﻿PROXIMATE ANALYSIS OF COALS. 49 



14 23 34 3° 



These give the ratios -~ = 1A3 and Tjr7g = l-79, and show that here 



also the variations in the percentage of ash and fixed carbon are not 

 exactly proportional. 



In the determinations made by the transition method the escape of 

 some of the incandescent particles was noticeable but it has been thought 

 that this was negligible. The above experiments show this to be a 

 false supposition; that for very accurate work more care must be exer- 

 cised than is outlined for that method. Since the variations between the 

 percentages of ash and fixed carbon as determined by the smoking-off 

 process and the transition method are exactly proportional the cause 

 must be solely mechanical loss. 



As previously stated, the shower of sparks driven off when the official 

 method is used is proof of a large mechanical loss; the analytical data 

 which corroborate this, show that it is several per cent. The variations 

 in ash and fixed carbon as determined by the smoking-off process and the 

 official method in samples I, II, and III are always in the same direction, 

 but are not exactly proportional and hence can not be accounted for solely 

 by the theory of mechanical loss. Other factors prevent proportional 

 variations. It has been suggested 10 that this -may be partially due to a 

 different breaking down of the hydrocarbon compounds when expelled 

 under different conditions of heat treatment, in the presence of variable 

 amounts of moisture, etc. This is true in the case of coking coals for it 

 has often been observed that both the qualitative and quantitative compo- 

 sition of coke depends not only on the nature of the coal used but also on 

 the conditions of the distillation that is, the temperature, pressure in the 

 retort, the time, the size of the coal, etc. However, in lignities and non- 

 coking coals, where the gases given off are mostly of comparatively very 

 simple composition, the variation between the official and the modified 

 methods is to a greater extent due to mechanical loss. 



Comparisons of a number of samples will show more clearly just how 

 much difference really exists in the breaking down of the hydrocarbon 

 compounds of the coal. The difference shown in the breaking down 

 between the transition method and the official one is the same as between 

 the smoking-off process and the official method, since the variations 

 in the percentage of fixed carbon and of ash as determined by the transi- 

 tion method and the smoking-off method are exactly proportional. The 

 transition method is an improvement over the official one in certain 

 cases but as there still are mechanical losses by its use, it is set aside 

 as less satisfactory than the smoking-off process. 



10 Somermeier E. E. : hoc. cit. 

 51214 4 



