Experimental and theoretical analysis of cracking in drying soils

The thesis focuses on the experimental and theoretical aspects of the process of cracking in drying soils. The results and conclusions were drawn from an exhaustive experimental campaign characterised by innovative multidisciplinary aspects incorporating Fracture Mechanics and classical Soil mechanics, aided with image analysis techniques. A detailed study of the previous works on the topic showed the absence of large scale fully monitored laboratory tests, while the existing studies were performed on already cracked soil, rather than focusing on the process of cracking. Also, the absence of a multidisciplinary approach was strongly felt. The soil used in the present work is a Barcelona silty clay, extensively studied previously for its Thermo-Hydro-Mechanical (THM) behaviour. Tensile strength and fracture toughness are two important fracture parameters that were not characterized previously and which were determined for the soil used in the investigation. The effect of moisture content on tensile strength and fracture toughness was established. The variation of tensile strength with the degree of saturation was theoretically explored considering the concept of unsaturated cohesion, while the concepts of activation energy and rate process theory were used to explain the relation between the fracture toughness and moisture content. Preliminary desiccation tests were conducted to gain basic knowledge of the cracking behaviour of Barcelona soil. Further experiments were conducted in a controlled laboratory environment with rectangular holding trays of similar geometry with five different surface areas. Crack initiation, temporal evolution of cracks resulting in the final crack pattern, meeting and bifurcation of cracks resulting in intersections, etc. were studied in detail. One of the objectives of the desiccation tests with similar geometry was to check the applicability of fracture mechanics. The results of the experiments showed existence of size-effect in soil cracking, thus justifying the use of fracture mechanics as a framework to model crack propagation. A novel equipment capable of simulating the different combinations of climatic conditions with monitoring by sensors and a digital camera was designed and constructed. This enivironmental chamber can hold large specimens up to 80 cm in diameter and 20 cm thick. The results from experiments carried out with the environmental chamber have provided new knowledge about the process and mechanisms of desiccation cracking. Some observations made during the experiments prompted more detailed experiments on some aspects of crack formation. The morphology of cracks, fissures, and the presence of spiral and ripple-like cracks on the bottom surface demanded to explore a different theoretical point of view into the mechanism of cracking in soils. The possibility of using a theory based on either classical or unsaturated soil mechanics is critically examined, with emphasis on the importance of surrounding ambient conditions and seasonal variations. Finally, detailed macro-morphology analysis of the crack patterns was performed. A theory was developed based on the morphology of crack patterns to explain the crack formation as a result of successive domain divisions or hierarchical nature of crack formation. To test the proposed theory, temporal evolution of cracks and crack pattern were studied in detail, showing the existence of hierarchy very clearly. An attempt has been made to explore a new way of analysing the cracking in soils as a process in transition from a disordered (at crack initiation) to a deterministic behaviour (at the end of drying resulting in a complex pattern). Finally, the possible mechanisms for the formation of cracks at the bottom of the specimens were examined, with emphasis on curling and synaeresis processes.