Experimental Researches in Chemistry and Physics/Passage of gases through tubes

by Michael Faraday

Experimental Observations on the Passage of Gases through Tubes^[1].

In a previous communication I have noticed briefly some curious effects which take place when gases are passed through tubes at low pressures. They consist in an apparent inversion of the velocities; those which traverse quickest when the pressure is high, moving more tardily as it is diminished, until they are among those which require the longest time in passing the tube; thus with equal high pressures equal volumes of hydrogen gas and olefiant gas passed through the same tube in the following times:—

Hydrogen in	57″
Olefiant gas in	135″·5;

but equal volumes of each passed through the same tube at equal low pressures in the following times:—

Hydrogen in	8′ 15″
Olefiant gas in	8′ 11″.

Again, equal volumes of carbonic oxide and carbonic acid gases passed at equal high pressures through the same tube,

occupied, the	Carbonic oxide	133″
	Carbonic acid	156″·5;
but at low pressures,	Carbonic oxide	11′ 34″
	Carbonic acid	9′ 56″;

I have lately had my attention again called to the subject, but have not yet been able to satisfy myself of the cause of this curious effect; nevertheless, as experiments do not always owe their value to the hypothesis which accompanies them, a few short observations on some made on this subject may be acceptable.

The effect is always produced by fine tubes at low pressures, but does not appear to belong to the mere obstruction by the tube to the passage of the gas, nor have I been able to produce it without a tube. A very fine needle-hole was made in a piece of platinum foil, and so arranged on a mercurial gasometer, that the pressure of a small column of mercury sent seven cubical inches of the following gases through in the times mentioned, namely—

Hydrogen	3′·8 nearly, taking a mean,
Olefiant gas	9′·2;

and when the pressure was increased, the same proportions in the times was observed. Other similar experiments gave similar results.

Slits, cut in platinum foil by the edge of a penknife, did not give so great a superiority to hydrogen as that mentioned above, and the proportion varied with different slits; still the hydrogen passed most rapidly, and a difference of pressure caused no difference in the relative times.

Three diaphragms were placed in different parts of the same tube, each being perforated with a small hole, but the effects produced in tubes were not observable here. Hydrogen passed in 3·8 minutes, and olefiant gas in 9·1 minutes. The gases were passed through discs of paper, and the number of discs was increased so as to increase the obstruction, the pressure and quantity of gas remaining the same.

With one disc
of drawing-paper 6·5 cubical inches of hydrogen passed in 7′
                 6·5    „      „    of olefiant gas in   18′;
with two discs the hydrogen  passed in 15′·4
   „      „        olefiant gas   „    38′;
with three discs the hydrogen     „    22′·5
   „      „          olefiant gas „    57′·75.

Lastly, for the effect of obstruction, I used a tube filled with pounded glass. This was uncertain, because on moving the tube it was impossible, almost, not to move some of the particles within, and then, of course, circumstances were changed; but by sending the gases through one alter the other, results were obtained,

the mean of which gave for hydrogen     3′·4
     „          „      for olefiant gas 4′·7.

It would seem from these experiments that mere obstruction is not the cause of the effect observed in tubes, for when the tubes are removed, and obstructions which retard much more placed for them, the effect is lost; and as the same aperture produces no difference of effect at high or low pressures, the variations between different apertures should probably be referred to some other cause.

I then endeavoured to ascertain some of the circumstances attending on tubes. Both glass and metal tubes produce the effect, and a metal tube, down which a wire had been thrust, did not seem to have this influence on the passage of gases through it altered. The effect is heightened as the gas is made to pass more slowly through the tube; and this, whether the increased time be caused by diminished pressure, increased length of tube, or diminished diameter. This may be well illustrated by putting several very fine tubes together, for the particular effect is thus increased whilst the time is shortened. Two brass planes were ground together, and a few scratches made down one of them so as to form very fine tubes; through these olefiant gas passed in 26′·2, and hydrogen in 32′·5.

Three glass tubes were taken of different diameter, and cut into such lengths that they passed nearly equal quantities of hydrogen gas in equal times hy the same pressure; their lengths were 42, 10·5, and 1·6 inches. The longer tube passed the

hydrogen in	.  .  .	3′·7,	the olefiant	gas in	2′·75
the second in	.  .  .	3′·5	„	„	2′·5
the smaller in	.  .  .	3′·45	„	„	2′·8;

and in several other experiments there seemed to be nearly an equal effect, when the quantity of gas passed in the same time was the same.

I imagined that the specific gravity of the gases might have some constant influence, but this does not seem to be the case; carbonic oxide and olefiant gas are nearly of the same density; and if the effect depended upon their weight, it should be nearly the same for both of them; but this is not so; seven cubical inches of carbonic oxide required 4·6 minutes to pass through a tube which was traversed by the same quantity of olefiant gas under the same pressure in 3·3 minutes, each gas having been placed over caustic lime for some time previously; and oxygen required to pass through the same tube 5·45 minutes of time.

I placed three gauges in different parts of a tube, of such a size that it passed olefiant and hydrogen gas in nearly equal times; the gauges were very obedient to the pressure of the gas in the different parts of the tube, but I could not perceive any difference between the effect of the different gases.

Such are some of the circumstances which affect and produce this curious effect: that the velocity of gases in passing through tubes should be in some proportion to the pressure on them is nothing particular; but the singularity is, that the ratio for the same gas varies with the pressure, and that this variation differs in different gases; thus the one which passes with the greatest facility at low pressure, passes with the least at high pressure.

It may be deduced from the experiments at high pressures and on obstructions, that the fluidity of the gas has little or nothing to do in this case, for where it alone can have an influence, the indications are the same at all pressures, and the gas of least density passes in the shortest time; thus comparing hydrogen with olefiant gas, and considering its time I, the time of the latter will be in the experiments already mentioned, as 2·38, 2·42, 2·4, 2·57, 2·46, 2·57 ratios, which do not differ much from each other, though the times, pressure, obstructions, and quantities of gas used vary very considerably.

Neither is the variation among the different gases between the ratio of the velocity and pressure, connected with specific gravity, at least I have not been able to observe such a connexion. I have quoted an experiment, or rather the general result of several, on carbonic oxide and olefiant gases, and it is adverse to the supposition; and in others, made on sulphurous acid gas and ammoniacal gas, still further departures from the order of the densities were observed.

If a tube sufficiently fine and long be connected with a portion of gas under high pressure, so that the time occupied in its passage through it be considerable, the effect will be produced, i. e. the times of different gases will vary from each other, but not according to their specific gravities; if the tube, however, be cut off so that the gases pass quickly, then the times will be as the specific gravity. Now, in the long tube, the pressure and velocity will vary throughout its length, the pressure being greatest at the internal or connected end, and least at the other extremity, whilst the velocity is least at the end towards the reservoir, and greatest at the other. But the ratio by which the pressure and velocity decrease and increase, appears different for and peculiar to each gas. At the end of the long tube the olefiant gas issues more rapidly than hydrogen, though the pressure st the reservoir is the same; but shorten the tube, and let that part in which high pressures only exist confine the gases in their passage, and the hydrogen gas will surpass the olefiant gas in velocity as far as 4 or 5 does 2. It would seem, therefore, that in the long tube the pressure or elasticity of the olefiant gas diminishes less rapidly than that of the hydrogen, or that its velocity increases more rapidly.

Perhaps these effects may be accounted for by the supposition of some power of expansion peculiar to each gas, which, if existing, a tube would for many reasons be well calculated to exhibit. The experiment requires numerous repetitions and much time, and I have not yet had sufficient to satisfy myself on the subject. I will therefore refrain from mixing up crude notions with facts, and at some more convenient opportunity endeavour to supply what is wanting in this paper.

1. Quarterly Journal of Science, vii. 106.