PanAm Unsaturated Soils 2017: Swell-Shrink and Tropical Soils

This paper summarizes the design, installation, and data collection from an autonomous field station located close to a new toll highway south of Austin, TX, that is underlain by more than 10 m of expansive clay. Given the remoteness of the site, the system was designed to be fully automated, requiring very low maintenance. The field station combines a novel string pot system for measuring vertical deformations at five depths in a central borehole, together with volumetric water content, and temperature data from multi-parameter reflectometer probes installed at similar depths in a series of surrounding boreholes. In addition, there is a conventional surface weather station, which records air pressure, temperature, rainfall, relative humidity, and solar radiation. The system has been active for more than 1 year, and the data collected over this period shows its robustness.

Desiccation cracks play a crucial role in the stability of slopes by impacting mechanical and hydraulic soil properties in upper layers of the slope. Research involving a bench scale device for studying desiccation crack development is presented in this paper. The bench scale device contained a compacted clay soil bed approximately 0.9 m deep and instrumented with five soil moisture sensors at different depths. In plan, the soil bed was 0.6 m long and 0.1 m wide, and constrained at the ends so that transverse desiccation cracks would form. The sides were constructed of clear acrylic so that crack depths could be monitored. A soil water characteristic curve was determined so that total suction variations in the soil bed could be estimated using moisture contents. After 1.5 months the soil moisture loss became gradual and exhibited a similar steady rate at all depths. Measured crack depths at the end of 5 months compared favorably with those predicted using a simple analytical model.

Low plastic expansive soils are widely distributed and commonly used as engineering material in arid and semi-arid regions. This paper addresses the swell-shrink behavior of a low plastic expansive soil. Cyclic swell-shrink tests were carried out in modified fixed-ring oedometers, in which samples were allowed to swell and either shrank fully in each wetting-drying cycle with constant surcharge pressure applied. Cumulative shrinkage or expansive deformations with increasing wetting-drying cycles were observed until an equilibrium state was reached. The number of wetting-drying cycles required for the low plastic expansive soil to reach equilibrium was much larger than that for highly expansive soils. On the other hand, the final deformations at equilibrium state after several wetting–drying cycles were larger than the deformations obtained from conventional oedometer test. The difference between them represents the additional deformation induced by cyclic wetting-drying, and is denoted as wetting-drying deformation. With increasing surcharge pressure, the wetting-drying deformation increases first and then declined after reaching its maximum at the preconsolidation pressure. The procedure used to derive the wetting-drying deformations in wetting cycles and drying cycles is also presented. It is found that more than 70% of the wetting-drying deformation occurred during the drying process. This phenomenon was attributed to the matric suction which greatly increased during the process of dehydration.

This paper provides an assessment of the influence of fabric on the swelling characteristics of clays. Specifically, the evaluation involves clay specimens compacted wet and dry of optimum (to achieve a comparatively dispersed and flocculated structures, respectively). In addition, characterization involved specimens collected in the field using Shelby tubes driven into an expansive soil deposit. The clay specimens were moisture-conditioned to the same dry state after compaction and before testing using a centrifuge-based inundation approach. The results from moisture-conditioned specimens were compared against those from specimens that were initially prepared dry prior to inundation. The results from specimens prepared to a dispersed structure show primary swelling of the same magnitude as those prepared to a flocculated structure. However, the specimens prepared to a dispersed structure took longer to reach the end of primary swelling. The results from field-collected undisturbed specimens show similar time response as the compacted, moisture-conditioned specimens, but reveal slightly higher swelling than specimens reconstituted to the same initial conditions. Overall, the experimental results show that the soil fabric of highly plastic clays affects primarily the time needed to reach the end of primary swelling but not the magnitude of swelling for laboratory reconstituted specimens.

An experimental investigation to analyze the effect of suction on the Poisson’s ratio of claystone is presented. For this, the vapor transfer technique is used to impose suction on rock specimens. To evaluate the mechanical behavior of the claystone, two conditioned specimens are tested at each suction value through uniaxial compressive tests. Axial and radial strains are analyzed using the moving point regression technique. Axial and radial stiffness increase with suction and exhibit a non-linear behavior during the loading process. Poisson’s ratio was found to decrease with suction and to evolve in a non-linear manner during loading.

Pavements constructed on expansive soils suffer from distress in response to climatic and environmental conditions. Drying and wetting cycles result in volumetric changes in expansive subgrade soils. Suction compression index (γ_h) is an important parameter in characterizing expansive soils and predicting their volume change (shrinkage and swelling) behaviors. The γ_h defines the changes in volumetric strain with respect to the changes in suction. Several methods are available for estimating the γ_h values. This paper presents a comparison study for estimating the suction compression index using the results obtained from the coefficient of linear extensibility (COLE) tests and soil index properties such as liquid limit, plasticity index, and clay content. Shelby tube soil specimens were obtained from different locations across Oklahoma for laboratory testing. In addition, the soil property values were extracted from the USDA-NRCS database for utilizing in the comparison study. All the data is also used to classify the Oklahoma soils according to expansion potential based on the COLE and index properties of the soils. Finally, contour maps of the γ_h parameter for the Oklahoma soils are prepared based on the COLE and soil index properties. The outcome is a set of reliable evaluation of γ_h and prepare contour maps of γ_h for shrinkage and swelling soils in Oklahoma for practicing engineers to use.

This paper presents an experimental investigation on the enhancement of swelling behaviour and soil water characteristic curve (SWCC) of bagasse fibre and lime stabilised expansive soil. Lime stabilisation is commonly used to improve the engineering properties of expansive soil. Bagasse fibre, an industrial waste by-product left after crushing of sugarcane for juice extraction, was used in this study as reinforcing component in combination of lime for expansive soil stabilisation. The expansive soil used in this investigation was collected from Queensland, Australia. In order to investigate the influences of combination of bagasse fibres and lime on the engineering behaviour of unsaturated expansive soil, a variety of stabilised soil samples were prepared by changing proportions of randomly distributed bagasse fibres combined with different lime contents. An array of experimental tests was performed including free swell potential, swelling pressure, and one-dimensional consolidation tests. Soil suction tests were conducted using the contact filter paper technique on natural and stabilised expansive soil samples. The results revealed that lime-bagasse fibre treatment of expansive clay has a significant effect on swelling behaviour and SWCC response of treated soils. Combination of hydrated lime and bagasse fibre resulted in more improvement on swelling behaviour of soil samples when compared to that treated with lime only. The air entry value of stabilised expansive soil increased with an increase in the stabiliser content.

Photogrammetry techniques have been widely used in geotechnical engineering both in the field and laboratory. In this study, photogrammetry is used to determine the shrinkage curves of an expansive soil. The shrinkage curve is a plot of volume of soil with water content as a soil dries. While, it is relatively easy to determine the gravimetric water content during drying, it is difficult to determine the volume of the soil specimen during drying. Using photogrammetry, the volumetric change of a soil specimen can be obtained by constructing a 3D model of the soil specimen using 2D images at various times. The difference in volumes of the soil specimen between two different times gives the volumetric change. Unlike previous studies, low-cost camera, and freeware are used to make the proposed method easily accessible. Moreover, this approach is non-contact and non-destructive and hence the soil specimen remains undisturbed during drying. In this study, a novel setup is developed which allows a handphone camera to move 360° around the soil specimen for capturing images. Considerations in the setup include lighting, number of images, and camera position. Proper lighting is essential to avoid shadow and distortion in the 3D model. A minimum number of images are required to ensure that the 3D model is constructed successfully. The camera can be fixed at one orientation (elevation and tilt). Two different handphone cameras of different specifications are evaluated. It is observed that 5-megapixel and 8-megapixel cameras gave similar results which show that camera having lower specification (5-megapixel) can be used to obtain an accurate volume measurement of the soil specimen. The absolute error in volume determination given by the two handphone cameras is about 0.3%. Generally, the volume of soil specimen with asymmetric deformation can be determined more accurately by photogrammetry compared to the conventional method.

This paper presents a new laboratory test method for determining the soil index of expansive clays. This method is suggested to replace the shrink-swell, loaded shrink, and unloaded shrink tests methods presently being used in Australia standard 1289 in an effort to solve some problematic issues identified in testing highly expansive clays. The research for this test has been carried out by the authors in a particularly high expansive clay area in Melbourne, Australia (liquid limits 70–120%+) derived from the in-situ weathering of neogene age basalts. The main aim of the investigation was to determine the values and variability of the soil indices in a large residential basaltic province and develop a speedier and more accurate method which can be widely used commercially to plot suction/shrink relationships and test the soil indices. A secondary aim was to develop a new model to calculate and the ground characteristic movement (y_s) for site characterization.

The focus of this paper is laboratory soil tests for estimating one-dimensional heave and settlement of compacted fills composed of expansive soils. Three types of tests, Methods A, B, and C, are described for use in estimating wetting-induced heave/settlement of a new fill, forensic evaluation of an existing partially wetted fill, and load-induced settlement after wetting. These tests have been used in California for decades and have been adopted by ASTM as standard tests for one-dimensional swell and collapse of soils (ASTM D4546-08 and -14).

One of the biggest challenges for researchers and engineers is to properly understand unsaturated expansive soil. Moreland clay which is composed mostly of Montmorillonite minerals causes the major structural damages in Northern Louisiana. The objective of this paper is to expand further knowledge and understanding of the soil volume change behavior of Moreland clay. According to the USCS soil classification, Moreland clay is a highly expansive fat clay (CH) with a significant moisture content absorption capacity (LL = 79 and I_P = 51). An in-depth soil testing is done to characterize Moreland clay including the identification of its swell-shrink properties, soil index property measurement, and soil heave predictions using five different methods. The average swelling percent (SP) is calculated as 7.2%. It indicates that Moreland clay has one of the highest SP values once compared with other expansive soil around the world.

During the 1960s and 70s as the awareness of the potential for clay-based soils to change volume with variations in soil moisture, practitioners had difficulty with quantifying these ground movements. Also in Australia new housing construction changed from the more movement tolerant flexible clad dwellings with timber floors, to slab on ground (raft slabs) and brick and/or blockwork (cinderblock) walls, which are very sensitive to all types of ground movement. Initially practitioners used the Atterberg limit suite of tests or one or more of them. Over time, there was one school of thought that contended that in preparing a sample for the Atterberg limit tests, the natural structure of the soil would be destroyed; therefore the outcome would not correlate well with the performance of the natural insitu clay. The proponents of this theory developed the shrink/swell test.

Lime-induced heave is a process whereby calcium from lime chemically reacts with sulfates within the soil, producing expansive minerals which result in expansion of the limed layer causing surface heave. This paper presents a case study of subgrade heave and offers an alternative explanation of the mechanism causing heave other than development of expansive minerals. After final grading, the site was subject to heavy rainfall. Relatively uniformly-spaced ridges developed within the limed layer in less than 4 days. The study included physical inspection of the ridges, elevation surveys, field density and moisture content tests, Atterberg Limits, soluble sulfate tests, and x-ray diffraction. None of the characteristics generally attributed to development of heave-related minerals were observed. Approximately 4 months of dry weather preceded the heavy rainfall event. It is hypothesized that development of the observed ridges is associated with closing of shrinkage cracks in the native subgrade which resulted in lateral stress and shear failure within the stabilized layer.

In this paper, an elastoplastic framework for the volumetric behavior of expansive soils based on effective stresses is proposed. As proposed by other authors, this constitutive model separates the contribution of the microstructure and the macrostructure on the volumetric behavior including their interaction. The hydraulic behavior of the soil is simulated using a porous-solid model able to reproduce wetting-drying cycles in deforming porous materials. This porous-solid model is based on the current pore size distribution (PSD) of the soil. Therefore, the volumetric deformation of the sample can be related to the evolution of the PSD. The result is a fully coupled elastoplastic framework based on effective stresses able to reproduce the volumetric behavior of expansive soils. Some experimental and numerical comparisons show the ability of the model to simulate the behavior of expansive soils under different stress paths. These results also indicate that the precision of the model is highly dependent on the correct determination and modeling of the retention curves.

Prior to this paper, the most recent work on TMI/Hs for Queensland was published by Fox in 2002. The Thornthwaite moisture index (TMI) was taken over a minimum of twenty years and this 2002 work generally complied with this principle. When AS2870 was revised in 2011, with respect to the TMI, they asked that it be calculated over a minimum period of 25 years, and at the same time gave a new relationship between the TMI and the Hs. The net result of these changes in AS2870 was that some of the 2002 data points were inside the 25 years time window and the changes in relationship between TMI/Hs meant that existing maps were no longer relevant for use by practitioners. This paper was commissioned to provide consistency for practitioners, using up-to-date data, as well as utilising advances in technology (e.g., Google Earth) to enable the practitioner to determine the Hs for a particular geographical area of the state of Queensland.

The volume variation of expansive soils due to alterations in water content causes socioeconomic and environmental damages to geotechnical engineering works. The expansiveness of a soil is determined by intrinsic (mineralogical composition, texture, and structure) and extrinsic (groundwater movement, vegetation, and anthropic occupation) characteristics. The addition of lime to a swelling soil improves its geotechnical properties through physical-chemical interactions capable of stabilizing the volume variation. This paper investigates the effects of lime treatment on the properties of Paulista’s swelling soil, comparing the natural soil’s behavior with 1, 3, 5, and 7% dry weight soil-lime mixtures. Physical, chemical, mineralogical, and swelling characterization tests were performed and the natural soil presented high expansiveness. The determination of expansion pressure was influenced by the stress trajectory followed during the test. The addition of lime to the soil caused the aggregation of particles or flocculation, and reduced the plasticity index. The addition of 5% hydrated lime content to Paulista’s soil stabilized both free and pressure swelling tests.

Pavements constructed over expansive subgrade often experience volumetric deformation due to seasonal moisture fluctuation. In addition to causing shrinkage and swelling in expansive subgrade, moisture variation alters resilient modulus, permeability, and shear strength of soil, which ultimately affects pavement performance. Monitoring the real-time variation of moisture underneath pavement can enhance the understanding of moisture patterns, which can be useful in improving the serviceability of pavement. The objective of this paper is to develop moisture variation model based on real-time monitoring data due to precipitation. A two-lane state highway in Ellis County, TX was selected for field monitoring and data acquisition. Moisture sensors were installed at different depths up to 4.5 m. The soil was found to be high plastic clay in the selected site. Based on 2 years of monitoring, the amplitude of moisture variation ranged from 1 to 14% from the existing moisture content due to rainfall. Collected data were analyzed in a non-parametric environment for temporal variation to develop real-time model development. The developed model can be used to determine the moisture fluctuation within similar pavement subgrade due to rainfall events.

The effect of different degree of saturation and compaction on the diffusive properties of soil is studied. Change in the diffusion of nonreactive chloride (Cl⁻) through the barrier soil sample at different degree of saturation is experimentally obtained. Samples prepared at different densities and degrees of saturation are subjected to single reservoir diffusion testing for the determination of diffusion coefficient. Filter-paper method is used for obtaining soil-water characteristics curve (SWCC). Experimental results are used to calculate the values of diffusion coefficient by the method of curve fitting. The results show that for a particular density, the diffusion coefficient of the completely saturated sample is higher than that of unsaturated sample. However, when the samples are not fully saturated, rate of contaminant transport is less for the sample prepared at the higher water content. The results demonstrate the possible mechanism of transport of the contaminants in the soil when initial suction also adds to the mechanism.

Lightly loaded structures underneath expansive soils encounter severe damage due to the swell/shrink nature of expansive soils resulting from moisture variations. Billions of dollars are spent every year to repair the damages caused by these soils in the U.S. and worldwide. Designing structures to accommodate the swelling strains is a major challenge as predicting the swelling potential of these soils accurately is not easy. A wide variety of swell prediction models have been introduced by various researchers to predict the behavior of these often-problematic expansive soils. These models include various properties of soils such as, plasticity characteristics, compaction conditions, consolidation characteristics, moisture content variations, matric suction, and clay mineralogical characteristics. However, these models are generally developed with typical moderate to high plastic soils in mind whose plasticity indices range from 25 to 45. Their applicability to soils that have liquid limits in the order of 200% is not well understood. In this paper, the ability of these models to predict the behavior of excessively high plastic soils with plasticity indices ranging from 45 to 85 were evaluated. For this purpose, four existing analytical prediction models that use combinations of above-mentioned properties were selected and used to predict the one-dimensional and three-dimensional swelling strains on three high swelling soils. These predictions were verified by conducting one-dimensional and three-dimensional swell tests on the three soil types. The swell tests were conducted at three different initial moisture contents to observe how well the models could predict different levels of moisture absorption. The ability of each of the four selected methods in predicting both 1D and 3D swell strains was discussed and their relative merits and demerits are highlighted. In addition, finite element modeling was performed to simulate one-dimensional and three-dimensional swell tests by using material models that use volumetric and suction changes with moisture contents to simulate expansive soil behavior within the finite element model. The results indicated that while the analytical prediction models gave reasonable results the finite element analysis predicted results were closest to the laboratory measure soils in case both 1D and 3D analyses. Among other analytical models the ones that incorporated mineralogical and suction data exhibited better results.

Liquid ionic soil stabilizers, such as diluted acids, are often used to treat expansive foundation soils for residential projects through pressurized deep injection in Texas. In practice, ionic liquid stabilizers generally work effectively to reduce swell potential and can also slightly increase soil strength. Due to the proprietary nature of the liquid stabilizers, there is a lack of comprehensive laboratory engineering studies of the stabilization mechanism to evaluate the efficiency of such treatment. In this study, a liquid ionic soil stabilizer (LISS) was selected for the laboratory study. The LISS consists of sulfuric acid, phosphoric acid, citric acid, surfactant, and water mixed at a proprietary ratio. Bulk samples of a high expansive soil with PI of 58 were collected in Caddo Mills, TX and tested for particle size distribution, Atterberg limits, standard proctor compaction, and specific gravity. Unconfined compressive and 1D swell tests were performed on specimens with and without treatment. The treated specimens were obtained by hand mixing the dry soils with the LISS at three dilution ratios. Results show that the stabilizer effectively reduces swell potential and decreases the PI of the tested soils.

Expansive soil problem is one of the major problems in the field of geotechnical engineering. The change in water content in unsaturated expansive soil is the major reason of heave or shrinkage of the soil. Numerical simulation is a valuable tool for analysis of water content or suction variation in unsaturated soil. Numerically solving unsaturated soil problems needs the assessment of moisture variation in time and space as a response to the variation of environmental factors such as rainfall and evaporation rate. This study involves a comparison of measured water contents of an expansive clay subgrade beneath a flexible pavement in the country road FM 2 project by TXDOT with numerical simulations conducted using VADOSE/W (Geo-Slope 2012). With the input of soil properties, climate data, and proper initial and boundary conditions, the numerical simulation results gave a good approximation of the water content variations as compared with the field measurements. Both numerical results and field measurements show that significant moisture fluctuations occurred in the clay subgrade under the unpaved shoulder and in the soil below the ditch, while moisture fluctuation was negligible in the subgrade below the pavement. This difference in water content change indicates the probable cause of the volume change as well as the longitudinal crack. After the model calibration and validation on the water content change, the heave/shrinkage of the subgrade soil was evaluated following the empirical equations based on the one-dimensional deformation assumption. For heave or shrinkage prediction the extreme weather (extreme dry and extreme wet period) conditions were considered. This study is of value in the sense that, the numerical results reached a good agreement with the measured moisture fluctuations. The research outcome indicates that VADOSE/W is a robust tool, capable of predicting the moisture content fluctuations below pavement with proper soil properties and initial and boundary conditions specified. This study highlights the convincing capacity of numerical simulations for predicting the moisture fluctuations in expansive soils. The analysis is simple but efficient for use in conventional engineering practice for designs of pavement and shallow foundation on expansive soils.

Potassium chemical injection stabilization has been effectively used to stop the volumetric change of unsaturated clays caused by the drying or wetting of expansive soils below existing buildings experiencing significant foundation settlement. This type of chemical injection is performed inside existing structures with equipment capable of treating 10–15 ft of expansive clays in low and elevated headroom conditions. This paper presents case studies of completed injection projects, including the post-injection laboratory soil swell test results on an existing historical railroad depot located in Temple, TX, a distribution center and a warehouse building both located in Irving, TX. This paper presents an overview of the potassium injection stabilization operations and performance requirements used for various projects, in addition to discussing the methodology used to properly treat the underlying clays while limiting further deterioration to the existing building’s structural stability and serviceability while being cost effective.

Changes in volume of clayey soils during periods of drought lead to considerable repair costs in France and around the world. Individual dwellings are particularly damaged and, so far, technical repairing solutions used are expensive and can be sometimes traumatic for owners (as they are sometimes undertaken inside the house…). Cerema proposes an innovative approach, which consists in watering the foundation soil of a house in order to neutralize natural shrinkage during critical droughts. A damaged house was chosen and equipped with a device, that recovers rainwater and injects it under the foundations. A soil moisture monitoring was also set in order to follow evolution of soil suction. Works took place in Fall 2016 during a severe drought (in France). Early observations suggests encouraging results with, on one hand, suction reduction as the device was used, and on the other hand, closing of existing wall’s cracks. The monitoring is planned for 3 years long.

The engineering properties of base course materials play an important role in the overall stability of a pavement system. The stabilization of sulfate-rich base course materials has been a challenge using traditional calcium-based stabilizers (i.e., cement, lime, and fly-ash) because soluble sulfates present in soils can induce chemical reactions between cement or lime and residual soils, resulting in significant loss of strength and heaving. Introduced as a non-traditional stabilizer, polymer-based products have become popular due to cost efficiency, ease of application, and fast curing times. To explore the feasibility of using polymer-based products for the stabilization of high sulfate soils, a series of unconfined compression strength tests were performed. A sulfate concentration of 20,000 parts per million (ppm) was used in the laboratory testing program. As compared to the strength without soluble sulfate present in the soil, the results revealed a loss of strength in cement, and lime treated specimens but an increase of strength in polymer treated specimens.

The active zone (AZ) constitutes the depth below the ground where soil movements can occur due to moisture/energy changes in unsaturated soils. The upper seasonal zone is affected by changes in moisture contents at the edges of foundation systems which can cause cyclical upward and downward movements as a function of the direction of moisture flux due primarily to climate. Another phenomenon creating soil expansion is the process of moisture equilibration after the construction of the foundation or pavement system at the surface. This equilibration phenomenon can affect soils in the upper seasonal zone as well as in the deeper zone of desiccation.

In the past 30 years approximately 6 million homes have been built in Australia in a time of chaotic weather conditions. The Australian Bureau of Statistics (ABS) figures show that from 2001 to 2015, 93,000 houses were built in the Werribee Plains in the Western and Northern suburbs of Melbourne. This period included the “Millennium” drought and wet periods. Reports commissioned by a number of government organisations estimated that approximately 12% of the houses built in the drought in this part of Melbourne did not meet the footing performance requirements of the Australian footing standard no 2870 after the breaking of this drought. The main factor discussed in this paper is the surface suction which, is recommended as 1.2pF but was greatly exceeded. This paper briefly discusses the weather conditions, the footing designs and the performance of houses with concrete slab footings and the mixed success of the Australian standard characterisation method during this period.

The loads from the civil engineering systems are ultimately transferred to the subsurface soil through various foundations and their load carrying capacity is greatly influenced by the degree of saturation (dos) of subsurface soil and matric suction within the influence zone. The dos of the subsurface soil and matric suction vary with climatic and hydrologic parameters such as rainfall, water table location, and temperature. In this study, a framework is proposed to evaluate the ultimate bearing capacity and elastic settlement of shallow foundation incorporating the site specific hydrological parameters such as historical rainfall and water table records. The results show that the ultimate bearing capacity obtained using the proposed coupled geotechnical-hydrological method is approximately 30% greater than that computed using the conventional method which assumes that the subsurface soil is fully saturated. Furthermore, the elastic settlement computed from the proposed method is approximately 60% smaller than that of conventional method. From this and the parametric studies, an increase in ultimate bearing capacity and a decrease in elastic settlement were observed for the site considered. It is also observed that the benefits are significant when the size of the footing increases.

For retaining walls and pile foundations constructed in expansive soil area, mechanical behaviors of both these geo-structures are significantly influenced by the lateral swelling pressure (LSP) mobilized in the infiltration process. The currently used design methods for retaining walls and pile foundations are mostly based on saturated soil mechanics. Mechanical behavior changes with respect to water content or suction change are not given due consideration. A theoretical framework is developed for determining lateral earth pressure (LEP) behind the retaining wall and uplift friction along the single pile shaft in expansive soil using unsaturated soil mechanics. An example problem is illustrated to highlight the influence of LSP on a retaining wall and a single pile, constructed in typical expansive clay from Regina, Canada. The studies presented in this paper are useful to the practitioners who routinely design geo-structures such as the retaining walls and pile foundations in expansive soils.

Collapsible soils possess a considerable strength when dry, and experience a significant and sudden loss in its strength and excessive settlement upon wetting. In the field, wetting may take place due to the rise of the groundwater level or flooding on the ground surface. End-bearing piles are widely used to penetrate these collapsible layers, in order to transfer the building load to a stronger deeper soil layer. While in the literature, reports can be found dealing with evaluating the interaction between piles and collapsible soil when fully inundated (full saturation), the research on the behavior of this soil when partially saturated is lagging behind. This is due to the complexity of modeling and the lack of field data. This paper presents the results of experimental modeling of an end-bearing pile penetrating collapsible soil subjected to partial inundation due to surface running water. The objective is to determine the drag load acting on the pile’s shaft when the degree of saturation increases along the collapsible soil layer surrounding the pile shaft. The measured drag load is presented as function of the degree of saturation, the collapse potential of the soil layer and pressure acting on the soil.

In this paper numerical analyses are performed to estimate bearing capacities of unsaturated soil shallow foundations by implementing an effective stress based finite element formulation. It is shown that the bearing capacity of these foundations increases due to suction. The effects of hydraulic hysteresis and different drainage conditions are investigated. It is found that the bearing capacity is dependent on the hydraulic loading history of the soil. It is also shown that a constant contribution of suction to the effective stress and strength condition (also referred to as a constant χs condition) provides a good approximation to a constant suction condition and a constant moisture content condition. When foundation loading occurs quickly, meaning a constant moisture content condition prevails, the χs magnitude can be assumed to be equal to the initial in situ value to simplify the calculation of bearing capacity.

This paper presents a new numerical analysis method for one-dimensional, large strain compression of a single undrained, unsaturated soil element. The method builds on the framework of the consolidation model, CS2, which has been used successfully for many consolidation-related engineering applications. In this work, soil properties and relationships are used to model undrained compression from first principles. A numerical study is performed to investigate the effect of increasing applied total stress on element compression and corresponding soil properties. Results indicate that volume change is non-linear and, as the element approaches saturation, an increase in total stress yields a nearly equivalent increase in pore air pressure. Ultimately, this work will serve as a point of departure for the development of a new large strain consolidation model for unsaturated soil.

The unsaturated soil behavior is one of the most significant areas of research in the geotechnical engineering. Desiccation cracking is one of such issues which results in variation of permeability, strength, and creation of zones of weakness. The cracks develop an irregular and/or preferential flow pattern for contaminant transport, thereby affecting the functionality of containment structures, such as for nuclear waste and hazardous wastes. Earlier studies depicted that, cracking characteristics of fine grained soils are mainly influenced by environmental conditions, soil type (mineralogy), evaporation rate, thickness of soil, and suggested various methods to arrest cracking using surfactants and controlling the evaporation rate in the soils. In lieu of the situation, where the cracks initiate due to poor tensile strength and high suction values, the above techniques may not be any more valid. In such situation, various other methods such as arresting plasticity, increasing the tensile strength of the soil by fiber reinforcement has been found effective. To demonstrate this, investigations were conducted in this study on soil specimen subjected to drying and thereby cracking. The present study counters the cracking by adding cement with the addition of sand and fiber. The cracking index of the soil is characterized by the parameter crack intensity factor (CIF), computed resorting to image analysis. Based on the present study, it has been observed that the least crack intensity factor have been succeeded for fiber-reinforced cemented compared to fiber-reinforced un-cemented soil.

Pavements constructed on expansive soil subgrades experience longitudinal cracks due to the desiccation of the subgrade soil from the environmental loads. During the soil desiccation, cracks initiate as a consequence of the soil shrinkage. These cracks can propagate to the overlaying pavement layers. Further, cracks decrease the soil strength and increase its hydraulic conductivity and compressibility. This paper presents a laboratory testing protocol and a simple analytical approach in identifying the crack initiation criteria. Four different laboratory tests have been undertaken to investigate the shrinkage cracks in expansive soils. These tests are suction measurements, free shrinkage test, constrained shrinkage test, and splitting tensile strength test (STT). The constrained shrinkage test was conducted using the restrained ring test (RRT). The soil water characteristic curve (SWCC) in terms of the degree of saturation was predicted by utilizing the gravimetric water content measurements and the soil shrinkage characteristic curve. The suction measurements and the tensile stress measurements from the RRT were tracked from the beginning of the shrinkage test until the crack initiation stage. The results confirm that a crack first initiates when the suction becomes close to the air entry value (AEV). A qualitative comparison between the tensile stress from the RRT and the tensile strength from the STT was carried out. It was found that the tensile stress increases with the soil shrinkage while the tensile strength starts to decrease at a suction stage close to the AEV.

The study of crack generation and propagation in soils is a very complex problem and possesses several challenges. In this paper a novel approach is proposed to model the formation and propagation of cracks in drying soils. This technique consists of introducing finite elements with high aspect ratio between the regular elements of the original mesh. The main components of the mathematical formulation of the proposed approach are presented, with special emphasis on the constitutive models. Three drying tests were simulated considering circular plates with different heights in order to study the development of desiccation cracks. Thus, the drying process was reproduced by imposing a suction gradient at the top of the soil specimen. The obtained crack patterns were compared against experimental tests to validate the proposed method. The results indicated that the technique was able to capture the main phenomena related to the desiccation crack.

The formation and propagation of desiccation cracks in soil is extremely complex because of the coupling between hydraulic and mechanical phenomena in soil. This paper presents a study on the desiccation cracking by a numerical simulation approach. The initiation and propagation of cracks are investigated using a finite element code including damage-elastic cohesive crack elements to describe the behaviour of cracks. The simulation of multi cohesive cracks shows that the cracking occurs successively to form the different cracks families. But the cracks in the same family can appear simultaneously. In addition, the cracks can propagate unstably in the initiation moment and propagate after progressively.

A locally available expansive clay was reinforced with polypropylene fibers of cut-lengths 6, 12, and 18 mm and a fiber content of 0.5% by dry weight of the clay. The unreinforced and fiber-reinforced clay specimens mixed at a state close to liquid limit were subjected to five subsequent drying and wetting cycles. The desiccation tests were performed at a constant temperature and relative humidity. Crack propagation, moisture loss due to evaporation, and crack feature measurements were monitored. Digital image analysis (DIA) was used to measure the surfacial crack feature measurements. The results indicate that the desiccation cracking behavior was considerably affected by the subsequent drying and wetting cycles. The severity of cracks in unreinforced specimens increased with an increase in the number drying-wetting cycles and reached a stable value after three cycles. However, in the case of polypropylene fiber-reinforced clay specimens, bridging effect was predominant, and in turn integrity of soil matrix was maintained. The smaller crack widths and reduced cracked areas indicate the significant influence of discrete and randomly distributed fiber reinforcement in expansive clay layers.

Soil-structure interfaces occur at the contact surfaces between the soils and the engineering constructions, such as retaining wall, foundations, and others, where reciprocal transition of stresses takes place at interfaces. An intrinsic motivation of this work stemmed from observations of problematic behavior of natural deposited soils and engineered soils during drying-wetting processes due to variations in annual thermal conditions. In response to variations in environmental conditions, unsaturated soils may experience different hydraulic, and mechanical behavior under wetting and drying processes due to hysteresis. This may cause evolution and propagation of cracks in the soil. In this study, four drying tests (shrinkage tests) were performed using identical circular plates. Two plates were fabricated; one with smooth-base and the other one with circular-grooved-base. The smooth plate was employed to allow the soil sample to shrink freely without crack formation, while the groove plate was used to provide restraints during shrinkage where cracks may develop during drying process. Furthermore, soil samples were tested under saturated and unsaturated in order to investigate the influence of initial moisture conditions of soils on the initiation and propagation of cracks during one drying and wetting cycle. Crack patterns were quantitatively obtained using image analysis techniques. The results of the drying tests refer to the important impact of soil-interface and the initial saturation conditions on the initiation and propagation of cracks during wetting and drying cycle.

Three Brazilian tropical soils are studied: residual soil of diabase, granite, and gneiss. The water retention curves of these soils were determined using compacted specimens under the optimum moisture conditions of their respective compaction curves. For the residual soil of gneiss, water retention curve was obtained with the use of suction plate, pressure plate, and filter paper. Granite and diabase residual soils have the water retention curves determined using filter paper technique following the drying path after initial saturation of the specimen. Characterization tests, determination of the grain size distribution curve, and resistance parameters were made to complement the experimental studies. The higher compaction density of the granite residual soil, associated with its higher clay fraction, is reflected in a higher air entry value. For water moisture contents of less than 25%, a stabilization of volumetric variations of the tested specimens during the drying procedure is observed in order to determine the water retention curve.

Dynamic properties such as shear modulus (G) and damping ratio (ξ) are important to understand the dynamic and non-linear stress-strain behaviours of soil from small to large strains. Limited studies have been done to understand dynamic and static properties of residual soil. The objective of this paper is to present the dynamic properties of saturated and unsaturated residual soil obtained using a modified cyclic triaxial apparatus. Undisturbed residual soil from Bukit Timah Granite in the northern part of Singapore was used. The test includes wave propagation test and cyclic triaxial test. Bender element based wave propagation test was used to determine the very small-strain shear modulus, G_max and small-strain damping ratio, ξ_min. Effects of varying net confining pressure and matric suction on P-wave, S-wave velocities, stiffnesses, and damping ratios (V_p, V_s, G_max,, and ξ_min, respectively) for the residual soil are discussed. Very small-strain dynamic properties from bender element tests are compared with those in the literature. Shear modulus and damping ratio at strains ranging from 0.05 to 1% were estimated using strain-control cyclic triaxial test. The number of cycles applied on the soil specimen was 40 and the loading frequencies were 0.2 and 0.5 Hz. Shear modulus and damping ratio for a wide range of strain were obtained at various effective mean confining pressures for saturated soil and various matric suctions at constant net confining pressure for unsaturated soil. It is observed that effective mean confining pressure and matric suction are crucial parameters which affect shear modulus and damping ratio of soil. Finally, the damping ratio and normalized shear stiffness of the residual soil (ξ and G/G_max) for a wide range of strain are found to be consistent with previous studies.

Soil structure depends on many variables in the system particles and surrounding soil water, thus it is reasonable to suppose that the magnitude of soil collapse strain can vary, under certain use conditions, like following sewer and different chemical effluent leakage. This paper deals with the influence of liquids of various compositions and chemical properties on the collapse behavior of two typical tropical soils, considering the magnitude of collapse strain, its development over time, and the variation in soil porosity as a result of the volumetric reduction after soil collapse. The paper shows that the pH of liquids is related to soil collapse to a limited extent, as is the case for organic liquids and sodium hydroxide solutions used in this study. However, pH is not an efficient unifying property in indicating the magnitude of soil collapse. Collapse strain also increases with the dielectric constant of the organic liquids used in the tests. Soil collapse strain reduces the amount and diameter of soil macropores, modifying soil pore distribution.

This paper presents the results of an experimental study about the variation of the small strain stiffness under K₀ conditions which means that the lateral deformations are restricted, for two lateritic soils with different weathering intensities. The specimens followed loading and unloading K₀ stress paths in an oedometer apparatus equipped with piezoceramic transducers capable to generate and receive shear (S) and compression (P) waves simultaneously. The influence of the loading process on the elastic small strain modulus (E₀ and G₀) was studied. The compressibility curve of the soils was established for different structural soil conditions (natural, unstructured, compacted, and saturated). For the samples at natural conditions (i.e., undisturbed), the more weathered soil exhibits a greater stiffness than the underlying Horizon; however, when removing the structure by reshaping, the stiffness of the upper Horizon decreases below of the lower Horizon. The addition of water softens cementation and the effect of suction. The results confirm that weathering provides a structure that increases the stiffness of soils.

Some tropical soils have distinct behavior due to their structure. In the last decades, new techniques improved the analysis of soil structure and enabled the correlation between the mentioned structure and the hydromechanical behavior of soils. In this work, a hydromechanical and microstructural characterization of slurry state of Brasilia soil was performed. The hydromechanical characterization was made with consolidation and water retention curve tests. The analysis of the structural characteristics of these soils was obtained from mercury intrusion porosimetry (MIP) test. It is assumed that different samples such as undisturbed soil, compacted soil and slurry have the initial structure affected in a distinct way by external loading. Thus, the main objective was to evaluate how the slurry sample had its structure affected by external loading and, also compare with undisturbed and compacted samples. Furthermore, the results showed that both the water retention curve and the pore size distribution curve are bimodal and that microstructure remained virtually unchanged. Moreover, as loading was increased there was a reduction of the macro pores, either in saturated and unsaturated samples.

In this paper, the change of electrical resistivity and matric suction with volumetric water content is investigated using the electrical resistivity measurement and pressure plate test for different densities compacted granite residual soils. The results show that the resistivity value decreases with increasing water content and dry density, and tends to constant at higher values. The shapes of the soil-water characteristic curve (SWCC) and electrical resistivity-water characteristic curve (RWCC) of the investigated soils are very similar. Both curves of RWCC and SWCC can be well described by the van Genuchten model and observed a better correlation i.e., the fitting parameters of RWCC and SWCC are linearly proportional to each other. Based on these two characteristic curves mentioned above, the modification of the van Genuchten model is proposed to develop a direct relationship between electrical resistivity and matric suction for the compacted residual soils. Moreover, the comparison of fitting parameters of the resistivity-matric suction characteristic curve with RWCC and SWCC parameters is also linearly proportional to each other. It can be concluded that the matric suction is strongly depend on the electrical resistivity of the soil, after knowing the fitting parameters of the RWCC, it may be possible to predict the SWCC from the resistivity data.

The interpretation of in-situ tests for site characterization of unsaturated tropical soil sites should consider both the spatial and the seasonal variability caused by the soil suction on the strength and stiffness parameters. This paper presents and discusses the suction-controlled and conventional triaxial with bender elements and in-situ tests data together with water content profiles, as well as soil water retention curves (SWRC) from an unsaturated tropical sandy soil site, to examine the influence of the soil suction on the strength and stiffness soil parameters. The results showed the influence of soil suction on the data interpretation of CPT and DMT to define strength and stiffness parameters for the top layer. This paper point out the importance of considering the seasonal variability caused by suction on the characterization of the study site.

To evaluate the stress-strain behavior of soils under loading in engineering practice it’s usually required to know the natural, initial state of stresses acting in the soil mass. This is not an easy task when dealing with sedimentary soils. When it comes to tropical residual soils, this challenge is even greater due to variations in the weathering degree and soil profile moisture. The natural state of the soil is usually expressed by the coefficient of at-rest earth pressure (K₀). To determine K_0, three classes of methods can be used: correlation methods, in situ determination and laboratory tests. On this research, one rigid cell with a system of suction control was used to obtain values of K₀ in laboratory tests, three samples of gneissic residual soil with different degrees of weathering were analyzed, and each one tested in three different levels of moisture. In general, the results attended the expected trend.