Laboratory results from triaxial deformation tests on sandstone are presented where changes in the elastic P-wave and S-wave velocity have been measured and compared, and in addition the rate and amplitude distribution of acoustic emission events has been recorded. The experiments were conducted on 15 mm diameter specimens of Darley Dale sandstone, at a nominal strain rate of 10–5/5, room temperature and confining pressures up to 200 MPa. The results show that during the brittle fracture of sandstone both P-wave and S-wave velocity initially increase during closure of pre-existing cracks and pores, but then decrease during dilatant microcracking. The acoustic emission event rate is initially very low during quasi-elastic deformation, but then increases rapidly up to failure. The seismic b-value, which characterizes the amplitude distribution of acoustic events, decreases prior to brittle failure.
The brittle failure of rocks is strongly dependent on both shear and normal stresses, pore pressure, temperature and strain rate (at high homologous temperatures). Rock deformation is usually accompanied by acoustic emissions which occur due to stress relaxation resulting from crack growth, dislocation motion or twinning. In addition, changes in elastic wave velocity in rocks occur because of changes in effective moduli and density caused initially by pore closure and then by microcracking during deformation. By measuring the acoustic emission event rate and amplitude distribution (characterized by the seismic b-value) and the changes in elastic wave velocity the course of rock deformation maybe followed. A number of laboratory studies have been conducted on acoustic emission activity during the brittle fracture of rock (see for instance Scholz, 1968; Meredith and Atkinson, 1983; Fonseka et al., 1985; Jones, 1988). These investigations have studied the acoustic emission event rate, amplitude distribution, frequency component, changes in b-value and source location and characterization. Laboratory studies of changes in elastic wave velocity (both compressional and shear wave velocity) during deformation have been conducted by Nur and Simmons, 1969; Masuda et al., 1987; and Jones, 1988. The present programme of work in our laboratory however aims to measure both the changes in elastic wave velocity (both compressional and shear) and the acoustic emission activity simultaneously during deformation under triaxial stresses, at non-ambient temperatures and under conditions of pore pressure control. The apparatus used in this laboratory investigation is a high pressure triaxial cell of conventional design, which is described in an accompanying paper (Murrell et al., 1989). The conventional triaxial cell provides probably the simplest and most commonly used method for achieving a combined stress state in a laboratory, by superimposing a uniaxial stress on a hydrostatic pressure. It has the advantage that it can be readily adapted to undertake complex physical property measurements (such as wave velocity changes and monitoring acoustic emissions) at high pressures, high temperatures and under pore pressure control. The development of the experimental techniques to measure wave velocity and acoustic emissions under triaxial stresses in our laboratory was carried out by Jones (1988). In this paper some initial results characterizing rock failure are presented.
