On Dec. 1, 2022, Prof. Jay Fineberg from The Racah Institute of Physics, The Hebrew University of Jerusalem was invited to the 92^nd Science Lecture of College of Science, SUSTech. He gave a lecture themed “The Fundamental Physics of the Onset of Frictional Motion: How does Friction Start?”, which was chaired by Asst. Prof. Shiqing Xu of the Department of Earth and Space Sciences, SUSTech. More than 100 audience participated in this lecture online. 

In this lecture, Prof. Fineberg not only presented a review about the history of understanding friction, but also introduced the achievements of his research group in probing friction over the past twenty years. Especially, Prof. Fineberg pointed out that the traditional view based on force balance could no longer characterize frictional motion satisfactorily; rather, the relation of energy balance based on fracture mechanics could more accurately describe when frictional motion will start, how it will evolve, and where it will stop.

First of all, Prof. Fineberg guided the audience through several concepts, such as the apparent contact area and the real contact area. On the basis of these concepts, he elaborated the importance of studying frictional motion from a microscopic point of view. Next, he combined transparent materials, reflection and transmission of light, and high-speed-camera photography, to show the mechanism of rupture propagation behind frictional motion. In addition, he utilized the strain data recorded near the fault to uncover the similarity between frictional motion and material fracture. Furthermore, he used several examples to illustrate the power of fracture mechanics in predicting the trajectory of the rupture front during frictional motion, including the spatiotemporal evolution of the rupture front and when and where it will stop. Based on these results, Prof. Fineberg confirmed the applicability of fracture mechanics for interpreting the physics behind frictional motion. Finally, Prof. Fineberg shared the recent achievement of his research group, that is, using artificially introduced barrier to study how rupture starts to nucleate and evolves into the spontaneous propagation stage.  

During the interaction part, the audience asked questions about the setup of friction experiments, the comparison between transparent materials and rocks and so forth. Prof. Fineberg answered those questions one by one.

Q: How to understand the boundary effects of the experimental sample?

A: By far there are theories and numerical simulation tools for evaluating the boundary effects of the sample, including the effects of free surface and finite thickness. As long as the boundary-reflected waves do not reach the fault, then the frictional motion along the fault can still be analyzed by fracture mechanics for an infinite space.

Q: Under a great amount of background normal stress, would there be any difference between the apparent contact area and the real contact area?

A: Experience tells that such hypothesis may be correct, because when the background normal stress is very large, for example approaching the failure strength of the material, the entire contact area is expected to fail and “flow”. As for the actual situation, the current experiments show that even under a background normal stress of several megapascals, the real contact area is still significantly smaller than the apparent contact area.

Q: Why did rupture nucleation start near the edge in the most recent experiments?

A: The edge happened to be the weakest point, so that rupture chose to nucleate there.

Q: The environment for natural earthquakes is very complex while the setting for laboratory experiments is quite simple, how to reconcile the two?

A: In the lab, even if we try our best to homogenize the loading condition, the actual stress distribution along the fault is still very inhomogeneous.

Q: The currently used samples in the lab are transparent materials, what if you change the samples to rocks?

A: If one uses rocks, one would expect to observe fault gouge near the fault, which may dramatically change the slip behavior of the fault, because fault gouge can serve as a third body material and can lubricate the fault.

Q: Recently some researchers suggest that fracture mechanics can also be applied to explain slow earthquakes, any comments on this?

A: Existing experiments show that, near or under the theoretically predicted nucleation length, rupture propagation speeds are very slow, which may satisfy the criteria for slow earthquakes. On the other hand, we still need extra mechanisms to develop the nucleation zone in the first place. The mechanisms inside and outside the nucleation zone are not necessarily connected.