Journal of the American Chemical Society, Vol.123, No.23, 5518-5526, 2001
Oxygen-carbon bond dissociation enthalpies of benzyl phenyl ethers and anisoles. An example of temperature dependent substituent effects
For some time it has been assumed that the direction and magnitude of the effects of Y-substituents on the Z-X bond dissociation enthalpies (BDE's) in compounds of the general formula 4-YC(6)H(4)Z-X could be correlated with the polarity of the Z-X bond undergoing homolysis. Recently we have shown by DFT calculations on 4-YC6H4CH2-X (X = H, F, Cl, Br) that the effects of Y on CH2-X BDE's are small and roughly equal for each X, despite large changes in C-X bond polarity. We then proposed that when Y have significant effects on Z-X BDE's it is due to their stabilization or destabilization of the radical. This proposal has been examined by studying 4-YC6H4O-X BDE's for X = H, CH3, and CH2C6H5 both by theory and experiment. The magnitudes of the effects of Y on O-X BDE's were quantified by Hammett type plots of Delta BDE's vs sigma+ (Y). Calculations reveal that changes in O-X BDE's induced by changing Y are large and essentially identical (rho (+) = 6.7-6.9 kcal mol(-1)) for these three classes of compounds. The calculated pf values are close to those obtained experimentally for X = H at ca. 300 K and for X = CH2C6H5 at ca. 550 K. However, early literature reports of the effects of Y on O-X BDE's for X = CH3 with measurements made at ca. 1000 K gave rho (+) approximate to 3 kcal mol(-1). We have confirmed some of these earlier, high-temperature O-CH3 BDE's and propose that at 1000 K, conjugating groups such as -OCH3 are essentially free rotors, and no longer lie mainly in the plane of the aromatic ring. As a consequence, the 298 K DFT-calculated Delta BDE for 4-OCH3-anisole of -6.1 kcal mol(-1) decreases to -3.8 kcal mol(-1) for free rotation, in agreement with the ca. 1000 K experimental value. In contrast, high-temperature O-CH3 Delta BDE's for three anisoles with strongly hindered substituent rotation are essentially identical to those that would be observed at ambient temperatures. We conclude that substituent effects measured at elevated temperatures may differ substantially from those appropriate for 298 K.