The boson peak in silicate glasses: insight from molecular dynamics

In the low-frequency regime, approximate to 1 THz, glasses show an anomalous excess in their vibrational density of states called the boson peak (BP). The origin of BP has been a subject of debate since its first discovery a few decades ago. Although BP has been the focus of numerous studies, no conclusive answers have been found about its origins, which remained elusive to date. Here, we present results based on molecular dynamics of several binary and ternary silicate glasses with different network intermediates and modifier oxides. The vibrational density of states and the BP are reported for all the studied glasses. Their correlation with the elastic constant C44, structural, and dynamical properties are extensively discussed in terms of Voronoi atomic volume and the vibrational mean square displacement of Q4 species specifically. We also question the classical classification of alkali oxides as modifiers, and we suggest that Li2O plays the role of pseudo-intermediate oxide in lithium silicate glasses. This claim is supported by the effect of Li on various vibrational modes, and this effect differs from the other alkali metals. Furthermore, we demonstrate a correlation between the BP intensities and both the Voronoi volume of the Q4 and Q3 units and vibrational mean square displacements. The correlation between the boson peak intensities, the structural and elastic properties of silicate glasses (ternary and binary systems) are discussed in addition to an anomalous behavior of lithium compared to the other alkali oxide modifiers.

Automated summary

In this study, molecular dynamics simulations were used to investigate binary and ternary silicate glasses with different network intermediates and modifier oxides. The vibrational density of states and boson peak were analyzed for all glasses, and their correlation with elastic constants, structural properties, and dynamical properties were discussed. The classification of alkali oxides as modifiers was questioned, and it was suggested that Li2O acts as a pseudo-intermediate oxide in lithium silicate glasses. This claim was supported by the effect of Li on various vibrational modes. The study also found a correlation between boson peak intensities and the Voronoi volume of specific species and vibrational mean square displacements.