The origin of sound damping in amorphous solids: Defects and beyond

Comprehending sound damping is integral to understanding the anomalous low temperature properties of glasses. After decades of theoretical and experimental studies, Rayleigh scattering scaling of the sound attenuation coefficient with frequency Γ∼ωd+1, became generally accepted when quantum and finite temperature effects can be neglected. Rayleigh scaling invokes a picture of scattering from defects. However, it is unclear how to define glass defects, or even if defects are necessary for Rayleigh scaling. Here we determine a particle level contribution to sound damping in the Rayleigh scaling regime. We find that there are areas in the glass that contribute more to sound damping than other areas over a range of frequencies, which allows us to define defects. We show that over a range of glass stability, sound damping scales linearly with the fraction of particles in the defects. However, sound is still attenuated in ultra-stable glasses where no defects are identified. We show that soun..., These data were computed from simulations of a model glass forming liquid. The simulated systems were quenched to their nearest potential energy minimum, an inherent structure. We calculated the eigenvalues and eigenvectors of the Hessian matrix that was calculated at the inherent structure. These eigenvalues and eigenvectors were used to calculate sound attenuation using a microscopic theory. The theory was compared to sound attenuation calculated using harmonic simulations of the same glasses. 
The inherent structures were found using LAMMPS simulation code. The eigenvalues and eigenvectors were determined using ARPACK. The calculation of the Hessian matrix and sound attenuation from the microscopic theory was determined from in in-house code. The simulations used to determine sound attenuation were performed using in-house code., , # Data for The Origin of Sound Damping in Amorphous Solids: Defects and Beyond

[https://doi.org/10.5061/dryad.cz8w9gjd8](https://doi.org/10.5061/dryad.cz8w9gjd8)

## Description of the data and file structure

All the data was created using simulation and C/C++ code. SWAP Monte Carlo simulations were performed to equilibrate liquids at a parent temperature. The liquids were then quenched to their nearest potential energy minimum, i.e. an inherent structure, using LAMMPS. The eigenvalues and eigenvectors of the Hessian matrix calculated at the inherent structures were calculated using ARPACK. These eigenvalues and eigenvectors were used to calculate sound damping using a microscopic theory. The microscopic theory is compared to harmonic simulations of sound damping. 

### Files and variables

#### File: B3pw\.txt

**Description:** The plane wave calculation of sound damping as a function of parent temperature. 

**Variables**

* T_p: The parent temperature
* B3pw: The Rayleigh scaling c...,