One may effect the combustion of charcoal, like that of phosphorus, under a glass bell jar, plate IV, figure 3, [see below] filled with oxygen gas and inverted over mercury : but as the heat of even red hot iron does not suffice to ignite it, one adds to the charcoal a small piece of tinder and a bit of phosphorus. Once can easily light the phosphorus with red hot iron; and the flame travels successively to the tinder, then to the charcoal.

Details on this experiment may be found in Mémoires de l"Académie, 1781, page 448. One sees there that 72 parts of oxygen by weight are required to saturate 28 of charcoal, and the the gaseous acid produced has exactly the same weight as the sum of the weights of the charcoal and the oxygen from which it formed.

This gaseous acid was named fixed air by the first chemists who discovered it. They did not know then whether it was an air like that of the atmosphere or another elastic fluid, vitiated and corrupted by combustion; but since it is clear today that this aeriform substance is an acid, that it is formed like all other acids by oxidation of a base, it is easy to see that the name fixed air is not at all appropriate. [Lavoisier and his colleagues decided to call it carbonic acid gas.]

Thus in this operation 45 grains of phosphorus combined with 69.375 grains of oxygen and, since no substance with mass could pass through the glass, one has the right to conclude that the weight of whatever substance resulted from this combination and which collected in white flakes, must total the sum of the weighs of oxygen and of phosphorus, that is 114.375 grains. One will soon see that these white flakes are none other that a solid acid. [which he calls solid phosphoric acid.]

1) How accurate was Lavoisier's work in these two experiments? Check his numbers using modern values of atomic weights.

2) How do you suppose he calculated the weight of oxygen? Suggest a plausible source of experimental error
(see his detailed experimental instructions below).

3) Assuming reasonable dimensions of the apparatus (6 pint jar), how many significant figures do you think he should have used?

4) Why do you think Lavoisier used so many digits in "69.375 grains"?

Part III , Chapter VI, Paragraph I On the Combustion of Phosphorus and Charcoal To effect combustion of phorphorus or charcoal one begins by using the pneumatochemical apparatus over water to fill a bell jar holding at least six pints with oxygen gas. When it is completely full and the gas begins to bubble out, one takes the jar A to the mercury apparatus, Plate IV, fig. 3, using a very flat glass or china dish beneath it. Then one carefully dries the surface of the mercury with blotting paper both inside and outside the jar. This requires several precautions : if one is not careful to put the paper completely under the mercury for a certain time before introducing it under the jar, one transfers common air that sticks tightly to the paper. One then takes a flat shallow dish D of iron or porcelain containing the substance that one wants to burn after weighing it very exactly on the balance ; one then covers this dish with a slightly larger one P,which serves as a diving bell, one transfers the assembly through the mercury ; after which one brings back P, which served only as a cover. One can avoid the challenge and difficulty of passing materials through the mercury by lifting one side of the bell jar for a tiny instant and thus introducing the dish with the combustible material through the gap. Using this second approach a little common air mixes with the oxygen gas ; but this mixing, which is slight, damages neither the success nor the precision of the experiment.

When dish D has been introduced under the bell jar, one sucks out part of the oxygen gas that it contains to raise the mercury to EF. Without this precaution, once the combustible substance is lit, the heat expands the air ; part of it would escape the bell jar and one could make no exact calculation on the quantities. To suck out the air one use the siphon GHI, which one sticks under the bell jar ; and to keep it from filling with mercury one twists a small piece of paper around its end I.

There is a way of raising a column of mercury several inches above its level by sucking [Note: this approach is not sanctioned by current standards of safe laboratory practice] : if one sucks air with the lungs, one can achieve only a very mediocre elevation, for example an inch or an inch and a half at most ; and then only with extraordinary effort ; while by the action of the muscles of the mouth one can raise the mercury to six or seven inches without tiring, or at least without risk of trouble. A more convenient method still is to use a little pump attached to the siphon GHI ; then one can raise the mercury to whatever appropriate height as long as it does not exceed 28 inches.

If the combustible substance is very inflammable, like phosphorus, one lights it with a curved iron rod, plate IV, fig. 16, heated red in the fire, and poked quickly under the bell jar : as soon as it touches the phosphorus, it lights. For less combustible substances, such as iron, several other metals, charcoal, etc. one uses a bit of tinder on which one places a speck of phosphorus : and lights the last in the same way with the curved rod ; flame travelling to the tinder and then to the combustible substance.

In the first instant of combustion the air expands and the mercury descends; but when no elastic fluid [gas] is formed, as in combustion of iron or phosphorus, absorption soon becomes observable and the mercury climbs high in the jar. Thus one must be careful not to burn too much of the combustible substance in a given quantity of air; otherwise near the end of combustion the dish approaches the dome of the jar too closely and the great heat could make it crack.

In Chapter II, paragraphs V and VI, I presented ways of measuring the volume of gases and necessary corrections of the volume relative to the height of the barometer and the temperature ; I will add nothing in this regard. The example cited on p. 381 was taken from the combustion of phosphorus. [Measuring the amount of oxygen consumed is easier for phosphorous, which gives a solid, than for carbon, which gives a gas of the same volume as that of the oxygen consumed. Lavoisier could have absorbed the carbon dioxide with solid potassium hydroxide to determine the reduction in volume of oxygen.]

...this method is not without danger for substances which are susceptible to vaporizing with modest heat, such as ether, wine spirits, and essential oils. These volatile substances dissolve in a rather great quantity in oxygen gas ; when one lights them, there is a sudden detonation which blows the bell jar to a great height and shatters it. I have observed two of these detonations, of which members of the Academy have, like myself, seen themselves to be the victims. [obviously this must also have scattered mercury far and wide.]