An Experimental Investigation of the Relationship between Flow Turbulence and Temperature Fields in Turbulent Non-Premixed Jet Flames
General Material Designation
[Thesis]
First Statement of Responsibility
McManus, Thomas Andrew
Subsequent Statement of Responsibility
Sutton, Jeffrey
.PUBLICATION, DISTRIBUTION, ETC
Name of Publisher, Distributor, etc.
The Ohio State University
Date of Publication, Distribution, etc.
2019
GENERAL NOTES
Text of Note
390 p.
DISSERTATION (THESIS) NOTE
Dissertation or thesis details and type of degree
Ph.D.
Body granting the degree
The Ohio State University
Text preceding or following the note
2019
SUMMARY OR ABSTRACT
Text of Note
In this dissertation, two sets of experiments were performed to improve the understanding of the relationship between flow turbulence and the temperature field in turbulent non-premixed flames. Independent high-speed temperature and velocity measurements were performed to examine flow and flame dynamics with a focus on spatio-temporal statistical analysis. Subsequently, simultaneous three-component velocity and high-resolution temperature measurements were performed to examine the interaction between fluid kinematics properties and the temperature field. Temperature field dynamics were first examined using high-repetition-rate (10 kHz) planar Rayleigh scattering in a series of turbulent non-premixed CH4/H2/N2 flames at two different Reynolds numbers, Re = 15,200 and 22,800. Additional high-speed particle image velocimetry (PIV) measurements were acquired in the same series of flames in order to facilitate a statistical comparison between the turbulent flow and a reactive scalar. Since the temperature and velocity data were resolved in both space and time, temporal auto-correlations, two-point spatial correlations, and two-point space-time correlation functions were derived as a function of spatial position and Reynolds number. The two-point space-time correlation maps for both the velocity and temperature fluctuations were used to better understand the mechanisms governing decorrelation for temperature and velocity fluctuations in turbulent non-premixed flames. Results show that the decorrelation of both temperature and velocity fluctuations is largely governed by both convection and turbulent velocity fluctuations, although reaction also appears to play a role, especially in the case of the temperature fluctuations. In order to examine the direct interaction between flow turbulence and temperature fluctuations, simultaneous velocity and temperature measurements are required. A critical part of the current dissertation research involved the development of a novel implementation of filtered Rayleigh scattering (FRS) as a thermometry approach that can be performed simultaneously with PIV in turbulent non-premixed flames. A detailed assessment of the most common Rayleigh-Brillouin scattering (RBS) spectra model, the Tenti S6 model, was performed for a number of gas species and gas mixtures at combustion-relevant temperatures. Overall, the results show that the Tenti S6 model produces accurate predictions of the RBS spectra for a wide range of combustion-relevant conditions and is suitable for use in FRS applications. Subsequently, a series of CH4/H2/Ar turbulent non-premixed flames were designed that facilitate quantitative temperature imaging using only a single FRS measurement. Simultaneous planar temperature and three-component velocity measurements were performed using FRS and stereo PIV in a series of piloted, turbulent non-premixed flames at Reynolds numbers of 10,000, 20,000, and 30,000. The joint temperature and velocity measurements were used to, among other things, generate detailed statistics characterizing their coupled relationship. Statistical results show that both the extensional and compressional principal strain rates play a significant role in generating large-magnitude thermal dissipation, with the most compressive principal strain rate playing a larger role for higher Reynolds number cases. The thermal scalar flux was calculated for both the axial and radial directions. It was observed that the gradient transport hypothesis appears to be satisfactory for describing transport in the radial direction; however, in the axial direction, clear indications of counter gradient transport are present.