Carbon oxidation in turbulent premixed jet flames: A comparative experimental and numerical study of ethylene, n-heptane, and toluene

Abstract

An experimental and numerical investigation of the thermochemical structure of piloted premixed jet flames was conducted, encompassing laser absorption tomography measurements and large-eddy simulations (LES). The investigation was performed holding laminar flame speed, jet Reynolds number, and surrounding flow conditions constant while considering three different fuel types, namely an alkene, a normal alkane, and an aromatic fuel. Quantitative spatially-resolved thermochemical profiles of carbon monoxide (CO), carbon dioxide (CO₂), and temperature obtained from laser absorption tomography were compared against profiles predicted by the simulations for premixed ethylene-, toluene-, and n-heptane-air flames. Variations in flow structure are observed for the different fuels, highlighting fuel-specific chemical effects on the spatial evolution of the flames. Quantitative agreement of laser absorption tomography and LES results is generally observed for all flames, with larger deviations observed in the nozzle-near region for the higher molecular-weight fuels, indicating potential deficiencies in the turbulent mixing models. To the authors’ knowledge, these measurements represent the largest molecular-weight fuels for which quantitative thermochemical data have been reported in a canonical piloted premixed jet-flame configuration. The spatially-resolved experimental measurements of CO, CO₂, and gas temperature provide valuable data which can be used as validation targets for the development of turbulent combustion models.

Introduction

Turbulent combustion has been the focus of extensive research efforts over the last several decades, with particular attention devoted to the investigation of hydrogen and light hydrocarbon fuels such as methane [1], [2]. Although these studies provide valuable knowledge about highly turbulent flames in the thin and broken reaction zone regimes, relatively few investigations have assessed the importance of finite-rate chemistry in the context of fuel specific effects, particularly those caused by the variety of functional groups encountered in practical fuels. Considering that many energy conversion devices rely on turbulent combustion of liquid fuels comprising numerous high molecular weight components, the investigation of fuel effects is of particular importance.

Unlike lighter fuels, heavy hydrocarbons are susceptible to thermal decomposition in regions of the flame that precede oxidation [3], [4], [5], [6]. In particular, at high-turbulence intensities, small-scale eddies penetrate these regions and modify its thermal and chemical structure [1], [7], [8]. Therefore, for heavy liquid fuels, the local flame structure could exhibit substantially different extinction and propagation behaviors compared to small-hydrocarbon fuels which are resistant to decomposition [9]. These phenomena especially depend on the mixture of smaller molecular fragments created from heavy fuel pyrolysis, which are eventually transported to the reaction zone [10]. Additionally, the diffusivities of heavy fuels and the products of their decomposition are substantially different than those of lighter fuels. Experimental investigations provide evidence that preferential diffusion effects—which could be enhanced for heavy-hydrocarbon fuels—affect the local flame structure and its overall response to hydrodynamics [9], [11], [12].

Recent developments in canonical burner designs, namely, the Hi-Pilot configuration developed by Driscoll and coworkers [13] and the Piloted Premixed Jet Burner (PPJB) developed by Dunn et al. [14], [15] have allowed the study of premixed jet flames in high Reynolds (Re) and Karlovitz (Ka) regimes of turbulence. Investigations with a similar PPJB at USC [16], [17], [18], shown in Fig. 1, have examined highly turbulent lean and near-stoichiometric premixed jet flames to explore fuel effects within the thin and broken reaction zones. Carbone et al. [16] captured time-averaged and instantaneous CH* chemiluminescence as well as the behavior of the mean and fluctuating velocity components for a wide range of C₁–C₈ jet flames at Re of 12,500 and 25,000. The authors observed qualitative and quantitative deviation between flames of methane and other liquid fuels, and explored the potential of scaling parameters such as the laminar flame speed (S_L) and the adiabatic flame temperature (T_ad) to scale the flame observables. Only S_L showed reasonable success in scaling the flame heights derived from CH* chemiluminescence. Paxton et al. [19] used the PPJB with an ignited coflow to study the effects of heat loss on flames in the broken reaction zones regime at Re up to 50,000. Although the flame heights were shown to scale reasonably well with S_L, the differences between various fuels were not entirely suppressed in the broken reaction zones regime, where heat loss has been found to significantly affect the jet reactivity. These differences become more prominent at higher Re and less pronounced for stronger burning flames.

Evidently, accurate and sufficiently resolved experimental measurements of thermochemical properties in reacting flows help distinguish physical behaviors of different fuels allowing for comparison with high-fidelity models, particularly for high Re and Ka number flames in the thin and broken reaction zones regimes. As such, several non-intrusive optically-based measurement techniques have been utilized to study turbulent flames. These include Rayleigh scattering [14], [20], [21], [22], Raman scattering [14], [23], [24], laser-induced fluorescence (LIF) [25], [26] and chemiluminescence [16], [18]. With the exception of Raman scattering, these spectroscopic methods are generally not well-suited for quantitative species detection without extensive calibration [27]. Moreover, relatively weak Raman interactions pose practical difficulties due to the size and power of the required light sources. In contrast, laser absorption spectroscopy (LAS) provides for a calibration-free quantitative method to discern gas properties using compact low-power light sources [28]. Though traditionally limited in non-uniform flows due to the line-of-sight nature of the technique, the integration of tomographic methods has expanded applicability  [29]. Recently, laser absorption tomography (LAT) was demonstrated to provide two-dimensional temperature and mole fraction measurements of CO and CO₂ in turbulent premixed jet flames using mid-infrared semi-conductor lasers [30]. This method—employed in the present work—is suitable for small diameter ( ~ 1 cm) axially-symmetric reacting flows and utilizes tomographic reconstruction techniques [31], [32] to extract time-averaged radial thermochemical profiles from spatially-resolved line-of-sight absorption measurements.

In this study, the thermochemical structure of turbulent jet flames of ethylene, n-heptane, and toluene, was experimentally and computationally examined using a piloted premixed jet flame burner. In this work, we define thermochemical structure as the spatially-resolved temperature and species concentration scalar fields in the reacting flow. The canonical piloted premixed jet flame burner configuration is widely used for turbulent combustion model validation [18], [24], [25], [33], [34], [35]; this represents an opportunity for comparing quantitative LAT measurements with numerical models, specifically large-eddy simulations (LES). The measurements in this study provide spatially-resolved profiles of CO, CO₂, and temperature, targeting regions of carbon oxidation. These carbon oxides are chosen for their roles as critical combustion intermediates and products and their relevance in determining a boundary of heat release associated with the kinetically slow oxidation of CO to CO₂. The novel experimental dataset is accompanied by a series of LES using finite-rate chemistry models to examine the predictive accuracy of current models in capturing fuel effects in these flames, as well as to quantify the influence of turbulent flow-field behavior on the measurements.

The remainder of the manuscript has the following structure: The burner configuration, operating conditions, experimental techniques and simulations methods are presented in Section 2. The results and comparisons between experiments and simulations are discussed in Section 3, followed by a detailed uncertainty analysis of the experimental measurements in Section 4. The manuscript concludes with a summary of the major findings.