Experiments on a reacting plume—1. Conventional concentration statistics

Abstract

Abstract Experiments have been carried out under laboratory conditions to study turbulent mixing with chemical reaction. Such flows are common in the environment but there are few high-resolution measurements of them under controlled conditions. The reaction NO + O 3 → NO 2 + O 2 + 200 kJ (mol) −1 (in the absence of UV radiation) has been used by introducing a non-buoyant reactive plume of NO into a turbulent grid flow doped with O 3. The experiments have been conducted over a wide range of conditions by varying the Damko¨hler number (the ratio of the flow timescale to that of the chemical reaction) by a factor of 6 and varying the ratio of initial reactant concentrations by a factor of 30. The data are presented in two park. Part 1 (present paper) presents conventional concentration statistics. Part 2 presents concentration statistics conditionally averaged by a mixing parameter and discusses their usefulness for investigating; the reactive behaviour of the plume. Specially developed chemiluminescent analysers are used to measure NO and O 3 concentrations with high-frequency response and a spatial resolution of four Kolmogorov scales (in this flow). The mixing field of the plume is studied using conserved scalar theory and is compared. to that obtained in other non-reactive flows. Results for the means, variances and joint statistics of the reactive species are presented and compared with frozen and equilibrium limits obtained from the conserved scalar theory for reacting flows. Statistics for NO show no significant trend with Damko¨hler number while those for O 3 and NO 2 have increased reaction at higher Damko¨hler number. The reactive scalar means are bounded by the frozen and equilibrium limits as they should be according to conserved scalar theory. Other reactive scalar statistics are not necessarily bounded by the limits. A simple relation between moments of O 3 and N0 2 is found in the experimental results and is confirmed by conserved scalar theory. Contribution to the mean reaction rate from the reactant covariance and product of means is compared and it is found that both are significant under the present conditions. Intensity of segregation varies over a wide range and is sensitive to initial reactant concentrations.