The measurement of soil suction
The development of soil suction measuring techniques has been slow relative to the rapid progress in theories for unsaturated soils. The indirect measurements of soil suction generally suffer from such problems as long equalization time, poor accuracy, hysteresis, and equipment deterioration. The direct insitu measurements of matric suction have, in general, been confined to less than 100 kPa. It is highly desirable to extend the range of direct measurements to cover matric suction values normally encountered in geotechnical and geo-environmental engineering. The tensile strength of water was used to extend the range of direct measurement of matric suction. A tensiometer-type suction probe was developed using the principle of the high tensile strength of water. High positive pressures pre-applied to the suction probe were found to significantly increase the tension that could be sustained by the water in the suction probe. Various factors that may affect the tensile strength of the water in the suction probe were identified. The mechanism of cavitation in the suction probe was investigated. It was found that for a particular pre-pressurization procedure, a given suction probe has an approximate "cavitation threshold" of tension. If a tension does not exceed the cavitation threshold, the tension can be maintained for a relatively long period of time (e.g., several days). The cavitation threshold appears to be dependent on the nature of the ceramic used in the suction probe. The highest sustainable tension observed in this research program was about 1250 kPa. Measurements of matric suction on different soils proved that the suction probe can be used to measure matric suctions up to at least 1000 kPa with an accuracy of about ±20 kPa. The principle of axis-translation was proved to be correct for negative pore-water pressures as low as at least $-$500 kPa. The suction probe appeared to be most suitable for measuring wet and clayey-type soils. Difficulties were encountered in measuring soils with a degree of saturation less than 40%. The developed suction probe provides a direct means by which the measurements of soil suction made by other conventional methods can be evaluated. Reasonable agreements were observed among the measurements using the suction probe, and those made by the filter paper method and the thermal conductivity sensor for soils with a relatively high degree of saturation. The discrepancies became more pronounced with decreasing degree of saturation. On the other hand, the null-pressure plate gave higher values of matric suction than the suction probe for soils near or above the optimum water content. However, the discrepancies became relatively smaller as the degree of saturation was reduced. In addition, efforts were also made to investigate the possibility of using the ability of a saturated clay-water system to transfer high tensions for the measurement of soil suction. The results of this research, however, were less rewarding. The use of a saturated clay for the measurement of soil suction does not appear to be a feasible approach because of the considerable hysteresis in the mechanical behavior of the clay. Nevertheless, it may be possible to use the water-absorbing properties of other materials, such as certain water-absorbing polymers, in the measurement of soil suction.