The feeling has gradually been gaining ground in the public mind that, when atmospheric burners and other devices for burning coal gas are employed for heating purposes, certain deleterious products of incomplete combustion find their way into the air, and that this takes place to a considerable extent is shown by the facts brought forward in a paper read by Mr. William Thomson before the last meeting of the British Association.
Mr. Thomson attempted to separate and determine the quantity of carbon monoxide and hydrocarbons present in the flue gases from various forms of gas stoves and burners, but, like every other observer who has attempted to solve this most difficult problem, he found it so beset with difficulties that he had to abandon it, and contented himself with determining the total amounts of carbon and hydrogen escaping in an unburned condition, experiments which showed that the combustion of gas in stoves for heating purposes is much more incomplete than one had been in the habit of supposing, but his experiments give no clew as to whether the incompletely burned matter consisted of such deleterious gases as carbon monoxide and acetylene, or comparatively harmless gases, such as marsh gas and hydrogen. After considerable work upon the subject, I have succeeded in doing this by a very delicate process of analysis, and I now wish to lay some of my results before you.
If a cold substance, metal or non-metal, be placed in a flame, whether it be luminous or non-luminous, it will be observed that there is a clear space, in which no combustion is taking place, formed round the cool surface, and that as the body gets heated so this space gets less and less until, when the substance is at the same temperature as the flame itself, there is contact between the two. Moreover, when a luminous flame is employed in this experiment the space still exists between the cool body and the flame, but you also notice that the luminosity is decreased over a still larger area although the flame exists.
This meaning that, in immediate contact with the cold body, the temperature is so reduced that the flame cannot exist, and so is extinguished over a small area; while over a still larger space the temperature is so reduced that it is not hot enough to bring about decomposition of the heavy hydrocarbons with liberation of carbon to the same extent as in hotter portions of the flame. Now, inasmuch as when water is heated or boiled in an open vessel, the temperature cannot rise above 100°C., and as the temperature of an ordinary flame is over 1,000°C., it is evident that the burning gas can never be in contact with the bottom of the vessel, or, in other words, the gas is put out before combustion is completed, and the unburned gas and products of incomplete combustion find their way into the air and render it perfectly unfit for respiration.
The portion of the flame which is supposed to be the hottest is about half an inch above the tip of the inner zone of the flame, and it is at this point that most vessels containing water to be heated are made to impinge on the flame; and it is this portion of the flame, also, which is utilized for raising various solids to a temperature at which they radiate heat.
In order to gain an insight into the amount of contamination which the air undergoes when a geyser or cooking stove is at work, I have determined the composition of the products of combustion, and the unburned gases escaping when a vessel containing water at the ordinary temperatures is heated up to the boiling point by a gas flame, the vessel being placed, in the first case, half an inch above the inner cone of the flame, and in the second, at the extreme outer tip of the flame.
GASES ESCAPING DURING CHECKED COMBUSTION.
| Bunsen flame. | Luminous flame. | |||
| Inner. | Outer. | Inner. | Outer. | |
| Nitrogen | 75.75 | 79.17 | 77.52 | 69.41 |
| Water vapor | 13.47 | 14.29 | 11.80 | 19.24 |
| Carbon dioxide | 2.99 | 5.13 | 4.93 | 8.38 |
| Carbon monoxide | 3.69 | Nil. | 2.45 | 2.58 |
| Marsh gas | 0.51 | 0.31 | 0.95 | 0.39 |
| Acetylene | 0.04 | Nil. | 0.27 | Nil. |
| Hydrogen | 3.55 | 0.47 | 2.08 | Nil. |
| 100.00 | 100.00 | 100.00 | 100.00 | |
These figures are of the greatest interest, as they show conclusively that the extreme top of the Bunsen flame is the only portion of the flame which can be used for heating a solid substance without liberating deleterious gases; and this corroborates the previous experiment on the gases in the outer zone of a flame, which showed that the outer zone of a Bunsen flame is the only place where complete combustion is approached.
Moreover, this sets at rest a question which has been over and over again under discussion, and that is whether it is better to use a luminous or a non-luminous flame for heating purposes. Using a luminous flame, it is impossible to prevent a deposit of carbon, which is kept by the flame at a red heat on its outer surface, and the carbon dioxide formed by the complete combustion of the carbon already burned up in flame is reduced by this back to carbon monoxide, so that even in the extreme tip of a luminous flame it is impossible to heat a cool body without giving rise to carbon monoxide, although acetylene being absent, gas stoves, in which small flat flame burners are used, have not that subtile and penetrating odor which marks the ordinary atmospheric burner stove, with the combustion checked just at the right spot for the formation of the greatest volume of noxious products.