Research effort around cracking is significant, but the focus is generally on the propagation of cracking. Research into the basic criterion for cracking was relatively limited, with the majority of papers supporting a total stress based tensile failure mechanism, but providing little evidence to support this. As a soil dries, a negative pore water pressure develops, affecting the effective stress experienced by the soil skeleton. The negative pore-water pressure had been largely ignored in experimental investigations, which is difficult to justify as effective stress-based analysis is a basic pillar of geotechnics. The only effective stress based experimental procedure was limited to heavily over consolidated samples with a relatively small maximum pore water pressure (»100kPa) using a non-direct testing method (Thusyanthan et al, 2007).
To facilitate the research a new experimental apparatus and testing methodology was developed to test specimens in direct uniaxial tension with the facility to monitor the suction (pore-water tension) using high-capacity tensiometers, developed at Strathclyde University, throughout the test to the point of specimen failure. This allowed for the analysis of the data from experiments in terms of effective stress for specimens across a significant range of negative pore water pressures (up to approximately 1500kPa of suction). The tests showed that the stress state at failure was in agreement with a shear failure envelope for specimens in both saturated and unsaturated states.
The programme of tensile stress experiments provided a number of difficulties that resulted in a number of additional research outcomes that were not expected. During testing in the saturated range failures were occurring at a lower tensile stress than was predicted and were not reaching the shear failure envelope. It was proposed that this was due to cavitation of naturally occurring small air bubbles within the pore water. As the suction increased during loading the cavities could expand, creating a localised weakness where there was no porewater pressure to hold the soil particles together. By de-airing the water prior to reconstituting the soil, the stress state at rupture was found to realign with the failure envelope. This provided the first experimental indication of cavitation affecting the failure behaviour of fine-grained soils.
It had also been speculated by Haigh et al (2013) that cavitation could affect the interpretation of plastic limit of soils, with cavitation causing cracking and fracture of soil threads before the true plastic limit moisture content was reached. In order to test this the same de-aired and non-de-aired soil were subjected to plastic limit tests. It was shown that the plastic limit moisture content was higher for non-de-aired soil, confirming the theory of Haigh et al, (2013). This was the first time to my knowledge that the effect of cavitation had been demonstrated experimentally.
My overall experience of my time doing a PhD was mixed, but generally positive, and I enjoyed working with my supervisor and fellow researchers. I was able to achieve the objectives set out at the start of the process, and additionally found some extra outcomes in doing so. I gained a significant amount of knowledge and experience working in a geotechnical research laboratory using advanced material testing techniques & equipment. There were lots of challenges and long periods of frustration where results were not as expected and significant time passed without progress being made despite of significant effort.
Haigh, S.K., Vardenaga, P.J. and Bolton, M.D. (2013). The plastic limit of clays. Géotechnique. 63, No.6, 435-440.
Thusyanthan, N.I., Take, W.A., Madabhushi, S.P.G. and M. D. Bolton (2007). Crack initiation in clay observed in beam bending. Géotechnique 57, No.7, 581-594.