Conduction-related polarization in glasses was analyzed, using an independent jump hypothesis, in terms of the relaxation time. This relaxation was based upon the assumption that each mobile ion relaxed independently of the other. The relaxation time at the peak temperature, as evaluated at a heating rate of 0.1K/s, had an average value of about 80s for the glasses which were considered. A similar treatment was applied to the glass transition, where weak geometrical interference between the fragments which formed mobile units after melting was assumed to provide sufficient space for structural disorder. Their behavior was also expected to be governed by the relaxation time. Melting and subsequent viscous flow of the fragments could be assumed to take place independently because, in a disordered material, geometrical interference between fragments could be weaker in some places than in others.

A.Doi: Physics and Chemistry of Glasses, 1994, 35[1], 34-7