Glass Manufacture

by Walter Rosenhain

The question which now naturally presents itself to us is, what is the essential difference between, for instance, a piece of red glass and a piece of “white” glass that confers upon the former the power of absorbing blue light? A perfectly complete and satisfactory answer to this question is not, in the writer’s opinion, available in the present state of our knowledge, but to a certain extent the difference between the two kinds of glass can be explained. The difference is produced, in the first instance by introducing into the colourless glass some additional chemical element or elements, the substances in question being generally known as “colouring oxides,” although they are by no means always introduced in the form of oxides, and are frequently present in the glass in entirely different forms. To a certain extent the colour of the glass may be ascribed to a definite “colouring” property of the chemical elements concerned; thus most of the chemical compounds of such elements as nickel, cobalt, iron, manganese and copper are more or less deeply coloured substances, and it would seem as if the atoms or “ions” of these elements had the specific power of absorbing certain varieties of light-waves while not materially affecting others. But this specific “colouring” property is not so easily explained when we recollect that the colours of iron compounds, for example, may be green or red according to the state of combination in which that element is present, and that iron has also the power of imparting either a green or a yellow colour to glass according to circumstances. The detailed discussion of these questions, however, lies outside our present scope, and we must confine ourselves to the broad statement that colouring substance in glass may be roughly divided into two kinds or groups; the first and probably the largest group are those bodies which occur in glass in true solution, the element itself being present in the combined state as a silicate or other such compound (borate, phosphate, etc.) which is soluble in the glass. In this class, the colouring effect upon the glass is specifically that of the element introduced, and is brought about in the same way as the colouring of water when a coloured salt—such as copper sulphate—is dissolved in it. The second class of colouring substances, however, behave in a different manner; they are probably present in the glass in a state of extremely fine division, and held not in true solution, but really in a sort of mechanical suspension that approximates to the condition of what is known as a “colloidal solution.” The point which is known beyond doubt, thanks to the researches of Siedentopf and Szigmondi on ultra-microscopical particles, is that in certain coloured glasses, of which ruby glass is the best example, the colouring substance, be it gold or cuprous oxide, is present in the form of minute but by no means atomic or molecular particles suspended in the glass. The presence of these particles has been made optically evident, although it can hardly be said that they have been rendered visible, and it is at all events probable that these suspended particles act each as a whole in absorbing the light-waves characteristic of the colour which they produce in glass. This being the case, it is easy to understand how readily the colour of such glasses is altered or spoilt by manipulations which involve heating and cooling at different rates—too rapid a rate of cooling producing a different grouping of the minute particles, altering their size or shape, or even obliterating them entirely by allowing the element in question to go into or to remain in solution in the glass.

While it would be entirely foreign to the purpose of this volume to give in this place a series of recipes for the production of various kinds of coloured glass, it will be desirable to state in general terms the colours or range of colours which can be produced in various kinds of glass by the introduction of those chemical elements which are ordinarily used in this way. In general terms it may be said that the lighter elements do not as a rule tend to the production of coloured glasses, while the heavier elements, so far as they can be retained in the glass in either solution or suspension, tend to produce an intense colouring effect. The element lead appears to form a striking exception to this rule, but this is due to the fact that while the silicates of most of the other heavy elements are more or less unstable, the silicate of lead is very stable, and can only be decomposed by the action of reducing agents. When lead silicates are decomposed in this way, however, the resulting glass immediately receives an exceedingly deep colour, being turned a deep opaque black, although in very thin layers the colour is decidedly brown. On the other hand, glasses very rich in lead are always decidedly yellow in colour, and it has been shown that this coloration is due to the natural colour of lead silicates and not to the presence of impurities. What has just been said of lead applies, with only very slight modification, also to the rare metal thallium and its compounds, which have been introduced into glass for special purposes. Leaving these two exceptional bodies on one side, we now pass to a consideration of the elements in the order of their chemical grouping. The rare elements will not be considered except in certain cases where their presence in traces is liable to affect results attained in practice.

The Alkali Metals, sodium, potassium, lithium, etc., and their compounds, have no specific colouring effect, although the presence of soda or of potash in a glass affects the colours produced by such substances as manganese, nickel, selenium, etc.

Copper, as would be anticipated from the deep colour of most of its compounds, produces powerful colouring effects on glass. Cupric silicates produce intense green, to greenish-blue tints. Copper, either as metal or oxide, added to glass in the ordinary way, always produces the green colour; but when the full oxidation of the copper is prevented by the presence of a reducing body, and the glass is cooled slowly, or is exposed to repeated heating followed by slow cooling, an intense ruby coloration is produced. In practice this colour is produced by introducing tin as well as copper into the mixture, and so regulating the conditions of melting as to favour reduction rather than oxidation of the copper. Under these circumstances the copper is left in the glass in a finely divided and evenly suspended state; if exactly the right state of division and suspension is arrived at, a beautiful red tint is the result, although the coloration of the glass is so intense that it can only be employed in very thin sheets, being “flashed” upon the surface of colourless glass to give it the necessary strength and thickness for practical use. It is further very easy to slightly alter the arrangement of the copper in the glass, with the result of producing an opaque, streaky substance resembling sealing-wax in colour and appearance, this product being, of course, useless from the glass-maker’s point of view. Finally, by exceedingly slow cooling, and under other favouring conditions which are not really understood, the particles of suspended colouring-material—be it metallic copper or cuprous oxide—grow in size and attain visible dimensions, appearing as minute shimmering flakes, thus producing the beautiful substance known as “aventurine.”

Silver is never introduced into glass mixtures, the reason being that it is so readily reduced to the metallic state from all its compounds that it cannot be retained in the glass except in a finely-divided form, causing the glass to assume a black, metallic appearance resembling the stains produced by the reduction of lead in flint glasses. On the other hand, silver yields a beautiful yellow colour when applied to glass as a surface stain, and it is widely used for that purpose.

Gold is introduced into glass for the production of brilliant ruby tints; its behaviour is very similar to that of copper, except that the noble metal has a great tendency to return to the metallic state without the aid of reducing agents. No addition of tin is therefore required, but the rate of cooling, etc., must be properly regulated, since rapidly cooled glass containing gold shows no special colour, the rich ruby tint being only developed when the glass is re-heated and cooled slowly. The colouring effect of gold is undoubtedly more regular and uniform than that of copper, and it is accordingly possible to obtain much lighter shades of red with the aid of the noble metal. “Gold ruby” can therefore be obtained of a tint light enough to be used in sheets of ordinary thickness, and the process of “flashing” is not essential.

The elements of the second group, such as magnesium, calcium, strontium, barium, zinc and cadmium, exert no strong specific colouring action on glass, with perhaps the exception of cadmium, and that element only does so to any considerable extent in combination with sulphur, sulphide of cadmium having the power of producing rich yellow colours in glass. The sulphur compounds of barium also readily produce deep green and yellow colours, and the formation of these tints is, indeed, very difficult to avoid in the case of glasses containing much barium. A colouring effect has sometimes been ascribed to zinc, but this is not in accordance with facts.

Of the elements of the third group, only boron and aluminium are ever found in glass in any notable quantity. Boron is present in the form of boric acid or borates, and as such produces no colouring effect, nor does there seem to be any tendency for the separation of free boron. The compounds of aluminium also possess no colouring effect, although certain compounds of this element are utilised for imparting a white opacity to glass for certain purposes—such glass being known as “opal.”

The elements of the fourth group are of greater importance in connection with glass. Carbon is capable of exerting powerful colouring effects when introduced into glass. These effects are of two kinds, viz., indirect in consequence of the reducing action of carbon on other substances present, and direct from the presence of finely-divided carbon or carbides in the glass. The latter are similar in kind to those produced by the presence of other finely-divided elementary bodies (copper, gold, lead, etc.) except that the lightness of the carbon particles tends to the production of yellow and brown colours rather than of red and black, while the chemical nature of carbon renders the glass in which it is suspended indifferent to rapid cooling, so far as the carbon tint is concerned. The indirect effects of carbon, in reducing other substances that may be present in the glass, become evident with much smaller proportions of carbon than are required to produce visible direct effects. As we have seen above, carbon, in the form of coke, charcoal or anthracite coal, is regularly introduced, as a reducing medium, into glass mixtures containing sulphate of soda. If even a slight excess of carbon be used for this purpose, the formation of sulphides and poly-sulphides of sodium and of calcium results, and these bodies, like all sulphides, impart a greenish-yellow tint to the glass, at the same time bringing other undesirable results in their train.

Silicon, in the form of silicic acid and its compounds, is a fundamental constituent of all varieties of glass, and in this form is in no sense a colouring substance; on the other hand, there is no doubt that under some conditions silicon may be reduced to the metallic state at temperatures which normally occur in glass-furnaces, and it is practically certain, that if present in glass in this condition, silicon would colour the glass. It is just possible that some of the colouring effects produced in ordinary glass by powerful reducing agents, such as carbon, either in the solid form or as a constituent of furnace gases, may be due to the reduction of silicon in the glass.