Attenuation of waves in glasses better understood through computer simulations
DOI: 10.1063/10.0003616
Attenuation of waves in glasses better understood through computer simulations lead image
It’s well known that long-wavelength elastic waves are attenuated in glasses, due to glasses’ intrinsic structural disorder. However, the physical mechanism that controls wave attenuation rates remains highly debated.
Kapteijns et al. took a detailed look at the issue using extensive computer simulations.
They carefully tuned the strength of the mechanical noise featured by model glasses. This allowed them to validate Fluctuating Elasticity Theory (FET), one of the leading theoretical frameworks, which predicts that wave attenuation rates are controlled by the spatial fluctuations of a glass’s coarse-grained elastic stiffness fields.
Instead of considering the spatial statistics of elastic stiffnesses, the authors studied their sample-to-sample statistics, which were found to be anomalous. They argued this anomaly is expected to follow a universal form, independent of the type of glass considered. This universality is explained to emerge from the presence of low-energy, localized excitations that inevitably exist in any structural glass made by cooling a liquid.
“This is a conceptual advancement in understanding how glasses behave,” said author Edan Lerner. “I hope that the support our results provide for FET will incentivize further work toward understanding the correct way to assess the spatial statistics of coarse-grained elastic stiffness fields.”
While their results show evidence that substituting spatial statistics of coarse-grained elastic stiffnesses with sample-to-sample statistics is valid, they have not yet fully proven it. However, they hope to prove such equivalency in future work.
Source: “Elastic moduli fluctuations predict wave attenuation rates in glasses,” by Geert Kapteijns, David Richard, Eran Bouchbinder, and Edan Lerner, Journal of Chemical Physics (2021). The article can be accessed at https://doi.org/10.1063/5.0038710