It has already been indicated that, for glass-making purposes, the raw materials are required in a state of reasonably fine division. The exact degree of fineness required depends very much upon the nature of the ingredient in question, the general rule being that the more refractory and chemically resistant materials require to be most finely ground, while substances which melt and react readily, such as soda ash and salt-cake, do not require very fine grinding.

Assuming that the materials are available in a suitable state of fineness, the first step in the process of glass melting consists in securing their admixture in the proper proportions. This may be done by hand entirely, by hand aided by some machinery, or entirely automatically. The process of hand mixing is only available for relatively small quantities of material and requires very careful supervision if inadequate mixing is to be avoided. In most cases the actual weighing out is done by hand, while the mixing is done by machinery. In this process the separate ingredients are weighed out from barrows or skips and are tipped into a large hopper whence each batch, as soon as it is completed, passes into the mixing chamber of the mixing machine. This may consist of nothing more than a cylindrical chamber in which steel arms revolve and stir up the contents, but more modern appliances take the form of rotating barrels or cylinders, set up on an inclined axis and provided with suitable shelves and baffles; in these the materials are very thoroughly shaken over and mixed. Where hand mixing is adopted, the various ingredients of each batch are thrown into a large bin and are there turned over several times with shovels, the entire material being ultimately sieved through a wire sieve of suitable mesh. In all cases the resulting mixture should be perfectly uniform in colour and texture, and analyses of different samples should show only small variations. With the mixture thus prepared the “cullet” or broken glass which is to be re-melted is now incorporated; ideally this should also be uniformly distributed, but this is rarely attempted in practice on the large scale.

The next step in the process is the introduction of the mixture into the furnace. In the case of tank furnaces this is a simple matter, since in these the temperature is kept as nearly constant as possible, and raw materials may, therefore, be introduced at almost any time, the amount introduced being so regulated as to keep the level of the molten glass or “metal” as nearly constant as possible. The actual introduction is managed by means of a large opening or door at what is known as the “melting end” of the furnace. Normally this opening is covered by a large fire-brick block suspended by a chain running over pulleys and counterbalanced by a counterpoise weight. When charging is to begin, this block is raised and the opening is uncovered. The raw materials are then introduced either by hand, by the aid of long-handled shovels, or they are first filled into a long scoop moved by mechanical means forward into the furnace, where it is given a half-turn, which empties the contents out, and is then rapidly withdrawn.

This charging process may be repeated every half-hour, or larger quantities may be introduced once every four hours, according to the practice that may be adopted at any particular furnace.

In the case of pot furnaces the charging process is not so simple. Here the first charge of raw materials has to be introduced into a pot which has been almost entirely emptied during the working-out process, and the temperature of the furnace has also fallen very considerably during this time. Before new material is introduced, the heat of the furnace must first be adequately restored. If this is not done, the fusion of the glass takes an abnormal course and very imperfect results arise. Further, the quantity of material introduced at one time must be carefully adjusted to the capacity of the pot. During the earlier stages of fusion most glass mixtures form large masses of foam, and if the crucible has been too heavily charged this foam overflows, with the result that valuable material is lost and the floor and passages of the furnace are clogged with glass. A certain amount of overflow, as well as leakage from defective crucibles, is, however, unavoidable, and for this purpose every pot furnace is provided with a chamber so placed that the glass will flow into it and so be prevented from finding its way into the regenerators or other parts where its presence would hinder the working of the furnace. These receptacles or “pockets” must, however, be periodically cleared of their contents from outside, and this constitutes one of the most irksome operations connected with glass manufacture. Owing to the occurrence of foaming and to the fact that the raw materials occupy much more space than the glass formed from them, it is necessary to fill the pot with fresh batches of raw materials several times, the quantity which can be introduced decreasing each time. The number of times that this must be done depends upon the particular circumstances, but from four to eight “fillings” are commonly used for various kinds of glass and size of pot. The precise stage at which a fresh batch of raw materials should be introduced is another matter requiring careful attention. For some purposes it is necessary to wait until the previous batch is completely melted, while in other cases raw material may be added whilst some of the previous batch is still floating on the surface of the glass in the pot.

We have now to consider the chemical reactions which take place in the mixture of raw materials that are introduced into the hot furnace. The exact course of these reactions is not known in very great detail, as this could only be ascertained by an elaborate research on the nature of the intermediate products that result under various circumstances. A research of this kind would throw much light on the whole of the melting processes but is in itself so difficult that it has not yet been carried out at all fully. We can therefore only give an account of the chemical changes from our knowledge of the end-results and of a few intermediate products that are known. To take the simplest case, we may consider a mixture consisting of sand, carbonate of lime and carbonate of soda mixed in suitable proportions. In such a case we know that the mere action of heat alone will produce two changes—the carbonate of soda will melt and the carbonate of lime will lose its carbonic acid and be “burnt” or converted into caustic lime. The first stage of the fusion process thus probably results in a mass consisting of sand grains and grains of carbonate of lime undergoing decomposition, all cemented together by molten carbonate of soda. This mass will be full of bubbles, some derived from the air enclosed between the grains of the original mixture and thus trapped by the melting mass, and others formed by the carbonic acid which is being driven off in the form of gas by the decomposition of the carbonate of lime. At the temperature of the furnace, however, silica has the properties of a strong acid, and not only attacks the carbonate of lime much in the same manner as, for instance, hydrochloric acid would do in the cold, but the silica also attacks the carbonate of soda, which heat alone can scarcely decompose. The exact order in which these reactions take place will depend upon the temperature of the furnace and the degree of mixing attained in the preparation of the raw materials. Although in the long run the final result will probably be the same as regards purely chemical constitution, much of the technical success of the process must depend upon the exact sequence of the changes involved, as this must govern the number and size of the bubbles that are formed in the glass and the fluidity of the mass from which these bubbles have to free themselves. In the present state of our knowledge, however, we can only say that the final result is the complete expulsion of all carbonic acid from the compounds present (although it may remain entangled in the glass in the form of bubbles) and the formation of silicates of both lime and soda which remain in the finished glass in a state partly of mutual chemical combination, partly of mutual solution.

The description of the process of fusion just given applies, with slight modifications, to the melting of ordinary flint-glass mixtures as well as to lime glasses, with the one modification that the carbonate of lime of the lime-soda glass is replaced by red-lead, and the gas evolved by the decomposition of the red-lead is oxygen in place of the carbonic acid evolved from the decomposition of the carbonate of lime. In the case of both lime and flint glasses, however, certain other substances besides those mentioned are usually introduced in small quantities. Although these substances do not very materially affect the end-products of the chemical reactions, they very materially affect the intermediate stages, and thus serve the purpose for which they are introduced by affecting the course of the chemical changes in a favourable manner. The substances usually employed for this purpose are arsenic and nitrate of either soda or potash. The manner in which the arsenic acts is very obscure and cannot be discussed in detail here; the chief factors in its action are, however, its volatility and its power of either absorbing oxygen or parting with it according to circumstances. The action of the nitrates is chiefly dependent upon the oxygen which they yield on decomposition by heat. This oxygen is in some cases stored up by other ingredients of the mixture and only given off at a much later stage, when the evolution of this gas assists in the removal of the last small bubbles of inert air or carbonic acid gas still left in the glass. The oxidising action of the nitrates, however, serves chiefly for the destruction of organic matter and the full oxidation of any iron present; both processes which tend to improve the colour of the glass, while in the case of flint glasses the presence of these oxidising additions is necessary to avoid all risk of reduction of lead, since this would result in the complete blackening of the glass.

A much more complicated set of reactions occur when the alkali of a soda-lime glass is introduced either partly or wholly in the form of sulphate of soda (salt-cake). We have already pointed out that the unaided action of heat and of silica is not sufficient to bring about the rapid decomposition of sulphate of soda which is required for successful glass manufacture, and that the intervention of reducing agents is required. For this purpose a certain amount of carbon in the form of coke, charcoal or anthracite coal, is introduced into all salt-cake mixtures, but the reducing gases of the furnace atmosphere also play an important part in the reactions that take place. Here again it is not possible to give anything but an incomplete account of what takes place. The rationale of the whole process lies, no doubt, in the fact that sulphite of soda (Na₂SO₃) is much more readily decomposed by the action of hot silica than the sulphate (Na₂SO₄) itself, so that the essential action of the reducing agents consists in robbing the sulphate of part of its oxygen, thus reducing it to the condition of sulphite and rendering it accessible to the attack of silicic acid. But if we attempt to express such a reaction in the usual manner by a chemical equation from which the quantity of carbon required to effect the reduction in question can be calculated, we find that the amount of carbon required in practice is very considerably less than that given by this theory; it follows therefore that either this very large amount of reducing action must be ascribed to the furnace gases, or that the actual reactions are not strictly of the kind we have described. Both explanations are probably partly correct, and in practice the amount of carbon to be used in a given mixture and furnace can only be found by actual trial, in which the manufacturer is, of course, guided by the results obtained with other furnaces of a similar type. The end-product of the reactions is again a mixture of silicates, but a certain amount of undecomposed sulphate is always found in such glasses, while gaseous oxides of sulphur escape from these furnaces in considerable quantity. Under exceptional circumstances the glass may even contain sulphides of soda or of lime, and sometimes even suspended carbon, but these are abnormal constituents and result in the serious discolouration of the glass.

It is obvious that to a mixture containing carbon as a reducing agent such oxidising materials as nitrates cannot be added, but small quantities of arsenic and of manganese dioxide are added because their other properties are sufficiently valuable to outweigh their disadvantages as oxidising agents.

Having now briefly considered the process of fusion proper, we pass to the second stage in the melting of glass. In a properly conducted glass-furnace, when the last trace of undecomposed raw materials has disappeared, we find the glass as a transparent mass throughout which gas bubbles are thickly disseminated. For the majority of purposes it is necessary to free the glass as perfectly as possible from these bubbles before it is worked into its final form. This freeing or “fining” process is carried out by further and more intense heating of the molten glass, which is thereby rendered more fluid and allows the bubbles to disengage themselves by rising to the surface. This occurs much more readily when the bubbles are large; very minute bubbles, in fact, show no inclination to rise through the fluid mass. The glass-maker accordingly compounds his mixtures of raw materials in such a way as to yield large bubbles, or, failing that, he adds to the molten mass some substance that evolves a great many large bubbles, and these in their upward course through the glass sweep the small ones away with them. The added substance may be an inorganic volatile body, such as arsenic, or more frequently some vegetable substance containing much moisture is introduced into the glass. The most usual method is to place a potato in the crook of a forked iron rod and then to dip the rod with the attached potato into the molten glass; the heat at once begins to drive off the moisture and to decompose the potato, so that there is a violent ebullition of the whole mass. This “boiling up” process assists the fining considerably and also serves to mix the whole contents of the pot very thoroughly, but it has some attendant disadvantages, such as the introduction of oxide of iron into the glass from the rod which is used in the operation, while the contaminated material adhering to the walls of the pot itself is dragged off and mixed with the rest of the glass by the violent stirring action that takes place. It is, of course, further obvious that this process can only be usefully applied to glass melted in pots, since the bulk of the molten glass in a tank furnace could not be reached at all in this manner. Mixtures that are to be melted in tanks must therefore be capable of freeing themselves of their enclosed bubbles without such outside aid. In a tank, in fact, the whole melting process proceeds on somewhat different lines, since the temperature of the furnace is never intentionally varied, while on the other hand the melting glass travels down the furnace into regions whose temperature can be regulated to favour the various stages of the process that take place in each part of the furnace. On the whole, however, it is an undoubted fact that while the running of a pot furnace can be varied, within wide limits, to suit the requirements of whatever mixture it is desired to melt, in the case of tank furnaces the mixture must be closely adjusted to the requirements of the furnace, whose general “run” cannot be very readily altered.

The completion of the “fining” process is generally determined by taking samples of the glass out of the pot or tank and examining them for enclosed bubbles. Such samples may be obtained in a variety of ways, the most usual method being to dip a flat iron rod just below the surface of the glass and to lift it out vertically upwards, thus retaining on the flat surface of the rod some of the glass that lay there at the moment when the rod was immersed. These test samples or “proofs” are examined very carefully, and if no trace of bubbles can be observed the glass is generally regarded as “fine,” but it is by no means certain that the absence of bubbles from such a small sample will prove that the whole mass is free; that, however, is a point where the melter’s experience enables him to judge how far he may rely upon the indications given by the “proofs.” When the glass is “fine” it frequently happens that the surface of the molten mass is contaminated by specks of foreign matter floating on the glass; for the purpose of removing these, the surface of all glass is skimmed before work is begun upon it. This is done by removing the surface skin of glass by means of suitably shaped iron rods, upon which small masses of molten glass are first “gathered.” Finally, it only remains to reduce the temperature of the glass from that of the melting and fining process to the much lower temperature at which the various methods of working the glass are carried out. In pot furnaces this is accomplished by lowering the temperature of the entire furnace, while in tank furnaces the fine glass flows into the working chamber of the tank which is always kept at the working temperature.