Search

This thesis is concerned with the use of the optic-acoustic effect for the measurement of vibrational relaxation times in gases. A brief review is given of the theoretical background to vibrational relaxation and vibrational energy transfer during collision?, followed by a survey of experimental techniques used in the investigation of this phenomenon. A theoretical analysis is presented of the physical process occurring in the spectrophone, extending earlier work to consider spatial variation of the radiation input to the gas, and the case of a 'two-step' relaxing vibration. Details are given of the re-designed apparatus, together with the experimental procedure adopted for measuring relaxation phase shifts, and hence calculating vibrational relaxation times. The results are given of tests and precautions taken to eliminate or evaluate instrumental phase shifts, including the use of a calibrating gas. Measurements have been made on the infra-red active modes of carbon dioxide and nitrous oxide. Results strongly suggest that the asymmetric valence modes of these molecules relax by a complex process involving the bending modes. Measurements on carbon dioxide-inert gas mixtures show that inert gas molecules are relatively inefficient at de-exciting both the carbon dioxide vibrations. The results indicate that for these mixtures, the transition probability depends in a regular way upon the collision reduced mass, the dependence being close to that postulated by current theory.

This thesis describes the use of light emission from sulphur dioxide, to measure vibrational relaxation times (vibrational energy transfer rates) in shock heated gas mixtures. Chapter 1 surveys the principles underlying the theory of vibrational energy transfer in gases and summarises the practical methods available for the measurement of relaxation times in the shock tube. The description of the shock tube and the experimental techniques used in this work are given in Chapter 2. The analysis of the sulphur dioxide emission to give vibrational relaxation times is set out in Chapter 3. It is shown that the observed decrease in the relaxation time of nitrogen behind incident shock waves, contrary to the expected behaviour, is due solely to boundary layer formation in the shock tube. The results have been corrected accordingly; they then fall in line with the determinations of other workers.

The vibrational relaxation of HCl, HBr, HI or DCl has been studied either as a pure gas or in mixtures with argon, nitrogen or carbon monoxide. Infrared radiation from the second or third vibrational levels was used to follow the relaxation process. Measurements on pure HCl and HBr only allowed upper limits for the Napier times to be given because of the rapidity with which vibrational energy is taken up. A lower limit for the Napier time of HCl was obtained by diluting it with argon. Collisions of argon with HCl were assumed to be completely ineffective in causing T~V energy transfer. Subsequently published values fell within the range obtained (0.2 - 0.6 atm s at 1850 K). Dilution of HCl by nitrogen was only moderately effective in slowing down the rate of relaxation. Analysis showed that the Napier time for the V~V process is about 50 times longer than that for the T,R~V process in pure HCl. The value agrees with the recently published Napier time for the N2/HI system. For mixtures of hydrogen halides with CO, emission from the predominant component, CO, was monitored. Even with small amounts of halide, the relaxation was considerably faster than that of pure CO. The measurements, analysed using the equations of Bauer, showed that the V~V and T,R(CO)~V(halide) processes were approximately equally efficient. Limiting values of these rate constants were set for CO/HCl, CO/HBr, CO/HI and CO/DCl systems. They indicate that the smaller the energy difference, the faster the rate for v~v energy exchange. All the measurements are compared with theoretical predictions for the systems. The Napier times for pure HCl and HBr agree with the predictions of the Moore theory better than with those of SSH and support the idea of R~V energy transfer being important for molecules containing hydrogen. For mixtures of halide with CO, the rates of vibrational energy exchange are predicted moderately well by the SSH near resonance theory. However values of THCl-CO are poorly predicted by both theories for they both neglect the rotation of the activated molecules.