Ground improvement using vacuum preloading together with prefabricated vertical drains

The use of vacuum preloading as an alternative to or in combination with fill preloading has become widely acceptable throughout the world. A number of successful case histories of vacuum preloading using different techniques have been reported in the literature. The vacuum pressure, which is applied to the soft soil through the vertical drains, results in an isotropic increase in effective stress within the soil mass, and therefore, can be applied with full intensity in a single stage. On the other hand an equivalent fill preload is applied in stages to avoid instability and bearing capacity failure. The additional stability achieved due to application of vacuum and associated consolidation also helps in speedy construction of fill preload and thus reduces the overall duration of any preloading operation. Despite the increasing use of vacuum as a preload and development of a number of innovative techniques for its application, the exact mechanism of vacuum preloading remains largely unknown. Field trials, laboratory studies, and actual case histories are used to support contradicting concepts about the soil behavior during vacuum or combined vacuum-fill preloading. There is no consensus among the engineers and researcher about the distribution of vacuum pressure along the vertical drains, rate and magnitude of induced settlements, interpretation of porewater pressures, nature of lateral displacements, and increase in undrained shear strength due to vacuum preloading. As a consequence, modifications to existing consolidation theories and separate design procedures have been proposed to incorporate the effects of vacuum pressure during the preloading operation. The present study aims at evaluating the performance of vacuum preloading as compared to that of fill preloading. Data from 40 case histories and 8 laboratory studies of vacuum consolidation were collected to study different aspects of vacuum preloading as compared to those of fill preloading. Time-dependent behavior of soft soil subjected to vacuum or vacuum-fill preload, distribution of vacuum pressure with depth and time, performance of vertical drains, lateral displacements, increase in undrained shear strength, and effectiveness of different techniques of applying vacuum in the field are examined carefully to explain the soil behavior during vacuum preloading operation. Applicability of existing consolidation theories to explain soil behavior during vacuum or vacuum-fill preloading was studied using the computer program ILLICON (based on the ILLICON theory of consolidation) by modeling the applied vacuum as an equivalent fill preload. The settlement analyses of 11 sections from 8 different case histories of vacuum, vacuum-fill, and fill preloading show that the applied vacuum can be modeled as an equivalent fill preload to predict the settlements under vacuum or vacuum-fill preload, i.e., principle of superposition can be used with confidence to predict or to back-analyze the field behavior during vacuum or vacuum-fill preloading. The porewater pressures due to a vacuum or vacuum-fill preload can also be predicted using principle of superposition by converting the positive porewater pressure response (as a result of assuming applied vacuum as an equivalent fill load) to a porewater pressure response due to vacuum by using a unique definition of excess porewater pressure and by considering the relative magnitudes of applied vacuum and fill preloads as well as the distribution of vacuum pressure with depth and time. The distribution of vacuum pressure which should be constant with depth under ideal conditions, may not develop to full intensity at different depths due to ‘leakage’ of vacuum from specific sublayers; therefore it is absolutely important to ascertain actual distribution of vacuum pressure with depth and time to carry out a meaningful settlement analysis. Hence, with correct interpretation of field data, the existing consolidation theories can be used as such without any modifications to fully explain and predict the soil behavior during vacuum or a combined vacuum-fill preloading operation. The study also shows that the increase in undrained shear strength due to vacuum, vacuum-fill, and fill preloads can be predicted using the available empirical correlations developed for fill preloading. The ratio of initial vane shear strength to preconsolidation pressure, suo(FV)/σ’p, which remains constant in the compression range, can be used to predict the increase in undrained shear strength at the end of preloading operation; however, it is important to consider the degree of consolidation achieved as a result of preloading and the distribution of vacuum pressure with depth and time to correctly access the consolidation pressure which is responsible for the increase in undrained shear strength. Lateral displacements due to vacuum preloading can be better explained in terms of consolidation behavior of soil; however, contribution of other soil parameters cannot be ignored. Among other factors, the distribution of vacuum pressure with depth and time appears to be a major factor governing the magnitude of lateral displacements. Analysis of lateral displacement data from different case histories of vacuum consolidation shows that the lateral displacements are maximum at the ground surface and reduce with depth. Moreover, the magnitude of lateral displacements (at the boundary of treatment area) is independent of the width of treatment area as well as the depth of penetration of vertical drains; therefore, no specific relation should be expected between these parameters. An empirical method to predict lateral displacements with depth and time is also proposed in the study. A simple procedure for design of preloading is also suggested in the present study. The proposed design procedure is based on ascertaining the increase in effective stress at a given time using the computer program ILLICON and predicting the increase in shear strength using the empirical correlation for undrained shear strength proposed by Terzaghi et al. (1996) together with a suitable method to compute factor of safety at a given time. Application of the proposed procedure to actual case histories shows that it can be used with confidence for design of preloading