Diffusional anomaly and network dynamics in liquid silica

The present study applies the power spectral analysis technique to understand the diffusional anomaly in liquid silica, modeled using the Beest-Kramer-van Santen (BKS) potential. Molecular-dynamics simulations have been carried out to show that power spectrum of tagged particle potential energy of silica shows a regime with 1f(alpha) dependence on frequency f which is the characteristic signature of multiple time scale behaviour in networks. As demonstrated earlier in the case of water [J. Chem. Phys. 122, 104507 (2005)], the variations in the mobility associated with the diffusional anomaly are mirrored in the scaling exponent alpha associated with this multiple time scale behavior. Our results indicate that in the anomalous regime, as the local tetrahedral order decreases with temperature or pressure, the coupling of local modes to network reorganizations increases and so does the diffusivity. This symmetry-dependence of the vibrational couplings is responsible for the connection between the structural and diffusional anomalies.