Flame structure and flame reaction kinetics III. Measurement of temperature profiles in flames at atmospheric pressure

Abstract

The temperature profile in a slow burning, flat hydrogen-nitrogen-oxygen flame at atmospheric pressure has been measured by means of (1) a thermocouple probe which could be moved through the flame zone, and (2) an optical method which involves measurement of the deflexion of rays of light passing through the flame in a direction normal to the temperature and refractive index gradient. In the case of the thermocouple measurements the effect of thermocouple configuration on the measured profile was examined, and the optimum configuration was found to be one in which the immediate leads to the thermojunction lay in an isothermal plane in the flame. Using such a configuration, the methods of schlieren and shadow photography were used to give visual evidence about possible local disturbance of the flame by the probe. However, although a small local disturbance was observed, a number of alternative methods of locating the flame during the temperature measurements were unable significantly to alter the resultant thermocouple temperature profile. The optical data permitted a direct derivation of the distribution of refractive index gradient in the flame, and hence of the overall refractivity profile. These profiles result from changes in both gas composition and temperature on passing through the flame. An iterative computational procedure is described which makes accurate allowance for composition effects. Although these effects are not very large in the flame studied, they do modify the measured deflexions on the cold side of the flame, and must be considered in an accurate temperature profile determination. Comparison of the two methods of temperature profile measurement on the same flame showed some discrepancy between the results. Although the observed discrepancy was small it was, nevertheless, appreciable in the context of flame studies, and in this particular case produced almost a twofold difference between the maximum rates of heat release calculated from the two profiles. From the nature of the discrepancy and comparison with other thermocouple temperature profiles for similar flames published elsewhere, it was concluded that the thermocouple profile was probably slightly in error, and that fine thermocouples are not reliable for precise temperature profile measurement in flames at atmospheric pressure.