1018
COMMUMCATIONS TO THE EDITOR
case, i t may cyclize in two possible manners to form methylvinylcyclobutanols [111] as well as methylcyclohexenol [IV] according to the following scheme: When a solution of 6-hepten-2-one in pentane was irradiated with a low pressure mercury resonance quartz lamp,6 acetone (32%), butadiene (23%) and an alcohol fraction (7.5%) isomeric with the starting ketone (found: C, 74.63; C, 10.48) were formed among some unreacted ketone (20%) and high molecular weight material. T h e major alcohol (6yG), separated by gas chromatography, was identified to be l-methyl2-vinylcyclobutanol [111] (found: C, 74.88; H, 10.77; 13% 1.4598; T m a x 3320, 3060, 1635, 990 and 910 cm.-I). The n.m.r. spectrum exhibits a singlet a t 8.85 r (- CHJ, a multiplet at 7.90-8.70 T (-CH2), a quartet a t 7.13 r (allylic), a singlet a t 6.36 7 (-OH), a triplet a t 5.52, 5.10 and 4.93 r (terminal vinyl) and a multiplet a t 3.504.45 T (non-terminal vinyl) and the relative intensities of these peaks are 3 : 4: 1: 1 : 2 : 1, respectively. Its double bond region in the n.m.r. spectrum was essentially identical with t h a t of a known allyl system.' T h e other alcohol component ( I .4y0)had been identified tentatively to be 1-methyl-3-cyclohexenol [IV], Y~~~ 3320 and 3005 cm.-l; 8.80 T (singlet, -CHJ and 4.42 T (singlet, vinylene). Upon hydrogenation the alcohol fraction absorbed one mole equivalent of hydrogen, and the hydrogenated mixture was separated by gas chromatography into 1-methyl-2-ethylcyclobutano18and 1methylcyclohexanol in 4 : 1 ratio, identical in all respects with the corresponding authentic samples. The isolation of both 111 and IV in the present investigation conclusively substantiates the step-wise mechanism.' It also indicates that the reactive state responsible for this reaction may be a triplet since the intermediate radical was free to delocalize over an allylic system.g Acknowledgment.--A part of this work was carried out in the Department of Chemistry a t the University of Illinois where N.C.Y. was a visiting lecturer, during the summer of 1962. The hospitality and stimulation provided by various members of t h a t Department, particularly by Professor Stanley Smith, are gratefully acknowledged. This work is supported by a grant from the Petroleum Research Fund, Grant No. 726.
Yol. 85
tr~meter.~ The detection of free radicals and other products is carried out b y ionization of a portion of the product stream with electrons of low energy, such t h a t only parent ions are formed. The thermal decomposition of 1,2-diiodobenzene in this reactor a t 960' (residence time sec., pressure -lo-? mm.) leads to the formation of the following products: iodine atoms, iodophenyl radicals, phenyl iodide, phenyl radicals, benzene, a product of parent mass 76 and a product of parent mass 152. The reactions giving rise to these products are thought to be
2
0
1
-
78 152
These reactions are similar to those reported in the photolysis of 1,2-diiodobenzene in ~ o l u t i o n . ~A comparison of the ionization efficiency curves for benzene, the product of mass 76 and an added xenon standard leads to the vertical ionization potentials : benzene, 9.50 v . ; mass 76, 9.75 v . Identification of this species of mass 76 as benzyne by its parent mass alone is not sufficient, since three other structures for C6H4 hydrocarbons can be written: HC=CCH=CHC=CH (I), HsC=C=C=C=C=CH2 (11) and H,C=C=-C= CHCECH (111). Formation of any of these three species from 1,%diiodobenzene appears quite improbable. Formation of I would require one H-atom to migrate to a next-but-one carbon, I1 and I11 require the migration of two H-atoms. Moreover, the identity of the product of mass 76 with any of these can be ruled out on the basis of the expected ionization potentials. I1 and 111, being substituted butatrienes, will have ionization potentials less than that of butatriene itself, 9.28 v.j The ionization potential of I can be estimated sufficiently closely for the present purposes by comparing the changes in ionization potential along the two [ f i ) If.S . Kharasch a n d H S Friedlander, J . Oyg, C h e m . , 14, 21; (1949). series : ethylene, propylene, %butene ; and ethylene, ( 7 ) Varian S . m . r Spectra CataloR, spectrum S o 13ii. ( 8 ) Prepared by t h e photochemical reaction of 2 - h e p t a n a n e , I). H . Y a n g , vinylacetylene and I. Substitution of H in ethylene unpublished result. by -C=CH decreases the ionization potential slightly ( 9 ) G . S . H a m m o n d , Abstracts of P a p e r s , 17th S a t i o n a i Organic Chemisless (0.72 v.) than substitution by -CH3 (0.78 v . ) . ~ t r y Symposium o f t h e American Chemical Society, 1901, p p , 48-58. Comparison with 2-butene, derived from ethylene by a (10) Fellow of t h e Alfred P . Sloan Foundation. (11) C. S. Public Hzalth Service Predoctoral Fellow. 1,2-disubstitution, gives the ionization potential of I DEPARTMEST OF CHEMISTRY S . C. 'I-AsG'O to be 9.40 v. The observed value for mass 76 is clearly THEUNIVERSITY OF CHICAGO .I.MORDUCHOWITZ"larger than expected values for I, I1 or 111. CHICAGO 3i, ILLISOIS DISG-DJUNG H . YANG The thermal decompositions of 1,4- and 1,3-diiodoRECEIVED FEBRUARY 7 , 1963 benzene under the same conditions were also examined. The former produced iodophenyl radicals, benzene, and IONIZATION POTENTIAL OF BENZYNE again a compound of mass f G , b u t no product of mass Sir : 132. The ionization potential of the species of mass 76 The chemistry of benzyne (1 ,z-dehydrobenzene) in in this case, however, was 9.46 v . , in good agreement solution has received considerable attention, and has with the estimated value for I. The course of this been reviewed by a number of authors.' Recently, the reaction is evidently transient spectrum of an intermediate has been observed in the flash photolysis of o-iodophenylmercuric iodide and other precursors, which is almost certainly attributable to benzyne. V-e wish to report the de1 1 tection of benzyne in the thermal decomposition of 1,2diiodobenzene in a reactor coupled to a mass spec(3) J . B F a r m e r a o d F P Lossing, ( ' n i t J . C k e m , 33, 8ii1 (195.5); F. P. (1) G. Wittig, Angew. C h e w . , 6 9 , 213 ( 1 9 5 7 ) ; J. 11. R o b e r t s , Chem. S O C . (1,tindon) Sper. Prtbi., 12, 11.5 (1'3,761, R. Huisgen, in "Organometallic Chemistry." H . H. Zeisi, E d . , Reinhold Publishing Co , S e w York, N.Y.. 1960, H. Heaney, Chem. R e a , 62, 81 (1962). ( 2 ) K . S. Berry, G. S . Spokes a n d R X I . Stiles, J . A m C h e m Soc., 82, ,5240 ( I H i i f l j , 84, :3;170 (1962).
Lossing, P. Kebarle a n d J. B . de S I u s a in "Advances in hfass Spectrometry," Pergamon Press, I.undun, 1'3.59, p 4,31, t1) J . A . Kampmeier a n d IC. Hoffmeister, J A m C h c m . Sur , 84, 3767 (19tiZ). ( 5 ) A , Streitwieser, J r . , ibid . 82, 412d (19iiO). (6) R . E . Htinig, J Chenz. P h j s , 16, 10.5 (1946).
COMMUNICATIOSS TO THE EDITOR
.2pril 5 , lXi3
The decomposition of 1 ,:
