Spectrophotometric Study of the Colours Given by Iron in Alkali–Lime–Silica and Alkali–Lime–Borosilicate Glasses. Part II. Reduction–Oxidation Equilibria in Glasses, as Revealed by the Colours Given by Iron. The Influence of Alumina on the Colour
Pt. II-Reduction-Oxidation Equilibria in Glasses, as Revealed by the Colours...
JSGT 1950 V34 T173-T192
work is described. The ferrous–ferric balance, based on the colour analysis of each of the experimental glasses, has been calculated, the" ferrous "proportion being taken to include the ferrous iron corresponding with the amount of ferrous colour, together with the appropriate proportion of the calculated amount of ferroso-ferric iron present. The results confirm the conclusions drawn in Part I, namely, that the state of oxidation is higher (a) as the iron content is progressively increased, (b) as potash is introduced to replace soda partially or completely, either weight for weight or in equivalent amounts, and (c) in glasses containing boric oxide replacing soda. The ferrous–ferric ratios obtained by colour analysis and by chemical analysis show marked differences which suggest that, even when all precautions are taken to prevent oxidation during the chemical analysis, some oxidation is inherent in this process, probably during the initial taking up of the glass into solution in the hydrofluoric acid mixture. The effects of prolonged founding have been investigated and show that, under given melting conditions, the ferrous–ferric balance in the glass is not usually greatly affected, and in some cases not at all, by prolongation of the founding period, the balance being determined primarily by the composition of the glass, i.e. by the proportions and nature of the alkali constituents of the glass. Increased reduction with prolongation of founding time was noticeable in glasses containing the mixed alkalis, soda and potash, and in the all potash glasses containing potash molecularly equivalent to 17% soda. A number of glasses have been melted containing lithia replacing some or all of the original soda. With small amounts of lithia, the reduction is greater than in the corresponding all-soda glasses, as would be expected from the greater mobility of the lithium ions. With lithia contents of the order of 8–9%, however, there is some evidence of a swing back towards a higher state of oxidation, and this tendency becomes marked when the proportion of lithia is still further increased. With high proportions of lithia there is an appreciable increase in the ferric colour and a very considerable increase in the grey absorption attributed to ferroso-ferric oxide. No satisfactory explanation of the effects produced by lithia can be put forward without further and more detailed investigation. The pronounced development of the ferroso-ferric grey colour in glasses containing high proportions of B2O3 suggests that the presence of B2O3 lessens the chances of Fe2O3 becoming incorporated into the glass structure as “colourless” iron. If this were so the ferric oxide in a glass containing B2O3 would tend to remain as a colloidal dispersion in the glass; it would thus be more likely to encounter an alkali ion and to be reduced either to the ferrous or to the forroso-ferric form. It appeared possible, therefore, that a similar effect might be obtainable by introducing alumina in place of boric oxide, on the assumption that these trivalent oxides all tend to enter the glass structure in the same way. Experimental glasses were therefore melted in which Al2O3 was substituted for silica in the standard soda–lime–silica glass on which all the experimental glasses were based. The effect of this substitution was closely similar to that produced by introducing boric oxide, insofar as the development of the grey ferroso-ferric colour is concerned. The increase in the ferroso-ferric component appeared to be in approximately the same ratio as the molecular proportions of boric oxide and alumina added, but this cannot be regarded as satisfactorily established because the number of results available is too small to provide sufficient evidence on which such a conclusion could definitely be based.
H. Moore & S. N. Prasad
