Size effect on the mechanical behavior of intact rock

Determination of the rock mass strength is a challenge for geotechnical engineers. Some part of this problem is related to complex interaction of rock mass components, intact rock and discontinuities, and the other part is related to difficulty in determination of the mechanical properties of intact rock and discontinuities. The common approach to determine the mechanical properties of rocks is undertaking laboratory experiments and extrapolate the insitu properties based on these laboratory experiments. This extrapolation, or in the other word size effect, has been remained as a challenge for practical rock engineers for decades. Size effect studies can be divided into two groups of one dimensional scale effect, in which the diameter of samples is fixed and the length of samples are changed, and three dimensional scale effect, in which the ratio of length to diameter, called slenderness ratio, is fixed and the diameter is changed. Review of previous studies on the effect of slenderness ratio on the strength of intact rock shows that in some studies as the slenderness increases up to specific value, the strength decreases while in other studies the slenderness ratio has no significant effect on the strength. Results of previous studies on the effect of sample diameter on the strength of intact rock are also inconclusive. Some studies show that as the diameter of sample increases, the strength decreases while other studies show the increase of strength up to specific diameter or no scale effect. This paper aims to study both one dimensional and three dimensional size effect on intact rock properties. To this end, around 120 sandstone samples with diameter between 19 to 145 mm and slenderness ratio of 1 to 4 were prepared and uniaxial compression tests were carried out on these samples. Uniaxial compression tests were carried out using MTS 815. Axial and circumferential extensometers were used to measure deformation of samples during experiments. Results of this study show that the increase of diameter up to 50 mm resulted in the increase of uniaxial compressive strength and no scale effect was observed for diameters greater than 50 mm. This trend was observed for all slenderness ratios. No scale effect was observed on the elastic modulus and the Poisson’s ratio. Therefore, the minimum diameter of around 50 mm, which is suggested by ISRM and ASTM standards, is the size that scale has no effect on the results of uniaxial compression tests. It was found that the increase of slenderness ratio up to 2.5 results in the decrease of strength and no significant effect was observed for greater slenderness ratios. To find out the reason of observed behavior, failure mechanisms of samples at different slenderness ratios were studied. Four types of failure mode were observed in the experiments; generation of a single shear plane, multiple shear planes, shear plane with some cracks in the center of specimens and tensile cracks parallel to the loading direction. It was found that at low slenderness ratios, samples are failed by propagation of tensile cracks which results in higher strength compared to the other failure modes. Therefore, the higher strength of samples with slenderness ratio of 1 is related to this failure mechanism.