STUDIES IN THE CAMYRANE SERIES. PART V. 1003 CVL-Studies in the Camphane Series, Paint I? Halogen Devivatives of p- Cymene fyom substituted Nit rocamphanes. By MARTIN ONSLOW FORSTER and WILLIAN ROBERTSON, A.R.C.S. IT has been'mentioned that when 1 : l-bromonitrocamphane is dissolved in concentrated sulpburic acid, the anhydride, C,,H,,ONBr, is always accompanied by a viscous oil having a fragrant odour (Forster, Trans., 1899, 75,1145). Although the yield of this material does not exceed 10 per cent. of the halogen compound employed, an investigation of the anhydride and its derivatives has placed at our disposal a quantity sufficient for systematic examination ; this has been undertaken, therefore, in the hope that ideatification might afford some explana- tion of the obscure change taking place when 1 : l-bromonitrocam- phane is converted into the anhydride.Suspecting that the oil, which contains bromine, might arise from some impurity in the crude bromonitrocamphane, we ascertained that it is also produced when the material is recrystallised from acetic acid previous to dissolution in sulphuric acid, Moreover, a similar com- pound, containing chlorine in place of bromine, attends the conversion of 1 : l-chloronitrocamphnne into the anhydride, and consequently that substance has been included in the investigation. From this, we have ascertained that the oils in question are halogen derivatives of p-cymene, in which substitution has occurred in the ortho-position relatively to the methyl group. This result is noteworthy in being the second established instance of the production of substituted cymenes from camphor derivatives of known constitution.The transformation of camphor itself into cymene has been the subject of numerous investigations, but the first record of the observation of this change in derivatives of camphor occurs in a paper by Marsh and Hartridge (Trans., 1898, 73, 852), who found that on preparing carvenone by the action of concentrated sulphuric acid on 1 : l-dichlorocamphane, the product is contaminated with a small proportion of a chloro-derivative, which they believed to be chlorocymene [Me : C1: PrS = 1 : 2 : 41 ; the formation of this compound is not recorded by Bredt (Anncden, 1901, 314, 369), who repeated the experiments of Marsh and Hartridge. It will be noticed that the position of the halogen atom as regards the methyl and isopropyl groups is in complete agreement with the requii*ements of Bredt's formula for camphor, assuming no alteration to have taken place in the attachment of the bromine or chlorine.1004 FORSTER AND ROBERTSON : Klr CH,-YH-CH, CH,-CMe-CBr *NO, 1 : 1-Rromonitrocamphane.Bromo-p-cymene. \/ I YMe2 I -+ CH3 It is scarcely possible, however, to furnish an explanation of the change so simple as that given by Bredt (Zoc. cit.) for the conversion of cam- phor into cymene. Most likely the nitro-group plays an important part in the reaction, as the liberation of nitrous fumes has been observed during the production of the substance. In this connection, it is noteworthy t h a t 1-nitrocamphane does not yield cymene when dissolved in cold concentrated sulphuric acid, and appears to be quite indifferent towards that agent.Bromocymene [OH, : Br : G,H,P = 1 : 2 : 4 J from 1 : l-Brornonitrocm&zne. The accumulated bye-product from 500 grams of bromonitrocam- phane was washed several times with concentrated sulphuric acid, treated with water and dilute sodium carbonate, extracted with ether, dried with calcium chloride, and freed from ether, which deposited 50 grams of a pale brown, somewhat viscous oil, having a powerful, fragrant odour. We have not succeeded in isolating from this pro- duct a pure specimen of bromocymene, but there is every probability that the latter forms the chief constituent of the mixture. The fol- lowing evidence, coupled with the fact that 1 : 1-chloronitrocamphane yields chlorocymene under similar conditions, led to this conclusion.After the separation described above, the oil was distilled in steam, which separated i t from a less volatile, resinous material. A t this stage, it was optically active, giving aD - 13'31' in a 2-dcm. tube, and contained 33 *3 per cent. of bromine ; the substance was saturated, and did not contain nitrogen. It was then agitated with concentrated sulphuric acid, and after remaining in contact with that agent during some days was washed with sodium carbonate and distilled in an atmosphere of steam. The pale yellow oil decomposed when distilled under atmospheric presmre, evolving hydrogen bromide and becoming charred, It was, therefore, distilled under 26 mm. pressure, the major portion boiling at T28-130°, and then redistilled under the same reduced pressure, the final product boiling a t 129-130° ; although a mobile, colourless oil, the substance became pale yellow on exposure to sunlight.On analysis : 0.1614 gave 0.3230 CO, and 0.0888 H20. C = 54.58 ; H = 6.12. 0.2188 ,, 0.4361 CO, ,, 0.1216 H,O. C=54*35; Ht6917. 0.2509 ,, 0.2120 AgBr. Br = 35-95. CIoH,,Br requires C = 56.34 ; H = 6.10; BP = 37.56 per cent.STUDIES IN THE CAMPHANE SERIES. PART v. 1005 From these results, it appears that the bromocymene was still con- taminated with some oxygenated impurity, which is optically active, because the material analysed gave aD - 4'48' in a 2-dcm. tube. Such a substance could only arise by condensation of 2 mols. of bromonitro- camphane taking place according to the equation 2ClOHl,O2NBr - 2H20 - N20 = C2,H2,0Br,; a compound having this empirical formula would contain 54.0, 6.3, and 36.0 per cent.of carbon, hydrogen, and bromine respectively. These re- quirements mere so nearly met by our analyses, that a determination of the molecular weight of the substance in benzene was made. Molecular weight of ClOHI3Br, 213 ; C2,H,,0Er2, 444. Grams of benzene. 37-70 19.59 Grams of substance. 0.7082 0.1467 Grams of substance in 100 grams of benzene. 4.0011 0-7488 Depression of the freezing point. 0'938" 0.182 Molecular weight deduced. 213.3 215.4 This result pointed unquestionably to the substance being bromo- cymene, and was borne out by a determination of the density, which gave 1.257 at 16O, the value recorded by Landolph being 1.269 at 17-69 As it seemed likely that the oxygenated compound of the type indi- cated would be decomposed by alcoholic soda, the substance was sub- mitted to this treatment, because it is known that bromocymene is indifferent towards the agent in question.The resulting substance was almost colourless, and boiled at 2299 the temperature at which bromo- cgmene distils. On analysis : 0.1 836 gave 0.1588 AgBr. Rr = 36-82, CloHl,Br requires Br = 37.56 per cent. With the object of proving that this compound is derived from p-cymene, an attempt was made to reduce it in alcohol with sodium. Forty grams were dissolved in 300 C.C. of absolute alcohol, and heated in a reflux apparatus with 40 grams of sodium, which mas added in small quantities at a time, and soon gave rise to the separation of sodium bromide.When the metal had dissolved, water was added, and a current of steam was passed through the liquid until the alcohol had completely separated. On diluting the distillate with water, a colourless oil was precipitated, and this was collected, dried with1006 FORSTER AND ROBERTSON : calcium chloride, and distilled; i t boiled at 172' (uncorr.) under 770 mm. pressure. On analysis : 0.1226 gave 0.3990 00, and 0.1147 H,O. C,,H,, requires C = 89.54 ; H = 10.45 per cent, The identity of the hydrocarbon was placed beyond doubt by the operation recommended for this purpose by Widman ; oxidation with potassium permanganate in presence of caustic soda gave rise to a specimen of hydroxyisopropyl benzoic acid melting at 1579 C=88*76; H= 10.40.Anhydride of Cl~loronitrocu~phun~, CloH1,ONCI. In preparing the anhydride o€ chloronitrocamphane, it is of greater importance to keep the rise of temperature under control than even in the case of the corresponding bromo-derivative. One hundred grams of chloronitrocamphane were added in small quantities to 800 C.C. of concentrated sulphuric acid, cooled below 0' by a freezing mixture, and at no time was the temperature of the liquid allowed to rise above 5". The acid soon became dark brown, and a colourless, fragrant oil was produced, whilst towards the end of the operation hydrogen chloride was liberated. When all the chloronitrocamphane had been added, the liquid was allowed to remain in a separating funnel until the oil had risen to the surface, the acid being then run slowly on t o crushed ice; this precipitated a yellow solid, which was washed, drained on earthenware, and recrystallised twice from boiling alcohol.The yield under favourable circumstances amounts to 25 per cent. On analysis : 0.2345 gave 14.8 C.C. of nitrogen at 20° and 773 mm. N= 7.35. 0.2094 ,, 0.1521 AgCI. C1= 17.97. C,,H,,ONCl requires N = 7-02 ; C1= 17.79 per cent. The anhydride dissolves readily in hot alcohol, from which it crys- tallises in aggregates of opaque prisms; the melting point is not very definite, but the substance darkens above 200°, and melts at about 230' to a deep brown liquid which evolves gas. It is readily soluble in cold benzene and in ethyl acetate, crystallising from the latter in minute octahedra ; boiling water dissolves the compound very spar- ingly, but it is moderately soluble in boiling petroleum.Chlovocymerte [CH, : C1 : C,H,@ = 1 : 2 : 41 from 1 : 1-Chloronitrocclmphune. The oil referred to in describing the preparation of the anhydride of chloronitrocamphane waB decolorised by treatment with concentrated sulphuric acid, and washed with water, followed by sodium carbonate ; after being distilled in a current of steam, the product was collectedSTUDIES IN THE CAMPIIANE SERIES. PART V. 1007 Grams of benzene. with ether, dried with calcium chloride, and distilled, 26 grams being obtained from 300 grams of chloronitrocamphane. The substance obtained in this way was a colourless, mobile liquid boiling at 211-212O (uncorr.) under 768 mm.pressure; i t had a sp. gr. 1*0122 at 16', and although in these respects it agreed closely with pchlorocymene, the fact that it gave - 0'35' in a 2dcm. tube in- dicates that the substance was not quite pure. 0.1651 gave 0.4250 CO, and 0.1136 H,O. On analysis : C = 70-20; H=7.64. 0.2198 ,, 0,1863 AgC1. C1= 20.92. C,,H,,Cl requires C = 71.26 ; H = 7.72 ; C1= 21.02 per cent. A determination of the molecular weight in benzene gave the fol- lowing result : Molecular weight of CloH,,Cl, 168.5. Grams of substance Depression MoIecul ar of the of benzene. Grams cjf substance. in 100 grams freezing point. weight deduced. 25'03 $ 9 0'1358 0.5425 0.178" 152.4 0'2988 1 '1 938 0'380 157.1 Conversion of Chlovonitrocumphane Anlhydride into the Isomeride.Chloronitrocam phane anhydride was dissolved in 5 parts of alcohol and heated in a reflux apparatus with 2 parts of concentrated hydro- chloric acid during half an hour. The liquid was poured into cold water, and the precipitate crystallised from dilute alcohol, which deposits i t in needles. On analysis : 0.2038 gave 13.0 C.C. of nitrogen at 16O and 754 mm. 0.1500 ,, 0.1065 AgCl. C1= 17.56. C,,H,,ONCl requires N = 7.02 ; C1= 17-79 per cent. The substance is readily soluble in alcohol and is best recrystallised from boiling petroleum, in which it is only sparingly soluble ; it crys- tallises from the latter in transparent, six-sided plates, and melts a t 24SO. The benxoyl derivative, prepared by the Schotten-Baumann method, crystallises from alcohol in lustrous white leaflets, and melts at 166'.0.1546 gave 0.0712 AgCI. C1= 11.39. C17H,,0,NC1 requires Cl = 11 *69 per cent. N = 7.38.1008 FORSTEK AND ROBERTSON : Actim OJ HydroxyZamiw on th Anhydride of Chbrtitrommphane. Ten grams of chloronitrocamphane anhydride were dissolved in 50 C.C. of absolute alcohol, to which was added 8 grams of dry sodium carbonate. The mixture was then heated during 5 hours in a refiux apparatus with 10 grams of hydroxylamine hydrochloride dissolved in the minimum quantity of water. Alcohol was then distilled off on the water-bath, and the residue treated with a considerable quantity of water, filtered, and washed. The compound was recrystallised from alcohol, which deposited transparent, rectangular plates. 022951 gave 0.1805 AgC1. C1= 15.13.Cl,H170,N,CI requires C1= 15.27 per cent. melts a t 1 8 7 O , when it turns brown and evolves gas. ammoniacal silver nitrate readily on warming. which it crpstallises in lustrous, white needles, and melts at 1649 alcoholic solution is indifferent towards ammonincal silver nitrate. 0.2634 ,, 0.1625 AgCI. C1= 15.25. The hydroxylamino-derivative of chloronitrocamphane anhydride It reduces The benxoyl derivative dissolves somewhat sparingly in alcohol, from The Act$on of Nitric Acid on the Anhydrides of Bromonitrocamphane and Ch Zoronitrocarnphane. When the anhydride of bromonitrocamphane, obtained by dissolv- ing that substance in concentrated sulphuric acid, is treated with cold fuming nitric acid, it is dissolved immediately, and gas is liberated, but the diluted liquid does not yield a definite product.If, however, the isomeric anhydride is trwted in the same way, no change takes place, and the substance merely dissolves, but on heating the liquid a nitro-derivative is obtained. Ten grams of the isomeric anhydride were consequently heated with 40 C.C. of fuming nitric acid of sp. gr. 1.52 during several minutes, the liquid, when cold, being poured into a considerable volume of water, which precipitated a colourless oil. This solidified rapidly, and was then collected, washed, and crystallised from hot alcohol. On analysis : 0.2911 gave 25.1 C.C. of nitrogen at 17" and 763 mm. 0.3084 ,, 0.1997 AgBr. Brz27.56. The substance therefore appears to be a nitro-derivative. N= 10.23. C,,0,30,N2Br requires N = 9-69 ; Br = 27*68 per cent.It crys- tallises from alcohol in rosettes of transparent prisms melting at 103O, and is insoluble in sodium carbonate or caustic alkali. Unlike the substance from which it is derived, it gives Liebermann's reaction forSTUDIES IN THE CAMPHANE SERIES. PART V. 1009 nitroso-derivatives. Hot caustic soda decomposes it slowly, giving rise to infracampholenonitrile. The same difference in behaviour towards nitric acid is exhibited by the anhydrides of I : 1-chloronitrocamphane. Twenty-five grams of the isomeride melting at 248' were heated with 100 C.C. of fuming nitric acid during one minute only, the solu- tion being poured on crushed ice; the oil which separated soon solidi- fied, and was crystallised from alcohol, which deposited flat prisms resembling the corresponding bromo-derivative, and me1 ting at 7 1-72', 0.2204 gave 0.1309 AgCI.CI,H1,O,N,C1 requires C1= 14.53 per cent. The substance dissolves very readily in hot alcohol, and in cold benzene, ethyl acetate, or glacial acetic acid ; it is moderately soluble in boiling petroleum, from which it crystallises on cooling, and is also slightly soluble in boiling water. Decomposition with hot caustic soda is effected more readily than in the case of the corresponding bromo-derivative, infracampholenonitrile being produced. It gives Liebermann's reaction for nitroso-derivatives and does not reduce an ammoniacal solution of silver nitrate; it is distinguished from the original anhydride by indifference towards benzoyl chloride, which does not yield a derivative. Behaviour of the Nitro-compound on Reduction.-Ten grams of the nitro-derivative obtained from the isomeric anhydride of 1 : 1-chloro- nitrocarnphane were dissolved in 60 C.C. of glacial acetic acid, and treated with 20 grams of zinc dust. The metal was added in small quantities at a time, and the liquid was cooled after each treatment, but in spite of these precautions nitrous gases were liberated on each occasion. When all the zinc had been added, the product was allowed to remain at the ordinary temperature during 12 hours, then heated on the water-bath during 2 hours, and finally filtered. On diluting the acetic acid with water, a white, crystalline precipitate was obtained ; this weighed nearly 8 grams, and was readily identified as the anhydr- ide melting at 248'. The filtrate from this compound was rendered alkaline with caustic soda, and distilled in a current of steam, which carried over a small quantity of infracampholenonitrile ; the distillate was tested for hydrazine, but no reduction of ammoniacd silver nitrate occurred. From this experiment it appears probable that the nitro-groups in the derivatives of the anhydrides are attached to nitrogen. C1= 14-68. ~ Y A L COLLEGE OF SCIENCE, LONDON. SOUTH KLNSINGTON, S . W.