ABSTRACT

Seven, 15 cm diameter, cores of gneissic granite, four containing induced fractures and three containing natural fractures all oriented normal to the core axis, were tested in an uniaxial compression mode over a range of 0 to 30 MPa. At given normal stress increments, over two or three loading and unloading cycles, the flowrates and changes in fracture aperture were measured. The induced fractures gave lower initial (at low stresses) and lower final flowrates (at maximum stress) than the natural fractures. Both types of fracture exhibited permanent fracture deformation between loading cycles as well as highly nonlinear, with distinct hystersis, loading and unloading flowrate-stress curves. A significant finding of this study is the breakdown of the cubic law for fracture flow in induced fractures subjected to normal stresses greater than 20 MPa, with the breakdown occuring at much lower stresses in natural fractures.

INTRODUCTION

With the growing need to isolate and contain toxic wastes, such as nuclear waste material, and the suggestion that this can be achieved by storing them in fractured crystalline or argillaceous rocks, it is essential that we understand how fluids and contaminants move, and the factors that control such movement, in fractured rocks. Any attempt at understanding the factors that control the flow of fluids through fractured rocks must be based on a clear understanding of the structural nature of the rock mass. The main flow paths an fractured crystalline and argillaceous rocks are joints, fracture zones and shear zones. In this paper we are concerned with flow through the 'joint' member of the fracture family but the general term 'fracture' will be used throughout the rest of this paper. Gale (1982a) has discussed a number of factors that control flow through single and multiple fractures, such as stress, temperature, roughness, fracture geometry, etc, but as yet the relative importance of each factor has not been assessed. The discussion here will be restricted to the effects of changes in normal stress on fracture permeability as shown by the results of a laboratory study of induced and natural fractures in 15 cm diameter gneissic granite cores.

EXPERIMENTAL PROCEDURES

Sample Collection and Preparation Both induced and natural fractures were collected from surface exposures of granitic gneiss near Chalk River, Ontario, (Lumbers, 1974). Blocks of rock were obtained from outcrops by drilling overlapping boreholes. These blocks were cut to size, and a 1.5 cm diameter center hole was drilled along the length of the sample. Natural fractures were collected by first installing a rock bolt in a 1.5 cm diameter hole drilled normal to the fracture plane and then overcoring with a 15 cm diameter thin walled core bit to provide a 30 cm long core with a natural fracture located about halfway along the core length and oriented perpendicular to the core axis. Final preparation of the cores followed that outlined by Gale and Raven (1982) and included either grinding the ends flat or adding sulphur caps to ensure end parallelism for testing purposes.

This content is only available via PDF.
You can access this article if you purchase or spend a download.