Skip to main content

Constitutive and numerical modelling of saturated and unsaturated soils

Periodic Reporting for period 1 - Coupled (Constitutive and numerical modelling of saturated and unsaturated soils)

Reporting period: 2017-02-01 to 2019-01-31

The focus of the project COUPLED was to have a better understanding of the distinct directional dependent response of problematic clayey soils, which results from a preferential arrangement of their soil particles in a given direction (anisotropic fabric). To achieve this goal, the formulation and subsequent validation of a coupled anisotropic constitutive model for soils was developed.
Due to the one-dimensional loading conditions applied to the soil during the compaction process, it is typical in compacted soils to exhibit an anisotropic fabric. Furthermore, the compaction process is carried out under unsaturated conditions, which means that compacted soils are typically unsaturated. Very often, civil engineers use compacted soils for the construction of embankments for rails, roads and flood defences. When the soil of such embankments is subjected to rainfall, it progressively saturates, eventually becoming fully saturated. Subsequently, when the embankment is subjected to dryer conditions, evaporation of the pore water may occur, which restores the initial partially saturated condition of the soil.
A good understanding of how these alternations between saturated and unsaturated conditions of the soil occur as a consequence of seasonal variations is key to the safety of the geotechnical design. This is because both the mechanics and the hydraulics of the embankment are controlled by how water is stored within the pores of the embankment. Such understanding of the processes of saturation and desaturation of the in-situ soil is even more relevant today as climate change is manifesting itself year after year in the form of extreme weather events.
Surprisingly, even though compaction is a broadly used method in civil engineering to improve the mechanical properties of the in-situ soil, very few models for unsaturated soils (if any) account for fabric anisotropy. Therefore, the development of a constitutive model able to handle anisotropic behaviour under partially and fully saturated conditions becomes very relevant for a proper safety assessment of any geotechnical design involving anisotropic soils. Modelling realistically soil behaviour by means of advanced constitutive models is important not only for gaining engineering knowledge on soil’s response, but also because it provides a numerical tool from where soil responses under various loading or environmental conditions can be quantified in a consistent manner. In addition, application of such engineering knowledge to geotechnical design reduces the very large uncertainties commonly faced by geotechnical engineers, contributing to the overall safety of the infrastructure.
The overall objective of COUPLED was to give insight, by means of a novel anisotropic constitutive model, on the mechanical behaviour of soils exhibiting an anisotropic fabric (including fully and partially saturated conditions). The core of the new model is the Glasgow Coupled Model (GCM), an existing fully coupled elasto-plastic constitutive model for saturated and unsaturated soils. To account for anisotropic responses, the GCM is combined with the S-CLAY1, an existing critical state elasto-plastic model for saturated anisotropic soils. The newly proposed model has been named the Anisotropic Glasgow Coupled Model (A-GCM) and its abilities to handle anisotropic behaviour under partially and fully saturated conditions is demonstrated in COUPLED by comparing model simulations against high quality experimental data.
The first half of COUPLED focussed on understanding and modelling transitions between saturated and unsaturated soil conditions in the context of the GCM. Various existing experimental observations were compared against model simulations to challenge the capabilities of the model across these transitions. These were done not only in terms of volume changes but also in terms of how water is stored within the soil pores (water retention behaviour). The main findings of this specific part of the research are published in “From saturated to unsaturated conditions and vice versa” by Lloret-Cabot et al. (Acta Geotechnica, 2018), an article that has attracted a considerable interest within the research community, with one discussion & corresponding reply to the Editor of the journal, and about 2.2K downloads since its online open access publication in August 2017.
The second half of COUPLED concentrated on the formulation of the A-GCM. This task involved the combination of the GCM (a fully coupled elasto-plastic constitutive model for saturated and unsaturated soils) with the S-CLAY1 (a critical state elasto-plastic model for anisotropic saturated soils). In contrast to the GCM (which is formulated for isotropic material behaviour), the new A-GCM has the ability to represent the behaviour of soils with an initial fabric anisotropy as well as the development or erasure of this initial anisotropy during plastic straining. Additionally, the A-GCM is able to handle these anisotropic features of response under saturated and unsaturated conditions.
Once formulated, the relevant capabilities of the model were successfully validated against the high quality experimental data of Al-Sharrad (2013), who carried out his PhD research at the soil mechanics laboratory of the Host Institution (University of Glasgow). The development and validation of the new A-GCM model is currently being written up for submission to a leading international journal in geotechnical engineering.
In addition to the scientific dissemination (involving peer-reviewed journal publications, peer-reviewed conference articles and presentations at international conferences) the fellow visited and collaborated with the Centre of Excellence in Geotechnical Science and Engineering at the University of Newcastle (Australia). During his research visits, the fellow had the chance to disseminate further his research outside Europe, as well as to discuss and collaborate with well-renowned experts in the area of unsaturated soil mechanics.
Complementary to the research carried out within COUPLED, the fellow had the chance to be part of TERRE (ETN-GA-2015-675762) co-supervising one early stage researcher of the project. TERRE gave the fellow also the chance to interact, disseminate and discuss with a broader audience, including industry partners and academics, from very diverse fields of research.
Overall, COUPLED has enriched CV and research experience of the fellow, something that has certainly played a role on the fact that he has secured a permanent academic position at Durham University, UK.
The development of an anisotropic model for soils, such as the A-GCM proposed in COUPLED, provides a numerical tool from where to study, in a consistent manner, the complex anisotropic behaviour of soft soils when these are partially or fully saturated. Soft soils are present worldwide, and they are often considered problematic because of their poor mechanical properties. Even though soft soils are quite well-understood when fully saturated, very little research has been done on soft soils under partially saturated conditions, and even less on soft soils that are subjected to seasonal alternations between partially and fully saturated conditions.
Insights achieved through COUPLED form the basis for further research on finite element analysis to address practical geotechnical problems involving anisotropic behaviour covering saturated and unsaturated conditions.