Wave Propagation and Mechanical Properties of Weak Sands

Jon E. Olson

Abstract

The objective of the proposed study is to develop methodology for predicting strength and failure of particulate material, especially weakly cemented sand and gravel, from small-strain stiffness as measured with the velocity of stress waves. The project will be especially concerned with phenomena such as soil liquefaction in earthquakes and sand production in petroleum wells.

 The objective will be approached through the development of an improved, fundamental understanding of how particle cementation affects both the strength and the stiffness of cemented granular material. The program of studies will have two concerns:

  • How can the degree and nature of particle cement be determined from properties such as wave velocities and static moduli?
  • How can this characterization of cement be related to such engineering behavior as aggregate strength, liquefaction potential, and sand production?

A concurrent numerical and experimental investigation is planned. Experiments on cylindrical test specimens will be conducted in a triaxial test apparatus, while the numerical work will involve Distinct Element Method (DEM) simulations using commercially available software. The parallel pursuit of laboratory tests and numerical simulations is a powerful approach to developing fundamental insight into the behavior of geologic materials. Experimental data permit immediate calibration and verification of analytic results, while numerical simulations can be helpful in identifying underlying mechanisms responsible for observed behavior and in designing useful laboratory tests. Two- and three-dimensional DEM simulations will be used to compute bulk material strength and elastic wave velocities as a function of particle size distributions, intergranular cement and material porosity. Behavior observed from these simulations will be used to devise appropriate experimental tests and test conditions that are most likely to produce meaningful data. Samples for experimental testing will include naturally occurring granular material of varying degrees of cementation and strength, as well as artificially created samples. The artificial samples should allow for more controlled sensitivity analysis of particular rock properties and how they affect mechanical behavior.


Contact:

Jon E. Olson
Center for Petroleum and Geosystems Engineering
1 University Station C0304
The University of Texas at Austin
Austin, Texas 78712-0228
Phone: (512) 471-7375 FAX: (512) 471-9605
Email: jolson@mail.utexas.edu