FRANCIS 3 4-DI-P-NITROTETRAPHENYLFURAN. 1039 XCI .-3 4-Di-p-nitrotetrtxphen?/lfuran. By ARTHUR GORDON FRANCIS. THE formatioil of furan derivatives by the condensation of benzoins has been observed by several workers. Zinin ( Z e i t s c h . fiir Chemie, 1867 [ii] 3 313) by the action of concentrated hydrochloric acid a t 130° on benzoin obtained lepiden benzil and an oil. Limpricht and Schwanert (Ber. 1871 4 337) showed that dibenzoylstilbene (Zinin’s acicular oxylepiden) is obtained by digesting benzoin with dilute sulphuric acid and that when heated with benzoin i t yields lepiden aiid benzil. On reduction dibenzoylstilbene yields bidesyl 1040 FRANCIS 3 ~-DI-~~-NITROTETRAPHENYIAFURAN. Magnanini and Angeli (Ber. 1889 22 855) showed that when bidesyl is heated with concentrated Iiydrochloric acid a t 130-140° for two hours lepiden is formed.Dorn (Annalen, 1870 153 358) suggested and Japp and Klingemann (T. 1890 57 662) showed, that lepiden is tetraphenylfuran. Japp and Tingle (T. 1897 71, 11 38) proposed the formula : C Ph===CP h for dibenzoylstilbene . The relation of these compounds is shown in the following scheine : Benzoin. Bidesyl. Dibenzoylstibene. CPh===CPh I QHPh*OH YH Ph H Ph COPh COPh c0Ph m<O,>b€% \x lHC1 ../ 9 Yh-SPh CPh CPh \/ 0 Lepiden or tetraphenylfuran. More recently Irvine and McNicoll (T. 1908 93 950 1601) have shown that methoxyfuran derivatives can be obtained by the action of hydrochloric acid in methyl-alcoholic solution on benzoin, aiiisoin and furoin but that when ethyl alcohol is substituted for methyl alcohol no such condensation products are formed only the ethyl ester being obtlained as is usual in the Fischer esterification process.The author and Keane (T. 1911 99 344) described pnitro-acetylbenzoiii and attempted to hydrolyse this substance and also the nitrobenzoylbenzoin prepared by Zinin (Annalen 1857 104, 116). Zinin states that the position of the nitro-group in nitro-benzoylbenzoin was not determined. This compound is now shown t o be p-nitrobenzoylbenzoin NO,*C,H,*CHBz*OBz by the same methods that fixed a similar constitution for p-nitroacetylbenzoin Both p-nitroacetylbenzoin and pnitrobenzoylbenzoin yield on hydrolysis with hydrochloric acid under appropriate conditions the same condensation product together with p-nitrobenzil.The reac-tion is quantitative and f o r every three molecules of nitroacyl-(ZOC. cit.) FRANCIS 3 4-DI-p-XITROTETRAPHENYLFURAN. 1041 benzoin used one molecule of the condensation product and one molecule of p-nitrobenzil are formed. This condensation product is shown to be 3 4-di-p-nitrotetm~pheizyZf zcmn. The probable course of the reaction may be stated as follows: (1) The nitroacylbenzoin is first! hydrolysed : NO,*C,€I,*CHBz*OBz(Ac) -+ NOz*C,H,*CHBz*OH. (1.) (2) Two molecules of the nitrobenzoin thus formed are condensed and reduced to dinitrobidesyl by a third molecule of nitrobenzoin, which is itself oxidised t o p-nitrobenzil : NO2*C6H;CHBz*0H NO C,H C H Bz *OH HO*CH Bz.C,H;NO, N0,*C6H4*CH Bz*CHBz*C,H,*NO2 + N O,*C H4*CO13z (11.) (3) The dinit,robidesyl under the influence of hydrochloric acid loses water and is converted into dinitrotetraphenylfuran : N02*COH4* $!H-$!H *C,H; NO, N 0,- C,N,* E-- y*C,H **NO, + COP11 COPh CPh*OH CPh-OH -+ CPh CPh NO,* C H g-g* C,H N 0, \/ 0 From this view of the course of the reaction the constitution of the condensation product is regarded as 3 4-di-p-nitrotetraphenyl-furan.The intermediate compounds pnitrobenzoin (I) and dinitro-bidesyl (11) have not as yet been obtained although attempts were made t o isolate them by partial hydrolysis by means of hydrochloric acid and water. The investigation is accordingly incomplete in this respect but the results obtained so far are published as the author is likely t o be prevented by pressure of other work from continuiug the research for some time to come.E X P E R I M E N T A L . Hgdrolysis of p-Nit roace t yl b en aoi n . Forty grzms of p-nitroacetylbenzoin melting a t 1 2 5 O (uncorr. ) were dissolved in 500 C.C. of boiling 90 per cent. alcohol in a capa-cious flask fitted with a reflux condenser. When completely dis 1042 FRANCIS 3 4-DI-$l-NITROTETRAPHENYLFURAN. solved 30 C.C. of concentrated hydrochloric acid (D 1.16) were added slowly through the condenser so as not to stop the ebullition of the liquid. At the end of half an hour yellow needles began to be deposited. After six hours a voluminous mass of crystals had separated and the solution had a strong odour of ethyl acetate. It was then filtered while boiling and washed several times with boiling 90 per cent.alcohol to remove 2)-nitrobenzil. The crude product melted a t 207O (uncorr.) and after recrystallisation from glacial acetic acid the pure substance (21 grams) melted a t 2 1 1 O (uncorr.). From the alcoholic solution 10 grams of pnitrobenzil were reccwered. 3 4- D~-pnitrotetraphenyZfuran crystallises from glacial acetic acid in long pale yellow silky needles. It is very sparingly soluble in light petroleum ether cold or hot 90 per cent. alcohol or water, and moderately so in benzene toluene acetone or chloroform. It crystallises best from acetic anhydride or glacial acetic acid. One gram dissolves in 50 C.C. of boiling glacial acetic acid : 0.1002 gave 0.2672 CO and 0.0357 H,O. C= 72-72; H=3-95. 0.2744 , 15.0 C.C.N (moist) a t 1 9 O and 748.5 mm. N=6*28. 0.3384 in 17.677 benzene gave E =0*12O. C,,H,,0,N2 requires C = 72.73 ; H = 3.90 ; N = 6.06 per cent. M.W =462. M.W. =426. A c t i m of Alcoholic Hydrogen Chloride a t 42O on p-Nitroacetyl-benzoin. Two granis of pnitroacetylbenzoin were heated with 150 C.C. of absolute alcohol saturated with dry hydrogen chloride in a thermo-stat a t 42O for four hours. After some time a yellow solid sepa-rated which on crystallisation from glacial acetic acid melted a t 21 Oo and was dinitrotetraphenylfuran. From the portion soluble in alcohol only p-nitrobenzil melting at 140° could be isolated together with unchanged nitroacetyl-benzoin. Action of Wccter on p-Nitroncetylbenzoin. Five grams of nitroacetylbeiizoin were heated with 10 C.C.of water a t 180° for six hours. No pressure was observed on opening the tube. The aqueous layer was decanted through a filter and was found t o contain acetic acid. The solid product was a red resin, from which 23-nitrobenzil melting a t 140° and p-nitrobenzoic acid melting a t 238O were isolated. No dinitrotetraplienylfuran was iso-lated nor could any intermediate product be obtained PRANCIS 3 4-DI-~-NITROTETRAPHEXYLFURAN. 1043 Coiistit rhon of Nitroben zoylb eiazoin X’itrobenzoylbenzoin was prepared by Zinin’s method (loc. c i t . ) . The specimen melted a t 1 3 7 O (uncorr.). (Found N=4*3. Calc., S =3*9 per cent.) When hydrolysed and oxidised by nitric acid (D 1.4) i t yields benzoic acid and p-nitrobenzil. When oxidised with a mixture of suiphuric acid (20 C.C.of concentrated sulpliuric acid in 20 C.C. of water) and ar approximately normal solution of potassium dichromate as described for pnitroacetylbenzoin (Zoc. c i t . ) i t yields benzoic acid and p-nitrobenzoic acid. The constitution, (4)N02-C,H,( l)*CHBz*OBz, similar to that of pnitroacetylbenzoin is therefore deduced.. These reactions take place with more difficulty than in the case of pnitro-acetplbenzoin. Hydrolysis o f p-Nitro b enzoyl h e n zoin. (1) Five grams of p-nitrobenzoylbenzoin melting a t 1 3 7 O were heated under reflux with 150 C.C. of ethyl alcohol and 7 C.C. of aqueous hydrochloric acid (D 1.16). A t the end of two and a-half hours there was no separation of yellow needles no yellow colour in &he solution and no d o u r of ethyl benzoate. The p-nitrobenzoyl-bezoiii was recovered unchanged. (2) Four grams of pnitrobenzoylbenzoin were heated under reflux for twenty-four hours with a mixture of equal parts of aqueous hydrochloric acid (D 1.16) and ainyl alcohol more acid being added from time t o time. The insoluble product weighed 2.2 grams. After crystallisation once from toluene and twice from glacial acetic acid there remained 1.5 grams of dinitrotetraphenyl-f waii melting a t 2 1 0 O . When mixed with dinitrotetraphenylfui-an irielting a t 21 lo obtained by the hydrolysis of pnitroacetylbenzoin, the mixture melted a t 2 1 0 O . The substances were therefore iden-tical. CHEXICAL DEPARTMENT, SIR JOHN CASS TECHNICAL INSTITUTE, Loxno~ E.C. [Rereivecl September 7th 1917.