Model testing of conepenetration in silt with numerical simulations

Abstract

Interpretation of results from CPTU testing in silty soils is very challenging and it is hard to obtain reliable design parameters. The main reason is that penetration in silts may occur under partially drained conditions. Interpretation methods for sand are based on drained behaviour while for clay undrained behaviour is assumed. For silts there is an intermediate drainage condition that should be better understood.

The present PhD work focuses on studying: a) the failure mechanism and b) the pore pressure generation/dissipation during CPTU in silty soils by combining experimental laboratory tests and numerical simulations. The experimental work is performed in model scale in laboratories. The work applies techniques like freezing after penetration, x-ray imaging using a medical scanner and x-ray micro computerized tomography (x-ray micro CT) combined with 3D-volumetric digital image correlation (3D-DIC). These techniques allow the direct observation and quantification of the failure mechanism around a cone tip that has penetrated a soil sample. The non-destructive technique of combining x-ray micro CT and 3D-DIC gives the most accurate results in terms of volumetric and shear strains. The samples used have low moisture content. Silt is known to dilate during static shearing to failure. In spite of this, the analysis shows a well-developed compaction zone below the cone tip with significant dilation zones adjacent to the probe, below and outside the compaction zone.

Laboratory CPTU tests with variable penetration rates are further performed on saturated samples. The samples used are identical in size and boundary conditions to the ones used for the x-ray micro CT tests. Measurements of pore pressure on the penetrating cone as well as in the silt sample reflect how pore pressure is linked to dilation, contraction and internal drainage. Measurements of cone penetration resistance and soil stresses in the sample are also carried out. The cone resistance and cone pore pressure vary with penetration rate. Different behaviouristic responses are identified for varying sample density: (1) for high density, a more dilative response where cone resistance increases and cone pore pressure decreases with an increase in penetration rate; (2) for low density, a somewhat contractive response where cone resistance decreases and cone pore pressure increases with an increase in penetration rate and (3) for medium density, the presence of both patterns of behaviour where increasing the penetration rate from very slow rate (drained behaviour) gives a reduction in resistance, as the rate grew and for further increase in rate towards very fast rate (approaching undrained conditions) the resistance increases again.

A procedure is developed and applied to study the micromechanical properties of miniature silt samples extracted from medium density specimens. The saturated silt specimens are penetrated by the CPTU, frozen and cut through its centre to prepare thin sections of the zones around the probe. Electron probe micro analysis (EPMA) enable quantification of porosity and dominating particle orientation in various locations around the tip and indicate the presence of both compaction and dilation zones like found from 3D-DIC. Micromechanical studies also suggest that for slow penetration rate, compaction dominates the failure process while for fast penetration rate, dilation dominates. At an intermediate rate, the results indicate the effect of partial drainage.

Finite element (FE) analyses are performed to shed further light on the processes governing the experimental results. An advanced soil model with dilatancy control is used in a simplified simulation procedure. The numerical simulations involve a simplified FE technique that accounts for large deformations to a certain extent and seems to capture several features of the observed behaviour. Comparisons are made between measurements and simulations of penetration force, soil stresses and pore pressure. A short discussion on possible recommendations for interpretation of CPTU test on silty soils in the field supports the application of the NTH-method for the interpretation of effective stress parameters.