| Heat Test at 80° | Heat Test at 134° | |
| Original sample C | 10 | 41 |
| Treated with lead acetate | 67 | 45 |
| " zinc " | 91 | 45 |
In conclusion we may briefly resume the main points arrived at in these investigations.
Causes of instability of cellulose nitrates.—The results of our experiments so far as to the causes of instability in cellulose nitrates may be summed up as follows:—
(1) Traces of free nitrating acids, which can only occur in the finished products through careless manufacture, will undoubtedly cause instability, indicated strongly by the ordinary heat test at 80°, and to a less extent by the heat test at 134°.
(2) Other compounds exist in more intimate association with the cellulose nitrates causing instability which cannot be removed by exhaustive washing with either hot or cold water, by digestion in cold dilute alkaline solutions such as sodium carbonate, or by extracting with ether, alcohol, benzene, &c.; these compounds, however, are soluble in the solvents of highly nitrated cellulose such as acetone, acetic ether, pyridine, &c., even when these liquids are so diluted with water or other non-solvent liquids to such an extent that they have little or no solvent action upon the cellulose nitrate itself. These solutions containing the bodies causing instability are neutral to test paper, but become acid upon evaporation by heating. (This probably explains the presence of free acid when guncotton is purified by long-continued boiling in water without any neutralising agent being present.)
(3) The bodies causing instability are products or by-products of the original ester reaction, acid bodies containing nitroxy-groups, but otherwise of ill-defined characteristics. They combine with the oxides of zinc or lead, giving insoluble compounds. They are precipitated from their solutions in diluted acetone upon the addition of soluble salts of these metals.
(4) Cellulose nitrates are rendered stable either by eliminating these compounds, or by combining them with the oxides of lead or zinc whilst still in association with cellulose nitrates.
(5) Even the most perfectly purified nitrocellulose will slowly decompose with formation of unstable acid products by boiling for a long time in water. This effect is much more apparent at higher temperatures.
Dense structureless or non-fibrous cellulose nitrates can be industrially prepared (1) by nitrating the amorphous forms of cellulose obtained from its solution as sulphocarbonate (viscose). The cellulose in this condition reacts with the closest similarity to the original fibrous cellulose; the products are similar in composition and properties, including that of instability.
(2) By treating the fibrous cellulose nitrates with liquid solvents of the high nitrate diluted with non-solvent liquids, and more especially water. The optimum effect is a specific disintegration or breaking down of their fibrous structure quite distinct from the gelatinisation which precedes solution in the undiluted solvent, and occurring within narrow limits of variation in the proportion of the diluting and non-solvent liquid—for industrial work the most convenient solution to employ is acetone diluted with about 10 p.ct. of water by volume.