PRESIDENTIAL ADDRESS. 369 



range, whicli is, within the limits of experimental error, the same tor botli air 

 and oxygen ((.<'., in the case of hydrogen it is 580° to 590°, and for carbon 

 monoxide 640° to 658°) ; on the other liand, in cases where the mechanism of 

 combustion is known to be very complex (e.g., hydrocarbons), the ignition range 

 is either fairly wide or is materially lower in oxygen than in air (or both), thus : 



The explanation of such behaviour is probably to be sought in the known 

 complexity of the combustion, and the marked tendency for appreciable and 

 fairly rapid interaction between the inflammable gas and oxygen before the 

 actual ignition-point is reached. 



If, by any means, such preliminary interaction could be entirely suppressed, 

 or if, on the other hand, it be very rapid in character, the observed ' ignition- 

 range ' would be narrowed, as is actually the case with ethylene (542° to 547° in 

 air and 500° to 519° in oxygen). 



There are two other means by which an explosive mixture may be ignited ; 

 one is by adiabatic compression and the other, and most commonly employed 

 of all, is by the passage of an electric spark. 



The adiabatic compression of an explosive mixture was originally suggested 

 by Nernst as a means of determining its ignition-point, provided always 

 (1) that ignition is not produced locally whilst the main of the gas is still 

 far below the true ignition temperature, and (2) that the piston of the 

 apparatus does not move appreciably after the gas has been raised to its 

 ignition-point. At the time of my 1910 Report, Falk,^ in America, had applied 

 the method in the case of hydrogen and oxygen mixtures, with results which, 

 in the light of more recent work, would appear to have been misleading or 

 erroneously interpreted. Thus, for instance, he found that, of all the mixtures 

 of hydrogen and oxygen, the equimolecular H„-fO„ mixture has the lowest 

 ignition-temperature (514°), from whicli he concluded that the gases react initially 

 to produce hydrogen peroxide rather than steam. Such a conclusion, which I 

 believe to be erroneous, naturally directed attention to the experimental method 

 involved. 



The subject was promptly taken up here in Manchester by H. B. Dixon and 

 his co-workers,'' with the result that much new light has been thrown on the 

 phenomena accompanying ignition. The ratio of the ignition-temperature to the 

 initial temperature of the mixture before compression, both expressed in degrees 



T /V \ 



absolute ~, may be calculated from the compression ratio, (j^)» by means 



of the well-known formula for adiabatic compression : 



(?,) = (v;)^-' 



where f = the ratio of the specific heats at constant pressure and volume 

 respectively of the mixture compressed, and which for a mixture of diatomic 

 gases, such as hydrogen and oxygen, is usually taken as 1.40. 



Dixon's recent photographic analysis of the appearance of flame when 

 mixtures of carbon bisulphide and oxygen (CS, + 30„) are adiabatically com- 

 pressed, have proved that the flame, starting from a point or layer, always takes 

 an appreciable time to spread through the mixture, and that unless special 

 precautions are taken to arrest the piston at the moment of attainment of the 

 ignition condition, it may be driven in much further than the minimum distance 

 for ignition. The real ignition-point, as above defined, is not necessarily syn- 

 chronous with the actual appearance of flame; there may be, and usually is, an 



* Journ. Amer. Chcm. Soc. 28, 1517; 29,1536. 



♦H. B. Dixon, L. Bradshaw. and C. Campbell, 7'/«/i.'. Chcm. Soc. 1914, 

 105, 2027 ; H. B. Dixon and J. M. Crofts, ibid. p. 2036. 



1915. ' B B 



