Search

The prediction of water quality from waste-rock stockpiles is an important aspect of mine planning and closure. These predictions are complex, and a well-documented mechanistic approach can provide greater confidence in the results. In this study, humidity-cell experiments (1-kg sample) conducted at 5° and 22°C were used to estimate the effluent water quality and release rates of oxidation products from small-scale (12 t) field-based lysimeters. A comparison between estimated total mass loadings from laboratory humidity cells and total mass loadings from field lysimeters at the Diavik diamond mine, Northwest Territories, Canada, was completed for a five-year period. Measured temperature, S content, and surface area values were used as the primary scaling variables in these estimates. The release rate of SO42− from humidity cells with varying S content (ranging from 0.02–0.18 wt % total S) were normalized to the surface area of sulfide (mol m–2 sulfide S–1 sec–1) from each humidity cell. The resulting variability in SO42− release rates for waste-rock types were within one order of magnitude. The release rates of SO42− from the humidity cells were used to estimate the release rates from waste rock in the field with an average of 0.053 wt % total S. The results suggest that the measured annual total mass loading in the field was within the range predicted from the normalized annual total mass loadings using the surface area of sulfide from the humidity cells, with some discrepancies. Variation in annual loading at the field scale, presumably due to variability in precipitation and flushing, resulted in year-to-year discrepancies. Predictions of Fe and Ni release rates did not capture the field lysimeter results well, likely as a result of variations in geochemical conditions between laboratory and field-scale experiments (i.e., pH, mineral solubility) and also physical and geochemical heterogeneity that is not fully captured in the material characterization at each scale. The results indicate that, at this scale, annual total mass loadings can be estimated using small-scale humidity cells with a mechanistic approach if a thorough characterization of the waste rock is completed to better refine the release rates.

A number of buildings in the Canadian Arctic communities have failed or showed signs of failure of pile foundations caused by climate-change-induced excessive settlements. Effects of climate change warrant alternative pile types and new installation methods. Installing a pile into frozen soils by the torque method is rare in current practice of the Canadian Arctic. The continuous-flight pile is a new foundation type that is installed by torque. This study is aimed at investigating the axial load transfer of continuous-flight piles in frozen soils under long-term loads. A primary objective is to examine the effects of soil temperature, water content, and soil salinity on the resistance and the failure mode of continuous-flight piles in frozen soils. A series of axial load tests of model continuous-flight piles with a shaft diameter of 89 mm were carried out in the lab. Results show that primary and secondary creeps were observed during the constant load tests. The adfreeze resistance increased with the applied load, and the plate bearing resistance increased as the normalized pile displacement rate increased. The mobilized adfreeze and plate bearing resistances suggest a potential increased load capacity of the continuous-flight pile compared to conventional smooth piles of similar diameter.

To investigate the effect of the pore matrix (i.e., ice and unfrozen water) on the strength of partially frozen sand, a series of triaxial compression tests with internal pore water pressure measurements were performed. Both dense and loose saline sand samples were prepared by rapidly freezing the samples and then slowly warming the samples to, and subsequently shearing at the temperature of -3 ℃. Test results indicate the temperature only affects the effective cohesion, not the effective friction angle of sand. The pore ice stress was estimated using Ladanyi and Morel’s (1990) postulate on the internal stresses within frozen soil; the stress (hydrostatic) change in pore ice was reflected by an equal change in pore water pressure in the partially frozen loose sand. Based on Ladanyi and Morel’s (1990) concept of internal confinement in frozen sand, a Mohr–Coulomb model that uses effective failure and residual friction angles from unfrozen sand to estimate the strength of partially frozen sand is presented. The proposed model reflects how the pore matrix contributes to the strength of partially frozen sand. For loose sand, the peak strength is enhanced by the internal confinement and cohesion resulting from the pore water (suction) and pore ice, respectively. For dense sand, only pore ice affects the peak strength by incorporating the internal confinement and cohesion.

Assessment of the potential for static or cyclic liquefaction of loose sand requires information on the undrained response of the deposit. Undisturbed samples of loose sand deposits are necessary to properly obtain this undrained response, which is a function of the in-situ state of the deposit and the applied stress path during loading. In-situ ground freezing for sampling which preserves the in-situ void ratio and fabric is described in this paper. Its application at six sites during the Canadian Liquefaction Experiment (CANLEX) are described to illustrate the challenges in the field and to outline how the different site conditions were managed. The techniques to freeze the ground, to obtain and preserve samples and to prepare laboratory test specimen are described along with a recommended procedure to thaw the frozen laboratory test specimen without disturbing the in-situ condition. The frozen core and in-situ field tests at the sites illustrate the extreme hetergeniety of sands placed hydraulically by man or nature. This variability is a challenge for the experimentalist to capture the material behavior and then incorporate this into an economical design.

Temperatures were measured along instrumented geogrids to determine thermal strains and their changes with seasonal temperatures. It was observed that the application of temperature correction to t...

Laboratory testing of rectangular beams using a synthetic rock was used to investigate the onset of dilation and spalling. The beams are axially compressed and subjected to 4-point bending to provide non-uniform compressive stresses. The maximum tangential stress occurs at the top of the beam and rapidly decreases with distance from the top of the beam. This stress distribution was used to simulate the maximum tangential stress distribution found around circular excavations. The results showed using this beam test configuration that the onset of dilation based on beam displacement and visually observed spalling began at approximately the same stress level. Discrete element numerical analyses (particle flow code) were used to evaluate the stress path at various locations in the beams. The analyses revealed that spalling and dilation in the beams occurred well below the peak strength failure envelope determined from conventional laboratory tests. The findings suggest that the onset of dilation in laboratory tests appears to be a good indicator for assessing the stress magnitudes required to initiate spalling.

Laboratory direct shear tests were carried out on a brittle synthetic rock at various normal loads. The observed fracture patterns changes as the normal load increased. A discrete element code was used to simulate the direct shear tests. The simulation results clearly showed that cracks that developed during the test were predominantly tension cracks. These cracks started to grow from the upper edge of the shear box due to the highly non-uniform stress distribution along the shear plane. The fracture development observed in discrete element simulation models showed that inclined fractures developed parallel to the maximum principal stress direction. En echelon type extensile fractures were also generated and assemblies of theses fractures formed shear bands similar to the process observed in conventional triaxial test using brittle rocks. The discrete element simulations showed that the macro-scale shear band in the direct shear tests resulted from progressive failure of tension-induced micro-cracks.

The use of a diffusion sampler made from regenerated cellulose dialysis membrane was investigated in this study to resolve issues from previous studies on the integrity of the sampler when deployed in the field. The dialysis samplers were deployed in monitoring wells at two upstream oil- and gas-contaminated sites. The average ambient temperature in the monitoring wells was 4 6 1� C over the 6 month duration of the test. Burst pressure and tensile strength tests were used to determine the integrity of the samplers at two field sites over time. The test results showed no adverse impacts on the samplers’ integrity after 6 months. Therefore, diffusion samplers from regenerated cellulose dialysis membrane show promising results when used for long-term monitoring associated with natural attenuation assessment under the conditions tested.

Discrete element (particle) modeling is now often used to simulate the behavior of rock. While the method is attractive because it does not require the formulation of complex constitutive models, it requires extensive calibration with measured macro-scale results to determine the particle–contact parameters that will predict the macro-scale response. This calibration is usually carried out using the laboratory uniaxial compression test. The behavior of intact rock in compression is complex, resulting in a nonlinear failure envelope with high frictional resistance and a ratio of tensile to uniaxial compression strength of approximately 0.05–0.1. An extensive series of calibrations were carried out using a particle code to determine if the micro-parameters determined from uniaxial compression tests could be used to predict the intact rock response regardless of stress path, i.e., tension to triaxial compression. It is determined that the particle code cannot predict the failure envelope measured on laboratory samples, unless significant modifications are made. This study shows that adjusting the particle parameters had little effect on the macro-scale properties of compression and tension tests, but using a clumped-particle geometry improves the predictive capabilities of the particle code significantly. For both Lac du Bonnet granite and a weak synthetic rock, the particle code calibrated to uniaxial tests using the clumped-particle geometry predicts both the stress–strain behavior and the complete nonlinear failure envelope.

This study investigated the feasibility of using trickle-freeze separation as an alternative method of reducing the salinity of oil sands process water. Using a specially designed flume housed in a cold room, an experiment was conducted to determine the degree of separation and subsequent concentration of salts during freezing and melting of saline water. During the freeze – thaw cycle, the majority of salts were concentrated into less than one quarter of the original frozen volume. Utilizing results from the laboratory-scale experiments, a trickle-freeze separation system was designed to treat 20 million m3/year of saline process water. The capital investment for construction of the system was Can$127 million or Can$6.36/m3 capacity, slightly higher than for a conventional desalination processes. Annual operating costs of Can$0.13/m3 of waste water are significantly lower than for conventional desalination. The design and cost estimate provided insight into the feasibility of using trickle-freeze separat...

Horizontal wells installed by directional drilling in the highly permeable river bottom sediments can improve the quality of water delivered to water treatments plants by filtration of suspended solids and dilution with groundwater. Predicting the flow rates available from horizontal wells is an important step in evaluating the technology. A series of three-dimensional finite element models were developed to simulate a horizontal well located beneath a river and a parametric analysis of predicted flow as a function of pipe length, pipe diameter, depth below river bottom, and sediment hydraulic conductivity was conducted to determine optimum horizontal well configurations. The results indicated that the frictional head losses in the well screen have a significant impact on predicted withdrawal rates. Analysis of well length and pipe diameter indicates that for each specific well configuration there exists an optimum length, beyond which little increase in flow occurs. The simulations indicated maximum flows occur when horizontal wells are placed as deep as possible until they are within 0.5 to 2.5 m of an impermeable lower boundary.Key words: directional drilling, water supply, horizontal wells, river filtration.

A field pilot scale experiment was conducted to evaluate the efficiency of spray freezing to remove dissolved chemicals from the tailings lake water at the Colomac Mine, NWT. Cores of the spray ice were then taken to the University of Alberta. The distribution of chemicals in the ice cores were determined, then a 1 m long core sample was thawed in a controlled manner to simulate thawing in the field to examine the concentrations of chemicals in the melt water as the columns thawed. For the pilot scale project approximately 30% of the water pumped was frozen, with the remaining water returned to the tailings pond as runoff. Analysis of the water collected from an ice core melted under controlled laboratory conditions showed dissolved chemical removal of 87–99% (depending on the chemical species) after 39% of the spray ice column had melted. Arsenic concentrations were reduced from approximately 19 μg/L to approximately 5 μg/l. Cyanide had 99.2% removal but still remained at a concentration of approximately 350 μg/L. Approximately 60% of the treated water released at the end of the melt contained only 1–17% of the dissolved species. This melt water at the end of thaw would only require minor further treatment, which may significantly reduce overall treatment costs.

In situ ground freezing with liquid nitrogen can be an excellent technique for obtaining high quality undisturbed samples from a wide range of loose sand deposits, with their in situ conditions preserved. This paper describes the techniques used during the CANLEX project to carry out in situ ground freezing and undisturbed sampling of the loose sand deposits encountered at six test sites. Specific details, with respect to the effects of difficult subsoil or ground-water conditions encountered at the sites, are provided. Modifications made to the ground freezing and sampling system to accommodate difficult ground conditions are described. The actual advance of the freezing front and the consumption of liquid nitrogen are compared with theoretical predictions. The dry coring technique utilized to recover undisturbed frozen sand samples is described. The void ratios measured for the frozen core samples are compared with void ratio measurements obtained using geophysical logging in boreholes located within a 5-m radius of the freeze pipe. The costs associated with in situ ground freezing utilizing liquid nitrogen and coring of the frozen sand at six test sites are presented.

The Canadian geotechnical engineering community has completed a major collaborative 5 year research project entitled the Canadian Liquefaction Experiment (CANLEX). The main objective of the project was to study the phenomenon of soil liquefaction, which can occur in saturated sandy soils and is characterized by a large loss of strength or stiffness resulting in substantial deformations. The intent of this paper is to compare, interpret, and summarize the large amount of field and laboratory data obtained for six sites in Western Canada as part of the CANLEX project. The sites are compared in terms of both flow-liquefaction and cyclic-softening considerations. The paper presents a number of conclusions drawn from the project as a whole, in terms of both fundamental and practical significance.Key words: sand, flow liquefaction, cyclic softening, CANLEX.

A major aim of the Canadian Liquefaction Experiment (CANLEX) was to verify analysis procedures for pre- dicting liquefaction phenomena. Towards this purpose, two loading events were carried out: a field event comprising a clay embankment built over a loose sand foundation layer, and a centrifuge test performed on a model of a sand em- bankment structure. Both the field event and the centrifuge model were planned so as to induce a static liquefaction failure and were instrumented to observe their response in terms of displacement and pore pressure. The fundamental mechanical characteristics of the foundation layer were determined from laboratory element tests (triaxial and simple shear). These tests formed the basis for the stress-strain modelling used in the numerical analyses. Two fundamentally different modelling techniques were used. One involved a fully coupled plasticity model, and the other involved a model based on a collapse-surface approach. The model and prototype structures were then analyzed and the predicted results in terms of displacements and pore pressures were compared with the measured values. The results from both approaches were found to be in reasonable agreement with the measurements, provided allowance was made for direc- tion of loading and drainage effects were accounted for. Resume : Un but principal du Canadian Liquefaction Experiment(CANLEX) etait de verifier les procedures d'analyse pour predire les phenomenes de liquefaction. A cette fin, deux cas de chargement ont ete realises: un chargement sur le terrain comprenant la construction d'un remblai d'argile sur une couche de fondation de sable meuble, et un essai au centrifuge realise sur un modele d'une structure de remblai de sable. Tant le remblai sur le terrain que le modele dans le centrifuge ont ete planifies de facon a induire une rupture par liquefaction statique et ont ete instrumentes pour ob- server leurs reponses en termes de deplacement et de pression interstitielle. Les caracteristiques mecaniques

The Canadian geotechnical engineering community has completed a major collaborative 5 year research project entitled the Canadian Liquefaction Experiment (CANLEX). The main objective of the project was to study the phenomenon of soil liquefaction, which can occur in saturated sandy soils and is characterized by a large loss of strength or stiffness resulting in substantial deformations. In many areas of Canada, large structures are constructed on or comprise sandy soils, e.g., some major hydroelectricity earth dams and many tailings impoundments in the mining industry. The behaviour of loose sandy soils can be difficult to predict, but can have a significant financial impact on these types of engineering structures. Consequently, the intent of the CANLEX project was to improve the overall understanding of soil liquefaction. This paper provides an overview of the CANLEX project, outlining the project objectives, major achievements, and conclusions. Four companion papers describe different aspects of the project; thus, together, the five papers provide a summary of the CANLEX project.Key words: geotechnical, liquefaction, CANLEX, collaborative project.

The freezing behavior of freely suspended piggery wastewater, pulp mill effluent and oil sands tailings pond water droplets was investigated in a specially designed vertical wind tunnel. The entire freezing process was recorded by video cameras and observed visually. The droplets were frozen under different ambient air temperature conditions while floating in the cold air stream. The video images revealed that ice nucleation in the freely suspended water droplets started at the edge of the bottom of the droplets and propagated over the entire surface enclosing the drop in an ice shell under all temperature conditions. The speed of the droplet surface freezing was a function of the ambient air temperature and the chemical characteristic of the water. Most droplets decreased in terminal velocity after freezing. Fracture of ice shell with concentrate discharge was observed for the frozen pulp mill effluent and oil sands tailings pond water droplets.

Freezing characteristics of pulp mill effluent, piggery wastewater and oil sands tailings pond water droplets was investigated. The ice nucleus concentrations of the wastewater droplets were calculated with the nucleus spectra model of Vali [Vali, G., 1971. Quantitative evaluation of experimental results on the heterogeneous freezing nucleation of supercooled liquids. J. Atmos. Sci. 28 (1971) 402–409]. The “freezing ability” of the wastewaters was compared based on the derived ice nucleus concentrations. Variation in freezing was found to be dependent on pH, impurity concentration and droplet volume.

Fluorescence spectrophotometry has been proposed as a quick screening technique for the measurement of naphthenic acids (NAs). To evaluate the feasibility of this application, the fluorescence emission spectra of NAs extracted from three oil sands process water sources were compared with that of commercial NAs. The NAs resulting from the bitumen extraction process cannot be differentiated because of the similarity of the fluorescence spectra. Separation of the fluorescent species in NAs using high performance liquid chromatography with fluorescence detector proved unsuccessful. The acidic fraction of NAs is fluorescent but the basic fraction of NAs is not fluorescent, implying that aromatic acids in NAs give rise to the fluorescent signals. The concentrations of NAs in oil sands process water were measured by Fourier transform infrared spectroscopy (FTIR), fluorescence spectrophotometry and ultra high performance liquid chromatography-time of flight/mass spectrometry (UPLC-TOF/MS). Commercial Merichem and Kodak NAs are the best standards to use when measuring NAs concentration with FTIR and fluorescence spectrophotometry. In addition, the NAs concentrations measured by fluorescence spectrophotometry are about 30 times higher than those measured by FTIR and UPLC-TOF/MS. The findings in this study underscore the limitation of fluorescence spectrophotometry in the measurement of NAs.

Shear wave velocity (Vs) measurements were carried out in a triaxial testing program on three different cohesionless soils. The Vswas measured using bender elements during consolidation and at ultimate steady state. After consolidation the soil samples were loaded in shear under constant strain rate triaxial compression either drained or undrained to determine their ultimate steady or critical state (USS) at large strains. The Vsmeasurements were used to develop relationships between the void ratio (e), mean normal effective stress (p′), and Vs. The shear loading results were expressed within the framework of critical state soil mechanics. The results of the Vsand USS information were combined with the state parameter concept to develop an equation to use field measured Vsto estimate the in situ consolidation state within a soil. Thus, the contractive–dilative boundary with respect to vertical effective stress for large strain loading can be determined from in situ measurements of Vs. These can then be used as a design aid to determine if a soil deposit is potentially susceptible to flow liquefaction. Worked examples to illustrate the procedure are given. Key words : shear wave velocity, cohesionless soil, in situ state, state parameter, liquefaction, laboratory testing.

This paper summarizes an extensive laboratory program undertaken to study the influence of soil type, temperature, and salinity on the strength of three different frozen soils under conditions of unconfined constant strain rate tests. Since the effects of temperature and salinity can be unified by studying the variation of unfrozen water content, measurements of unfrozen water at different temperatures were carried out using the time-domain reflectometry (TDR) method. The stress–strain behavior is influenced by the presence of fine particles in the soil, and an increase in temperature and salinity (unfrozen water content) causes a significant loss of strength. For each soil tested, a predictive model of its strength in terms of salinity and temperature (unfrozen water content) is presented. Key words : frozen soil, saline, unfrozen water, strength.

In situ ground freezing to obtain undisturbed samples of a loose sand beneath the Duncan Dam in British Columbia is described. This was the first known use of ground freezing to assist in obtaining samples at depth greater than 10 m in Canada. Once frozen, the sand was cored and brought to the surface using a Cold Regions Research and Engineering Laboratory core barrel, which is used extensively in permafrost regions. The design, installation, and performance of the liquid nitrogen freezing systems are described in detail. The drilling, sampling, and preparation of the frozen core for transport to the testing facility are also outlined. The quality of the core retrieved during the sampling was judged to be excellent, and good agreement was found between the void ratios measured using both the frozen core and high-quality gamma–gamma density logging techniques, which were performed adjacent to ground-freezing and sampling locations. Key words : ground freezing, cohesionless soils, undisturbed sampling.

Loose cohesionless saturated materials have proved responsible for a number of serious or catastrophic flow slides. Liquefaction failures with no obvious triggering mechanism have also been recorded. This phenomenon of sudden liquefaction without a presence of cyclic shear stresses is often referred to as spontaneous or static liquefaction. Results from previously published studies suggest that liquefaction is triggered not by the undrained loading and generation of pore pressures but by the collapse of the metastàble sand structure, which in turn generates the driving pore pressures in a saturated material. Hence, the collapse is a characteristic response of a material to certain stress states rather than a result of some enforced undrained loading. This theory is evaluated on very loose dry Ottawa sand. It is shown that the very loose dry sand when subjected to a constant deviatoric stress path significantly changes its behavior at a certain discreet stress state, increases compressibility, and becomes increasingly unstable. This results in collapse – vigorous contraction of the specimen. This structural collapse appears to be equivalent to the pore-pressure generation in collapsing, very loose saturated sand. Key words : dry sand, collapse, liquefaction, stress path, triaxial cell.

Au cours de l'hiver 1990-1991, un programme de recherche en nature a ete entrepris pour etudier les proprietes in situ de la glace giclee compte tenu que les informations sur les proprietes mecaniques et physiques des champs de glace intacte giclee en nature sont peu abondantes. Le principal objectif de cette recherche a ete de prelever en nature des echantillons intacts pour les caracteriser en laboratoire, et de faire des mesures sur le terrain pour aider a l'etude des processus de formation et des proprietes physiques apres formation. A cette fin, une planche de glace giclee a ete construite dans les eaux douces du Port de Tuktoyaktuk, Territoires du Nord-Ouest, Canada. La planche d'une epaisseur de 4,3 m etait partiellement echouee dans des profondeurs d'eau de 3-6 m, et avait une hauteur de revanche maximale de 2,2 m. La planche de glace giclee a ete construite au moyen d'un systeme mobile de giclage reposant sur le couvert de glace de la premiere annee. Le rendement en glace du systeme a ete mesure et l'on a trouve qu'il produisait une teneur en glace superieure a 50 %. Un total de 11,8 m de carottes intactes de glace giclee ont ete prelevees en trois endroits sur la planche et transportees vers une chambre froide pour des mesures de densite, de vitesse de l'onde cisaillement, et du comportement mecanique. De plus, l'on a realise des mesures de l'onde de surface dans la glace giclee in situ et dans le couvert de glace de la premiere annee. Cet article decrit l'etude en nature et les methodologies utilisees pour realiser cette recherche

The findings of a laboratory study on the time-dependent displacement of model piles in saline frozen soil are reported. The short-term time-dependent pile deformation in ice-poor saline silty sand was best described using a simple power law of time, whereas the long-term time-dependent deformations were best described using a flow law formulation similar to that used to describe the long-term time-dependent deformation of ice or ice-rich permafrost. The use of cementitious grout as a backfill resulted in doubling of the pile load carrying capacity for a given displacement rate. The laboratory test results compare well with other laboratory and field studies. Key words : frozen soil, saline, model pile, time-dependent deformation, sand backfill, grout.

A 12 m high geogrid reinforced cohesive soil test embankment with 45 degree side slopes has been built near Devon, Alberta, Canada. The embankment has four test sections: three are reinforced with different geogrids, and one section is unreinforced. The test sections were instrumented with strain gauges and inductance coils on the geogrids) and extensometers, inclinometers and piezometers in the soil. The design, construction and instrumentation of the test embankment is described. Field measurements, taken from the slopes over a seven year period during and after construction, are presented and discussed. The development and the distribution of geogrid strains, soil deformations and pore pressures are presented to show the performance of the reinforced cohesive soil slopes.

Loose cohesionless materials can collapse during either static or dynamic loading, resulting in a rapid buildup of pore pressure and associated reduction in shear resistance. As the cohesionless material collapses, it rapidly looses resistance until the acting shear stress decreases to the available residual or steady-state strength. Specially designed stress-path testing has been performed on sand to investigate this collapse process. Results from this test program and previously published data show that a state boundary can be defined when a cohesionless material moves from peak to steady state along a constant void ratio stress path regardless of whether it is loaded drained or undrained. Further, it is demonstrated that the state boundary represents a surface in the effective mean normal stress–deviator stress–void ratio space. Hence, flow slides and liquefaction can be initiated when the stress path followed during either drained or undrained loading attempts to cross this state boundary surface. Key wordy : sand, collapse, liquefaction, stress path, state boundary, triaxial test.

Saline permafrost has a detrimental impact on the performance of foundations. Increased pore-water salinity decreases the strength and increases the deformation susceptibility of permafrost and foundations installed in permafrost. A data base of the distribution of saline permafrost in the Northwest Territories shows high pore-water salinities beneath coastal communities and at locations along the upper Mackenzie Valley and within the Mackenzie Delta. Salinity of inland sites is usually low. High salinities are usually associated with ice-poor materials and are linked with previous marine submergence associated with glacial events. Key words : permafrost, saline distribution, measurements, Quaternary geology.

Laboratory flume deposition tests were carried out to study the physical phenomena associated with the deposition of a sand slurry to form a hydraulic fill. The experimental apparatus was carefully designed to minimize flume-wall effects on the flow and to allow discharge of slurry of constant composition for an indefinite period of time. Slurry concentration and flow rate were varied independently to study their effects on characteristics of the fill such as geometry, grain-size distribution, and density. Three different sands were used to evaluate the influence of the mean grain size. Key words : hydraulic fill, flume tests, sand, profile, slope, density.

Several flume deposition tests carried out in different parts of the world to study hydraulic fills are compared and discussed. The results of all test programs are coherent and consistent with field observations of hydraulic fills and natural alluvial deposits, which suggests that, at least qualitatively, flume tests are adequate to simulate the physical phenomena associated with hydraulic deposition. Key words : hydraulic fill, flume tests, sand, grain size, profile, slope, density.

An experimental program that measured the electro-osmotic conductivity variation with temperature in a frozen soil showed that the conductivity remained constant at the value measured at 22°C until about 0.4°C below the freezing point depression. Below this temperature the electro-osmotic conductivity decreases rapidly.

Using triaxial test data, a time-independent incremental plasticity model has been developed for normally-consolidated broken ice. For this material, both the yield criterion and potential function are found to depend on the void ratio. The inclination of the incremental plastic-strain vector is approximately constant in the triaxial (p-q) plane. It appears that the initial yield of the broken ice can be approximated by a Mohr-Coulomb-type yield surface with the cohesion and fractional parameters varying with the void ratio of the material. Extrapolating into three-dimensional stress space, a Drucker-Prager-type yield surface may be adopted. The hardening law is assumed to be isotropic and can be described by a unique relationship in the mean normal stress-deviatoric stress-void ratio (p-q-e) space. A plastic potential function is established using the observation that the ratio of the incremental shear strain to the incremental volumetric strain is approximately constant. The analytical solution of these plasticity equations has been used to reproduce several triaxial test results.

The findings of a laboratory study on the time-dependent displacement of model piles in saline frozen soil are reported. The short-term time-dependent pile deformation in ice-poor saline silty sand was best described using a simple power law of time, whereas the long-term time-dependent deformations were best described using a flow law formulation similar to that used to describe the long-term time-dependent deformation of ice or ice-rich permafrost. The use of cementitious grout as a backfill resulted in doubling of the pile load carrying capacity for a given displacement rate. The laboratory test results compare well with other laboratory and field studies. Key words : frozen soil, saline, model pile, time-dependent deformation, sand backfill, grout.

The flow law of polycrystalline ice is extended to a general state of stress and the second stress invariant accounts for the stress dependence. Experimental evidence supports this generalized flow law. Punch indentation experimental results are analyzed using this generalized flow law and the expansion of a hemispherical cavity at the punch tip. A finite element analysis of the punch experiments supports the use of a cavity expansion zone beneath the punch.Analytical, as well as experimental, evidence of a lack of influence of depth of penetration on the punch resistances is presented. This evidence conflicts with data presented by S. S. Vyalov and co-workers in 1973 and reasons for the conflict are presented. These findings suggest that higher bearing capacity factors are applicable at the ice surface. Key words: ice, creep, flow law, punch penetration, multiaxial stress states.

This study confirms the existence of two flow laws for polycrystalline ice. One describes the behaviour of ice before the inflection point of a constant stress test, and one the behaviour well into the tertiary portion of the strain–time curve. Each flow law may be represented by a power law with an exponent of 3.0.The inflection point in constant stress experiments and the peak in constant strain rate experiments is shown to occur at about 1% strain in experiments conducted in the ductile material behaviour range. A Cottrell–Aytekin relationship has been used to fit the strain–time material behaviour up to the inflection point.The flow law of ice valid below 1% strain is shown to be dependent on the grain size ratio of the ice. The grain size ratio is a measure of the size effect of the sample tested. The temperature influence on the flow law is best described by using the inverse temperature relationship described by Voytkovskiy. Keywords: ice, creep deformation, long-term, failure strain.