Glass Manufacture

by Walter Rosenhain

While the range of bodies that can be obtained in the vitreous state is very large, only a comparatively small number of substances are ordinarily incorporated in industrial glasses. With the exception of certain special glasses used for scientific purposes, such as the construction of optical lenses, thermometers and vessels intended to resist unusual treatment, all industrial glasses are of the nature of mixed silicates of a few bases, viz., the alkalies, sodium and potassium, the alkaline earths, calcium, magnesium, strontium, and barium, the oxides of iron and aluminium (generally present in minor quantities), and lead oxide. The manner in which these various elements enter into combination and solution with one another has been much investigated, and the more general conclusions have been anticipated in what has been said above. It is abundantly evident that glasses are not definite chemical compounds, but rather solutions, in varying proportions, of a series of definite compounds in one another. In many cases the actual constitution of industrial glasses is so complex as, for the present at all events, to baffle adequate chemical expression.

One of the factors that limit the range of possible compositions of glasses has already been indicated, and two others must now be discussed. For industrial purposes, the cost and rarity of the ingredients becomes a vital bar at a certain stage; thus the use of such elements as lithium, thallium, etc., is prohibitively costly. In another direction the glass-maker is very effectively restrained by the limitations of his furnaces as regards temperature. The presence of excessive proportions of silica, lime, alumina, etc., tends to raise the temperature required for the free fusion of the glass, and when this temperature seriously exceeds 1600° C., the manufacture of the glass in ordinary furnaces becomes impossible. Thus pure silica can be converted into a glass possessing very valuable properties, but the requisite temperature cannot be attained in regenerative gas-fired furnaces such as are ordinarily used by glass manufacturers. The production of this glass has accordingly been carried on upon a small scale only by means of laboratory furnaces heated by oxy-acetylene flames, while latterly a less perfect variety of silica glass-ware has been produced on a large scale by the aid of electric furnaces. Such methods are, however, obviously limited to very special products commanding special prices.

A further limitation in the choice of chemical components is placed upon the manufacturer by the actual chemical behaviour of the glass both during manufacture and in use. As regards chemical behaviour during manufacture, it must be borne in mind that, although glasses are of the nature of solutions rather than of compounds, yet these solutions tend towards a state of saturation; thus a glass rich in silica and deficient in bases will readily dissolve any basic materials with which it may come in contact, while, on the other hand, a glass rich in bases and poor in acid constituents such as silica, boric acid or alumina, will readily absorb acid bodies from its surroundings. During the process of melting, glass is universally contained in fire-clay vessels. These are chosen, as regards their own chemical composition, so as to offer to the molten glass a few of those materials in which the glass itself is deficient; yet a limit arises in this respect also, since glasses very rich in bases, such as the very dense lead and barium glass made for optical purposes, rapidly attack any fire-clay with which they may come in contact. The finished glass also betrays its chemical composition by its chemical behaviour towards the atmospheric agents, such as moisture and carbonic acid, with which it comes in contact; glasses containing an excessive proportion of alkali, for example, are found to be seriously hygroscopic and to undergo rapid decomposition, especially in a damp atmosphere.

Within the limits set by these considerations, the glass manufacturer chooses the chemical composition of his glass according to the purpose for which it is intended; for most industrial products the cheapest and most accessible raw materials that will yield a glass of the requisite appearance are employed, while for special purposes the dependence of physical properties upon chemical composition is utilised, as far as possible, in order to attain a glass specially suited to the particular requirements in question. Thus the flint and barium glasses used for table and ornamental ware derive from the dense and strongly refracting oxides of lead and barium their properties of brilliancy and weight. The fusibility and softness imparted to the glass by the presence of these bases further adapts it to its purpose by facilitating the complicated manipulations to which the glass must be subjected in the manufacturing processes.

Taking our next example at almost the opposite extreme, the hardest “combustion tubing,” which is intended to resist a red heat without appreciable softening, is manufactured by reducing the basic contents of the glass to the lowest possible degree, especially minimising the alkali content, and using the most refractory bases available, such as lime, magnesia, and alumina in the highest possible proportions. Such glass is, of course, difficult to melt, and special furnaces are required for its production, but on the other hand this material meets requirements which ordinary soda-lime or flint glass tubing could never approach. Another instance of these refractory glasses is to be found in the Jena special thermometer glasses and in the French (Tonnelot) “Verre dur”; the best of these glasses show little or no plasticity at temperatures approaching 500° C., and have thus rendered possible a considerable extension of the range of the mercury thermometer. Further modification of chemical composition has resulted in the production of glasses which are far less subject to those gradual changes which occur in ordinary glass when used for the manufacture of thermometers—changes which vitiated the accuracy of most early thermometers. A still more extensive adaptation of chemical composition to the attainment of desired physical properties has been reached primarily as a result of the labours of Schott and Abbé, in the case of optical glasses. The work of these men, and the developments which have followed from it, both at the works founded by them at Jena and elsewhere, have so profoundly modified our knowledge of the range of possibilities embraced by the class of vitreous bodies, that it is not at all easy at the present time to realise the former narrow and restricted meaning of the term “glass.” The subject of the dependence of the optical properties of glass upon chemical composition will be referred to in detail in [Chapter XII.] on “Optical Glass,” but the outline of the influence of composition on properties here given could not be closed without some reference to this pioneer work of the German investigators.

The chemical behaviour of glass surfaces, to which we have already referred, is of the utmost importance to all users of glass. The relatively neutral chemical behaviour of glass is, in fact, one of its most useful properties, and, next to its transparency, most frequently the governing factor in its employment for various purposes. Thus the entire use of glass for table-ware depends primarily upon the fact that it does not appreciably affect the composition and flavour of edible solids or liquids with which it is brought into contact—a property which is only very partially shared even by the noble metals. Again, the use of glass windows in places exposed to the weather would not be feasible if window-glass were appreciably attacked by the action of water or of the gases of the atmosphere. For these general purposes, it is true, most ordinary glasses are adequately resistant, but this degree of perfection in this respect is only the outcome of the centuries of experience which the practical glass-maker has behind him in the manufacture and behaviour of such glass. When, however, a higher degree of chemical resistance is required for special purposes, as for instance when glass is called upon to resist exposure to hot, damp climates, or is intended to contain corrosive liquids, the rules which are an adequate guide to the glass-maker in meeting ordinary requirements are no longer sufficient, particularly when the glass is expected to meet other stringent requirements as well. It has, in fact, frequently happened that a glass-maker, in striving to improve the colour or quality of his glass, as regards freedom from defects, brilliancy of surface, etc., has spoilt the chemical durability of his products. The reason lies in the fact, long known in general terms, that an increased alkali content reduces the chemical resistance of glass, while at the same time such an increase of alkali is the readiest means whereby the glass-maker can improve his glass in other respects by making it more fusible and easier to work in every way.

This subject of the chemical stability of glass surfaces attracted much attention during the later part of last century, and careful investigations on the subject were carried out, particularly at the German Reichsanstalt (Imperial Physical Laboratory) at Charlottenburg. Here also the labours of Schott and Abbé proved helpful, until at the present time such glass as that used by the Jena firm in the production of laboratory ware, and certain other special glasses of that kind, are fitted to meet the most stringent requirements.

Leaving aside the inferior glasses, containing, generally, more than 15 per cent. of alkali, the behaviour of glass surfaces to the principal chemical agents may be summed up in the following statements. Pure water attacks all glass to a greater or lesser extent; in the best glasses the prolonged action of cold water merely extracts a minute trace of alkalies, but in less perfect kinds the extraction of alkali is considerable on prolonged exposure even in the cold, and becomes rapidly more serious if the temperature is raised. Superheated water, i.e., water under steam pressure, becomes an active corroding agent, and the best glasses can only resist its action for a limited time. For the gauge-glass tubes of steam boilers working at the high pressures, which are customary at the present time, specially durable glasses are required and can be obtained, although many of the gauge-tubes ordinarily sold are quite unfit for the purpose, both from the present point of view and from that of strength and “thermal endurance.”

In certain classes of glass, the action of water, especially when hot, is not entirely confined to the surface, some water penetrating into the mass of the glass to an appreciable depth. The exact mechanism of this action is not known, but the writer inclines to the view that it arises from a partial hydration of some of the silica or silicates present in the glass. If such glasses be dried in the ordinary way and subsequently heated, the surface will be riddled with minute cracks, some glass may even flake off, and the whole surface will be dulled. As such penetrating action sometimes takes place—in the poorer kinds of glass—by the action of atmospheric moisture when the glass is merely stored in a damp place, it is often mistaken for “devitrification.” This latter action, however, is not known to occur at the ordinary temperature, although glass when heated in a flame frequently shows the phenomenon; it is, however, entirely distinct from the surface “corrosion” just described. Water containing alkaline substances in solution acts upon all glasses in a relatively rapid manner; it acts by first abstracting silica from the glass, the alkali and lime being dissolved or mechanically removed at a later stage. Water containing acid bodies in solution—i.e., dilute acid—on the other hand acts upon most varieties of glass decidedly less energetically than even pure water, and much less vigorously than alkaline solutions; this peculiar behaviour probably depends upon the tendency of acids to prevent the hydration of silica, this substance being thereby enabled to act as a barrier to the solvent action of the water upon the alkaline constituents of the glass. The better varieties of glass are also practically impervious to the action of strong acids, although certain of these, such as phosphoric and hydrofluoric, exert a rapid action on all kinds of glass. Only certain special glasses, containing an excessive proportion of basic constituents and of such substances as boric or phosphoric acid, are capable of being completely decomposed by the action of strong acids, such as hydrochloric or nitric, the bases entering into combination with the acids, while the silicic and other acids are liberated.

In connection with the action of acids upon glass, mention should be made of certain special actions that are of practical importance. The dissolving action of hydrofluoric acid upon glass is, of course, well known. It is used in practice both in the liquid and gaseous form, and also in that of compounds from which it is readily liberated (such as ammonium or sodium fluoride), for the purpose of “etching” glass, and also in decomposing glass for purposes of chemical analysis. Next in importance ranks the action of carbonic acid gas upon glass, especially in the presence of moisture. The action in question is probably indirect in character; the moisture of the air, condensing upon the surface of the glass, first exerts its dissolving action, and thus draws from the glass a certain quantity of alkali, which almost certainly at first goes into solution as alkali hydrate (potassium or sodium hydroxide); this alkaline solution, however, rapidly absorbs carbonic acid from the air, and the carbonate of the alkali is formed. If the glass dries, this carbonate forms a coating of minute crystals on the surface of the glass, giving it a dull, dimmed appearance; this, however, only occurs ordinarily with soda glasses, since the carbonate of potassium is too hygroscopic to remain in the dry solid state in any ordinary atmosphere. Potash glasses are, as such, no more stable chemically than soda glasses, but they are for the reason just given less liable to exhibit a dim surface. If the dimming process, in the case of a soda glass, has not gone too far, the brightness of the surface of the glass may be practically restored by washing it with water, in which the minute crystals of carbonate of soda readily dissolve, while separated silica is removed mechanically. An attempt made to clean the same dimmed surface by dry wiping would only result in finally ruining the surface, since the small sharp crystals of carbonate of soda would be rubbed about over the surface, scratching it in all directions.