Binary alkali silicate glasses were synthesized as beads by aerodynamic levitation coupled to laser heating to test the applicability of the method to this compositional range. While bubble-free lithium disilicate beads could be easily obtained, sodium and potassium silicates proved more challenging to melt without significant alkali evaporation: the final samples contained bubbles and exhibited compositional drifts compared to the starting stoichiometry, especially at high SiO 2 content. The risk of volatilization from the melts was evaluated empirically: the volatility of each oxide component scaled to the ratio between its melting temperature T m and the T m of the target composition (r evap ), while the difference between such ratios (Δ evap ) provided a qualitative estimation of the risk of differential evaporation. The formulated approach enables to evaluate the suitability of aerodynamic levitation synthesis for a given target glass composition: while low melting temperature and low liquidus viscosity (η < 10 0 Pa s) represent the primary optimal conditions, more viscous materials can still be prepared without major compositional drifts using a more careful melting procedure, especially if r evap and Δ evap are minimized.

No abstract