1284 LAPWORTH AND LENTON: THE CONSTITUTION OF CXXXV.-The Constitution of Camphanic Acid and of Bromocccmphoric Anhydride. By ARTHUR LAPWORTH and WALTER HENRY LENTON. ONE of the most interesting points in connection with the problem of the constitution of camphor and its derivatives is to be found in the behaviour of ordinary bromocamphoric anhydride on hydrolysis. When that substance is boiled with water, especially in presence of sodium carbonate, the anhydride ring is broken, and the bromine atom is simultaneously removed. The product consists for the most part of the lactonic product, camphanic: acid, but a certain amount of an un- saturated monobasic acid, lauronolic acid, is also produced, its mode of formation being represonted empirically by the equation, co C8H1,Br<CO>0 + H20 = G,H,,*CO,H + HBr + GO,.The researches of Fittig and his pupils on brominated acids have brought to light the fact that the removal of carbon dioxide and hydrogen bromide from such compounds by the aid of dilute aqueous alkalis is associated exclusively, or nearly so, in open-chain com- pounds, with P-bromo-acids, and, generalising from this fact, Aschan (Ber., 1894, 27, 2114) and others have concluded that in bromo- camphoric anhydride the bromine atom must occupy the @position with regard to one of .the two carboxyl groups. Acceptance of this view, however, is rendered difficult, for the fol- lowing reasons. Saturated carboxylic acids, on bromination, invariably yield a-bromo-acids. Fittig’s observations have reference to the behaviour of /3-bromo-acids. Combining these conclusions, we arrive at the view that bromocamphoric acid must be both an a- and a /3-bromo-acid, that is to say, it must contain the grouping CO,H 6Br*c*CO,H, and be a bromosuccinic acid.Hence it mustCAMPHANIC ACID AND OF BROMOCAMPHORIC ANHYDRIDE. 1285 follow that camphanic acid belongs to that unstable group of com- pounds, the p-lactones, of which but few are known, a view which it is almost impossible to reconcile with the great stability of the lactone ring, for hydroxycamphoric acid is unable t o exist in the free state, being at once converted into camphanic acid. Another fact which militates against the probability of the existence of the group CO2H*kBr*o*C0,H in bromocamphoric acid is the fact that lauronolic acid, which, according t o this view should contain the grouping CO,H*&b, is not an ap- but a &unsaturated acid, as is shown by its ready conversion into and formation from campholactone and the corresponding hydroxy-acid.Other chemists, desiring t o maintain that camphoric acid was a derivative of succinic acid, and either ignoring the work of Volhard and others, or attaching less significance t o it than to Fittig's observa- tions, have assumed that camphanic acid is a y-lactonic acid. This assumption necessitates the additional one that camphoric acid either becomes brominated at a point which is not in the a-position with regard to a carboxyl group, or that the brominat'ion follows the usual course, and that the hydrolysis of the bromo-anhydride has involved a change of a type otherwise quite unknown, and representable as follows : *?H *q yBr*CO,H + YH*CO,H 1 + HBr.I n the hope of being able to prove that the mistake had arisen in generalising too hastily from the behaviour of brominated open chain acids, we have made a long series of experiments with bromocamphoric anhydride and camphanic acid. It was clear from the first that the investigation would be more than usually diacult, for had any straightforward method been available, so important a problem would scarcely have remained for so long a time unsolved. It was soon found that any views based on the assumption that the change of bromocamphoric anhydride to camphanic acid involved an unusual change of configuration must be set aside, as the reverse change may be effected with great ease by merely warming camphanic acid on the water-bath with phosphorus pentnbromide. The next point t o be proved was that camphanic acid was derived from an a-hydroxy-acid.For this purpose, and for a, long time without success, we sought a method of removing the carboxyl group which was presumed to be attached to the same carbon atom as the hydroxyl group. The removal of a carboxyl group from an a-hydroxy-acid is, as a rule, a fairly easy process, and in a large number of cases may be achieved by such simple means as heating with strong acids, or with *9*C02H *q--co-o VOL. LXXIX. 4 T1286 LAPWORTH AND LENTON: THE CONSTITUTION OF I CH,*Q*CO*NH, I ?Me2 chromic acid or lead peroxide in presence of diliite acetic acid. Tn the present case, however, it would appear that the absence of the actual hydroxyl group itself, which is altered by its participating in the formation of the stable lactone ring, renders the usual oxidation processes entirely ineffectual.Attempts were also made to oxidise the product in which the lactone ring is opened (obtained by boiling the acid with excess of alkali); these were more effectual, but as the oxidation went too far, the experiments gave no useful result. Eventually, however, the nitrile of camphanic acid was prepared in the hope that, like the a-hydroxynitriles, it would prove to be unstable towards alkalis. For this purpose camphanamide was required as an intermediate product and it was obtained by the following methods. Camphanic acid was heated with phosphorus trichloride and the product poured into ammonia; a small quantity of the amide mas thus obtained, but the process was tedious and the yield unsatisfactory.The amide was obtained much more easily, however, by a modification of Wreden’s method. Wreden heated bromocamphoric anhydride with ammonia in closed tubes, and obtained a substance which he took to be the imide of hydroxycamphoric acid (Amalen, 1872, 163, 339), but which is identical with the amide made by the above process, behaves neither as a hydroxy-compound nor as an imide, and, as our subsequent work showed, must be regarded as camphanamide. The action is represented by the equation CH,+CN CH,*F:O + H 2 0 + I ?Me2 $?Me, co 0 C,H,,Br<CO>O + ZNH, = NH,Br + NH2*CO*C,H,,<b0. The heating in sealed tubes was found to be unnecessary, as it was sufficient to leave the finely-powdered bromoanhydride with strong ammonia for several hour6 at the ordinary temperature in order to effect its almost complete conversion into the amide.The amide was then converted into the nitrile by heating it with a mixture of phosphorus trichloride and pentachloride. A waxy nitrilolactone was thus produced, which, on treatment with strong alkalis, .was resolved into hydrogen cyanide and camphononic acid (Lapworth and Chapman, Trans., 1899, 75, 1000). As this is a y-ketonic acid, the investigation has afforded an apparently incontro- vertible proof of the views which we hold of the relationship of bromo- camphoric acid and camphanic acid to camphoric acid itself. The course -of the changes involved may be represented by the scheme,CAMPHANIC ACID AND OF BHOMOCAMPHORIC ANHYDRIDE.1287 and i t may be worth while to point out that a combination of the above facts with those already known about camphononic acid (Zoc. c k ) is in itself sufficient to decide almost conclusively the true constitution of camphoric acid. Thus, it has been shown that camphononic acid has one or other of the two formulae, CH,*$!O CO--C)H, I p e 2 and I $!Me2 UH,*CMe*CO,H CH,* CMe*CO,H and of these, in reality, only the former can be deemed worthy of serious consideration, for the acid yields hydrazones, a semicarbazone, and an oxime only with difficulty, does not unite with hydrogen cyanide, and cannot be reduced by such energetic treatment as with sodium and boiling amyl alcohol.Such a behaviour is altogether unknown in simple ketones and ketonic acids which contain the group- ing *CH,*CO*CH,*, present in the second of the above formule; the properties of such an acid would approach more nearly those of cam- phonic acid (Trans., 1900, 77, 454), which has a constitution of that kind. The first of the above formulae, therefore, apart from any views as to the constitution of camphor or camphoric acid, is the only possible which is character- ised by an exceptionally feeble tendency to form additive complexes (compare Trans., 1901, 79, 379). I n extending the foregoing observations, we have succeeded in devising a second method of converting camphoric acid into cam- phononic acid, through camphanamide as the intermediate stage.In this method, advantage was taken of the fact that the additive com- pounds of ammonia with aldehydes or ketones are easily broken down under the influence of alkalis or acids. In order to obtain the analogous compound of cnmphononic acid, the amide of camphanic acid was subjected to the action of sodium hypobromite and excess of sodium hydroxide. It was expected that the CO-NH, group would, as usual, be converted into the NH, group, and that an unstable aminohy droxy-acid of the for mula one, containing as it does the grouping "\ C-C*C:O, c/ I $Me2 CH,*CMe*CO,H would be produced which would at once break down into ammonia and camphononic acid. As a matter of fact, by this process, large quan-1288 LAPWORTH AND LENTON: THE CONSTITUTION OF tities of the ketonic acid may be prepared, and in a shorter time than by any of the methods hitherto discovered.EX PE R IJI EN TAL, The camphanic acid used in these experiments was prepared from bromocamphoric anhydride by boiling it with potassium acetate dis- solved in glacial acetic acid, and removing the potassium bromide from time to time by filtration, in accordance with Aschan's directions (Actu SOC. scient. fenn., 21, No. 5, 1). Dry camphanic acid is best made by heating the hydrated crystals in a flask a t ZOO0, a stream of air being occasionally drawn through the flask to remove the steam. On cooling, i t is noticed that the whole sets to a semi-transparent mass of crystals, and that this after- wards becomes opaque; this change is due t o dimorphism, and is most easily observed in the following way.The camphanic acid is melted on a microscope slide beneath a cover glass; on cooling, the substance sets rapidly to R transparent isotropic mass the structure of which is difficult to distinguish, but consists of n conglomeration of crystals resembling ammonium chloride ; almost immediately afterwards, the appearance of the second, doubly refract- ing, ordinary modification is noticed, the field now becoming distinctly crystalline, Camphanic acid is very stable towards most, and especially towards acid, oxidising agents. Thus, when heated with lead peroxide and acetic acid, it does not afford any appreciable quantity of carbon dioxide. It is very slowly attacked by nitric acid, however, and is practically converted into camphoronic acid, the intermediate com- pounds, if any, being less stable towards the agent than camphsnic acid itself.Attempts were made to obtain a ketonic acid. by boiling the cam- phanic acid with excess of potash to hydrolyse the Iactone ring, and the resulting solution of the potassium hydroxycamphorate was diluted and partially neutralised by means of a stream of carbon dioxide. A 2 per cent. solution of potassium permanganate was then run in, and :M it was found t h a t decolorisation occurred only very slowly, excess of the oxidising solution was added, and the whole allowed to remain for several days at O o ; the solution was then decolorised by means of sulphurous acid, heated, filtered, and evaporated. On acidifying and extracting the residue with ether, a n oily mixture was obtained, which soon deposited a considerable quantity of crystals of uualtered cam- phanic acid ; the remaining portion gave a very small deposit when warmed with a solution of phenylhydrazine acetate, but the amount of t h i s was iubufficient to be of any value for further investigation.CAMPHANIC ACID AND OF BROMOCAMPHORIC ANHYDRIDE.1288 Experiments were then tried with campholactoniz acid, which is usually supposed to be closely related t o camphanic acid, as it was possible that it might be the hydroxy-acid corresponding with cam- phononic acid. It was made by hydrolysing purified campholactone, and formed beautiful, white crystals melting a t 146'. The purified acid was dissolved in sodium carbonate solution, cooled to 0' and treated in the usual manner with dilute potassium permanganate, which a t first was instantly decolorised, but when n comparatively small quantity had been added, retained its colour for a long time.The products, isolated in the usual way, contained substances of a ketonic character, but the insoluble phenylhydrazones obtained from it, curiously enough, mere insoluble in dilute sodium carbonate, and were, therefore, not simple hydrazones of ketonic acids. From this it would appear that campholactonic acid is not genetically related to cam- phononic acid in the simple manner commonly supposed.* I n attempting to obtain camphanamide, the first experiments were made with the object of preparing it from the chloride of camphanic acid. For this purpose, the dry, finely powdered acid was heated in a flask with a large excess of pbosphorus trichloride for several hours, the excess of the latter boiled off, the residue dissolved in ether to render it more easy to manipulate, and the ethereal solution gradually poured into strong aqueous ammonia, which was kept a t Oo and shaken con- tinuously.A small quantity of a white, pulverulent substance separated, which was collected by filtration, washed, and crystallised from methyl alcohol, when i t separated in brilliant prisms melting at 160°. When warmed with dilute soda, the substance slowly dissolved, and on acidifying the resulting solution, a sparingly soluble acid separated. This melted a t 160°, with evolution of water vaponr, solidified im- mediately, and afterwards melted a t the same temperature as the original amide.The properties of the two foregoing substances resembled very * The author has suggested the formula 1 as a probable one for lauronolic acid (Brit. Assoc. Report, 1900, 327). Jf this is correct, then the formula for the corresponding y-hydroxy-acid, namely, campholactonic acid, would be CH,'CMe ZMe C H , ~ M ~ *CO,H """~,"G'j 1 1 CH,*UMe--CO1290 LAPWORTEI AND LENTON: THE CONSTITUTION OF closely those of the compound obtained by Wreden from bromocam- phoric anhydride by heating i t with ammonia in closed tubes (Zoc. cit.), and it was thought, therefore, that a better yield might be obtained by his process. On repeating his experiments, it was found that the componnds were, in fact, identical, and, moreover, that it was un- necessary to employ heat during the process, as, when the hromo- anhydride was sufficiently finely powdered, a marked development of heat occurred spontaneously on adding it t o ammonia. We have, therefore, used the following process in preparing the amide.The anhydride is powdered as finely as possible, passed through a fine sieve, and projected, in small quantities a t a time, into a flask containing strong aqueous ammonia (sp. gr. O*SSO), which is agitated during the whole time to prevent the formation of lumps. I n the course of 24 hours, the powder is converted into a mass of minute needles, which may be collected at the pump, washed, and crystallised from methyl alcohol or acetone. On analysis : 0.1947 gave 0.4377 CO, and 001371 H,O. C=60*8 ; H=7.8.0.2390 ,, 13.9 C.C. moist nitrogen a t 13O and 750 mm. N = 6.8. C,,H,,O,N requires C = 60.9 ; H = 7.6 ; N = 7.1 per cent. Camphanamide has all the properties ascribed to it by Wreden (Zoc. cit.). It dissolves sparingly in water, cold alcohol, or ethyl acetate, more readily in hot methyl or ethyl alcohol or acetone, and separates from the last-named solvent in beautiful, transparent prisms of calcite-like lustre. The crystals are apparently of rhombic symmetry ; in crushed frag- ments, in convergent polarised light, a wide biaxial figure is occasion- ally seen. The melted substance solidifies readily in long needles, flattened in a direction perpendicular to the optic axial plane. The directions of greatest length and elasticity are at right angles to one another, That the compound is in reality an amide, and not, as Wreden sup- posed, a hydroxyimide, is shown by the fact that it does not behave like an imide in any way, and gives no silver compound on treatment with ammonia and silver nitrate.Its identity with the substance obtained from the chloride of camphanic acid tells in favour of the view that it is an arnide. Further, and perhaps more conclusive, evidence is afforded by the fact that it was not found possible, even with Franchi- mont’s mixture of acetic anhydride and sulphuric acid, to prepare an acetyl derivative, the substance remaining unattacked after treatment for several hours at 130’. Camphnamic Acid, NH,*CO*C,H,,(OH)*CO,H.-This compound is easily prepared by warming powdered camphanamide with 10 per cent.sodium hydroxide solution, and, after precipitation with hydrochloricCAMPHANIC ACID AND OF BROMOCAMPHORIC ANHYDRIDE. 1291 acid, may be crystdlised from dilute alcohol. A specimen was analysed, after exposure a t 100' for 9 hours. 0-2172 gave 0.4529 CO, and 0.1613 H,O. The substance, when crystallised from dilute alcohol, contains 1 ruol. of water of crystallisation. It dissolves somewhat readily in alcohol, is sparingly soluble in ethyl acetate, and ingoluble, or nearly so, in benzene or light petroleum. From a mixture of methyl alcohol and ethyl acetate, it separates in beautiful, transparent prisms of consider- able size, which have a brilliant lustre, a well-marked plane of cleavage, and readily break up into thin plates. The double refraction is weak.The melting point of the substance is difficult to determine with any certainty, owing to the difficulty of getting rid of the water of crystallisation and the readiness with which the acid is converted into the lactone. When slowly heated, it melts at 160', as stated by Wreden, but when the tube containing it is plunged into sulphuric acid at 155O, fusion does not occur until a temperature of 165-166O is reached. I n all cases, an effervescence, due to the escape of water- vaponr, accompanies the fusion, and the residue soon solidifies, the mass melting once more a t 20S0, the melting point of camphanamide. A dilute solution of the ammonium salt gives no precipitate with solutions of calcium or barium salts; a white precipitate is produced with warm basic lead acetate, and with ferric chloride a nearly white precipitate is formed, which, on warming, disappears, the whole becoming reddish-brown.C=56.S ; H=8.2. CI,,Hl70,N requires C = 56.0 ; H = 7.9 per cent. Action of Behgdrating Agents on Camphanamide. 0 Camphanonitvile, CN*C,H,,<b, . Camphanamide is not easily converted into the corresponding nitrile ; zinc chloride, phosphoric oxide, or phosphoric sulphide either pro- duced no effect on it or converted i t into a charred mass. By the following process, however, n small quantity of the desired product may be obtained. The finely powdered amide is placed in a flask, uncovered, with a large excess of phosphorus trichloride, one molecular proportion of phosphorus pentachloride is added, and the whole heated on the water- bath until the particles of the amide are no longer distinguishable. The flask is ther, cooled, and the contents are poured slowly on to powdered ice which is kept well stirred.Chloroform is then added, the whole well shaken, and the chloroform extract shqken repeatedly with very dilute sodium carbonate solution, dried with calcium1292 THE CONSTITUTION OF CAMPHANIC ACID. chloride, and evaporated to dryness. The green or greenish-black, oily residue is extracted repeatedly with large quantities of light petroleum, and the waxy material which separates from the hot solu- tions crystallised several times from the same solvent, until it melts at 135'. A specimen dried in a vacuum was analysed: 0.2345 gave 0.5777 CO, and 0.1506 H20. C = 67.2 ; H = 7.1.CI,H,302N requires C = 67.0 ; H = 7.3 per cent. Camphanonitrile forms soft, fern-like crystals resembling those of ammonium chloride, which aggregate on pressure to a waxy, camphor- like mass. Melted on a slip of giass, it solidifies rapidly and com- pletely, forming skeleton crystals whose branches are rectangularly arranged, the whole being completely isotropic. The highest melting point observed for the substance was 135-137', but was never quite sharp, and it is possible that this temperature is a few degrees too low. When heated slowly at a higher temperature, the substance distils, apparently without much decomposition, and the vapour has an odour resembling that of raspberries. The nitrile dissolves in about five times its weight of sulphuric acid without decomposition, and if the solution is mixed with an equal bulk of fuming acid (25 per cent, SO3) and the mixture poured into ice, a small quantity of nearly pure camphanamide is obtained, showing that no structural change has been involved in the action of dehydrating agents on the amide.Action of Alkalis on Camphanonitrile. Tormution, of Canaphononic Acid. When camphanonitrile is covered with 20 per cent. aqueous potassium hydroxide, no immediate effect is observed, but if the whole is warmed or allowed to stand for some time the solid slowly disappears and a clear solution is finally obtained. On acidifying this solution, efferves- cence occurs and a strong odour of hydrogen cyanide becomes perceptible, the actual presence of this substance being easily shown by the usual tests. When the strongly acidified solution is extracted with chloroform in the usual way, a transparent, waxy mass remains on evaporating the extract.This substance crystallises from ethyl bromide in small prisms melting at 228' and has all the properties of camphononic acid. On analysis : 0.3713 gave 0.8583 CO, and 0.2760. In order to obtain icirther proof of the identity of the acid, it was This crystallised from metbyl alcohol in C = 63.1 ; H = 8.1. C,HI,O, requires C = 63.5 ; H = 8.2 per cent. converted into the oxirne.THE CHLORODIRROMO- AND DICHLOROBROMO-BENZENES. 1293 large, transparent rhombohedra which rapidly became opaque on ex- posure to the air, and as the substance melted at 186-187' the identity of the acid with camphononic acid was certain. Action of Sodium Hypobvomite and Sodium Hydroxide 0% Camphanamic Acid. Eighteen grams of ths amide were dissolved in 100 C.C. of 10 per cent. sodium hydroxide solution by the aid of warmth ; t o the resulting liquid 13 C.C. of bromine dissolved in 350 C.C. of sodium hydroxide were added, and the whole was heated on the water-bath for an hour. The excess of hypobromite was removed by means of sodium sulphite, and the liquid then acidified and evaporated to dryness, any unaltered amide which separated during this process being removed from time t o time. The residue was then extracted with dehydrated methylated spirit, and the extracted matter shaken with chloroform several times, the extract being evaporated to dryness, A waxy or oily residue remained, which was first triturated with ethyl bromide and afterwards crystallised from the same solvent. The substance thus obtained consisted almost entirely of pure carnphononic acid, as was proved by the methods j u s t described. By the treatment of 55 grams of camphanamide in this may, the recovered amide being repeatedly subjected to the same process, a quantity of pure camphononic acid weighing 17 grams mas finally obtained, and the process occupied only a few days. CHEMICAL DEPAKTMENT, SCHOOL O F PHARMACY, BLOOMSBUILY sQUAItE, W.C.