SCEC Award Number 12062 View PDF
Proposal Category Individual Proposal (Integration and Theory)
Proposal Title The Effect of Off-Fault Damage on the Propagation of an Earthquake Rupture
Investigator(s)
Name Organization
Charles Sammis University of Southern California
Other Participants Harsha S. Bhat
Aris J. Rosakis
SCEC Priorities 3c, 4b, 3e SCEC Groups FARM, Seismology, GMP
Report Due Date 03/15/2013 Date Report Submitted N/A
Project Abstract
 The primary objective of this research has been to identify the dynamic processes that produce fault zone structures observed in the field and to understand the effect of fault zone structure on the dynamics of an earthquake rupture. The approach has been a combination of experiments in Professor Ares Rosakis’ high-speed digital photography lab at Caltech and a theoretical expansion of the Ashby and Sammis (1990) quasi-static micro-mechanical damage mechanics to include the physics of high-speed fracture propagation. Experimental results include:
1) A rupture propagating in a fracture damaged medium travels at a velocity below that expected based solely on the associated reduction in the elastic moduli, even if the rupture does no new damage. The implication for fault mechanics models is that it is not sufficient to simply account for the reduction in moduli associated with fracture damage.
2) A rupture propagating along the edge of a damaged fault zone propagates at a higher velocity in the direction that puts the compressive lobe of the crack tip field in the damaged material. In some cases the propagation is unilateral in this direction and may even jump to supershear. This is the opposite direction from that expected based on the reduction in modulus and is observed to dominate the velocity effect.
Theoretical results include:
1) Expansion of the Ashby and Sammis damage mechanics to include the physics of high-speed crack propagation and incorporation of this new dynamic damage mechanics as a user defined rheology in the ABAQUS dynamic finite element code.
2) A verification of the dynamic damage mechanics by accurately prediction the failure envelope of marble over 14 orders of magnitude in loading rate.
3) Prediction of the damage pattern and elastic waves generated by an explosion in a “candy-glass” plate. This is a further test of the dynamic damage mechanics code in a geometry that is a bit simpler than a propagating rupture. We are currently working to generate dynamic fractures on a frictional interface in candy glass. We have switched from photoelastic polymers (Homalite and polycarbonate) to candy-glass because we were unable to generate new damage in the polymers by either stick-slip events or explosions. Explosions in candy-glass generated extensive fracture patterns. We have identified the difference by measuring the critical stress intensity factor in the candy glass which turned out to be only 0.04 MPa m1/2 as compared to 1 MPa m1/2 in the polymers.
Intellectual Merit A major objective of the SCEC research program is to model the source mechanics of an earthquake with enough physical reality to predict the resultant seismic waves in the near and far field. The micro-mechanical damage mechanics being developed and tested in this project is the most physically based approach to study the effects of the generation of on- and off-fault fracture damage during an earthquake.
Broader Impacts The dynamic damage mechanics being developed here has applications beyond the earthquake source. Many fracture problems in the geosciences involve very high loading rates (eg.underground nuclear explosions and meteorite impacts). It is also finding wide range of engineering applications. The collaboration with Professor Rosakis at Caltech has allowed us to use expensive and sophisticated instrumentation to pursue SCEC objectives. The Post Doctoral Student trained on the project (Dr. Harsha Bhat) has accepted a faculty position at IPGP in Paris.
Exemplary Figure Figure 3. Simulation of an earthquake rupture on the boundary of a damage fault zone. Note the asymmetry in damage and rupture velocity. (From Bhat, H. S., A. J. Rosakis, and C. G. Sammis (2012), A micromechanics based constitutive model for brittle failure at high strain rates, J. Appl. Mech., 79(3), 031016, doi:10.1115/1.4005897)