ABSTRACT
Cone penetration testing (CPT) is a standard method for evaluation of soil strength for construction purposes, in which a rod with a conical tip is inserted into the ground for measurement of resistive forces, pore pressure, etc. Classic theories in soil mechanics such as the bearing capacity and cavity expansion theories offer possible mechanisms by which the motion of soil particles exert forces on the penetrometer (Yu & Mitchell 1998; van den Berg 1994). With sufficient number of fitting parameters these theories offer reasonable predictability of soil strength. Microscopic validation of these theories require, in addition to information obtained by the standard penetrometer, the following: 1) soil material which allows imaging of the interior of the bulk, 2) imaging techniques with subparticle resolution, 3) computer algorithms to track the motion of individual particles. Past attempts by others to directly image the soil movement during CPT have been limited to a quasi 3D geometry in which a penetrometer in the shape of a half cylinder, pressed against a glass side window, was lowered into a real soil sample as illustrated in van den Berg (1994). Since the introduction of transparent soil simulant, new imaging techniques have been tried with much success. For example, vertical cross sections of transparent soil illuminated by a laser sheet have been analyzed by digital image correlation (Sadek et al. 2003). Difficulties in imaging the motion of particles through volume scans, however, have yet to be overcome. Certain anisotropies in the displacement field of particles, for example, can be observed only when one has data from a full volume scan. We present a new imaging technique using confocal and two photon laser scanning microscopy
with enough resolution to make possible the full 3D analysis of particle motion during CPT.
