Mechanical behaviour of unsaturated kaolin (with isotropic and anisotropic stress history). Part 1: wetting and compression behaviour

Over the last 40 years considerable progress has been made in understanding the complex behaviour of unsaturated soils. Research using constitutive modelling has extended the critical state framework and the concept of yielding in saturated soils to encompass unsaturated soils experiencing suction. However, validation testing of the framework for unsaturated soils has shown disagreement with the basic propositions. The main reason for this disparity is the anisotropic properties of the soil specimens tested as a result of preparation using one-dimensional compaction. The paper describes the detailed testing carried out to justify this statement. As part of the work presented, samples of unsaturated kaolin were prepared using isotropic compression. The suctions in these samples were reduced to predefined values by wetting under low isotropic loading. The pore size distributions, the pressure–volume relationships and yielding under subsequent isotropic loading are compared with tests on samples prepared by statically compressing kaolin into a one-dimensional compaction mould. The anisotropically compressed samples had initial water contents and specific volumes similar to those of the isotropically prepared samples and were also tested under reducing suctions; they exhibited distinctly different behaviour when tested under similar conditions. The results obtained from the isotropically prepared and tested samples have shown, probably for the first time, the existence of a unique normal compression surface that is not dependent on the initial conditions of the samples. The shape of the loading–collapse (LC) yield locus is shown to be different from the generally accepted form.

Mechanical behaviour of unsaturated kaolin (with isotropic and anisotropic stress history). Part 2: performance under shear loading

Validation of a framework for unsaturated soil behaviour has frequently resulted in disagreement with basic propositions. A primary reason for this disparity is considered to be attributable to the anisotropic properties of the soil specimens tested as a result of preparation using one-dimensional compaction. As part of the work presented, comparison is made between tests on samples of unsaturated kaolin prepared at identical specific volumes and specific water volumes using isotropic compression and one-dimensional compression. The suctions in the samples were reduced to predefined values by wetting under low isotropic loading in a triaxial cell. The samples were then taken through various stress paths to failure, defined as the critical state strength, while the suctions were held constant. Stress path tests were also performed on samples without reducing the suction to predefined values. In the latter, constant water mass tests, the suctions were allowed to vary and were measured using a psychrometer. The results of the tests at critical state are compared with the propositions of Wheeler and Sivakumar. The shear strengths of samples with isotropic previous history are shown to be significantly greater than those of samples with one-dimensional stress history when plotted against the mean net stress. The normal compression lines, critical state lines and yield characteristics are also shown to be significantly influenced by the previous stress history and are shown to be different for isotropically and one-dimensionally prepared samples.

Saturated-unsaturated flow through leaky dams

This research investigated seepage flow through leaky dams using the well known finite-element method. Different areas, locations, and hydraulic conductivities of leaks were examined. An area of leak, equal to 4.4% of the core area, increased the seepage flow through the dam to be about 9.5 times the seepage flow through tight (nonleaky) core. This happened for a dam having a downstream horizontal drainage filter. When the drainage filter did not exist, the increase of flow because of the same area of leak was about seven times the flow through a tight core. When the leak existed at the centerline of the core in the out-of-plane direction, its impact was slightly greater than when it existed at the edge of the core. Moreover, as the location of the leak moved up vertically, its impact was observed to be less. It was also observed that when the leak existed in curtain wall driven into underneath the dam, its impact was not significant compared with the case when it existed in the core.

Experimental observations of the stress regime in unsaturated compacted clay when laterally confined

Construction processes often involve reformation of the landscape, which will inevitably encompass compaction of artificially placed soils. A common application of fill materials is their use as backfill in many engineering applications, for example behind a retaining wall. The post-construction behaviour of clay fills is complex with respect to stresses and deformation when the fills become saturated over time. Heavily compacted fills swells significantly more than the lightly compacted fills. This will produce enhanced lateral stresses if the fill is laterally restrained. The work presented in this paper examines how the stress regime in unsaturated clay fills changes with wetting under laterally restrained conditions. Specimens of compacted kaolin, with different initial conditions, were wetted to various values of suction under zero lateral strain at constant net overburden pressure which allowed the concept of K _{0 (the ratio between the net horizontal stress and the net vertical stress) to be examined. Tests were also carried out to examine the traditional concept of the earth pressure coefficient ‘at rest' under loading and unloading and its likely effects on the stress–strain properties. The results have shown that the stress regime (i.e. the lateral stress) changes significantly during wetting under laterally restrained conditions. The magnitude of the change is affected by the initial condition of the soil. The results have also indicated that the earth pressure coefficient ‘at rest' during loading (under the normally consolidated condition) is unaffected by suction and such loading conditions inevitably lead to the development of anisotropic stress–strain properties}

An elasto-plastic critical state framework for unsaturated soil.

Data from a series of controlled suction triaxial tests on samples of compacted speswhite kaolin were used in the development of an elasto–plastic critical state framework for unsaturated soil. The framework is defined in terms of four state variables: mean net stress, deviator stress, suction and specific volume. Included within the proposed framework are an isotropic normal compression hyperline, a critical state hyperline and a state boundary hypersurface. For states that lie inside the state boundary hypersurface the soil behaviour is assumed to be elastic, with movement over the state boundary hypersurface corresponding to expansion of a yield surface in stress space. The pattern of swelling and collapse observed during wetting, the elastic–plastic compression behaviour during isotropic loading and the increase of shear strength with suction were all related to the shape of the yield surface and the hardening law defined by the form of the state boundary. By assuming that constant–suction cross–sections of the yield surface were elliptical it was possible to predict test paths for different types of triaxial shear test that showed good agreement with observed behaviour. The development of shear strain was also predicted with reasonable success, by assuming an associated flow rule.

Influence of compaction procedure on the mechanical behaviour of an unsaturated compacted clay, Part 1: Wetting and Isotropic Compression

The influence of compaction pressure, compaction water content and type of compaction (static or dynamic) on subsequent soil behaviour during wetting and isotropic loading has been investigated by conducting controlled-suction tests on samples of unsaturated compacted speswhite kaolin. The results are interpreted within the context of an elastoplastic framework for unsaturated soils, to examine which compaction-induced effects can be explained simply by variation in the initial state of the soil and which require that soils produced by different compaction procedures are modelled as fundamentally different materials. The compaction pressure influences initial state, by affecting the initial position of the yield surface, but it also influences, to a limited degree, the positions of the normal compression lines for different values of suction. The compaction water content influences the initial suction, but also has a significant influence (greater than does compaction pressure) on the positions of the normal compression lines. A change from static to dynamic compaction has no significant effect on subsequent behaviour