Dissociation of Steam

From physico-chemical considerations it is probable that even under ordinary conditions liquid water contains an exceedingly minute though definite proportion of uncombined hydrogen and oxygen in equilibrium with the compound molecules. This state of equilibrium is outside the scope of experimental detection unless disturbed in some way, as by the influence of ultra-violet light, when the decomposition may become appreciable.

It is more easily observed at higher temperature, because with rise of temperature the position of the equilibrium moves in favour of a higher proportion of dissociated molecules. In 1847 Grove noticed the formation of some free hydrogen and oxygen when platinum, heated almost to fusion, was dropped into water, the experiment being repeated by Deville a little later, with an even more decisive result. The main difficulties in detecting the thermal dissociation are the smallness of its extent, and especially the tendency of the gases to recombine on cooling. The latter can be obviated by cooling the gases very rapidly; for example, by passing a current of steam over a white-hot platinum wire or over a gap between sparking electrodes in such a manner that the gas is removed so rapidly from the heated area that a portion of the products fails to recombine. A similar effect was achieved by Deville by means of his "hot and cold" tube, in which steam was passed through the space between an externally heated porcelain tube and a water-cooled inner silver tube; hydrogen and oxygen produced by dissociation at the hot surface become cooled below the temperature of rapid recombination by the inner cold surface. The difference in the velocity of diffusion of hydrogen and oxygen can also be applied (Deville) by passing steam through a heated unglazed tube when, on account of the more rapid passage of the hydrogen, an excess of oxygen is to be found in the vapour issuing from the tube.

Lowenstein's apparatus for determining the dissociation of steam.

Lowenstein's method consisted in passing a slow current of water- vapour through a tube heated to various temperatures, and containing a closed platinum vessel, P, attached to a manometer, M (fig.). The water-vapour dissociated, and hydrogen passed through the platinum, causing a rise in pressure, which became constant for any one temperature. Since platinum functions as a semipermeable membrane in that it is impermeable to oxygen and water-vapour, it was easy to calculate from the pressure registered by the manometer the extent of dissociation of the steam.

Holt in his later experiments used a glass globe of 4 litres capacity containing a short length of platinum wire which was electrically heated. The globe was evacuated and then connected with a vessel containing water, the vapour from which passed into the globe and underwent partial dissociation in contact with the wire. When equilibrium was attained for the particular temperature chosen, the vessel was cooled, the gases pumped off and measured after the water-vapour had been frozen out. From these data the dissociation pressure was calculated.

Holt also endeavoured to determine the lowest temperature at which water-vapour appreciably decomposed in contact with heated platinum wire. Minute quantities of gas were collected at about 750°, C. No doubt, had it not been for the solubility of the gases in the condensed vapour at the conclusion of the experiments, traces would have been detected at even lower temperatures.

Bjerrum employed an explosion method and obtained the following results for 1 atmosphere pressure:

Temperature abs	1705	2257	2642	2761	2834	2929
Percentage dissociation	0.108	1.79	4.3	6.6	9.8	11.1

Nernst and Wartenberg have given several formulae, based on thermodynamic considerations, whereby the percentage degree of dissociation, x, of steam at any temperature may be calculated. The first one was used in a somewhat simplified form both by Langmuir and by Holt, namely:



By neglecting x compared to 1 and dividing by 3,



where T is the absolute temperature. The value for k given by Nernst and Wartenberg is 3.83. Calculated from the experimental data of Langmuir, k = 3.79; a slightly higher value is found by Holt, namely, 3.806, whilst Lowenstein's results yield the value, 3.80 (up to 1968° T). This latter may therefore be accepted as the most probable value for the equilibrium constant of water-vapour, oxygen, and hydrogen.

In the following table are given the percentage degrees of dissociation of water-vapour at temperatures ranging from 1000° to 3500° abs., and under pressures varying from 0.1 to 100 atmospheres.

Percentage degree of dissociation of water-vapour at various temperatures and pressures

Absolute Temperature	0.1 Atm.	1.0 Atm.	10 Atm.	100 Atm.
1000	0.04556	0.04258	0.04120	0.05556
1500	0.0433	0.0202	0.00935	0.00433
2000	1.25	0.582	0.270	0.125
2500	8.84	4.21	1.98	0.927
3000	28.4	14.4	7.04	3.33
3500	53.1	30.9	16.1	7.79

Steam is slightly decomposed when subjected to an electric discharge, hydrogen and oxygen being liberated, curiously enough, sometimes at one terminal and sometimes at the other. The nature of the spark appears to be the determining factor. With long sparks the hydrogen appears at the negative and the oxygen at the positive pole; with short sparks the positions are reversed, and the process is opposite to that taking place in the ordinary electrolysis of liquid water.