Electron spin-lattice relaxation times of transient free radicals

Abstract

The saturation recovery method has been used to measure the electron spin relaxation times of short-lived free radicals in solution. The radicals were produced by continuous in situ radiolysis with a 2.8 MeV electron beam (except for two cases where stable radicals were formed by air oxidation). The ESR spectrometer used broad-band amplification of the detected ESR signal (time constant of 100 ns) and signal averaging by means of a transient recorder and minicomputer. The magnetic field was stepped off the ESR line for alternate saturating pulses; the phase of the microwave saturating pulse could be changed by 180° to allow separation of the saturation recovery and free induction decay. The sensitivity was such that T₁ values greater than 1 ms could be determined for samples of 1014 spins (i.e., 3×10⁻⁶ M radical in 50 ml) if the ESR linewidth was 100 mG and transient curves from 40000 saturating pulses were averaged. Somewhat smaller values of T₁ could be determined with higher radical concentrations. To achieve this sensitivity, it was necessary to use observing microwave powers which significantly perturbed the recovery signal so that an analysis in terms of an exponential decay of the ESR signal to the steady state value was not possible. Curve fitting of solutions to modified Bloch equations was used. Values of T₂ were determined from the damping of Torrey oscillations. Values of T₁ were determined for p-benzo-semiquinone anion (2.0), the 2,5-di-tert-butyl analog (11.5), ascorbate anion radical (2.3), chelidonic acid trianion radical (5.0), 1,2-dicarboxylvinyl radical (9), and H atoms (13.5), all values in ms and accurate to about ±10%. The radicals were produced in aqueous solution except for the semiquinones which were in ethanol. The values of T₂ were similar to the corresponding T₁ except for ascorbate radical and 2,5-di-tert-butyl-p-benzosemiquinone anion, where unresolved splittings made the apparent values smaller.