Granular Matter, February 2015, Volume 17, Issue 1, pp 1-20.
Yahya Nazhat, David Airey.
School of Civil Engineering, University of Sydney, Sydney, Australia.
The paper reports new observations of strain localisation following dynamic impact. Two dimensional physical model tests have been performed to investigate the response of dry granular soils to the impact of a free falling steel plate. The tests have been performed to study the mechanics of the densification of soils during dynamic compaction, which is a widely used process to improve the performance of deep soil deposits by repeated dropping of large weights onto the ground surface. High speed photography and digital image correlation techniques have enabled the deformation patterns, soil strains and strain localisations to be observed. Tests have been performed on sand with a range of densities and a sand–silt mixture. The results have shown that the soil deformations are comprised of a conventional bearing capacity mechanism at the surface and a series of compaction bands that propagate downwards beneath the impacting plate. Similar patterns are observed in all tests, however as the compressibility of the soil increases the contribution from the bearing capacity mechanism decreases.
Despite the wide use of dynamic compaction as ground improvement technique in granular soil sites, there are no established design procedures given that there is no clear understanding of the unseen kinematic mechanism of the densification process taking place beneath the ground surface. It has been thought that dynamic compaction increases relative densities in a bulb of soil the shape of a semi prolate-spherical with the effect of compaction reducing as one moves away from the impact footprint.
To the authors’ knowledge, it is the first comprehensive body of research in the field of dynamic compaction that has investigated the fundamental physics of the dynamic compaction process and, notably, has included a study of the micromechanics of particle movement. This research work has investigated the kinematics occurring during lab-based dynamic compaction tests using high-speed photography and image correlation techniques. This has enabled the displacement and strain fields to be determined from the digital images. It is very original and innovative, and has led to significant new findings and data which should be of use and interest to the geotechnical community at large, particularly designers of ground improvement by dynamic compaction.
The results reveal a distinctive mechanism that should affect the direction of how we understand the kinematics of the dynamic compaction process, towards better design, and it is likely to pose a significant challenge to theoreticians and modellers for some time to come. The finding that the resulting compaction does not vary continuously through the soil with distance from the point of impact (as one might expect from classical mechanics), but rather is the result of the progression of discrete compaction shock bands, is highly significant.
The main output of this research calls into question what has been long believed about ground improvement by dynamic compaction. The results also suggest that the degree of densification due to dynamic compaction can not be simply quantified by the common practises of before and after subsurface field testing and surface measurements. This is likely to change the way we think about densifying granular soils.