This research studied the effects of the addition of ground granulated blast furnace slag (ggbs), activated by 2% calcium hydroxide (Ca(OHh, on the strength, permeability and porosity development of a laboratory prepared clay mix (kaolinite with and without 6% gypsum) and a natural sulphide-bearing clay soil, Lower Oxford Clay. Based on shear, compressive and indirect tensile strength testing, it was found that an increase in the stabiliser slag/lime ratio results in substantial strength increase even after short curing periods (up to 12 weeks). This increase in strength is more pronounced if curing is at elevated temperatures (30 °C). The presence of sulphates (6% gypsum=2.73% SO3) resulted in an accelerated increase in the strength development for stabilised kaolinite, which was comparable to that of stabilised Lower Oxford Clay. In the absence of sulphates, large ggbs additions were only activated effectively at higher curing temperatures (20 and 30 °C) after curing periods of 24 weeks and beyond, although it is suggested that 2% lime creates a sufficiently alkaline environment for activation. The degree of slag activation and thus the subsequent cementation process was reflected by an increase in the percentage of the pore volume occupied by pores with a radius ::s;0.0Sμm, which is usually associated with the pore fraction characteristic of cementitious gels. The increase in slag addition, for kaolinite mixes, was accompanied by a reduction in total porosity. Specimens made from Lower Oxford Clay exhibited a significant increase in pore volume at higher slag additions. This is interpreted as being due to the creation of pore space resulting from restrained shrinkage of gels by inert particles during drying in this coarser, natural clay. No significant trend in the effect of curing temperature on the pore size distribution could be identified from the data. The development of permeability, however, showed some sensitivity to curing temperature. Results from specimens cured at 20 and 30 °C showed an accelerated reduction in their k-values in comparison to samples which had been cured at 10 °C. However, little correlation between measured permeability and exhibited pore size distribution could be established which is believed to be due to the strong influence of shrinkage during drying prior to mercury intrusion porosimetry in the dimensionally semi-stable soil system. The volume stability of stabilised specimens during frost action was assessed in a series of 12 freeze-thaw cycles, which were carried out in accordance to the German proposal for a European Pre-Standard. Generally an increase in the curing period prior to frost action and higher overall sample porosity resulted in relatively better performance during frost action. The influence of the slag/lime and slag/gypsum ratio on the swelling potential upon soaking was assessed in long-term soaking tests and the underlying causes were identified by findings from microstructural investigations including SEM and TG analysis. These results contributed to a better understanding of the slag activation process. In an alkaline environment slag hydration appears to be triggered earlier by sulphate, due to the more intensive disturbance of a thin protective layer of cementitious products on the slag grains. Disruption of this layer, for example by ettringite formation, exposes more unreacted slag grain surface, which will subsequently start to hydrate. Findings were complemented by two case studies, one which investigated the cause of substantial heave on a German highway on a microscale and the other which assessed the technical performance and the economic implications of a full-scale trial utilising the stabilisation technique with lime and ggbs for a temporary diversion. The overall findings from the projects indicate that soil stabilisation with lime and ggbs is, particularly for soils with significant sulphate/sulphide content, a feasible and environmentally friendly alternative to the classic soil stabilisation methods.

Global concern about the environmental impacts of waste disposal and stringent implementation of environmental laws lead to numerous research on recycled materials. Increased awareness about the inherent engineering values of waste materials, lack of landfill sites and strong demand for construction materials have encouraged research on composite materials, which are either fully or partly made of recycled materials. This trend is particularly strong in transportation and geotechnical projects, where huge quantities of raw materials are normally consumed. Owing to the low mass-to-volume ratio, disposal of Expanded Polystyrene (EPS) is a major problem. In addition, EPS recycling methods are expensive, labour intensive and energy demanding. Hence, this thesis is focused on the development of a new soil composite made by mixing recycled EPS with expansive clays. Given the high cost of damage to various buildings, structures and pavements caused by the unpredictable ground movements associated with expansive soils, it has been considered prudent to try and develop a new method of soil modification using recycled EPS beads as a swell-shrink modifier and desiccation crack controller. The innovative application of recycled EPS as a soil modifier will minimise the quantity of waste EPS destined to the landfill considerably. An extensive experimental investigation has been carried out using laboratory reconstituted expansive soils - to represent varied plasticity indices - consisting of fine sand and sodium bentonite. Three soils notated as SB16, SB24 and SB32 representing 16%, 24% and 32% of bentonite contents respectively were tested with four EPS contents of 0.0%, 0.3%, 0.6% and 0.9%. The tests performed include compaction, free swell, swell pressure, shrinkage, desiccation, shear strength and hydraulic conductivity. All the tests have been performed at the respective maximum dry unit weight and optimum moisture content of the mixes. It has been observed that by mixing of recycled EPS beads with the reconstituted soil, a lightweight geomaterial is produced with improved engineering properties in terms of dry unit weight, swelling, shrinkage and desiccation. The EPS addition depends on the moulding moisture content of the soil. With increasing moisture content, additional EPS can be added. Also, there is a reduction in dry unit weight with the addition of EPS. Furthermore, the reduction of swell-shrink potential and desiccation cracking in soils, for example, is related to the partial replacement of soil particles as well as the elasticity of the EPS beads. There is a reduction in shear strength with the addition of EPS to soils. However, mixing of chemical stabilisers along with EPS can enhance the strength in addition to improved overall properties.

A numerical formulation for two-dimensional electro-osmotic consolidation in soft clays was derived from the basic equations of fluid flow, current flow and virtual work law. And, a well known elasto-plastic soil model, Modified Cam Clay was embedded into the formulation and implemented into the finite element program AFENA. The formulation was evaluated by comparing the predicted settlement and pore water pressure response with the values obtained from laboratory tests. The tests were conducted in an electro-osmotic triaxial apparatus, which was modified from a standard triaxial apparatus to facilitate electro-osmotic consolidation and required measurements. A series of electro-osmotic consolidation tests under different initial stress conditions were conducted to evaluate the finite element model and very good agreements between the observed and predicted results were observed. Another set of electro-osmotic tests were conducted with the similar initial stress, but different boundary conditions to examine the effects of electrochemical changes during the electro-osmosis. The study showed an apparent increase in the preconsolidation pressure and alteration in the coefficient of consolidation as a result of the electrochemical changes. A one-dimensional electro-osmotic consolidation problem was simulated and analysed, and the settlement and pore water pressure responses were compared with the solutions obtained from Esrig's (1968) one-dimensional theory. Another problem involving combined electro-osmotic and direct loading consolidation was also analysed and the results were compared with the solutions obtained from Wan and Mitchell's (1976) theory. Excellent matches were observed in both cases mentioned above for constant values of electro-osmotic and hydraulic permeabilities. However, the effects of varying electro-osmotic and hydraulic permeabilities, which are practically far more significant and not adopted in the theories mentioned above, were also analysed and the results presented. An attempt was made to simulate and analyse one of the successful field trial of electro-osmotic stabilisation conducted by Bjerrum et al (1967) on Norwegian quick clay. Good agreement between the predicted and reported settlement was observed for the first 50days of the treatment period. However, the model over-predicted the settlement after this period and the possible causes for this variation are discussed.

Microbial-Induced Calcium Carbonate (CaCO3) Precipitation (MICP) is a biological process in which microbial activities alter the surrounding aqueous environment and induce CaCO3 precipitation. Because the formed CaCO3 crystals can bond soil particles and improve the mechanical properties of soils such as strength, MICP has been explored for potential engineering applications such as soil stabilisation. However, it has been difficult to control and predict the properties of CaCO3 precipitates, thus making it very challenging to achieve homogeneous MICP-treated soils with the desired mechanical properties. This PhD study investigates MICP at both micro and macro scales to improve the micro-scale understandings of MICP which can be applied at the macro-scale for improving the homogeneity and mechanical properties of MICP-treated sand. A microfluidic chip which models a sandy soil matrix was designed and fabricated to investigate the micro-scale fundamentals of MICP. The first important finding was that, during MICP processes, phase transformation of CaCO3 can occur, which results in smaller and less stable CaCO3 crystals dissolving at the expense of growth of larger and more stable CaCO3 crystals. In addition, it was found that bacteria can aggregate after being mixed with cementation solution, and both bacterial density and the concentration of cementation solution affect the size of aggregates, which may consequently affect the transport and distribution of bacteria in a soil matrix. Furthermore, bacterial density was found to have a profound effect on both the growth kinetics and characteristics of CaCO3. A higher bacterial density resulted in a quicker formation of a larger amount of smaller crystals, whereas a lower bacterial density resulted in a slower formation of fewer but larger crystals. Based on the findings from micro-scale experiments, upscaling experiments were conducted on sandy soils to investigate the effect of injection interval on the strength of MICP treated soils and the effects of bacterial density and concentration of cementation solution on the uniformity of MICP treated soils. Increasing the interval between injections of cementation solution (from 4 h to 24 h) increased the average size of CaCO3 crystals and the resulting strength of MICP-treated sand. An optimised combination of bacterial density and cementation solution concentration resulted in a relative homogeneous distribution of CaCO3 content and suitable strength and stiffness of MICP-treated sand. This thesis study revealed that a microfluidic chip is a very useful tool to investigate the micro-scale fundamentals of MICP including the behaviour of bacteria and the process of CaCO3 precipitation. The optimised MICP protocols will be useful for improving the engineering performance of MICP-treated sandy soils such as uniformity and strength.

During a survey of black cotton soil zones of Karnataka, indigenously stabilized black cotton soil deposits were encountered in Belgaum, Bijapur, Bagalkot and Gadag Districts of Karnataka. These modified black cotton soils have low swelling and negligible shrinkage tendencies. Owing to their low volume change potential on moisture content changes, these soils are widely preferred in earth construction activities. The exact origin of these modified black cotton soil deposits is not known. According to anecdotal references, these soils were prepared by mixing unknown proportions of wood ash, organic matter and black cotton soil and allowing them to age for unknown periods of time. As wood-ash was apparently used in their preparation, these modified black cotton soils are referred to as ash-modified soils (AMS) in the thesis. The practice of preparing ash-modified soils is no longer pursued in black cotton soil regions of Karnataka and the available supply of this indigenously stabilized soil is being fast depleted. Also, attempts have not been made to characterize the physico-chemical and engineering properties of AMS deposits of Karnataka. Given the widespread utilization of ash-modified soils in black cotton soil areas of Karnataka, there is a need to understand their physico-chemical and engineering behaviour and the physico-chemical mechanisms responsible for their chemical modification. Swelling and shrinkage of expansive soil deposits are cyclic in nature due to periodic climatic changes. Chemically stabilized black cotton soil deposits are also expected to experience cyclic wetting and drying due to seasonal climatic changes. The impact of cyclic wetting and drying on the swelling behaviour of natural expansive soils is well-documented. However, the impact of alternate wetting and drying on the swelling behaviour of admixture stabilized expansive soils (these include natural - ash-modified soils and laboratory - lime stabilized black cotton soils) has not been examined. Such a study would be helpful to assess the long term behaviour of admixture stabilized soils in field situations. To achieve the above objectives, experiments are performed that study: 1.The physico-chemical and engineering properties of ash-modified soils from different Districts of Karnataka. The physico-chemical and engineering properties of natural black cotton soil (BCS) specimens from locations adjacent to ash-modified soil deposits are also examined to understand and evaluate the changes in the engineering characteristics of the ash-modified soils due to addition of admixtures. 2. Identify the physico-chemical mechanisms responsible for the chemical stabilizationof ash-modified soils. 3.The influence of cyclic wetting and drying on the wetting induced volume changebehaviour of admixture stabilized black cotton soils, namely, ash-modified blackcotton soils and lime stabilized black cotton soils.

Trifolium repens, Lotus pedunculatus and Holcus lanatus were oversown on two recent soil slip surfaces at AgResearch’s Ballantrae pastoral hill‐country farm near Woodville. The two slip surfaces were located on (Manamahu steepland soil) sedimentary mudstone. One slip had a north aspect and the other had a south aspect. Both slips were located on a land class 6 with slope 28‐330. The pasture species were oversown during early spring and the percentage seedling emergence and early establishment from viable seeds oversown was analysed at early spring (Day 15), late spring (Day 45), early summer (Day 90), and late summer (Day 120). The slip surfaces showed micro‐climatic extremes in terms of both soil moisture and surface temperatures during the summer period. Significant differences (P < 0.05) were found in soil moisture between north and south facing slip surfaces. Higher soil moisture and lower soil mean temperature were recorded on the south aspect slip surface. Significant differences (P < 0.05) were found between the three pasture species in terms of seedling emergence and early establishment. Significant differences (P < 0.05) were also found with aspect. The south aspect slip surface had a higher percentage of seedling emergence and earlier establishment for all the species. Interaction between species by aspect became significantly different (P < 0.05) at Day 90 and Day 120. The main effects of time and species were also significantly different (P < 0.05) illustrating seedling emergence and establishment as a race against time. Trifolium repens was a more successful pasture specie, than L. pedunculatus and H. lanatus due to its higher consistency on both north and south slip surfaces. Oversowing T. repens during early spring is a viable option for rehabilitation of recent soil slips in Manawatu.