The choice of raw materials for all branches of glass manufacture is a matter of vital importance. As a rule all “fixed” bodies that are once introduced into the glass-melting pot or furnace appear in the finished glass, while volatile or combustible bodies are more or less completely eliminated during the process of fusion. Thus while the chemical manufacturer can purify his products by filtration, crystallisation or some other process of separation, the glass-maker must eliminate all undesirable ingredients before they are permitted to enter the furnace, and the stringency of this condition is increased by the fact that the transparency of glass makes the detection of defects of colour or quality exceedingly easy. For the production of the best varieties of glass, therefore, an exacting standard of purity is applied to the substances used as raw materials. As the quality of the product decreases, so also do the demands upon the purity of raw materials, until finally for the manufacture of common green bottles, even such very heterogeneous substances as basaltic rock and the miscellaneous residues of broken, defective and half-melted glass forming the refuse of other glassworks may be utilised more or less satisfactorily. For the best kinds of glass the most desirable quality in raw materials is thus as near an approach to purity as possible under commercial conditions, and next to that, as great a constancy of composition as possible. For instance, the quantity of moisture contained in a ton of sand appreciably affects the resulting composition of the glass, and if the sand cannot be obtained perfectly dry, it should at least contain a constant proportion of moisture, otherwise it becomes necessary to determine, by chemical tests, the percentage of moisture in the sand that is used from day to day, and to adjust the quantity used in accordance with the results of these tests, a proceeding which, of course, materially complicates the whole process. In other cases, variable composition is not so readily allowed for, and uncontrollable variations in the composition of the glass result—at times the quality falls off unaccountably, or the glass refuses to melt freely at the usual temperature. The systematic employment of chemical analysis in the supervision of both the raw materials and of various products will frequently enable the manufacturer to trace the causes of such undesirable occurrences; but however necessary such control undoubtedly is, it cannot entirely compensate for the use of raw materials liable to too great a variation in composition or physical character. For not only the chemical composition, but also the physical condition and properties of the material are of importance in glass manufacture. Thus it is essential that materials to be used for glass-melting should be obtainable in a reasonably fine state of division, and in this connection it must be remembered that both exceedingly hard bodies and soft plastic substances can only be ground with very great difficulty. Further, where a substance occurs naturally as a powder, this powder should be of uniform and not too fine a grain, more especially if it belongs to the class of refractory rather than of fluxing ingredients. In that case the presence of coarser grains will result in their presence in the undissolved state in the finished glass, unless excessive heat and duration of “founding” be employed to permit of their dissolution. This applies chiefly to siliceous and calcareous ingredients, but hardened nodules of salt-cake may behave in a similar manner.

A further consideration in the choice of raw materials is facility of storage. Thus limestone in the shape of large lumps of stone which are only ground to powder as required, is readily stored, and undergoes no deleterious change even if exposed to the weather; on the other hand, sulphate of soda (salt-cake), if stored even in moderately dry places, rapidly agglomerates into hard masses, at the same time absorbing a certain percentage of moisture. Such properties are not always to be avoided, salt-cake for example being at the present time an indispensable ingredient in many kinds of glass-making, but the value of a substance is in some cases materially lessened by such causes.

The raw materials ordinarily employed in glass-making may be grouped into the following classes:—

(1) Sources of silica.
(2) Sources of alkalies.
(3) Sources of bases other than alkalies.

(1) Sources of Silica.—The principal source of silica is sand. This substance occurs in nature in geological deposits, often of very considerable area and depth. These deposits of sand have always been formed by the disintegration of a siliceous rock, and the fragments so formed have been sifted and transported by the agency of water, being finally deposited by a river either in the sea (marine deposits) or in lakes (lacustrine deposits), while the action of the water, either during transport or after deposition, has frequently worn the individual particles into the shape of rounded grains.

In consequence of this origin, the chemical composition of sand varies very greatly with the nature of the rock whose denudation gave rise to the deposit. Where rocks very rich in silica, or even consisting of nearly pure silica, have been thus denuded, the resulting sand is often very pure, deposits containing up to 99·9 per cent. silica being known. More frequently, however, the sand contains fragments of more or less decomposed felspar, which introduce alumina, iron and alkalies into its composition. Finally, “sands” of all ranges of composition from the pure varieties just referred to down to the clay marls, very rich in iron and alumina, are known.

For the best varieties of glass, viz., optical glass, flint glass and the whitest sheet-glass, as well as for the best Bohemian glass, a very pure variety of sand is required, preferably containing less than 0·05 per cent. of iron, and not more than 0·05 per cent. of other impurities such as alumina, lime or alkali. As a matter of fact, sands containing so little iron rarely contain any other impurity except alumina in measurable quantities. The best-known deposit of such sand in Europe is that at Fontainebleau near Paris, but equally good sand is found at Lippe in Germany, whence sand is delivered commercially with a guaranteed silica content of 99·98 per cent. Sand of excellent quality, although not quite so good as the above, is obtained at Hohenbocka in Germany (Saxony) and at a few other places in Europe. In England no deposit of sand of such purity is at present being exploited.

Next in order of value to these exceedingly pure sands, come the glass-making sands of Belgium, notably of Epinal. These usually contain from 0·2 to 0·3 per cent. of iron and rather more alumina, but they are used very largely for the manufacture of sheet and plate-glass. When the standard of quality is further relaxed, a large number of sand deposits become available, and the manufacturers of each district avail themselves of more or less local supplies; thus in England the sands of Leighton in Bedfordshire and of Lynn on the East Coast, are largely used. Finally, for the manufacture of the cheapest class of bottles, sands containing up to 2 per cent. of iron and a considerable proportion of other substances are employed.

Silica, in various states of purity, occurs in nature in a number of other forms than that of sand. By far the commonest of these is that of more or less compact sedimentary rock, known as “sandstone.” As far as chemical composition is concerned, some of these stones are admirably suited for making the best kinds of glass, although as a rule a stone is not so homogeneous as the material of a good sand-bed. The stone has the further disadvantage that it requires to be crushed to powder before it can be used for glass-making, and the crushed product is generally a mixture of grains of all sizes ranging from a fine dust to the largest size of grain passed by the sieves attached to the crushing machine. The presence of the very fine particles is a distinct objection from the glass-maker’s point of view, so that it would probably be necessary to wash the sand so as to remove this dust—a process that in itself adds to the cost of the crushed stone and at the same time leads to the loss of a serious percentage of the material. Objections of the same kind apply, but with still greater force, to the use of powdered quartz or flint as sources of silica for the glass-maker; further, these materials are exceedingly hard and therefore difficult to crush, so that the price of the materials is prohibitive for glass-making purposes. The use of ground quartz and flint is therefore confined to the ceramic industries in which these substances serve as sources of silica for both bodies and glazes; in former times, however, ground flint was extensively used in the manufacture of the best kinds of glass, as the still surviving name of “flint glass” testifies.

Minerals of the felspar class, consisting essentially of silicates of alumina and one or more of the alkalies, are extensively used in glass-making and should be mentioned here, since their high silica-content (up to 70 per cent.) constitutes an effective source of silica. As a source of this substance, however, most felspars would be far too expensive, and their use is due to their content of alumina and alkali.

(2) Sources of Alkali.—Originally the alkaline constituents of glass were derived from the ashes of plants and of seaweed or “kelp”; in both cases the alkali was obtained in the form of carbonate and was ordinarily used in a very impure form; at the present time, however, the original source of alkali for industrial purposes is found in the natural deposits and other sources of the chlorides of sodium and potassium. At the present time it is not yet industrially possible to introduce the alkalies into glass mixtures in the natural form of chlorides. The principal difficulty in doing this arises from the fact that the chlorides are volatile at the temperature of glass-melting furnaces and are only acted upon by hot silica in the presence of water vapour. Introduced into an ordinary glass furnace, therefore, these salts would be driven off as vapour before they could combine with the other ingredients in the desired form of double silicates.

Alkalies are, therefore, introduced into the glass mixture in less volatile and more readily attackable forms. Of these the carbonate is historically the earlier, while the sulphate is at the present time industrially by far the more important. The Carbonate of Soda, or soda ash, which is used in the production of some special glasses, and is an ingredient of English flint glasses, is produced by either of two well-known chemical processes. One of these is the “black ash,” or “Le Blanc” process, in which the chloride is first converted into sulphate by the direct action of sulphuric acid, and the sulphate thus formed is converted into the carbonate by calcination with a mixture of calcium carbonate and coal. The sodium carbonate thus formed is separated by solution and subsequent evaporation. A purer form of sodium carbonate can be obtained with great regularity by the “ammonia soda” process, in which a solution of sodium chloride is acted upon by ammonia and carbonic acid under pressure. Soda ash produced by this process is now supplied regularly for glass-making purposes in a state of great purity and constancy of composition. It is upon these qualities that the great advantages of this substance depend, since its relatively high cost precludes its use except for special kinds of glass, and for these purposes the qualities named are of great value.

For most purposes of glass-making, such as the production of sheet and plate-glass of all kinds, the alkali is introduced in the form of salt-cake—i.e., sulphate of soda. This product is obtained as the result of the first step of the Le Blanc process of alkali manufacture—i.e., by the action of sulphuric acid on sodium chloride; salt-cake is thus a relatively crude product, and its use is due to the fact that it is by far the cheapest source of alkali available for glass-making. There are, however, certain disadvantages connected with its use. The chief of these is the fact that silica cannot decompose salt-cake without the aid of a reducing agent; such a reducing agent is partly supplied by the flame-gases in the atmosphere of the furnace, but in addition to these a certain proportion of carbon, in the form of coke, charcoal or anthracite coal must be added to all glass mixtures containing salt-cake. The use of a slightly incorrect quantity of carbon for this purpose leads to disastrous results, while even under the best conditions it is not easy to remove all traces of sulphur compounds from glass made in this way. A further risk of trouble arises in connection with salt-cake from the fact that it is never entirely free from more or less deleterious impurities. According to the exact manner in which it has been prepared, the substance always contains a small excess either of undecomposed sodium chloride or of free sulphuric acid, or the latter may be present in the form of sulphate of lime. A good salt-cake, however, should contain at least 97 per cent. of anhydrous sodium sulphate, and not more than 1·0 per cent. of either sodium chloride or sulphuric acid. While pure sodium sulphate is readily soluble in water, ordinary salt-cake always leaves an insoluble residue, consisting frequently of minute particles of clay or other material derived from the lining of the furnace in which it was prepared, or from the tools with which it was handled; and these impurities are liable to become deleterious to the glass if present in any quantity. The insoluble residue should not exceed 0·5 per cent. in amount, and in the best salt-cake is generally under 0·2 per cent.

Salt-cake possesses certain other properties that make it somewhat troublesome to deal with as a glass-making material. Thus, on prolonged exposure, particularly to moist air, the powdered salt-cake absorbs moisture from the atmosphere and undergoes partial conversion into the crystalline form of “Glauber’s Salt,” a process which results in the formation of exceedingly hard masses. Ground salt-cake, therefore, cannot be stored for any length of time without incurring the necessity of regrinding, and this accretive action even comes into play when mixtures of glass-making materials, containing salt-cake as one ingredient, are stored. In practice, therefore, salt-cake can only be ground as it is wanted, and its physical properties make it difficult to grind it at all fine, while the dust arising from this process is peculiarly irritating, although not seriously injurious to health.

Potash is utilised in glass-making almost entirely in the form of carbonate, generally called “pearl-ash.” Originally derived from the ashes of wood and other land plants, this substance is now manufactured by processes similar to those described in the case of soda, the raw material being potassium chloride derived from natural deposits such as those at Stassfurth. The pearl-ash thus commercially obtainable is a fairly pure substance, but its use is complicated by the fact that it is strongly hygroscopic and rapidly absorbs water from the atmosphere. Where it is desired to produce potash glasses of constant composition, frequent analytical determinations of the moisture contents of the pearl-ash are necessary, and the composition of the glass mixture requires adjustment in accordance with the results of these determinations.

The alkalies are also introduced into glass in the form of nitrates (potassium nitrate, or saltpetre, and sodium nitrate, or nitre); but although these substances act as sources of alkali in the glass, they are employed essentially for the sake of their oxygen contents. Such oxidising agents are not, of course, added to glass mixtures containing sulphates and carbon, but are employed to purify the mixtures containing alkali carbonates, and more especially to oxidise the flint glasses. Since these substances are only introduced into glass in small quantities their extreme purity is not of such great importance to the glass-maker, and the ordinary “refined” qualities of both nitrates are found amply pure enough to answer the highest requirements.

A certain number of natural minerals which contain an appreciable quantity of alkali are sometimes utilised as raw materials for glass manufacture. The most important of these are the minerals of the felspar class already referred to. These, however, contain a considerable proportion of alumina, while all but the purest varieties also contain more or less considerable quantities of iron. Some glass-makers regard alumina as an undesirable constituent, while others take the opposite view, and upon this view their use of felspathic minerals will depend. For the cheaper varieties of glass, however, such as bottle glass, felspathic minerals and rocks, such as granite and basalt, are freely used as raw materials. Another mineral in which both alkali and alumina are found is cryolite. This mineral is a double fluoride of soda and alumina, whose properties are particularly valuable in the production of opal and opalescent glasses. As a mere source of alkali, however, cryolite is much too expensive.

(3) Sources of Bases other than Alkalies.—The most important of these are lime and lead oxide, the former being required for the production of all varieties of plate and sheet-glass, as well as for bottles and a large proportion of pressed and blown glass, while lead is an essential ingredient of all flint glass. The only other base having any considerable commercial importance in connection with glass-making is barium oxide, while oxide of zinc, magnesia, and a few other substances are used in the manufacture of special glasses for scientific, optical or technical purposes, where glass of special properties is required. The metallic oxides which are used for the production of coloured glass are, of course, also basic bodies. These will be treated in connection with coloured glasses, with the exception of manganese dioxide, which is used in large quantities in the manufacture of many ordinary “white” glasses.

Calcium Oxide (lime) is generally introduced into glass mixtures in the form of either the carbonate or the hydrated oxide (slaked lime). The carbonate may be derived either from natural sources, or it may be of chemical origin, while the hydrate is always obtained by the calcination of the carbonate, followed by “slaking” the lime thus produced. Natural calcium carbonate occurs in great quantities in the form of chalk and limestone rocks. Both varieties are used for glass-making. Chalk is a soft friable material which is apt to clog during the grinding operations, particularly as the natural product is generally somewhat moist. As regards the greater part of its mass, chalk is often found in a state of great purity, but it is frequently contaminated by the presence of scattered masses of flint. Chemically this impurity is not very objectionable to the glass-maker, since it merely introduces a small proportion of silica whose presence need scarcely be allowed for in laying down the mixture. On the other hand, if any fragments of flint remain in the mixture when put into the furnace, they prove very refractory, and are apt to be found as opaque enclosures in the finished glass. Natural limestone can also be obtained in great purity in many parts of the world. It is generally a hard and rather brittle rock that can be readily ground to powder of the requisite degree of fineness. Flint concretions are not so frequently found in this material, but, on the other hand, it is often contaminated with magnesia and iron. The former ingredient, when present in small quantities, tends to make the glass hard and viscous, so that limestone of the lowest possible magnesia content should be used, especially for the harder kinds of glass, such as plate and sheet-glass, etc. The iron contents of the limestone used must also be low where a white glass is required; but since a smaller quantity of limestone is used for a given weight of glass produced than the quantity of sand used for the same purpose, the presence of a somewhat higher percentage of iron is permissible in the limestone as compared with the sand; for the better varieties of glass, however, the iron should not exceed 0·3 per cent. of the limestone.

Slaked lime is sometimes used as the source of lime for special glasses where the process of manufacture renders it desirable to avoid the evolution of carbonic acid gas which takes place when the carbonate is heated and attacked by silica. When slaked lime is used only the water vapour of the hydrate is driven off, and this occurs at a much lower temperature. For the production of slaked lime, an adequately pure form of limestone, preferably in the form of large lumps, is burnt in a kiln until the carbonic acid is entirely driven off; after cooling, the lime so formed is slaked by hand. The product so obtained is, however, apt to vary both as regards contents of moisture and carbonic acid, which latter is readily absorbed from the atmosphere; the use of this material, therefore, requires frequent analytical determinations of the lime contents and corresponding adjustments of the mixture if constant results are required.

It is possible to introduce lime into glass mixtures in the form of gypsum or calcium sulphate, but the decomposition of this compound, like that of sodium sulphate, requires the intervention of a reducing agent such as carbon, and the difficulties arising from this source in connection with the use of salt-cake are still further increased in the case of the calcium compound. Since limestones of considerable purity are more or less plentiful in many districts, the commercial value of calcium sulphate for glass-making is probably slight.

The Compounds of Barium may best be dealt with at this stage, since they are chemically so closely allied to the compounds of lime just described. Barium occurs in nature in considerable quantities in the minerals known as barytes (heavy spar) and witherite respectively. The former is essentially sulphate of barium, while the latter is a carbonate of barium. The use of the sulphate meets with the same objection here as in the case of calcium sulphate discussed above, except that the barium compound is much more easily reduced and decomposed than the lime compound. The natural mineral witherite is used to a considerable extent in the production of barium glasses, and these have been found capable of replacing lead glasses for certain purposes. On the other hand, for the best kinds of barium glasses, viz., those required for optical purposes, the element is introduced in the form of artificially prepared salts. Of these the most important is the carbonate, commercially described as “precipitated carbonate of barium”; this precipitated compound, however, does not ordinarily correspond to the chemically pure substance, but contains more or less considerable quantities of sulphur compounds. The question whether these impurities are or are not objectionable can only be determined for each particular case, since much depends upon the special character of the glass to be produced. Both the nitrate and the hydrate of barium are commercially available, but they are very costly ingredients for use in the production of even the most expensive kinds of glass; these substances are, however, obtainable in a state of considerable purity, although the hydrate has the inconvenient property of rapidly absorbing carbonic acid from the atmosphere, thus becoming converted into the carbonate.

Magnesia is another glass-forming base that is closely related, chemically, to calcium and barium. This element is usually introduced into glass mixtures in the form of either the carbonate or the oxide. The carbonate occurs in nature in a more or less pure state in the form of magnesite, and by calcination, the oxide is obtained. The natural mineral and its product are, of course, by far the cheapest sources of magnesia, but as the element is only used in comparatively small quantities, the artificial precipitated carbonate or calcined magnesia are frequently preferred. Magnesia is only introduced intentionally in notable quantities in special glasses where the properties it confers are of special value; in ordinary lime glasses this element, as has already been mentioned, is to be regarded as an undesirable impurity.

Zinc oxide lies, chemically, between the bases already discussed on the one hand, and lead oxide on the other. This element is only introduced into special optical glasses, a special “zinc crown” having found some application. Chemically prepared zinc oxide is almost the only form in which the element is used, but the very volatile character of this substance must be borne in mind when it is introduced into glass mixtures.

Lead is one of the most widely-used ingredients of glass; the glasses containing this substance in notable quantity are all characterised to a greater or less degree by similar properties, such as considerable density and high refractive power, and are classed together under the name “flint glasses.” Lead is now almost universally introduced into glass mixtures in the form of red lead, although the other oxides of lead might be employed almost equally well. Red lead is a mixture of two oxides of lead (PbO and Pb₂O₃) in approximately such proportions as to correspond to the formula Pb₃O₄. It is prepared by the roasting of metallic lead in suitable furnaces, where the molten lead is exposed to currents of hot air. The product is obtainable in considerable purity, very small proportions of silica, derived from the furnace bed, and of iron derived from the tools with which the lead is handled, being the principal foreign substances found in good red lead. Silver would be an objectionable impurity, but owing to the modern perfect methods of de-silvering lead, that element is rarely found in lead products. Analytical control of red lead as used in the glass mixtures, and consequent adjustments of the mixture, are, however, necessary where exact constancy in the glass produced is desired. The reason for this necessity lies in the fact that the oxygen content, and therefore the lead-oxide (PbO) content, varies decidedly from batch to batch, while the material as actually delivered and used frequently contains notable proportions of moisture.

A word should perhaps be said here as to methods of handling red lead on account of the injurious effects which the inhalation of lead dust produces upon the workmen exposed to it. For glass-making purposes it is not feasible to adopt the method adopted by potters of first “fritting” the lead and thus rendering it comparatively insoluble and innocuous; even if this were done, the difficulty would only be moved one step further back, and would have to be overcome by those who undertook the preparation of the frit. The proper solution of the problem, in the writer’s opinion, is to be found in properly preventing the formation of lead dust, or at all events in protecting the workmen from the risk of inhaling it. Where only small quantities of lead glass are made, and therefore only small quantities of lead are handled and mixed at a time, it is no doubt sufficient to provide the workmen engaged on this task with some efficient form of respirator to be worn during the whole of the time that they are engaged on such work, and to take the further precautions necessary—by way of cleanliness and the provision of proper mess-rooms—to avoid any risk of lead dust either directly or indirectly contaminating their food. Where, however, large quantities of flint-glass are made every day, it is possible and proper to make more perfect arrangements for the mechanical handling and mixing of the lead with the other ingredients by the provision of suitable mixing and transporting machinery, so arranged as to be dust-tight. It is only fair to state, however, that partly under their own initiative, partly under pressure from the authorities, glass makers in this country are complying with these requirements in an adequate manner.

Aluminium.—There are several varieties of glass into which alumina enters in notable quantities, the principal examples being certain optical and many opal glasses, while most ordinary glasses contain this substance in greater or less degree. In the latter, the alumina is derived by the inevitable processes of solution, from the fire-clay vessels or walls within which the molten glass is contained, while in some cases the element is intentionally introduced in small proportions (about 2 per cent. to 3 per cent. of Al₂O₃) by the use of felspar as an ingredient of the mixture. Where larger proportions of alumina are required, the substance is introduced in the form of the hydrate, which is obtainable commercially in a state of almost chemical purity, but of course at a correspondingly high cost. In opal glasses alumina is derived partly or wholly from felspars, or in some cases from the use of the mineral cryolite. This is a double fluoride of aluminium and sodium which is found in great natural masses, chiefly in Greenland. Owing to the high price of this mineral, however, artificial substitutes of nearly identical composition and properties have been introduced and are used successfully in the glass and enamelling industries.

Manganese.—Although the oxides of this element really belong to the class of colouring compounds, they are so widely used in the manufacture of ordinary “white” glasses that it is desirable to deal with them here. The element manganese is most usually introduced into glass mixtures in the form of the per-oxide (MnO₂), although the lower oxide (Mn₃O₄) can also be used. The material ordinarily used is the natural manganese ore, mined chiefly in Russia; the purest forms of this ore consist almost entirely of the per-oxide, but “brown” ores, containing more or less of the lower oxide, are also used with success. These ores always contain small amounts of iron and silica, but provided the iron is not present in any considerable quantity, the value of the ore is measured by the percentage of manganese which it contains. The colouring and “decolourising” action of manganese will be discussed in a later chapter. Certain other substances, which have been suggested as either substitutes for, or improvements upon, manganese for this purpose need only be mentioned here, viz., nickel, selenium and gold.

Arsenic is another substance frequently introduced into “white” glass mixtures. This element is universally introduced in the form of the white arsenic of commerce (i.e., arsenious acid, As₂O₃) which is obtained in a pure form by a process of sublimation. Owing to the very poisonous nature of this material, special precautions must be taken in its use for glass-making purposes to avoid all risk of poisoning.

Carbon.—As has already been indicated, an admixture of carbon in some suitable form is essential in the case of certain glass mixtures. The carbon for this purpose may be used in the form of either charcoal, coke, or anthracite coal. Of these, charcoal is undoubtedly the purest form of carbon, but it is excessively expensive in this country. Coke varies very much in quality according to the coal from which it has been produced, but it always contains notable proportions of ash rich in iron, and also some sulphur. Anthracite coal can be obtained in a very pure form, containing considerably less ash than that found in most kinds of coke, and this is therefore probably the most convenient form of carbon for this purpose.