Interpretation of the loading–wetting behaviour of compacted soils within the “MPK” framework. Part I: Static compaction

Depending on the state paths, loading–wetting of compacted unsaturated soils can exhibit complex volumetric behaviour, such as swelling, collapse, collapse followed by swelling, swelling followed by collapse, and swelling pressure development. Microscopically, these behaviours arise from complex interactions among applied stresses, air–water pressure deficit or suction at the water menisci, moisture content or degree of saturation in the voids, and the nature of the micro- and macrosoil aggregates of compacted soils that depend on the level of suction. While significant advances have been made in modelling hydromechanical behaviour of compacted unsaturated soils taking these interactions into account, input parameter determination requires advanced testing equipment and the testing processes can be very time-consuming. In 2012, a relatively simple and practical framework within the void ratio – moisture ratio (water volume / solid volume) – net stress space (referred to as the MPK framework) has been proposed by Kodikara to explain–predict these state paths. A desirable feature of this framework is that it identifies a direct link between the well-known compaction curve and the compacted soil constitutive behaviour. This paper presents a comprehensive series of tests on statically compacted soils, the results of which are in close agreement with this framework. Two soil types, namely lightly reactive kaolin and more reactive clay (referred to as Merri Creek soil), were used in the testing. The soils were prepared with different moisture contents from the dry state and statically compacted at constant water content to obtain void ratio – moisture ratio – net stress constitutive surfaces, as well as soil specimens for state path tests. The state path test results of yielding under loading, collapse under wetting, swelling pressure development, and change in yield pressure due to wetting are explained within this framework. In addition, some published data on a silty soil mixture were also analysed, highlighting that the framework is valid, regardless of the degree of reactivity of the soil. Suction was not measured in the authors’ experiments, as it was not required to explain the above state paths according to this framework. However, it is recognised that suction is the conjugate state variable to the moisture content. Therefore, in future experiments, suction will be measured and its role will be fully explained within the framework, adding more generality.