The study of atom reactions in gases by electron spin resonance

Abstract
Electron spin resonance provides a specific and sensitive method of studying para-magnetic free atoms in gases at low pressures. We are using this technique to study the reactions of free oxygen atoms and hydrogen atoms with simple hydrocarbons using an X–band superheterodyne spectrometer (Coates 1964; Brown 1966) with a TE102 mode cavity. With this arrangement, only atoms (which have magnetic dipole transitions) can be detected, the smallness of the magnet available has prevented the use of a cavity which was also suitable for the electric dipole transitions of diatomic radicals. Magnetic field modulation at 33 c/s is used and the signal recorded in differential form by phase sensitive detection at 33 c/s. Transitions of molecular oxygen are used to calibrate the sensitivity at regular intervals (Krongelb & Strandberg 1959; Westenberg & de Haas 1964), this calibration was checked by measuring the oxygen atoms produced in the stoichiometric titration of active nitrogen with known flows of nitric oxide. All the reactions studied to date are first order in free atom concentration, therefore only relative concentrations are needed except in determining the overall stoichiometry. It was found that the peak-to-peak height of the differential display and its first moment were strictly proportional to each other for both oxygen atoms and hydrogen atoms at constant total pressure (figure 1) providing the usual precautions were taken to avoid line distortion and power saturation. Nitrogen atoms, which give very narrow lines and have a long spin relaxation time, need particular care. The design and operation of the flow system call for some comment. A 10 mm diameter quartz flow tube is used with a fixed distance of about 1 m between the discharge and the measuring cavity (figure 2). For typical total pressures around 2 mmHg, a 50 W microwave discharge gives about 1% of free atoms (i. e. 10-9 g atom cm-3 or 6 x 1014 atoms cm-3). Concentrations of hydrogen atoms are measured down to 10-11 g atom cm-3 (6 x 1012 atoms cm-3), the limit of detection being smaller than this by about a factor of 10. A movable probe permits the introduction of varying amounts of reactants at different points upstream from the measuring cavity. The change of atom concentration at the cavity is then wholly due to the reaction being studied, and the loss of atoms by parallel processes such as re­combination on the walls of the flow tube is automatically cancelled, providing it and the reaction studied are both first order in free atoms (Clyne & Thrush 1963). This condition is fulfilled for the surface recombination of oxygen atoms and hydrogen atoms. The above relation holds even if the surface recombination is not uniform (Hartley & Thrush 1967). Since all runs can start from the same measured atom concentration irrespective of the reaction time and reactant concentration chosen, the reproducibility and drift of the spectrometer and flow system can be checked frequently. Under favourable conditions, spectrometer reproducibility approaches 3%.

This publication has 0 references indexed in Scilit: