MARCH
1962
1037
NOTES
stirred with an efficient mechanical stirrer and brought to the reflux temperature. Pinacolone (80 g., dried over magnesium sulfate and distilled) in 100 ml. of dimethoxyethane wm added from a dropping funnel over a 2-hr. period. Gas evolution (96% of the theoretical amount) ceased about 20 min. after addition waa completed. Concentrated hydrochloric acid (180 ml.) was added aa rapidly as possible to the mixture. Just before addition wm complete the BUSpension became clear and a fine white precipitate formed. The mixture was then cooled and poured into 2 1. of water. Five hundred milliliters of pentane waa added and the organic layer was separated and washed with five 500-ml. portions of water dried over magnesium sulfate and concentrated by distillation. The concentrate waa fractionated by distillation through a 10-in. Vigreux column at 36 mm. A forerun of 80 ml., b.p. 60-100°, was discarded. The product was collected at 100-102". Yield, 85 g., 63%. Recovery by this procedure is not maximum since both the forerun and the pot residue contain dipivaloylmethane.
anthracene as is shown by the exclusive addition of the vinyl double bond to form adduct I in high yields.
Vinylcyclohexane from 4-Vinylcyclohexene-1
EXPERIMENTAL
H
This is consistent with previously reported6 additions of 4-vinylcyclohexene-1 to butadiene and cyclopentadiene, in which the vinyl group, rather than the ring double bond, added to the diene. Hydrogenation of I followed by pyrolysis gave high yields of vinylcyclohexane and recovered anthracene, indicating a selective reduction of the CONTRIBUTION NO. 2763 FROM THEGATESAND CRELLINLABORATORIES OF CHEMISTRY "cyclohexene" double bond. Mass spectrometric CALIFORNIA INSTITUTE OF TECHNOLOGY and gas chromatographic analyses failed to show PASADENA, CALIF. impurities in the vinylcyclohexane, while the infrared and NMR spectra were those to be expected. CONTRIBUTION No. 1100 FROM THEAMESLABORATORY O F THE Specifically, it was shown that ethylidenecycloATOMICENERGY COMMISSON hexane, ethylcyclohexenes, and vinylcyclohexenes AbfES, IOWA were not present as impurities.
LYNNH. SLAWCIA A N D EUQENE F. MAGOON Received October 11, 1061
Because vinylcyclohexane has received considerable attention in recent polymerization studies,' we are prompted to report a new convenient synthesis of this monomer, which we believe offers several advantages over those previously rep~rted.~-~ Since 4-vinylcyclohexene-1 is readily obtainable from the thermal dimerization of l,&butadiene, it is a convenient starting material. It is difficult to hydrogenate 4-vinylcyclohexene-1 to vinylcyclohexaneb; however, it may be indirectly converted via the following three steps: (1) The Diels-Alder addition of the former to anthracene, (2) hydrogenation of the adduct, (3) pyrolysis of the hydrogenated adduct. There is a marked difference in reactivity of the two double bonds in 4-vinylcyclohexene-1 toward (1) See for example, T. W. Campbell and A. C. Haven, Jr., J. Appl. Poly. Sci., 1, 73 (1959); C. G. Overberger and J. E. Mulvaney, J . Am. Chem.Soc., 81,4697 (1959). (2) R. Ya. Levina and N. W. Meeentsova, Org. Khim., 7 , 241 (1950). (3) J. R. van der Bij and E. C. Kooyman, Rec. trau. chim., 71, 837 (1952). (4) N. A. Rozanov and Belikov, J . Russ. Phys. Chem. Soc., 61, 2309 (1929); Chem. Abstr., 24, 3766. (5) Direct hydrogenation of Pvinylcyclohexene-1 over a nickel catalyst gives ethylcyclohexene in good yields. 0. C. W. Allenby, U. S. Patent 2,576,743, November 27, 1!a61
-1--.
Diels-Alder addition of 4-vinz~lcyclohexene-1to anthracene. Anthracene (300 g., 1.7 moles), Pvinylcyclohexene-1 (1000 g., 9.2 moles), and toluene (800 ml.) were shaken in a 3-1. stainless steel autoclave a t 225' for 16 hr. The pressure during the experiment did not exceed 150 p.8.i.g. After removal of most of the excess 4vinylcyclohexene-1 and toluene via vacuum distillation the crystalline residue was filtered and washed with toluene, yielding 413 g. (85% yield basis anthracene used) of I. A second crop, 71 g. (14.6%) of less pure, semicrystalline product was obtained by removing the remaining solvent from the filtrate. Gas chromatographic analyses of the liquid removed by distillation indicated a 100% recovery of the unchanged 4vinylcyclohexenel, while ultraviolet analysis indicated < 0.1 yo unconverted anthracene in the solid adduct which melted a t 164-165" (uncorr.) (from benzene) and 176-178' (uncorr.) [from bis(2methoxyethyl) ether]. Anal. Calcd. for C Z ~ H ~ C,Z92.26; : H, 7.74; mol. wt., 286; bromine no., 56. Found: C, 92.39; H, 7.75; mol. wt. (ebul. benzene), 280; bromine No., 57. Only a slightly lower yield of pure adduct was obtained when the toluene was omitted in a similar experiment. Shorter reaction times gave incomplete conversions of anthracene. Hydrogenation of the anthracene adduct. Adduct I (150 g.) in isopropyl alcohol (300 ml.) with Raney nickel (-5 g.) was hydrogenated to completion within 1-2 hr. a t 125O, and 500 p.8.i.g. of hydrogen. The yield of white crystalline solid, separated from the catalyst by extraction with acetone, was 98%. Toluene, benzene, and bis(Zmethoxyethyl) ether were also good solvents for the hydrogenation. The product, m.p. 163" (uncorr.), was free of anthracene aa shown by ita ultraviolet spectrum. Anal. Calcd. for GPHZ.: C, 91.61; H, 8.39; mol. wt., 288. Found: C, 91.72; H, 8.34; mol. wt. (ebul. benzene), -280. Attempts to hydrogenate the molten adduct I, in the absence of a solvent, over copper-chromite catalyst a t 120180' were not successful. Pyrolysis of the hydrogenated adduct. Pyrolysis (330490') (6) K. Alder and H. F. Rickert, Ber., 71, 373 (1938).
1038
VOL.
NOTES
of a portion of the above hydrogenated adduct (24.3 9.) in a flask equipped with a heated 6-in. distillation head attached to a cold trap gave 9.1 g., vinylcyclohexane (98% yield). The ultraviolet spectrum of the solid (14.9 g., -100% recovery) remaining in the flask indicated it to be 104 f 5% anthracene melting a t 218" (m.p. of an authentic sample was 217-218'). The pyrolysis proceeds rather slowly below 300' but smoothly and rapidly a t temperatures 2 330". Distillation of the liquid product gave pure vinylcyclohexane; b.p. 125.7-126", 7 ~ 2 1.4458, 0 ~ di0 0.80178. Lit.,l b.p. 127", n Z o J1.4462, ~ d i 0 0.8012. Pyrolysis of semicrystalline addud I. The second crop of adduct I obtained above was pyrolyzed without hydrogenation to give a 7 0 4 0 % recovery of 4vinylcyclohexene-1 and 5 15% of c 0 - C ~ hydrocarbons. A near quantitative recovery of anthracene was obtained.
-
27
tulated from organomagnesium compounds for some time,4 none seems to have been reported for compounds containing active hydrogen. A possible equation for the reaction based on the evidence in Table I is as follows: 0
Acknowledgment. The authors wish to express their appreciation to P. A. Wadsworth, C. A. Reilly, and J. M. Gordon for mass spectrometric, KMR, ultraviolet, and infrared analyses. SHELLDEVELOPMENT Co.
EMERYVILLE, CALIF.
The Reaction between 2-Bromobenzamide and Benzylmagnesium Chloride' MARVINKORALAND ERNEST I. BECKER~ Received September 21, 1961
I n attempting to synthesize a-phenyl-2-bromoacetophenone by the reaction of excess benzylmagnesium chloride with 2-bromoben~arnide,~ it was surprisingly found that the product was a-phenylacetophenone (desoxybenzoin), the bromine having been removed. A study of the reaction was undertaken to determine conditions for removal of the halogen and whether it indeed could be retained. The most important variable appeared to be the ratio of benzylmagnesium choride to 2-bromobenzamide and, accordingly, a series of experiments was carried out in which this ratio was varied from 1: 1 to 13.8:1 (see Table I). The data show that with a ratio of Grignard to amide of from 1:1 to 3 : 1, the products are the starting amide, 2-bromobenzamide1 the corresponding acid, 2-bromobenzoic acid, or the magnesium salt of the latter. With a ratio of 4: 1 the desired product, cu-phenyl-2-bromoacetophenone, was obtained in good yield. However, with a large excess of Grignard dehalogenation took place. The dehalogenation is most readily understood as a metal-halogen exchange reaction. Although metal-halogen exchange reactions have been pos(1) Submitted to the Faculty of the Polytechnic Institute of Brooklyn in partial fulfillment of the requirements for the degree of doctor of philosophy, 1956. ( 2 ) To whom inquiries should be addressed. (3) For analogous reactions see E. I. Shapiro and E. I. Becker, J. A m . Chem. Soc., 75,4769 (1953).
TABLE I PRODUCTS O F THE REACTION BETWEEN 2-BROMOBENZAMIDE AND BENZYLMAQNESIUM CHLORIDE
Grignard Ratio Reagent, Grignard: Moles5 Amide 0.042 0.084
1:1 2 :1
0.125
3: 1
0.168
4: 1
0.368 0.580
9: 1 13.8:1
Yield, Products
%
2-Bromobenzamide PBromobenzamide 2-Bromobenzoic acid Magnesium 2-bromobenzoate 2-Bromobenzamide PBromobenzoic acid Magnesium 2-bromobenzoate a-Phenyl-2-bromoacetophenone a-Phenylacetophenone a-Phenylacetophenone
89.3 64.3 16.7 14.0 19.0 29.8 39.7 77.1 83.9 81.6
The yield of Grignard reagent waa not determined in each case, but estimated from the quantities of benzyl chloride taken. In standard runs 92 f 29" yield of bensylmagnesium chloride waa obtained. 5
It is of interest that 2-chlorobenzamide was not found to undergo any exchange with benzylmagnesium chloride.3 This is in accord with the facts known for metal-halogen exchange reactions with organolithium compounds,5 namely, that chlorine atoms are less reactive than bromine atoms. EXPERIMENTAL
A general procedure was followed. To the Grignard reagent prepared in ether from the stated quantities (see Table I) of benzyl chloride and sublimed magnesium was added with stirring the stated quantity of 2-bromobenzamide (m.p. 154.5-156.5O). Reflux with stirring was maintained for 2 (4) E. Urion, Compt. rend., 198, 1244 (1934).
(5) R. G. Jones, Org. Reactions, VI, 339 (1951).
