure 2.6 shows a desiccated specimen out of the desiccator, ready to be installed in the triaxial cell.    Figure 2.6 Desiccated BSB specimen before installing in the triaxial cell   2.3.1 Suction Measurements   The RH sensors (Rotronics Hygroclip Relative Humidity sensors - model # HC2-S) were used to measure the total suction values after removing the specimens from the desiccators. The  38sensors were calibrated using six calibration points including five different oversaturated binary salt solutions and distilled water. The RH sensor was first connected to a suction tip as shown in Figure 2.7. and then inserted into a drilled hole at the bottom of the specimen.     Figure 2.7 Suction tip, RH sensor and signal amplifier for measuring internal total suction levels in the specimens (Tabiatnejad et al. 2016)  The amount of desorbed water molecules was calculated by measuring the weight difference of the samples before and after the desiccation process. The Soil Water Characteristic Curves (SWCCs) were obtained by plotting the total suction values against the post-desiccation bulk water contents and degree of saturation. SWCC is known as the primary constitutive relationship for interpreting behavior of unsaturated soils. Once the initial RHs in the soil specimens were measured, total suctions were calculated using Kelvin`s equation (Fredlund and Rahardjo, 1993).   ߰ ൌ	െ	 ோ்ெೢቀ భഐೢቁ. lnሺܴܪሻ        (3.1)   where:   39Ψ = Total suction in soil (kPa) R = Universal (molar) gas constant (8.314 J/mol.K) T = absolute temperature (°K) Mw = Molecular weight of water (18.016 kg/kmol) ρw = Unit weight of water in kg/m3   2.3.2 High Pressure triaxial Equipment   Custom-made, high-pressure cells with maximum pressure bearing capacity of 10 MPa were used for the triaxial experiments. Axial and radial displacements were measured and monitored using five Linearly Variable Displacement Transducers (LVDT). The axial LVDT was connected at the top of the top-cap, and four radial LVDTs were mounted with each being 90 degrees apart. All LVDTs were carefully calibrated with calibration blocks prior to installation. The volume strain was measured from the point measurements of axial and radial displacements. Axial load was measured using a submerged compression load cell with a maximum load capacity of 100 KN. An Omegadyne pressure transducer was calibrated and used to measure the cell pressure.  After compacting specimens, two commercial latex membranes and eight Viton O-rings were used to properly seal them from cell fluid when installing them on the triaxial cell. Two porous stones and filter papers were used at the top and the bottom of a specimen to ensure that the friction was the same at both ends. Figure 2.8 shows an installed BSB specimen in the triaxial cell.  40  Figure 2.8 A BSB sample installed in the high-pressure triaxial apparatus   Specimens were compressed by gradually incrementing cell pressures up to a maximum cell pressure of 6 MPa. This pressure level is selected to ensure that the specimens are sheared at a pressure higher than one that they experienced during the compaction process (approximately 1.5 MPa). To confirm that a specimen’s volume and suction are in equilibrium, pressure increments were applied at 24 hours intervals (Blatz et al. 1999). Following the last loading step, the specimens were unloaded to the cell pressure equal to half of the maximum applied pressure. This unloading step is followed to establish an over consolidation ratio (OCR) of 2. The OCR = 2 is selected to compare data from this study with the available shearing data for specimens prepared with distilled water in Blatz et al. (2007). As described in as-compacted samples, a similar approach was followed for desiccated specimens.    412.4 Results and Discussions 2.4.1 Constant Suction Drying  Figure 2.9 and Figure 2.10 shows the SWCC plots in terms of total suction versus bulk water and total suction versus degree of saturation for both the 100 and 250 g/L CaCl2 specimens, respectively.   Figure 2.9 Total suction versus bulk water content for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  42 Figure 2.10 Total suction versus degree of saturation for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2   In both plots, results are compared with data for specimens prepared with distilled water obtained from Blatz et al. (2007). The results show notable differences between these SWCC curves, as the specimens with saline pore fluids experienced higher total suction ranges in comparison to distilled water ones. This behaviour was expected, as higher concentrations of salt resulted in higher osmotic suction and consequently higher total suction.  For example, within the bulk water content range of approximately 5% to 18%, BSB specimens prepared with 100 and 250 g/L CaCl2 showed total suction ranges of 18 to 210 MPa and 46 to 251 MPa, respectively. Whereas the BSB specimens prepared with DW showed total suction ranges between 9 to 125 MPa within the same range of bulk water content.   43In unsaturated soils, total suction in the soil has two components: matric and osmotic suction. Matric suction is related to the capillary phenomenon in the soil structure and osmotic suction is related to the dissolved salts in the soil-water system (Nelson and Miller 1992; Fredlund and Rahardjo 1993). When the salt concentration in soil becomes significant, the osmotic suction comes into account (Yong 1999). Higher concentrations of salt in soil induce higher osmotic suctions, therefore, at the same bulk water content or degree of saturation, the measured total suction for the BSB specimens with 250 g/L CaCl2 was higher than the ones prepared with 100 g/L CaCl2. Also, a higher osmotic suction in the specimens with 250 g/L CaCl2, makes the expulsion of water molecules more difficult, therefore, at a same suction level, specimens with 250 g/L CaCl2, had higher bulk water content and degree of saturation values. This indicates a higher water retention capability of the specimens prepared with 250 g/L CaCl2.  Volumetric strain is plotted against measured total suction for BSB in Figure 2.11. The volumetric strain calculations were based on the dimension measurements with 10-5 m precision, immediately after the specimen received compaction, and their equilibrium was reached in the desiccators.   44  Figure 2.11 Volumetric strain versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  Results depicted that the volumetric strain after the constant suction drying are lower for specimens with higher salt concentrations. The higher volumetric strain for specimens with distilled water can be attributed to higher suction gradients that the specimens were exposed. This is because the total suction gradient i.e. difference between the initial suction of as-compacted specimens and suction generated in the desiccators, is greater for the specimens prepared with distilled water.  Figure 2.12 presents dry density with respect to total suction and Figure 2.13 presents dry density against degree of saturation for BSB. Changes in dry density for the specimens prepared with 250 g/L CaCl2 salt concentration turned out to be the least significant. At the  45same suction (for example 120 MPa), data for DW specimens from Blatz et al. (2007) have the highest dry density, whereas, BSB specimens with 100 g/L and 250 g/L CaCl2 have lower values, respectively.    Figure 2.12 Dry density versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  46  Figure 2.13 Dry density versus degree of saturation for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  2.4.2 Isotropic Compression  The volumetric strain with respect to the logarithm of mean stress (p) is illustrated in Figure 2.14 for the as-compacted and desiccated BSB specimens prepared with DW, 100 and 250 g/L CaCl2 solutions. The plots are grouped in four different initial conditions: as-compacted, target suction levels of 80, 160 and 300 MPa.   47  Figure 2.14 Volumetric strain versus mean stress for BSB specimens with distilled water, 100 g/L and 250 g/ L CaCl2  The specimens with DW show the lowest compressibility and the 250 g/L CaCl2 ones show the highest for both as-compacted and desiccated tests. Results are also presented in terms of effective montmorillonite dry density (EMDD) with respect the logarithm of mean stress (p) for all BSB specimens discussed above. EMDD is calculated by the ratio the mass of montmorillonite and the volume of bentonite within the barrier material (Siddiqua et al. 2011). This figure clearly indicates that all the as-compacted specimens’ show a large change in EMDD with an increase of stress; the 250 g/L CaCl2 specimen demonstrates a higher EMDD at a same stress level as compared to other two. The desiccated specimens show a very little  48change in EMDD with respect to the change of stress; however, both 80 MPa and 160 MPa distilled water samples demonstrate a higher EMDD as compared to CaCl2 specimens.    Figure 2.15 EMDD versus total suction for BSB specimens with distilled water, 100 g/L and 250 g/L CaCl2  Compression moduli were calculated on the compression curves using the first three pressure increment points for loading modulus () and the last three increment points on the loading path for virgin compression modulus (λ). The unload modulus (u) was determined considering the unloading stage of the plot. Loading modulus () with respect to total suction is presented in Figure 2.16. This data does not show any clear relationship. Whereas, Figure 2.17 presents data for loading modulus () against degree of saturation.   49  Figure 2.16 Loading modulus versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  50  Figure 2.17 Loading modulus versus degree of saturation for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  This figure indicates a decreasing trend for loading modulus with an increase of saturation for DW samples and an increasing trend for CaCl2 samples. As presented in Figure 2.18 and Figure 2.19, virgin compression values (λ) are generally higher for the 250 g/L CaCl2 specimens than the 100 g/L CaCl2 one. The values tend to decrease with an increase of total suction (Figure 2.18) and they tend to increase with an increase of the specimens’ saturation (Figure 2.19). BSB specimens prepared with 250 g/L CaCl2 showed a lower unloading modulus (u) as compared to 100 g/L ones (Figure 2.20).   51  Figure 2.18 Virgin compression modulus versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  52  Figure 2.19 Virgin compression modulus versus degree of saturation for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  53  Figure 2.20 Unloading modulus versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  The values also decrease with an increase in total suction. However, distilled water samples show a higher unloading modulus which indicates that the presence of salt made the material stiffer. Both DW and 100 g/L CaCl2 show a decreasing trend for unloading modulus (u) with an increase of degree of saturation, whereas, specimens prepared with 250 g/L CaCl2 demonstrated an increasing trend for the same relationship.  54  Figure 2.21 Unloading modulus versus degree of saturation for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  Preconsolidation pressures were interpreted using the Casagrande construction method using the volumetric strain versus mean stress plots shown in Figure 2.14. Preconsolidation values in Figure 2.22 show that an increase in the salt contents in the specimens decreased the preconsolidation pressures. Preconsolidation pressures for DW, 100 g/L and 250 g/L CaCl2 specimens show an increase with an increase of total suction. It is important to note that the preconsolidation pressures for the specimens prepared with DW are the highest as compared to 100 g/L and 250 g/L CaCl2 ones.  55  Figure 2.22 Preconsolidation stress versus total suction for BSB specimens with distilled water (Blatz et al. 2007), 100 g/L and 250 g/ L CaCl2  2.4.3 Triaxial Shearing  BSB specimens under triaxial cells were sheared along the stress path of 3:1 (q:p) in which the cell pressure was  kept constant and the deviator stress was developed by lowering the load ram and applying an axial force on the specimen. Figure 2.23 shows the stress-strain relationship for both types of BSB specimens prepared with DW, 100 and 250 g/L CaCl2 solutions. Due to some technical limitations, not all the specimens were sheared at same cell pressure. However, all three as-compacted specimens were sheared at a cell pressure of 3 MPa.    56  Figure 2.23 Deviator stress versus axial strain for BSB specimens with distilled water and 250 g/L CaCl2  Results of as-compacted tests, indicate a higher strength for the 250 g/L sample followed by the 100 g/L and then the DW one. Two specimens were prepared using one of each concentration of CaCl2 solution and desiccated to achieve a total suction of 80 MPa. Both specimens were sheared at 2 MPa cell pressure. Results of these two tests demonstrate that the 100g/L CaCl2 sample shows higher strength properties as compared to the 250 g/L CaCl2 one. Finally, the last two samples were prepared at 160 MPa suction using two different concentrations of CaCl2 solution as seen for 80 MPa samples. These two were sheared at, cell pressure = 1.5 MPa for the 100 g/L specimen and cell pressure = 1 MPa for the 250 g/L CaCl2 specimen. Similar to the 80 MPa suction samples, the 100 g/L specimen shows higher strength  57values than the 250 g/L specimen. This figure shows a higher qeot/pc for the 100 g/L CaCl2 desiccated samples and a higher qeot/pc for the 250 g/L CaCl2 as-compacted sample.  To compare the shear strength properties of specimens prepared with different salt concentrations, secant Young’s moduli of E1% and E50 as well as critical stress ratio M were calculated. (Table 2.3 and Table 2.4).  Table 2.3 Shear strength parameters for BSB specimens prepared with 100 g/L CaCl2  Target Suction Cell Pressure (MPa) E1% (MPa) E50 (MPa) M As-compacted 3 91.4 139 0.36 80 2 200 291 1.27 160 1.5 267 230 1.57   Table 2.4 Shear strength parameters for BSB specimens prepared with 250 g/L CaCl2  Target Suction Cell Pressure (MPa) E1% (MPa) E50 (MPa) M As-compacted 3 100.9 162 0.44 80 2 170 155 1.50 160 1.0 206 233 1.69   As it is shown, all measured parameters for the as-compacted 250 g/L CaCl2, are higher than the as-compacted 100 g/L CaCl2. Also, M values for the higher salt concentration of pore fluids, are generally higher. It can be mentioned that the increase in the pore fluid salinity, has resulted in higher shear strengths. Similar behaviour was previously studied by Siddiqua et al. (2014) for saturated low density bentonite-sand material. As documented in Siddiqua et al.  58(2014), presence of salt reduces the diffuse double layer (DDL) thickness (Yong et al. 1992; Mitchell 1993) and results in an increase of strength properties.  2.5 Experimental Limitations  Numerous unfortunate and frustrating minor and major experimental difficulties and limitations were faced. Although dealing with these difficulties and trying to find alternative ways to solve them took a lot of time, the frustration caused the author to discover a new strategy, adding some parts to the original plan. Here just the limiting factors that significantly held the project back is brought to attention. During the experiments on the high-pressure cells at RMC’s geotechnical lab, the confining fluid, silicone oil, was leaking into the specimen because of numerous reasons. This made a huge burden in the research progress. In the beginning, pressurizing the specimens to cell pressures of 6 MPa the oil was leaking in to the specimens and was saturating the sample with silicone oil.  (Figure 2.24a and Figure 2.24b).        (a) (b) Figure 2.24. Leaked oil into (a) an as-compacted specimen (b) a desiccated specimen   59A series of more than 20 leak tests was carried out to figure out and fix the cause of the leaks. In each leak test, one parameter was carefully isolated and tested to find the reasons. Some of the major actions in the leak tests are as follows: ‐ Wrapping the sharp edges of the pedestal cap with duct tape (Figure 2.25a)  ‐ Wrapping the sharp edges with geotextile wrap (Figure 2.25b) ‐ Smoothing any sharp edges on the membrane stretcher (Figure 2.25c) ‐ Adding two hoses to the bottom of the cell to be able to observe the oil leak during the test ‐ Changing the plugs on the top cap (Figure 2.25d)   ‐ Switching the top cap and using one without scratches on the side ‐ Using a high vacuum grease on the side of the top cap and the pedestal ‐ Replacing the O-rings with new ones ‐ Contacting the O-ring supplier and determining if the O-rings might be the problem ‐ Purchasing new O-rings with new dimensions and properties ‐ Increasing the number of O-rings   60     (a)      (b)                  (c)      (d) Figure 2.25. Fixing and improving the sealing of different parts of the setup such as (a) sharp edges of the pedestal cap (b) wrapped sharp edges with geotextile (c) smoothing sharp edges on the membrane stretcher (d) using new plugs on the top cap  In the end, the problem was solved and the experimental tests on the as-compacted specimens were completed. Unfortunately, another leak problem was faced in the triaxial tests on the desiccated specimens. Shrinkage and desiccation in the desiccators was causing a great number of cavities with sharp edges on the side of the BSB specimens. Under the high pressure, the membranes were punctured when drawn into the cavities. To fix the problem, double thick membranes were purchased and used to better seal the specimens at higher pressures. However,  61even with doubling the thickness the problem was not solved. Despite the still existing problem, it was decided to continue the tests.   2.6 Summary   The engineered clay barrier system of Canadian Deep Geological Repository environment will face thermal-hydraulic gradients as well as the high concentration of ground water salinity at the selected host rock location. Under these complex processes, clay barriers closer to the used fuel containers will start to lose moisture due to elevated temperatures of the containers while at the same time local saline groundwater begins to infiltrate into the barriers. Initially, the saline concertation will be higher at the rock-barrier interface than the barrier-container interface. Therefore, both as-compacted and desiccated barriers’ performance need to be examined under the saline ground water condition to capture both initial and long term performance.      Tests were performed on the bentonite-sand buffer (BSB) prepared with 100 g/L and 250 g/L of CaCl2 salt solutions with two different initial states: as-compacted and desiccated. As-compacted specimens were prepared in the laboratory following the designed properties same as that will be used in the DGR vault. The desiccated specimens were prepared using the vapor equilibrium technique to achieve three different target total suctions. Results of these tests are interpreted in terms of total suction, volumetric strain, loading-unloading modulus and stress-strain-strength behaviour, and this data was also compared against distilled water tests performed by Blatz et al. (2007).     62Chapter 3: Mechanical behavior of Light Backfill  3.1 Overview  LBF is one of the materials designed to be used as an engineered barrier within the deep geological repository vaults under the ground. The infiltration of the high saline groundwater into the clay-based barriers can affect the suction state and mechanical behavior of the LBF material. Therefore, it is important to study those in saline pore fluid conditions. This chapter presents a series of total suction measurements in saline and pure water conditions of the material to investigate the changes in the total suction in the presence of salt. Moreover, the soil water characteristic curves for both saline and distilled water conditions is obtained.  In continue, the laboratory results of a series of triaxial experiments on the material prepared with distilled water and saline solution are illustrated, and the engineering parameters are calculated and interpreted.    3.2 Materials and Methods  3.2.1 Light Backfill Material LBF, provided by AECL, is a 50:50 by dry mass mixtures of Na-bentonite and silica sand. The bentonite component in the mixture is batched and re-blended Saskatchewan bentonite with the general properties presented in Table 3.1.   63Table 3.1 Properties of the bentonite component of Light backfill (after Graham et al. 1997) Property Value Montmorillonite Content, % 80 Liquid Limit, % 214 ± 6 Plasticity Index 182 ± 5 Cation Exchange Capacity, meq/100g 88 Specific Surface Area, m2/g 520 – 630  Figure 3.1 illustrates a close view of the material. As it is shown, the material does not have a powder shape and the mixture of sand and clay are formed in small and big clumps. LBF is designed to be placed in the repository with the target dry density of 1.24 ± 0.02 Mg/m3 and water content of 15 ± 0.2 % (Siddiqua et al. 2014). The degree of saturation of the material at the given target moisture content is 34% approximately.     Figure 3.1 Close view of the LBF material as provided by AECL  643.2.2 Solution Preparation The type of distilled water used to prepare the saline solutions in this study is Double de-ionized water with the electrical 18.2 Ω resistance. CaCl2 salt was purchased from Fisher Scientific Co. to prepare 250 g/L concentration solutions. The concentration of the solutions was measured and checked by taking several samples to confirm.   3.2.3 Specimen Preparation  Light backfill material used to prepare the specimens for suction measurement and triaxial testing, was provided by AECL. The material was oven dried for 48 hours at 105°C. Following removal from the oven, the material was sealed to prevent moisture absorption, and allowed to reach room temperature for at least 90 minutes. The required amount of dried LBF based on the calculations in an Excel spreadsheet was placed in a stainless-steel bowl. Both the bowl and material were put on a scale to monitor the amount of added distilled water/solution. The distilled water/solution was added to the soil by spraying. After every two sprays, the mixture was gently mixed until the required amount of solution was added. After this, the mixing was continued for two minutes, then transferred and sealed in two Ziploc bags for 48 hours to reach moisture equilibrium.   3.2.4 Specimen Compaction  The moisture content of the mixtures was measured after the 48-hour moisture equilibrium period and was put into the Excel spreadsheet to calculate the required masses for compaction. Triaxial specimens were compacted in five layers to dimensions of 100 and 50 mm, in height and diameter, respectively. The Instron machine in the structural lab at the University of British  65Columbia was used to compact the specimens in a metal mold statically. Specimens for the suction measurement series were compacted in two layers to a height of 40 mm.  Typical compacted specimens for both test series are shown in Figure 3.2a and Figure 3.2b.     (a)                                                                                       (b) Figure 3.2 Typical compacted LBF specimen for (a) Triaxial experiments (b) suction measurements  3.2.5 Triaxial Equipment  The triaxial experiments were performed in a single walled GDS triaxial cell with a maximum pressure tolerance of 1.7 MPa. The setup included a cell pressure controller mounted vertically on the wall. Axial deformation was controlled by an LVDT attached to the loading ram outside of the cell. Radial deformation was measured using the Hall effect on specimen transducers mounted on the sample. The triaxial cell was equipped with a 50-mm diameter stainless steel  66pedestal at the bottom with two hoses running through it for drainage. Figure 3.3 shows an installed LBF specimen in the triaxial cell with the on-specimen Hall Effect transducers mounted on.                                                    (a)                                                                            (b) Figure 3.3 Installed specimen in the GDS Triaxial cell with on specimen Hall Effect transducer (a) front view (b) side view  3.3 Experimental Program To obtain the soil water characteristic curve, a total number of 14 specimens were prepared, and seven samples were prepared with each type of pore fluid. An RH sensor attached to the suction tip was used to measure the total suction values. A hole was hand drilled at the bottom  67of the specimen for inserting the suction tip.  The specimen was completely wrapped with shrinkage wrap to seal the specimen and prevent any moisture loss. The amplifier was attached to a Hygropalm – HP22-A, a handheld data logger, which was used to acquire data to show the measured RH and temperature values on its screen. The set up for measuring the total suction is shown in Figure 3.4.    Figure 3.4 Suction measurement setup for LBF specimens  A triaxial test matrix was designed to cover a range of degree of saturation starting from the original compaction parameters of the material in the repository vault. Since the material was not holding together after being extracted from the mold in degrees of saturation less than 34%, it was decided to prepare the specimens with degrees of saturation higher than the mentioned value. Also, the specimens with degrees of saturation greater than 70% were so  68cohesive that it was difficult to extract them from the mold without distortion.  Therefore, the degree of saturation of the specimens in the test matrix was bounded by the upper limit of 70%.   3.3.1 Results and Discussions 3.3.1.1 Soil Water Characteristic Curve  The soil water characteristic curves for LBF specimens prepared with distilled water, and CaCl2 250 g/L solution were generated by plotting measured suction values versus degrees of saturation. Figure 3.5 and Figure 3.6 show the obtained curves for the specimens prepared with both pore fluid types. In lower degrees of saturation, both specimen series showed higher total suction values.    Figure 3.5 Soil water characteristic curves for LBF specimens prepared with distilled water and 250 g/L solution versus degree of saturation  69 The total suction values decrease with the increase in the degree of saturation, and the obtained curves both followed a linear decreasing trend. Moreover, the total suction values were distinctly higher for the specimens prepared with saline pore fluid. This is related to the osmotic suction generated in the soil structure due to the presence of salt in the constituent pore fluid. Since the total suction was changed by altering the initial moisture content prior to compaction and no drying had been carried out, the obtained SWCCs are a wetting path.   Figure 3.6 Soil water characteristic curves for LBF specimens prepared with distilled water and 250 g/L solution versus bulk water content  Since the moisture content of the engineered barriers varies due to change in the environmental conditions, the suction value of the materials varies. Also, suction in unsaturated soils is an important parameter governing compressibility and shear strength of the material.  70Because it would be difficult to measure the suction in the barriers directly, it would be helpful to develop a simple mathematical equation based on a fitted trend line through the experimental data to be able to calculate the total suction from the water content and degree of saturation. The resulting equation is a power law function relating the degree of saturation and water content to total suction in the material. For the material with distilled water as pore fluid, the equations are as below:  ܵ ൌ 42.23	expሺെ0.042	ܵݎሻ        (3.1)  ܵ ൌ 41.28	expሺെ0.095	ݓሻ        (3.2)  Moreover, for the specimens prepared with saline solution:  ܵ ൌ 68.70	expሺെ0.008	ܵݎሻ         (3.3)  ܵ ൌ 68.77	expሺെ0.018	ݓሻ	        (3.4)  Where S is the total suction in MPa, Sr and w are the degree of saturation and bulk water content, respectively. These equations are valid over the total suction ranges from 3 to 10 MPa and 44 to 52 MPa, for specimens prepared with distilled water and CaCl2 250 g/L solution, respectively. Using the presented equations, total suction values within the mentioned  71ranges can be determined by knowing the bulk water content or degree of saturation of the material.  3.3.1.2 Isotropic Compression LBF specimens were statically compacted into the dimensions of 50 and 100 mm, in diameter and height, respectively. Following the compaction and measurement of the dimensions, specimens were installed in the GDS cell. The isotropic loading increments were selected to be able to calculate the initial loading modulus, compression modulus, and unloading modulus.  As mentioned before, the Light backfill specimens were prepared at different target degrees of saturations by changing the specimen moisture contents with two different pore fluids of distilled water and 250 g/L CaCl2 solution. All the specimens were compressed to a maximum cell pressure of 300 kPa in increments of 30, 60, 90, 120, 150 and 300 kPa. These cell pressures were selected to ensure that the pre-consolidation pressure of material is passed and a clear normal consolidation line is distinguishable. The specimens were kept at each cell pressure for 24 hours to ensure both suction and volume strain equilibrium throughout the sample. Following the equilibrium of the specimen at the final loading pressure, the specimens were unloaded to half of the maximum applied pressure to calculate the unload modulus of the material. At this stress state, all the specimens would have the over-consolidation ratio (OCR) equal to 2. Drainage was permitted during the isotropic compression stage.  Figure 3.7 shows volume strain versus the logarithm of mean stress (p) for LBF specimens prepared at different degrees of saturation with distilled water and 250 g/L CaCl2 solution. All the specimens are pressurized in increments up to 300 kPa and unloaded to 150  72kPa. As is illustrated, the material undergoes larger volume strains as the degree of saturation improves. Specimens showed an initial stiffness in both pore fluid situations, which was followed by yielding and forming the normal consolidation line (NCL).   Figure 3.7 Isotropic compression curves for LBF specimens with distilled water and 250 g/L CaCl2 solution   73The compression moduli for initial loading and normally consolidated regions were measured from the first and last three points along the compression curves. The values for the initial loading modulus (κi) are shown in Figure 3.8.    Figure 3.8 Loading modulus (κi) for LBF with distilled water and 250 g/L CaCl2 solution at varying degrees of saturation  The calculated initial loading modulus values show an increasing trend for the specimens with both distilled and saline pore fluid, while the values for salty specimens are generally higher. Figure 3.9 shows the calculated virgin compression modulus (λ) values. They are also generally higher than for specimens prepared with saline pore solution. This means that the presence of salt in the constituent pore fluid has increased the compressibility of the material.   74 Figure 3.9 Compression Loading modulus (λ) for LBF with distilled water and 250 g/L CaCl2 solution at varying degrees of saturation  Figure 3.10 presents the values for the unload modulus (κu). Values for the specimens prepared with distilled water show a decreasing trend with an increase in the degree of saturation. Additionally, the unload modulus increases as the degree of saturation increases for the specimens prepared with saline solutions.    75 Figure 3.10 Unload modulus (κu) for LBF with distilled water and 250 g/L CaCl2 solution at varying degrees of saturation  The pre-consolidation pressure for the test series was defined using the Casagarande construction method on the compression curves, and are plotted versus the degree of saturation in Figure 3.11. The values for specimens with 250 g/L CaCl2 are clearly lower than for the specimens with distilled water. Both trends show an increase in the pre-consolidation pressure with the increase in the degree of saturation.   76 Figure 3.11 Pre-consolidation pressure values for LBF with distilled water and 250 g/L CaCl2 solution at varying degrees of saturation  3.3.1.3 Triaxial Shear Strength  The triaxial shearing was undertaken by raising the load frame at a constant rate of 0.0139 mm/min for 24 hours, which results in an axial deformation of approximately 20% in the specimen. The stress path q:p = 3:1 was followed during the shearing stage by keeping the cell pressure constant. All the specimens were sheared at the same cell pressure of 150 kPa. The stress-strain curves for the specimens prepared with distilled water and 250 g/L CaCl2 solution are illustrated in Figure 3.12 and Figure 3.13, respectively.   77 Figure 3.12 Stress-strain curves for LBF specimens with distilled water at different degrees of saturation  Figure 3.13 Stress-strain curves for LBF specimens with 250 g/L CaCl2 at different degrees of saturation  78The stress-strain for most of the tests showed an initial stiff behavior at the beginning of the shearing, and the strain hardening of the material was observed during the whole stage. Since the deviator stress was developing continuously towards the end of the test, a distinctive peak was not observed for most of the curves. In Just one of the samples prepared at a low degree of saturation of around 34% with distilled water, a peak value is observed followed by a strain-softening behavior in the material.  Just a few tests seem to have reached the critical state as most of the curves were sloping upward through the end of the test. Ductile behavior was detected for all the sheared specimens. Figure 3.14 shows the typical type of ductile behavior that was observed in the shearing of the LBF specimens.      Figure 3.14 Typical failure mode of sheared specimens   79Two secant moduli of E1% and E50 were used to evaluate the stiffness of the samples. Secant modulus E1% was measured from the beginning of the shearing stage to the axial strain of 1%, and the secant modulus E50 was measured from the beginning of shearing to the axial strain where half of the peak deviator stress occurs. The calculated values for the moduli are plotted versus the degree of saturation in Figure 3.15 and Figure 3.16.    Figure 3.15 E1% for the specimens with distilled water and 250 g/L CaCl2 solution   80 Figure 3.16 E50 for the specimens with distilled water and 250 g/L CaCl2 solution   Comparing the values for specimens of both pore fluid types it shows the values for CaCl2 specimens are generally higher than distilled water samples. This indicates that the increase in the salinity of pore water has resulted in an increase in the initial stiffness of the material, although the increase percentage is small. Same trend was observed in the E50 secant modulus values. It can be stated that the increase in the pore fluid salinity has resulted in higher shear strength, and stiffness of the material in unsaturated conditions. Same behaviour for the LBF material in saturated conditions is reported in Siddiqua et al. 2014. The presence of the cations in the constituent pore fluid directly influences the thickness of the diffusive layer in bentonite component of the light backfill and decreases the thickness of the DDL (Barbour and Yang 1993). Reduce in the DDL results in a tighter micro-structure of the material. Also, with the increase in the degree of saturation, a decrease in the stiffness of the material is observed.  81The end of test shear strength values shown in Figure 3.17 confirms the increase in the shear strength of the material with the increase in the salinity of the pore fluid.    Figure 3.17 qeot for the specimens with distilled water and 250 g/L CaCl2 solution   3.4 Summary   Light backfill material is an engineered barrier considered to be used in the Canadian concept of deep geological repository for used nuclear fuel disposal. This material will be compacted at a low dry density of 1.24 Mg/m3. Regarding the infiltration of the groundwater and saturation the barrier with saline pore fluid, the possible impact of the salinity on the water retention and  82mechanical behavior of the material needs to be studied. To fulfill this objective, series of suction measurements and triaxial tests is carried out.  The suction measurements resulted in two soil water characteristic curves for the material in two distilled water and saline pore fluid situations. Results showed that the total suction values in the saline specimens were higher than the pure water samples. Also, it has been demonstrated that increase in the degree of saturation in both specimen series decreased internal total suction values. Moreover, the mathematically derived equations for both situations were presented. Mechanical behavior of the material including the compression and shear behavior was studied in a series of triaxial experiments. From the obtained data from the isotropic compression stage, the compression moduli were calculated and presented in plots versus degree of saturation. It was shown that increase in the salinity of the pore fluid increased the loading modulus and virgin compression modulus, while decreased the unloading modulus.  Following the isotropic compression stage, the specimens underwent shearing stage. Stress-strain curves for both specimen series were obtained and analyzed to calculate the stiffness moduli. The analysis showed that the increase in the salinity of the pore fluid resulted in higher stiffness of the material.    83Chapter 4: Effect of pore fluid salinity on thermal conductivity of bentonite-sand mixtures and evaluation of prediction models2  4.1 Overview   The following chapter presents the study on the impact of pore fluid salinity on the thermal conductivity of mixtures of bentonite clay and silica sand. Distilled water and two saline solutions with two different salt types were used to prepare the cylindrical shaped specimens. The needle probe technique was used to measure the thermal conductivity values. Experimental data obtained were compared to each other to illustrate the impact of the pore fluid chemistry. In continue, some selected thermal conductivity prediction models were used to calculate the values based on the properties of the mixtures. The calculated outcomes were compared to the laboratory results and the most suitable model for each situation is presented.  4.2 Materials  Laboratory experiments were carried out on 50:50 mixtures by dry mass of bentonite clay and sand. The tested clay was sodium rich Wyoming sodium bentonite with a Plastic limit (wp) of 45%, and liquid limit (wL) of 625%. The dominant mineral in the bentonite clay is                                                  2 A version of chapter five has been published in the conference proceedings of 69th Canadian Geotechnical Conference, GeoVancouver 2016, Vancouver, Canada, 2 -5 October, 2016; Tabiatnejad, B., Siddiqua, S., and Siemens, G. “Thermal properties of an engineered barrier in the presence of saline pore fluid”.    84montmorillonite (~77%) as documented in Sarkar and Siddiqua (2016). The sand component is prepared based on a recipe from different size ranges of Frac sand so the particle size distribution fits between the proposed boundaries by Dixon et al. (1994). The proportioning used for sand component preparation is presented in  . The recipe is prepared based on the particle size distribution of each size range.  The specimens were prepared at two different dry density values of 1.67 and 1.24 Mg/m3, with three pore fluids of distilled water, 100 g/L NaCl and 250 g/L CaCl2. Table 4.1 presents the experimental test series in this study.  Table 4.1 Test Series for Thermal Conductivity Measurements Series ID Number of Specimens Dry Density (Mg/m3) Constituent Pore Fluid S1 16 1.67 100 g/L NaCl S2 16 1.67 250 g/L CaCl2 S3 16 1.24 Distilled water S4 16 1.24 100 g/L NaCl S5 16 1.24 250 g/L CaCl2  The specimens were prepared to cover a wide range of degrees of saturation (30% to 85%) for each test series. The mixtures were prepared by combining equal dry mass of oven dried clay and sand with the target pore fluids. The prepared mixtures were sealed in plastic bags for 48 hours to ensure even moisture distribution. Afterwards, the mixture was used to prepare compact specimens. Prior to compaction, the moisture content of the mixtures was carefully measured. The specimens at higher dry density of 1.67 Mg/m3 were statically compacted to a diameter of 50 mm and a height of 100 mm. Following the compaction, specimens were extracted and the measurements were carried out on the specimens. On the  85other hand, the measurements for lower target dry density i.e. 1.24 Mg/m3 material were taken without extracting these specimens from the mold. This method was adopted to avoid any disturbance on measurements of the low-density materials. The compaction mold used for these specimens had the dimeter and height of 38 and 101 mm, respectively. After compaction, the weight and dimensions of specimens were carefully logged to make sure that the target density is achieved.   4.3 Thermal Conductivity Measurement  In this study the KD2 Pro (Decagon Devices Inc.) with a TR-1 sensor were used for the thermal conductivity measurements. The device along with the figure is shown in Figure 4.1.    Figure 4.1 KD2 Pro thermal meter and TR-1 Sensor (from Rozanski and Sobotka 2015)  86This device operates based on the hot wire technique and the thermal conductivity is calculated via observing the heat dissipation from a linear heat source through the material in two cycles of heating and cooling (ASTM D 5334‐14). As the heat flows radially around the probe the equation for heat conduction in the cylindrical coordinates is (Rozanski and Sobotka 2013):  డ்డ௧ ൌ ܦ ቂడమ்డ௥మ ൅ଵ௥డ்డ௥ቃ          (4.1)  where  T     =    temperature at time t (ºK) r        =   distance from the heated needle (m)  When a constant amount of heat Q is supplied to the needle, the temperature change ΔT over time t would be calculated as:  ∆ܶ ൌ െ ொସగఒ ܧ݅ ቀି௥మସ஽௧ቁ 0< t ≤ th       (4.2)  where:  	ΔT    =    change in temperature from time zero (K) Q      =    heat input per unit length of heater (W/m) λ       =    thermal conductivity (W/ (m.K)) Ei      =   exponential integral   87r        =   distance from the heated needle (m) D       =   thermal diffusivity (m2/s) th       =    heating time  Also, the change in the temperature after switching the heat off is calculated by:  ∆ܶ ൌ െ ொସగఒ ቂെܧ݅ ቀି௥మସ஽௧ቁ ൅ ܧ݅ሺି௥మସ஽ሺ௧ି௧೓ሻቃ     t > th    (4.3)  The exponential integral (Ei) is defined as:  ܧ݅ሺݔሻ ൌ 	െ	׬ ௘షೠ௨ஶି௫ ݀ݑ        (4.4)  Although using Eq. (4.2) and (4.3) are the most precise method to compute the thermal conductivity values, but referring to ASTM D5334-14 they cannot be solved for λ and D clearly. Therefore, using mathematical techniques (i.e. series expansion) they are simplified to Eq. (4.5) and (4.6):  ∆ܶ ≅ ொସగఒ lnሺݐሻ  0< t ≤ th      (4.5)  ∆ܶ ≅ ொସగఒ ln ቀ௧௧ି௧೓ቁ  t > th       (4.6)   88The temperature measurements are plotted versus ln(t) and ln(t/(t-th)) for both heating and cooling phases. In the beginning of both phases the effect of the thermal capacity and contact resistance of the probe are overcome and the graph is nonlinear. However, the logarithmic graph becomes linear after a certain time and the thermal conductivity can be computed. The quality of the contact between the probe and the soil material changes the time it takes for the logarithmic graph becomes linear (Low et al. 2014). ASTM D 5334-14 suggests to visually select the linear portion of the graph. An example for the temperature measurements of the probe in intervals for a specimen is shown in Figure 4.2. The linear portions of the temperature measurements plot per ASTM D5334-14, for heating and cooling phases, are illustrated in Error! Reference source not found.(a) and Figure 4.3 (b), respectively.    Figure 4.2 Typical temperature measurements by KD2 Pro Decagon thermal meter for both heating and cooling cycles   89 (a)  (b) Figure 4.3 Linear portion of the temperature measurement plot for (a) heating and (b) cooling phases   90 The TR-1 probe has dimensions of 24 mm in diameter and 100 mm in length, and it has measurement capacity within the range of 0.10 to 4.00 W/(m.°K). The calibration of the probe was checked prior to every scheduled test per day using the TR-1 verification cylinder provided by the Decagon. A testing protocol was developed for thermal conductivity measurement. Following the protocol, a hole was drilled at the center of the specimen through the entire height. The drilled hole was coated with a thin layer of thermal compound (i.e. Arctic Silver 5) to ensure the maximum thermal contact between the probe and the specimen. The setup of the thermal conductivity measurement for the specimens is schematically shown in Figure 4.4..    Figure 4.4 Schematic illustration of the thermal needle probe inserted in the cylindrical specimen  After inserting the probe in the soil, through the drilled hole, it was kept in the specimen for a minimum period of 15 minutes to ensure the temperature equilibrium.Figure 4.5Error! Reference source not found.a and Error! Reference source not found.b show two samples at high and low density, respectively.   91        (a)                  (b) Figure 4.5 Thermal Conductivity measurement of the bentonite-sand specimen prepared at (a) 1.67 Mg/m3 dry density with (a) 250 g/L CaCl2 solution at ~ 40% degree of saturation and (b) completely dry specimen  The measuring time for TR-1 is five minutes and during this time the sample should not be moved. By pressing the button on the pad, the measurement starts.   4.4 Results and Discussions  Figure 4.6 illustrates the results obtained for the high density (1.67 Mg/m3) bentonite-sand material prepared with 100g/L NaCl and 250 g/L CaCl2 solutions. These tests are denoted as S1 and S2 in Table 1. The results are compared with the data from Graham et al. (1997) for the same density bentonite-sand material prepared with distilled water as constituent pore fluid.  92  Figure 4.6 Thermal conductivity values for the specimens with dry density of 1.67 Mg/m3 prepared with distilled water and saline solutions   It is clear that the presence of the salt component in the material have lowered the thermal conductivity ability of the mixture in general. The material with distilled water has the highest thermal conductivity values comparing two other series. Moreover, the specimens with 250 g/L CaCl2 which are the samples with highest amount of salt concentration among three, are showing the lowest values. This trend is almost consistent throughout the covered range of degree of saturation. The thermal conductivity measurements for the samples prepared at dry density of 1.24 Mg/m3 with three pore fluids are illustrated in Figure 4.7. In these three test series (S3, S4 and S5), similar to as observed trend for the specimens with higher dry density,  93the highest and lowest thermal conductivity values are obtained for the specimens with distilled water and 250 g/L CaCl2 solution, respectively.   Figure 4.7 Thermal conductivity values for the specimens with target dry density of 1.24 Mg/m3 prepared with distilled water and saline solutions  Since the presence of salt in the constituent pore fluid lowers its thermal conductivity (Jamieson and Tudhope 1970, Ramires et al. 1994) it was expected to have the lower thermal conductivity for the specimens prepared with saline solutions.   944.5 Calculated Thermal Conductivity Values using Prediction Models  As mentioned before, five different prediction models (Bruggeman 1935, Johansen 1977, Cote and Konrad 2005, Lu et al. 2007, and Chen 2008), which are suitable for soil mixtures have been evaluated. The calculated values from the models are plotted versus the experimental values to analyze the suitability of the model for the soil mixture with the specific pore fluid conditions. For a better quantitative evaluation of the correlation models on calculating the thermal conductivity values two factors were introduced:  Mean value of R (ߤ ൌ ଵே∑ ܴ௜ሻே௜ୀଵ        (5.7)  Root mean square error of R ሺߜ ൌ ට	ଵே∑ ሺܴ௜ െ 1ሻଶே௜ୀଵ 	)    (5.8)  Where R is defined as the ratio of the predicted value to the experimentally measured value of thermal conductivity. Figure 4.8 - Figure 4.12 show the calculated thermal conductivity values using various models for the specimens prepared at the density of 1.67 Mg/m3. For a better observation, three lines for R=1.2, 1, and 0.8 are shown in each plot. These lines are marked with +20%, 1:1 and -20%, respectively. Any plotted data that falls within the two created regions, their experimental values are less than 20% off the predicted values.   95 Figure 4.8 Evaluation of. Johansen (1977) model for the BSB specimens prepared at 1.67 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions   96 Figure 4.9 Evaluation of. Cote and Konrad (2005) model for the BSB specimens prepared at 1.67 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions   97 Figure 4.10 Evaluation of. Lu et al. (2007) model for the BSB specimens prepared at 1.67 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions   98 Figure 4.11 Evaluation of. Chen (2008) model for the BSB specimens prepared at 1.67 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions   99 Figure 4.12 Evaluation of. Bruggeman (1935) model for the BSB specimens prepared at 1.67 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions  Studying these figures, a suitable model would be the one with the plotted points within the boundaries of R=1.2 and R=0.8 line and closer to the R=1 line. The values for μ and δ are illustrated in Table 4.2.     100Table 4.2 Mean value of R (μ) and root mean square error (δ) values for the selected thermal conductivity prediction models for the specimens prepared at target dry density of 1.67 Mg/m3       Graham et al. (1997) 1.67 Mg/m3  CaCl2 250 g/L 1.67 Mg/m3  NaCl 100 g/L    μ  δ  μ  δ  μ  δ Johansen (1977) 0.94 0.21 1.20 0.27 1.18 0.23 Cote and Konrad (2005) 0.90 0.15 1.13 0.20 1.09 0.14 Lu et al. (2007) 0.99 0.15 1.35 0.42 1.24 0.31 Chen (2008) 1.04 0.15 1.34 0.44 1.28 0.38 Bruggeman (1935) 0.92 0.25 1.36 0.53 1.41 0.42     101Overall, all the selected models showed good performance for the specimens prepared with distilled water (data from Graham et al. 1997). Particularly, calculated values using Lu et al. (2007) model for the distilled water specimens showed good agreements with the experimental data (Mean value of R μ = 0.99) and Root mean square error (δ) = 0.19). The values form Chen (2008) model resulted in a good mean R value of μ = 1.04. Bruggeman (1935) and Cote and Konrad (2005) slightly underestimated the thermal conductivity values for the specimens prepared with distilled water (data from Graham et al. 1997). For the specimens prepared with 100 g/L NaCl solution and compacted at the dry density of 1.67 Mg/m3, the predicted thermal conductivity values for Cote and Konrad (2005) were in good agreement with the measured values (μ = 1.09, δ = 0.14). Whereas, Johansen (1975) slightly overestimated the thermal conductivity of the material and Lu et al. (2007), Chen (2008) as well as Bruggeman (1935) calculated the values with higher than 20% overestimation. Cote and Konrad (2005), also, showed the best performance for the specimens prepared with 250 g/L CaCl2 solution (μ = 1.14, δ = 0.20). Johansen (1977) slightly overestimated the values, but Lu et al. (2007), Chen (2008) and Bruggeman (1935) showed higher overestimated thermal conductivity values.   Figure 4.13 – Figure 4.17 show the calculated thermal conductivity values using various models for the specimens prepared at the density of 1.24 Mg/m3 .  102 Figure 4.13 Evaluation of. Johansen (1977) model for the BSB specimens prepared at 1.24 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions   103 Figure 4.14 Evaluation of. Cote and Konrad (2005) model for the BSB specimens prepared at 1.24 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions   104 Figure 4.15 Evaluation of. Lu et al. (2007) model for the BSB specimens prepared at 1.24 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions    105 Figure 4.16 Evaluation of. Chen (2008) model for the BSB specimens prepared at 1.24 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions   106 Figure 4.17 Evaluation of. Bruggeman (1935) model for the BSB specimens prepared at 1.24 Mg/m3 dry density with 100 g/L NaCl and 250 g/L CaCl2 solutions  Table 4.3 presents the mean values of R (μ) and root mean square error of R (δ) for the specimens prepared at the target dry density of 1.24 Mg/m3 with three pore fluids of distilled water, 250 g/L CaCl2, and 100 g/L NaCl, respectively.    107Table 4.3 Mean value of R (μ) and root mean square error (δ) values for the selected thermal conductivity prediction models for the specimens prepared at target dry density of 1.24 Mg/m3     1.24 Mg/m3  Distilled water 1.24 Mg/m3  CaCl2 250 g/L 1.24 Mg/m3  NaCl 100 g/L    μ  δ  μ  δ  μ  δ Johansen (1977) 1.23 0.38 1.12 0.23 1.12 0.28 Cote and Konrad (2005) 1.17 0.31 1.08 0.18 1.08 0.23 Lu et al. (2007) 1.33 0.52 1.18 0.30 1.19 0.38 Chen (2008) 1.32 0.50 1.18 0.29 1.18 0.37 Bruggeman (1935) 1.14 0.34 1.43 0.50 1.31 0.41     108Based on these values, the best performance for prediction of the thermal conductivity values for distilled water specimens are for Bruggeman (1935) and Cote and Konrad (2005), with mean values of μ = 1.14 and 1.17, respectively. Chen (2008) and Lu et al. (2007) noticeably overestimated the values for the distilled water specimen series. For the bentonite-sand mixtures prepared with 250 g/L CaCl2, Cote and Konrad (2005) model showed best performance in calculation of the values and Bruggeman (1935) showed a significant overestimation for the mentioned specimen series. Cote and Konrad (2005) is the most suitable model for predicting the thermal conductivity values for the specimens prepared with 100 g/L NaCl (μ = 1.08, δ = 0.23), and Bruggeman (1935) overestimated the values (μ = 1.31, δ = 0.41). The rest of the models, (Johansen (1977), Lu et al. (2007), and Chen (2008)), showed an acceptable estimation (1 < μ < 1.2). Cote and Konrad (2005) model has been developed by analyzing nearly 220 experimental results available from the literature. Also, they used different soil-type dependent factors to correlate more than 650 test results for different types of moist soils. Regarding the massive number of results that the correlation is developed on and implementing the mentioned factors, it is a very comprehensive thermal conductivity prediction model.    4.6 Summary  This chapter investigates the impact of the pore fluid chemistry on the thermal conductivity of the mixtures of bentonite clay and silica sand. Experiments were performed on two different density material, named as: low and high dry density bentonite-sand. Results showed that increase in the pore fluid salinity results in decrease of the thermal conductivity values. Results  109also showed that the soil dry density is an important parameter in the thermal conduction ability of the material. In continue, five different thermal conductivity prediction models were examined for the material prepared with distilled water, 100 g/L NaCl and 250 g/L CaCl2 solutions. The model with the closest predicted values to the experimental data is presented.      110Chapter 5: Conclusions and Recommendations  5.1 Summary and Conclusions In the Canadian concept of DGR the used fuel bundles will be placed in special containers and isolated form the environment through a multi-barrier sealing system. This system consists of a series of clay-based materials designed to support the containers in place, minimize the groundwater infiltration and dissipate the thermal energy generated by the used fuel bundles. In the repository, the engineered barriers undergo coupled processes of infiltration of saline groundwater and desiccation due to the existing thermal gradient. Therefore, the material experiences different saturation degrees with the increased salt content, which affects the behavior of the material  This research studied the effect of pore fluid salinity on the shrinkage, compressibility and shear strength of the clay-based materials. It was shown increase in the pore fluid salinity decreased the shrinkage, and increased the water storage capability, compressibility and shear strength of the materials.      This study investigated the impact of pore fluid salinity on the thermal conductivity of mixtures of bentonite and sand and concluded that increase in the salt concentration of the constituent pore fluid decreased the thermal conduction ability of the material. Also, evaluated some of the thermal conductivity prediction models and presented the most suitable model to use for the mixtures with saline pore fluid.    1115.2 Originality and Contribution Previous studies on the clay-based materials, studied the impact of pore fluid salinity on their mechanical behavior in saturated conditions. The coupled process of increased salt contents and unsaturated conditions within a wide range of degrees of saturation had not been considered in the previous works. Moreover, all the past researches on the thermal conductivity measurements were focused on the materials with distilled water as the constituent pore fluid and the data for the saline pore fluid conditions which is the realistic situation that occurs in the repository, was not available. This study investigated the mixtures of bentonite and clay prepared with saline solutions within a wide range of saturation degrees. In addition, few studies had evaluated thermal conductivity prediction models for the specimens prepared with distilled water as pore fluid, it was necessary to find a suitable prediction model for the mixtures prepared with saline pore fluids to calculate the thermal conductivity values without performing laboratory experiments. This research evaluated several thermal conductivity prediction models for the bentonite-sand mixtures compacted at high and low dry densities prepared with both distilled water and saline solutions; and presented the most suitable model for calculating the thermal conductivity values for the mixtures prepared with saline pore fluid.   5.3 Applications The results from this study will be used in the modeling of the behavior of the engineered barriers within the multi-barrier sealing system of the Canadian concept of DGR for their life time operation of hundreds and thousands of years. The findings from this research made it clear that in the computer modeling of the behavior of the barriers, the impact of pore fluid chemistry cannot be ignored.   112The presented suitable thermal conductivity prediction models for the mixtures prepared with distilled water and saline solutions can be used to calculate the thermal parameters of the materials in both pore fluid conditions without performing laboratory experiments.  5.4 Limitations and Recommendations The limitations associated with this study are listed in Table 5.1 as well as the recommendations for future studies   113Table 5.1 Limitations and recommendations in this study No. Limitation   Recommendation Impact of pore fluid salinity on mechanical behavior of BSB material 1 The total suction is measured prior to installation of the specimen in the triaxial cell.   Measuring the total suction changes during the different stages of the triaxial experiments would provide insight for the suction changes of the material in the compression and shearing.   2 Pressure limit of the triaxial equipment  Pressurizing the specimens to higher confining pressures in the triaxial isotropic compression stage will help to observe a clear NCL in the compression stage.   3 Triaxial cell equipped with device for control and changing the temperature in the material.  Temperature changes has an impact on the behavior of the clay component in the engineered barriers. Change and control the temperature of the specimens prepared with saline solutions in unsaturated condition would contribute to the knowledge. 4 Other clay-based materials  One of the barriers in the Canadian DGR concept is the highly-compacted bentonite. It is beneficial to perform the same experimental program on this material. 5 Membrane type in the triaxial tests  It is recommended to use silicone membranes for the experiments in high pressure triaxial equipment.  Thermal conductivity prediction model 6 None of the thermal conductivity prediction models take the pore fluid salinity in to account.   It is recommended to develop a thermal conductivity prediction model based on experimental data on different soil types prepared with saline solutions and define a parameter accounting for the presence of salt component in the pore fluid.     114References Abu-Hamdeh, N. H., & Reeder, R. C. (2000). Soil thermal conductivity effects of density, moisture, salt concentration, and organic matter. Soil Science Society of America Journal, 64(4), 1285-1290. Adams, B., & Wulfsohn, D. (1997). Variation of the critical‐state boundaries of an agricultural soil. European Journal of Soil Science, 48(4), 739-748. Anderson, D.E.S. 2003. Evaluation and comparison of mechanical and hydraulic behaviour of two engineered clay sealing materials. M.Sc. Thesis, Department of Civil Engineering, University of Manitoba, (Winnipeg, Manitoba). ASTM, C. (1997). 177-85. Standard test method for steady-state heat flux measurements and thermal transmission properties by means of the guarded hot plate apparatus. 1992 annual book of ASTM standards, 4, 20-31 ASTM D 5334- 14 (2014). Standard test methods for determining of thermal conductivity of soil and soft rock by thermal needle probe procedure, vol. 04.08, ASTM, 100 Barr Harbor Dr., West Conshocken, PA 19428- 2959. Bishop, A. W. (1959). The principle of effective stress. Teknisk Ukeblad I Samarbeide Med Teknikk, Oslo, Norway, 106(39), 859-863. Blatz, J., Tang, G., Graham, J. & Wan, A. (1999), "Psychrometer techniques for measuring suction in the triaxial test", Proceedings of the 52nd Canadian Geotechnical Conference, Regina, Sask, pp. 25. Blatz, J. A., & Graham, J. (2003). Elastic-plastic modelling of unsaturated soil using results from a new triaxial test with controlled suction. Géotechnique, 53(1), 113-122.  115Blatz, J., Anderson, D. E., & Siemens, G. (2007). Evaluation of the transitional inelastic behaviour of unsaturated clay-sand mixtures. Canadian geotechnical journal, 44(4), 436-446.  Brewer, R. (1964). "Fabric and Mineral Analysis of Soils". John Wiley & Sons, Inc., New York. Briaud, J. L. (2013). Geotechnical engineering: unsaturated and saturated soils. John Wiley & Sons. Börgesson, L., A. Fredrikson and L-E. Johannesson. 1994. Heat conductivity of buffer materials. SKB Technical Report TR 94-29. Stockholm, Swedish Nuclear Fuel and Waste Management Co. Bouyoucos, G. J., & Mick, A. H. (1947). Improvements in the plaster of paris absorption block electrical resistance method for measuring soil moisture under field conditions. Soil Science, 63(6), 455-466.  Budhu, M. (2008). Soil mechanics & foundations. John Wiley & Sons, Inc. New York. 656 pages. Casás, L. M., Pozo, M., Gómez, C. P., Pozo, E., Bessieres, L. D., Plantier, F., & Legido, J. L. (2013). Thermal behavior of mixtures of bentonitic clay and saline solutions. Applied Clay Science, 72, 18-25.  Chen, S. X. (2008). Thermal conductivity of sands. Heat and Mass Transfer, 44(10), 1241-1246.  Côté, J., & Konrad, J. M. (2005). A generalized thermal conductivity model for soils and construction materials. Canadian Geotechnical Journal, 42(2), 443-458.   116Crilly, T. 1996. Unload-reload tests on saturated illite specimens at elevated temperature. M.Sc. Thesis, Department of Civil Engineering, University of Manitoba, (Winnipeg, Manitoba). Das, B. M. (1979). Introduction to soil mechanics. Iowa State University Press. Delage, P., Howat, M. D., & Cui, Y. J. (1998). The relationship between suction and swelling properties in a heavily compacted unsaturated clay. Engineering geology, 50(1), 31-48.  DELAGE, P., & CUI, Y. J. (2000). L'eau dans les sols non saturés. Techniques de l'ingénieur. Construction, 1(C301), C301-1.  DeVries, D. A. (1963). Thermal properties of soils. In ‘Physics of plant environment’.(Ed. WR van Wijk) pp. 210–235.  Dixon, D. A. (1994). Sodium bentonites of Canada, the United States and Mexico: sources, reserves and properties. Engineering materials for waste isolation. Engineering Material Division Special Publication. Edited by AM Crawford, HS Radhakrishna, M. Goutama, and KS Lau. Canadian Society for Civil Engineering (CSCE), Toronto, Ont, 37-65.  Farouki, O. T. (1986). Thermal properties of soils.  Ferris, G. W. 2000. An Elastic Plastic Approach: Modeling Deformation of Dense Sand. PhD thesis, University of Manitoba, Winnipeg. Fredlund, D. G., & Rahardjo, H. (1993). Soil mechanics for unsaturated soils. John Wiley & Sons. Fredlund, D. G., & Morgenstern, N. R. (1977). Stress state variables for unsaturated soils. Journal of Geotechnical and Geoenvironmental Engineering, 103(ASCE 12919).   117Fredlund, D. G., Rahardjo, H., & Fredlund, M. D. (2012). Unsaturated soil mechanics in engineering practice. John Wiley & Sons. Gallipoli, D. 2000. Constitutive and Numerical Modeling of Unsaturated Soils. Ph.D. Thesis, Department of Civil Engineering, University of Glasgow. Garcia-Bengochea, I., Altschaeffl, A. G., & Lovell, C. W. (1979). Pore distribution and permeability of silty clays. Journal of the Geotechnical Engineering Division, 105(7), 839-856. Gascoyne, M., Davison, C. C., Ross, J. D., & Pearson, R. (1987). Saline groundwaters and brines in plutons in the Canadian Shield. In Saline Water and Gases in Crystalline Rocks (Vol. 33, pp. 53-68).  Graham, J., Saadat, F., Gray, M. N., Dixon, D. A., & Zhang, Q. Y. (1989). Strength and volume change behaviour of a sand-bentonite mixture. Canadian Geotechnical Journal, 26(2), 292-305. Graham, J., Oswell, J. M., & Gray, M. N. (1992). The effective stress concept in saturated sand-clay buffer. Canadian Geotechnical Journal, 29(6), 1033-1043.  Graham, J., Tang, X., Wiebe, B., Rajapakse, R.K.N.D., and Zhou, Y. 1995. Progress report on second year of contract no. WS 102679: Modelling of Sand-bentonite Buffer for use in the Canadian Nuclear Fuel Waste Management Program. AECL-Research., July 1995.170 pp. Graham, J., Chandler, N.A., Dixon, D.A., Roach, P.J., To. T., and Wan, A.W.l. 1997. The Buffer/Container Experiment: results, synthesis, issues. AECL Whitesheil laboratories, Pinawa, Manitoba. AECL- 11746.  118Greenspan, L. (1977). Humidity fixed points of binary saturated aqueous solutions. Journal of research of the national bureau of standards, 81(1), 89-96.  Grim, R. E. (1968). Clay mineralogy. Hoffman, C., Romero, E. and Alonso, E.E. (2006). Hydro-mechanical small-scale tests on compacted bentonite. Geotechnical Special Publication No. 147: Proceedings of the Fourth International Conference on Unsaturated Soils, Carefree, Arizona, 2-6 April 2006, 1: 973-984. Jamieson, D. T., & Tudhope, J. S. (1970). Physical properties of sea water solutions: thermal conductivity. Desalination, 8(3), 393-401.  JNC (2000). Safety assessment of the geological disposal system. Supporting report 3. Project to establish technical basis for HLW disposal in Japan. JNC Tech. Rep., JNC TN1410 2000-004. Johansen, O., 1977. Thermal conductivity of soils. Ph.D. thesis, Trondheim, Norway. (CRREL Draft Translation 637, 1977). ADA 044002. Kayadelen, C., Tekinsoy, M. A., & Taşkıran, T. (2007). Influence of matric suction on shear strength behavior of a residual clayey soil. Environmental geology, 53(4), 891-901. Kjartanson, B. H., Dixon, D. A., & Stroes‐Gascoyne, S. (2003). Effects of container/buffer gap fill on buffer performance. Ontario Power Generation, Toronto, Ont. Nuclear Waste Management Division Report. Lingnau, B. E., Graham, J., Yarechewski, D., Tanaka, N., & Gray, M. N. (1996). Effects of temperature on strength and compressibility of sand-bentonite buffer. Engineering Geology, 41(1), 103-115.   119Low, J., Loveridge, F. A., & Powrie, W. 2014. Thermal conductivity of soils by the needle probe method, for energy foundation applications. In 32nd International Thermal Conductivity Conference and the 20th International Thermal Expansion Symposium. Lu, N., Likos, W.J. 2004 Unsaturated Soil Mechanics. John Wiley & Sons, New York. Lu, S., Ren, T., and Gong, Y. 2007. “An improved model for predicting soil thermal conductivity from water content at room temperature.” Soil Sci. Soc. Am. J., 71(1), 8–14. Maatouk, A., Leroueil, S. & La Rochelle, P. 1995. Yielding and critical state of a collapsible unsaturated silty soil. Geotechnique, 45(3):465-477. Man, A., Garinger, B., Alfaro, M., Graham, J., & Goh, T. B. (2003, May). Effect of pore fluid chemistry on the triaxial behaviour of a high-plastic clay. In Proc., Symposium on Geotechnical Engineering for Geoenvironmental Applications. Vancouver Geotechnical Society, Vancouver BC. Man A. and Martino J.B. 2009. Thermal, hydraulic and mechanical properties of sealing materials. NWMO TR-2009-20. December 2009, Atomic Energy and Canada Limited Mazurek, M. 2004. Long‐term used nuclear fuel waste management. Geoscientific review of the sedimentary sequence in southern Ontario. Institute of Geological Sciences, University of Bern, Technical Report TR04‐ 01, Bern Switzerland, available from Nuclear Waste Management Organization, Toronto, Canada. Mitchell, J. K. 1993, Fundamentals of Soil Behavior, John Wiley and Sons, Inc., New York. Mitchell, James K. Soga, Kenichi 2005. Fundamentals of Soil Behavior (3rd Edition). John Wiley & Sons  120Mohsenin, N.N. (1980) 'Thermal properties of foods and agricultural materials', Gordon and Breach, Science Publishers, Inc.: New York, p 407. Murray, H. H. (2006). Applied clay mineralogy: occurrences, processing and applications of kaolins, bentonites, palygorskitesepiolite, and common clays (Vol. 2). Elsevier. Nelson, J. D. and Miller, D. J. 1992. Expansive soils—Problems and practice in foundation and pavement engineering, Wiley, New York. NWMO. 2005. Choosing a way forward: The future management of Canada’s used nuclear fuel, final study. Annual Report, Nuclear Waste Management Organization, Toronto, Canada. Pan H, Quing Y, Pei-Yong L. 2010. Direct and indirect measurement of soil suction in the laboratory. Electron J Geotech Eng (EJGE) 15:1–14 Priyanto, D.G., and Baumgartner, P. 2008. Volume-mass relationships of soil with fresh water and saline solutions. Nuclear Waste Management Organization, Toronto, Ont. Appendix D: Technical Report TR-2008-04. Ramires, M. L., Nieto de Castro, C. A., Fareleira, J. M., & Wakeham, W. A. 1994. Thermal conductivity of aqueous sodium chloride solutions. Journal of Chemical and Engineering Data, 39(1), 186-190.  Richards, B. G. (1965). Measurement of free energy of soil moisture by the psychrometric technique, using thermistors (No. pp 39-46).  Romero E (1999) Perturbation of the heat transfer and the friction factor of a rib roughened surface in an annular passage due to localized removal of the ribbing. Nucl Eng Des 188(1):85–96  121Saadat, F. 1989. Constitutive modeling of the behaviour of a sand-bentonite mixture. Ph.D. Thesis, Department of Civil Engineering, University of Manitoba, (Winnipeg, Manitoba). Sarkar, G. and Siddiqua, S (2016). Effect of fluid chemistry on the microstructure of light backfill: An X-ray CT investigation. Engineering Geology, vol. 202, 153–162. Schofield, A., & Wroth, P. (1968). Critical state soil mechanics. Shaw, B. and Baver, L.D. (1939). An electro-thermal method for following moisture changes of the soil insitu. In Proceedings of Soil Science SocietyofAmerica,4:78-83. Siddiqua, S., Blatz, J., and Siemens, G. (2011a). Evaluation of the impact of pore fluid chemistry on the hydro mechanical behaviour of clay-based sealing materials. Canadian Geotechnical Journal, 48(2), 199-213. Siddiqua, S., Blatz, J., & Siemens, G. (2011b). Experimental study on the performance of light and dense backfills. Canadian Geotechnical Journal,48(2), 214-225. Siddiqua, S., Siemens G., Blatz, J., Man, A., and Lim B.F. 2014. Influence of pore fluid chemistry on the mechanical properties of day-based materials. Geotechnical Geology Engineering, 32:1029–1042. Skinner, A., Hignett, C., & Dearden, J. (1997). Resurrecting the gypsum block for soil moisture measurement. Australian Viticulture, October/November.  Spanner, D. C. (1951). The Peltier effect and its use in the measurement of suction pressure. Journal of Experimental Botany, 145-168.  C. Steefel, J. Rutqvist, C.-F. Tsang, H.-H. Liu, E. Sonnenthal, J. Houseworth and J. Birkholzer, Reactive transport and coupled THM processes in engineering barrier systems (EBS), LBNL Report, 2010.  122Tabiatnejad, B., Siddiqua, S. and Siemens, G. (2016), "Impact of pore fluid salinity on the mechanical behavior of unsaturated bentonite-sand mixture", Environmental Earth Sciences, vol. 75, no. 22, pp. 1434.  Tanaka, N. 1995. Thermal elastic plastic behaviour and modelling of saturated clay. Ph.D. Thesis, Department of Civil Engineering, University of Manitoba, (Winnipeg, Manitoba). Tang, G.X. 1999. Suction characteristics and elastic-plastic modeling of unsaturated sand-bentonite mixture. Ph.D. Thesis, Department of Civil Engineering, University of Manitoba, (Winnipeg, Manitoba). Tang A-M, Cui Y-J (2005) Controlling suction by the vapour equilibrium technique at different temperatures and its application in determining the water retention properties of MX80 clay. Can Geotech J 42:287–296 Tang, G. X., Blatz, J. A., & Graham, J. (1998). Volume changes in unsaturated sand–bentonite caused by suction-controlled drying and wetting. InProceedings of the 51st Canadian Geotechnical Conference, Edmonton, Alta (pp. 4-7). Terzaghi, K. V. (1936). The shearing resistance of saturated soils and the angle between the planes of shear. In Proceedings of the 1st international conference on soil mechanics and foundation engineering (Vol. 1, pp. 54-56). Harvard University Press Cambridge, MA.  Tessier D (1984) E´tude expe´rimentale de l’organisation des mate´riaux argileux: hydratation, gonflement et structuration au cours de la dessiccation et de la re´humectation. Ph.D. thesis, Universite´ de Paris VII, Paris, France. Toll, D. G. (1990). A framework for unsaturated soil behaviour. Géotechnique, 40(1), 31-44.   123Villar, M.V. 2000. Caracterizacio´n termo-hidro-meca´nica de una bentonita de Cabo de Gata. Ph.D. thesis, Universidad Complutense de Madrid, Madrid, Spain Villar Galicia, M. V. (2002). Thermo-hydro-mechanical characterisation of a bentonite from Cabo de Gata A study applied to the use of bentonite as sealing material in high level radioactive waste repositories. Publicación técnica.(Empresa Nacional de Residuos Radiactivos), (4), 15-258.  Wan, A. W. L., Gray, M. N., & Graham, J. (1995). On the relations of suction, moisture content, and soil structure in compacted clays. In Proceedings of the international conference on unsaturated soils, UNSAT 95, Paris, France, 6-8 September 1995, Vol. 1.  Wheeler, S. J., & Sivakumar, V. (1995). An elasto-plastic critical state framework for unsaturated soil. Géotechnique, 45(1), 35-53.  Wiebe, B.J. 1996. The effect of confining pressure, temperature, and suction on the strength and stiffness of unsaturated buffer. M.Sc. Thesis, Department of Civil and Geological Engineering, University of Manitoba, (Winnipeg, Manitoba). Yarechewski, D.S. 1993. Constant mean effective stress tests on sand-bentonite specimens at elevated temperature. M.Sc. Thesis, Department of Civil Engineering, University of Manitoba, Winnipeg, MB. Yong, R. N., Mohamed, A. M. O., & Warkentin, B. P. (1992). Principles of contaminant transport in soils. Elsevier Science Publishers. Yong, R. N. (1999). Soil suction and soil-water potentials in swelling clays in engineered clay barriers. Engineering geology, 54(1), 3-13.   124Yu, X., Zhang, N., Pradhan, A., & Puppala, A. J. (2016). Thermal conductivity of sand-kaolin clay mixtures. Environmental Geotechnics.  Zhang, N., Yu, X., Pradhan, A., & Puppala, A. J. (2015). Thermal Conductivity of Quartz Sands by Thermo-Time Domain Reflectometry Probe and Model Prediction. Journal of Materials in Civil Engineering, 27(12), 04015059. Zhao, H. F., Zhang, L. M., & Fredlund, D. G. (2013). Bimodal Shear-Strength Behavior of Unsaturated Coarse-Grained Soils. Journal of geotechnical and geoenvironmental engineering, 139(12), 2070-2081.  Zheng, L., Rutqvist, J., Birkholzer, J. T., & Liu, H. H. (2015). On the impact of temperatures up to 200° C in clay repositories with bentonite engineer barrier systems: A study with coupled thermal, hydrological, chemical, and mechanical modeling. Engineering Geology, 197, 278-295. Zhou, Y. (1998). Thermo-hydro-mechanical models for saturated and unsaturated porous media. Ph.D. Thesis, Department of Civil Engineering, University of Manitoba, (Winnipeg, Manitoba).  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0340683/manifest

Comment

Related Items