The formation of the a-substituted radicals 1-9 with the 4pyridyl (?), l-naphthyl(8), and the 2-naphthyl(9) substituent by thermolysis of their dimers 1-3 in solution was investigated. The activation parameters of the homolysis of the central CC bond were obtained from kinetic measurements of the thermal cleavage of 1-3 in the presence of radical scavengers. These data were compared with those obtained

Thermolabile Hydrocarbons, 34 [I]. -Thermal Stability of 9,9' -Bixanthene and 9,9', 1 0,lO' -Tetrahydro-1 0 , 1 0 ,lo ' , 1 0' -tetra- methyl-g,g'-bianthracene. -Stabilisation Energies of Planar Benzhydryl RadicaldZlThe thermal cleavage of the central CC bonds in the title ASDiss, as expected. The radical stabilisation enthalpies RSE compounds was investigated. The activation parameters obtained from these data (RSE: ?a = 17.9 k 1.3; ?b = 15.6 t were obtained from kinetics and the equilibrium enthalpies 1.4 kcal/mol) correspond closely to twice the RSE of a-phenyand entropies from the temperature dependence of the equilethyl radicals (8.0 k 1.5 kcal/mol) determined previously. librium radical concentrations, determined by quantitative The smaller RSE of the benzhydryl radical (12.9 t 0.9 kcall ESR. AH* and AHD,,, are nearly identical but not AS* and mol), therefore, must be due to its nonplanarity.

The title radicals were generated by thermal decomposition from the enthalpies of combustion and enthalpies of vaporiof (E,E)-meso-and (E,E)-(f)-5,6-di-tert-butyl-2,2,9,9-tetra-zation, those of 6 and 7 by MM2 calculation. From these data methyl-3,?-decadiene (4a, b), (E)-5-( 1,l-dimethylethy1)-and their comparison with those of comparable saturated hy-2,2,6-trimethyl-6-phenyl-3-heptene (6), and (E)-5-(diphenyl-drocarbons of similar strain the radical stabilization enthalpy methyl)-2,2,6,6-tetramethyl-3-heptene (7). The activation of the title radical RSE = 11.4 f 1.5 kcal/mol was determined, parameters of these reactions were calculated from the rate which is in good agreement with results obtained from other constants which were followed over a 40°C temperature model compounds (12.6 kcal/mol) previously. range. The enthalpies of formation of 4a, b were determinedIn a series of papers[la1 we have shown that CC bond energies may be obtained from a kinetic study of the homolytic decomposition of hydrocarbons having one CC bond which is particularly weak for steric or electronic reasonst21 and which is, therefore, selectively cleaved at elevated temperature in a homolytic fashion into two radicals. The activation enthalpies A P of these thermolysis reactions are equivalent to the bond dissociation enthalpies BDE. It has been found that there exists a linear relationship between the BDE and the change in strain Ds during the dissociation processL21 as long as CC bonds of the same type are compared, e.g. C,C, or C,C, bonds, i.e. bonds between tertiary or quaternary carbon atoms, respectively. When substituents are attached to these CC bonds which are capable of stabilizing radical centers by conjugation, radical stabilization energies RSE can be calculated from the decrease in BDE resulting from the replacement of a methyl group by the substituent S3] at the CC bond which is cleaved during thermolysis[21. Recently, we have determined the RSE of a-vinylalkyl radicals, i.e. allyl radicals, in this way by following the kinetics of decomposition of mesoand DL-3,4-di-1 -cyclohexen-1 -yl-2,2,5,5-tetramethylhexane (1) into 1-(1-cyclohexen-l-yl)-2,2-dimethylpropyl radicals (2)L4]. The analysis of the kinetic data has been hampered, however, by a competing Cope rearrangement of 1 to 3, which has required a detailed study of the products and rates for obtaining the activation enthalpies of the homolytic process 1 -2r41.From this analysis, finally, a RSE of 12.6 k 1.5 kcaVmo1 has been obtained for the allylic radicals 2 in reasonable Lo] Part 33: agreement with the allylic stabilization energies of 13.4-14.5 kcaVmol reported by Roth and co-workers recently for several methyl-substituted allyl radicalsf51.In order to avoid this complication of a competing Cope rearrangement we have now investigated (E,E)-meso-and (E,E)-~~-5,6-di-tert-butyl-2,2,9,9-tetramethyl-3,7-d~ene 4a, b because their Cope rearrangement should be degenerate except for the stereochemistry of the products 4c-f. From force-field calculations (vide supra) it is e...

Thermolabile Hydrocarbons, 32 [' I. -Competing Cope-Rearrangement and Homolytic Decomposition of meso-and ~~-3,4-Di(l-cyclohexen-l-yl)-2,2.5.5-tetramethylhexane meso-and ~~-3 , 4 -d i ( l-cyclohexen-l-yl)-2,2,5,5-tetramethylhex-calculations of the diastereomers of 3a and the Cope products ane (3a) were prepared by an improved reduction procedure 8a were performed by using MM2, MMZERW, and MM3 in (Li/C,H,NH,) from meso-and ~~-2,2,5,5-tetramethyl-3,4-di-order to analyse the conformational situation. The calculated phenylhexane. Products of two competing thermal transfor-A@(g) values for rneso3a are higher than the experimental mations, C -C cleavage vs. Cope rearrangement, were isolated value by 4.4 (MMZ), 5.2 (MMSERW), and 14.5 (MM3) kcal/mol. and identified. From a kinetic analysis of this system of re-Obviously, the three computational methods require reliable actions, characteristic activation enthalpies and entropies parameters for highly branched olefins like 3a and 5a. From were obtained for the competing transformations. The stan-the experimental heat of formation A@(g) and strain enthalpy dard heat of formation of meso-3a (Ak$(g) = -58.4 k 1.2 kcal/ H, of meso-3a and its AH* of homolysis the stabilization enmol) was determined by combustion calorimetry and by meas-ergy of the allylic radical 4 was determined to be 12.6 f 1.5 urement of its heat of sublimation. A strain enthalpy H, (meso-kcal/mol. 3a) = 10.2 f 1.2 kcal/mol was derived therefrom. Force-field Durch das kinetische und reaktionsmechanistische Studium der thermisch induzierten Homolyse von C -C-Bindungen in sterisch gespannten, substituierten Kohlenstoffgerusten, z.B. 1, konnten die fur die Bindungsstarke verantwortlichen Faktoren erkannt und weitgehend quantitativ analysiert ~erden['*~]. Der sterische Effekt wurde durch lineare Beziehungen zwischen den Aktivierungsenthalpien der Homolyse und der Abnahme der Spannungsenthalpie im DissoziationsprozeD quantitativ erfaDt. Aus dem uberlagerten elekronischen Effekt der Substituenten S auf die Aktivierungsenthalpie der Dissoziation laDt sich die Stabilisierungsenergie der Substituenten S auf ein Radikalzentrum in 2 ableiten. Dabei haben wir den durch Austausch von Methyl gegen S festgestellten kinetischen Effekt zugrundegelegt r3 -51. IEin zusatzlicher Punkt des Interesses war die Moglichkeit, neben dem homolytischen Zerfall zu 4 a auch Cope-Umlagerung zu 5a zu beobachten und die Konkurrenz dieser beiden Reaktionen zu studieren. In einer fruheren Studie hatten wir festgestellt, da13 die sterisch weniger belastete

rearrangements rearrangements O 0140 -084Thermolabile Hydrocarbons.Part 32. Competing Cope-Rearrangement and Homolytic Decomposition of meso-and DL-3,4-Di(1-cyclohexen-1-yl)-2,2, 5,5-tetramethylhexane -(kinetic analysis; force-field calculations using MM2, MM2ERW, and MM3) .-(HERBERG, C.; BECKHAUS, H.-D.; KOERTVELYESI, T.; RUECHARDT, C.; Chem.