Coal-gas Flame

The coal-gas flame is practically a replica of the candle flame, for it may be divided into three analogous zones, and the phenomena attendant upon its luminosity are closely similar to those already detailed.

Examination of the spectrum of illuminating gas leads to the conclusion that two superimposed effects must be considered. Firstly, there is the continuous temperature spectrum, caused by the incandescent particles floating in the Coal-gas Flame, and secondly, a band spectrum caused possibly by the luminous particlcs in the act of burning in the outer mantle of the flame, or perhaps by their products of combustion.

If air is admitted into the heart of the flame in sufficient quantity, the luminosity suddenly disappears almost entirely, and a flame resembling that of pure hydrogen results. Owing to the rapid combustion of the hydrocarbons, no luminous particles are separated, and the flame, being intensely hot, is a convenient one to employ for heating purposes, since it yields no soot when made to impinge upon a cold object. This is the principle of the Bunsen burner, the flame of which consists of two parts only, both of which are non-luminous. The inner zone is essentially a mixture of unburned gas with air. It is warmed by radiation from the outer zone, and ranges in temperature from about 300° C. a little above the nozzle, to 1000° C. at the apex, where rapid combustion commences. The outer zone is a hot mantle of burning gas, the temperature of which may rise to about 1550° C., or even as high as 1760°±50° C. Combustion is here fairly complete.

Carbon monoxide may appear amongst the products of combustion, principally as the result of sudden cooling of the flame. It appears that an incandescent mantle tends to act in this way, and cause appreciable amounts of carbon monoxide to escape. The amount depends upon the type of burner, and is independent of the amount of carbon monoxide in the original combustible gas between the limits of 0 and 60 per cent.

The Smithells separator

When a mixture of a combustible gas with air or oxygen is burning at the mouth of a tube, there is a constant tendency for the flame to "strike back," this explosive tendency being counteracted by the velocity with which the gaseous mixture passes along the tube. By adjusting the relation between the proportion of oxygen in the gas and the rate of flow, it is possible to effect the separation of a flame into its component parts, because when the rate of flow of the gas is just insufficient to maintain the combustion at the end of the tube, the inner portion of the flame recedes towards the origin of the supply of the gaseous mixture, whilst the outer mantle of the flame, although enfeebled, remains in the original position. By providing a constriction inside the tube, a local increase is caused in the rate of flow of the gas, and the receding portion of the flame can be arrested at this point. This is the principle of the Smithells Separator, which may be conveniently made by fixing, with the aid of a piece of rubber tubing, a short glass tube some 6 inches in length, on to the nozzle of a Bunsen burner. A second, wider tube is slipped over and fixed in position by means of a cork. The tops of the two glass tubes are protected with metal foil to prevent cracking at A and B. The gas is turned on full and ignited at the top, yielding the ordinary Bunsen flame. By turning down the gas and regulating the air supply, the flame becomes unstable and the inner portion recedes down the tube taking up a fixed position at A. At this point partial combustion only ensues, the products being CO, CO₂, H₂, and H₂O. In the upper flame combustion is rendered complete.

The effect is very beautiful if carried out in a darkened room.

Price's experiment (1920)

A curious experiment has been described by Price in which a flame is obtained very similar to the inner flame of the Smithells separator. The experiment consists in fixing a glass chimney over the base of a Bunsen burner, and a horizontal wire an inch or two above its upper end (fig.). If the height of A above the burner and the rate of flow of gas are correctly adjusted, upon applying a light at C the gas burns with a pale blue flame, and bears a resemblance to the " gas cap " of a miner's lamp.

When sodium chloride is sprinkled on a glowing coal fire a characteristic blue flame results, which is attributable to the presence of salts of copper originating in the pyrites in the coal. This was demonstrated by Salet who, in 1890, not only identified the flame spectrum as that of copper chloride, but succeeded in separating metallic copper from the fuel ash. The colour is not due to carbon monoxide, the two flames being quite distinct.

It has recently been observed by de Forest that flames can be employed for direct production of telephonic currents by sound-waves.

Mention has already been made of the fact that the introduction of air into the heart of the flame enables rapid combustion to take place without the separation of luminous particles, so that the Bunsen flame tends to lose the luminosity characteristic of coal gas. This, however, is not the entire cause 6; there are several contributory factors. For example, the air introduced into the flame is cold and thus tends to cool the whole. Again, the formation of intermediate luminous bodies is retarded by the nitrogen which serves as a pure diluent and elevates the temperature necessary to effect the partial decomposition of the hydrocarbons.