LAPWORTH AND HAKN : MENTEIYI, ACETOACETATB. 1409 CXLVII1.-Optically Active Esters of &Ketonic and P- A lclehydic Aczds. Purt I". Menthyl Acetoacetate. By A. LAPWORTH and A. C. OSBORN HANN. IN continuation of researches on the tautomerism of P-ketonic esters, the study of an optically active ester of acetoacetic acid was undertaken, as little is yet known as to the mode in which intercon- version of the enolic and ketonic Forms of ethyl acetoacetate is brought about, for although it is already well established that bases in small quantities are effective in establishing equilibrium, tke effect of acids has not been properly investigated. After a number of attempts had been made to obtain the menthyl ester from menthyl acetate by the action of metallic sodium-a method which was found to afford the desired compound only in small quantity-the action of menthol on the ethyl ester in presence of hydrogen chloride was tried. A preliminary experiment indicated that alcohol was evolved in Considerable amount before the hydrogen chloride was introduced, and we afterwards prepared some quantity of menthyl acetoacetate by the action of heat on a mixture of menthol and ethyl acetoacetate, a method which afforded excellent results.Shortly afterwards, we found that Cohn had obtained the compound by this process (Mornatsh., 1900, 21, 200), but as he shortly after- wards notified that he had abandoned the work (Bar., 1900, 33, 734), with his concurrence we proceeded with the investigation. The ease with which the ester is prepared renders it possible to produce large quantities of it in a short time, and we propose to employ the compound as the starting point in the preparation of a number of other active compounds which may possess interest from various standpoints.1500 LAPWORTH AND HA”: OPTICAT,LY ACTIVE ESTERS OF Cohn does not state that he had formed any definite opinion as to the constitution of the solid ester which is obtained by this process.We are disposed t o believe that it must be the ketonic form, and for the following reasons, which, it must be admitted, are not altogether conclusive. The greater proportion of ethyl acetoacetate at the ordinary tempera- ture exists as the ketonic form (Trans., 1892, 61, 808). As a rule, compounds are stable in the enolic condition when they exhibit strongly acid properties, whereas the acidic properties of menthyl acetoacetate are exceeding weak, as is shown by the fact that it dissolves only very sparingly in cold alkalis, although these bring about equilibrium between the two forms very rapidly.Again, in the case of substituted acetoacetic esters, in which the substituent is not an acyl group, it is invariably the ketonic form which is the stable one. Lastly, there may be mentioned the behaviour of the compound towards ferric chloride, which indicates that the maximum quantity of enolic ester is not present a t the first moment of dissolution in a solvent. Menthyl acetoacetate exhibits mutarotation in non-oxygenated solvents, the change occurring with a speed which renders it easy to study the effects of agents with advantage. It is interesting to observe that hitherto only those types of com- pounds which have been shown by independent chemical or phyrsical methods to exhibit desmotropy or isodynamic isomerism possess the property of mutarotation, a fact which serves to support Lowry’s con- clusion (Trans., 1899, 75, 213) that the two phenomena are very closely related.Thus the compounds in which mutarotation has hitherto been observed with certainty are the aldoses and pentoses, nitro- and r-bromonitro-camphor, nitrocamphane, a-benzoylcamphor, cam phorquinonepb enyl hydrazone, and men thy1 acetoacetate, and in all these cases there is every reason t o believe that isodynamic change is the cause. I n the case OE many other active menthyl esters obtained from the formylacetate and acetoacetate, no observable change of rotation occurs, and this was t o be expected, as there is no suflicient reason t o suppose them capable of desmotropic change, although in some instances their constitutions are capable of two modes of representation, differing only in the position of a hydrogen atom.The most. striking feature of the mutarotation of menthyl aceto- acetate is the fact that traces of acids as well as of bases aczelerate the change, The significance of this observation is referred t o else- where (p. 1513) and need not again be discussed. Menthyl acetoacetate has apparently all the chemical characters of the ethyl ester, but exhibits certain qualitative differences, inasmuch as its reactions are more sluggish, requiring in most cases much longerP-KETONIC -4ND P-ALDEHYDIC ACIDS.PART 11. 1501 periods for their completion, A similar sl.uggishness is observed in the crystallising power of the higher substituted menthyl acetoacetates which we have recently obtained, for where the ethyl esters crystallise very readily, the menthyl esters tend to form jellies, or remain liquid for many weeks. R x P ERIM E NT AL. Preparation and Properties of Menthyl Acetoacetate. -Cohn (Monutsh., 1900, 21, ZOO) heated ethyl acetoacetate with menthol in a reflux apparatus at about 150' for 5 hours, and finally fractionated the pro- duct by distillation under diminished pressure. We prefer t o heat the materials in a Wurtz flask over an Argand burner, allowing the alcohol produced to distil OE as soon as it is formed, and, when the action be- comes very slow, t o boil off the unaltered menthol and ethyl acetoacetate under a pressure of about 15 mm., the distillation being interrupted when a thermometer immersed in the vapour indicates a temperature of about 1 6 0 O .The residue is poured into a beaker, and on cooling it slowly sets to a nearly solid mass, which is ground into a paste, freed from most of the oily matter by the aid of the pump, and finally washed in the funnel with a very small quantity of ether. The pro- duct so obtained forms a white, crystalline mass. It may be recrystal- lised by dissolving it in dry ether and cooling the solution by a freezing mixture. A further quantity of the substance may be obtained from the lower boiling portions of the distillate and the attendant oil by refractionation or by conversion into the copper derivative.By very slow crystallisation, the ester may be obtained in the form of large, transparent prisms, in which, however, there is very little tendency to the development of good end faces. Through some of the larger faces, which are usually very bright, a biaxial figure of very small angle may be observed in convergent polarised light, emerging obliquely to the field; the crystals appear to belong t o the monoclinic system, in which case the bisectrix is in the plane of symmetry. The double refraction is positive in sign and weak. The melting point of the compound is given by Cohn as 30--32O, which is the melting point we have usually observed, but on one or two occasions we have noticed that a small quantity of the material in a thin-walled capillary tube remained for a considerable time at 38-40° without melting, and if the temperature were raised somewhat rapidly, fusion did not occur until a temperature of 43-45O was reached.On cooling, no matter how long the material had been kept in the fused condition, rapid solidification always occurred, the whole setting to a mass of flat needles identical in crystalline character with the large prisms above described. The solid substance is probably the ketonic form of the ester, for1602 LAPWORTH AND .€€Ah": OPTICALLY ACTIVE ESTERS OF when it is dissolved in anhydrous ether, benzene, or chloroform, and a drop of an ethereal solution of ferric chloride is added, no marked coloration is developed unless the liquid is boiled or is allowed to remain for some time; a coloration is at once developed in the cold, however, if a trace of an organic base, such as pyridine, is previously added. Solutions in alcohol, immediately after preparation, gave with ferric chloride a faint coloration which rapidly deepened, whilst if the solution had been allowed to remain for a few minutes in the cold the maximum colour appeared to be produced at'once.A violet coloration is also produced when a drop of alcohol is added to the nearly colourless solution of the ester in benzene or ether containing ferric chloride. I n fact, the phenomena observed were very similar to those which we had noticed in the caw of menthyl formylphenylacetate, and, as in that case, all attempts to prepare the desmotropic form of the ester were unsuccessful.The specific rotation of the compound was determined in several solvents. I n 1.5 per cent. solution in absolute alcohol, it had [a], - 68.5'. I n ethyl acetate, its rotation at 1.5 per cent. concentration was [a]D - 59*8', and remained constant. I n these liquids, no very marked alteration of the specific rotation with time was noticed. Menthyl acetoacetate exhibits the phenomenon of mutarotation in aon-oxygenated solvents in a much more marked manner than does the formylphenylacetate. The change, moreover, appears to be much more rapid, and the velocity is greatly affected by the presence of catalytic agents, and varies cousiderably with the degree of purity of the solvent. In the cases cited below, we have employed media which have been subjected to careful purification.I n chloroform, an initial rotation of about [a], - 68.5' was observed, and in the course of some hours this had attained a constant value of -70.7". I n light petroleum, the rotation was initially [.ID 65*1°, and rapidly reached a maximum value of -74.4'. I n benzene, with which we have made most of our experiments, the rotation changed from about [a], -61.5' to -67.4' or -68.4' for concentrations between 1 and 2 per cent., the duration of measurable change varying from 20 minutes to 2 hours, depending, no doubt, on the purity of the specimen of solvent used. The effect of catalytic agents on the change could be conveniently studied in this instance, and we made numerous experiments on the point.The mode of procedure was similar to t h a t adopted in the case of camphorquinonehydrazone (this vol. , p. 15 17), a series of readings with three or more tubes containing the same solution being made simultaneously in every case. It was observed that, as usual, traces of all bases, including evenB-KETONIC AND B-ALDEHYDIC ACIDS. PART 11. 1503 weak tertiary bases, increase the velocity of change, and that traces of acids produce a similar effect. The acids tried included acetic, trichloro- acetic, benzoic, salicyclic, and crotonic acids, and were found to be effective even in minute quantities, particularly so in the case of the fitrongest one-trichloroacetic acid. I n this instance, too, it was found that marked acceleration was produced even by such weak bases as the monobromo- and mononitro- anilines, although the effect was distinctly smaller than when more powerful bases were used in equivalent quantities.The effects of various other substances' soluble in benzene were also examined, but hitherto only those compounds the effect of which might be ascribed t o acidic or basic properties have been found effective when added in minute amount : certain compounds, such as alcohol, produce very marked acceleration, but only when present in comparatively large quantities. As an instance of the sensitiveness of the compound toward acids, it may be mentioned that in one experiment four tubes were filled with the same solution of the ester in benzene; the initial rates of change of rotation in each mere measured and found to be nearly the same.Into t w o of these tubes traces of trichloroacetic acid were introduced by touching a moist crystal of the acid with a platinum wire which was then dipped into the solutions. I n the case of the two unopened tubes, the mutarotation continued t o be observable for more than an hour, whilst in the two others the maximum rotation was reached in less than ten minutes. Derivatives of Menthy2 Acetoacetale. With the exception of the copper compound, the metallic derivatives of the ester have not been obtained in characteristic forms. The com- pounds obtained by the action of amines and hydrazines on menthyl acetoacetate are, as a rule, easily isolated in a crystalline form. The phenylhydrazide has already been described by Cohn (Ber., 1900, The copper derivative, (C,,H,303),Cu, is precipitated as an oil on adding copper acetate dissolved in twenty times i t s weight of water to an alcoholic solution of the ester.When left in contact with the alccholic liquid, i t slowly becomes crystalline, and was purified by cry stailisation from hot alcohol. It separates almost completely from cold 90 per cent. ethyl alcohol in dark green, transparent prisms, which contain alcohol of crystallisation ; the latter is gradually lost when the substance is exposed to the air, the compound then forming a bluish-green powder. When heated rapidly, the crystals soften at 78-80", but do not melt below 106'; the dry substance, on the 33, 735.)1.504 LAPWORTH AND HA"; OPTICALLY ACTIVE ESTERS OF other hand, melts sharply at 117-118' and slowly decomposes.analysis : On 0.8139 gave 0.1173 CuO. Cu= 11.5. (C14H2303)2Cu requires Cu = 11 6 per cent, The sernicarbazide, NH,*CO*N,H,* C(CH,)~CH*CO,-C,,H,,, was made by adding semicarbazide hydrochloride dissolved in a little water to a solution of the ester in alcohol; on warming gently, the product separated rapidly as a paste of glistening, flat needles, and was crystal- lised from hot alcohol. On analysis : 0.1985 gave 0.4423 CO, and 0.1635 H,O. C,,H2703N3 requires C = 60.6 ; H = 9.0 per cent. It dissolves readily in benzene, acetone, ethyl acetate, or hot alcohol, somewhat less readily in ether, and is nearly insoluble in light petrol- eum. It crystallises from alcohol in small, flat needles, from ether in elongated plates, and melts a t 143-144'.A 1.5 per cent. solution in benzene had the rotation [a], - 56.1' and no mutarotation was observed. The crystals from ether are usually four-sided, rhomboidal plates, and on these the extinction directions in polarised light are inclined at varying angles to ths longer sides ; thsy frequently show a biaxial interference figure, the optic axis emerging neaxly perpendicular to the field. The p-nit.lo~~enyZhyd4.ccxide, NO,*C,H,*N,H,*C( CH3) :CH*CO,*C,oHIS, was prepared by heating together the requisite quantities of the ester and p-nitrophenylhydrazine in alcoholic solution for several hours on the water-bath and precipitating the product with water. The oily product, which did not show any sign of crystallising after some days, was fractionally extracted with cold light petroleum, when a consider- able proportion was obtained, on evaporation of the solvent, in a crystalline form.C = 60.7 ; H = 9.1. It was purified by recrystallisation from alcohol : 0.2275 gave 0.5320 CO, and 0.1626 H,O. C = 63.8 ; H = 7.9. C20H290,N, requires C = 64.0 ; H = 7.7 per cent. The hydrazide is readily soluble in warm alcohol and most of the usual media, but is only sparingly dissolved by light petroleum and is insoluble in water. It melts at 105-106'. A solution of 0.3533 gram in 25 C.C. of benzene gave [ u ] ~ - 42*5', no mutarotation being observable, even after addition of an acid or a base. The crystals from alcohol are transparent, brownish-yellow prisms or pyramids. When melted on a microscope slide beneath a cover- glass, the substance solidifies very slowly on cooling, forming small,,@-KETONIC AND ,&ALDEHYDIC ACIDS.PART 11. 1505 flat, isolated, four-sided plates, in which the extinction in polarised light is parallel to the smaller pair of sides. When the crystals are crushed and examined in convergent polarised light, many fragments show a biamial interference figure of wide angle. The double refraction is unusually strong. NH,*C( CH,): C H*C02*C,oH,,. - When ammonia is passed into menthyl acetoacetate at a temperature slightly above the melting point of the ester, it is only very slowly absorbed, and even after some dayathe action does not appear to be complete. If, however, a small quantity of ammonium acetate is introduced into the liquid, reaction occurs much more quickly, globules of water separate, and the whole finally sets to a semi-solid mass.I n order to purify the product, it was spread on porous earthenware and then crystallised Erom hot alcohol. Menthyl P-Arnirtocrotonate, On analysis : 0.2006 gave 0.5179 CO, and 0.1906 H,O. C = 70.4 ; H= 10.6. C14H2502N2 requires C = 70.3 ; H = 10.5 per cent. The compound dissolves fairly readily in most of the usual organic media and separates from alcohol on spontaneous evaporation in large, brilliant,, transparent prisms, with one good plane of cleavage. It melts at 88-89', and after solidification melts once more at the same temperature. A benzene solution containing 0.3895 gram of the substance in 25 C.C. of benzene was examined ; this gave [a], - 105.2', and no altera- tion of this value was observed.Small, elongated crystals of the compound are sometimes obtained ; these have straight extinction in polarised light, the directions of greatest length and elasticity being parallel. Crushed fragments of the crystals examined in convergent light frequently show a biaxial interference figure of moderate angle; this figure is probably that corresponding with the acute bieectrix ; the diqpersion is strong, the angle for red light being greater than that for blue. Menthy2 /3-Benxy Zaminocrotonate, C,H,*CH,* NH*C(C€€:,): CH*C02*C,oH1,. -A mixture of menthyl acetoacetate and benzylamine in molecular pro- portion was warmed gently until the solid had disappeared and was then allowed to remain at the ordinary temperature for some time.A milkiness, due to separation of water, soon became noticeable and at the end of six hours the whole was solid. The mass was dissolved in a little hot absolute alcohol and the crystalline separation purified by recrystallisation from absolute alcohol. On analysis : 002006 gave 0.5638 CO, and 0.1732 H,O. C = 76.7 ; H= 9.6 C2,H,,02N requires C = 76.6 ; H = 9.4 per cent. VOL. LXXXI. 5 11506 LAPWORTH AND HANN: OPTICALLY ACTIVE ESTERS OF The substance dissolves readily in benzene, chloroform, carbon tetra- chloride, ethyl acetate, acetone, or light petroleum, but is less readily dissolved by methyl or ethyl alcohol. It separates from hot absolute alcohol in flat needles and melts a t S5-86O. A solution containing 0.4032 gram in 25 C.C. of benzene gave [.ID -59.8O and this value remained constant, The crystals are elongated rectangular plates or flat needles, which show straight extinction in polarised light ; the directions of greatest length and elasticity are parallel. Menthyl p- Anilinocrotonnte, C,H,*N H*C( CH,): CH CO,* CloH19.- This compound was made by the interaction of aniline and menthyl acetoacetate in molecular proportion in the cold; the action is complete in 24 hours, and the product, which is somewhat oily, may be drained on porous earthenware and finally crystallised several times from methyl alcohol.On analysis : 0.2022 gave 0.5670 CO, and 0,1679 H,O. C=76*5 ; H ~ 9 . 2 . C,,H,,O,N requires C = 76.2 ; H = 9.2 per cent, It dissolves somewhat readily in most of the usual organic media and separates from methyl alcohol or absolute ethyl alcohol in thin plates or flat needles.It melts sharply a t 89-90', and after solidifica- tion fuses once more a t the same temperature, A solution containing 014045 gram in 25 C.C. in benzene gave [a], -98*2", and this value remained constant. The crystals from absolute alcohol are usually flat, rectangular plates which probably belong to the monoclinic system ; in polarised light, the extinction is straight and the directions of greatest length and elasticity are parallel. In convergent polarised light, the obtuse bisectrix of a bixial interference figure emerges through the larger face a t an angle inclined to the perpendicular; the axial plane is coincident with the plane of symmetry and the double refraction is moderately strong and negative in sign.When the substance is melted on a glass slide beneath the cover-slip, i t solidifies very slowly as the temperature falls, to masses of large, irregular plates identical in character with those above described. Action of Acid Ciiloyides on Xenthyl Acetoacetate and its Metallic Derivatiues.-Acetyl chloride acts slowly on the ester, hydrogen chloride being evolved, but nothing of definite character could be isolated from the product ; when, instead of the ester, the sodium or copper derivative is used, very rapid action occurs, but from the pro- duct nothing but the copper compound of the unaltered ester could be isolated. Unsatisfactory results were also obtained when attempts were made to prepare the 0-acetyl derivative from the ester by treat- ment with a mixture of acetyl chlaride and pyridine by Claisen's method.&KETONIC AND &ALDEHYDIC ACIDS.PART 11. 1507 More definite results were obtained by acting with benzoyl chloride on the sodium derivative of the ester suspended in ether. The action, which was apparently a slow one, was completed by the aid of heat and it should be mentioned that, even after several days' heating on the water-bath, no deposit of sodium chloride was observed, although the odour of the benzoyl chloride had entirely disappeared; we are un- able to offer any explanation of the fact. The ethereal solution was finally shaken with ice-cold water to remove sodium chloride and was then extracted repeatedly with a solution of sodium carbonate.On evaporation, the residual ethereal solution gave an oil which resisted all attempts at purification. From the sodium carbonate solution, however, an oil was deposited on acidification'; this was dissolved in alcohol and the solution treated with aqueous copper acetate, when a copious deposit of an insoluble copper derivative was obtained. The copper compound crystallised from hot ethyl acetate in slender needles of a greenish-blue colour. On analysis : 0,5053 gave 0-0527 CuO. Cu = 8.4. (C21H2704)2 Cu requires Cu = 8.4 per cent. The substance is sparingly-soluble in alcohol or cold ethyl acetate or acetone, but dissolves somewhat feadily in cold chloroform or benzene. When heated slowly, it darkens at 210--215O and melts and decomposes a t 226O, but if heated suddenly to 230' i t melts some degrees above this point.I n order to obtain the free ester, the copper compound was sus- pended in purified ether and shaken with dilute hydrochloric acid until the crystals had disappeared ; on evaporation of the washed and dried ethereal solution, a substance, which formed a very thick oil at the ordinary temperature was obtained. This decomposed when dist,illed under the lowest pressures which we were able to obtain, and was therefore dried at 100' and analysed : 0.23'75 gave 0.6358 CO, and 0.1802 H,O. The compound has therefore the composition, as well as the 'general properties of menthyl benzoylacetoacetate, COMe-CHBz*CO2*C,,Hl9, and is, in all probability, a mixture of the enolic and ketonic forms of that substance ; its alcoholic solution gave a deep brownish-purple solution with ethereal ferric chloride.A 25 per cent. solution of the compound in benzene gave [a]= -44.3' and this value remained constant during a week. In order to prove that a benzoyl group was present in the substance, it was cxidised with boiling alkaline permanganate solution and a con- siderable quantity of benzoic acid was obtained ; during the oxidation, a C=73.0; H=8-4. C,1H280, requires C = 73.2 ; H = 8.1 per cent. 5 1 21508 LAPWORTH AND HA”: powerful odour of menthol was observed. That the acetyl group was still present as well as the benzoyl group was evident from the strongly acidic properties of the compound, which dissolved readily in sodium carbonate solution. Menthyl acetoacetate reacts only slowly, if at all, with phenyl- carbimide; a mixture of 1 mol. of the latter with 2 mols. of the former remained unaltered in appearance during several months, and even after further heating on the water-bath retained the odour of the carbimide. The bromo-derivatives of the ester do not crystallise readily, nor do they appear to afford crystalline copper derivatives. The products obtained by the actioh of 1, 2 or 3 mols. of bromine on the ester were carefully examined, but in all cases they remained oily and afforded oily copper derivatives; no better results were obtained on attempting t o prepare the y-bromo-derivatives. The products obtained by the action of aldehydes and of diazonium compounds on the ester are a t present under examination. We pro- pose also to extend this study of optically active esters toactive esters of other ,&?-ketonic and P-aldehydic acids, and also to derivatives of menthyl cyanoaee tate. I n conclusion, me desire to express our thanks to the Research Fund Committee of the Chemical Society for a grant defraying much of the cost of this work. CHEMICAL DEPARTMENT, THE GOLDSMITHS’ INSTITUTE, NEW CROSS, S.E.