Tunneling depth is gradually increasing around the world due to the improvement of construction technology and the requirement of deep underground space. At depths where the high in-situ stresses might result hard rock failure with violent post-peak strength degradation. The characteristic of post-peak strength degradation may impact deep underground excavation stability. However, there is no a generally acceptable method in estimating the post-peak strength degradation up to now. For this, a serial of tri-axial compression tests for different hard rocks, including marble, gneiss, and granite, were carried out. The confining stresses from 0.1σc to 0.8σc were applied for the tri-axial compression test. It is found the post-peak strength degradation is strongly affected by confining stress condition. The evaluation of post-peak strength degradation should not neglect the effect of confining stress. According to the experiment data and the post-peak form of the Hoek-Brown failure criterion, the relationships between confining stress and strength loss parameter are established. The proposed relationships can be used to reasonably evaluate the post-peak strength degradation and further applied to assess the tunneling stability at deep depth.
Severe climate change is common around the world. Variety of damages to ground structures frequently takes place due to extreme climate impact, and therefore, utilization of underground space has become more and more important. Underground excavation depth exceeding 1000 m has increased gradually with improved construction technology and the requirement of deep underground space. At depths where the high in-situ stresses and the stress concentration near the excavation boundary might result brittle failure occurrence due to hard rock overstressed. The brittle failure of hard rock is different from the squeezing failure of weak rock, that is, relatively small deformation occurs prior to failure during loading, and under continued loading, the integrity of rock would be destroyed with degrading rock strength. Except for violent rock burst, the post-peak strength degradation may impact the stability of deep underground excavation. The evaluation of rock strength around an underground opening is a difficult task. Many researches have been proposed on the determination of peak strength of rock. However, only few attempts have been made to estimate the post-peak strength degradation of hard rock under high stress condition. Actually, there is no a generally acceptable method in estimating the post-peak strength degradation up to now. Experiences from case studies or results from material tests are the usually adopted method. For instance, Read and Chandler (1997) proposed using the constant values of s=0.16, mb=28, sr=0.01 and mr=1 to optimize the tunnel shape on granite in Canada, where s and mb are the Hoek-Brown peak strength parameters and the subscript "r" indicates residual values. The relations of sr=0.04s and mr=0.65mb were suggested to estimate the post-peak strength of jointed rock masses by Ribacchi (2000). Furthermore, the relations of sr=s and mr=0.1mb were presented to portray the post-peak strength of muscovite schist by Crowder et al. (2006), which is obtained from the back analysis of underground mining. The Mohr-Coulomb parameters of cr=0.1c and ϕr-=0.9ϕb, were adopted to simulate the post-peak strength of sandstone in a hydraulic tunnel by Kumar et al. (2008). These suggestions are essentially obtained from specific rock type or individual case. The appropriateness for different rock type or for different site needs to be re-examined.
