NOVEMBER 1957

%CARBOMETHOXYTROPTNON~

1385

[CONTRIBUTION FROM THE NATIONAL INSTITUTE OF ARTHRIIIS N D METABOLIC A DISEASES, NATIONAL INSTITUTES O F HEALTH, PUBLIC HEALTH SERVICE, 8. DEPARTMENT HEALTH, OF EDUCATION WELFARE] AND

u,

Concerning 2 Carbomethoxytropinone*
STEPHEN P. FINDLAYt

-

Received June 25, 1967
Racemic 2-carbomethoxytropinone is obtainable by the partial saponification of 2,4-dicarbomethoxytropinone and, more conveniently, by the condensation of monomethyl p-ketoglutarate, got from P-ketoglutaric anhydride, with succindialdehyde and methylamine. 2-Carbomethoxytropinone can conceivably exist in three racemic and six optically active forms. Of these only one has been obtained heretofore. The configurational relation of d-(2-~arbomethoxytropinone)and its E antipode to 1-cocaineis established by the Kiliani chromic acid oxidation of pseudoecgonine methyl ester to the former. Previous methods of preparing racemic 2-~arbomethoxytropinone, properties of 2,4dicarbomethoxytropinone, and incidental experimental the data are discussed.

In any synthetic scheme for cocaine (11) and its stereoisomers, 2-carbomethoxytropine (I) is an obvious and probably the best key intermediate and has indeed been used to prepare d- and 1-cocaine and racemic pseudococaine.'J In connection with the synthesis of the other two isomers, allococaine and allopseudococaine, unreported until recently,a relatively large amounts of this @-ketoesterwereneeded, and it was accordingly desirable to ascertain the most efficient method of its preparation. Of the four synthetic methods recorded, one (Sequence A) is rather involved, and the cyclization of l-methyl-2,5-dicarboethoxypyrrolidine(111, R = CzH6)to 2-carboethoxytropinone (IV, R = GH5) diffi~ult.'~~J Another (Sequence B), in principle the simplest, is reported to give only a twenty per cent yield of product.6 Hence, only the remaining two 7 (Sequences C and D2),which are relatively simple and claimed to proceed in good yield, came into consideration. On the basis of the yields reported and the availability of the starting materials Sequence C seemed preferable. Tropinone (V), the components for the Robinson Synthesis of which are now obtainable commercially (see Experimental), can be made in high yield (- 75%)8,9 and, according to Preobrashenski,' is readily convertible (70-80%) by sodium or potassium and dimethyl carbonate in boiling benzene or xylene to 2-carbomethoxytropinone. It was here found, however, that, although the re-

ported yields of tropinone are not exaggerated, this base is thus transformed to 2-carbomethoxytropinone only in rather low yield (- 35%). For such a process many side reactions are conceivable, and in fact some or all of them do occur and produce large amounts of resinous by-products.
COzCH:,

I 0

I1

1 ) €In-Pt 2)Na in
p-RONa -cymene

&'""
NCHI IV
0
NCH?
I\

B.

+ CH&HO)z +CHZNH2+ CH~COCHZCO~CH~ I x0
Na (or K),

c. tC H ~ C H O ) ~ + C H ~ N (H ~O ~ C C H ~ ) ~ C O H +

-

*This paper is a contribution in honor of Lyndon F. Small, former Editor of the Journal. (1) R. Willstatter and M. Bommer, Ann., 422, 15 (1921). (2) R. Willstatter, 0. Wolfes, and H. Miider, Ann., 434, 111 (1923). (3) S. P. Findlay, J. Org. Chem., 21, 711 (1956). (4) R. Willstiitter and A. Pfannenstiehl, Ann., 422, 1 (1921). (5) The iise of the dimethyl ester (R = CHs) would presumably not improve the over-all yield of sequence A appreciably. (6) German Patent 345,759. ( 7 ) N. A. Preobrashenski, M. N. Schtschukina, and R. A. Lapina, Ber., 69,1615 (1936). (8) C. Schopf and G. Lehmann, Ann., 518, l(1935). (9) L. C. Keagle and W. H. Hartung, J. Am. Chem. Soc., 68,1608 (1946).

0

V

Certain variations of the reaction were also tried. Substituting sodium methoxide in toluene did not alter the outcome appreciably. Sodium methoxide in a small volume of methanol gave a somewhat lower yield. By increasing the relative amount of methanol and prolonging the time of reaction correspondingly the product is markedly reduced and

t Present address: 206 Branch Avenue, Red Bank, New Jersey.

(12) R. have nearly of the carboethoxyphloroglucinols. methyl P-ketoglutarate by dissolving it in methanol. J. London. SOC. Its formation in high yield is conised to be more satisfactory. H. The number of isomers. . which precipitated slowly and of diethyl carbonate which may be a decomposition product appeared to be only partly crystalline. l 3 The large of the keto ester was realized in this manner. effected more readily than that of the parent compound. racemic %-carbomethoxytropinone was thus obtained in satisfactory quantity (52-5970). failure of this representative to exhibit either reBy a reversal (E) of Willstatter's sequence one producible or sharply defined physical properties is therefore not surprising. 1874 (1911). 205 newly liberated anhydrous base and melting about (1906).. 111. Anhydrous 2. Chem. Ormerod. of cycloheptane. 2 2 .chloride or similar salt: the greater the proportion of potassium salt of monomethyl @-ketoglutarate. 5277 (1946). CI2I-I. (b) H. however. Am. also. Geserick. @-Ketoglutaric anhydride (VII).4-dicarbomethoxytropinone (VI) appears to proD. and the ory to the low harvest of pure dipotassium salt. -NaHC03 Although the optinium conditions for its condensa(64%) tion subsequently with succindialdehyde and 0 VI methylamine were not determined. 68. Cornelius and H. SOL.Fdicarbo(IO) The strength bf the indicated bonds. Ber. The chief difficulty of tingent on the use of methylamine as its hydroWillstatter's method lies in preparing the pure di. J . essential im. is converted to the requisite mono1 -CH. This procedure is. Robinson prepared a solumethanolic potassium hydroxide affords a high yield tion of the 2. the larger the quantity of dark. 63.the free amine used. For both simplicity and VI1 VI11 yield this sequence isverylikely superior to all others CHgNHe. 762 (1917). tained in excellent amount by a modification of KauCOZCHJ . Soc. and D. . excess might hope to avoid some or all of the foregoing solvent reversing the reaction presumably. 2252 (1941)l excess hydrate and melting about 146". Leuchs and A. -COT (5?7 2.procedure. Robinson. SOC.p. consisting in the use of monomethyl &ketoglutarate as its methylammonium salt. were obsodium ethoxide do give under certain conditions about 10% tained.. (CHsOzCCHifiCO ceed in high yield. The binoxalate salt seems to exist in two modifications H. 19. both geometric and tautomeric. formation of other potassium salts of lesser solu. Homeyer.. Unlike the condensation. A.5-84') and the other tan (m. H. 88-91'). Ber. 1446 (1886). also convert.OH. Leuchs and F. be readily interconvertible.4-Dicarbomethoxytropinone(VI) which may be the first isolated dicarboalkoxy derivative of troI n these circumstances Willstatter's synthesis2 pinone is a sufficiently unusual compound to merit (Sequence D) or some modification thereof prom. Fieser and AI. 4171 (1908).count for the apparently low yield of tropinone obno provement of Willstatter's over-all yield (-30%) tained therefrom by acid hydrolysis. Although it was found polymeric by-products. Chem.p. (13) R. the other from the methanol lowers the yield in such reactions. tedious. and another variation (F) of Willstatter's synthesis.12 which was oblNaOH NCH. (11) (a) E.. yield being less than 50%.17. Pechet. the chief products of this reaction. . was therefore investigated. Wallingford.Indian Chem. but two possibly distinct modifications of a solid one white (m. Ber. lo complications. Indeed the Robinson Synthesis of 2.4-dicarboethoxy analog by a similar of somewhat impure dipotassium salt.138 (1940). I n his classical article on the that treatment of dimethyl &ketoglutarate with synthesis of tropinone. and its partial saponification is. malonic ester and 82. 44.~N0~~1/2H~0. J .some description. H. one being obtainable directly from the hemiM. These forms.. Jones [ J . Chem. one need not expect sodium alkoxides or enolates to effect acid cleavage of .1386 FINDLAY VOL. partly because of the basic nitrogen undergo self-condensation in the presence of basic atom and partly because of their presumably similar catalystsll are probably the chief factors contribut. Kaushall.stabilities. (c) I F.. His use of free methylamine may acible to 2-~arbomethoxytropinone. According to V. 22 much unreacted tropinone is recoverable. Soc. being different. moreover. NaO-R. Chem. 41. NaO-H and methoxytropinone seems to exist only as a liquid. Proc. of dimethyl ioketoglutarate this proceeds with fewer side reactions in alkaline solution. observed to yield a dialkyl carbonate. While no ester of acetoacetic acid itself has seemingly been hemihydrate.(n70) shall's procedure.%keto esters. Am. Simion. Pechmann.*. discussed in more detail below. presently known and probably represents also a conCCH~CHO)~ ------+I venient means of synthesizing certain derivatives -2H90. should. M.which are possible for compounds of such strucbility and the ease with which 0-ketoglutarates ture. indistinguishable infrared spectra in S u j 01. v.

One concludes that. which appeared to be largely a monohydrate. The inhomogeneity of anhydrous 2-carbomethoxytropinone was confirmed by infrared measurements. The L-bitartrate dihydrate has a specific 107.6". and a molecular model of 2-carbomethoxytropinone suggests that the bridged-ring structure characteristic of the tropane alkaloids contributes to the stability of this tautomer. Although 2-carbomethoxytropinone (I) has three asymmetric carbon atoms and hence. which crystallizes well and melts at 176". Furthermore. the high and An analytically pure sample.3". is got from acetone solution. t o be expected.that of the pure anhydrous enol.5' It is noteworthy that Preobrashenski' reported the Dihydrate 93-96' ( 100°2) 98. of course. and. From Racemic d. melted a t 103. similarly. it exists largely or entirely as the enol when None of the numerous batches of anhydrous hydrated or dissolved in a hydroxylic solvent.4" were obtained. In. as usually obtained. In one instance needles melting at 104. enol form te recrystallize the base. and from TABLE I aqueous methanol a granular modification slowly MELTING POINTS OF 2-CARBOMETHOXYTROPINONES A N D separated which melted at 97. which is granular or powdery and melts about 167". The ultraviolet spectra of the two esters in absolute alcohol demonstrate the predominance of the enol IX X form of such compounds in polar solvents. SS"). (164") These indications that racemic P-carbomethoxy137' and 145' 142' Binoxalate tropinone.4-dicarbomethoxytropinone hemihydrate appears to exist exclusively as the enol.5' hydrate to melt at 98-98". 104. Both binoxalates regenerate one and the same form of racemic 2-carbomethoxytropinone. and melts at 145" with little or no decomposition. generated from the sharp melting point reported by him may represent pure picrate and then sublimed.oxytropinone) has a specific rotation in methanol of tained in this work crystallized from aqueous ace18. separates from solutions in methyl alcohol. is also possible and the benzoyl derivative thereof has indeed been prepareda2 According to Willstatter.2 racemic 2carbomethoxytropinone melts at 111 " and its monohydrate at 100". In aqueous sodium sulfate or in aqueous acetone equivalent amounts of 2. Another form (m. seemingly. able mixture of easily interconvertible epimers or tautomers were supported by the isolation of two picrate salts. two picrate salts.5-104. One. Like the racemic modification it crystaltone or from aqueous methyl acetate as colorless lizes in the presence of water as a dihydrate and needles melting a t 93-96" and having the com. although this @-ketoester in the anhydrous state is usually obu OCOCsHs tained as a mixture of ketone (IX and/or X and U. the formation of only two of these (IX and X and their antipodes) is.5-101.and more distinctly at 108. Since racemic 2-carbomethoxytropinone prepared in this Willstatter employed methanol which favors the investigation melted sharply or as high as 111".tl XI1 their optical antipodes) and enol (XI and its antiXI pode).Recovered from the L-bitartrate.5-98" and had a comSOME THEIR OF SALTS position approximating that of a trihydrate. four racemic forms. On the other hand the dihydrate in Nujol gives no evidence of the presence of the keto form. the other. Anhydrous Base 103.rotation in water of +15. It melts higher position of a dihydrate.NOVEMBER 1957 2-CARBOMETHOXYTROPINONE 1387 155".4-dicarbomethoxytropinone and sodium hydroxide react slowly to give a yellow precipitate which may be the sodium salt of the half ester and which is decomposed by mineral acids to carbon dioxide and racemic 2-carbomethoxytropinone.water. Neither he nor Preobrashenski' looked for or found more than one racemic form.5-109.and D-bitartrates from higher.or 1all of these forms the anhydrous base can be obtained by sublimation or distillation in high vacuum. d-(2-carbomethdeed the only analytically pure hydrated form ob.5" (Table I). was also obtained from aqueous acetone. is readily dehydrated.5. The other crystallizes from water as a dihydrate. is less readily dehydrated.6' Picrate 167-168'and176" 174and 176. 2. Another kind of racemate. In chloroform solution the base absorbs in regions characteristic of both keto and enol structures. is a somewhat vari- f i II + + . but whether these two forms have the same molecular constitution has not been established before. 81. that of the enol form (XI and its antipode). no monohy.p. and chromatography of a somewhat less 2-Carbomethoxytropinone was readily resolved pure sample furnished material which melted no by recrystallizing its L.4" ([M]g 5898 140).5.6' (111 '2) 108.gives. drate melting at or near 100" could be isolated. theoretically.5'-109. Two binoxalate salts were also found. One form may be converted to the other merely by recrystallization from the other solvent. One crystallizes from water as a sesquihydrate. and melts about 135" with decomposition.

L e . Schaub. and photochemical oxidation experiments compound in good yield by partial saponification were unsuccessful. NCHa Mannich likewise obtained better yields of keto diesters by using methylamine as the hydrochloride rather than as the free base.. 1954. 97 (1952). cocaine.rather laboriously separated from large quantities tute for supplying this reference. liquid by-product. reported:23 n. Carbide and Carbon 2. Findlay. the latter is neuexperiment. .21Distillation had the exisomers. Ber. 17. Kiliani. In the seven examples BO far examined only one exception to this rule has been found. rather than their mirror images. IX. he was unable to obtain this last mination. Exploratory experiments to determine whether esterd which was isolated as the optically active racemic 2-carbomethoxytropinone can be got from hydroace tate. R.. R.p.1388 nich concerning the preparation and properties of the analogous 1. Soc.~ (XII)” and must therefore have the structures.and n-bitartrates of d-(2-carbomethoxytropinone) (see Experimental). and L. known and unknown. Vol.12 The low tity of a high-boiling. Progress in Stersochmistry. more positive. the oxidation of I-ecgonine methyl ester (XIII) and its Cz-epimer.869(1950). which is yield of the p-keto ester may be owing both to the presumably the other possible form of 2. Although catalytic dehydrogenation. J.312 ( 1 955)).5-diethoxytetrahydrofuran or succindialdehyde.On the other hand. Ber . Am.5-diethoxytetrahydrofuran (na: 1. H.Landolt. E. (18) W. appeared to be figurations of the Cs-hydroxyl group.2. which were indebted to Dr. catalytic pected effect on the keto-enol equilibrium. Walter A. Long. a small yield of the keto ester as and found that catalytic hydrogenation of its hydrothe d-antipode was a t length obtained by oxidizing chloride resulted in the formation of racemic peither epimer in acetone with Kiliani’s chromic acid hydroxy esters. 72. 2855 (1954).the products. (23) J. noted the instability of such compounds and their salts in hydroxylic solH OH vents.5-diacetoIt resistance of equatorial hydroxyl group^'^^^^ to oxi. By using the value of the molecular rotation lor ammonium bitartrate [H. Baker.4208. Fakstorp. The results here set forth are in essential agree. A’oc.yield by adding dry methanol to the mixture re0x. Am. (14) 8. tral.20 Miscellaneous Observations. X I . New York.2 Inasmuch as racemic 2-carbomethoxytropinone dride. The former is basic and (1 6) Dr. Pharm. 676 (1913). also represents the absolute configuration of 2-cocaine [E. Yoshio Sato of this Institute kindly called this method to my attention and condurted the first oxidation gives a well-defined binoxalate. of viscous by-products. isolated as the picrate. EXPERIMENTAL^^ Materials. Soc. J. Chem. Chem. (22) All melting points recorded herein are corrected and were observed in Pyrex glass capillaries. than the calculated molecular rotation of the corresponding ammonium ion.. J . Indeed. I-cocaine. 22 ( 1 7 ) The etructure. Chem. J. pp. and the readily available methyl acetoacet.methylamine.tropinone just described is tantamount to the sulting from the reaction of anhydrous citric acid direct synthesis of all the optically active cocaine and fuming srxlfuric acid. Dimethyl P-ketoglutarate was prepared in 65% been converted in turn to the four possible racemic c o c a i n e ~ . 119-124”). 815 (1901)] or that of bitartrate calculated from the rotations of the L. one finds that the molecular rotation of &(2-~arbornethoxytropinone)i s more dextrorotatory. P. A s will be set forth in mor? detail elsewhere. Hardegger and H.6-trialkyl-3. which have racemic anhydroecgonine methyl ester. Schniepp. in two seemingly new compounds.the ~ ~ ? resolution of racemic 2-carbometh. was undertaken.. Chem.2. CllH18N202 (m..16s16Accordingly.6-trimethyl-4-keto-3-carboXI11 XIV methoxypiperidine by means of the bitartrate salts. Raleigh. 23. Chim.l 4 d-pseudoecgonine methyl ester (XIV).g 1. prepared solution. (21) Cf. The acreduction of the d antipode in aqueous acetic acid tion of fuming sulfuric acid on dimethyl citrate profurnished what is presumably alloecgonine methyl duces this ester directly in small yield.. Org. D.Acta. Iilyne. 1073 (1873). 76.4164). Querry. d-(2-carbomethoxytropinone) is derived configurationally from 1-cocaine approximately according to the procedure of Keagle . 272. and J.p. Helv. Am. .and H a r t ~ n gwas contaminated by a small quanXI. dehydrobro. *” FINDLAY VOL. Ott..ate resulted.2. ment with the long and excellent memoir of Man. 6 . Z-cocaine and its derivativep are more dextrorotatory than the corresponding salts. M. Williams. according to the conditions. One of with Catalytic hydrogen3 to the two possible con. Jacobs of The Rockefeller Insti. I am 166’) and C1GHzrN06(m.8 was characterized as its dation and to the ease with which this product is picrate and bis-p-nitrophenylhydrazone.5~-dicarboalkoxy-4piperidones from which mere obtained “open” tropine. and their derivative^. V. 323 (1934).'^ A few of the similarities of and the differences between these two similar classes of compounds are noteworthy. Arch. Mannich. Tropinone. 9 To ascertain its configurational relation to the naturally occurring. Charles Pfiaer (19) C. (20) 2-Carbomethoxytropinone hydrochloride is under similar conditions nearly inert (unpublished observations of the author). racemic 2-carbomethoxytropinone gave a has been reduced both with sodium amalgam2 and mixture which could not be readily purified. 1. and resolved 1. E. J . H. The Academic Press. 60 and 63. 46. parent alkaloid. (15) H. Reduced with ethereal lithium aluminum hyfurther oxidized.nyl-l-methylpyrrolidine. footnote 1. 38. B.

Anal. ( b ) Using sodium methoxide. It was recrystallized by dissolving in hot acetone. C. 202 ml. which is extremely soluble in acetone but sparingly so in methanol and alcohol. Mixed with water (1.P. heated quickly to boiling. 10 (1946). 14th Revision.) and a residue of trimethyl citrate (3.70 g. . gave 1.). which gradually darkened during the subsequent heating.0 g.) was added sodium (0. Unlike pure 2-carbomethoxytropinone this material soon darkened. N. conducted the same way except by using less methanol (2 ml.. 0.) of the high-boiling residue from the first distillation (above) gave a small amount (-0.59. Kranig. 153".5 9. The initially nearly colorless solution..7 g. Kogerman and J.6 g.48 g.5 ml.p. L.p.4452 (after several days). (27) The refractive index of the freshly distilled ester changed much more rapidly between the prisms of the refractometer than i the container. Wis. b. 1. under nitrogen with 0. J.p. d2: 1. Pa. cooling.4538 After a week Fraction 1 had n' 1.) containing ammonium chloride (2. 50 g. of refluxing. 133'1. of refluxing were used.5 hrs. The mixture was stirred 1 hr.. S.8. A small fore-run (5 g.. @-Ketoglutaric acid itself is perhaps obtainable in better yield than reported. the extracts dried (NalS04).35 g. b. A t the height of the reaction the foaming mixture fills a 2-1. (29) The Pharmacopeia of the United States of America.60 mm. D-tartaric acid was obtained from Aldrich Chemical Company. which is efflorescent (Merck Index. Tropwnone.. The initial and final pH's were about 6. 13. m. 3943.7-78"/0. It.) and washed twice with a mixture of saturated aqueous potassium carbonate (6 ml.4463) afforded two fractions: + nz (Immediately Fraction B. 107-110°/23 mm. 5. When 31% fuming sulfuric acid (265 9.p. 47. its bis-p-nitrophenylhydrazone was obtained as small.0 ml. L-Tartaric acid (the naturally occurring.27 . The amount of fuming sulfuric acid employed is that calculated to convert all the water theoretically formed to sulfuric acid. yellowish-orange.). conducted like the first except that toluene (10 ml. 2. p. N. Distillation of nearly pure dimethyl @-ketoglutarate (100 g.250 mole) in water (1800 ml.Hz2NaO~: 47. portions to fuming sulfuric acid (383 g.. Sodium methoxide (1. ny 1.) was added dropwise a t -5 to 0" (3 hrs. basic oil which became red on keeping.) in dry methanol (5. m. 1. From two such preparations was obtained after two distillations tropinone (40 g. under nsrogen with occasional % agitation. (24) P. (36%) of yellowish deposit.. The Mack Printing Company. initially colorless. (25) Cf. 18. and methylamine hydrochloride (16. The base.025 mole) and dimethyl carbonate (4.. Inasmuch as the base is scarcely darker after a month than after the first day. J. of 2-carbomethoxytropinone. protected against moisture and carbon dioxide. dextrorotatory tartaric acid. 1950. a t 25". 40 g. 18. N. was refluxed 20 min. The reaction of tropinone with dimethyl carbonate. The mixture was kept 13-20 hrs. After distillation) 1 77-77. picrate.4539 2 77. the sodium dissolving with effervescence within about 13 min.65-0.5 g. the mixture was extracted with cold water (2 X 10 ml.02: N. yellow mixture absolute methanol (500 ml. was added in 32-g.) and 3N potassium hydroxide (3. gave 0.202).) (ng 1.) which was collected after several hours and which darkened on keeping overnight. Eastman Kodak dimethyl carbonate was dried over potassium carbonate and distilled from silver carbonate: b. Anal.0 ml. 252). 6. and 17 a t 25". Yang. 1. 21%) with stirring. and purified as described above: 0.8 g. and the solvent removed on the water bath and then in vacuo.p..00 mole).p. the discoloration is no doubt caused by a small quantity of unstable impurity.).2551.) were used without further purification. and G. 1928 11. The oil which separated a t the interface was taken up in saturated aqueous ammonium chloride. That of the equilibrated n ester does not change between the prisms. and the dried extracts concentrated in vacuo to a brown oil (5. N. 3 a t 35O.. The procedure in Organic Syntheses** prescribes U.2N sulfuric acid (1 f ml.0 ml. loses its moisture by exposure to air in dry weather. Prepared in and purified from methanol. m.3687). was purified from the last solvent: slender prisms. stout prisms having hexagonal faces.0 g. . Chem. flask.*g but (26) Citric acid.. ice (500 g.). The evolution of carbon monoxide is not complete even after 24 hr. 200". Calcd. 13. 0. of 2-carbomethoxytropinone. Soc.8".) in small pieces. 0. 112114" (softening at-102"). I .) and 10 min. COIL Vol.45. 56..91: H. anhydrous citric acid*E (192 g. 97-102' (reported9 111").48. also known as bthrearic acid) was a Mallinckrodt Chemical Works product.) and 30 min. C.200 mole) was stirred 15 min. 0.).2 g. (28) Org. Another experiment.) were obtained from the first distillation.. 167. which contained some water. a t 25" and added to a stirred mixture of sodium bicarbonate (700 g. the aqueous extracts mixed with ammonium chloride (2.7"/0.6 g.). The dried extracts were concentrated to an oil which was taken up in ether (100 ml. more readily by heating in vacuo at 60".) of colorless.40. 72%).7 mm. 1.83.p.) was added to a solution of disodium phosphate (142 g.14. Biichi. citric acid. Evaporative distillation (80°/l mm. Dimethyl @-ketoglutarate. N. ( a ) Using sodium in xylene? To a warm solution of tropinone (3. Fisher d i m methoxide. m. Powdered.) and extracted with chloroform (4 x 50 ml. H. Zentr.) at 10" and the whole kept 3 days at room temperature.4466. Meinwald. The residual oil.) and kept 4-5 hr. Am.). and rubbing to induce crystallization.) and extracted with chloroform (5 X 50 ml. was distilled in uacuo: after two distillations 112 g.p. Easton.26 The homogeneous solution was then stirred with BaCOs (8 9. the yield was 10% lower. H. Chem.65-0.) and dimethyl carbonate (6. Calcd. na. acquires a brown coloration in a matter of hours.88.P..025 mole) was added to a solution of tropinone (3. m. After cooling to O".20. 88"/760 mm.) and the mixture heated at reflux temperature 1 hr. The pale yellowish-green solution was worked up according to the directions of Keagle and Hartung.. which is a monohydrate.6-168.05 g.84. for C1.p.5 g. From the ether solution was recovered tropinone (3. @-ketoglutaric acid (36..02.7 g. p.). The mixture gradually thickened and became dark red. 85"/1 mm. Found: C.p. and Fisher methylamine hydrochloride (C.). Syntheses. 143. A third experiment. The precipitated salts were immediately filtered off and washed with 50% aqueous methanol (150 mi. the first two cautiously at 0' and the other four at 15-20' during 1.9 g.5 g. 77. of nearly pure 2-carbomethoxytropinone and unreacted tropinone recovered as picrate (2. 4401 (1955). 0. 56.. Found: C. m. It was mixed at 0" with water (15 ml. 6th Ed.P. Wt.(64%. 7. The yellowish-brown crystals were then freed of remaining gummy by-products by sublimation in vacuo alto: 1. To the stirred.NOVEMBER 1957 2-C ARBOMETHOXYTROPINONE 1389 0-ketoglutaric acid (m. Emerman.1 ml.). for CzsHzgN7Oc 2CHaOH: C. Milwaukee. adding a little water.).250 mole).0 ml.6'.. and water (200 g.).00 mole). 1.3684 (reported:%' ny 1.).) a t 0" this gave the crude hydrate (3.S. The slightly alkaline yellow filtrates were extracted with ether (7 X 400 ml. The filtrate (-200 ml. At 0" the reaction between anhydrous citric acid and fuming sulfuric acid is slow even after half of the citric acid has been added. extracted into chloroform.) until nearly neutral and the barium salts collected and washed.) in dry xylene (7. HI 4.) waa substituted in the foregoing procedure. and the reaction mixture.8 g.5-Diethoxytetrahydrofuran (32 g.

and allowed to warm slowly.) The dialdehyde solution was mixed successively with methl 1amine hydrochloride (15. 4. (a) From B. Z O : CHsO. Found: C.): 56.. and washed with cold methanol: 52. potassium bicarbonate (35. 27.5. When about one fifth of the alkali had been added a granular precipitate began to s e p arate. 21. The powdered binoxalate (55. It appears not t o interfere appreciably with the subsequent condensation reaction. m.50. Brown. 48. or 80%). Another experiment differing only in that the alkali was admitted a t room temperature during 1 hr. 54. (32) S. dipotassium salt. Found: C.5 hr. filtered through Super-cel.) added. This was dissolved in methanol (50 ml.1390 FINDL.). The yield of 2.5-84' and giving some extinction under the polarizing microscope. 23. 172-173".)j and dimethyl P-ketoglutarate (34. Anal. adding about 10% of water.&dicar- . was kept 2 hr. for CI8H20N4012.. for C ~ Z H I ~ N O V ~ / ~ H54.09. 22 the yield reported there (85-90%) is calculated on the assumption that the anhydrous acid was used. 48. CHsO.06 after 60.5 ml. with chloroform (100 ml. Ber.) and mixed a t room temperature with powdered oxalic acid dihydrate (24. 644 (1956).HN0~: 45. : (31) C. 45.34.5 ml. 2596 (1925).5-151 (total yield: 55.51.) added cautiously.54.. and a yellow color appeared when the addition was about half complete.200 mole) in methanol (50 ml. no change occurred on longer keeping]. --+ -(-CH2--CHO)z Was04 + + + + 2HzNzOz Hyponitrous acid.2 g.p. C. The higher-melting binoxalate described above could not be thus obtained.0 g. 43. 0. 167-167. seeding.0 g.4Y VOL. Calcd. neutralization of only 10% of the methylamine hydrochloride with potassium hydroxide lowered the yield about 10%. and in alkaline solution a t any dilution some ester appeared always to be lost by hydrolysis. to a mixture of dimethyl b-ketoglutarate (43. 5. for ClrHleNOe: C.84. 23.. The base generated as described above melted as low as 85-88' and as high as 88-91" and was obtained in over-all yields of 69-79%. Org. P.1 g. The ethanol necessarily present did not seem to contaminate the product appreciably through ester exchange. 5.5 '.p.69. Crystallized by dissolving in the minimum quantity of warm acetone.87.. 0. and the mixture seeded and stirred 8 hr.30 a t the beginning. and J.69. Calcd.p. Anal. s diethoxytetrahydrofuran as described above and appropriately diluted was employed..300 mole) and methanol (10 ml. 4. at room temperature. voluminous precipitate.600 mole) in absolute methanol (150 ml.. Bailey.0 g. Powdered succindialdoxime32 (23. m. 0.00 1. which was augmented by storing 2 days a t O'. stout prisms. acquired a superficial pink coloration during drying. 'The picrate was prepared in aqueous acetone and purified from acetone: needles. and the base recovered approximately as before: 45.20 after 10 min.55. for C12H17S06. (33) The equation for this reaction appears to be as follolls: +CH2--CH=NOH)t 2NaN02 H#O. The light tan. H. Schroeter and L. €1. Anhydrous 2 .29. By using lower temperatures and greater dilutions in the experiments with free methylamine the tendency to by-product formation was somewhat reduced. In the first experiments. I n later experiments succindialdehyde generated from 2 . Found: K.p.6 g./t-Dzcarbomethoxytropinone (VI).1 g. Am.43. and is partially (but not completely) liquefied by storing in vacuo over potassium hydroxide. 0. It was dissolved in ether (100 ml.200 mole). Under similar conditions dimethyl citratemva' and fuming sulfuric acid gave dimethyl B-ketoglutarate directly in about 10% yield.2085 (1902).5'. The stirred. protected from air. is relatively high melting. Propertaes of 2. The crystalline binoxalate which slowly separated was stored overnight a t O'. which had become yellowish orange. seems to be largely amorphous. for ClQH17Pu'Os.55: CHIO. and not notably favored by low temperatures.75. methylamine hydrochloride was always neutralized by adding potassium bicarbonate. day a t O'. 4.6 g. slowly decomposes liberating nitrous oxide. and was probably a mixture of a t least two isomeric binoxalates. H. 150. Chem.222 mole).). (By treating an aliquot with a slight excess of aqueous p-nitrophenylhydrazine hydrochloride and recovering the p-nitrophenylhydra~one. 6. and storing in the cold. and a solution of sodium nitrite (27. 4. 54.. After storing overnight in the refrigerator. I. Conduct of the reaction a t room temperature indicated that the condensation is fairly rapid. The aqueous phase was extracted with more chloroform (3 X 100 ml. 5. Found: C. at room temperature. the salt was collected and washed with cold methanol (15 ml.d-dicarbomethoxylropinone.) and Super-cel. Found: C.2. 17 54.60. from another dialdehyde solution.82. a second crop was obtained: 3. Schmitz. Calcd. Calcd.160 mole) was stirred (30) I.6 g. 4. The nitrate was prepared from 3 8 nitric acid a t 0" and purified from methanol-ether and from acetone: minute. H. By diluting the mother liquor and washings with an equal volume of ether and storing N. J .. A. 0.5'. but the isolation became correspondingly more laborious.26. 17. of light orange oil.09 after 40. (75%). Findlav. 5. very mildly exothermic..28.). was collected and washed with ether: 39.*~3~ yield the of succindialdehyde was estimated to be 96-97%. water (2.).58.12 after 30. H. H. H. 6.) was added dropwise a t 0' during 35 min.p..). 35.1/2H20: C.).). and extracted with chloroform (5 X 200 ml. colorless prisms melting at 143-145" or a t 146' and remaining colorless. Chwn. similarly prepared. the anhydrous oil afforded a white product melting at 82.88. C. and the evolution of gas for a t least 24 hr.) and water (150 ml. H. polymeric by-products. H. Naw. 47.87. C.96. which is known to result from the reaction of nitrous acid and hydroxylamine. B.) [a small sample removed a t the beginning had a pH of -4. 3. R. for CL4H19N09: 48. Anal. Found: C.1/2H20:C.8 g. 4. 150. 4 dicarbomethoxytropinone was obtained only as an oil which was a t least moderately soluble in all common solvents.0 g. H.). 0.90 after 12 hr. Isolated as described above the hemihydrate has usually a tan coloration. Anal. H. 43. 85-88. 6.. 6..54. m. methanol (100 ml.70. On cooling the solution and scratching the binoxalate crystallized out: aggregates of minute. m. Racemic I-carbomethoxytropinone (I).). 54. Anal.2 g.5-151' (dec. polar and non-polar.2 g. afforded the same yield of salt of the same potassium content. J.8 g. I n the foregoing procedure. 48. Anal. SOC. 146-146.29. 5. A solution composed of equal parts of the hemihydrate and methanol was mixed with the theoretical amount of powdered oxalic acid dihydrate. -4fter 20 hr. and diluted to 1. was mixed with potassium bicarbonate (30.400 mole) in water (250 ml.' The mixture was kept 3 hours a t room temperature during which period it became thicker. 0. The combined extracts were dried over sodium sulfate and concentrated zn vacuo to a reddish orange oil (49. CH30.41. H. pale yellow mixture.5 g. Calcd..4 g. 0. Mimomethyl B-ketoglutarate. The intermediate binoxalate salt usually melted about 152" (sometimes a t 155"). indicates its presence in thip reaction too. Calcd. E. the solution. collected. Anal. Found: C.5 g. for CJL06K2: K.6ti.) was added dropwise a t 2-3" with mechanical stirring during 3-3..p.Ea stirred with BaC03 (5. was suspended in 1N sulfuric acid (410 ml. 5. and the mixture shaken.4dicarbomethoxytropinone in these experiments was comparatively low owing to the formation of large quantities of brown. Calcd. m. Kept in a closed jar the salt slowly turned orange and after a year contained 2'3" more potassium. 33. m. A solution of potassium hydroxide (33.300 mole).

67. and dried in vacuo over potassium hydroxide 2-3 hr.1 g. of granular.). and is especially soluble in hot methanol. Anal.p.) and saturated aqueous sodium sulfate (25 ml. Calcd. was suspended in water (50 ml. The crude product was collected.4-dicarbomethoxytropinone hemihydrate (21.5 ml. The base gives an intense color with ferric chloride. m. which. Calcd. . H. a t room temperature. the salt lost 8.18. Recovered as above described racemic 2-carbomethoxytropinone was obtained as a yellowish brown oil which.0 ml. small prisms slowly formed. Further elution afl'orded a p proximately the same amount melting. was colorless and melted a t 103. partly yellow and partly white.) the light orange mixture became turbid. It is unstable in hot water. after sublimation. pale yellow product. after two sublimations. of crude product. 48.) and acetone (4 ml.). H. the high-vacuum sublimation of the crude base gave frequently a mixture of solid and liquid. washed with benzene.8N) stirred in. This was purified by dissolving it in hot methyl acetate (30 ml.6 g. m. The yellow crude product from the partial saponification was chromatographcd on alumina.47. From the mother liquors rosettes of needles melting at 81. (64%). slowly crystallized as yellowish brown warts: 16. ( c ) From 8-ketoglutaric anhydride. Dried t o constant weight in vacuo at 40°..5'.0 g.5-86" separated.) and 6N aqueous sulfuric acid (-27.51.17. . a t 8 -15". and a proportionately greater quantity of resinous by-products was obtained.1 g.0800 mole) was dissolved a t room temperature with stirring in a solution (41 ml. 99. with potassium car- bonate (2. H.97. Anal. 93-100". By evaporation of an aqueous solution a t room temperature the salt was obtained in two forms: porous mushroom-like growths and colorless prisms. From concentrated aqueous solutions either form could be obtained by seeding.p. this mixture. the solution was concentrated in vacuo to a yellow oil which slowly solidified when seeded: -16 g. 61. The former melted about 87".) was suspended in a mixture of glacial acetic acid (60 ml. 167-168"). for ClaHl&Os.3 g. The binoxalate was prepared in and purified from water in which it is quite soluble even at room temperature. (86%).lz 8-Ketoglutaric acid (40 g. Similarly recrystallized from methyl acetate the dihydrate was obtained as colorless prisms melting at 93-97'.0. 7. Found: C. filtering off the potassium picrate.11.4". By evaporative distillation about 100°/l mm. Found: C. The particular phenomena observed appeared to be related to the history of the preparation: whether ammonia or potassium bicarbonate had been used to liberate the base and whether the chloroform extracts had been concentrated in vacuo or not. m. 7.93.p. From the 61trate a more finely divided precipitate separated: 3 g. Anhydrous racemic 2-carbomethoxytropinone (3.0 9. for CloH16N03. and acetone. m. Calcd.) and aqueous succindialdehyde (125 ml. Sublimed about 90°/1 mm. One part of diester dissolved in two parts of acetone and the stoichiometric amount of 4N aqueous sodium hydroxide gave a lower yield of the keto ester. the latter crystallizing only when scratched. Elution with benzene gave about one third which was colorless and.4 g. 0. and almost a t once the anhydride began to crystallize. m.4" from the upper walls of flasks containing the anhydrous base previously generated as noted above from pure salts. and keeping several hours at 0". H. After 24 hr. Furthermore. Found: C. 0.80. m.0 g. It was also obtained as prisms melting a t 104-107.0 g. 137. The initially obtained crude anhydrous base did not always solidify readily when seeded or scratched. After removing about 75% of the solvent from the dried (NazSOd) extracts on the steam bath.6". Calcd. 47.148 mole) and sodium hydroxide (4. (59%).3 g. washed with chloroform (35 ml. was obtained: 10. 8. Found: C.5-138. after sublimation. 51. 92-93". 6. 97-101".100 mole) in water (850 ml. m.5-98 ' Anal. 60. During the stirrirg (5 hr.) and extracted with chloroform (5 X 50 ml.: 30 g. (86%).5-104. were combined with methylamine approximately according to the directions of Willstatter. for CloH1&03: C.p. A crust of rosettes of colorless. By sublimation in w c u o this and the material from the mother liquors were freed of water and small amounts of resinous impurities: 11. basified with sodium hydroxide (20 ml. the former separating as a voluminous precipitate. 8. melted a t 90-98". and evolved carbon dioxide slowly. Small mounds or warts separated slowly from the cold solution.29. The mixture warmed almost imperceptibly. C. the yield of 2-carbomethoxytropinone was 45%.) were collected. and isolating the base as described above: 2. The analytically pure base was obtained by mixing the picrate (6.5' ( r e p ~ r t e d 138-140").) to 100 ml.21. Found: C. H.a Based either on succindialdeyhde or on the dipotassium salt. turned slowly yellow. The mixture was made basic with potassium bicarbonate (12 9.) and water (1 ml. which melted a t 93-96' and a t 101-104" after cooling and reheating.. a t room temperature the yellow solution was brought to pH 4. 102. When no sodium hydroxide was used to generate free methylamine.p.). ( b ) From the dipotassium salt of dimethyl &ketoglutarate. H. After several hours these (-1 8. H.) and the cooled solution poured into water (8. :~ Crude 0-ketoglutaric anhydride (13. when seeded.56. Another sample from aqueous acetone having the same composition melted a t 96-98". Anhydrous racemic Zcarbomethoxytropinone (4. Usually the acid dissolved in about an hour. m. (52%). When the reactants were combined near 0" and the mixture kept in the refrigerator 24 hr. 0.p. Calcd.3H~O:C. Properties of racemic 2-carbomethoxytropinone. and extracted with chloroform (9 X 100 ml. No tropinone was isolated unless excess base was used in the saponification.52.) and the solution kept 1 hr. the anhydrous base collects as white flakes and warts which electrify readily. The yellowish product was collected and washed with cold methyl acetate: 10.50% of its weight (theory for l l / p H20: 8. described below. After keeping 24 hr. 96-98. 0.15.p.3 g. 8. prepared as described above.00 wit.) added to the cooled solution. a solid.60%) and melted about 137" (bubbling and discoloration).. From the mother liquors additional material was recovered..42..). m. methyl acetate.) t o remove neutral and colored byproducts. 8. for CUH&O~: C. the yield was somewhat lower: 10.). a t 101-103".89. The color and about one fifth of the sample were ret8ained by the adsorbent.p.) at ca. The anhydrous base dissolves in all the common solvents.. H. Anal.) was dissolved in hot acetone (10 ml.9 g. 55. After keeping 48 hr. 54. 7. This material was recrystallized from acetone-water as described above: 12.5-105". of ca..) made by diluting sodium hydroxide (8. a t room temperature the crystals were washed with acetone and dried: 2. Finely divided 2. the yield of the keto ester was much smaller. Secondary butyl alcohol and methanol removed gummy material and a trace of benzene-insoluble substance. of 4N) and potassium bicarbonate (4 g.80. 97. 50. 0. H.2 g.p.h 6N mineral acid (effervescence).5-104.0 g.) was dissolved in hot absolute methanol (4.NOVEMBER 1957 2-CARBOMETHOXYTROPINONE 1391 bomethoxytropinone.13.1 g.). Anal.lHzO: C.96. H. of anhydrous base. which had become a yellowish white semi-solid..105 mole) was dissolved in cold methanol (50 ml.49.) and acetic anhydride (43 ml. It was added to a solution of methylamine hydrochloride (10. at room temperature. of aqueous solution.) added cautiously (effervescence!) t o pH 4.5 g. polar and non-polar..) a t 0". for C10HlbN0~2Hz0: 51. 8.. 5.0 g. The potassium salt and succindialdehyde. dissolves in aqueous alkali.49.) in water (50 ml. adding t o the cold solution water (4 ml. 10" and stirred 3 hrs. and is precipitated therefrom by either carbon dioxide or ammonium chloride.0 ml.

15 (1941).1." -15. and subliming the residue: colorless. Anal. 4.49.C4H~Oe: 48. and keeping 2 days a t 6". the base from one salt was indistinguishable from that from the other.0 g. Calcd. . H.) dissolved in methanol (1ml.73.) of the picrate melting a t 175176" separated overnight." -7. H.64 g.16. H.83.2H?O: C. powdery picrate. for C1fiH18S4010: 45.). 48.6-109. [a]'. CH30.1392 FINDLAY VOL. Anal. augmented by cooling to 0". a t 80". The mother liquors from the L-bitartrate were evaporated to dryness.25. for C10H15NOa 2Hz0: C. It was similarly dried for analysis.13. gives with this solvent supersaturated solutions from which both varieties of the picrate can be obtained. Anal. 4.07. These were converted to the anhydrous salt by dissolving in methanol and removing the solvent at room temperature.). The hydrated form was obtained by mixing with a slight excess of distilled water and drying in vacuo over anhydrous sodium sulfate: map.16. m. 4. ecgonine methyl ester. 44. 51. 165-167'. methanol).) added. The d-base was obtained by mixing the salt with a slight excess of potassium bicarbonate. Under the polarizing microscope it appeared either t.407'). Dried a t room temperature bot'h appeared to retain tenaciously a third of a mole of water per mole of salt. C. No 2-carbomethoxytropinone was obtained from pseudoecgonine methyl ester by the action of N-bromosuccinimide in chloro- + (34) R. Found: C. Mozingo. 6. rather stout prisms by evaporation at room temperature. H . 6. Found: C.p. Calcd. 1-(2-Carbomethoxytropinone). which is rather insoluble in hot methanol.7' (c. Found: C. 5.p. S . Calcd. Recrystallized from acetone in which it dissolves rather easily. Seeded with impure picrate recovered from another experiment. [a]'$ -36.08. the salt was obtained as a yellow powder melting a t 167. 8.7.p. water). 6. decomposed with ethereal acetone. [a]': -18. 8. water).43 g. 4. yellow prisms melt. Lithium alzirninum h.7 g. oxalate. H. Recrystallized from methanol the salt was obtained either as rhombohedra or as small.). 44. 13.).). 44. the weight loss was 9. Calcd. Calcd. The high-melting picrate.. To a solution of lithium aluminum hydride (0. This was combined with an equivalent amount of D-taitaric acid.46.8 g. 4. for C I ~ H ~ ~ X ~ O45. Dried a t $7" in high vacuum t o constant weight (5 hr. They are not completely dehydrated by heating in vacuo first a t 57" and then 12 hr.0.). Near the end o the addition a precipitate was obtained.07. extracting with chloroform.p. 142-143'. for C10H15N03.) in butanone ( 5 ml. transparent. 47. filtered.67: H. Rather large polyhedra (0. removing the solvent.98. d-(d-Carbomethoxytropinone). 108. It.93.79. methanol).) in methanol (-1 ml. 45. These were recrystallized to constant rotation from water: 4. The binoxahte was prepared in methanol and crystallized therefrom by adding ether: m.) added.p.ydride reduction. H.. C.) in ether (100 ml. Anal. A hydrated form (0. H. [a]'.. H.). 100-104".25: I . slowly and only partially solidified. Regenerated in the customary manner. To a solution of the hydrate (0. Calcd.) was dissolved in methanol ( 5 ml. 45.) in warm methanol ( 5 ml. 1 . 51. 9.5 g.46 g. 0. The aqueous extract was diluted with water (25 ml. for ClsH18N. water).) which could be distilled in uucuo alto.p.0 g. The lowermelting modification may be the same as that reported by Preobrashenski t o melt at 163-164'. A n a l . 22 The prisms melted about 98" and.07. 11.4' ( e .6". Anal. 0.57.4 g. m. 4. water).7 g.55.99.10. it is converted t.) was dissolved in butanone ( 5 ml.23. after the evaporation of the solvent.0" ( e . when partially dehydrated..6 g. and the resulting D-bitartrate purified as above noted for the L-bitartrate: white slender prisms.46 g.).95.32. 45. weight a t 125" in vacuo alto for analysis (weight loss: 1. 7. or binoxalate.). it. a solution of anhydrous racemic 2-carbomethoxytropinone (2.p. Calcd. 86-91'. -16.100 mole) was finely ground and dissolved in warm water (80-100 rn1. 7.29. did not give a crystalline hydrochloride.85.) and a solution of picric acid (0. for ClsH. Calcd. 145-146' (dec. 13.42.98.4" (c.3' (c.5-101. m.ed recrystallization from methanol the solid portion was obtained as a rather deep yellow substance which dissolved in methanol with difficulty but readily gave supersaturated solutions therein. m.25.41. of feathery needles of the bbitartrate separated: m.65. and basified with sodium hydroxide (6 g. Efforts to dehydrogenate ecgonine. for C1~H18N4010: 45.C4Hfi0fi: 48. 91-94' (bubbling). H .5' (dec. and L-tartaric acid (15. 3.67. 7. f After refluxing the mixture I hr. I t was necessary to dry both salts several hours in vacuo alto about 125" before analysis. 21. This form was recrystallized from absolute methanol in which i t is noticeably less soluble than the hydrate: rosettes of prisms (which were dried over CaC12). H..25. [ C Y ] ? f15. N. The anhydrous base (0. 2. The dried sample was analyzed. Anal. C. C. C. A solution of this form (0.43. Found: C. 176-177" (bubbling). 1. 2.p.58. and the base was regenerated in the usual manner. The sample was dried to constant. H. These were recrystallized from methanol: minute. 44. Neutralized with methanolic picric acid it was converted to an orange gum which. but little or no discoloration). again a t 145" (bubbling.0 g. gave the high-melting form (0. however. a light yellow granular precipitate was obtained: 0.3" (e. m. for C12H1&01. 60.14. H. for C10H1~N03: 60.7 The hydrochloride was prepared in methanol and purified from methanol-ether: m. C. On keeping a t room temperature or on seeding a large quantity.11. and pseudoecgonine methyl ester in alcoholic solution with freshly prepared Raney nickel3' did not succeed. 6. A n a l .80. 2. H.). which electrify readily.5". Found: C.100 mole). Recrystallized to constant melting point from acetone. slender prisms. HzO). Dehydrogenatzon experiments.5-168'.43 g.41. Found: C.95.16. 159.). to be indistinguishable from the first in crystalline form.) was added picric acid (0.p.89: H. 8. and extracted with 6N sulfuric acid (25 ml.8 g. Continuous extraction with chloroform removed an orange oil (1. H. Calcd. 48. H. Evidence of the transitory existence of a more dextrorotatory L-bitartrate was obtained. 12. A mixture of anhydrous racemic 2-carbomethoxytropinone (19. I O : T Found: C. It melted a t 206-207" and had a composition approximating that of the picrate of racemic anhydroecgonine methyl ester. Found: C. Anal. Found: C. for CIaH16NO3. after reffuxing 4 hr. 108.22. 4. The picrate of the 1-base separated suddenly from the supersaturated solution as balls of prisms: m. 4. j. By repeat. K.p.00. 4. Recrystallized from methanol the pure dihydrate was converted to stout prisms of the anhydrous a-bitartrate: m.79.9" (c. The base nTas generated in the manner already described: m. 176-176. [ a ] ? +18.14. m.60 g.) in methanol (50 ml. 1.ing a t 176-177 ". Found: C.21. The D-bitartrate prepared in water from stoichiometric amounts of the d-base and a-tartaric acid was obtained as colorless. Silver carbonate in boiling benzene had no effect on either of the foregoing esters. H. Org.o the lowmelting. mixed with Rochelle's salt (10 g.89% (theoretical loss for converting the dihydrate t o the anhydrous bitartrate: 9. was kept 2 days.497. Recrystallized to constant melting point from acetone this form gives another modification melting a t 174" (bubbling) which seems. H. Calcd.) and a solution of picric acid (0. 180' (reported? 180"). Anal.p.07. Syntheses.o be amorphous or to consist of aggregates of minute crystals.a:C.10. minute prisms.p.5.5'. 2.5-109. 51.BN&: C.) in dry ether (200 ml.5' (bubbling). 46. 6.) was added dropwise during 2 hr.08: CH30.

17 g. Found:C.05.N.20. Calcd. SOC.5-84' and at 88-91' had strong of acetic acid.17: i . extracting with ether (4 X 30 ml. Removal of the dropwise during 1hr.83 p and two strong bands a t 6. The writer is indebted t o the (200 ml.) was added hot ligroin (4 X 100 ml.).). and keto groupings indicates strongly that theBe two bands are concentrating the dried extracts on the water bath and then for the hydroxyl and the conjugated ester groups of the enol in vacuo. . After stirring another hour the acetone solution was filtered.11.5-diethoxytetracrystalline material consisting mainly of unchanged ester. H. Anal. limation it melted a t 118-119'.p.. 60-70". Found: C. and extracted with chloroform (3 X 50 d ) did not become noticeably darker. Calcd. 51.).09 p . The first third of the reagent produced an orange from methanol: -2 g.) which partially methyl ester (1. 255 mp (E 7882). The remainder caused a color change to greenish which retains some methanol. . Soc. W. There was little or no evidence of ketone dibenzoyl-&bitartrate was prepared in methanol and thrice absorption near 5. and extracted with chloroform (3 X 50 ml.0 g. methanol). 5. d-($'-Carhomethoxytropinone) from pseudoecgonine methyl then in vacuo. Kiliani's chromic acid solutionl6 (4.) were warmed and agitated until the mixture was homogeneous.61. showed noticeable Alloecgonine methyl ester. 2. N. The absence of any bands for unconjugated est. H. The base was recovered by adding aqueous po. treated with excess saturated ammo.) and treated with an acetoneIn Nujol samples of 2. methanol). turbidity. 17' (c. and the mix.99.. The keto ester (0. The binoxalate was prepared in and purified from methaalkaline with potassium bicarbonite. This was taken up in aqueous potassium weight of the unsublimed material to be 191.74 and 5. and 6.45 p.er and tassium carbonate. 1or more hr. It tially crystallized on keeping.).80 p. lnal.).H.Kuratani.). dissolved in ether (1 ml. 193-194'. Found: C.7 ml. 9. conhydrogen (at approximately atmospheric pressure).3 From the ether solution 0..125 g. 60. potassium acetate (29. 5.0 ml. 111.p. L. For the same carbonate.7%. 6.) in methanol trated aqueous solution of its potassium salt. cf.0 ml.09. was sublimed for analysis. 8. to a stirred solution of pseudoecgonine ligroin afforded an orange brown oil (11 9. and with a Carp Recording Spectrophotometer (Model 11). The jugated ester.bands a t 2.5-diethoxytetrahydrofuran(16. H. Calcd. 119was sublimed in vacuo alto for analysis: m. The hydroacetate which separated was re.33. (1949).) and 6.).80. and conjugated ester groupm.0 p region and gave two medium bands acetic acid (45 ml. H. ture extracted with chloroform (3 X 50 ml. unconjugated ester. 2. 1. gave a much 5.p. N. After 2 weeks it of pseudoecgonine methyl ester and benzophenone con.9 g. N. In tthe same solvent methyl the dibenzoyl-&bitartrate (m. 166. It was washed (350 ml.95 p.0 p. Found: The infrared measurements were made with a PerkinC. water (25 ml.105 g. droxyl.).. concen. 7.6.@ or by the then methyl acetoacetate (10. Purified from methanol it was obtained as minute. C. It is neutral and gives no ferric chloride test. 6. by the action of N-bromoacetamide. It waa filtered and con.R. 5.). Ecgonine methyl ester.02: C.80.5. for C16H21N06: 61. for CllH18N202. the + (35) R.7. tropinone had Xmax 255 p (E 6190) and 2. 60. and at ish gum. 60.5-167'. 1165 . Am.p.Jones and F.59. excess ammonium chloride added. The ings. and Carbide and Carbon Chemicals Company. N." was dia mixture of saturated aqueous potassium carbonate (3 ml. This material. it was form. stout prisms: 5. prepared from 2. -104') natural sources was dissolved in Mallinckrodt A. Calcd. Inasmuch as the base did not crystallize. b. Y Anal. ande 5.14.74 ~ . It was diluted with glacial acetic acid Acknowledgment. 159.H Sarett. The green salts were mixed with aqueous sodium sulfate. During the next two weeks the solution nium chloride.5-106. recrystallized therefrom.60.3682 (1954).. and extracted Sunlamp produced no Zcarbomethoxytropinone as indicated with chloroform (3 X 100 ml.6" (c. for CdI16NOa: C.64.85 p . emulsions being broken by centri.67. The solid material was filtered off and purified (250 ml.). 160') similarly obtained acetoacetate had Amax 240 mp (E 1513 after 5 min.material the methoxyl content was 15.78.76 M.) as described under "Tropinone.5'. N. Am. 76. 6.63. 7. H.03 p. In chloroform solution the sublimed material exhibited pronounced absorption a t of pseudoecgonine methyl ester was recovered. Crude d-(a-carbo. Elmer Recording Spectrophotometer (Model 21) having a Experiments with methyl acetoacetate.). [a]: -3. and concentrated in ester groupings.6. The mixture.07 g. Sci. 1667 after from the synthetic d-base.69.4-dicarbomethoxyThis melting point was not depressed by admixture with tropinone A . m. Anal.0 g.p.) added.5 g.5-160'. 25 (1951).within a few hours..). Anal.. After subLike the synthetic base.4-dicarbomethoxytropinone hemiether solution (1 ml. Drying and removal of the solvent in the usual manner centrated in vacuo to a mixture of water and oil which parafforded crude d-(2-~arbomethoxytropinone): 0. with platinum oxide (0. 5.31. mixed with methylamine hydrochloride and 3N potassium hydroxide (1. Chem. E. treated taining a small amount of glacial acetic acid with a G. initially Oppenauer oxidation. it gives an intense red color when treated with ferric chloride. The dried (NazS04) extracts were combined.).33.0 9.7 ml. Spertral measurements.22. catalyst-free solution was concentrated in vacuo to a yellowIn Nujol the dihydrate gave strong bands at 2.5-1113. J .. but not to constant melting point: I n absolute alcohol anhydrous racemic 2-carbomethoxysmall crystals having hexagonal faces.13. Chem. similarly oxidized.97 p characteristic of hy~ crystallized from acetone to constant melting point: 0.NOVEMBER 1957 Z-CARBOMETHOXYTROPINONE 1393 form.). 103. Kocher.75 g. respectively. of Found: C.4. and the mix.). water was added (9.71. Rept. The irradiation of a fused mixture yellow.0 ml. Technol. had become brown within 24 hr.95 p .p. (36) K. I n chloroform solution it had strong bands a t 3. (20%). (b) Succindialdehyde. Anal. H. gave a dark thick oil which was leached with ester. glacial absorption in the 3. and the residue recovered in the customary manner: -0. fuging. Tokyo. An ether solution of this mixture was extracted twice with hydrofuran (16 9.finally with methyl acetoacetate (10. 8. and ethylenic groupings. orange and the separation of a granular.4 g.was filtered.18 p characteristic of unconjugated ester. Univ. 9.p.p. J . and methoxytropinone) was thus obtained as a yellow. The Rast method indicated the molecular vacuo to a brown oil. m. H. respectively.E. concentrated in vacuo to a dark syrup.02 p . made slightly 13.7'. H.58. Calcd.). 63. and 6.74.p. Abstr.).83. m. ketone. with excess aqueous potassium carbonate. The initially pale green solution had become brown with ether (10 ml.3 g.). (a) Methylamine sodium chloride prism and the ultraviolet adsorption spectra hydrochloride (20. 60. 1354 (1952).. greenish deposit. H.nol: irregularly shaped crystals. 7.) dissolved in Fisher Reagent acetone crystallized. for CaH81NOll: C. m. I n chloroform solution racemic 2lower yield of the keto ester. Removal of the solvent from the dried (NaaS04)extracts on the steam bath and by a negative ferric chloride test.) luted to 125 ml. 7. 71.(6.00. for ClrHzoNzOa: 51.a t 5. 46.4. then with potassium bicarbonate (10.82. 62.) from carbomethoxytropinone (m.89. N.6 g. 6. Chem. 124'.) containing the stoichiometric amount hydrate melting at 82.54.04 p36 and ture was shaken 12 hr. C..27 p but none for saturated ketone or unconjugated made just alkaline with ammonia.

The object of the present studywas to determine whether such a ke€o Grignard reagent could be produced by the action of a dibromomagnesium compound on a hindered o-methoxyaryl ketone. 1967 SHACHAT2 1-Mesitoyl-2-methoxynaphthalene I ) reacts with the dimagnesium derivatives (11) of l. Similarly. and Mr. and 1. Speck.. The amount of reduction product decreases progressively as the chain of the Grignard reagent is lengthened.. Nenitzescu and I. respectively. determined the ultraviolet absorption spectra. a vinylog of an ndar bU""1. William C.4-dibromobutane gave 2-( 7-butenylpheny1)durylcarbinol (XII). and tropinone. [CONTRIBUTION FROM THE N O Y E S CHEMICAL LABORATORY. The terminal position of the double bond in the side chain of the carbinol was established by ozonization. Evelyn G. 19561957. Small. This unexpected product was formed presumably by way of the keto Grignard reagent I11 (n = 2 ) . U. Chem. MesC=O MesC=O Oxidation with dichromate in glacial acetic acid. . (2) Allied Chemical and Dye Corporation Fellow. J . which may undergo intramolecular reduction by way of a transition state resembling the quasi-cyclic system IV (n = 2) . 22 Charles Pfiser Company. 3483 (1950).5-diethoxytetrahydrofuran. Byron Raer of the Microanalytical Service Laboratory directed by Dr. idly and quantitatively. Harold K.1394 FUSON AND SHACHAT VOL. Am. Soc. Fuson and S. 14. In addition to formaldehyde.. Soc. Nes for helpful interpretations of the infrared absorption spectra. Intramolecular Reduction"' REYNOLD c. B. which was *This paper is a contribution in honor of Lyndon F. and Merck and Company for generous samples of 2. (1) This investigation was supported in part by a grant from the Office of Ordnance Research. 64.S. 3994 (1950). Army (Contract NO. n = 2). Alford. Am. La Combe.6-dibromohexane to yield products that could be derived from intermediate keto Grignard reagents (111). William R. J .n = 2) to give the olefinic carbinol V (n =2) in a yield of 65%. (5) R. former Editor of the Journal. I1 & MesCHOH s H ~ (cH~) I-Mesitoyl-2-methoxynaphthalene(I) has been found to react with the Grignard reagent from 1. the principal product is 2-( 7-buteny1)naphthyl-1-mesitylcarbinol(V. J .4V dibromobutane (11.4-butanedimagnesium dibromide. With 1.6 ihev intramolecular reduction of the keto hydrogenated catalytically to give 2-n-butyl-lGrignard reagent I11 (n = 2 ) probably occurs rap. Chem. Wright. duryl2-methoxyphenyl ketone (X) and the Grignard reagent from 1. & I MesC=O OCHs BrMgCHdCHmCHsMgBr .l-dibromobutane. a larger frag- . UNIVERSITY O F ILLINOIS] Keto Grignard Reagents. 4 Since n-butylmagnesium bromide displaces the methoxvl mouR of the methoxy ketone I in 55% yield. Mosher and E. S. Miller made the infrared measurements.5-dibromo( pentane. The condensation of esters with certain aliphatic bifunctional Grignard reagents t o give cyclic carb i n o l ~appears to involve the formation of keto ~ Grignard reagents as transient intermediates. and Mrs.. Miller and Dr. BETHESDA MD. C. for example. He is indebted to Mr. (3) C. Necsoiu.mesitoylnaphthalene (VII. n = 2). (3-ketoglutaric acid. D. Parisius. by preferential attack of the carbinol function. Mr. prc&cH2)ncH=CH. VI Mea \ VI1 & VI11 CH /o\ YH-OH (CHz). 72. Peake. Am. SOC. 72. FUSON AND NORMAN Received June 26. The microanalyses herein reported are principally by Miss Paula M. Chem. &(CH21n-CHzCHJ duced the olefinic ketone VI (n = 2). DA-11-022-ORD-874). Charles I. 1. 2446 (1942). (4) H. Mrs.

Sign up to vote on this title
UsefulNot useful