Particle cluster routines were tested using PFC software. The numerical specimens matched the mechanical properties of Westerly granite. The results from PFC2D and PFC3D simulations were compared against the actual laboratory data. After the mechanical tests the PFC3D cubic specimens were heated up to 450 °C. During the heating cracking, AE activity and temperature evolutions were monitored. P wave velocity measurements were conducted for each numerical specimen. The responses were compared against similar thermal laboratory data. The results showed similar P wave velocity decrease due the heating up to temperature of about 250°C.
The paper describes steps taken to implement and test enhanced particle clustering routines in PFC (Itasca Consulting Group, Inc., 2005). The study relates to the underground storage of high-level nuclear waste and the concept of placing fuel canisters in underground cavities. The canisters produce heat due to fission. The thermal changes in rock could potentially lead to the opening of existing cracks and the initiation of new cracking. Such damage may have serious implications such as increasing the permeability of the rock designed to function as a barrier to groundwater contamination for example. Measured macroscopic properties of rock tend to change with changing temperature. (Mahmutoglu 1998, Jackson et al. 1999). Elastic velocity measurement is a way to monitor some of the change, for example, David et al. (1999) used velocity techniques to study La Peyratte granite. Davidge (1981) and Homand-Etienne and Houpert (1989) among others studied thermal cracking at the grain scale and concluded that it is due to the mismatch in the thermal expansion coefficient between adjacent minerals and produces intergranular cracking. The work presented in the paper is part of the research to better understand mechanical damage induced by thermal loading. The numerical results are compared against results from thermal laboratory experiments onWesterly granite. The presented thermal modeling builds on the PFC2D thermal numerical experiments reported by Wanne and Young (2006).
Commercially available PFC was used in the particle cluster development and thermal simulations. The numerical approach models the movement and interaction of circular/ spherical particles by the distinct element method. The method represents solid rock by an assembly of particles joined by breakable bonds.The damage occurs by bond breakages, thus the material can evolve from solid to granular. PFC has been used to simulate hard rock in several studies. A thorough description of the method is given in (Potyondy & Cundall 2004).
PFC uses an explicit time-marching calculation scheme to simulate material behavior. This allows dynamic simulations to be performed in which seismic waves propagate across material at a speed that depends on the material properties. The approach permits realistic acoustic emission simulations (Hazzard & Young 2002). The numerical seismic monitoring technique has been used in studies reported by Young et al (2000) and Hazzard andYoung (2004). Seismic velocities of the numerical specimens were measured by propagating pressure waves through them.
