The conditions for initial tensile cracking around a rounded blunt and a sharp notch and subsequent propagation were investigated by means of wedge splitting tests on structurally isotropic granite. In addition, the crack initiation and propagation on anisotropic gneiss specimens with a blunt notch were investigated in two different material directions. The wedge splitting test specimens which has a straight notch is normally used to determine mode I fracture properties for concrete but not for rock materials. The use of a straight notch in the specimens instead of a chevron type of notch, which is commonly used for fracture mechanics tests, was found to be well suited for the actual type of investigation.
The development of fractures from the notch into the specimens was monitored by measuring the deformation field on the specimen surface by digital image correlation (DIC) and by the crack mouth opening displacement. The fracture patterns were different in the granite and the gneiss. The DIC illustratively showed how cracks branched along the main crack path in the granite whereas no branching occurred in the gneiss material. This could also be verified by studying thin sections taken along the fracture path. It was seen that the granite behaved more ductile than the gneiss.
A finite element model of the wedge splitting test was made where the crack propagation was modelled using a cohesive-zone model and calibrated using the results from the experiments. The fracture energies were computed and it was found that the energies were about 50–100 percent higher in the granite than in the gneiss. The tensile stress cracking initiation in granite for the blunt and the sharp notch was slightly different. The gneiss specimens were all with a blunt notch and a comparison of the two notch types could not be made in this case.
Subdividing large primary blocks to smaller secondary blocks and slabs is part of the manufacturing chain in a natural stone quarry. This is often done by drilling holes in a row in the primary block, along the line where the block is going to be divided, where after a splitting force is achieved along this line by driving in mechanical wedges into the holes using a sledgehammer. An alternative splitting method, where the heavy labor using a sledgehammer is replaced by a splitting system based on a flattened hydraulic hose that is pressurized to achieve a splitting force, was partly developed and scrutinized within the EU project HYDRASPLIT (Grant Agreement No. FP7-SME-2012–1-315530). The required force to split blocks in a quarry is needed to be known in order to evaluate the design and performance of the alternative splitting method. This was investigated in an experimental program and a finite element model was developed to simulate the experiments. Results from the project are presented in this paper.
Wedge splitting test (WST), see e.g. Linsbauer and Tschegg (1986) and Brühwiler and Wittman (1990), is a method that was developed to determine the fracture energy for mode I fracturing of concrete, but has been very little used in the rock mechanics community. The specimens are rectangular blocks with a sawn straight slot. The sides of the slot is forced apart using a mechanical system which uses wedges. This type of test was found suitable since customized specimens having a borehole at the end of the slot as well as a sharp sawn notch could be tested. Moreover, the experimental set-up provides a stiff mechanical system which makes it possible to follow the crack propagation in semi-brittle materials. The standard laboratory methods suggested by the International Society for Rock Materials and Rock Engineering (ISRM) to determine mode I fracture mechanics properties all uses a chevron type of notch (e.g. Ouchterlony, 1988; Fowell, 1995; Kuruppu et al., 2014).
