Fire Safety Science Digital Archive

The production of smoke in fires constitutes a serious hazard. Smoke can drastically reduce visibility and has unpleasant physiological effects. In addition the heat radiated by flames is related to the concentration of radiating solid particles within the flame. It would be useful to have some information on those factors which might influence smoke produotion by flames. A major component of smoke from flames is soot, a solid material generally of high carbon content. This material makes a large contribution towards its obscuring power. Information on the rate of soot production by diffusion flames under various combustion conditions would therefore be appropriate to an investigation of the smoke production problem. At present there is surprisingly little quantitative information on the rate of soot production, even under laboratory conditions. In enclosed compartments the composition of the atmosphere in which flaming combustion is taking place will not, in general, correspond to that of air. The composition will be modified by the presence of combustion products and decomposition products; the use of inerted atmospheres or of chemical extinguishants to control the fire will cause further modification of the atmospheric composition. Smoke production might be expected to vary within such changes in combustion conditions and it would therefore be useful to investigate rates of smoke production when different diluents are added to a flame. Measurement of sooting rates, rather than smoking rates, would probably be experimentally simpler. The literature on soot formation in flames is extensive and the results are often conflicting. In part this is due to the variety of systems which have been investigated. Reference to Appendix 2 and Section 3 of this report will show that the conflict is particularly pronounced as far as the effects of additives and diluents are concerned. Most investigations have been concerned with measurements of "smoke points", i.e. the critical fuel flow (diffusion flames) or mixture strength (premixed flames) at which soot just begins to leave the flame. Quantitative information is confined, for the most part, to measurements of carbon inside the flame. In the present context we are concerned, rather, with the quantities of unburnt carbon which leave the flame. It should be made clear at this point that soot produced by flames is not elementary carbon. It may contain hydrogen and oxygen in atomic percentages in excess of 30%, particularly under mild combustion conditions. Briefly, soot produced under mild combustion conditions has the characteristics of a polybenzenoid hydrocarbon; it may contain volatiles of lower molecular weight, some of which may possess oxygenated functional groups. Under stronger combustion conditions a soot of higher carbon content, and with interstitial hydrogen and oxygen set in a strained graphite lattice, is formed. In this report the term "soot" is used to cover solid material given off by flames. The word "carbon" will be used to describe such material inside the flame, on the understanding that such a term may not adequately describe its chemical nature. It will be appropriate to begin with a brief mention of current ideas on the mechanism of soot formation in flames. Relevant information on those factors which are known to influence soot production will then be summarised. Both diffusion and premixed flames will be discussed.