Quantitative technique for imaging mixture fraction, temperature, and the hydroxyl radical in turbulent diffusion flames

Rate-ratio asymptotic analysis of the influence of stoichiometric mixture fraction on structure and extinction of laminar, nonpremixed methane flames with comparison to experiments

A new imaging technique was developed that provides two-dimensional images of the mixture fraction (ξ), scalar dissipation rate (χ), temperature (T), and fuel consumption rate \((\dot{\omega}_{\rm F})\) in a turbulent non-premixed jet flame. The new method is based on “seeding” nitric oxide (NO) into a particular carbon monoxide–air flame in which it remains passive. It is first demonstrated that the mass fraction of NO is a conserved scalar in the present carbon monoxide–air flame configuration, using both laminar flame calibration experiments and computations with full chemistry. Simultaneous planar laser-induced fluorescence (PLIF) and planar Rayleigh scattering temperature imaging allow a quantitative determination of the local NO mass fraction and hence mixture fraction in the turbulent jet flame. The instantaneous mixture fraction fields in conjunction with the local temperature fields are then used to determine quantitative scalar dissipation rate fields. Advantages of the present technique include an improved signal-to-noise ratio over previous Raman scattering techniques, improved accuracy near the stoichiometric contour because simplifying chemistry assumptions are not required, and the ability to measure ξ and χ in flames experiencing localized extinction. However, the method of measuring ξ based on the passive NO is restricted to dry carbon monoxide–air flames due to the well-controlled flame chemistry. Sample imaging results for ξ, χ, T, and \(\dot{\omega}_{\rm F}\) are presented that show high levels of signal-to-noise while resolving the smallest mixing scales of the turbulent flowfield. The application, accuracy, and limitations of the present technique are discussed.

Turbulent non-premixed hydrogen-air flame structure in the pressure range of 1–10 atm

A lagrangian simulation of flamelet extinction and re-ignition in turbulent jet diffusion flames

Stretched laminar flamelet analysis of turbulents H 2 and CO/H 2/N 2 diffusion flames

A Consistent Flamelet Formulation for Non-Premixed Combustion Considering Differential Diffusion Effects

A quantitative imaging method using 2D single-pulse laser- induced predissociative fluorescence (LIPF) of OH concentrations is developed to study the flame structure in a swirling methane jet flame. A narrowband tunable KrF excimer laser is used to excite the P2(8) rotational line of the A2(Sigma) implied by (Chi) 2(Pi) (3,0) transition at (lambda) equals 248.46 nm. Though this transition produces a relatively weak signal, LIPF is much less sensitive to collisional quenching in atmospheric flames and suitable for generating quantitative data. OH concentration data are obtained by careful calibration against the flat flame burner data of known fuel-air equivalence ratio and temperature using identical optical setup. In this experiment, the OH fluorescence signal is imaged onto an intensified CCD camera. Since the distribution of OH concentration has good correspondence with the flame, the measured 2D imaging of OH indicates the instantaneous shape of the reaction zone. In the present experiment, the flame structure is first identified from the flame visualization. In the upstream of the swirling flame, combustion is found to take place in two regions; the recirculation vortex inside the recirculation bubble and the thin layer between and the thin layer between the recirculation zone and the ambient air. Higher intensity of OH is found early in recirculation zone. It is superequilibrium OH, because its intensity is higher than that from the recirculated burnt gas and the measured temperature is low. The measured OH concentrations have a direct influence on flame stabilization and NOx pollutant formation.© (1999) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

A premixed flame assisted by the burning of a diffusion flame is used in gas-turbine combustors to reduce NOx emissions. A united model that can be applied to the premixed and diffusion flames is therefore required to simulate the combustion phenomena. This paper proposes such a united model based on the author’s premixed combustion model for reactive progress variable equation. The proposed model has the following features. 1) It includes the laminar flame speed and the gradient of the mixture fraction as parameters. When the gradient of the mixture fraction is close to zero, the model is also close to the previous premixed combustion model as an asymptotic form. 2) It considers the effects of pressure in the combustor, unburned gas temperature, and flame stretch on combustion based on the laminar flame speed. 3) It can be applied to all types of turbulence models like the k-ε model, large eddy simulations, and direct simulations in the case of wrinkled laminar flames. The effect of turbulence is considered through the turbulent eddy viscosity of all turbulence models. To verify the accuracy of the model, the opposed diffusion flame presented by Tsuji and Yamaoka was numerically simulated, as an example of a laminar diffusion flame. Further, a turbulent diffusion flame, which was assisted by the burning of a pilot jet, was demonstrated using the united combustion model as an example of the turbulent diffusion flame discussed by Barlow and Frank. The flame was known as Sandia Flame D. Model results were in good agreement with the experimental data and this agreement confirmed the proposed united model was able to accurately simulate both diffusion flames.

Laminar triple flames are investigated numerically using detailed and reduced chemical reaction mechanisms. Triple flames are believed to play an essential role in flame propagation in partially premixed systems. In order to reduce the computational cost of the simulations, the flamelet-generated manifold (FGM) method is used to simplify the chemistry model. The FGM method is based on a library of premixed laminar flamelets. Mixture fraction variations in partially premixed flames are taken into account by using the mixture fraction as an extra degree of freedom. A comparison of the results computed with FGM and with detailed chemistry shows that FGM predicts the structure of triple flames very accurately. The gradient of the mixture fraction in the unburnt mixture is varied, and its influence on the structure and propagation of triple flames is studied. For decreasing gradients, the curvature of the premixed flame branch decreases and the propagation velocity increases. Due to the diffusive nature of the flow and the use of symmetry boundary conditions in the lateral direction, high mixture fraction gradients could not be realized at the position of the flame. As a result, the trailing diffusion flame branch is only weakly present. The heat release in the premixed flame branches is two orders of magnitude higher than in the diffusion flame branch. The premixed and diffusion flame branches are studied using a flamelet analysis. Flame stretch and curvature appear to be significant in the premixed flame branch of a triple flame. These effects cause a decrease in the local mass burning rate of almost 15%. The structure of the diffusion flame branch appears to be similar to the structure of a counterflow diffusion flame with the same scalar dissipation rate.

Auto-ignition processes of hydrogen, diluted with nitrogen, in heated air are numerically investigated by means of an unsteady laminar flamelet approach in mixture fraction space. The focus is on the auto-ignition delay time and the most reactive mixture fraction as obtained with five chemical mechanisms. Two strongly different levels of dilution, corresponding to experiments in the open literature, are considered. This concerns low-temperature chemistry at atmospheric pressure. The temperature of the air stream is much higher than the temperature of the fuel stream in the cases under study. We extensively investigate the effect of the co-flow temperature, the conditional scalar dissipation rate and the resolution in mixture fraction space for one case. With respect to the conditional scalar dissipation rate, we discuss the Amplitude Mapping Closure (AMC) model with imposed maximum scalar dissipation rate at mixture fraction equal to 0.5, as well as a constant conditional scalar dissipation rate value over the entire mixture fraction value range. We also illustrate that an auto-ignition criterion, based on a temperature rise, leads to similar results as an auto-ignition criterion, based on OH mass fraction, provided that the hydrogen is not too strongly diluted.

Relationships among nitric oxide, temperature, and mixture fraction in hydrogen jet flames

A lagrangian model for predicting turbulent diffusion flames with chemical kinetic effects

The determination of the concentration of OH in the Earth's troposphere is of fundamental importance to an understanding of the chemistry of the lower atmosphere. Many experiments to measure OH concentration have been performed in recent years; very few of these experiments have produced significant results. In particular, because of the extremely low amount of OH in the clean troposphere (several tenths parts per trillion by volume at summertime local noon), none of the experiments performed have attained the sensitivity limit necessary to test the photochemical theories of OH and its temporal behavior. Described here is an experiment utilizing a laser source and very high resolution ultraviolet absorption spectroscopy to measure the concentration of OH in a clean environment at a measured sensitivity limit of approximately 5 × 105 cm−3 (0.01 pptv) with an integration time of several minutes. This limit is substantially below predicted noontime OH concentrations and should be low enough to provide a rigorous test of photochemical theories of hydroxyl formation. This paper describes the instrumentation developed to make the measurements.

Laminar counterflow nonpremixed flames with local extinction caused by nonuniform inflows are numerically investigated in two-dimensions in a system governed by Lewis numbers of unity and assuming a single-step, irreversible, finite-rate reaction. A numerical method based on the SIMPLE algorithm is used. Edge flame structures are investigated through the introduction of a new progress variable defined as the normalized integral of the mass fraction of a product in the mixture fraction Z space and which expresses the progress of the chemical reaction. Calculation results reveal that the edge flame structure can be classified into three regions; fully burning flame, weakened flame with a low reaction rate, and a non-reacting preheat region. Under these conditions, and in terms of the energy balance, the edge flame structure is dependent on the Z-directional heat diffusion transport, heat consumption (given by the progress variable), and heat production by chemical reaction. It is found that each type of flame structure can be well characterized by three parameters; the mixture fraction, the stoichiometric scalar dissipation rate, and the progress variable. These three parameters are identified as important factors affecting the edge flame structure of nonpremixed flames with local extinction.

Prediction of autoignition in a lifted methane/air flame using an unsteady flamelet/progress variable model