substance produced is explosive, and is called from its composition tri-nitro-cellulose. This is the purest form of gun-cotton. If weaker acid is used, less hydrogen is displaced, and the product is called di-nitro-cellulose or mono-nitro-cellulose, according as it contains two or only one part of nitroxyl. These derivatives are either feebly explosive or not explosive at all. Such are the compounds known as photographic collodion and soluble gun cotton—the latter name distinguishing it from pure gun-cotton, which is not soluble in a mixture of ether and alcohol.
The Schultze powder contains both the explosive and the non-explosive varieties of nitro-cellulose.
If the wood fibre, after being carefully purified according to the method described in Schultze’s patent of 1864, were thoroughly desiccated and allowed to cool out of contact with air, and then dipped in acid of the strength mentioned in the specification, there seems no theoretical reason why an explosive powder containing at least 90% of true tri-nitro-cellulose should not be produced. As, however, I find on experiment that nothing like that per-centage is arrived at, I can only conclude that, in order to moderate the violence of the explosion, the Schultze Company secure the formation of a large per-centage of “soluble” or less explosive nitro-compounds by merely air-drying their wood.
If this supposition be generally true, it seems probable that the sample of Schultze powder supplied by Messrs Blissett may owe its extra explosive force to exceptional care being taken, during the interval between the drying and the dipping, to prevent the absorption of moisture—with the addition, perhaps, of an increased length of exposure to the action of the acid.
That some such variation of the ordinary procedure was carried out seems evident from the different proportions of soluble and insoluble gun-cotton in the specimens of Schuitze powder supplied by Messrs Blissett and Messrs Bland; for it was found that on the washed wood fibre from each being submitted to the action of a mixture of alcohol and ether, about one half of the former powder and two thirds of the latter were dissolved out. This shows that while the “Blissett” specimen contained about one half its weight of insoluble or explosive nitro-cellulose, the “Bland” contained only about one third—a difference which confirms the result obtained by analysis as stated below.
The soluble gun-cotton, ordinarily non-explosive, may, however, be rendered explosive by saturating it with bodies rich in oxygen, which promote the decomposition and complete the combustion of the fibre. Nitre is used for that purpose, because it parts with its oxygen readily; and nitrate of baryta is also used, because, being more stable than the nitre, it renders the combustion more gradual than would be the case if nitre were alone employed. When both are used, the nitre, I should think, would start, and the nitrate of baryta continue and finish the combustion of the powder. The amount used is, I suppose, the result of calculation and experiment; but a powder containing little true tri-nitro-cellulose should require more of these salts than one containing much tri-nitro-cellulose; and an excess of the salts would lower the rate of burning of the powder.
I will now give my analysis in full of the three powders, viz.—(1) the ordinary powder issued last season, being part of a supply obtained from Messrs Bland, gunmakers, of the Strand; (2) some powder furnished by Messrs Blissett, of Holborn, and alluded to in their letter in the ‘Field’ of Jan. 19th last, as having damaged a gun made by them; and (3) some of the new powder of 1878, as used at the ‘Field’ trial of explosives in May last.
| 1877 Bland’s. | Blissett’s. | 1878 Trial or New. | ||
| Moisture, per cent. | 2·18 | 2·39 | 2·97 | |
| Extracted by water. | Nitrate of baryta, per cent. | 21·50 | 16·59 | 22·32 |
| Nitrate of potash, per cent. | 11·46 | 10·46 | 6·47 | |
| Yellow coloured organic substance, trace of chlorides, &c., undetermined | ||||
| Insoluble in water. | The converted wood fibre (nitro-cellulose) then remaining contained the following per-centage of mineral matter | 5·0 | 6·0 | 2·95 |
The converted wood fibre (after allowing for extraneous mineral matter) possessed the following per-centage composition. I place for comparison Professor Abel’s determination of the composition of tri-nitro-cellulose, and two of the impurities found along with it, in a parallel column.
| Bland’s. | Blissett’s. | Trial or New. | Tri-nitro- Cellulose. | Impurities. | ||
| Carbon | 28·75 | 28·07 | 28·12 | 24·24 | 29·20 | 30·50 |
| Hydrogen | 3·49 | 3·65 | 3·54 | 2·36 | — | 2·91 |
| Nitrogen | 10·80 | 15·60 | 11·66 | 14·14 | 11·85 | — |
| Oxygen | 56·06 | 52·68 | 56·68 | 59·26 | — | — |