1212
ERNESTF. PRATT, RIPG. RICEAND RAYMOND W. LUCKEXBAUGH
n ater, and used for paper chromatography and radioactivity assay. For the large-scale production of the urinary metabolites (in non-radioactive form), three rats were placed in metabolism cages and injected at 48-hour intervals with 1 ml. of vitamin X acetate (2.5 mg.) emulsion. Crine mas collected under toluene and frozen daily. After 3 weeks, the collected urine was lyophilized and extracted with anhydrous butanol. T h e extract wa5 reduced in volume and applied t o the base-line of a 3-MM IVhatman chromatographic p'iper. The bands were developed with butanol (Table IV). The marginal portion of the sheets was cut off and sprayed with dinitrophenylhydrazine reagent to locate the band of the metabolite (Table V). This band was then cut out, the metabolite eluted with water, ether-soluble material removed by ether cxtraction and the dinitrophenylhydra-
[CONTRIBUTION FROM
THE
Vol. 79
zone derivative prepared. After extraction into toluene, it was crystallized from ethanol. An aliquot of a solution of the radioactive dinitrophenylhydrazone of JYS was added (15,868 counts per minute) prior to the first crystallization and recrystallization was carried out t o constant specific activity. About one-half of the activity added (8,625 c./min.) remained in the mother liquor. The reqt ma5 crystallized, yielding a dark red solid, m.p. 182-186O (specific activity, 55 c./min./mg.); yield 108.8 mg. Anal. Calcd. for C17H1807S4: C, 53.4; H , 4.6; K, 14.4. Found: C, 53.0; H,4.8; N, 14.5. Upon saponification of this dinitrophenylhydrazone derivative, a n acidic compound was obtained, soluble in alkali, insoluble in water, acid, ether and alcohol. URBANA, ILLISOIS
DEPARTMENT O F CHEMISTRY, UNIVERSITY O F MARYLASD]
Reactions of Naphthoquinones with Malonic Ester and its Analogs. 111. Phthaloyl- and Phthaloylbenzopyrrocolines' BY ERNEST F. PRATT, RIPG. RICE.4KD
RAYMOND
1-Substituted
w.LUCKENBAUGH~
RECEIVED AUGUST20, 1956 A variety of 1-substituted 2,3-phthaloylpyrrocolines(I1 to V I ) have been prepared by the extension to other active methylene compounds of the previously reported condensation of acetoacetic ester with pyridine and 2,3-dichloro-1,4-naphthoquinone. T h e carbethoxy-, acetyl- and cyanopyrrocolines have been interrelated by conversion t o the same acid or amide. Similarly substituted 2,3-phthaloyl-7,8-benzopgrrocolines(XI11 t o X X I ) have been smoothly prepared by using isoquinoline in place of pyridine. The pyrrocoline structures have been confirmed by practical new syntheses of 1-acetyl- and l-benzoyl2,3-phthaloylpyrrocolines (I1 and 111) from 2-acetonyl- or 2-phenacylpyridine and 2,3-dichloro-1,4-naphthoquinone.I n the same way l-acetyl-2,3-phthaloyl-5,6-benzopyrrocoline ( X X I I I ) has been obtained from 2-acetonylquinoline.
I n the preceding study of this series3 i t was The same by-product was isolated in the synthesis found that acetoacetic ester reacted with 2,3- of the cyanopyrrocoline from ethyl cyanoacetate dichloro- lJ4-naphthoquinone and pyridine to give and the dichloroquinone. a product for which the structure l-carbethoxyThe hydroquinone diacetates of the cyano- and 2,3-phthaloylpyrrocoline(I) was proposed. This phenylpyrrocolines (IV and V) were readily prereaction has now been extended to a number of pared. By the Rast method, using camphor, the analogs of acetoacetic ester. The active methyl- molecular weights of the acetylpyrrocoline (11) ene compounds employed, the formulas of the prod- and the hydroquinone diacetate of IV were found ucts and the yields were as follows: acetylace- to be 296 and 365 as compared to the calculated tone, 11, 5S%; benzoylacetone, 111, 10%; ethyl values of 289 and 358. The infrared absorption cyanoacetate, IV, 53% ; methyl cyanoacetate, IV, spectrum of the cyanopyrrocoline (IV) showed a 50%; phenylacetone, V, 17% ; benzyl ethyl ke- peak a t 2227 cm.-' indicating a cyano group contone, V, 1y0; and nitroethane, VI, 23%. Com- jugated with a carbon-carbon double bond.8,Y Upon treatment of this nitrile IV Kith sulfuric pounds 11,4 IV5 and V5 were also obtained, in somewhat lower yields, from unsubstituted 1,4- acid the amide VI1 was obtained. This crystallized as an allotropic form of the amide previously naphthoquinone. A 74% yield of the above cyanopyrrocoline (IV) prepared3 from the carbethoxypyrrocoline (I) resulted upon the treatment of ethyl (2-chloro-l,4- via the acid (VIII), and the same acid was obnaphthoquinonyl-3)-cyanoacetate6 with ~ y r i d i n e . ~tained from the acetyl analog I1 via the pyridin.I by-product formulated as X on the basis of the ium iodide (IX).'O Compounds which on the basis of their melting results of ultimate analysis and the formation of a precipitate with aqueous silver. nitrate was also points, elementary composition and color are idenisolated. Both this synthesis of the pyrrocoline tical with the carbethoxy- and acetylpyrrocolines and the formation of the by-product are consistent (I and 11) have been prepared in another laboratory with the route of reaction previously p r o p o ~ e d . ~from the dichloroquinone by essentially the same method, but the structures proposed in the initial (1) From t h e Ph.D. thesis of R . 1%'. Luckenbaugh. M a y , 1952, and independent reports were those of the ionic comRip G Rice t o be submitted in May, 1957. ( 2 ) Research Corporation Fellow 1950-1951 a n d Sational Institutes pounds I a and IIa." The results of quantitative < > f Health Fellow 1951-1952. analysis for hydrogen furnish significant support ( 3 ) E . IT. I'ratt, R . W. Luckenbaugh a n d R. L . Erickson, J . Org. for the pyrrocoline structure since they contain two ( ' h e m . , 1 9 , 178 (1954). (1) I t is a pleasure t o acknowledge our indebtedness t o Mr. Raymond L. Erickson, who carried o u t this experiment. ( 5 ) I t is a pleasure t o acknowledge our indebtedness t o Mr. Robert \\'. Storherr w h o carried o u t this experiment. (6) Prepared by t h e method of C. Liebermann, Ber., 32, 916 (1899). 17) Snhsequent t o the completion of this synthesis (ref. 12) closely related r e a c t i o n s were reported in ref. 11.
( 8 ) R. E. Kitson a n d X. E . Grilfith, A n a l . Chem.. 2 4 , 334 (1952). (9) W e wish t o t h a n k Dr. Robert Spurr and his staff for t h e determination of t h e infrared absorption spectra. (IO) L. King, THISJOCRNAL, 66, 894 (1944). (11) B. Suryanarayanaand B. D. Tilak, Current Sci. ( I n d i a ) ,22. 171 (1953): C. A , , 48, 14212 ( 1 9 5 4 ) ; P r o c . I n d i a n Acod. Sci., S9A, 185 (1954); C. d., 49, 12411 (1055).
1-SUBSTITUTED PHTHALOYLBENZOPYRROCOLINES
Ifarch 5 . 1957
I, R = COOCzHj 11, R = COCH3 111, R COC&, IV,R = C S V, R = CsHs VI, R = CHI VII, R = COSH2 VIII, R = COOH I X , R = COCH2X’CjHjIOH
A
,
Ia, R = COOC2Hb IIa, R = COCH3 Va, R = C6H5
OAc
‘:)3’:.c&5
I SI
QyyR 0 X X I I I , R = COCHa XXIV, R = COOH XXV, R = H
I
OAc
R R R R R R XIX, R XX, R XXI, R
XIII, XIS‘, XI’, XVI. XLrIIl X\‘III;
0,
3
= = = = = =
COOCzHs COOCH, COCH, COCrHh . CN
CsH5 CH3 = H = C2H5
XXII
less hydrogen atoms and the total hydrogen content is unusually low. The products of reductive acetylation are readily formulated as diacetates of the pyrrocoline h y d r o q u i n ~ n e s ,but ~ ~ ~a~ reasonable formulation on the basis of the ionic structures is often impossible; on this basis, for example, the product from the phenyl analog Va would be X I which contains a free carbanion. The ionic formuIations recently have been withdrawn in favor of our pyrrocoline structures primarily on the basis that the product of decarboxylation of the acid (VIIT) was identical with that isolated in about lyOyield from the reaction of 2,3-dichloro1,4-naphthoquinone with a large excess of a-picolineal3 Further confirmation of the pyrrocoline structures has now been obtained by preparation of the acetyl and benzoyl analogs (I1 and 111) from 2(12) R. W. Luckenbaugh, Ph.D. Thesis, Univ. of Maryland, May, 1952, pp. 34-44. (13) R. V. Acharya, B. Suryanarayana and B. D. Tilak, J . Sci. Indust. Res. ( I n d i a ) , 14B, 394 (1955): C. A . , SO, 12971 (1956). The preparation of 111 from acetophenone or dibenwylmethane is also described in this paper.
1213
acetonyl- and 2-phenacylpyridine with 2,3-dichloro-1,4-naphthoquinone. The results of melting point and mixture melting point determinations showed that identical products were obtained by the two routes. Since the yields were 4Oy0or better the alternative route is of preparative value. It is suggested that an intermediate such as XI1 is formed by a route closely analogous to that proposed for the first steps of the earlier synthesis3; loss of the highly activated hydrogen as a proton followed by electron shifts as indicated in XI1 would give the acetylpyrrocoline (11). When isoquinoline is used in place of pyridine together with 2,3-dichloro-1,4-naphthoquinoneand an active methylene compound, the corresponding 1-substituted 2,3-phthaloyl-7,8-benzopyrrocolines are obtained.14 The yields are in general superior to those obtained with pyridine and are especially satisfactory when it is considered that a t least five successive reactions appear to be i n ~ o l v e d . ~ The active methylene compounds used, the formulas of the products and the yields were as follows: ethyl acetoacetate (XIII) 64%; ethyl benzoylacetate (XIII), 61%; methyl acetoacetate (XIV), 64% ; acetylacetone (XV), 727, ; benzoylacetone (XVI), 43%; ethyl cyanoacetate (XVII), 80%; methyl cyanoacetate (XVII), 72yc; benzoylacetonitrile (XVTI), 63% ; phenylacetone (XVIII), 19% ; nitroethane (XIX), 26% ; and nitromethane (XX)?9%. The results of mixture melting point determinations showed that the two carboalkoxyl analogs X I I I and XIV were identical with those previously obtained from coal tar “quinoline” and which had been, therefore, formulated as 5,6-benzopyrrocolines3 Since i t has now been found that synthetic quinoline gives no crystalline products under these conditions while synthetic isoquinoline does give the products in good yield, it is apparent that the isoquinoline impurity in the large excess of coal tar quinoline is the actual reactant and that the products should, therefore, be formulated as 1carboalkoxy - 7,8 - benzopyrrocolines (XIII and XIV). Most of the isoquinoline derivatives (XIII through XX) were also obtained in fair yields from coal tar “quinoline” and the reaction may well prove useful for removing isoquinoline from the coal tar product. With 1-nitropropane and coal tar “quinoline” a 29y0 yield of the ethylbenzopyrrocoline (XXI) was obtained while with isoquinoline a difficultly separable mixture of this product and the ionic by-product X X I I apparently resulted. The only crystalline product isolated from the attempted reaction of dimethyl malonate, ethyl benzyl ketone or acetonitrile with the dichloroquinone and isoquinoline had the carbon, hydrogen and nitrogen content of the ionic byproduct XXII, and the same compound was made directly from isoquinoline and the dichloroquinone in 77y0 yield. The pyridine analog is well kn0wn.3.1~ The parent 2,3-phthaloyl-7,8-benzopyrrocoline (14) The ordinarily safe assumption that the reaction occurs at the 1 and 2 rather than the 2 and 3 positions of the isoquinoline is made here. (15) F. Ullman and M. Ettisch, Ber., 84, 259 (1921).
1214
ERNESTF. PRATT,RIP G. RICEAND RAYMOKD IT.LCCKEKR.
