"I put two ounces and a half of Walnut Oil into the bottom part of a broken retort, having the figure of a cap, or concave hemisphere; and poured thereon two ounces of smoking Spirit of Nitre. It was scarce put in when a considerable ebullition arose, with a very thick smoke. As I found it continually increasing, and very fast too, I retired a little, that I might observe the event without danger. This caution was not unnecessary: for immediately the whole mixture blew up as high as the ceiling, with a noise like the discharge of a musket. Nothing was left in the vessel but a black matter, which still continued to boil a little and run over, and at last remained very rare, spungy, and as full of holes as a honeycomb: its consistence also was such that it did not stick to my fingers when I handled it.
"As Mr. Geoffroy, who first found the means of firing the natural Balsams, observed in them a similar explosion on that occasion, it appears that my Oil was very near taking fire in this experiment: which makes me presume that we may at last succeed in firing Fat Oils likewise, and consequently all others; seeing these have always been looked upon as the most unlikely to produce that phenomenon. I imagine that, to accomplish this, nothing more is necessary than to make use of sufficiently great quantities, and to order it so that the surfaces of the liquors, where they come into contact, may be of a large extent."
Afterwards, in 1747, Mr. Rouelle read before the Academy a Memoir on the accension of Oils by Acids. That Memoir contains a great number of curious experiments, and peculiar manual operations described very distinctly, from which there results a general method of firing without fail, not only Essential Oils, but even any Fat Oil whatever: so that my conjecture, concerning the possibility of firing these latter Oils, mentioned in my above-cited Memoir of 1745, is now changed into a certainty. I shall proceed to explain how I conceive these accensions are brought about, and endeavour to account for the phenomenon from such causes as to me seem the most probable.
A due attention to the phenomena produced by mixing Oils with Acids will enable us, I imagine, to discover the natural cause why the Oils take fire. It is certain, and demonstrated by the most decisive experiments, that the friction of several bodies rubbing against each other produces heat; and that when these bodies are combustible, and the heat produced by their friction rises to a certain degree, they take fire. This, in my opinion, is what happens to Oils when mixed with concentrated Acids. When these two sorts of substances rush into union with rapidity, as in the experiments under consideration, there must necessarily be a great friction among their parts. This friction produces the heat observed at the time of their union. The more concentrated the Acids are, with the greater violence and rapidity do they act upon the Oils, and the greater is the heat raised. If the Acids be concentrated to such a degree as to produce, by uniting with the Oils, a heat equal to that of an ignited body, the combustible substances that are exposed to it, which in this case are Oils, must needs take fire and flame.
The heat produced on this occasion is so great, that, even when the inflammation doth not take place, if you touch the surface of the Oil with your finger, as soon as the Acid hath had its effect, you will find it burn you like a live coal.
Two pieces of wood, rapidly and violently rubbed against each other, take fire. What is it that is kindled in this case? It can be nothing but their Oil: for they contain no other combustible principle. Why doth this Oil take fire? I do not think it possible to assign any reason for it, but the heat produced by the friction of the pieces of wood containing the Oil. If, when Oil is dispersed in a body, of which it is only one component principle, and consequently mixed with many saline, aqueous, and earthy parts, that are not inflammable, but, on the contrary, make the Oil less so, the Oil nevertheless takes fire, and burns when agitated by a sufficient degree of heat; why shall not this very Oil, when separated from the mixt of which it made a part, when united into one distinct mass, and entirely, or almost entirely, freed from the heterogeneous, incombustible parts with which it was combined, and consequently now more inflammable than before; why, I say, shall it not take fire, when exposed to a degree of heat equal, or rather superior, to that which is produced by rubbing two pieces of wood together?
Let us now examine the phenomena produced when Oils are fired by Acids, all the circumstances that favour or hinder their accension, and see if they agree with the explanation here offered.
First, no sort of Oil will take fire with any Acid whatever that is not highly concentrated; for weak Acids act but feebly on Oils, and dissolve them slowly; so that the friction is neither quick nor violent, and consequently produces too faint a heat, far below the degree of ignition.
Secondly, no inflammation is produced when Acids and Oils are mixed in too small quantities; but the more Acid and Oil you mix together, the greater is the certainty of succeeding: for the heat is exactly in proportion to the friction that produces it; and the total quantity, or amount, of this friction is so much the greater, as there are more particles rubbing against each other at the same time. So that if a very small quantity of Acid and Oil be mixed together, there will be but a very small quantity of friction, and consequently a very small quantity of heat; and in that case no inflammation. It was with a view to avoid these inconveniencies, and to procure the opposite advantages in as great a degree as possible, that, in the passage above quoted from my Memoir of Oils, I proposed mixing together large doses of Acid and of Oil, as one of the means by which we might succeed in the accension of Fat Oils.
Thirdly, the figure of the vessel, in which the two liquors are mixed together, is not a matter of indifference. A wide-spreading vessel, of a large diameter with respect to the quantity of liquor it is to contain, favours the inflammation much more than one of a small diameter. Nay, it may not succeed at all in too narrow a vessel, though all other circumstances be properly attended to.