Chapter III: NITRO-CELLULOSE, &c (1)

Cellulose Properties--Discovery of Gun-Cotton--Properties of Gun-Cotton-- Varieties of Soluble and Insoluble Gun-Cottons--Manufacture of Gun-Cotton-- Dipping and Steeping--Whirling out the Acid--Washing--Boiling--Pulping-- Compressing--The Waltham Abbey Process--Le Bouchet Process--Granulation of Gun-Cotton--Collodion-Cotton--Manufacture--Acid Mixture used--Cotton used, &c.--Nitrated Gun-Cotton--Tonite--Dangers in Manufacture of Gun-Cotton-- Trench's Fire-Extinguishing Compound--Uses of Collodion-Cotton--Celluloid-- Manufacture, &c.--Nitro-Starch, Nitro-Jute, and Nitro-Mannite.

~The Nitro-Celluloses.~--The substance known as cellulose forms the groundwork of vegetable tissues. The cellulose of the woody parts of plants was at one time supposed to be a distinct body, and was called lignine, but they are now regarded as identical. The formula of cellulose is (C_{6}H_{10}O_{6})_{X}, and it is generally assumed that the molecular formula must be represented by a multiple of the empirical formula, C_{12}H_{20}O_{10} being often regarded as the minimum. The assumption is based on the existence of a penta-nitrate and the insoluble and colloidal nature of cellulose. Green (_Zeit. Farb. Text. Ind._, 1904, 3, 97) considers these reasons insufficient, and prefers to employ the single formula C_{6}H_{10}O_{5}. Cellulose can be extracted in the pure state, from young and tender portions of plants by first crushing them, to rupture the cells, and then extracting with dilute hydrochloric acid, water, alcohol, and ether in succession, until none of these solvents remove anything more. Fine paper or cotton wool yield very nearly pure cellulose by similar treatment.

Cellulose is a colourless, transparent mass, absolutely insoluble in water, alcohol, or ether. It is, however, soluble in a solution of cuprammoniac solution, prepared from basic carbonate or hydrate of copper and aqueous ammonia. The specific gravity of cellulose is 1.25 to 1.45. According to Schulze, its elementary composition is expressed by the percentage numbers:--

Carbon 44.0 per cent. 44.2 per cent. Hydrogen 6.3 " 6.4 " Oxygen 49.7 " 49.4 "

These numbers represent the composition of the ash free cellulose. Nearly all forms of cellulose, however, contain a small proportion of mineral matters, and the union of these with the organic portion of the fibre or tissue is of such a nature that the ash left on ignition preserves the form of the original. "It is only in the growing point of certain young shoots that the cellulose tissue is free from mineral constituents" (Hofmeister).

Cellulose is a very inert body. Cold concentrated sulphuric acid causes it to swell up, and finally dissolves it, forming a viscous solution. Hydrochloric acid has little or no action, but nitric acid has, and forms a series of bodies known as nitrates or nitro-celluloses. Cellulose has some of the properties of alcohols, among them the power of forming ethereal salts with acids. When cellulose in any form, such as cotton, is brought into contact with strong nitric acid at a low temperature, a nitrate or nitro product, containing nitryl, or the NO_{2} group, is produced. The more or less complete replacement of the hydroxylic hydrogen by NO_{2} groups depends partly on the concentration of the nitric acid used, partly on the duration of the action. If the most concentrated nitric and sulphuric acids are employed, and the action allowed to proceed for some considerable time, the highest nitrate, known as hexa-nitro- cellulose or gun-cotton, C_{12}H_{14}O_{4}(O.NO_{2})_{6}, will be formed; but with weaker acids, and a shorter exposure to their action, the tetra and penta and lower nitrates will be formed.[A]

[Footnote A: The paper by Prof. Lunge, _Jour. Amer. Chem. Soc._, 1901, 23[8], 527-579, contains valuable information on this subject.]

Besides the nitrate, A. Luck[A] has proposed to use other esters of cellulose, such as the acetate, benzoate, or butyrate. It is found that cellulose acetate forms with nitro-glycerine a gelatinous body without requiring the addition of a solvent. A sporting powder is proposed composed of 75 parts of cellulose nitrate (13 per cent. N.) mixed with 13 parts of cellulose acetate.

[Footnote A: Eng. Pat. 24,662, 22nd November 1898.]

The discovery of gun-cotton is generally attributed to Schoenbein (1846), but Braconnot (in 1832) had previously nitrated starch, and six years later Pelouse prepared nitro-cotton and various other nitro bodies, and Dumas nitrated paper, but Schoenbein was apparently the first chemist to use a mixture of strong nitric and sulphuric acids. Many chemists, such as Piobert in France, Morin in Russia, and Abel in England, studied the subject; but it was in Austria, under the auspices of Baron Von Lenk, that the greatest progress was made. Lenk used cotton in the form of yarn, made up into hanks, which he first washed in a solution of potash, and then with water, and after drying dipped them in the acids. The acid mixture used consisted of 3 parts by weight of sulphuric to 1 part of nitric acid, and were prepared some time before use. The cotton was dipped one skein at a time, stirred for a few minutes, pressed out, steeped, and excess of acid removed by washing with water, then with dilute potash, and finally with water. Von Lenk's process was used in England at Faversham (Messrs Hall's Works), but was given up on account of an explosion (1847).

Sir Frederick Abel, working at Stowmarket and Waltham Abbey, introduced several very important improvements into the process, the chief among these being pulping. Having traced the cause of its instability to the presence of substances caused by the action of the nitric acid on the resinous or fatty substances contained in the cotton fibre, he succeeded in eliminating them, by boiling the nitro-cotton in water, and by a thorough washing, after pulping the cotton in poachers.

Although gun-cottons are generally spoken of as nitro-celluloses, they are more correctly described as cellulose nitrates, for unlike nitro bodies of other series, they do not yield, or have not yet done so, amido bodies, on reduction with nascent hydrogen.[A] The equation of the formation of gun-cotton is as follows:--

2(C_{6}H_{10}O_{5}) + 6HNO_{3} = C_{12}H_{14}O_{4}(NO_{3})_{6} + 6OH_{2}.
Cellulose. Nitric Acid. Gun-Cotton. Water.

The sulphuric acid used does not take part in the reaction, but its presence is absolutely essential to combine with the water set free, and thus to prevent the weakening of the nitric acid. The acid mixture used at Waltham Abbey consists of 3 parts by weight of sulphuric acid of 1.84 specific gravity, and 1 part of nitric acid of 1.52 specific gravity. The same mixture is also used at Stowmarket (the New Explosive Company's Works). The use of weaker acids results in the formation of collodion- cotton and the lower nitrates generally.

[Footnote A: "Cellulose," by Cross and Bevan, ed. by W.R. Hodgkinson, p. 9.]

The nitrate which goes under the name of gun-cotton is generally supposed to be the hexa-nitrate, and to contain 14.14 per cent. of nitrogen; but a higher percentage than 13.7 has not been obtained from any sample. It is almost impossible (at any rate upon the manufacturing scale) to make pure hexa-nitro-cellulose or gun-cotton; it is certain to contain several per cents. of the soluble forms, i.e., lower nitrates. It often contains as much as 15 or 16 per cent., and only from 13.07[A] to 13.6 per cent. of nitrogen.

[Footnote A: Mr J.J. Sayers, in evidence before the court in the "Cordite Case," says he found 15.2 and 16.1 per cent. soluble cotton, and 13.07 and 13.08 per cent. nitrogen in two samples of Waltham Abbey gun-cotton.]

A whole series of nitrates of cellulose are supposed to exist, the highest member being the hexa-nitrate, and the lowest the mono-nitrate. Gun-cotton was at one time regarded as the tri-nitrate, and collodion-cotton as the di-nitrate and mono-nitrate, their respective formula being given as follows:--

Mono-nitro-cellulose C_{6}H_{9}(NO_{2})O_{5} = 6.763 per cent. nitrogen. Di-nitro-cellulose C_{6}H_{8}(NO_{2})_{2}O_{5} = 11.11 " " Tri-nitro-cellulose C_{6}H_{7}(NO_{2})_{3}O_{5} = 14.14 " "

But gun-cotton is now regarded as the hexa-nitrate, and collodion-cotton as a mixture of all the other nitrates. In fact, chemists are now more inclined to divide nitro-cellulose into the soluble and insoluble forms, the reason being that it is quite easy to make a nitro-cellulose entirely soluble in a mixture of ether-alcohol, and yet containing as high a percentage of nitrogen as 12.6; whereas the di-nitrate[A] should theoretically only contain 11.11 per cent. On the other hand, it is not possible to make gun-cotton with a higher percentage of nitrogen than about 13.7, even when it does not contain any nitro-cotton that is soluble in ether-alcohol.[B] The fact is that it is not at present possible to make a nitro-cellulose which shall be either entirely soluble or entirely insoluble, or which will contain the theoretical content of nitrogen to suit any of the above formulae for the cellulose nitrates. Prof. G. Lunge gives the following list of nitration products of cellulose:--

[Footnote A: The penta-nitrate C_{12}H_{15}O_{5}(NO_{3})_{5} = 12.75 per cent. nitrogen.]

[Footnote B: In the Cordite Trial (1894) Sir F.A. Abel said, "Before 1888 there was a broad distinction between soluble and insoluble nitro- cellulose, collodion-cotton being soluble (in ether-alcohol) and gun-cotton insoluble." Sir H.E. Roscoe, "That he had been unable to make a nitro-cotton with a higher nitrogen content than 13.7." And Professor G. Lunge said, "Gun-cotton always contained soluble cotton, and _vice versa_." These opinions were also generally confirmed by Sir E. Frankland, Sir W. Crookes, Dr Armstrong, and others.]

Dodeca-nitro-cellulose C_{24}H_{28}O_{20}(NO_{2})_{12} = 14.16 per cent.
nitrogen. (= old tri-nitro-cellulose)
Endeca-nitro-cellulose C_{24}H_{29}O_{20}(NO_{2})_{11} = 13.50 per cent.
nitrogen.
Deca-nitro-cellulose C_{24}H_{30}O_{20}(NO_{2})_{10} = 12.78 per cent.
nitrogen.
Ennea-nitro-cellulose C_{24}H_{31}O_{20}(NO_{2})_{9} = 11.98 per cent.
nitrogen.
Octo-nitro-cellulose C_{24}H_{32}O_{20}(NO_{2})_{8} = 11.13 per cent.
nitrogen. (= old di-nitro-cellulose)
Hepta-nitro-cellulose C_{24}H_{33}O_{20}(NO_{2})_{7} = 10.19 per cent.
nitrogen.
Hexa-nitro-cellulose C_{24}H_{34}O_{20}(NO_{2})_{6} = 9.17 per cent.
nitrogen.
Penta-nitro-cellulose C_{24}H_{35}O_{20}(NO_{2})_{5} = 8.04 per cent.
nitrogen.
Tetra-nitro-cellulose C_{24}H_{36}O_{20}(NO_{2})_{4} = 6.77 per cent.
nitrogen. (= old mono-nitro-cellulose)

It is not unlikely that a long series of nitrates exists. It is at any rate certain that whatever strength of acids may be used, and whatever temperature or other conditions may be present during the nitration, that the product formed always consists of a mixture of the soluble and insoluble nitro-cellulose.

Theoretically 100 parts of cotton by weight should produce 218.4 parts of gun-cotton, but in practice the yield is a good deal less, both in the case of gun-cotton or collodion-cotton. In speaking of soluble and insoluble nitro-cellulose, it is their behaviour, when treated with a solution consisting of 2 parts ether and 1 of alcohol, that is referred to. There is, however, another very important difference, and that is their different solubility in nitro-glycerine. The lower nitrates or soluble form is soluble in nitro-glycerine under the influence of heat, a temperature of about 50 deg. C. being required. At lower temperatures the dissolution is very imperfect indeed; and after the materials have been left in contact for days, the threads of the cotton can still be distinguished. The insoluble form or gun-cotton is entirely _insoluble_ in nitro-glycerine. It can, however, be made to dissolve[A] by the aid of acetone or acetic ether. Both or rather all the forms of nitro-cellulose can be dissolved in acetone or acetic ether. They also dissolve in concentrated sulphuric acid, and the penta-nitrate in nitric acid at about 80 deg. or 90 deg. C.

[Footnote A: Or rather to form a transparent jelly.]

The penta-nitrate may be obtained in a pure state by the following process, devised by Eder:--The gun-cotton is dissolved in concentrated nitric acid at 90 deg. C., and reprecipitated by the addition of concentrated sulphuric acid. After cooling to 0 deg. C., and mixing with a larger volume of water, the precipitated nitrate is washed with water, then with alcohol, dissolved in ether-alcohol, and again precipitated with water, when it is obtained pure. This nitrate is soluble in ether-alcohol, and slightly in acetic acid, easily in acetone, acetic ether, and methyl-alcohol, insoluble in alcohol. Strong potash (KOH) solution converts into the di-nitrate C_{12}H_{18}O_{8}(NO_{3})_{2}. The hexa-nitrate is not soluble in acetic acid or methyl-alcohol.

The lower nitrates known as the tetra- and tri-nitrates are formed together when cellulose is treated with a mixture of weak acids, and allowed to remain in contact with them for a very short time (twenty minutes). They cannot be separated from one another, as they all dissolve equally in ether-alcohol, acetic ether, acetic acid, methyl-alcohol, acetone, amyl acetate, &c.

As far as the manufacture of explosive bodies is concerned, the two forms of nitro-cellulose used and manufactured are gun-cotton or the hexa- nitrate (once regarded as tri-nitro-cellulose), which is also known as insoluble gun-cotton, and the soluble form of gun-cotton, which is also known as collodion, and consists of a mixture of several of the lower nitrates. It is probable that it chiefly consists, however, of the next highest nitrate to gun-cotton, as the theoretical percentage of nitrogen for this body,. the penta-nitrate, is 12.75 per cent., and analyses of commercial collodion-cotton, entirely soluble in ether-alcohol, often give as high a percentage as 12.6.

We shall only describe the manufacture of the two forms known as soluble and insoluble, and shall refer to them under their better known names of gun-cotton and collodion-cotton. The following would, however, be the formulae[A] and percentage of nitrogen of the complete series:--

Hexa-nitro-cellulose C_{12}H_{14}O_{4}(NO_{3})_{6} 14.14 per cent.
nitrogen.
Penta-nitro-cellulose C_{12}H_{15}O_{5}(NO_{3})_{5} 12.75 per cent.
nitrogen.
Tetra-nitro-cellulose C_{12}H_{16}O_{6}(NO_{3})_{4} 11.11 per cent.
nitrogen.
Tri-nitro-cellulose C_{12}H_{17}O_{7}(NO_{3})_{3} 9.13 per cent.
nitrogen.
Di-nitro-cellulose C_{12}H_{18}O_{8}(NO_{3})_{2} 7.65 per cent.
nitrogen.
Mono-nitrocellulose C_{12}H_{19}O_{9}(NO_{3}) 3.80 per cent.
nitrogen.

[Footnote A: Berthelot takes C_{24}H_{40}O_{20} as the formula of cellulose; and M. Vieille regards the highest nitrate as (C_{24}H_{18}(NO_{3}H)_{11}O_{9}). _Compt. Rend._, 1882, p. 132.]

~Properties of Gun-Cotton.~--The absolute density of gun-cotton is 1.5. When in lumps its apparent density is 0.1; if twisted into thread, 0.25; when subjected, in the form of pulp, to hydraulic pressure, 1.0 to 1.4. Gun-cotton preserves the appearance of the cotton from which it is made. It is, however, harsher to the touch; it is only slightly hygroscopic (dry gun-cotton absorbs 2 per cent. of moisture from the air). It possesses the property of becoming electrified by friction. It is soluble in acetic ether, amyl acetate, and acetone, insoluble in water, alcohol, ether, ether-alcohol, methyl-alcohol, &c. It is very explosive, and is ignited by contact with an ignited body, or by shock, or when it is raised to a temperature of 172 deg. C. It burns with a yellowish flame, almost without smoke, and leaves little or no residue. The volume of the gases formed is large, and consists of carbonic acid, carbonic oxide, nitrogen, and water gas. Compressed gun-cotton when ignited often explodes when previously heated to 100 deg. C.

Gun-cotton kept at 80 deg. to 100 deg. C. decomposes slowly, and sunlight causes it to undergo a slow decomposition. It can, however, be preserved for years without undergoing any alteration. It is very susceptible to explosions by influence. For instance, a torpedo, even placed at a long distance, may explode a line of torpedoes charged with gun-cotton. The velocity of the propagation of the explosion in metallic tubes filled with pulverised gun-cotton has been found to be from 5,000 to 6,000 mms. per second in tin tubes, and 4,000 in leaden tubes (Sebert).

Gun-cotton loosely exposed in the open air burns eight times as quickly as powder (Piobert). A thin disc of gun-cotton may be fired into from a rifle without explosion; but if the thickness of the disc be increased, an explosion may occur. The effect of gun-cotton in mines is very nearly the same as that of dynamite for equal weights. It requires, however, a stronger detonator, and it gives rise to a larger quantity of carbonic oxide gas. Gun-cotton should be neutral to litmus, and should stand the Government heat test--temperature of 150 deg. F. for fifteen minutes (see page 249). In the French Navy gun-cotton is submitted to a heat test of 65 deg. C. (= 149 deg. F.) for eleven minutes. It should contain as small a percentage of soluble nitro-cotton and of non-nitrated cotton as possible.

The products of perfectly detonated gun-cotton may be expressed by the following equation:--

2C_{12}H_{14}O_{4}(NO_{3})_{6} = 18CO + 6CO_{2} + 14H_{2}O + 12N.

It does not therefore contain sufficient oxygen for the complete combustion of its carbon. It is for this reason that when used for mining purposes a nitrate is generally added to supply this defect (as, for instance, in tonite). It tends also to prevent the evolution of the poisonous gas, carbonic oxide. The success of the various gelatine explosives is due to this fact, viz., that the nitro-glycerine has an excess of oxygen, and the nitro-cotton too little, and thus the two explosives help one another.

In practice the gases resulting from the explosion of gun-cotton are-- Carbonic oxide, 28.55; carbonic acid, 19.11; marsh gas (CH_{4}), 11.17; nitric oxide, 8.83; nitrogen, 8.56; water vapour, 21.93 per cent. The late Mr E.O. Brown, of Woolwich Arsenal, discovered that perfectly wet and uninflammable compressed gun-cotton could be easily detonated by the detonation of a priming charge of the dry material in contact with it. This rendered the use of gun-cotton very much safer for use as a military or mining explosive.

As a mining explosive, however, gun-cotton is now chiefly used under the form of tonite, which is a mixture of half gun-cotton and half barium nitrate. This material is sometimes spoken of as "nitrated gun-cotton." The weight of gun-cotton required to produce an equal effect either in heavy ordnance or in small arms is to the weight of gunpowder in the proportion of 1 to 3, i.e., an equal weight of gun-cotton would produce three times the effect of gunpowder. Its rapidity of combustion, however, requires to be modified for use in firearms. Hence the lower nitrates are generally used, or such compounds as nitro-lignose, nitrated wood, &c., are used.

The initial pressure produced by the explosion of gun-cotton is very large, equal to 18,135 atmospheres, and 8,740 kilogrammes per square centimetre for 1 kilo., the heat liberated being 1,075 calories (water liquid), or 997.7 cals. (water gaseous), but the quantity of heat liberated changes with the equation of decomposition. According to Berthelot,[A] the heat of formation of collodion-cotton is 696 cals. for 1,053 grms., or 661 cals. for 1 kilo. The heat liberated in the total combustion of gun-cotton by free oxygen at constant pressure is 2,633 cals. for 1,143 grms., or for 1 kilo. gun-cotton 2,302 cals. (water liquid), or 2,177 cals. (water gaseous). The heat of decomposition of gun- cotton in a closed vessel, found by experiment at a low density of charge (0.023), amounts to 1,071 cals. for 1 kilo. of the substance, dry and free from ash. To obtain the maximum effect of gun-cotton it must be used in a compressed state, for the initial pressures are thereby increased. Wet gun-cotton s much less sensitive to shock than dry. Paraffin also reduces its liability to explode, so also does camphor.

[Footnote A: "Explosives and their Power," trans. by Hake and M'Nab.]

The substance known as celluloid, a variety of nitro-cellulose nearly corresponding to the formula C_{24}H_{24}(NO_{3}H)_{8}O_{12}, to which camphor and various inert substances are added, so as to render it non-sensitive to shock, may be worked with tools, and turned in the lathe in the same manner as ivory, instead of which material celluloid is now largely used for such articles as knife handles, combs, &c. Celluloid is very plastic when heated towards 150 deg. C., and tends to become very sensitive to shock, and in large quantities might become explosive during a fire, owing to the general heating of the mass, and the consequent evaporation of the camphor. When kept in the air bath at 135 deg. C., celluloid decomposes quickly. In an experiment (made by M. Berthelot) in a closed vessel at 135 deg. C., and the density of the charge being 0.4, it ended in exploding, developing a pressure of 3,000 kilos. A large package of celluloid combs also exploded in the guard's van on one of the German railways a few years ago. Although it is not an explosive under ordinary circumstances, or even with a powerful detonator, considerable care should be exercised in its manufacture.

~The Manufacture of Gun-Cotton.~--The method used for the manufacture of gun-cotton is that of Abel (Spec. No. 1102, 20. 4. 65). It was worked out chiefly at Stowmarket[A] and Waltham Abbey,[B] but has in the course of time undergone several alterations. These modifications have taken place, however, chiefly upon the Continent, and relate more to the apparatus and machinery used than to any alteration in the process itself. The form of cellulose used is cotton-waste,[C] which consists of the clippings and waste material from cotton mills. After it has been cleaned and purified from grease, oil, and other fatty substances by treatment with alkaline solutions, it is carefully picked over, and every piece of coloured cotton rag or string carefully removed. The next operation to which it is submitted has for its object the opening up of the material. For this purpose it is put through a carding machine, and afterwards through a cutting machine, whereby it is reduced to a state suitable for its subsequent treatment with acids, that is, it has been cut into short lengths, and the fibres opened up and separated from one another.

[Footnote A: The New Explosive Co. Works.]

[Footnote B: Royal Gunpowder Factory.]

[Footnote C: Costs from L10 to L25 a ton. In his description of the "Preparation of Cotton-waste for the Manufacture of Smokeless Powder," A. Hertzog states that the German military authorities require a cotton which when thrown into water sinks in two minutes; when nitrated, does not disintegrate; when treated with ether, yields only 0.9 per cent. of fat; and containing only traces of chlorine, lime, magnesia, iron, sulphuric acid, and phosphoric acid. If the cotton is very greasy, it must be first boiled with soda-lye under pressure, washed, bleached with chlorine, washed, treated with sulphuric acid or HCl, again washed, centrifugated, and dried; if very greasy indeed a preliminary treatment with lime-water is desirable. See also "Inspection of Cotton-Waste for Use in the Manufacture of Gun-cotton," by C.E. Munro, _Jour. Am. Chem. Soc._, 1895, 17, 783.]

~Drying the Cotton.~--This operation is performed in either of two ways. The cotton may either be placed upon shelves in a drying house, through which a current of hot air circulates, or dried in steam-jacketed cylinders. It is very essential that the cotton should be as dry as possible before dipping in the acids, especially if a wholly "insoluble" nitro-cellulose is to be obtained. After drying it should not contain more than 0.5 per cent. of moisture, and less than this if possible. The more general method of drying the cotton is in steam-jacketed tubes, i.e., double cylinders of iron, some 5 feet long and 1-1/2 foot wide. The cotton is placed in the central chamber (Fig. 10), while steam is made to circulate in the surrounding jacket, and keeps the whole cylinder at a high temperature (steam pipes may be coiled round the outside of an iron tube, and will answer equally well). By means of a pipe which communicates with a compressed air reservoir, a current of air enters at the bottom, and finds its way up through the cotton, and helps to remove the moisture that it contains. The raw cotton generally contains about 10 per cent. of moisture and should be dried until it contains only 1/2 per cent. or less. For this it will generally have to remain in the drying cylinder for about five hours. At the end of that time a sample should be taken from the _top_ of the cylinder, and dried in the water oven (100 deg. C.[A]) for an hour to an hour and a half, and re-weighed, and the moisture then remaining in it calculated.

[Footnote A: It is dried at 180 deg. C. at Waltham Abbey, in a specially constructed drying chamber.]

It is very convenient to have a large copper water oven, containing a lot of small separate compartments, large enough to hold about a handful of the cotton, and each compartment numbered, and corresponding to one of the drying cylinders. The whole apparatus should be fixed against the wall of the laboratory, and may be heated by bringing a small steam pipe from the boiler-house. It is useful to have a series of copper trays, about 3 inches by 6 inches, numbered to correspond to the divisions in the steam oven, and exactly fitting them. These trays can then be taken by a boy to the drying cylinders, and a handful of the cotton from each placed in them, and afterwards brought to the laboratory and weighed (a boy can do this very well), placed in their respective divisions of the oven, and left for one to one and a half hours, and re-weighed.

When the cotton is found to be dry the bottom of the drying cylinder is removed, and the cotton pushed out from the top by means of a piece of flat wood fixed on a broom-handle. It is then packed away in galvanised- iron air-tight cases, and is ready for the next operation. At some works the cotton is dried upon shelves in a drying house through which hot air circulates, the shelves being of canvas or of brass wire netting. The hot air must pass under the shelves and through the cotton, or the process will be a very slow one.

~Dipping and Steeping.~--The dry cotton has now to be nitrated. This is done by dipping it into a mixture of nitric and sulphuric acids. The acids used must be strong, that is, the nitric acid must be at least of a gravity of 1.53 to 1.52, and should contain as little nitric oxide as possible. The sulphuric acid must have a specific gravity of 1.84 at 15 deg. C., and contain about 97 per cent. of the mono-hydrate (H_{2}SO_{4}). In fact, the strongest acids obtainable should be used when the product required is gun-cotton, i.e., the highest nitrate.

The sulphuric acid takes no part in the chemical reaction involved, but is necessary in order to combine with the water that is liberated in the reaction, and thus to maintain the strength of the nitric acid. The reaction which takes place is the following:--

2(C_{6}H_{10}O_{5}) + 6HNO_{3} = C_{12}H_{14}(NO_{3})_{6} + 6 H_{2}O.
324 378 = 594 108.
Cellulose. Gun-Cotton.

Theoretically,[A] therefore, 1 part of cellulose should form 1.8 part of gun-cotton. Practically, however, this is never obtained, and 1.6 lb. from 1 lb. of cellulose is very good working. The mixture of acids used is generally 1 to 3, or 25 per cent. nitric acid to 75 per cent. sulphuric acid.

[Footnote A: (594 x 1)/324= 1.83.]

The dipping is done in cast-iron tanks (Fig. 11), a series of which is arranged in a row, and cooled by a stream of cold water flowing round them. The tanks hold about 12 gallons, and the cotton is dipped in portions of 1 lb. at a time. It is thrown into the acids, and the workman moves it about for about three minutes with an iron rabble. At the end of that time he lifts it up on to an iron grating, just above the acids, fixed at the back of the tank, where by means of a movable lever he gently squeezes it, until it contains about ten times its weight of acids (the 1 lb. weighs 10 lbs.). It is then transferred to earthenware pots to steep.

~Steeping.~--The nitrated cotton, when withdrawn from the dipping tanks, and still containing an excess of acids, is put into earthenware pots of the shape shown in Figs. 12 and 13. The lid is put on, and the pots placed in rows in large cooling pits, about a foot deep, through which a stream of water is constantly flowing. These pits form the floor of the steeping house. The cotton remains in these pots for a period of forty-eight hours, and must be kept cool. Between 18 deg. and 19 deg. C. is the highest temperature desirable, but the cooler the pots are kept the better. At the end of forty-eight hours the chemical reaction is complete, and the cotton is or should be wholly converted into nitro-cellulose; that is, there should be no unnitrated cotton.

~Whirling Out the Acid.~--The next operation is to remove the excess of acid. This is done by placing the contents of two or three or more pots into a centrifugal hydro-extractor (Fig. 14), making 1,000 to 1,500 revolutions per minute. The hydro-extractor consists of a machine with both an inner cylinder and an outer one, both revolving in concert and driving outwardly the liquid to the chamber, from which it runs away by a discharge pipe. The wet cotton is placed around the inner cone. The cotton, when dry, is removed, and at once thrown into a large tank of water, and the waste acids are collected in a tank.[A]

[Footnote A: Care must be taken in hot weather that the gun-cotton does not fire, as it does sometimes, directly the workman goes to remove it after the machine is stopped. It occurs more often in damp weather. Dr Schuepphaus, of Brooklyn, U.S.A., proposes to treat the waste acids from the nitration of cellulose by adding to them sulphuric anhydride and nitric acid. The sulphuric anhydride added converts the water liberated from the cellulose into sulphuric acid.]

~Washing.~--The cotton has now to be carefully washed. This is done in a large wooden tank filled with water. If, however, a river or canal runs through the works, a series of wooden tanks, the sides and bottoms of which are pierced with holes, so as to allow of the free circulation of water, should be sunk into a wooden platform that overhangs the surface of the river in such a way that the tanks are immersed in the water, and of course always full. During the time that the cotton is in the water a workman turns it over constantly with a wooden paddle. A stream of water, in the form of a cascade, should be allowed to fall into these tanks. The cotton may then be thrown on to this stream of water, which, falling some height, at once carries the cotton beneath the surface of the water. This proceeding is necessary because the cotton still retains a large excess of strong acids, and when mixed with water gives rise to considerable heat, especially if mixed slowly with water. After the cotton has been well washed, it is again wrung out in a centrifugal machine, and afterwards allowed to steep in water for some time.

~Boiling.~--The washed cotton is put into large iron boilers with plenty of water, and boiled for some time at 100 deg. C. In some works lead-lined tanks are used, into which a steam pipe is led. The soluble impurities of unstable character, to which Sir F.A. Abel traced the liability of gun- cotton to instability, are thereby removed. These impurities consist of the products formed by the action of nitric acid on the fatty and resinous substances contained in the cotton fibres. The water in the tanks should be every now and again renewed, and after the first few boilings the water should be tested with litmus paper until they are no longer found to be acid.

~Pulping.~--The idea of pulping is also due to Abel. By its means a very much more uniform material is obtained. The process is carried out in an apparatus known as a "Beater" or "Hollander" (Fig. 15, _a, b_). It consists of a kind of wooden tank some 2 or 3 feet deep of an oblong shape, in which a wheel carrying a series of knives is made to revolve, the floor of the tank being sloped up so as to almost touch the revolving wheels. This part of the floor, known as the "craw," is a solid piece of oak, and a box of knives is fixed into it, against which the knives in the revolving wheel are pressed. The beater is divided into two parts--the working side, in which the cotton is cut and torn between the knife edges in the revolving cylinder and those in the box; and the running side, into which the cotton passes after passing under the cylinder. The wheel is generally boxed in to prevent the cotton from being thrown out during its revolution. The cotton is thus in constant motion, continually travelling round, and passing between the knives in the revolving cylinder and those in the box fixed in the wooden block beneath it. The beater is kept full of water, and the cotton is gradually reduced to a condition of pulp. The wheel revolves at the rate of 100 to 150 times a minute.

When the gun-cotton is judged to be sufficiently fine, the contents of the beater are run into another very similar piece of machinery, known as the "poacher" (Fig. 16, _a, b, c_), in which the gun-cotton is continuously agitated together with a large quantity of water, which can be easily run off and replaced as often as required. When the material is first run into the poacher from the beater, the water with which it is then mixed is first run away and clean water added. The paddle wheel is then set in motion, and at intervals fresh water is added. There is a strainer at the bottom of the poacher which enables the water to be drawn off without disturbing the cotton pulp. After the gun-cotton has been in the poacher for some time, a sample should be taken by holding a rather large mesh sieve in the current for a minute or so. The pulp will thus partly pass through and partly be caught upon the sieve, and an average sample will be thus obtained. The sample is squeezed out by hand, bottled, and taken to the laboratory to be tested by the heat test for purity. It first, however, requires to be dried. This is best done by placing the sample between coarse filter paper, and then putting it under a hand-screw press, where it can be subjected to a tolerably severe pressure for about three minutes. It is then rubbed up very finely with the hands, and placed upon a paper tray, about 6 inches by 4-1/2 inches, which is then placed inside a water oven upon a shelf of coarse wire gauze, the temperature of the oven being kept as near as possible to 120 deg. F. (49 deg. C.), the gauze shelves in the oven being kept about 3 inches apart. The sample is allowed to remain at rest for fifteen minutes in the oven, the door of which is left wide open. After the lapse of fifteen minutes the tray is removed and exposed to the air of the laboratory (away from acid fumes) for two hours, the sample being at some point within that time rubbed upon the tray with the hand, in order to reduce it to a fine and uniform state of division. Twenty grains (1.296 grm.) are used for the test. (See Heat Test, page 249.)

If the gun-cotton sample removed from the poacher stands the heat test satisfactorily, the machine is stopped, and the water drained off. The cotton is allowed some little time to drain, and is then dug out by means of wooden spades, and is then ready for pressing. The poachers hold about 2,000 lbs. of material, and as this represents the products of many hundred distinct nitrating operations, a very uniform mixture is obtained. Two per cent. of carbonate of soda is sometimes added, but it is not really necessary if the cotton has been properly washed.

~Compressing Gun-Cotton.~--The gun-cotton, in the state in which it is removed from the poacher, contains from 28 to 30 per cent. of water. In order to remove this, the cotton has to be compressed by hydraulic power. The dry compressed gun-cotton is packed in boxes containing 2,500 lbs. of dry material. In order to ascertain how much of the wet cotton must be put into the press, it is necessary to determine the percentage of water. This may be done by drying 2,000 grains upon a paper tray (previously dried at 100 deg. C.) in the water oven at 100 deg. C. for three hours, and re-weighing and calculating the percentage of water. It is then easy to calculate how much of the wet gun-cotton must be placed in the hopper of the press in order to obtain a block of compressed cotton of the required weight. Various forms of presses are used, and gun-cotton is sent out either as solid blocks, compressed discs, or in the form of an almost dry powder, in zinc- lined, air-tight cases. The discs are often soaked in water after compression until they have absorbed 25 per cent. of moisture.

At the New Explosives Company's Stowmarket Works large solid blocks of gun-cotton are pressed up under a new process, whereby blocks of gun- cotton, for use in submarine mines or in torpedo warheads, are produced. Large charges of compressed gun-cotton have hitherto been built up from a number of suitably shaped charges of small dimensions (Fig. 17), as it has been impossible to compress large charges in a proper manner. The formation of large-sized blocks of gun-cotton was the invention of Mr A. Hollings. Prior to the introduction of this method, 8 or 9 lbs. had been the limit of weight for a block. This process has been perfected at the Stowmarket factory, where blocks varying from the armour-piercing shell charge of a few ounces up to blocks of compressed gun-cotton mechanically true, weighing 4 to 5 cwts. for torpedoes or submarine mines, are now produced. At the same time the new process ensures a uniform density throughout the block, and permits of any required density, from 1.4 downwards, being attained; it is also possible exactly to regulate the percentage of moisture, and to ensure its uniform distribution. The maximum percentage of moisture depends, of course, upon the density. By the methods of compression gun-cotton blocks hitherto employed, blocks of a greater thickness than 2 inches, or of a greater weight than 9 lbs., could not be made, but with the new process blocks of any shape, size, thickness, or weight that is likely to be required can be made readily and safely. The advantages which are claimed for the process may be enumerated as follows:--(1.) There is no space wasted, as in the case with built-up charges, through slightly imperfect contact between the individual blocks, and thus either a heavier charge--i.e., about 15 per cent. more gun- cotton--can be got into the same space, or less space will be occupied by a charge of a given weight. (2.) The metallic cases for solid charges may be much lighter than for those built-up, since with the former their function is merely to prevent the loss of moisture from wet gun-cotton, or to prevent the absorption of moisture by dry gun-cotton. They can thus be made lighter, as the solid charge inside will prevent deformation during transport. With built-up charges the case must be strong enough to prevent damage, either to itself or to the charge it contains. For many uses a metal case, however light, may be discarded, and one of a thin waterproof material substituted. (3.) The uniform density of charges made by this process is very favourable to the complete and effective detonation of the entire mass, and to the presence of the uniform amount of moisture in every part of the charge. (4.) Any required density, from the maximum downwards, may be obtained with ease, and any required amount of moisture left in the charge. These points are of great importance in cases where, like torpedo charges, it is essential to have the centre of gravity of the charge in a predetermined position both vertically and longitudinally, and the charge so fixed in its containing case that the centre of gravity cannot shift. The difficulty of ensuring this with a large torpedo charge built up from a number of discs and segments is well known. Even with plain cylindrical or prismatic charges a marked saving in the process of production is effected by this new system. The charges being in one block they are more easily handled for the usual periodical examination, and they do not break or chafe at the edges, as in the case of discs and cubes in built-up charges. A general view of the press is given in Fig. 19. The gun-cotton in a container is placed on a cradle fixed at an angle to the press. The mould is swivelled round, and the charge pushed into it with a rammer, and it is then swivelled back into position. The mould is made up of a number of wedge pieces which close circumferentially on the enclosed mass, which is also subjected to end pressure. Holes are provided for the escape of water.

~The Waltham Abbey Process.~--At the Royal Gunpowder Factory, Waltham Abbey, the manufacture of gun-cotton has been carried out for many years. The process used differs but little from that used at Stowmarket. The cotton used is of a good quality, it is sorted and picked over to remove foreign matters, &c., and is then cut up by a kind of guillotine into 2-inch lengths. It is then dried in the following manner. The cotton is placed upon an endless band, which conducts it to the stove, or drying closet, a chamber heated by means of hot air and steam traps to about 180 deg. F.; it falls upon a second endless band, placed below the first; it travels back again the whole length of the stove, and so on until delivered into a receptacle at the bottom of the farther end, where it is kept dry until required for use. The speed at which the cotton travels is 6 feet per minute, and as the length of the band travelled amounts to 126 feet, the operation of drying takes twenty-one minutes. One and a quarter lb. are weighed out and placed in a tin box; a truck, fitted to receive a number of these boxes, carries it along a tramway to a cool room, where it is allowed to cool.

~Dipping.~--Mixed acids are used in the proportion of 1 to 3, specific gravity nitric acid 1.52, and sulphuric acid 1.84. The dipping tank is made of cast iron, and holds 220 lbs. of mixed acids, and is surrounded on three sides by a water space in order to keep it cool. The mixed acids are stored in iron tanks behind the dipping tanks, and are allowed to cool before use. During the nitration, the temperature of the mixed acids is kept at 70 deg. F., and the cotton is dipped in quantities of 1-1/2 lb. at a time. It is put into a tin shoot at the back of the dipping tank, and raked into the acids by means of a rabble. It remains in the acids for five or six minutes, and is then removed to a grating at the back, pressed and removed. After each charge of cotton is removed from the tank, about 14 lbs. of fresh mixed acids are added, to replace amount removed by charge. The charge now weighs, with the acids retained by it, 15 lbs.; it is now placed in the pots, and left to steep for at least twenty-four hours, the temperature being kept as low as possible, to prevent the formation of soluble cotton, and also prevent firing. The proportion of soluble formed is likely to be higher in hot weather than cold. The pots must be covered to prevent the absorption of moisture from the air, or the accidental entrance of water, which would cause decomposition, and consequent fuming off, through the heat generated by the action of the water upon the strong acids.

The excess of acids is now extracted by means of hydro-extractors, as at Stowmarket. They are worked at 1,200 revolutions per minute, and whirled for five minutes (10-1/2 lbs. of waste acids are removed from each charge dipped). The charge is then washed in a very similar manner to that previously described, and again wrung out in a centrifugal extractor (1,200 revolutions per minute). The gun-cotton is now boiled by means of steam in wooden tanks for eight hours; it is then again wrung out in the extractors for three minutes, boiled for eight hours more, and again wrung out; it is then sent to the beater and afterwards to the poacher. The poachers hold 1,500 gals. each, or 18 cwt. of cotton. The cotton remains six hours in the poachers. Before moulding, 500 gals. of water are run into the poacher, and 500 gals. of lime water containing 9 lbs. of whiting and 9 gals. of a caustic soda solution. This mixture is of such a strength that it is calculated to leave in the finished gun-cotton from 1 to 2 per cent. of alkaline matter.

By means of vacuum pressure, the pulp is now drawn off and up into the stuff chest--a large cylindrical iron tank, sufficiently elevated on iron standards to allow room for the small gauge tanks and moulding apparatus below. It holds the contents of one poacher (18 cwt.), and is provided with revolving arms to keep the pulp stirred up, so that it may be uniformly suspended in water.

Recently a new process, invented by J.M. and W.T. Thomson (Eng. Pat. No. 8,278, 1903), has been introduced at the Waltham Abbey Factory. The object of this invention is the removal of the acids of nitration from the nitrated material after the action has been completed, and without the aid of moving machinery, such as presses, rollers, centrifugals, and the like. The invention consists in the manufacture of nitrated celluloses by removing the acids from the nitrated cellulose directly by displacement without the employment of either pressure or vacuum or mechanical appliances of any kind, and at the same time securing the minimum dilution of the acids. It was found that if water was carefully run on to the surface of the acids in which the nitro-cellulose is immersed, and the acids be slowly drawn off at the bottom of the vessel, the water displaces the acid from the interstices of the nitro-cellulose without any undesirable rise in temperature, and with very little dilution of the acids. By this process almost the whole of the acid is recovered in a condition suitable for concentration, and the amount of water required for preliminary washing is very greatly reduced. The apparatus which is used for the purpose consists of a cylindrical or rectangular vessel constructed with a perforated false bottom and a cock at its lowest point for running off the liquid. Means are also provided to enable the displacing water to be run quietly on to the surface of the nitrating acids.[A]

[Footnote A: In a further patent (Eng. Pat. 7,269, 1903, F.L. Natham), J.M. Thomson and W.T. Thomson propose by use of alcohol to replace the water, used in washing nitro-cellulose, and afterward to remove the alcohol by pressing and centrifuging.]

The apparatus is shown in Fig. 2O, side elevation, and in Fig. 21 a plan of the nitrating vessel and its accessories is given. In Fig. 20 is shown in sectional elevation one of the trough devices for enabling liquids to be added to those in the nitrating vessel without substantial disturbance.

In carrying out this invention a rectangular lead-lined or earthenware tank _a_ is employed, having a false bottom _b_, supported by ribs _c'_, over the real bottom _c_, which slopes down to a draining outlet pipe _d_, provided with a perforated grid or plate _e_, adapted to prevent choking of the outlet. Suitably supported near the top of the vessel _a_ are provided two troughs, _f_ having depending aprons _g_, a pipe _h_ has two branches _h'_, leading to the troughs, _f_. This pipe _h_ is adapted to be connected by a rubber pipe either to the outlet pipe _k'_ of the sulphuric acid tank _k_ or the water supply pipe _l_. The nitrating acids are supplied through the pipe _m_. A charge of mixed nitrating acids is introduced into the vessel _a_ say up to the level _n_, and the dry cellulose thrown into the acids in small quantities at a time, being pushed under the surface in the usual way.

A thin layer, say half an inch, of a suitable liquid, preferably sulphuric acid, of a gravity not exceeding that of the waste acid to be produced, is run carefully on the top of the acids by means of the troughs _f_, which are perforated as shown at _o_, so that the sulphuric acid runs down the aprons _g_, and floats on the nitrating acids. The whole is then allowed to stand till nitration has been completed. Water is then supplied to the troughs by way of the pipes _l_, _h_, and _h'_, and is allowed to float very gently over the surface of the sulphuric acid, and when a sufficient layer has been formed, the cock _p_ at the bottom of the apparatus is opened, and the acid slowly drawn off, water being supplied to maintain the level constant. It is found that the rate of displacement of the acids is a factor which exerts a considerable influence on the properties of the resulting nitro-cellulose, and affords a means of regulating the temperature of displacement. A rate of displacement which has been found suitable is about two inches in depth of the vessel per hour when treating highly nitrated celluloses, but this rate may, in some cases, be considerably increased. The flow of water at the top of the apparatus is regulated so that a constant level is maintained. By this means the water gradually and entirely displaces the acids from the interstices of the nitro-cellulose, the line of separation between the acids and the water being fairly sharply defined throughout. The flow of water is continued until that issuing at the bottom is found to be free from all trace of acid. The purification of the nitro-cellulose is then proceeded with as usual, either in the same vessel or another.

In the process above described, the object of the introduction of a small layer of sulphuric acid is mainly to prevent the fuming which would otherwise take place, and is not essential, as it is found it can be omitted without any deleterious effect. In order to use the mixed acids in the most economical manner, the waste acid from a previous operation may be used for a first nitration of the cellulose; being afterwards displaced with fresh acids which carry the nitration to the required degree before they are in turn displaced by water. The apparatus may be used merely for the removal of the acid, in which case the nitration is carried out in other vessels in the usual way, and the nitro-cellulose removed to the displacement apparatus where it is just covered with waste acid, and the displacement then proceeded with as above described. In some cases the process is carried out in an ordinary nitrating centrifugal, using the latter to effect preliminary drying after acid extraction. This gives a great advantage over the usual method of working ordinary centrifugal nitrating apparatus, because the acid being removed before the centrifugal is run, practically all danger of firing therein disappears, and a greater proportion of the waste acid is recovered.

In some cases the acids and water may be supplied by perforated pipes, lying along the edges of the nitrating vessel, and these edges may, if desired, be themselves made inclined, like the sides of the troughs _f_. In the case of effecting nitration in centrifugals as above, the displacing sulphuric acid and water may thus be supplied round the edges of the machines, or removal troughs such as _f_ may be used. It will be obvious that any inert liquid of suitable specific gravity may be used instead of sulphuric acid, as a separation layer.

~Moulding.~--By means of the small measuring tank above referred to, the gun-cotton pulp is drawn off from the stuff chest, and run into moulds of the shapes and sizes required. Thence a large proportion of the water is drawn off by means of tubes connected with the vacuum engine, the moulds having bottoms of fine wire gauze, in order to prevent the pulp from passing through. Hydraulic pressure of about 34 lbs. on the square inch is then applied, which has the effect of compressing the pulp into a state in which it has sufficient consistency to enable it to be handled with care, and also expels a portion of the remaining water.

~Compressing.~--The moulded gun-cotton is now taken to the press house, which is situated at some distance from the rest of the factory. Here the moulds are subjected to powerful hydraulic pressure, from 5 to 6 tons per square inch, and is compressed to one-third of its previous bulk. The slabs or discs thus formed are kept under pressure for a short time, not exceeding a minute and a half, to give the requisite density. It should, when removed, be compact, and just sink in water, and should perceptibly yield to the pressure of the fingers. There are perforations in the press blocks, to allow of the escape of gases, if formed, by reason of sufficient heat being generated. The men working the press are placed under cover, behind strong rope mantlets having eye tubes which command a view of the press.