Validation of microcrack models by experiments on real and replicated rocks

This paper presents a systematic approach to validate microcrack models by experiments. Although only experiments on rocks were done, similar approach should be applicable to concrete, ceramic, or other brittle materials. A brief review of microcrack models will be presented. Microscopic observations and ultrasonic wave measurements were done on natural and replicated rocks. The results from microscopic studies provide the necessary information to quantify the amount or degree of microcracking, in terms of a parameter called `crack density'; while the change of ultrasonic wave speeds was used to interpret the degree of microcracking. In particular, preliminary results on replicated rocks suggests that both `self-consistent method' and `non-interacting method' give reliable prediction if the crack density is less than 0.21; similar results on natural rocks also seems to suggest that both theories are good only for crack density up to 0.2. Nevertheless, more refinement of the testing technique is still required.