Two different relationships have been proposed by the authors, individually, for the calculation of the densities of glasses from their compositions. These relationships are here compared with regard to accuracy, composition limitations, relationships between their constants, etc. The Stevels equation is useful in supplying values (accurate to within about 2%) of the density of glasses having a high proportion of network formers (e.g., Si, B, and Al). Whether or not it is applicable, with different constants, when the content of network modifiers (e.g., Na, Ca, and Pb) is high, has not been determined. The Huggins relationship involves more empirical constants, including two which have different values, depending on which of four composition ranges is pertinent. It is considerably more accurate for well‐annealed glasses of accurately known composition. Difficulties arise in applying it to glasses having a high content of network formers other than silicon or to glasses having components for which accurate values of the necessary constant have not previously been deduced. It is interesting to note that the limiting conditions of the Stevels equation lead, at least for the sodium silicate system, to the same breaks of the volume‐concentration curves as found by Huggins. Comparison of the two relationships yields a better understanding of the reasons for the long‐realized fact that the presence of certain elements, such as Li, Be, and Ti, invalidates the Stevels equation.
|Journal of the American Ceramic Society
|Nummer van het tijdschrift
|Gepubliceerd - okt. 1954