April 18, 1913. THE COLLIERY GUARDIAN. 811 developments. There is considerable congestion at the loading berths. The Durham coal trade generally is easier, especially as regards steam coals, bunkers, and coking sorts. Lancashire house coal continues in fair request. All other varieties are firm. Business in West Yorkshire is good. Stocks are light, and delivery has been regular. Trade in South Yorkshire is uniformly active, and the export trade is now showing more vivacity. The Derbyshire house coal trade continues satisfactory and there is no surplus. The demand for manufacturing fuel is as brisk as ever. The market at Cardiff is very firm, many fresh orders having come to hand. Small coals remain in strong demand and it is expected that the Belgian strike, if prolonged, will speedily affect this branch of the trade. Monmouthshire coals are firm. The Scottish coal trade is active in all departments. THE INFLUENCE OF INCOMBUSTIBLE DUSTS ON THE INFLAMMATION OF GASEOUS MIXTURES. Third Report of the Explosives in Mines Committee. The Explosives in Mines Committee have issued their Third Report. It contains the results of experiments made with a view of testing the conclusions published by the late Sir Frederick Abel on the danger of incombustible dusts in promoting the inflammation of otherwise uninflammable mixtures of firedamp and air. With the report are three appendices comprising respectively an abstract of Messrs. Wheeler and Burgess’s paper on “ The Lower Limit of Inflammation of Mixtures of the Paraffin Hydrocarbons with Air,” read before the C hemical Society in 1911 ; a note on the sampling and analysis of mine gases ; and another on the effect of inert dust upon explosions of gas and air in a small tube. We give an instalment of the Report itself below :— (1.) The Limits of Inflammation. It is well known that when a flame is introduced into a mixture of inflammable gas and air containing but little inflammable gas, a “cap” or aureole is formed’ around the flame; and that this cap is larger, the greater the proportion of inflammable gas present. But the formation of such a cap, though it may be very large, does not necessarily involve the combustion of all the inflammable gas present. In our view the proper definition of an inflammable mixture is a mixture in which flame can spread to any distance, independently of and away from the original source of ignition. It has often been observed that in a current of gas and air, passing a lamp flame, a cap or aureole may be separated from the burning body by which it is originated and carried along for a limited distance, giving a “ cloud ” of flame (which must not be mistaken for a general inflammation); and it is possible for flame conveyed in this way to ignite gases or other combus- tible substances in its path and thereby, perhaps, ultimately effect an explosion. But, unless this “ cloud” were fed by some mixture richer in combustible gas than that in which it originated, it would gradually die away. In a uniform mixture of gas and air the flame could only be carried on automatically if the inflam- mable gas were present in quantity above what is termed the “ lower limit of inflammation.” The rate at which a flame is self-propagated through a mixture just richer than this “lower limit” is initially slow; therefore, a current of gas and air may pass the lamp-flame faster than the flame can travel through the mixture. A “ flare ” or cone of flame will then spread outwards from the source of heat in the direction of the current, but will not pass back against the current. When a source of heat is introduced into a still mixture of a combustible gas and air, two things are necessary to ensure propagation of flame throughout that mixture; (i) an adequate means of ignition, and (ii) a sufficient proportion of inflammable gas. (i) We know by experience that minute sparks can be passed through even such an explosive mixture as elec- trolytic gas (2 H2 + O2) without exploding it, although the temperature of the spark is above the “ignition temperature ” of the mixture, and although some com- bustion indeed takes place in the path of the spark. When the intensity of the spark is increased (for instance, by including a condenser in the circuit), the gases will explode. Imagine a small sphere of hydrogen and oxygen 1 mm. in diameter at atmospheric pressure heated up by an electric spark ; the complete combustion of the gases in this sphere (whose volume is about J a cubic millimetre) would yield less than the one-thousandth of a (small) calorie of heat.* In order to spread the flame this heat has to be communicated to the spherical shell of unburnt gas in contact with the sphere (whose surface is 3T square millimetre) and bring a certain thickness of that shell to its ignition temperature. It may fail to do so. Let us suppose that the diameter of the heated sphere is doubled ; the volume of gases fired is now eight times as large, while the surface is four times as large as before. By increasing the size (or the intensity) of the spark a point is soon reached at which the combustion produced by the spark evolves enough heat to fire per se the adjacent gas mixture, and the flame is propagated. Gases, of course, differ in their ignition temperatures, and a gas mixture diluted with an inert gas, or under reduced pressure, requires a hotter spark to fire it than when under normal conditions. Methane (the chief constituent of firedamp) differs from hydrogen in having a higher ignition temperature and a more prolonged process of oxidation for complete combustion. It is for this reason, probably, that in the case of inflammable mixtures containing methane the source of heat necessary to inflame them must be more intense or must be longer in contact with them than in the case of hydrogen mixtures. (ii) Supposing that too small a proportion of com- bustible gas is present, only a small quantity of heat per unit volume of mixture is generated when the sphere surrounding the initial source of heat is inflamed, and the products of combustion have to impart heat to a considerable volume of “ inert ” gases. The number of collisions between molecules of combustible gas and of oxygen that are chemically effective is therefore small. Such collisions, resulting in combination, will occur only in the neighbourhood of the initial source of heat, around which an aureole or cap will form of a size dependent on the nature and quantity of the combustible gas present. As the proportion of combustible gas is increased, a greater quantity of heat is evolved per unit volume of mixture, and a smaller volume of inert gases is present to absorb it; until a point is reached when the amount of heat contained in the products of combustion is just sufficient to raise to its ignition temperature the adjacent gas mixture. Flame is then propagated progressively throughout the mixture without any necessity for the continued presence of the source of heat which started the inflammation, and the mixture is said to “inflame” or “explode” according to the rapidity of the propagation. To ensure propagation of flame, therefore, throughout the mixture, it is necessary (1) that the initial source of heat should be of a volume, intensity, and duration sufficient to raise a certain mass of gases in its immediate vicinity to the ignition-temperature of the mixture; and (2) that the heat contained in the products of com- bustion of this mass should be sufficient to raise the adjacent gas-mixture to its ignition-temperature. When a flame or other source of heat is introduced into a mixture of combustible gas and air containing more combustible gas than is required for a lower-limit mixture, inflammation will take place and become general, even though the igniting body is quite small, provided it is sufficient to start the action. On the other hand, the size of the aureole or cap formed in a non-inflammable mixture of gas and air depends largely on the size and temperature of the heated body which enables it to form. Just as the smallest quantity of any combustible gas which, when mixed with a given quantity of air (or oxygen), will enable self-propagation of flame to take place after ignition has been effected, is termed the lower limit of inflammation of the gas ; similarly, there is a higher limit of inflammation corresponding with the greatest quantity of inflammable gas, or, rather, with the smallest quantity of oxygen, that will allow self- propagation of flame. Experiments that have been made by different observers with a view to determining these limits have given rather discrepant results (vide Appendix I.), chiefly, we think, because different conceptions have been formed regarding the phenomena indicative of true inflammation; and sometimes because explosion-vessels of insufficient size have been used. As stated above, we regard a “lower-limit mixture” as one in which the self-propagation of flame can take place throughout, whatever the volume of the mixture, after the original source of ignition has been removed. It is possible, by introducing into a small vessel containing a mixture of methane (for example) and air a source of heat large in proportion to the vessel, to produce what appears to be a general inflammation of | * A small calorie is the amount of heat required to raise 1 gramme of water through 1 deg. Cent. the mixture when quite small percentages of methane are present. A lambent flame may appear to fill the vessel completely, but it is by no means certain on that account that the flame will travel throughout a mixture of the same composition but of larger volume. The definition of a lower-limit mixture which we give affords a clear distinction in kind between a local and a general inflammation; moreover, so far as we can ascertain, it conforms to the meaning of an “ inflammable mixture ” as understood by those engaged in the practical work of mining. A lower-limit mixture can be distinguished with certainty from one containing just insufficient com- bustible gas; for the momentary passage of an electric spark in a lower-limit mixture suffices to promote the inflammation of all the gas contained in the containing- vessel, however great its size, and further sparking reveals no further signs of combustion. Whereas with a mixture containing just insufficient combustible gas, although on first sparking the flame may appear to travel through the whole mixture when the containing- vessel is small, on passing the spark a second time a cap ” will appear above it, showing that the mixture still contains combustible gas. In connection with experiments referred to later, it appeared to us very important that the lower limits of inflammation of the gases that may be found in coal- mines should be known with as great accuracy as possible. An investigation into this matter had been carried out for the Mining Association Committee, and the results published in the Transactions of the Chemical Society. A reprint of this paper appears in Appendix I. It is shown in this paper that the percentage of each of the gases experimented with required to form a lower- limit mixture with air, can be calculated from a formula, and that the calculated values approximate very closely with the results of experiment. This formula is as follows:— Let L be the percentage of the combustible gas necessary to form a lower-limit mixture. Let 0 be its calorific value. And let & be a constant. Then— L = &0. The value obtained by experiment for the lower limit of inflammation of methane when mixed with air was 5’6. The calorific value of methane is 189T (large calories per gramme-molecule*). Substituting in the above equation we get a value for k of 1,059, and the relative values of L for other gases of the paraffin series can then be calculated. Thus, for ethane the calorific value is 336’6; whence L = = 315. 336’6 The experiments show that the following are the lower limits of inflammability for the lower members of the paraffin series of hydrocarbons :— Per cent. Methane ........................ 5 60 Ethane ......................... 310 Propane ........................ 2 17 n-butane........................ 1 65 n-pentane ...................... 137 iso-pentane..................... 1*32 The corresponding limits of inflammability of hydrogen and carbon monoxide are, as generally given:— Lower limit. Higher limit. Hydrogen............... 10 00 63 00 Carbon monoxide ....... 15 90 74 50 In view of the fact that the presence of other paraffin hydrocarbons beside methane in firedamp might render its lower limit of inflammability lower than that usually accepted (about 5 J per cent.) we considered it advisable to procure from different coalmines specimens of such gases as were believed to be of a particularly inflam- mable nature. In nearly every case the only paraffin hydrocarbon found was methane. Thus the results for 10 different samples were:— Sample No. Methane, per cent. Other paraffins, per ct nt. 1 8475 1 85 o 86’50 045 3 90 40 040 4 78 00 nil 5 90 65 nil 6 89 45 nil 7 9510 nil 8 86’20 nil 9 84 80 nil 10 8875 nil It is possible, though not probable, that some samples might be found to contain a higher proportion of other * A large calorie is the amount of heat required to raise a kilogramme of water through 1 deg. Cent, when the tempera- ture is between 18degs. and 20degs. The gramme-molecule is the weight in grammes of any substance which is equal numerically to its molecular weight. Thus, for methane the gramme-molecule is 16 grammes.