SPONTANEOUS COMBUSTION.

Progress, Volume II, Issue 3, 2 January 1907, Page 97

SPONTANEOUS COMBUSTION.

By Prot Easterfield, Victori \ Collegi", Wellington

O the large numbei of fires which occur annually it is probably correct to say that the majority may be classed under one of three heads, fires due to incendiarism, fires due to ordinary carelessness, and fires due to spontaneous combustion. It is of course easy to imagine cases, for instance, fires following earthquakes which fall outside of the above classification ; or again cases may occur in which the fire belongs partly to one and partly to another of the above three classes The classification suggested will, however, be found to cover the majority of cases Of these three types it is the spontaneous fires to which the greatest interest attaches They are, to say the least, uncanny, and they are generally regarded as entirely beyond control. Let us see whether we can learn something of this class of phenomena. In the first instance we must recognise tha£ chemical changes are taking place around us perpetually, of which we take no heed until they are brought to our notice by some special circumstance Let us take an extreme case A block of granite, exposed to the weather, appears to be undergoing no change , nevertheless, the examination of any piece of granite that has been so exposed for years will show that great changes have taken place, and in particular that the feldspar is breaking down and yielding white china clay Iheie is reason to believe that the weathering of the feldspar is accompanied by heat evolution, but so slow is the change that the heat evolution cannot be measured ; for the equalising agencies of radiation and conduction dissipate the liberated heat so quickly that there is no perceptible rise of temperature Take again the case of the exposure of an non rail to weather even the most cateless observer knows that rusting will take place Now, careful observation has shown that ordinary lusting or atmospheric oxidation of iron requires the presence (a) of moisture, (b) of carbonic anhydride, (o) ot oxygen ; and that the rate, and this is important, at which rusting will occur is considerably greater in a warm atmosphere than in a cold one Now, if the atmospheric oxidation of the iron take place at a sufficiently high temperature, the rate of chemical change is so gieat that the metal glows and becomes hotter and hotter up to a definite limit, a fact which is made use of in the oidinaiy Bessemer steel converter Any circumstance then which hastens the rate of chemical action may be regarded as a possible source of fire risk The circumstances which affect the late of any particular chemical action are chiefly these (1) The temperature at which the reaction is carried out , (2) the presence of so-called catalytic substances -which though apparently unchanged, serve in some way to profoundly modify the course of the action , (3) the state of subdivision, and hence the area and nature of surface of the substances which are undergoing change. I Temperature In nearly all cases it is found that substances which enter slowly into chemical reaction in the cold will react vigourously at a higher temperature, the rate of action increasing in geometrical progression as the temperatuie rises in arithmetical progression When once, then, the temperature has risen to such a point that the rate of heat evolution is greater than the rate at which heat can be dissipated, the conditions for an ultimate flare are piesent Conversely, the cooling of a burning substance to a certain limit will cause the burning to stop, a fact well illustrated by the familiar experiment of extinguishing a burning candle by placing a spiral of cold copper in the flame At the tempeiature of liquid air very few substances appear capable of chemical action II The influence of foreign substances in bringing about chemical actions may be illustrated in

many ways. (1) A trace of spongy platinum will almost instantly cause a jet of coal gas to ignite in the air, the platinum remaining unchanged. (2) A small quantity of iron oxide may render chlorate of potash explosive, though the oxide appears to be itself unacted on (3) In the absence of a trace of moisture, phosphorus burns with difficulty, even in oxygen. 111. As illustrating the influence of state of subdivision upon the rate of chemical reaction, it may be mentioned that very finely divided iron and lead oxidise so rapidly in atmospheric air that they eventually catch fire — an exceedingly interesting experiment. If we once firmly grasp the idea that all combustible substances exposed to the air are really undergoing a slow combustion process, and that it is only necessary to hasten the process m order to get an outbreak of fire, we shall, I thmk, be filled with wonder that the number of such so called spontaneous (Latin sponte — of fiee will) fires is not much greater than we actually find it to be. Let us now turn to a few specially interesting cases of spontaneous combustion. I Firing of coal in a ship's bunkers This is perhaps the commonest of the spontaneous firings. Experience shows that such fires occur most readily if the coal be finely divided, if the coal be moist, if the quantity stocked in one bunker be large, and if the ventilation be inefficient. With small quantities and proper ventilation the heat generated by the slow combustion is carried away as rapidly as it is generated, and hence no dangerous rise of temperature can occur Highly sulphurous coals containing finely divided iron pyrites are particularly to be regarded wi<h suspicion. II Firing of haystacks This phenomenon is almost invariably due to the stacking of hay m an insufficiently dried condition The rise of temperature under such circumstances is often extraordinarily rapid, and in the first instance appears to be undoubtedly due to bacterial action. The tempeiature being thus raised to a point at which atmospheric oxidation of the vegetable matter can proceed more rapidly than the generated heat can be dissipated from the mass actual incandescence will eventually be brought about In the making of ensilage the original fermentation and rise of temperature occur, but, owing to the fact that the silo is under great pressure, the supply of air is insufficient for rapid oxidation to further increase the temperature of the mass. 111. Firing of greasy rags. Many mills have been destroyed owing to this cause. The initial rise of temperature is due to rapid absorption of oxygen by the large surface of grease exposed to the air in a thm layer Most animal and vegetable oils are liable to cause these accidents, but of all oils linseed oil appears to be the most dangerous on account of its great affinity for oxygen. IV. The cause of wool fires on ships appears to offer no special points for consideration The evidence appears to be overwhelming, that if wool is shipped in a clean and dry condition the risk is practically nil, but we shall await with interest the report of the Royal Commission which is now investigating the phenomena from the New Zealand standpoint. V Spontaneous combustion of explosives. This is a peculiarly interesting case. In an explosive we place certain substances which have a great tendency to chemical action in such close proximity that they will not, we hope, act upon one another rapidly at the ordinary temperature When we wish to bring about the explosion we hasten the action to begin either bv detonation or rise of temperature Sometimes the leaction begins " spontaneously" if it does there is little evidence for the coroner