NOTES
Feb., 1944 But even this material soon turned pink when expwed to a=. However, after several recrystsllmtions of the vacuumdistilled product from 70% alcohol the pyrrole was obtained in a form which remained colorless; m. p. 128-129O.5 (5) The melting point has been reported to be 129” (ref. 4).
COLLEGE OF PHARMACY UNIVERSITY OF MICHIGAN ANN ARBOR, MICHIGAN RECEIVED &3PTEMBER 30,1943
Reactions of &Butyl Cinnamate and &Butyl Benzoate with Phenylmagnesium Bromide’ BY FREDFROSTICK, ERWINBAUMGARTEN AND CHARLES R. HAWSER
305
in 160 ml. of eth?), contaiqed in a reaction h k equipped with a U-tube Lmmefsed m a dry ice-bath, was added during one-halfhour 0.3 mole of t-butyl benzoate’ in 100 ml. of ether. After standing overnight the reaction mixture was duxed for five hours. The liquid condensed in the U-tubedid not absorb an appreciable amount of bromine in carbon tetrachloride solution, indicating that no signidcant amount of isobutene was farmed. After daxnnposition of the n&ctloIl mixture with ammonium &laride, the ether solution was Wen with bicarbollpte (fromwhich was isolated a 10% yield of benzoic acid), dried and the solvent distilled. The residue was subjected to steam distillation until oily material (containing diphenyl but apparently no tbutylbenzene) ceased to pass over, and then re-crystaUized from alcohol, yielding triphenykarbinol (41%), m. p. 169-162”. In a similsr experiment, carried out with &butylbenzoate and methylmagnesium iodide, no isobutene appeared to be formed.
Kohler and Heritage’ showed that methyl (6) Nodm and Rigby. THIS JOURMAL, 64, 2097 (1932). cinnamate undergoes both 1,2- and 1,4ddition with phenylmagnesium bromide yielding a mix- DEPARTMENTOF CHEMISTRY 6.1943 ture of products. We have found that t-butyl D r w r U ~ ~ v ~ ~ s m RECEIVED DECEMBER cinnamate (in which the carbonyl group is rela- DURBAY.NORTHCAROLINA tively hindered) undergoes apparently only 1,4addition yielding t-butyl @,@-diphenylpropionate. Nature of Carotenes in Alfalfa -6MgBr BY A. R. KEXKERERAND G. S. Fruw
G”CH=CHCOOC(CHs),
(OH~)SCHCHICOOC(C&)~
To a stirred solution of phenylmagnesium bromide, prepared from 6.6 g. (0.23 mole) of magnesium and 46.1 g. (0.23 mole) of bromobenzene in 100 ml. of dry ether, was added during twenty minutes 23.5 g. (0.116 mole) of t butyl cinnamate.’ The reaction mixture d u x e d gently and a precipitate formed. After duxing for three hours, the muture was poured into an icecold saturated ammonium chloride solution. The ether layer was shaken with cold 10% sulfuric acid, and then with cold 6% po-um hydroxide solution. After drying with anhydrous wxhum sulfate followed by “Drierite,” the solvent was distilled, finally using the water aspirator. The solid residue (29.6p) was recrystallized from petroleum ether (b. p. 30-60 ), yielding. 16 g. (44%) of essent+lly pure.t-buty1 &&diphenylpropionate. Four recrystalhattons w e da product meltiug at 65.5-55.6”. Anal.‘ Calcd. for CI&~&: C, 80.86; H, 7.86. Found: C. 80.93;H,7.60. The product was further identified by hydrolysis in the presence of concentrated hydrochloric acid to fl,@-diphe-nylpropionic acid, which, after recrystallizationfrom acetonewater mixture, melted a t 152-154’ (reported, 166O)f
Analogous to methyl or ethyl benzoate, &butyl benzoate has been found to undergo the “normal” carbonyl addition reaction with phenylmagnesium bromide, yielding triphenylcarbinol. Some benzoic acid (which might have resulted from hydrolysis of the ester) was also isolated. Neither isobutene nor t-butylbenzene was found; these products would have resulted had the elimination or substitution reaction occurred.
According to Beadle and Zscheile,’ spinach and some other plants contain appreciable amounts of “neo-@-carotene,” stereoisomer of 8-carotene, which can be produced by heating @-carotenein petroleum ether. This pigment is the same as that termed pseudo-a-carotene by Gillam and El and neo-p-carotene B, by Polg6r and Zechmeister.8 Kemmerer and Fraps,‘ by chromatographic analysis with calcium hydroxide, found, besides @-caroteneand this “neo-@-carotene” in plants, another pigment which they termed “carotenoid X.” “Carotenoid X” did not possess vitamin A potency, while “neo-@-carotene” had one-half the potency of &carotene. PolgC and Zechmei~ter,~ by various treatments of @-carotene, secured about twelve neo-@-carotenes including neo-@carotenesB and U. Since the identification of the carotenes in plants is of both practical and scientiiic importance,both the “neo-0-carotene” and “carotenoid X”in alfalfa were studied.
Neo-B-carotene B was prepared by d d n g a solution of 20 mg. of crystalline carotene in 100 ml. of hexane for one hour.‘ The solution was chromatographed on calcium hydroxide and the neo-&carotene B was separated and extracted. Then it was purified by another chromatographic treatment. This pigment is just below the fl-carotene band in the chromatogram. Neo-&carotene U was prepared by dissolving about 10 mg. of crystalline car* tene in petroleum naphtha (Skellysolve P). adding a small crystal of iodine, and allowing the solution to stand an hour.’ The neo-&carotene U was separated and purified To a stirred solution of phenylmagnesium bromide (pre- by absorption on calcium hydroxide. The neo-&carotene pared from 0.5 mole each of magnesium and bromobeazene U is just above the ,%arotene in the chromatogram. “Carotenoid X” and “neo-~carotene”were prepared (1) This work was muppfnted by a grant from the Duke Univerfrom alfalfa leaf meal by extraction with alcholic potasCounal. aity R-ch sium hydroxide and petroleum naphtha (Skdyso!ve,F) (2) Kohlu and Heritage, A m C h . J., 88, 21 (IW5); see atro and chromatographic ScpSration on calcium hydroxide.
Allen and Blatt in ‘‘Organic Chemhry,” Gilman, Editar-in-Chid, John Wiley and So=, Inc., New York, N. Y.,1843, p. 681. (1) B. W. &uikandP. P.Zchdle,J. Bid. C h . ,144,21(1842). (8) Abramovitch, Shivers, Hudson and H a w , TEIS JOURNAL, (9 h E. Gilhm a d M.9. El RW,Biocksn. J., 80, 1735 (1986). 06, 986 (1943). (8) A. Pol& and L. ZLehrndSter, TEIS JOURNAL. 64,1856 (1942). (4) Analysis by T. S. Ma. Univemity of Chicago, Chicago,IUioOir (4) h R. Kemmaer and G. S. Fram I d . En#. C h . , Anal. Ed., (6) Baun~,Am. Chem. J . , 88, 84 (1905). 11,714 (1948).
300
Vul. (56
NOTES
Solutions of neo-@-caroteneU and “carotenoid X ’ were mixed and subjected to Chromatographic analysis on calcium hydroxide. There was no separation of bands, showing that the two pigments were identical. By a similar experiment neo-@-carotene B and “neo-&car+ tene” from alfalfa were found to be identical. Direct comparisonS of the absorption of light by the preparations were made by means of a quartz spectrograph. I n petroleum naphtha (Skellysolve F) solutions, photographed at dderent settings of the photometer, “carotenoid X ’ and neo-&carotene U both had absorption maxima at 468 and 443 mp; “necq3carotene” from alfalfa .and neo-#?-carotene B likewise had the same absorption curve with maxima at 473 and 446 mp. The pigments were also compared by dissolving approximately equal quantities in hexane, treating with iodine and analyzing the product by chromatographic separation on calcium hydroxide. The colorimetric ratio of the pigments formed from neo-@-caroteneU was, unknown pigment: neo-@-carotene U: &carotene: neo-#?-carotene E : neo-@-carotene E = 2 :22 :42:29:5 and from “carotenoid X” of alfalfa = 3:23:46:24:4. For n-8-carotene B the ratio was., 2:25:45:28:0andforneo-@carotenefromalfalIa 1 :22 :43:34:0. The ratios of stereoisomers produced by this treatment were nearly the same in the two comparisons, another evidence of the identity of the carotenes compared from alfalfa. The same carotenes have been found in the chromatographic analysis of other plants (spinach, collards, grasses, etc.). DIVISION OF CHEMISTRY TBXAS AGR~CWLTTJRAL EXPERIMENTSTATION COLLEGE STATION, TEXASRECEIVED NOVEMBER 22,1943
CNCH&OC(CHs)aCOOCHr I11
dl H
C(CN)C(CH3)2COOCH3 IV
In the present work it was demonstrated that IV is correct. This was done by showing that the hydrolysis product of the nitrile, previously considered to be formed through hydration of the enol of I11 and formulated as V, actually is VI. The hydrolysis product was identical with the acid obtained by adding hydrogen cyanide to VI1 and hydrolyzing the resulting cyanohydrin, procedures which can lead only to VI. OH CHs
OH CHs
HOOCA0/\O
VI1 2,4-Diketo-3,3-dimethyltetrahydrofuran*(VII), b. p. 200-210” (740mm.) or 103-107’ (16 mm.), was obtained The Action of Sodium Cyanide on Methyl in 86% yield when 25 g. of methyl y-acetoxy-P-keto-ap dimethylbutyrate’ was boiled for two hours with 5 g. of 10 7-Bromo-a,a-dimethylacetoacetate molar alcoholic hydrogen chloride (contrary to the statement of Conrad and Gast,E the acetoxy ester was unBY C. F. KOELSCH changed after it had been kept for three months; it also The reaction of sodium cyanide with an a- was undected when it was heated with potassium carbonhalogenated ketone usually leads to the formation ate). When 1.3 g. of VI1 was shaken for fifteen minutes with a solution of 0.6 g. of sodium cyanide and 0.8 ml. of of a @-ketonitrile; for example, bromopinacolone hydrochloric acid in 4 ml. of water, it furnished an oily yields pivaloylacetonitrile.’ A few a-haloketones, cyanohydrin. This was removed with ether and boiled however, furnish a,j3-oxidonitriles; for example, for one hour with 6 ml. of 20% hydrochloric acid; when desyl chloride ,yields a,p-epoxy-a,p-diphenyl- the solution was cooled, it deposited 0.5 g. of VI, colorless prisms from water, m. p. 213-217’ alone or mixed with the propionitrile.2 compound obtained from4 by the procedure of Lawrence. 3 On the basis of the polarizations indicated in When it was boiled with methanol and sulfuric acid, VI formulas I and 11, it was anticipated that methyl gave a methyl ester that melted at 104-105O alone or 7-bromo-a,a-dimethylacetoacetate (I) might not mixed with Lawrence’s ester.
Recognition of the correct structure (IV) for the nitrile previously considered to be I11 necessitates changes in the structures of substances derived from this compound. Simple esters, etc., need 0 CHa 0 CH3 not be specifically mentioned here, but the follow1I I It I BrCH-CcC-CHB BrCHt-C+CdCOOCH, ing revisions must be pointed out. “3-AceI I toxy-4,4-dimethyl-5-ketotetrahydro-furoicacid”4 CH, CHI is actually 3-acetoxy-4,4-dimethyl-5-ketotetaI I1 hydrofuran-3-carboxylic acid; “4,4-dimethyl-5The reaction had been studied long ag03*4.6 ketotetrahydrofuroic acid”4.6 is actually 4,4and the product had been assigned structure 111. dimethyl - 5 - ketotetrahydrofuran - 3 - carboxBut a survey of the old data indicated that the ylic acid; “ethyl a’-chloro-8-hydroxy-a,a-dimethproperties of the product were more in accord ylg1utarate”P is actually ethyl a-chloromethylwith structure IV. a-hydroxy-@,&dimethylsuccinate ; a,a-dimethyl glutaconic acid”6 is actually* a,a-dimethylita(1) Widman and Wahlherg. Bcr., 44, 2065 (1911). (2) Kohla and Brown, THISJOIJRNAG. 66, 4299 (1933). Other conic acid. Some doubt now attaches to the examples mag be found in the work of Justoni [Cas.. chim. ital:, structures formerly assigneda to homologs of I11 71, 41 (lQ.41). C k m . Abs., 86, 1016 (l942)l and of Delbaere [Bull. and V, vie., ethyl y-cyano-p-keto-a-ethyl-asoc. chim. &Is., 81, 1 (19421,CAam. Abs., ST, 5018 (194311. parallel bromopinacolone (11) in its behavior toward sodium cyanide, but might suffer attack a t its carbonyl group, leading to an oxide (IV).
“
(3) Lawrence, J . C k m . Soc., 76, 417 (1899). (4) Conrad and Gad, Ber., 8% la7 (1899). (5) Conrad, ibid.. 38, 1920 (1900)
(6) Conrad and Gast, abid.. 81, 2728 (1898). (7) Conrad and Kreichgauer, ibid.. 80, 867 (1897). (8) Cf.Perkin, J . Chcm. Soc., 81,249 (1902)
