Remarks Concerning the Formation and Crystallization of Vitreous Media

Abstract

The various characteristics of the liquid‐glass transition are examined in an attempt to establish the hypothesis that liquid or vitreous properties depend upon the relation between the thermal relaxation time and the time of measurement, and not upon any discontinuous allotropic change at a fixed temperature. In the course of this survey it is found that the relaxation time involved in the liquid‐glass transition for the heat capacity of glucose must be far less than that for the molecules, and is therefore probably that characterizing rotations of groups within the molecules, or certain modes of vibration of atoms or groups. No need is found from any of the measurements examined for the postulation of a ``vitreous state of aggregation,'' since all the properties of glasses appear to be those of undercooled liquids of high viscosity. Since the viscosity of laboratory glass is too high to permit its devitrification within a reasonable time, a mechanism involving surface solution is suggested by means of which observed phenomena may be explained.