These experiments are selected as the most accurate of nearly fifty, made on the inflammation of different quantities of nitrous oxide and hydrogene.
As Mr. Keir found muriatic acid in the fluid, produced by the inflammation of oxygene and hydrogene in closed vessels, in Dr. Priestley’s experiments, I preserved the residual gas of about 3 cubic inches of nitrous oxide, that had been detonated at different times with less than a cubic inch and half of hydrogene; but solution of nitrate of silver was not clouded when agitated in this gas, nor when introduced into the detonating tube in which the inflammation had been made.
From these experiments we learn that nitrous oxide is decomposable at the heat of ignition, by hydrogene, in a variety of proportions.
When the quantity of hydrogene very little exceeds that of the nitrous oxide, both of the gates disappear, water is produced, no nitrous acid is formed, and the volume of nitrogene evolved is rather greater than that of the nitrous oxide decomposed.
When the quantity of hydrogene is less than that of the nitrous oxide, water, nitrous acid, oxygene and nitrogene, are generated in different proportions; one part of the nitrous oxide is most probably wholly decomposed by the hydrogene, and the other part converted into nitrous acid and atmospheric air, in consequence of the ignition.
From experiments c, d, and e, the composition of nitrous oxide may be deduced. In experiment d, 39 of nitrous oxide were decomposed by 40 of hydrogene, and converted into 41 of nitrogene.
Now from b it appears that 40 of hydrogene require for their condensation about 20.8 of oxygene in volume; so that founding the estimation upon the quantity of hydrogene consumed, 100 parts of nitrous oxide would consist nearly of 63.1 of nitrogene, and 36.9 of oxygene. But 41 of nitrogene weigh 12.4, [Res. I. Div. I]. Consequently, deducing the composition of nitrous oxide from the quantity of nitrogene evolved, 100 parts of it would consist of 63.5 nitrogene, and 36.5 oxygene.
These estimations are very little different from those which may be deduced from the other experiments, and the coincidence is in favor of their accuracy.
From the following experiment it appears that the temperature required for the decomposition of nitrous oxide by hydrogene must be higher than that which is necessary to produce the inflammation of hydrogene with oxygene. I introduced into small tubes filled with equal parts of nitrous oxide and hydrogene, standing on a surface of mercury, iron wires ignited to different degrees, from the dull red to the vivid white heat. The gases were always inflamed by the white and vivid red heats; but never by the dull red heat, though the last uniformly inflamed mixtures of oxygene and hydrogene, and atmospheric air and hydrogene.
Dr. Priestley[170] first detonated together nitrous oxide and hydrogene; his experiment was repeated by the Dutch chemists, who found that when a small quantity of hydrogene was employed, the nitrous oxide was partially converted into a gas analogous to common air. Their estimation of its composition, which is not far removed from the truth, was founded on this phænomenon.[171]