From Structure Analysis to Glass Technology

Prof. Dr. Reinhard Conradt

What is the relationship in principle between the chemical composition of a glass and its properties? This is a question that has exercised the minds of scientists and technologists since the beginnings of systematic glass research in the 19th century. Otto SCHOTT recognized at an early stage the need to approach the question on the basis of a very broadly structured scientific concept:

“A systematic study of the phenomena of glass melting comprising its whole inorganic nature has not yet been attempted; we lack, therefore, much in this area before we will be in a position to determine the reactions with certainty on the basis of fixed laws as is the case with aqueous solvents at normal temperature.”
Otto SCHOTT, 1880, quoted by W. Vogel, Glaschemie (Springer Verlag, Berlin 1992)

The nature of glass, its transparency, its homogeneity and its isotropy, suggest simplicity – yet the opposite is the case. With metallic materials, the issue has been largely settled. However glass, as a “special state of matter,” has eluded a more fundamental quantification again and again. In deducing the properties of glass from its composition, one has to collect experimental data, link it, evaluate it statistically and then interpolate or extrapolate into unknown territory.

Thermodynamics: a New Method

In recent years a new, fundamental strategy has emerged. Attempts to understand the properties of a glass largely on the basis of its structure have been abandoned. This route is one which, in spite of enormous research efforts, has thus far only resulted in explanatory, qualitative evidence. Instead thermodynamic methods have been enlisted and applied systematically to the situations encountered in multi-component glasses. The method is, of course, not as transparent as a structural interpretation, but it does supply quantitative evidence. The glass is described via its energy difference to a crystalline object (or reference system) with the same chemical composition.

Complex Sequences can be Analyzed

Tracking this reference system is simple for mono-component glasses (e.g. fused silica and cristobalite); it can also be done for glasses with many components using geochemical methods. In principle this has finally opened up the use of comprehensive thermodynamic databases and books of tables for glass technology. Now it is possible to computer-model quite a few of the properties of glass, such as chemical resistance, before carrying out the first trial melt and the first corrosion test. With further development of the process it can be expected that it will be possible to significantly reduce the amount of work based on trial and error, thus saving time and costs.

The new modeling strategy is, however, not limited to just forecasting the properties of glass. Complex reaction sequences, in which a glass or its melt is involved as part of the reaction, can now be subjected to quantitative treatment. Technologically relevant examples are the determination of the energy required for the fusion of glass batches or the evaporation of individual components in a melt resulting from the effect of a furnace atmosphere.