Slow Combustion of Hydrogen

Owing partly to its familiarity, and partly to the ease with which it can be obtained in a highly pure condition, electrolytic gas has been studied by many investigators from the point of view of slow combustion of Hydrogen.

In 1803 Hooke observed that if the gas is allowed to stand for some months in the presence of water, the dissolved hydrogen and oxygen enter into combination. This has been confirmed by Marcacci in more recent years.

The presence of colloidal platinum in water in contact with electrolytic gas accelerates the reaction, the rate of formation of water being proportional to the concentration of the platinum and the pressure (i.e. the rate of solution) of the gases. Many other surfaces also accelerate the reaction.

It has further been shown that in the course of several months a mixture of hydrogen and oxygen when moist and exposed to daylight shows signs of chemical combination, although the action is inappreciable during the course of an ordinary experiment.

If the temperature is raised slightly in contact with certain metals, such as platinum, the rate of union of the gases is greatly accelerated. Thus, compact platinum acts at 50° C., the gases combining with measurable velocity. Finely divided silver acts at 150° C., thin gold leaf at 260° C., and even fragments of non-metallic bodies such as charcoal, pumice, porcelain, quartz, and glass are active at temperatures below 350° C. Angular pieces of glass are found to be more efficient than glass balls of equal superficial area.

Such being the case, it is clear that the walls of a containing vessel may exert an enormous influence upon the slow combustion of its gaseous contents. This is evidenced by the very varying results obtained for the lowest temperatures at which hydrogen and oxygen have been observed to unite with measurable velocity when heated in glass vessels. Thus, Bone and Wheeler kept electrolytic gas in seven different glass bulbs at 350° C. for several days, and found no combination had taken place in six of them after one week, although in the case of the seventh bulb, in which the glass had become devitrified at one end, the presence of water could be detected. At 400° C. no change was observable in three bulbs, but after a week one of the bulbs contained water, although the other two were apparently unchanged.

These results clearly indicate the influence of the glass, and it is interesting to compare them with those reached by Meyer and Raum, who obtained evidence of combination at considerably lower temperatures than the above. Their results were as follow:

Combination of electrolytic gas

Temperature, ° C.	Period (days).	Remarks.
100	218	No combination.
300	65	Water detected.
350	5	Water detected.
448	. . .	Slow combination.

Clearly the temperature of 400° C. may be regarded as the border-line temperature of the slow combustion of Hydrogen or electrolytic gas.

These data, however, are merely qualitative in character. In 1906 Bone and Wheeler published the results of a very thorough quantitative investigation of the reaction at about 450° C. in the presence of several different types of catalysers. These were as follow:

1. Refractory acidic oxide – porcelain.
2. A basic refractory - magnesite.
3. Easily reducible oxides - oxides of copper, iron, and nickel.
4. Compact metals - gold, nickel, platinum, and silver.

The catalyst chosen was packed into a hard glass combustion tube, heated to the desired temperature, and the gases, measuring some 1500 c.c. in toto, were continuously circulated throughout the system. Any combination to form water was measured by observing the fall in pressure. The majority of the experiments were carried out with porcelain as catalyst, and it was found that the rate of combination of hydrogen and oxygen in electrolytic gas is directly proportional to the pressure of the dry gas.

In other words, the reaction is monomolecular, although, from the equation

2H₂ + O₂ = 2H₂O

a reaction of the third order is to be expected. By increasing the proportion of either the oxygen or the hydrogen above that required for the foregoing equation it was found that the rate of the reaction was directly proportional to the pressure of the hydrogen. A result so opposed to that which might be expected indicates that the reaction is indirect and complicated; this conclusion receives support from the further observation that previous exposure to hydrogen appreciably enhances the catalytic activity of the porcelain, although chemical reduction of the porcelain by this preliminary treatment is out of the question. Indeed, if reduction did take place, a prolonged preliminary exposure to hydrogen might be expected to enhance the catalytic action. But experiment showed that such was not the case. Further, the hydrogen could easily be removed again by heating the porcelain to redness in a vacuum. A preliminary ignition in oxygen did not appear to influence the results. It may therefore be concluded that porous porcelain adsorbs both hydrogen and oxygen at rates which depend to some extent upon the physical condition and past history of the surface. In general, the adsorption of hydrogen is less rapid than that of oxygen, but the limit of saturation is higher. Combination between the adsorbed gases occurs at a rate either comparable with or somewhat more rapid than that with which the film of occluded oxygen is renewed, but at a rate considerably higher than that of the adsorption of hydrogen.

When magnesite - a typical basic refractory - was used as catalyst, the temperature being 430° C., closely similar results were obtained. Ferric oxide and nickel oxide showed no reduction during the process, but behaved in an analogous manner to the magnesite. Copper oxide, however, exhibited unique behaviour. The rate of formation of water was not only slow, but was proportional to the partial pressure, not of the hydrogen, but of the oxygen when the proportions of the twro gases did not correspond to that in electrolytic gas. The authors explain this by supposing that the surface of the oxide becomes coated with a film of '' active " oxygen, which protects it from the reducing action of the hydrogen. At low pressures water-vapour is formed because the hydrogen succeeds in penetrating the attenuated film.

Gold was studied at 250° C., and from the fact that its surface remained apparently unchanged throughout, the conclusion was drawn that the metal acted through adsorption or occlusion, and not in a chemical manner. The surface of silver gauze, however, when employed as catalyst, was observed to become frosted over, and its catalytic activity greatly increased. It would appear, therefore, that some chemical action takes place, such as the formation and subsequent decomposition of a hydride. An oxide is ruled out since silver oxides are unstable above 350° C.

As is well known, the noble metals are very powerful catalysts, and their activities have been carefully studied. Pre-treatment of platinum, palladium, or iridium with oxygen greatly enhances their effectiveness as regards the catalysis of mixtures of hydrogen and oxygen. Pre-treatment with hydrogen produces a similar effect but to a less marked extent.