Raoul Pictet - Exposition Nationale Suisse Genève 1896

THE LIQUEFACTION OF OXYGEN AND HYDROGEN, AND THE SOLIDIFYING OF HYDROGEN.

HEAT being known to man as a sensation, it is no wonder that philosophers should so long have mistaken it for some special agent. But, at last, the revelations of sensation were found insufficient to explain many phenomena connected with the manifestations of heat. Hence arose the science of calorimetry, whose business it is to treat of the relations existing between the elastic force, the volume, and the temperature of bodies. Heat that cannot be tested by sensation was then called latent heat, a felicitous expression, which was the harbinger of a host of fresh discoveries in every scientific field. The mechanical theory of heat, the immediate outcome of the theory of latent heat, seems destined to explain every difficulty connected with this hitherto abstruse subject.
The mechanical theory of heat does away with heat as a special agent, and declares that heat is nothing more than a molecular and atomic motion; so the study of heat has now become the study of molecular and atomic motion. It is no exaggeration to say that the moving atoms are now as clearly visible to the eye of scientific analysis as though they could be brought under microscopic investigation.
The theory of atomic motion is a new element in the long pending discussions on the physical constitution of bodies, an element of such paramount importance that atomic motion and the nature of matter may be considered as almost one and the same subject. M. Clausius’s theory of gases is perhaps the most telling deduction yet made from that of atomic motion.
My purpose is to take up some of the phenomena which have hitherto been considered exceptions to the Clausius theory or to accepted physical laws, and experimentally to prove that they form no exceptions to those laws. A few words concerning these phenomena may help the reader to apprehend the drift of this essay.
Almost all known bodies may assume the gaseous, liquid, or solid state. In the same body, these three states are brought about by

three different degrees of temperature; the solid state is the result of the lowest temperature, the liquid state of a higher temperature, and the gaseous of the highest temperature. The condition of a body depends on two forces often antagonistic—( 1) atomic cohesion the result of attraction, and (2) atomic vibration the result of heat. This may be considered as a general law ruling matter.
Boyle’s law on elastic fluids states that when the temperature remains constant, the elastic force of a gas is inversely proportional to the volume it occupies. Boyle’s law would only be accurate for a gas of ideal purity. Now, M. Regnault in his memorable experiments on the compressibility of gases has brought out an important fact, viz., that gases nearing the point of liquefaction decrease in volume more rapidly than Boyle’s law indicates. This accelerated decrease of volume at liquefying point is caused by the force of atomic attraction coming more rapidly into play as the gas is condensed into a liquid, and thus accelerating, in opposition to Boyle’s law, the decrease of volume.
All the vaporous forms of known liquids, such, Tor instance, as those of mercury, water, alcohol, sulphurous acid, and carbonic acid, undergo the same influence; they all decrease in volume more rapidly at liquefying point than would a perfect gas.
The gases hitherto called permanent, because they had not been liquefied, i.e. hydrogen, oxygen, nitrogen, seem also to form an exception to Boyle’s too absolute law; for at liquefying point their decrease of volume is retarded.
And this brings us to investigate the phenomena of cohesion which have led me to find out the laws wherewith I might liquefy and solidify the so-called permanent gases of oxygen, hydrogen, and nitrogen.
If, in permanent gases, the cohesiveness of gaseous molecules were of itself sufficient to cause cohesion, under strong pressures cohesion would not fail to take place.
M. Regnault’s suggestions induced M. Natterer, of Vienna, to try the effect of enormous pressures on hydrogen, oxygen, and nitrogen. In 1854, M. Natterer applied a pressure of 2,790 atmospheres to these permanent gases.
His mode of treatment was the following:-Into a receiver, measured quantities of hydrogen were successively introduced, say, ten measures of hydrogen, oxygen, or nitrogen at a time. A very delicate manometer registered the pressure for every fresh supply of hydrogen.
In the following table, the first columns show the measures of hydrogen, oxygen, and nitrogen introduced into the receiver, the second columns the pressures corresponding to those measures, and the third columns the differences between the atmospheric pressures for every ten measures of gas.

The foregoing table proves that Boyle’s law is not true as soon as a pressure of 100 atmospheres is reached.
For relatively moderate pressures, oxygen is closer than hydrogen to the principle of Boyle’s law, viz., that with a constant temperature the elastic force of a gas should vary directly as the quantity of that gas contained in a given receiver. But still, for high pressures, oxygen also is an exception to Boyle’s law, and when 657 measures have been compressed, the pressure registered by the manometer is 1,354 instead of 657 which Boyle’s law would bid us expect. Again, 657 measures of hydrogen show a pressure of 1,104 atmospheres, and
657 measures of nitrogen show a pressure of 2,156 atmospheres.
These results expressed by a curve whose abscisse represent the compressed gas-measures, and whose coordinates express the corre-

sponding pressures, show a manifest tendency towards a limit of compressibility which cannot be passed. This limit is the point where the curve is an asymptote to the vertical coordinate. Then, for any increased quantity of gas in the receiver, the pressure is infinitely increased. Such is the case when the gaseous molecules have been pressed down into absolute contact. The intermolecular spaces being reduced to nothing, the volume of the gas cannot be further reduced on account of the impenetrability of matter.
The above figures clearly prove that the molecules of the permanent gases must repel one another with considerable force, since ten measures of oxygen show an increased pressure of 70 atmospheres, and 10 measures of nitrogen show an increased pressure of 110 atmospheres, facts which seem to invalidate the hypothesis of universal molecular cohesiveness.
The inference to be drawn from these experiments is, according to M. Clausius, that the molecular cohesiveness of permanent gases is next to nothing, and that their departure from Boyle’s law comes from the infinite smallness of their molecules. But in vapours whose molecules are relatively large, cohesiveness would operate even under weak pressures, a fact which would explain their liquefying under pressure sooner than consonant with Boyle’s law.
I have shown in a foregoing work that, supposing the temperature constant, the molecular forces, which bind together two atoms or two molecules of a liquid, are the same as the molecular forces which would bind together two atoms or molecules of another liquid. A more technical statement of this theorem would run thus: If any volatile liquid be taken at a temperature T⁰, and an atom a be

taken from it, if moreover the cohesiveness of a be calculated, that cohesiveness will be found to be the same for all liquids. This law proves that the liquid state is restricted to a fixed power of cohesion K, which acts at a distance L between two molecules. This condensation or volatilisation at T⁰ must depend on K. No other theory of condensation stands investigation.
The only known force antagonistic to cohesion is heat, giving to molecules or their atoms a pendulous motion, the amplitude of which is a function of their temperature. Considering the phenomena of latent heat such as they have been tabulated by M. Regnault, considering the laws of vaporous tension and dilatation under heat, I venture on the hypothesis that temperature is directly proportional to the amplitude of the atomic heat-wave.
Shall we maintain that in a body standing at T all its component

elements will vibrate with equal amplitudes? Assuredly not, for there is an intermolecular interference of heat-waves which gives to this molecule a greater vibratory amplitude, and to that a less than my law would assign to them were they independent one from another. A mean vibratory amplitude taken from all the atomic vibrations of a body represents the temperature of a body. In other words, the mean vibratory amplitude expresses the dynamic resultant of the atomic vibrations constituting the sum of the vibratory forces in a body. This sum of vibratory forces is called the potential of a body.
The above theory may be easily proved experimentally. Take, for instance, any vapour under pressure P and at temperature T. The intermolecular space is here inversely proportional to the number of the molecules. In other words, double the number of the molecules, and you thereby halve the intermolecular spaces. Call K (see diagram) the fixed power of cohesion acting at a distance L between two molecules A B, and on which condensation must depend; T being the tem.. perature and 1 the amplitude of the vibration corresponding to T.
The diagram represents two molecules of the vapour under the pressure at the temperature T. The distance between A and B is A B, a distance which may be increased or decreased by an increase or decrease of pressure. Call L the minimum distance at which cohesiveness acts in order to bring A and B under the fixed cohesive power K, and let A B” equal 1 the amplitude of the heat-wave proportional to the temperature T. Evidently, when the pressure leaves B at B, liquefaction cannot take place by cohesiveness, which, in this case, is inferior to K.
Increase the pressure to P’, A B will be reduced to A B’. Then, 1 being equal or less than L, and cohesiveness equalling K, B rushes on to A, and forms with it a liquid molecule. The two molecules in their approach will develope much heat, since the first oscillation being A B’ will almost immediately afterwards be reduced to A B”. The vis viva lost by the molecules and given out to the walls of the receiver represents the latent heat of condensation, i.e. the work of cohesiveness between the limits A B’ and A B”.
The variations of volume of a gas and of the liquid into which it is condensed allow the observer to determine the relation of the lengths A B’/A B”. The change of volume is considerable for liquids of mean volatility.
Bearing in mind Boyle’s law, the relation A B’/A B” and the K of molecular cohesiveness, we can realise the conditions which bring about the liquefaction of a vapour.
Press the molecules A, B, to the limit of cohesion B’; the temperature being constant, the pressure will be constant too, whatever the

quantity of a vapour pressed down. P’ being the maximum pressure, the latent heat set free is a function of the lengths A B’, A B” ; is a function of the number of the condensable molecules and a function of K which tallies with the condensing temperature T. An experimental proof of this statement may be given. For such a proof, let us take a few liquids in the order of their volatility. Generally, the intermolecular cohesiveness of a body depends on the density of the latter. The more stable a liquid or the higher its boiling-point on the thermometric scale, the greater the cohesiveness of its molecules. Hence the fact that in a dense liquid, the temperature T being constant, the distance of molecular attraction would be greater than in a liquid of less density. Take, for instance, water and sulphuric ether, and press down their vapours at a temperature -30° C. The distance A B’ for water will be greater than for ether; and thus, according to Boyle’s law, the pressure of a vapour of water will be less than that of a vapour of ether. In short, vapour pressures vary as their volatile power.
Now, the length A B’ is registered by 31.548 millimetres of mercury for water, and 634.80 millimetres of mercury for sulphuric ether; so for both liquids the pressure is not identical. But, on the other hand, the length of their heat-waves A B” will be equal for both liquids after condensation. Thus, the latent heat set free in both these cases will vary (1) as the number of molecules liquefied by their cohesiveness, (2) as the constant K tallying with the temperature T, and (3) as the function binding increasing cohesiveness to the different intermolecular distances between A and B. If, in succession, we compare water to sulphuric ether, to sulphurous acid, to ammonia, to carbonic acid, we see that the distance A B’ at which condensation takes place gets smaller and smaller as a more volatile liquid is taken. At freezing point condensation pressures are registered by millimetres of mercury as follows: Water, 4 millimetres; sulphurous acid 1.165; methylic ether 1.879; carbonic acid is condensed by a pressure of thirty atmospheres. It is manifest that the less condensable a vapour the greater is the difference between its volume and the liquid into which it has been condensed, a fact proving that the essential difference between one vapour and another is that of molecular cohesiveness.
In order, therefore, to condense two gas molecules A, B, two conditions must be realised. 1. Press A towards B, so that the distance A B’ to which A and B are brought may be under the influence of K the minimum force of condensing cohesiveness. 2. It is absolutely necessary that the distance A B’ should be greater than A B” the amplitude of the heat-wave. For if the molecular cohesiveness of a gas be small and the amplitude A B” of the heat-wave tallying with the temperature T be greater than A B’, liquefaction becomes impossible, because in order to prevent the heat-wave counteracting K, A and B

must be kept out of reach of K, and thus cannot cohere, a dilemma both alternatives of which bring about non-cohesion.
The second condition of liquefaction fully explains all the phenomena relative to the permanent gases oxygen, hydrogen, and nitrogen. In order to condense these gases, it is not enough to subject them to enormous pressures; the amplitude 1 of the heat-wave must be made less than A B’ by lowering the temperature. Then A B” being small, as small as possible, A B’ the distance at which the molecules A, B, become condensable by K will be larger than A B”. Thus, and thus alone, can liquefaction be brought about.
Hitherto it was thought that pressure and temperature were so bound together that the one might be made to do the work of the other. This belief was, as I have shown, but partially true.
As all volatile liquids enable us to tabulate the tensions of their saturated vapours, we can ascertain the relation between the pressure P and the temperature T. But these tables cannot be made up for high temperatures, and experience shows that at a certain temperature the liquid suddenly passes into vapour without changing its volume. For water, the point of change from the liquid to the vaporous state lies between 400° and 500° C. For ether, the liquid limit is lower, and for sulphurous acid it does not reach 250°. For carbonic acid and the protoxide of nitrogen it is still less, and lastly, for hydrogen oxygen, and nitrogen the liquid limit is at a point lower than the temperature of the ambient air. The foregoing facts show the universality of the law of cohesion, and have proved to me that the liquefaction of the permanent gases was to be obtained by the simultaneous employment of two agents, (1) great pressure, (2) great cold. The analytical method which has opened my eyes to these phenomena will be rich in results in similar fields of inquiry. All the laws relative to the variations of latent heat both external and internal, to the tensions of volatile vapours, to the mixture of gases with vapours—in short, the whole of calorimetry and thermo-dynamics - must henceforth be examined in the light of the general principles I have laid down.
A description of the machinery used by me for the liquefying of hydrogen and oxygen and for the solidifying of hydrogen would be out of place here. An extract from the Journal de Genève, December 23, 1877, describes accurately enough what may interest the general reader:-

By a double circulation of sulphurous acid and of carbonic acid, this latter gas is liquefied at a temperature of sixty-five degrees of cold, under a pressure of from• four to six atmospheres. The liquefied carbonic acid is conducted in a tube four metres long; two combined pumps produce barometric vacuum over that acid which solidifies in consequence of the difference of pressure. Inside the first tube, containing, as aforesaid, solidified carbonic acid, passes a tube of smaller diameter, in which a current of oxygen is caused by a generator containing chlorate of potash, and in the form of a large-sized thick-walled shell. Pressure may be brought up

to 800 atmospheres. Yesterday morning, the 22nd of December, under a pressure of 300 atmospheres, a liquid jet of oxygen gushed out of the extremity of the tube at the very moment when the compressed and cooled-down gas was passing from its high pressure to atmospheric pressure. To the beholder the gushing liquid oxygen is very much like a great rush of hot water from the hot-water cock of a bath.

By a singular coincidence, M. Cailletet of Paris succeeded in vctpo~rising oxygen the same day as I liquefied it, and a few days later M. Cailletet vaporised hydrogen and nitrogen. The experiments were made in the laboratory of the Ecole Normale, in the presence of MM. Boussingault, Henri Sainte-Claire Deville, Berthelot, Mascart, &c. These eminent men, says the Journal des Débats, declared themselves satisfied that the nitrogen was reduced to the condition of little drops, while the hydrogen became visible in the form of a vapoury cloud.
The method employed by M. Cailletet is that of a sudden reducing of a great pressure on the oxygen to that of atmospheric pressure. This sudden slackening of pressure produces a great external work, and is thus the cause of a great lowering of temperature. The refrigeration consequent on the slackened pressure may easily reach -200° C. below the initial temperature of the gas. The sudden reduction of 260 atmospheric pressures condenses the gas into a vesicular or vaporous form. This misty state of the gas is extremely transitory; for radiating heat almost instantaneously makes the drops of vaporous gas pellucid—in other words, brings them back to their gaseous form. In my researches, I have aimed at converting oxygen into a relatively permanent liquid to be collected in a receiver, so as to be able to measure its density and maximum tension.
On the 11th of January 1 forwarded from Geneva the following telegram to M. Dumas, Secrétaire Perpétuel de l’Académie des Sciences:-

I have just liquefied hydrogen with a pressure of 650 atmospheres and 140 degrees of cold. The gas solidified under the effect of evaporation. The jet had a flashing bluish colour somewhat like steel. The gushing jet fell on the ground, making a noise not unlike a heavy charge of shot, accompanied by a strident hiss. Lumps of hydrogen were kept intact in the tube.

M. Dumas immediately laid my discovery before the Société d’Encouragement. From a report of a lecture of his on the subject, I take the following lines:-

M. Dumas, the illustrious chemist, began by reminding his hearers with legitimate pride, that he had foreseen some forty years ago, in his Traité de Chimie, that hydrogen was the gaseous form of a metal. After dealing at some length with the inductions which had led him to that conclusion, M. Dumas laid some stress on the distinction to be made between M. Raoul Pictet’s and M. Cailletet’s experiments.
M. Cailletet has proved the possibility of reducing all gases to the liquid and the solid states. M. Pictet has really reduced the permanent gases to the liquid and the solid states.
And now to conclude. I here beg to thank the scientific world

for the handsome welcome they have given my discovery. Still, the scientific world is not the world. A suppressed titter has rippled on the faces of the ignorant, followed by the query: ’What’s the use of it?’ Well, it is perhaps the fault of the scientific world if so grovelling an exclamation is all but universal. Books, and especially manuals treating of physics, chemistry, and other sciences, lay more stress on tangible results than on the workings of the creative mind. Is it astonishing that man, who is naturally prone to value none but paying facts, should, when left to the mercies of a practical manual, remain unsympathising whenever his attention is called to the laws which are at the root of creation? The philosophical temper which reverences God on account of the perfection of His thought ought to be the fruit of scientific education. Science is a religion which can and ought to make man God-loving, by sedulously turning his mind to the divine first principles which rule the world.

[Cet article a été rédigé par M. Robert Harvey de Genève d’après les conversations que nous avons eues ensemble. Je lui en témoigne ici ma sincère reconnaissance. - RAOUL PICTET.]