The arrangement of the mill is shown in plan in Fig. 33 and in elevation in Fig. 34. The seed or other material passes through the following course:—It runs from an upper floor through the roll frame A, by which it is crushed three or four times; it is then taken by the elevators B to the kettle C, where it is heated and damped. From beneath the kettle, it is drawn, in quantities sufficient to make a cake, by a box which conveys it to the moulding machine E. Here it undergoes preliminary compression, the objects of which are (1) to increase the number of cakes which may be inserted in the presses at one time, enabling 18 12-lb. cakes to be made where 4 8-lb. cakes were formerly made, and (2) to ensure uniform size and weight, and uniform density or consistence throughout.

The cakes are removed from the moulding machine, and put into the press F, 3 or 4 of which are required to each moulding machine. The pressure is applied either by means of hydraulic pumps, or by a high and low pressure accumulator; but unless extreme care is used with the latter, it gives too rapid a pressure, squirting out the seed at the side of the plates, and exercising a destructive effect upon the cloth employed. The pulsation caused by the pumps working directly to the press cylinder is more akin to the action of a wedge, and seems to extract the oil better than the dead pressure given by the accumulator. If the latter is used, a small cylinder may be applied to give the preliminary pressure in the moulding machine, in lieu of a cam. After remaining under pressure about 25 minutes, the cakes are withdrawn, and after being stripped of the cloth, are pared by the machine H, which completes the manufacture of the cakes. The parings fall under a very small pair of edge-running stones J, which automatically discharge them when sufficiently ground, into an elevator conducting to the kettle, where they are worked up with fresh seed. In a mill with 4 presses, 2 men and a boy in the press room can make 6 tons of cake in 11 hours, a rate of production requiring 6 men by the old process. The saving in steam power is about 30 per cent., chiefly due to the absence of the heavy edge runners, which also effects an economy of space. About 2 per cent. more oil is extracted, and the cakes are improved in appearance by not having the structureless texture caused by the trituration of the seed under edge-runners.

Having described the general routine of the process, some details may be added concerning the working of the several machines. The roll-frame, Fig. 35, consists of 4 or 5 chilled-iron rolls, each 3 ft. 6 in. long by 16 in. in diameter, placed one above the other. These rolls are used for crushing all the seed that passes through one set of presses, making 5½-6¼ tons linseed-cake per spell of 11 hours. The seed passes into the hopper in the usual manner, and is distributed to the crushing-rolls by a fluted feed roll the same length as the crushing-rolls, placed at the bottom of the hopper. When the seed passes the feed roll, it falls on a guide-plate that carries it between the 1st and 2nd roll. After passing between these rolls and being partly crushed, it falls on a guide-plate on the other side, which carries it back between the 2nd and 3rd rolls, where it is crushed more fully. It then falls on another guide-plate, which carries it between the 3rd and 4th rolls, where it is ground more fully. Then it falls on a 4th guide-plate, and is conveyed between the 4th and 5th rolls to receive the finishing touch. It is thus crushed four times.

The kettle is shown in Fig. 36, which represents one capable of heating sufficient seed to keep four 16-plate presses occupied, or to make 6 tons of cake per 11 hours. It is steam jacketed and furnished inside with a damping apparatus. The inside diameter is 5 ft., and the depth 2 ft. 6 in. The seed introduced is kept in motion by the stirring gear, and when sufficiently heated and damped, is withdrawn by the box A in quantities to form one cake, and transferred at once to the moulding machine, attached or separate.

This machine is illustrated in Figs. 37, 38. Its purpose is to measure the quantity of seed required to make each cake, to shape it as required, and to press it so much, without extracting any oil, as will enable the greatest number of cakes to be put into the press. The measure of seed is placed on a strip of woollen cloth, spread upon a thin iron tray, sliding on the guides B; the bottomless hinged mould C, having the exact shape of the intended cake, is closed upon it, and the measure A (Fig. 36), which is also bottomless, is drawn over guides in the upper surface of the mould C, thus accurately distributing the seed. The mould is next thrown upon its hinge (Fig. 37), and the ends of the strip of cloth are folded over the seed, the thickness of which is about 3½ in. The thin iron tray, with the mould of seed upon it, is then pushed along the guides B, beneath the die D. This action gives motion to a cam, shown above in the illustrations, but which may be placed beneath if necessary. This cam brings down the die and compresses the mould of seed to a thickness of 1¼ in.; its revolutions are so timed that the seed is under pressure long enough (about one-third of a minute) to let the workman have another cake ready.

When the die of the moulding-machine rises, the cake and tray are removed and placed in the press (Fig. 39), the tray being withdrawn. The plates of the press are slightly thickened towards the edges, and bear the name of the

manufacturer in reverse. The press is suitable for extracting oil from linseed, rape-seed, cotton-seed, hemp-seed, niger-seed, sunflower-seed, gingelly-seed, castor-seed, ground-nuts, coco-nuts, olives, &c. It is made in various sizes. The No. 1

double press (not shown) is furnished with 4 cake boxes, suitable for making 4 tapered cakes at one pressing, each about 2 ft. 5 in. long, by 10½ in. wide at one end, and 7½ in. at the other, when using linseed, 48 lb. of Bombay seed being required to charge the press, and giving a cake weighing about 8 lb.; the maximum and minimum weights of its charges are 60 lb. and 40 lb., of the cakes, 13 lb. and 6½ lb. The charges vary from 3 to 6 an hour, being 4 for cotton-seed and 5 for linseed; most other seeds are worked the same as linseed, but rape and gingelly are worked twice. By using 2 presses for the first time and 3 for the second, 3 presses will crush as much seed as 5. These presses are made of a capacity to take 270-320 lb. of seed at a charge, giving cakes of 9-15 lb., and requiring 30-45 minutes for the operation. In all these presses, the hair wrappers, weighing some 26 lb., used in the old process, are dispensed with.

A very complete account of oils and fats will be found in Spon’s ‘Encyclopædia of the Industrial Arts,’ to which the reader is referred for further information.

Dryers.—The maximum of drying power in oils is obtained by the addition of certain metallic oxides, which not only part with some of their own oxygen to the oil, but also act as carriers between the atmospheric oxygen and the heated liquid. This heating of the oil with oxides is known as boiling, although the liquid is not volatilised without decomposition, as is the case with water. At about 500° F., bubbles begin to rise in the oil, producing acrid, white fumes on coming into contact with the air. The gas thus given off consists chiefly of vapour of acrolein mingled with carbonic oxide. There is no advantage in heating the oil to a higher temperature than 350° F. Accurate experiments have shown that the drying properties of the oil are not increased by heating it beyond this point, while its colour is considerably darkened.

For the finer qualities of boiled oils, it is essential that the raw oil should have been stored for some time, so that it may be free from mucilage. This mucilage is the chief source of the dark colour of some boiled oils; when heated, it forms a brown substance, which is soluble in the oil itself, and extremely difficult to remove.

The oxides usually added to the oil during boiling are litharge or red-lead, the former being preferred on account of its lower price. About 2-5 per cent. by weight of the oxides or dryers is gradually stirred into the oil after it has been slowly raised to a temperature of about 300° F. The stirring should be continued until the litharge is dissolved, or it would cake on the bottom of the pan, and cause the oil to burn. Litharge may even be reduced to a cake of metallic lead when the fire is brisk. Some pans are furnished with stirrers and gearing by which the latter can be worked, either by hand or steam. The material of which the pans are made is either wrought or cast iron. Copper pans are sometimes used with the object of improving the colour of the oil.

Little is known respecting the chemical reactions which take place during the boiling of oil. Even when the air is excluded during the process, the drying properties are greatly increased, and, if boiled long enough, the oil is converted into a solid substance. The loss of weight which ensues is dependent upon the temperature and the time during which the operation continues. It is less when the air is freely admitted than if the pan is covered with a hood. The vapours given off by the oil are of an extremely irritating character, and should be destroyed by passing them through a furnace. As their mixture with air in certain proportions is explosive, this furnace should be situated at some distance, and the gases be conducted into it by means of an earthenware pipe.

Since zinc oxide has been introduced as a substitute for white lead in painting, researches have been made to replace litharge as a dryer, because it is not logical to discard the lead pigment and then use a lead dryer with a zinc pigment.

Several metallic oxides and salts, especially zinc sulphate, manganese oxide, and umber, have the property of combining with oils, which they render drying. To these may be added the protoxides of the metals of the third class, i. e. iron, cobalt, and tin. But these oxides are very unstable and difficult of preparation; hence it became desirable to discover some means by which they might be combined with bodies which would enable them to be prepared cheaply, and at the same time leave unimpaired their desiccating powers. Moreover, it is acknowledged that dryers in the dry state are preferable in many respects to drying oils. Following are some of the recently introduced dryers:—

Cobalt and Manganese Benzoates.—Benzoic acid is dissolved in boiling water, the liquid being continually stirred, and neutralised with cobalt carbonate until effervescence ceases. Excess of carbonate is removed by filtration, and the liquor is evaporated to dryness. The salt thus prepared is an amorphous, hard, brownish material, which may be powdered like rosin, and kept in the pulverulent state in any climate, simply folded in paper. Painting executed with a paint composed of 3 parts of this dryer with 1000 of oil and 1200 of zinc-white, dries in 18 to 20 hours. Manganese benzoate is prepared in the same way, substituting manganese carbonate for that of cobalt. Applied under similar circumstances, it dries a little more rapidly, and a little less is required. Urobenzoic (hippuric) acid is equally efficacious.

Cobalt and Manganese Borates.—These salts also, in the same proportions, are found to be of equal efficacy. The latter is extremely active, and requires to be used in much smaller proportions.

Resinates.—If an alkaline resinate of potash or soda be dissolved in hot water, and this solution be precipitated by a solution of a proportionate quantity of a cobalt or manganese chloride or sulphate, an amorphous resinate is formed, which, after being collected on cloth filters, washed, and dried, forms an excellent drier.

Zumatic (Transparent) Dryer.—Take zinc carbonate, 90 lb.; manganese borate, 10 lb.; linseed-oil, 90 lb. Grind thoroughly, and keep in bladders or tin tubes; the latter are preferable.

Zumatic (Opaque) Dryer.—Manganese borate, as a dryer, is so energetic that it is proper to reduce its action in the following way:—Take zinc-white, 25 lb.; manganese borate, 1 lb. Mix thoroughly, first by hand, then in a revolving drum; 1 lb. of this mixed with 20 lb. paint ensures rapid drying.

Manganese Oxide.—Purified linseed-oil is boiled for 6 or 8 hours, and to every 100 lb. boiled oil are added 5 lb. of powdered manganese peroxide, which may be kept suspended in a bag, like litharge. The liquid is boiled and stirred for 5 or 6 hours more, and then cooled and filtered. This drying oil is employed in the proportion of 5 to 10 per cent. of the zinc white.

Guynemer’s.—Take pure manganese sulphate, 1 part; manganese acetate, 1 part; calcined zinc sulphate, 1 part; white zinc oxide, 97 parts. Grind the sulphates and acetate to impalpable powder, sift through a metallic sieve. Dust 3 parts of this powder over 97 of zinc oxide, spread out over a slab or board, thoroughly mix, and grind. The resulting white powder, mixed in the proportion of ½ or 1 per cent. with zinc-white, will enormously increase the drying property of this body, which will become dry in 10 or 12 hours.

Manganese Oxalate.—A writer in the Moniteur de Produits Chimiques draws attention to the properties possessed by manganese oxalate as a drier. This salt has hitherto not had any important industrial uses, but it can be readily prepared in a state of purity from the native carbonate by the action of oxalic acid; the author is of the opinion that it will be found of use for this purpose. If prepared from carbonate free from iron and lime, it can be obtained as a fine crystalline white powder, and two-fifths per cent. suffices to bring about the change. The oxalate is resolved by heat into manganese oxide, carbonic acid and carbon monoxide, and in the presence of fatty acids the manganese oxide formed combines with them, the decomposition taking place at about 130°. The operation is carried out by mixing in a mortar the oxalate with two or three times its weight of oil, and then adding the mixture to the main portion of the oil. The heat should be applied gradually, and the decomposition is known to be complete when there is no further evolution of gas. The boiled oil, under this treatment, preserves its limpidity and also remains colourless. Manganese oxalate has the advantage over oxide of lead, which is commonly employed for this purpose, in causing the oil to remain transparent when exposed to sulphur vapours. Manganese acetate has also been used, but it likewise causes a darkening in the colour of the oil, and the nitrate is dangerous owing to the possible action of nitric acid on the fats present in the oil. Manganese borate appears to be next in value to the oxalate as an oil drier.

In a paper recently read before the Society of Arts, Prof. Hartley remarked that paint, such as is used for ordinary purposes, is essentially composed of three materials, without taking into account the coloured pigments.

(1) White lead, or sublimed zinc-white.

(2) An oil, generally linseed or poppy oil, which is ground up with the white lead or zinc-white until it becomes a soft paste. This is mixed with variable preparations of linseed oil and spirit of turpentine.

(3) A substance called dryers, or siccative materials; it may be linseed oil in which litharge is dissolved, or it may be linseed oil containing a compound of manganese.

Paint owes to the dryers its property of drying more rapidly than it would do without it; and it is considered indispensable in buildings in all cases where paint applied to wood, stone, or metal, would not be quite dry in 48 hours, or at most in 72 hours, after the first application.

The first question which requires an answer is, what chemical process takes place when a paint dries.

Boiled Oil.—Linseed oil absorbs oxygen; and, when the oil contains manganese, it absorbs oxygen much more greedily; and when a manganese oil—that is to say, a boiled oil containing manganese—is mixed with linseed oil, the substance absorbs oxygen, from a limited supply of air contained in a closed space, until no trace of any other gas but nitrogen remains. The power of absorbing oxygen possessed by 100 volumes of linseed oil, compared with that of 100 volumes of a mixture of linseed oil and so-called manganese oil, is as 9·4 to 100. This may be termed the measure of its drying power. A mixture of linseed oil, with a little more than one-fourth of its volume of manganese oil, has a power of absorbing oxygen four and a half times greater than either of the components of the mixture taken separately. In this case Chevreul argues that linseed oil may be considered as a “dryer” to manganese oil.

Linseed oil, without any addition whatever, if boiled for three hours, becomes a better drying oil than it was previous to the action of heat.

Oil boiled with 10 per cent, of litharge for three hours, is a much better dryer than when heated without this oxide.

Oil boiled alone for five hours is an inferior drying oil to one heated for only three hours.

Oil previously boiled alone for five hours, and boiled alone again for three hours, is scarcely altered in drying power, but it becomes a better drying oil if it is boiled the second time with litharge. It is inferior to a drying oil which has been boiled only three hours with litharge, without being submitted to a previous boiling.

Oil boiled alone for five hours, boiled for a further period of three hours with manganese dioxide which has already been used for one operation, is very nearly as strong a dryer as that which has been boiled with litharge under the same conditions; but it is superior to an oil which has been boiled with manganese dioxide for eight hours. This no doubt arises from the longer boiling with manganese having caused a larger quantity of manganese to dissolve, and that the quantity dissolved is in excess of that which yields the best result.

Finally, oil boiled for five hours, and then boiled alone once more for eight hours, becomes viscous, and the first coat requires a considerable time to dry. We thus see that the oxides of lead and of manganese in certain proportions concur with heat in increasing the drying power of linseed oil. This drying of oils is a process of slow oxidation.

The following points of Chevreul’s appeared to be difficult of satisfactory explanation, and suggested to Prof. Hartley an examination de novo of the facts, as well as an investigation of the chemistry of the subject generally:—

1. Linseed oil not boiled acted as a dryer to the same oil boiled with manganese dioxide.

2. Linseed oil, boiled with either litharge or manganese dried more rapidly when mixed with turpentine.

3. Oil, mixed with white lead, zinc white, antimony white, and arseniate of tin, acts differently, thus:—The white lead dries most rapidly, the zinc white next, but antimony white and arseniate of tin are incapable of acting as dryers, in fact, they retard the drying process.

4. Oil boiled alone for five hours, and boiled for a further period of three hours with manganese dioxide becomes a superior drying oil to one which has been boiled with manganese dioxide for eight hours.

From a series of experiments, which were continued for two years, on twenty-five weighed quantities of raw linseed oil, Prof. Hartley draws the following deductions:—

1. The chemical action of a manganese compound, when dissolved in linseed oil, is that of a carrier of oxygen from the atmosphere to the oil. Manganese oxide takes up oxygen from the air, and transfers it to the oil, and in so doing it suffers alternately the opposite processes of oxidation and reduction.

2. To obtain the best result, the amount of manganese present must not exceed a certain small proportion of the oil.

3. Oil to which turpentine has been added dries more rapidly than oil without such addition, because the oil being diluted and rendered thinner, it spreads over a larger surface, and is in contact, therefore, with a much larger quantity of oxygen.

4. Turpentine does not act as a dryer, that is, as a carrier of oxygen to linseed oil.

5. Different white pigments behave differently when drying, because the more powerfully basic the properties of the pigment, the more powerful is its action as a dryer. Lead oxide and white lead (basic lead carbonate) combine more easily with the acids of linseed oil than zinc oxide does. But zinc oxide dries better than antimony oxide, because it is a stronger base, while arseniate of tin has no basic properties, therefore does not act as a dryer.

Different substances, that is to say, those without chemical action on oil, such as lamp-black, sulphate of baryta, and sulphate of lead, cannot act as “dryers.”

Linseed oil is a glyceride of a peculiar acid, called linoleic acid. Whatever the exact constitution of linoleic acid may be, linseed oil for the most part is composed of trilinolein. Raw linseed oil contains the following constituents:—

1. Glyceride of linoleic acid or trilinolein

2. Water.

3. Mucilage, with the composition n(C₆H₁₀O₅). On boiling with dilute acids this yields a gum and a sugar.

4. An essential oil, present in minute proportions, and of unknown composition.

5. A mixture of colouring matters of intense tinctorial power, viz. blue and yellow chlorophylls and erythrophyll.

The only useful and desirable substance is the trilinolein.

The effect of oxidation upon linseed oil is to destroy all the glycerine, and to produce therefrom carbonic, formic, and acetic acids, together with some acrolein. When boiled at a high temperature without the addition of any metallic oxide, the glyceride is decomposed, acrolein is formed, and linoleic acid is set free. In fact, whether oil is oxidised by air or by metallic oxides, or whether it be simply heated, the action in each case first leads to the destruction of the glycerine and the liberation of linoleic acid. But linoleic acid very readily absorbs oxygen, and the oxidised substance becomes a tough elastic solid, which is essentially a varnish.

In fact, the process which an oil undergoes in drying is not desiccation, or depriving it of moisture or of glycerine, but solidification, and the technical term “drying” is a misnomer. That, however, is of little consequence if we really know what is the chemical action of the “drying” process. When oxidised even at a low temperature, the glycerine is destroyed, and the oxidised products form a tough varnish.

There are various methods of converting linseed oil into a drying oil or varnish:—

1. Heating it to a high temperature with litharge.

2. Heating with red oxide of lead.

3. Heating with metallic lead.

4. Heating to a high temperature with manganic oxide.

5. Heating with manganese borate.

6. Heating with manganese oxalate.

7. By the joint action of air and heat upon the oil and manganous oxide, or a solution of manganese dioxide or manganous oxide in the oil.

In the processes 1, 2, 3, there can be no doubt that a lead of linoleic acid is produced, and that this facilitates further oxidation in air, by forming salts with some of the acid products of such oxidation, while the oxidation of the linoleic acid continues. Heating with red lead favours oxidation, by the compound itself conveying oxygen to the oil. In the case of metallic lead, it must be noted that the metal is dissolved. Under certain circumstances, metals become dryers to oils; thus sheets of metallic lead are capable of acting as dryers to linseed oil.

Linseed oil is pre-eminent in its capacity for absorbing oxygen. This action of metallic lead as a dryer is due to the metal becoming oxidised at the expense of the glycerine of the oil, and so passing into solution by combining with the linoleic acid, or with acetic or formic acid, caused by the oxidation of the glycerine. It is the destruction of the glycerine with concurrent oxidation of the fatty acid which causes the drying or hardening of the oil.

When a drying oil which has been treated with metallic lead, or with litharge, is shaken up with a solution of zinc sulphate, all the lead is precipitated from the oil, and zinc passes into solution therein. By manganese sulphate or copper sulphate, the lead is removed by manganese or copper. Oil charged with lead dries in 24 hours when spread out in a thin layer on glass; it will dry completely in 5 or 6 hours if charged with manganese, in 30 or 36 hours with copper, zinc, or cobalt; and it requires more than 48 hours with nickel, iron, chromium, &c.

Although solidification of a drying oil charged with manganese takes place in from 5 to 6 hours when spread in thin films, the solidification of thicker films requires a longer time. A temperature of 122° to 140° F. accelerates the oxidation of the drying oils, partly because the oil becomes more fluid, and partly because the oxygen is more active at a higher temperature. Hence, oil which has been mixed with an equal volume of turpentine, or a light hydrocarbon, such as benzene, dries more rapidly than oil without such admixture.

When a boiled oil, prepared with manganese, is dissolved in an equal volume of benzene, and shaken up with air in a bottle, rapid absorption of oxygen occurs, especially about 120° F. If fresh air is repeatedly provided, the oxidation is sufficient to cause the liquid to become thick, and, on distilling off the sapient, a perfectly dry and elastic solid remains.

An oil containing manganese is a very superior drying oil to one which has been prepared with lead. This fact, however, is to be noted, that though a large proportion of manganese in an oil may hasten its drying, yet it is disadvantageous, because it does not form so tough a film. This arises from the film becoming hard upon the surface, and so protecting the oil underneath from absorbing oxygen from the air.

Though the oils containing large quantities of dryers dry, they afterwards lose weight, and become viscous under the same conditions.

Pure linoleates of lead and of zinc are not dryers; but if heated until it has turned brown, or begun to blacken, a lead dryer becomes effective, although it contains less of the lead compound.

In this case, some compound of lead is formed by absorption of oxygen, which either itself actually oxidises or causes the oxidation of ordinary linseed oil.

Having treated of the materials used for producing boiled oil, and of their action upon the oil, let us now consider how the operation is brought about.

Process 1.—Oil is boiled at a high temperature, that is to say, it is heated until frothing and bubbles of gas escape, when litharge or a manganese compound is added.

Process 2.—Oil is boiled at a steam heat, with litharge or a manganese compound, in conjunction with a blast of air.

Process 1.—The chemical action in the first process is doubtless one which takes place in three stages. It commences by depriving the oil of water; in the second stage, it destroys the mucilage, by charring it; in the third, it destroys, in part, the glycerine, and sets free the fatty acids. After the litharge or manganese compound is added, there is formed in the oil a solution of lead salts of the fatty acids, or a manganese salt of the fatty acids.

The oil then, at the high temperature, loses glycerine by oxidation caused by the air, such oxidation being greatly facilitated by the presence of manganese compounds, which are repeatedly oxidised by the air and reduced by the oil, that is to say, they absorb oxygen and pass it over to the oil with great facility.

It matters little, so long as the ultimate action is oxidation, what salt of manganese or what oxide is used, if it be capable of undergoing processes of an alternate character called oxidations and reductions.

It is, however, certain that some manganese compounds are more suitable than others, owing to their more or less complete solubility in the oil, and their more readily undergoing the two different processes of oxidation and reduction in presence of air and of oil.

Process 2.—The credit of being the first to boil oil without resorting to the dangerous expedient of using an open fire and a high, temperature in the manufacture, is due to Vincent. He used manganese compounds, or both manganese salts and litharge. His method of boiling oil for the manufacture of printing inks is, with some modifications in technical details, carried out on a large scale at the present time in the preparation of ordinary boiled oil. The essential parts of the plant are a steam-jacketed close boiler with agitating gear, and a pipe for conducting a current of air into the oil by means of a blowing engine. From the head of the boiler there passes a funnel under the back of the furnace fire, by which the disagreeable products of the chemical action are conducted to a place where they are destroyed. These products, as already mentioned, are volatile fatty acids and acrolein.

Oil boiling, as ordinarily carried out, is conducted by means of litharge along with compounds of manganese; in some processes these are mixed with salts of alumina and zinc. The oil so produced is brown and not clear, but it is clarified by keeping. Many samples of such boiled oil deposit insoluble matter when stored for some time, even although they may have become clear previously. This is not a desirable property. Sometimes rosin is added to hasten its drying.

The defects to be noticed, even in the best samples of boiled oil, are the following:—

1. The oil causes a brownish or yellow colour to be communicated to white lead or zinc white.

2. The oil darkens pigments containing brilliantly coloured metallic sulphides, such as vermilion, cadmium yellow, and ultramarine blue.

3. Delicate colours are darkened by the oil when exposed to ordinary town air, that is to say, air which, is not quite pure. This is the case even when the oils themselves may not injure the paints.

The causes of such alterations is, in nine cases out of ten, the use of lead dryers.

1. In the first place, boiled oil which contains litharge or other lead compounds takes a permanent brown colour, which affects the purity of white lead, zinc white, and delicate pale tints.

2. Lead forms, with extreme ease, lead sulphide, which, in very minute proportions, is yellow or brown; in larger quantity its colour is black. The lead sulphide is readily formed by contact with other sulphides, as, for instance, vermilion, cadmium yellow, and ultramarine.

3. Boiled oil, containing lead, is coloured brown by exposure to air, owing to the presence of minute quantities of sulphuretted hydrogen, which causes the formation of lead sulphide.

The remedy is obvious: no oil should be used which has been boiled with dryers containing lead. In other words, oil should be boiled with pure manganese compounds only.

In cases where it is desirable to have information of the presence or absence of lead in a boiled oil, the following test will be found most useful:—A mixture is made of 4 oz. of glycerine with 1 oz. of ammonium sulphide, the liquid being kept in a stoppered bottle. Or glycerine is mixed with an equal volume of water, and saturated with sulphuretted hydrogen. Half an ounce of the oil to be tested is placed in a white basin, with the addition of two or three drops of the glycerine solution. The two liquids are thoroughly incorporated, by stirring with a strip of glass. A brown or black colour, which gradually appears, indicates the presence of lead. A pure manganese oil simply becomes slightly yellow. It is true that, if iron is present, a black colour might appear, but iron is also an undesirable impurity. Should it be required to ascertain that the coloration is or is not caused by iron, two or three drops of glacial acetic acid may be stirred into the oil, when, if the black colour remains, it is certainly not caused by iron.

Under the old process of oil-boiling at a high temperature, the brown colour of the oil was, to some extent, an indication that the oil had been sufficiently heated—that is to say, properly boiled; but in the modern processes, so largely used, in which oxidation is aided by a blast of air, this coloration is no indication whatever of the excellence of the oil; it may be, in fact, the very reverse.

This fact appears to be unknown, or, at any rate, is not a matter of common knowledge among practical men in this country, who, being uninformed as to the methods of preparing the oils, consider that a brown colour is desirable, if not essential.

When oil-boilers were compelled to adopt some expedient to give a reddish-brown colour to the oil, they added a small amount of litharge, the introduction of which actually spoils the oil, and makes it unsuitable for many purposes to which it is otherwise applicable. Of late years, pale boiled oils have been more largely manufactured for special purposes. It is obvious that, for decorative house painting, in which delicate tints are a leading feature, they may be advantageously employed.

Notwithstanding that some of the brown oils, when mixed with white lead, do not entirely retain the brownish tint, but, to some extent, lose it upon drying, yet they never preserve the whiteness of white lead. It follows, therefore, that a pale colour in the oil, provided it is not the yellow colour of raw oil, is greatly to be preferred. Moreover, when paints are mixed with zinc white, no trace of lead should be contained in the oil, otherwise, one of the valuable properties of zinc white pigments is destroyed, namely, its power to retain its whiteness in the atmosphere of a town, because its colour is not affected by sulphuretted hydrogen.

Very generally, zinc white and white lead paints are not mixed with drying oils, but with refined linseed or bleached oil. This, at any rate, is the practice on the Continent. That is to say, the pigments are mixed with an oil from which the impurities, and the natural yellow and red colouring matter, have been removed, so that the colour of the paint is white. If ordinary oil be used, the paint is more or less yellow. In order to render such paint quick drying, a certain amount of dryers, in a solid or liquid form, is added. These dryers almost invariably contain lead, so that zinc white paint is contaminated by lead in another way, which may not be suspected, or which is overlooked.

Now as to the chemical action of dryers on oils. Raw oil contains water and mucilage; the former can be absorbed by anhydrous zinc salts and by dried alum, and solutions of the salts and the salts themselves are capable of precipitating mucilage from the oil; hence these substances cause the impurities to become insoluble, so that they are carried down as “foots.” Heat greatly facilitates this action, particularly by causing the oil to become more fluid; and by the action of the anhydrous salts water is withdrawn from the oil. On the “drying” or oxidation of the oil, they exert no chemical action whatever.

Zinc linoleate and lead linoleate do not act as dryers when simply added to the oil. Though the former is soluble in hot oil it is insoluble in cold oil, and it therefore separates from the oil as it cools. The latter is very soluble in linseed oil, but only adds to its drying power when heated therewith.

In conjunction with a high temperature, lead dissolves in oil at the expense of the glycerine, which is decomposed into acrolein, while lead linoleate is formed.

When litharge is heated with linseed oil, the action is somewhat similar, the substances formed being acrolein, lead linoleate, and linoleic acid.

If we consider the action of red lead on trilinolein, we have not only the formation of these lead linoleates, but an excess of oxygen available for the oxidation of glycerine to acrolein and acrylic acid, or to acetic and formic acids.

These equations serve to show the effect of lead and lead oxides in what may be termed the initiation of the chemical action upon the oil. Subsequent changes, no doubt, depend upon the conditions which obtain at the time, notably upon the temperature and upon access of air to the oil. It is probable that acid linoleates are formed, and that compounds formed from the polymerisation of linoleic acid result eventually.

Whatever doubt there may be as to the action of lead salts, there can be none whatever as to that of manganese compounds. In the first place, manganous oxide is a powerful base, which readily dissolves in oil; manganic oxide is also readily soluble, yielding fatty acid salts of manganese, and causing oxidation of glycerine. Manganese borate and manganese oxalate are both soluble in oil, the former much more readily than the latter, but they are both salts of little stability at high temperatures in contact with oils. They both dissolve by the aid of heat, forming fatty acid salts of manganese. Borate liberates boric acid under these circumstances, but oxalate yields a mixture of carbon monoxide and carbon dioxide.

Of manganese oleate and linoleate nothing more may be said than that both are extremely soluble in oil, and both easily oxidised from colourless to brown compounds when submitted to the action of air.

The chief adulterants of linseed oil and of boiled oil are cotton-seed oil, rosin oil, and linoleic acid. Cotton-seed, which is to some extent a drying oil, can act as such when mixed with linseed, but when added to olive oil, it behaves as a non-drying oil. In fact, its behaviour is anomalous, and of such a character that it greatly facilitates its extensive use as an adulterating material for the more expensive oils.

Rosin oil is a deleterious adulterant, but one which may be more readily detected than cotton-seed oil. Rosin is added to boiled oil to hasten its drying; this also is an injurious substance. Of late years glycerine has become an article of greater value than formerly, and this may account for the manufacture of linoleic acid and its use as an adulterant of oleic acid and of linseed oil.

Lastly, it may be mentioned that certain samples of “pale boiled oil” have been found to contain what is practically a raw oil mixed with dryers. Although such oils will dry, their efficiency is nothing like so great as that of an oil “boiled” with a blast of air at a suitable temperature, and, moreover, such oils are deficient in body.

In bleaching vegetable oils, it is necessary to consider the nature of the colouring matters naturally contained in them. These consist of a mixture in varying proportions of the colouring matters known to exist in the leaves of plants, but which, in the case of oils, are derived from the fruit or seeds from which the oils are expressed. There can be no doubt that these substances are closely allied in chemical constitution; they all possess an intensely powerful colouring property, by which is meant that though the colour of some of them may not be dark, yet a very minute weight is capable of imparting a tint to a very large quantity of material.

The names of these substances are:—

In some oils only the xanthophyll and yellow chlorophyll are present; in others, such as olive oil, the yellow and blue chlorophylls occur, and give the liquid a green tint; while in linseed erythrophyll is always present with more or less of the yellow and blue chlorophylls, and some xanthophyll. According to the different proportions of these colouring matters the oil varies in colour. For instance, linseed oil, when brown, contains a mixture of erythrophyll with yellow and blue chlorophylls; when greenish brown, the yellow and blue chlorophylls are present in somewhat larger proportion, but mixed with erythrophyll; while, generally speaking, a bright yellow or pale yellow oil contains xanthophyll only. These substances appear to be combined with the oils, or to be substances of a fatty nature. They are neither dissolved nor acted upon by water, nor by acids diluted with water, when naturally contained in the oils. They are freely soluble in alcohol, and an alcoholic solution is not only susceptible of being destroyed by the joint action of air and water, but by very dilute aqueous solutions of mineral acids, and by acetic acid. In aqueous and alcoholic solutions, light speedily modifies the blue, and eventually destroys all these colours. A solution in turpentine of the isolated colouring matters is also easily destroyed. But, on the other hand, a solution of the colours in melted paraffin wax is comparatively stable.

Zinc hydroxide, copper hydroxide, baryta, potash, and soda easily combine to form metallic salts with blue chlorophyll, less readily, though readily enough with yellow chlorophyll, but far less readily with xanthophyll and erythrophyll. The following facts will serve to show that this is the case. When a solution of the colouring matters contained in green leaves is made by extracting dry, but freshly-gathered, leaves with absolute alcohol, an addition of a saturated solution of baryta water, to the intensely green extract, precipitates first the compound of blue chlorophyll with baryta, then a further addition precipitates the yellow chlorophyll, also as a baryta salt; but xanthophyll and erythrophyll either remain in solution, or require a much larger addition of the base in order to be precipitated. A crystalline compound of blue chlorophyll with soda is comparatively stable. This substance is, no doubt, formed in green vegetables when they are boiled in water to which some carbonate of soda has been added to maintain their fresh appearance. The addition of a small trace of copper sulphate to peas and to pickles forms a very permanent copper compound with the colouring matter, which gives an attractive appearance to these articles. Such being an outline of the chief chemical properties of the natural colouring matters contained in oils, the facts mentioned will serve to render the processes for removing the colour from oils more intelligible than they otherwise would be.

Vegetable oils are decolorised, either partially or completely, by the application of one of the following processes:—

1. By the action of light, or by the joint action of light and air.

2. By acids.

3. By saponification.

4. By the action of chlorine.

1. By exposing raw linseed oil to the action of sunlight, it slowly becomes pale in colour, and finally colourless. It is in the highest degree probable that, as oxygen is absorbed by the oil and acid substances are thereby produced, these acids effect the destruction of the colouring matters. In such wise castor oil is bleached.

2. By treating linseed oil with moderately strong sulphuric acid. As the oil and sulphuric acid are of very different specific gravities, it is essential that they be very rapidly and thoroughly mixed by violent agitation. The impurities, such as mucilage and albuminous matters, are thus deprived of water, and more or less charred, and along with them the colouring matters are destroyed by the acid. It is essential for the success of the process that the oil and the acid be not long in contact without undergoing dilution, otherwise the oil itself may become charred. It is, however, possible to obtain oil by this process in a fairly colourless condition, after it has been thoroughly washed with water and allowed to settle.

3. Both rape oil and cotton oil may be rendered of a pale yellow, and even almost colourless, by a process of partial saponification with caustic alkali of a suitable strength. The colouring matters are saponified, and the resulting soap is of a dark yellow or brown colour, from the colouring matter having combined with the alkali.

4. By the action of chlorine produced in contact with the oil when, for instance, an aqueous solution of bleaching powder is acidified with a cheap mineral acid, such as dilute sulphuric. In this case rapid mixing and violent agitation are essential to the success of the process, otherwise chlorinised products are retained in the oil, which not only confer upon it a distinct flavour and odour, but also cause the oil to solidify with a very moderate lowering of the normal temperature. It is very questionable whether drying oils can with advantage be submitted to such treatment.

5. A variety of methods may be merely mentioned, such as treatment with sulphurous acid, with ferrous sulphate (green vitriol), and potassium dichromate and sulphuric acid.

6. Lastly, the method of Binks, to which reference will be made farther on.

Prof. Hartley next gives an account of certain improvements in the process of oil-boiling, designed with the object of producing a drying oil absolutely free from lead, and, as compared with ordinary oils, absolutely free from colour.

The operations have been carried out, on a manufacturing scale, by Mr. W. E. B. Blenkinsop and himself, and there is no doubt of the practicability of the process.

The process consists in, first, refining the oil, by the removal therefrom of water and mucilage; second, boiling and bleaching the oil at one operation.

It is a fact that water and mucilage can be removed from linseed oil by the action of certain dehydrating substances and solutions of metallic salts, as, for instance, by alum, by strong sulphuric acid, and by a solution of zinc chloride.

There are certain objections to each of these methods, which are of a practical nature: thus, in treating the oil with strong sulphuric acid, there is too frequently a charring of something, either the oil itself, or of some impurity therein, and this charring, though it may be very slight, has the effect of giving a pale brownish tinge to the oil, which cannot be completely removed by the bleaching process to which the natural colouring matters in the oil are amenable. It is quite true that this brown colour separates sometimes, but it is only after storage for a long period, when a finely divided flocculent matter separates by subsidence. Treatment with zinc chloride is satisfactory but expensive. Perfectly pure manganese sulphate, which is a neutral salt, has been used by Hartley and Blenkinsop in very strong solution, and where there is an objection to using an acid. For ordinary purposes, perfectly satisfactory results are obtained by the use of a dilute sulphuric acid containing about 30 per cent. of H₂SO₄, since, though it possesses the power of withdrawing water from the oil, it may remain in contact therewith without causing any charring, and at the same time it causes the precipitation in a complete and rapid manner of all the mucilage. A purified linseed oil is thus produced which is bright, clear, and slightly yellowish in colour, though somewhat paler than the ordinary oil. It is important that the strength of the oil should not exceed that degree of concentration which is sufficient for the purpose for which it is intended. The oil having been so treated, and the impurities separated by subsidence or otherwise, it is next submitted to the bleaching and oxidising treatment.

Binks bleached oils with oxides of manganese dissolved in the oil, but difficulty was experienced in carefully regulating the quantity of the manganese compounds which were to be introduced into the oil. For instance, he precipitated manganous hydroxide in contact with oil, and added the mixture to the bulk of the material, and he also modified the treatment by dissolving manganous hydroxide in ammonia, and added the solution to the oil.

Hartley and Blenkinsop prepare manganese linoleate, and dissolve this in a hydrocarbon, and add a sufficient quantity of the solution to the oil, whereby it dissolves easily and mixes completely. By this treatment, the colouring matter of the oil forms a compound with the manganese which, while it remains in solution, is very speedily oxidised in contact with air, especially when a current of air or oxygen is blown through. The oxidation destroys the colouring matter, and the manganese compound is deoxidised; subsequently it undergoes oxidation again, and the products of such oxidation taking place in the oil are acrolein, formic and acetic acids. After, or concurrently with, the oxidation of the colouring matters, the oil is oxidised, and, at a suitable temperature below 132° F., the oil is bleached, increased in density, and converted into a pale drying oil. By limiting the amount of the manganese linoleate to that which is capable of just oxidising the colouring matters, oils may be bleached with very little further oxidation.

Excellent drying oils have been produced by this process, of a very pale colour. The oil has been used for decorative house painting, for both indoor and outdoor work, on wood and on metal. It has also been used as a coating for iron work, without the addition of a pigment. The plant used in its production is the same as that employed in oil-boiling by the usual processes when a blast of air is used.

The advantages of a pale boiled oil, containing no lead, are the following:—

1. Zinc white retains its pure white colour.

2. Delicate tints, and colours containing sulphides, are not darkened in course of time.

It may be suggested that for indoor decoration, for the painting of ships, railway carriages, railway semaphores, signs, and stations, such oil is free from liability to alter the colours with which it is mixed, owing to its freedom from lead, which is darkened by traces of sulphuretted hydrogen in the air to which such paints are exposed.

Gasometers in gas-works may be painted an unalterable white with such oil and zinc white. But in this case also the zinc white must be free from lead carbonate or oxide.

In commenting on Prof. Hartley’s paper, Mr. Laurie said he had never used linoleate of manganese for boiling with oil, but by the use of borate one did get a boiled oil paler than the oil with which one started. If you take linseed oil which has been already bleached in the sun to a golden yellow, and convert it into boiled oil with manganese, a further bleaching process undoubtedly takes place. An oil prepared with manganese salts, spread on a glass plate, and allowed to dry in the dark, will remain almost colourless, whereas if it were boiled with a lead salt it quickly darkens, even if it is kept away from impure air. Even in a dark room, in pure air, a picture painted with oil boiled with lead will darken. That is another argument in favour of manganese, and he should say it ought always to be used in preparing oil for artistic purposes.

CHAPTER XIII.

PAINT MACHINERY.

In Fig. 40 is shown a complete set of paint grinding and mixing machinery, made by Wright & Co., 157 Southwark Bridge Road, London, S.E., which has given highly satisfactory results in efficiency and economy. It has a set of three granite rollers 30 inches by 15, and two mixing cylinders or pugs, 24 inches in diameter by 25 inches deep, the whole mounted on a strong cast-iron frame. It is made in the following five sizes:—

The utility of having the pug mills placed above the granite rollers is to save labour and space, and the roller mill is kept constantly at work. There are two pugs, one of which is always ready to deliver to the rollers, whilst the other is getting ready by the time the first is being emptied; by this means the output is always going on, and hence great saving both of time and labour. The pugs are