OF OLEFINIC ACIDS A New Synthesis of Unsaturated Acids. 11. a,p

Feb. 5, 1955. A NEW SYNTHESIS. OF OLEFINIC ACIDS. 60 1 furic acid and silica gel. ... 11. 111. R vious workers4 the acid having the @-substituent and...
0 downloads 0 Views 551KB Size
A NEWSYNTHESIS OF OLEFINIC ACIDS

Feb. 5, 1955

furic acid and silica gel. An orange crystalline solid formed (3.2 g., 76.3%, m.p. 70-75"). Recrystallization from a mixture of 30-60" and 60-90" ligroin (1:l) gave 2.6 g.

[COSTRIBUTION FROM

THE

60 1

(62%) of nearly white needles, m.p. 74-76' (1it.ll m.p. 76O), X%3'4.46 I.L (2242 cm.-l). AMHERST,MASS.

DEPARTMENT O F CHEMISTRY, UNIVERSITY

A New Synthesis of Unsaturated Acids.

OF

fiIABSACHUSETTS]

11. a,p-Olefinic Acids1V2

BY LOUISA. CARPINO RECEIVED SEPTEMBER 17, 1957 A new synthesis of or,P-olefinic acids is described which involves treatment of a 4-substituted-4-halo-2-pyrazolin-5-one with aqueous sodium hydroxide. While the yield of mixed acids is not exceedingly high (40-65%) the reaction promises t o be useful since the labile isomer of the ci5,trans pair predominates in the mixture. A simple method of separating the isomeric acids, fractional extraction with sodium bicarbonate solution, is described.

In the preceding paper it was shown that the the carboxyl group trans to one another is the more readily available 4,4-dihalo-2-pyrazolin-5-ones were stable isomer (IV) of a cis,trans pair. Such acids converted by an excess of aqueous alkali a t 0-5' to are easily obtainable by the Perkin and Refora,p-acetylenic acids in fair to good yields (75-90%). matsky reactions. On the other hand, it was found This reaction has now been extended to the corre- in the present case that a mixture of the &,trans sponding 4-substituted-4-halo compounds since i t pair was formed in which the labile isomer prewas expected that the process might lead to a con- dominated (see Table I). venient synthesis of a,@-olefinicacids and examinaTABLE I tion of a pair of isomeric 3,4-disubstituted-4-halo YIELDS OF LABILEAND STABLE ACIDSFROM HALOPYRAZOderivatives would yield evidence as to the posLONES (%)" sibility of cyclopropenone intermediates which was Acid Labile Stable previously considered. or-Phenylcinnamic 3 5 . 7 (44.6) 14.7 (17.4) I n order to determine whether cyclopropenones a-Phenylcrotonic 44.5 (55.5) 8 . 0 (10.6) intervened in the reaction, 3-methyl-4-phenyl-421.6 (28) 8 . 6 (11) chloro- and 3-phenyl-4-methyl-4-chloro-2-pyrazolin- a-Methylcinnamic 5-ones (IIIa and IIIb, respectively) were treated a The yield of crude material is given in parentheses. with an excess of sodium hydroxide. If cyclopropenones were involved, both IIIa and IIIb should In view of the ready availability5 of @-ketoesters yield the same unsaturated acid. The isolation of of all types it would seem that the present reaction IV-Va and IV-I'b, respectively, from these reac- represents a useful synthetic route to acids difficult Furthermore by isomtions disposes of the possibility of cyclopropenone to obtain by other intermediate^.^ As has been determined by pre- erization of the reaction mixture prior to isolation it should be possible to obtain solely the stable R' isomer although the Perkin and Reformatsky reactions would usually be more suitable providing the requisite aldehydes are available. The 4-haloH 2-pyrazolin-5-one precursors are available from the 11 I corresponding pyrazolidones8 as well as from the pyrazolones so that the reaction also can be used as a method of converting the stable isomer I V to

111

R

vious workers4the acid having the @-substituentand (1) Presented in part a t the 132nd National Meeting of t h e American Chemical Society, September 13, 1957, New York, N. Y. (2) Supported by the Office of Ordnance Research, U. S. Army. (3) Several self-evident rationalizations of this reaction are conceivable. Discussion is omitted a t the suggestion of the referee since no experimental evidence is available t o decide among various possibilities. (4) See L. Crombic, Quarf. Reos.. 6, 101 (1952), for a discussion and review.

(5) Recently several new convenient procedures for the synthesis of ,%keto esters have become available: (a) Ester acylations by means of sodium hydride, sodium amide and particularly diisopropylarninomagnesium bromide: E. E. Royals and D. G. Turpin, THISJ O U R N A L , 76, 5452 (1954): F. Y. Swamer and C. R. Hauser, { b i d , 7P, 1352 (1950); F, C. Frostick and C. R. Hauser, ibid., 71,1350 (1949); J. C. Shivers, M. L. Dillon and C. R . Hauser. ibid., 69, 119 (1947). (h) Condensation of an a-bromoester with an ordinary ester by means of magnesium: M. Montague, Burl. SGC. chim. France, 63 (1946) (c) The method of Blaise as modified by Cason, Rinehart and Thornton (ref. 20). (d) Acylation of ketone enamines: G. Stork, R. Terrell and J. Szmuszkovicz, THISJOURNAL, 76, 2029 (1954). (6) Previously two methods have been available for the synthesis of labile acids of type V. Isomerization of the stable acid (1V) by ultraviolet irradiation has been used, although the conversions generally are not high. For example, H. Burton and C. W. Shoppee [J. Chem. Soc.. 1156 (1935)l obtained a conversion of 8 % by a 70-hour irradiation of trans-a-methylcinnamic acid.7 Recently, a stereospecific method has become available [see e.g. (a) A. S. Dreiding and R . J. P r a t t , THISJOURNAL, 76, 1902 (1954), and (b) D. Y. Curtin and E. E. Harris, ibid., 73,2716, 4519 (1951)l. (7) In this paper c i s and trans refer t o the configurations of the parent a,B-unsaturated acids (c.g., V and I V , respectively). ( 8 ) Pyrazolidinone in Chem. A b s f r . terminology.

LOUISA. CARPINO

602 the labile form V.

For example, the methyl ester

of the stable a-phenylcinnamic acid upon treatment

with hydrazine hydrate yields the pyrazolidone VI which upon chlorination in nitromethane yields the pyrazolone I I I C . ~ Initially we experienced considerable difriculty in separating the isomeric &,trans-acids IV and V. Fractional crystallization (ligroin) was without effect. Even the 5 : l mixture IV-Va could not be separated by crystallization since the desired isomer was more soluble in all solvents tested. Previously the benzene-insoluble aniline saltlo was used to separate the labile a-phenylcinnamic acid from its stable isomer and this method was found to be a useful one for Vc, but IVc could not be obtained quantitatively by working up the filtrates. In the case of the a-phenylcrotonic acids (IV-Va) a modification of Sudborough's16partial esterification could be used but again IVa was not quantitatively recoverable as the ester. TABLE 11" CHARACTERISTIC

c=o

BANDS FOR

ASD

\.I1

1IN

\.I& i.1

\.I 11

IS

S

SI

r'

ExperimentalI3--E

\I1

4,5-Diphenyl-3-pyrazolidone (VI) .--..I sulutioii of X3.X

g.

o f methyl a-plien~-lcinnaniate'"ia u c l 6 g. of 11ydraziii~Iiydrate in 55 nil. of ethanol was refluxed for 2 1 hours :inti c:iutiously diluted with small amounts of water so as t o iiitluci. t'i,sition, t h e substance t o separate as a solid rather thnn a he'tv ~ ~ ~ I l l ~ J U U l l ~ ~ p (cm. -1) In all, about 200 nil. of water was added and the r i i i (K = CGII;,,R' IJ)' 3.80s (1724) stored in an ice-box 6-6 hours, filtered, \w5llCd well with 5.97s (1675) water and air-dried. T h e yield was 20.9 g., 1n.p. 116.125'. Recrystallization from about 50 ml. of be ( K = IJ!-O~SC&I~, I,vas 24 g . (4370), b.p. 113-116' (O.5mm.j. 3-Phenyl-4-methyl-4-chloro-2-pyrazolin-5-one (IIIbj -The ester I b was converted to the pyrazolone I I b by the method of v. -4uwers and Mauss?' which gave a 75y0yield of crude product, m.p. 212-217'. Xfter recrystallization from water 4 m e t h y l f o r m a m i d e (1:li the yield of long white needles, -1 m.p. 217.5-219.5', was 657, (lit.21m.p. 213-214.5'). suspension of 17 g . of the recrystallized pyrazolone I I b i n 50-60 ml. of nitroniethane was heated to boiling and chlorine was passed into the hot mixture until the solid dissolved and a definite excess of chlorine was present (15-20 min.). T h e mixture was heated occasionally on a hot-plate to keep the hnlopyrazolone from separating. After standing overnight in a n ice-box the solution deposited 14.5 g . of green(18) L. Edeleano, Be?., 20, 616 (1887). (19) M. Busch and B. Weiss, i b i d , , 33, 2701 (1900). (20) J. Cason, K. L. Rinehart a n d S. D. Thornton, J. O v g . Ciiem., 18, 1591 (1953); Org. Synlhcses, 35, 15 (1955). (21) I;. Y. h u w e r b and H . hIauss, J . pi.ni;l. L'lirrii , j21 110, 221 (1925). S o yield was reported.

603

yellow needles, m.p. 143-146'. A second recrystallization from nitromethane gave 13 g . of the pure substance, m.p. 144-146". T h e filtrates gave a n additional 2 g. of pure pprazolone so that the total yield was 15g. (70%). Methylene chloride (5 ml./g. of IIb) was also used as solvent for the chlorination. T h e analytical sample melted a t 145.5146.5" (nitromethane). Anal. Calcd. for CIoH&20Cl: C, 57.56; H , 4.35. Found: C, 57.20; H, 4.22. 3-Methyl-4-phenyl-4-chloro-2-pyrazolin-5-one (IIIa) .--A suspension of 18 g. of H a 2 ?in 100 ml. of methylene chloride was treated with a n excess of chlorine and the resulting solution evaporated either spontaneously or from a waterbath with the aid of a water aspirator. T h e residue was triturated with water, filtered a n d air-dried. T h e halopyrazolone (19 g., @%), m.p. 68-74', was recrystallized from benzene-60-90" ligroin (1:l) giving 15.5 g. (71.7y') of small white crystals; a second recrystallization gave 14.5 g. ( 6 i . l % ) , m.p. 81-82'. A n a l . Calcd. for C10H9K20Cl: C, 57.56; H , 4.33; C1, 16.99. Found: C, 57.40; H,4.43; C1, 16.80. Conversion of 3-Methyl-4-phenyl-4-chloro-2-pyrazolin-5one (IIIa) to cis- and trans-a-Phenylcrotonic Acids. Separation of the Isomeric Acids.--A solution of 2.0 g. (0.05 mole) of sodium hydroxide in 50 ml. of water was cooled in a n ice-bath a n d stirred mechanically. There mas then added during 1-2 min. 2.0 g. (0.01 mole) of finely powdered I I I a . T h e clear yellow solution soon became frothy due to gas evolution. Stirring was continued for two hours in the ice-bath and a n additional hour a t room temperature. T h e resulting slightly cloudy solution was decolorized with charcoal (J. T. Baker), cooled in a n ice-bath and acidified with concentrated hydrochloric acid t o congo red. T h e cream-orange solid which precipitated was extracted with four 25-ml. portions of ether. T h e combined etlicr extracts were washed three times with 5-ml. portions of water and fractional extraction with dilute sodium bicarbonate solution was begun. T h e first extraction was made with 10 ml. of 0.05 IM. sodium bicarbonate solution in order to remove traces of mineral acid and strongly acidic oily materials occasionally formed as by-products in the reaction. Fourteen additional extractions were then made with 10-nil. portions of 0.1 dl b i c a r b ~ n a t e . ~Each ~ fraction mas then acidified with 0.5 ml. of concentrated hydrochloric acid. Fraction

1 8 9

10-13 14 15

Result on acidification

No ppt. Altnost iiiiniediate cream-white p p t . Oil ppts. which solidifies in 3-4 niiiiutcs Same as no. 8 Almost imtiiedi:itc crem-\vliite p p t . s o ppt. s o ppt.

Based on the behavior on acidification, separatioii was judged t o be a t fractions 8 and 9 and these were therefore filtered a n d dried on a clay plate: 8 , a m t . 0.05 g., m.p. 79-81'; 9, a m t . 0.01 g., m.p. 85-110". Having established the separation point, fractions 2-7 were combined and the acid extracted with four 15-ml. portions of ether, the ether extracts washed once with 5 ml. of water, filtered by gravity to remove an>-trace of suspended inorganic material and allowed to evaporate. T h e residue of white-cream crystals m.p. 85-90'. Recrystallizaamounted to 0.9 g. (55.5y0,), tion from 90-120" ligroin gave 0.72 g . (44.5y6) of large creamy flakes, 1n.p. 94-96.5" (softening a t 93") . 2 4 (22) S. Veibel, K. Eggersen and S. C Linholt, Acfn C h e r n . S c a m . , 8 , 768 (1954). We obtained a yield of 81%; these authors report 307,. (23) T h e bicarbonate solutions were prepired b y delivering 0.50 and 1.00 ml. of a stock 1 M solution from a buret t o a graduated cylinder in which t h e volume was made u p t o 10 ml. After the separation of each bicarbonate extract 1&2 ml. of water was added a s a "chaser" t o remove t h e last portions o f each extract and t o a r o i d draining ether into t h e stopcock bore. Care was taken t o equilibrate t h e layers by shaking for about t w o minutes after gas evolution stopped. Fifteen extractions required about 1-1.5 hours for completion. (24) Comparison of this material by its infrared spectrum with t h e .inaiytical sample showed no difference. A strong band a t 14.83 u !,(j74.3cni. . - I ) , characteristic of the high melting isomer, ~ \ - i babbent.

11. K. S U M ~ ~ E RAND REL HANS L E. LUNK

604

Vol. SO

Three additional recrystallizations from 60-90" ligroin for the corresponding 3-methyl-4-phenyl derivative using gave a n analytical sample of pure cis-a-phenylcrotonic acid, 0.01 mole of the pyrazolone. However when fractions 12-14 in .p. 98.5-100 O . were reached i t appeared t h a t the calculated amount of acid A n a l . Calcd. for C L O H ~ O C, O ~74.05; : H, 6.22. Found: was no longer being extracted. Since the separation point had been passed, the extraction was completed with tlirce C, 74.25; H, 6.55. 10-ml. portions of 1 M bicarbonate. Evaporation of tile Fractions 10-13 were treated as indicated above for fracether extracts of fractions 2-8 gave 1.O g. (44.6Yc) of creamtions 2-7. Evaporation of the ether gave 0.17 g. (10.57,) white crystals, m.p. 123-131" (softening at 100"). Rcof cream colored solid, m.p. 120-131'. Recrystallization from 60-90" ligroin gave 0.13 g. (8.03%) of tiny cream- crystallization from nitromcthane gave 0.8 g. (35.7yc) of white crystals, m.p. 134-136.j3 (lit.25 m.p. 135'). The in- yellow-cream crystals, m.p. 137-139' (lit.*6 m.p. 136-137' Fraction 10 and the following were combined and work frared spectrum was identical with t h a t of a n authentic UP as usual giving 0.39 g. (17.47,) of cream-white crystals, sample of trans-a-phenylcrotonic acid and showed no signim.p. 160-167" (softening at 155'). Recrystallization froiii ficant contamination by the cis isomer by the absence of strong bands a t 12.80 p (751.3 cin.-lj and 11.70 (854.7 nitromethane gave 0.33 g. (14.770) of crcain colored xleedles, m.p. 173.5-175' (lit.26n1.p. 170-172'). cni.-l). Fractions 8 and 9 were mixtures containing mainly Conversion of 3-Phenyl-4-methyl-4-chloro-2-pyrazolin-5the cis- and trans- acids, respectively. Acids.-The initial In the present work we tentatively assign structure \'a to one to cis- and trans-~~Methylcinnamic reaction of 0.01 mole of the halopj-razolone with alkali was the compound m.p. 95.5-100" on the basis of method of carried out essentially as indicated for the 3-niethgl-4synthesis (cf. t h e other cases examined here wherein both phenyl derivative except that after acidification of the rclabile and stable acids have long been known), greater action mixture and extraction into ether the mixed acids acidity and lesser rate of esterification (see below) compared were then extracted together from the ether solution inti, with its isomer, m.p. 134-136.5". about six 20-ml. portions of 1 M sodium bicarbonate soluSeparation of the cis-acid was also possible by a modification of the method of Sudborough and Lloyd.'@ A solution tion. The bicarbonate extracts were then acidified and tlic of 0.58 g. (Byc)of a crude mixture of the cis and trans iso- mixed acids again taken into ether and the ether solution then fractionally extracted as: 1-5, 0.05 Af sodium bicarmers obtained from 0.98 g. of I I I a in 5 ml. of a 37, solution bonate; 6-10, 0.10 Ad sodium bicarbonatc; 11-15, 0.05 ilf of hydrogen chloride gas in commercial absolute methanol was allowcd t o stand a t room temperature for 1.75 hours and sodium bicarbonate; 16-19, 0.10 M sodium bicarbonate. Finally the extraction was completed with four 10-1111. then poured into water. Extraction of the precipitated material into ether followed by extraction of the ether solu- portions of 1 hl sodium bicarbonate solution. E\-aporatioii tion with several portions of 1 sodium bicarbonate solu- of the ether extracts of fractions 5-10 left a slightly o i l y 0.45 g. (2i.8yc) tion served to remove the cis- acid from the majority of the solid which was dried on a clay plate; trans-methyl ester. After acidification and isolation of the of m.p. 60-85". Recrystallization from ligroin (60-90') crude cis-acid (0.45 g . , m.p. 88-92') by means of ether the gave 0.35 g. (21 .6y0) of well-formed cream-colored crystals, A second recrystallization gave 0.3 g., m . p . crude solid was again added to 5 ml. of 3% hydrogen chlo- m.p. 86-91'. Evaporation of the ether exride in methanol and allowed t o stand for one hour at room 90-92" (lit.lo m.p. 91-92'). tracts of fractions 12-20 gave a cream-colored solid which temperature. IVorking up as before gave 0.4 g. (53.97,) was dried on a clay plate; amount 0.18 g. ( l l . 1 7 c ) , . m . p . of the cis-acid, m.p. 92-96' (softening a t 8 8 " ) . Recrys76-79.5" (softening a t 74'). Recrystallization from ligroin tallization from 90-120" ligroin gave 0.26 g. (35%) of snow(60-90") gave 0.14 g. (8.67,) of tiny cream-colored crystals, wliiteflakes, m.p. 94-98'. Conversion of 3,4-Diphenyl-4-chloro-2-pyrazolin-5-onern.1p. 79-80.5' (lit.10 rn.p. 81--82"). (IIIc) t o cis- and trans-a-Phenylcinnamic Acids.-The ini(28) AI. Bakunin, Cazr. chim. dfal., 27, 11, 48 (1807) tial reaction was carried out essentially as indicated above ( 2 5 ) H. Gilnian and S h Harris. ' T u i s

JOIJKNAI.,

63, 3541 (1931).

i h ~ € I m S rAIASS. ,

A Novel Elimination of Acetyl Chloride BY R. K. SUXMERBELL AND HANSE. LUNK RECEIVED .4UGCST 20, 1967 At temperatures below 200", 2-chlor0-3-acyl-~-dioxanes ( I ) decompose t o form acid chlorides and p-dioxanone. A mechanism is proposed for this reaction. The reaction of 2,3-dichloro-p-dioxane with anhydrous formic acid a t 170" serves as a convenient laboratory synthesis for p-dioxanone. 2-Acetoxy-3-ehlorotetrahpdropyran does not pyrolyze in a manner similar t o the 2-chloro-3-acyl-p-dioxanes, but eliminates acetic acid at about 200".

Discussion In the course of an investigation concerned with the stereochemistry of 2,3-disubstituted p-dioxanes, we were interested in obtaining examples of isomeric pairs of esters of p-dioxane-2,3-diol. Several methods of preparation of individual isorners of these compounds are described in the literature. Boeseken and co-workers' treated trans-2,3-dichloro-p-dioxane2 with potassium acetate or lead acetate in acetic acid and obtained a 2,3-diacetoxy+-dioxane, m.p. 104-105", of unknown configuration. Another method of preparation is described ( I ) j UoeseLen, I? Tellegen and P Cohen IIenllquez, 7 m s J O U R 66, 1285 (11133). (2) R K Summerbell and IInns E Lunk, r h d , 79, 480.2 (l'l57J

by Slagh, who treated various acids with t m m - 2 , S dichloro-p-dioxane in inert solvents such as tolucne and xylene at reflux temperature to prcparc the corresponding esters, and claims to have obtained a diacetate having 3 1n.p. of 79". Hoping that this compouiid would prove to be the second isomer we have repeated the of 2,3-diacetoxy-p-dioxane, work. However, the only compound obtained in several runs was the 2,3-diacetoxy-p-dioxane, 1ii.p. 104-105", identical with that obtained by the method of Boeseken.' The treatment of p-dioxenc: with hydrogen peroxide in t-butyl alcohol, followed by acetylation of the dihydroxy compound with .cetic anhydride and pyridine, resulted likewise iii

hA1

(3) ILiiold R Slagli, I, S Patent 2,164,355