194 



SCIENCE 



[N. S. Vol. XXV. No. 631 



few inches away from the seat of disturbance, 

 are uninjured. 



The sensitiveness of the orthotoluene de- 

 rivative is even greater than that of the above 

 benzene compound, because in the moist state 

 it explodes if it be touched with a porcelain 

 stirring rod. 



Chemically, these compounds are relatively 

 stable and can be retained in a cool place for 

 a day without change. 



A considerable number of other diazonium 

 perchlorates are described in the two papers 

 quoted; in general, they exhibit the interest- 

 ing contrast of being slightly less explosive 

 than the two salts described above, but con- 

 siderably more unstable chemically. 



PREPARATION AND PROPERTIES OF BENZOYL 

 NITRATE 



The formation of an acid chloride, such as 

 acetyl chloride, CHjCOCl, from an acid and 

 phosphorus pentachloride, is one of the most 

 generally applicable reactions known. The 

 corresponding bromides are available in mod- 

 erate numbers, but only a very few of the 

 iodides or fluorides have been prepared and, 

 with these exceptions, no compounds are 

 known containing an inorganic acid radicle 

 in place of chlorine. For this reason, as well 

 as on account of its properties which are de- 

 scribed below, the isolation of benzoyl nitrate, 

 OjHjCOONO., is a matter of considerable in- 

 terest. A preliminary announcement on the 

 subject some months ago by F. Francis* has 

 been followed up with a fuller account by him 

 self and T. H. Butler." The compound is 

 prepared by mixing together, at about — 15°, 

 benzoyl chloride and well-dried, finely divided 

 silver nitrate. The product is an oil which is 

 extremely sensitive to moisture. It may be 

 filtered through dry filter paper, but if the 

 latter contains the ordinary amount of moist- 

 ure the nitrate reacts with explosive violence. 

 In sealed tubes, at the ordinary temperature, 

 the nitrate changes gradually into benzoic 

 anhydride and, apparently, nitrogen pentoxide. 



^Jour. Ghem. Soc, 89, 1 (1906). 

 'Ber., 39, 3795 (1906). 

 'IMd., 39, 3804 (1906). 



At 100° the change is similar, except that 

 nitrogen peroxide and oxygen are formed, and 

 at higher temperatures the reaction is so rapid 

 as to produce an explosion. In nitrobenzene 

 solution the nitrate is transformed slowly, to 

 the extent of 60 per cent, in six months, into 

 metanitrobenzoic acid. Other non-ionizing 

 solvents bring about the same change, but less 

 rapidly. 



In its reactions benzoyl nitrate behaves as a 

 nitrating agent, as an oxidizing agent, or as a 

 means of introducing a benzoyl radicle. Al- 

 cohol yields ethyl nitrate; aromatic hydro- 

 carbons, phenols, and their ethers give, with 

 ease, chiefly mono-nitroderivatives. As ex- 

 amples of oxidation may be mentioned the 

 transformation of thiophenol and hydrazo- 

 benzene into diphenyldisulphide and azoben- 

 zene, respectively. Primary aromatic and 

 secondary aliphatic amines yield benzoyl 

 derivatives, OuHjCONHE, but secondary 

 aromatic amines form nitrosamines. With 

 the exception of paratolylmethylnitramine, 

 CH3C,H,N(N0,)CH„ these are unstable and 

 pass into nitro secondary amines, such as 

 CH AH3 (NO,) NHCH,. 



One of the chief points of interest in con- 

 nection with the nitrating action of benzoyl 

 nitrate is that it permits of this reaction 

 being realized in the complete absence of 

 water, a condition which, hitherto, has been 

 unattainable. The nearest previous approxi- 

 mation to it was by the use of diacetyl ortho- 

 nitric acid, (CH3COO),N(OH)3. One result 

 of such change in the conditions of reaction is 

 that orthonitroaniline, which can only be ob- 

 tained to the extent of about 10 per cent, in 

 admixture with its isomers by the ordinary 

 methods of nitration, is formed quantitatively 

 by the action of benzoyl nitrate on acetanilide. 



Metanitrohenzoyl nitrate, O^NOsHjCOONOj, 

 and hutyryl nitrate, CH3CH,CH,C00N0„ are 

 also formed by the action of the respective 

 chlorides on silver nitrate, hence it would 

 appear that such compounds are always pro- 

 duced, as intermediate steps, in the reaction 

 of an acid chloride on a metallic nitrate. 



J. Bishop Tingle 

 Johns Hopkins Universitt 



