﻿PROXIMATE ANALYSIS OP COALS. 63 



a coke oven. Probably, by the smoking-off method this is more nearly 

 attained, since seldom more than seven to nine minutes are required 

 to expel the volatile matter. The curves demonstrate clearly that the 

 smoking-off method is more reliable for general use than the official 

 method. Varying quantities of water have no effect with the former 

 while the curves obtained by the latter under similar conditions are more 

 or less variable; therefore, I heartily recommend the smoking-off method 

 for general adoption. 



In order to ascertain how much of the entire "seven minutes over 

 the full heat of a Bunsen burner" is necessary in the analysis of coal 

 by the official method, the following experiment was made on Australian 

 coal ISTo. 1. About one and a third minutes were required to expel the 

 gases which came off at a rate fast enough to burn. The results indicat- 

 ing the influence on the percentage of total volatile matter when the 

 crucible and sample were subjected to the regulation flame for varying 

 lengths of time are noted below. 



Time over 

 tie full flame 

 in minutes. 



Total 



volatile matter 



per cent. 



3 



38.65 



4 



38.50 



■ 7 



38.65 



Furthermore, to determine with the smoking-off process, whether or 

 not the subjection to the full heat of the Bunsen burner for seven minutes 

 is necessary or to what extent essential, a number of experiments were 

 made on Negros coal No. 21. The samples were carefully smoked off and 

 then heated over the regulation flame for varying lengths of time as 

 noted below, with the following results : 



Time over 



the full flame 



in minutes. 



Total 



volatile matter, 



per cent. 



1 



48.53 



3 



49.46 



4 



50.27 





5 

 6 



7 



50.24 



50.21 



19 50.24 



> Constant 



It has already been shown 20 that heating until all gases apparently 

 cease issuing from the crucible is not sufficient. In the experiments with 

 finely powdered coal such a sufficiency was most nearly attained. My 

 results show that by either method the gases are all expelled after four 

 minutes of heating, also that there is no loss on further heating. In the 

 analyses of these particular coals considerable latitude might bo allowed. 



w 



"The average of four results, the greatest variation of wliieli is 0.06 per cent. 

 20 Wright, L. T.: J. Soc. Chem. Ind. (1885), 4, 657. 



