200 MUNROE-HOWELI^PRODUCTS OF DETONATION OF TNT. 



or not until later, we will leave undecided. When analyzing the gas one natu- 

 rally has to absorb these hydrocarbons first by means of bromine and fuming 

 sulphuric acid, for otherwise the methane determination will be incorrect. If 

 the explosion gases are lead off through ammoniacal silver solution, a small 

 precipitate is obtained which we claim to be acetylide. Moreover, the explo- 

 sion gases contain in small quality products of the incomplete combustion, 

 which probably account for their disagreeable sharp odor. We presume, on 

 account of the color of fuchsine sulphurous acid, unsaturated aldehyde. 



We found further, in the explosion gases of trinitrotoluol significant 

 quantities of ammonia, about 6 per cent. We ascertained this while we were 

 introducing the gases and the ammonia set free from the residue of the ex- 

 plosion by potash-lye, into standardized muriatic acid. The ammonia could 

 have been formed, as Kast assumes, by direct combination of nitrogen and 

 hydrogen. We are not willing to subscribe to this opinion, for then we 

 should not be able to explain why only insignificant small quantities of am- 

 monia are formed when picric acid is detonated. We are much more of 

 the opinion, supported by Sabatier's experiments, that the ammonia is formed 

 from nitric oxide and hydrogen. The nitric oxide must be formed in quan- 

 tities which correspond to the equilibrium between oxygen and nitrogen. 

 According to Sabatier^ 2NO -f- 2H2 = No + 2H0O (I). But in the case of 

 greater hydrogen concentration, ammonia is formed, 2NO -\- 5H2 := 2H2O 

 -I-2NH3 (II). In the case of picric acid only small quantities of hydrogen 

 are contained in the explosion gases, therefore equation I is substantially real- 

 ized. But in the case of trinitrotoluol the nitrogen present is converted into 

 ammonia. These reactions occur with great speed; whether or not they are 

 to be considered in arriving at the disintegration equation, later experiments 

 will have to prove. 



Bichel in his "Table III. of 1904" definitely states that the data 

 presented there is given in percentages by weight. Smith in report- 

 ing the analyses of the different products of these experiments treats 

 them as if they were stated all in the same terms whereas the re- 

 sults in the second and third experiments he cites appear to be stated 

 in percentages by volume. Poppenberg and Stephan do not state 

 the terms in which they report their results, evidently assuming 

 that the results of the analysis of a gas will be understood as being 

 given in percentages by volume since this is the usual custom, and 

 an examination of the data justifies this opinion. 



Taking the data of Bichel's experiment and subtracting the solid 

 C the percentage by volume of the residual permanent gases under 

 normal conditions may be found. The following result is thus 

 obtained : 



8 Ann. de phys. et chem., 1905, 319. 



