No abstract

The ESR linewidths of dilute solution of di-tertiary-butyl nitroxide (DTBN) in n-pentane and propane have been studied. Corrections have been made for (1) spin-rotational effects, (2) the variation of concentration with temperature, (3) the line-shape distortion of overlapping “strongly exchanging” lines, (4) the exchange narrowing of inhomogeneous linewidth contributions arising from proton extra hyperfine splitting, and (5) the temperature dependence of the diffusional step length λ. Existing theories of spin exchange appear adequate to explain the low-concentration data; at high concentration it is not possible to make all the corrections discussed above and the comparison of theory and experiment is not meaningful. In n-pentane we found that (zT/z).Jλ02 = 5×1012 Å2 sec−1 and JzT = 4.9×1011 sec−1, where J is minus two times the exchange integral, λ0 is the diffusional step length at the freezing point, zT is the mean total number of molecular nearest neighbors, and z is the number of new radical nearest neighbors encountered in each jump. The results for DTBN in propane are similar.

Measurements have been made on the ESR linewidths of vandyl acetylacetonate (VOAA) in a number of nonhydrogen-bonded solvents. The X-band spectra were taken over a wide range of temperatures and at pressures ranging up to 4.5 kbar. The high pressure measurements were carried out in a high pressure bomb containing a slow-wave helix, instead of a resonant cavity, and a pair of 100 kHz modulation coils; the sample was placed in a Teflon holder which also surrounded the helix and could be sealed after the sample was deoxygenated. The reorientation correlation time τθ, is written as τθ=43π r03κ(η/kT), where η is the coefficient of shear viscosity, T the absolute temperature, r0 the hydrodynamic radius determined by translational diffusion experiments, and κ the anisotropic interaction parameter. κ is proportional to the ratio of the mean square intermolecular torques to the mean square intermolecular force where the interaction is between VOAA and solvent molecules. The values of κ determined from these experiments are independent of pressure, temperature, and density. However, κ varies from solvent to solvent and an analysis of the data indicates, but not unambiguously, that κ decreases with solvent size and it increases with solvent dipole moment. The results are compatible with a quasirotational diffusion model, with the assumption that the autocorrelation functions for intermolecular forces and torques have the same time dependence, and with the assumption that the dependence of torques on intermolecular separation is nearly equivalent to that of forces. The single adjustable parameter, κ, is sufficient to explain all the spectra at low (η/T) for a given solvent regardless of pressure and temperature, but at high values of (η/T) unexplained discrepancies between theory and experiment occur. A discussion of anisotropic diffusion is also given; however, the anisotropic contributions are probably negligible for VOAA.

No abstract

No abstract