2:30 PM - 2:45 PM
[SCG44-10] Contact Mechanics of Alpha Quartz Asperities in Nanoscale:
Inelastic Behavior Deviated from Hertz and Elastoplastic Models
Keywords:α-quartz, Nanoscale asperity, Contact mechanics, Molecular dynamics
The contact behavior of asperities on fault surfaces plays a crucial role in influencing both the probability and magnitude of seismic events caused by tectonic loading. Therefore, understanding the contact mechanics of asperities is essential for unraveling the complexities of earthquake source physics. Currently, the mechanical behavior at centimeter scales and above can be well investigated in physical experiments and theoretical modelling. However, observations at micrometers and below are practically difficult and very rare, resulting in persistent challenges in accurate modelling of the multiscale contact and friction properties of rocks. Therefore, in this study, to gain insights into the contact behavior of asperities at micro scale, we employed the molecular dynamics method to explicitly simulate a series of nanoscale contact processes. We comprehensively observed the microscopic damage evolution of asperities and examined the contact force evolution by comparing various contact models, including elastic Hertz model and elastoplastic KE, ZMC, JG, and CEB models. These four models initially follow the Hertz model at the elastic contact stage but become different later in the elastoplastic or fully plastic stage.
Our simulation resembles the loading stage of macro-friction experiments but focuses on a pair of identical semi-spherical α-quartz asperities positioned at the centers of the upper and lower contact surfaces. Five groups of simulations were considered respectively with different asperity radii: 8a, 12a, 16a, 20a, and 24a (where a = 8.4903 Å). While the top block was set in motion at a constant velocity of 10 m/s in the vertical direction, the bottom block remained fixed. Periodic boundary conditions were applied in the horizontal directions. The normal contact force was measured by summing the force of top block in the vertical direction. The interference distance ω was defined as the loading distance after the contact of two asperities.
When the interference distances ω was small, the nanoscale contact force of all simulation results followed well the Hertz model, indicating an elastic stage. However, as ω increased, an elastic limit point appeared where the nanoscale contact force began to leave the Hertz model. Notably, at the elastic limit point, the normal contact force of asperities exhibited a power-law relation to the radius of asperities. In addition, none of the elastoplastic KE, ZMC, JG, and CEB models can well predict the initial yielding point of asperities, which generally occurred after a large ω at nanoscale. With further increase of ω, the damage evolution of alpha quartz developed and formed a fracture network, similar to the brittle breaking observed in rocks at room temperature. This behavior was quite different from most metallic materials, which typically transition into elastoplastic or fully plastic stage. Therefore, the contact force curves of all models have experienced an obvious and long weakening or declining stage, which were significantly different from those predicted by KE, ZMC, JG, and CEB models.
These results show that the current contact models have obvious limitations in the accuracy of the nanoscale contact modeling of asperities, which contribute significantly to the multiscale modelling of contact behavior and friction properties in rocks.
Our simulation resembles the loading stage of macro-friction experiments but focuses on a pair of identical semi-spherical α-quartz asperities positioned at the centers of the upper and lower contact surfaces. Five groups of simulations were considered respectively with different asperity radii: 8a, 12a, 16a, 20a, and 24a (where a = 8.4903 Å). While the top block was set in motion at a constant velocity of 10 m/s in the vertical direction, the bottom block remained fixed. Periodic boundary conditions were applied in the horizontal directions. The normal contact force was measured by summing the force of top block in the vertical direction. The interference distance ω was defined as the loading distance after the contact of two asperities.
When the interference distances ω was small, the nanoscale contact force of all simulation results followed well the Hertz model, indicating an elastic stage. However, as ω increased, an elastic limit point appeared where the nanoscale contact force began to leave the Hertz model. Notably, at the elastic limit point, the normal contact force of asperities exhibited a power-law relation to the radius of asperities. In addition, none of the elastoplastic KE, ZMC, JG, and CEB models can well predict the initial yielding point of asperities, which generally occurred after a large ω at nanoscale. With further increase of ω, the damage evolution of alpha quartz developed and formed a fracture network, similar to the brittle breaking observed in rocks at room temperature. This behavior was quite different from most metallic materials, which typically transition into elastoplastic or fully plastic stage. Therefore, the contact force curves of all models have experienced an obvious and long weakening or declining stage, which were significantly different from those predicted by KE, ZMC, JG, and CEB models.
These results show that the current contact models have obvious limitations in the accuracy of the nanoscale contact modeling of asperities, which contribute significantly to the multiscale modelling of contact behavior and friction properties in rocks.