1032 JAPP AND ME1,DRUM: INTERACTION OF CI.--Inte?*action of Pilena~ztki.apuilzone, Acetophenone, and Ammonia. By FRANCIS It. JAPP, F.R.S., and ANDREW N. MELDRUM, B.Sc. JAPP AND STREATFEILD showed (Trans., 1888,41,270) that phenanthra- quinone, acetone, and ammonia interact according t o the equation CI4H,O, + C,H,O + NH, = CI7H,,NO, + H,O, and to the compound thus formed the constitution ?6H40 (?(oH>.cH~o co*cH, was assigned by C,H,* CNH Japp and Miller (Trans., 1885, 4'7, 18). The results of the present investigation, however, lead us to ascribe to it rather the constitution gous to that of diacetonnmine, (CH,),C(NH,)*CH,* CO*CH,. Thus me find that phenanthraquinone, acetophenone, and alcoholic ammonia interact according to the equation yielding di~?~nac?lldin772ino~~~?lc~Top?~~ant?~rene. It is evident that a constitution analogous to that ascribed by Japp and Miller to the acetone-ammonia derivative is inadmissible in the case of this com- pound.The acetophenone-ammonia derivative is hydrolysed by aqueous oxalic acid according to the equation yielding p?~enc~cyl?~?ldroz~~~henu?~t~~~~one (acetol,?&snone~?~ena?~t?~?~~uinone) which melts at 125-130'. By the interaction of phenanthraquinone, acetophenone, and aqueou4PHENANTHRAQUINONE, ACETOPHENONE, AND AMMONIA. 1033 C,H,* 7(NH2)* CH,* CO*C6H, b6H,* CO 9 ammonia, phenacyZan~inop~cnnt?&rone, analogous to the acetone-ammonia derivative, was obtained. The same substance is formed by the action OF ammonia on phenacyl- hydroxyphenanthrone. The ease with which these compounds break up into their generating substances rendered a detailed study of their reactions impossible.Other ketones, such as methyl ethyl ketone and diethyl ketone, ap- peared to act in a similar manner with phenanthraquinone and ammonia; but it was found impossible to obtain the resulting com- pounds in a condition suitable for analysis. EXPERIMENTAL. Action of AZcoholic Ammonia o?b a Mixture of Phenanthraquinone and Acetop?~enone.-20 grams of phenanthraquinone, 28 grams of acetophenone, and a large excess of alcoholic ammonia were heated in an open flask on the water-bath. I n proportion as the ammonia was expelled and the boiling point of the mixture rose, the greater part of the phenanthraquinone passed into solution. The filtered liquid deposited, on cooling, yellow, lustrous needles, which, by their appear- ance and melting point ( 158-159O), were recognised as phenanthra- quinonimide.Later on, a colourless substance began to separate. 'l'he liquid was therefore again filtered, and, on standing overnight, deposited a large quantity of this second substance in tufts of colour- less, silky needles, which, when heated in a capillary tube, decomposed about 150°, without showing a definite melting point. I n some pre- parations of this substance, the portions first deposited showed a pinkish tinge; but, by filtering at this stage, the colourless substance was obtained from the filtrate. As we found it impossible to recrystal- lise this substance without decomposing it, it was merely washed with cold alcohol, and dried for analysis i n a vacuum desiccator.It was perfectly homogeneous in appearance. Analysis gave figures agreeing with the formula of aminodilqd~ophenanthrene, (JNH,)*CH,* CO*C,H,. C,H,* C(NH,)*CH,* CO*C,H, 0.15'70 gave 04657 CO, and 0,0794 H,O. C = 80.90 ; 0.1454 ,; 0.4317 CO, ,, 0.0738 H20. C=80*98 ; H = 5.62. H = 5.64, 0.3049 0.2928 ,, 155 ,, ,, 13' ,, 754 mm. N= 6.20. C,,H,,N,02 requires C = 80.72 ; H = 5.83 ; N = 6.28 per cent. Hydrolycris of Dip?~enacyZdiamino~iAy~~~op~~na~~t?~rene.--In this ex- periment, the method employed by Japp and Miller (Zoc, cit., p. 17) in ,, 16.2 C.C. moist nitrogen at 11' and 740 mm. N=6*17. VOL. LXXV,. 3 21034 INTERACTION OF PHENANTHRAQUINONE AND AMMONIA. the hydrolysis of acetonylaminophenanthrone t o acetonephenanthra- quinone, was followed.The diphenacyldiaminodihydrophenanthrene was first moistened with alcohol-since otherwise it is not readily wetted by water-ground with water to a thin cream, and then poured into a large excess of a cold concentrated aqueous solution of oxalic acid, stirring well. Almost everything dissolved. The solution was quickly filtered through a large folded filter-an operation which must be rapidly performed, a6 the separation of the new compound begins almost immediately-and the filtrate, which speedily became yellow and turbid and smelt strongly of acetophenone, was allowed to stand over night. There was a large deposit of a pale yellow substance ; this was filtered off, thoroughly washed with water, dried in a vacuum desiccator, and purified by recrystallisation, first from ether, allowing the solvent to evaporate spontaneously, and afterwards from ethylic acetate, the latter giving the better result.It was thus obtained in fairly large thick prisms or six-sided plates which had a scarcely per- ceptible yellowish tinge, and are doubtless colourless when pure. It melted between 125O and 130°, turning dark yellow. Boiling the sub- stance with solvents must be avoided, as this treatment decomposes it, Analysis gave figures agreeing with the formula of phenacylhydroxy- phenccia t hone ( ccce t ophenonephenant hraquinone), ($H4*F(OH)*CH,* CO*C,H,, C,H,*CC) 0.1716 gave 0,5053 CO, and 0.0791 H,O. C=80.31 ; H=5.12. 0.1568 ,, 0,4622 CO, ,, 0.0717 H,O. C = 80.39 ; H=5*08. C2,H1,0, requires C = 80.49 ; H = 4-85 per cent.As acetonephenanthraquinone, when boiled with fuming hydriodic acid, yields diphenylenemethylfurfuran (Trans., 1890, 57, 663), we subjected the foregoing compound to the same treatment, in the hope of obtaining a diphenylenephenylfurfuran, but found that it broke up, before reduction, into acetophenone and phenanthraquinone, the latter then yielding phenanthrone and tetraphenylenefurfuran. By passing gaseous ammonia into an ethereal solution of phen- acylhydroxyphenanthrone, we obtained ~~~encccykal?ainop?~enant~~rone, 7GH4* ?(NH2)*CH,* C0*CGH5. This compound is,however, more readily C,H,* CO prepared by the method described in the next paragraph. Action of Aqueous Am.nzonicc on a Mixture of Phenanthraquinolee and Acetophenone.-20 grams of phenanthraquinone, 14 grams of acetophenone, and an excess of the strongest aqueous ammonia were introduced into a strong flask; this was tightly corked, and the mix- ture shaken vigorously, at intervals, during several hours.The organic substance solidified in the form of yellowish granules, whilst thePURFURAN DERIVATIVES FROM BENZOIN AND PHENOLS. 1035 aqueous portion was red. The solid substance was separated, ground with water, thoroughly washed, and dried over sulphuric acid. It was then digested with a small quantity of chloroform in the cold, to re- move unaltered phenanthraquinone, after which the residue was treated with sufficient cold chloroform to dissolve nearly the whole. On adding light petroleum to the filtered chloroform solution, a bulky, white precipitate was obtained ; this was filtered off and dried at the ordinary temperature, The substance, which was yellowish after drying, was dissolved in cold ether which had been previously satu- rated with ammonia (in order to check the tendency of the substance to decomposition), and the solution was allowed t o evaporate sponta- neously, In this way, the compound was obtained in small, colourless, six-sided plates, which melted with decomposition about 1 60°, turning green and evolving gas. The loss in purification by the above method is very great. Heating with solvents decomposes the substance completely. Analysis gave figures pointing to the formula of a phenacykumino- pl~nunthrme, cult to burn, and the value for carbon was lorn. *C(NH2)*cH2*Co*c6H5a The substance was dig. &r:* 60 0.1 175 gave 0.3449 C'O, and 0.0570 H20. 0.3090 ,, 11.7 C.C. moist nitrogen at 18" ancl 754 mm. N=4*33. C = SO*OS ; H = 5-39, C,,H17N0, requires C = 80.73 ; H = 5.20 ; N = 4.28 per cent. CHEMICAL DEPARTMENT, UNIVEESITY OF ABERDEEN.