Structural Origin of Thermal Transport in Industrial Borosilicate Glasses

ABSTRACT The thermal conductivity of oxide glasses is a key parameter for applications ranging from thermal insulation to high‐power optical components; however, identifying the length scale that limits vibrational heat transport in disordered networks remains an open challenge. Here, we combine experimental measurements with molecular dynamics (MD) simulations to examine vibrational transport in three industrial borosilicate glasses: Pyrex (PYX), a modifier‐rich borosilicate (BORO), and a wool glass (EGL). Density, heat capacity, and elastic moduli were measured experimentally, while MD simulations employing the SHIK potential and a quasi‐harmonic Green–Kubo (QHGK) formalism were used to evaluate vibrational spectra and thermal transport properties. When radial distribution functions are expressed in scaled coordinates , all three compositions collapse onto a common curve near , suggesting a common characteristic length scale governing vibrational transport within this glass family. A medium‐range correlation length extracted from the position of the first sharp diffraction peak in the static structure factor, , is found to be comparable in magnitude to the vibrational mean free path inferred from thermal transport analysis across all compositions. PYX exhibits the highest thermal conductivity (1.35 ) and the longest inferred coherence length (0.64 ), consistent with its more strongly connected network topology and enhanced medium‐range structural connectivity, whereas modifier‐rich networks show shorter coherence lengths and suppressed transport. Together, these results clarify how a finite vibrational coherence length, reflected in both real‐space pair correlations and diffraction‐derived metrics, constrains thermal conductivity in multicomponent oxide glasses.