1902.] Isomeric Change in Benzene Derivatives. 159 



cannot be got to crystallise from any solvent; they decompose at 85% 

 and in a moist atmosphere slowly give off nitrogen; this is very 

 noticeable when an attempt is made to estimate the nitrogen by 

 Dumas' method. They do not dissolve in aqueous alkalis or acids. 

 Cryoscopic determinations of the molecular weight in benzene show 

 that, whether prepared by means of sodium acetate or bicarbonate, the 

 molecule contains three benzeneazo-groups. There are three such com- 

 pounds possible, formed by condensation either of (1) 2 mois. of 

 •s-trichlorobenzenediazonium salt with 1 mol. of the diazophenol 



( (CoHsCls-N^.-CiClo^ 1 ' has a mol. weight = 601; chloride = 46'8 



per cent.) ; or (2) of 1 mol. of the diazonium salt with 2 mols. of the 

 diazophenol (mol weight = 585*5; chlorine = 42*2 per cent.); or 

 (3) of 3 mols. of the diazophenol (mol. weight = 567 ; chlorine 

 = 37 '5 per cent.). The material prepared by the use of sodium acetate 

 has the highest percentage of chlorine, 44*7 — 44-9 ; its molecular 

 weight was found to be 580 — 590; it is probably a mixture of (1) 

 and (2). When bicarbonate is used, the yellow substance contains 

 41 - 4— 43*7 per cent, of chlorine; its molecular weight was found to 

 be 560—580. 



The Transformation of s-Tribromobenzenediazonium Hydroxide. 



The transformation of s-tribromobenzenediazonium hydroxide re- 

 sembles in its main features that of the s-trichloro-compound. s-Tri- 

 bromobenzenediazonium nitrate, when first prepared, is pure white, 

 although Silberstein* has described it as yellow. On keeping, it 

 gradually acquires a yellow colour. Solutions both of the nitrate and 

 the hydrogen sulphate become yellow, and bromine can be detected in 

 the liquid. When either sodium hydrogen carbonate or sodium acetate 

 is added to solutions of the salts, the decomposition is far more rapid. 

 Thus in one experiment, in which sodium bicarbonate was added to a 

 solution containing 1 gramme of the diazonium hydrogen sulphate, 

 0*31 gramme of silver bromide was obtained, whereas 0*508 gramme 

 is required for the elimination of one atom of bromine. 



In the presence of sodium acetate, the solutions of the diazonium 

 salt rapidly become yellow and turbid. After 24 hours at 0°, a bulky 

 and, at first glance, homogeneous precipitate had separated. In one 

 experiment when 2 grammes of the sulphate were used, 0'58 gramme 

 of silver bromide was obtained from the filtrate from the yellow pre- 

 cipitate. Careful examination of the yellow precipitate showed that it 

 consisted of crystals completely covered and matted together by an amor- 

 phous yellow powder. These two substances could not be separated 

 by crystallisation from any solvent. It was, however, found possible 



* ' Journ. Pratt, Chem.,' 1883, [2], vol. 27, p. 113. 



