6212 resulting pentanoic acid-3,3-d2 was converted t o l-bromopentaneCondensation of bromoethane-2,2,2-d3 (5.6 g, 0.05 mol) with diethyl malonate, followed by hydrolysis and d e c a r b o ~ y l a t i o n , ~ ~ 3,3-d2 by the usual procedures. 25, 28 Conversion of the bromide t o the corresponding organocadmium gave 2.49 g ( 5 5 %) of butanoic acid-4,4,4-d3. reagent, followed by treatment with an equivalent amount of butyryl Treatment of the acid with thionyl chloride resulted in the formachloride, gave 4-nonanone-7,7-d*. Purification was accomplished tion of butyryl chloride-4,4,4-d3; a benzene solution of the crude by vapor phase chromatography. The resulting ketone was of acid chloride was added t o a solution containing an equivalent 98 dzisotopic purity. amount of dipentylcadmium. 2* The desired ketone, 4-nonanone4-Nonanone-3,3,5,5-d4 was prepared by the repeated equilibration l,l,l-d3 (97% d3),was isolated in high yield and purified further of the parent ketone with deuteriomethanol containing 10% heavy by repeated vapor phase chromatography on a 10-ft Carbowax 20M water and a catalytic amount of sodium deuterioxide. The product column operated at 170”. was recovered by removing the solvent under vacuum. The dis4-Nonanone-7,7-dz (IV). Propionic acid was reduced with lithcolored liquid was further purified by vapor phase chromatography ium aluminum deuteride in dry d i g l ~ m e . ~ 5The resulting propanolthrough a 10-ft Carbowax column (170”) preequilibrated with heavy l,l-& was converted t o the corresponding bromide by the procedure water.29 The resulting &ketone was 9 4 z pure. of Wiley, et a1.,26 and homologated by condensation with diethyl 1-Methylcyclobutanol and l-Methylcyclobutanol-1’,1’,1’-d3. The malonate, followed by hydrolysis and decarboxylation. 2 4 The preparation of these compounds has already been described.llJ3o (27) R. Adams and R. M. Kamm, “Organic Syntheses,” Coll. Vol. I, Wiley, New York, N. Y.,1932, p 250; E. B. Vliet, C. S. Marvel, and C. M . Hseuh, “Organic Syntheses,” Coll. Vol. 11, Wiley, New York, N. Y.,1943, p 416. (28) 3. Cason and F. S. Prout, “Organic Syntheses,” Coll. Vol. 111, Wiley, New York, N. Y., 1955, p 601.
(29) M. Senn, W. J. Richter, and A. L. Burlingame, J . Aifier. Client. W.J. Richter, M. Senn, and A. L. Burlingame, Terruhedrori Lett., 1235 (1965). (30) D.A. Semenow, E. F. Cox, and J. D. Roberts, J . Anier. Chem SOC.,78, 3221 (1956). SOC.,87, 680 (1965);
Heterocyclic Radical Ions. IV. Electron Spin Resonance Spectra of Conformationally Mobile Semifuraquinones’ Stephen F. Nelsen* and Errol1 D. Seppanen Contribution f r o m the Department of Chemistry, Unicersitj, of Wisconsiii, Madison, Wisconsin 53706. Received Jflniiflrj) 22, 1970 Abstract: The esr spectra of anions of dimethylmaleic anhydride, tetrahydrophthalic anhydride, 3,6-dihydrophthalic anhydride, and three methylated derivatives are reported and discussed. The methyl and vinyl splittings of 4-methyl-3,6-dihydrophthalicanhydride were about the same size, but it was concluded that the methyl splitting was not arising solely from r spin density in the 4,Svinyl A system.
he mechanisms involved in spin delocalization to protons y to spin-bearing p orbitals have been a subject of considerable discussion in recent years, especially since the discovery by the groups of RussellY and Stock4 that the esr splitting constants of such y protons are considerably enhanced in rigid, strained systems. One case of interest has been the interaction of a homoconjugated vinyl group with a spin-bearing A system, as that generalized in structure 1 below.b
T
X
1
h
2
CH2-CH2 (2a and 2b) than when X = CH, (3a and 3b) in seniiquinones and seniifuraquinones. Since the C2-CGdistance should be larger for 2 than for 3, we have argued that this is evidence against a mechanism i n which spin is transmitted by simple physical overlap of the label and vinyl A systems. Furthermore. the vinyl splitting of 2c is only 37% as large as that of 2a, although the Q spin density of seniidiones is greater than that of semifuraquinones. One possibility which has been ~ o n s i d e r e d l -for ~ the spin delocalization has been a homoallylic interaction of the sort depicted in la-b. If such delocalization or tautomerism were
3
The vinyl splitting is considerably enhanced when a bridging X group is present in the cyclohexadienyl system, but larger splittings are observed for X = (1) For Part I11 scc S. F. Nclscn nnd E. D. Seppcincn, J . Aotcr. Cheni. SOC.,89, 5740 (1967). (2) Alfrcd P. Slonn Fellow, 1968-1970. (3) G.A. Russcll, G.W. Holland, and I
