loading page

Design of sand-based, 3D-printed analogue faults with controlled frictional properties
  • Philipp Braun,
  • Georgios Tzortzopoulos,
  • Ioannis Stefanou
Philipp Braun
Ecole Centrale de Nantes

Corresponding Author:[email protected]

Author Profile
Georgios Tzortzopoulos
Ecole Centrale de Nantes
Author Profile
Ioannis Stefanou
Ecole Centrale de Nantes
Author Profile

Abstract

Laboratory experiments with surrogate materials play an important role in fault mechanics. They allow improving the current state of knowledge by testing various scientific hypotheses in a repeatable and controlled way. Central in these experiments is the selection of appropriate analogue, rock-like materials. Here we investigated the frictional properties of sand-based, 3D-printed materials. Pursuing further recent experimental works, we performed detailed uniaxial compression tests, direct shear and inclined plane tests in order to determine a) the main bulk mechanical parameters of this new analogue material, b) its viscous behavior, c) its frictional properties and d) the influence of some printing parameters. Complete stress-strain / apparent friction-displacement curves were presented including the post-peak, softening behavior, which is a key factor in earthquake instability. Going a step further, we printed rock-like interfaces of custom frictional properties. Based on a simple analytical model, we designed the a) maximum, minimum and residual apparent frictional properties, b) characteristic slip distance (d_c), c) evolution of the friction coefficient with slip and d) dilatancy of the printed interfaces. This model was experimentally validated using interfaces following a sinusoidal pattern, which led to an oscillating evolution of the apparent friction coefficient with slip. This could be used for simulating the periodical rupture and healing of fault sections. Additionally, our tests showed the creation of a gouge-like layer due to granular debonding during sliding, whose properties were quantified. The experimental results and the methodology presented make it possible to design new surrogate laboratory experiments for fault mechanics and geomechanics.