Hydrogen–hydrocarbon turbulent non-premixed flame structure

In this study, the structure of turbulent non-premixed CH₄-H₂/air flames is analyzed with a special emphasis on mixing and air entrainment. The amount of H₂ in the fuel mixture varies under constant volumetric fuel flow. Mixing is described by mixture fraction and its variance while air entrainment is characterized by the ratio of gas mass flow to fuel mass flow at the inlet section. The flow field and the chemistry are coupled by the flamelet assumption. Mixture fraction and its variance are transported by the computational fluid dynamics (CFD) code. The slow chemistry aspect of NO_x is handled by solving an additional transport equation with a source term derived from flamelet library.     

A study of flame spread along a droplet array at elevated pressures up to a supercritical pressure

Experimental investigations on flame spread along a droplet array have been conducted at elevated pressures up to supercritical pressures of the fuel droplet under normal gravity and microgravity. The flame spread rate is measured using high‐speed chemiluminescence images of OH radicals and direct visualization is employed to observe the images of the vaporizing fuel around the unburnt droplet. The mode of flame spread is categorized into two: a continuous mode and an intermittent one. There exist a limit droplet spacing and a limit ambient pressure in normal gravity, above which flame spread does not occur. It is seen that flame spread rate is dependent upon the relative position of flame to droplet spacing. In microgravity, the limit droplet spacing of flame spread and the droplet spacing of maximum flame spread rate are larger than those in normal gravity. In microgravity, the flame spread rate with ambient pressure decreases initially, shows a minimum, and then decreases again after taking a maximum. Flame spread time is determined by competing effects between the increased transfer time of the thermal boundary layer due to reduced flame diameter and the decreased ignition delay time in terms of the increase of ambient pressure. In normal gravity, the flame spread rate with ambient pressure decreases monotonically and there exists a limit ambient pressure, except at small droplet spacing, under which flame spread extends to the range of supercritical pressures of fuel. This is because natural convection induces the upward flow of hot gases into a plume above the burning droplets and limits the lateral transfer of thermal boundary layer. Consequently, it is found that flame spread behaviour under microgravity is considerably different from that under normal gravity due to the absence of natural convection. Copyright © 1999 John Wiley & Sons, Ltd.     

Numerical investigation on hydrogen-enriched methane combustion in a domestic back-pressure boiler and non-premixed burner system from flame structure and pollutants aspect

In this paper, the non-premixed hydrogen-enriched methane-air combustion was investigated numerically with the use of a CFD code. In the first part of the study, the combustion experiments were performed in a back- pressure boiler using natural gas. The intake rate of fuel was kept constant as 45 Nm³/h while the coefficient for the air excess ratio was changed between 1.2 and 1.35. After the experiments, the numerical analyses were performed. The Fluent code was utilized as the simulation instrument. The eddy dissipation combustion model was selected to be used in the numerical analyses, since it is known that this combustion model can save computational time and fairly predict the combustion flame structure and emissions. Pure methane and natural gas were taken as fuels in the numerical analyses. The obtained results from the numerical analyses were validated with the experimental flue gas temperature and emission measurements. Then, the hydrogen-enrichment of pure methane fuel was investigated numerically in such a way that the boiler capacity (432 kW) was kept constant. The coefficient for the air excess ratio was 1.2 for all the considered combustion simulation cases. The hydrogen addition ratio was 25%, 50% and 75% by mass, respectively. The thermal NO emissions and temperature distributions in the combustion chamber were obtained according to the different hydrogen-enriched methane fuel combustion cases. In addition, the emissions contained in the flue gas together with the temperature values were calculated. The obtained results from the numerical studies indicate that the hydrogen-enrichment of methane reduces the carbon emissions, while it substantially augments the formation of the thermal NO emissions. The calculated thermal NO emission value in the flue gas is 147 ppm for the pure methane combustion case, and it is 566 ppm for the combustion case with 75% by mass of hydrogen addition ratio. Therefore, it is determined that hydrogen fuel is a pollutant from the thermal NO emission aspect for the considered enrichment ratios in the studied domestic boiler-burner system.     

Numerical analysis of gaseous hydrogen/liquid oxygen flamelet at supercritical pressures

Supercritical conditions are typically encountered in high-pressure combustion devices such as liquid propellant rockets and gas turbine engines. Significant real fluid behaviors including steep property variations occur when the fluid mixtures pass through the thermodynamic transcritical regime. The laminar flamelet concept is a robust and reliable approach that correctly accounts for real fluid effects, the large variation in thermophysical properties, and the detailed chemical kinetics for turbulent flames at transcritical and supercritical conditions. In the present study, the flamelet equations in the mixture fraction space are extended to treat the flame field of general fluids over transcritical and supercritical states. Flamelet computations are carried out for gaseous hydrogen and cryogenic liquid oxygen flames under a wide range of thermodynamic conditions. Based on numerical results, the detailed discussions are made for the effects of real fluid, pressure, and differential diffusion on the local flame structure and the characteristics encountered in liquid propellant rocket engines.     

Large eddy simulation of methane non-premixed flame using the laminar flamelet model

The large eddy simulation (LES) using the steady laminar flamelet model is applied to a simple turbulent jet flame with 33.2% H2, 22.1% CH4 and 44.7% N2 at the Reynolds number of 15,200 in order to validate the numerical methods and to investigate the flame structure. For the validation, the detailed experimental data of DLR-A flame is used. The numerical results are in reasonable agreement with experimental results except mass fractions of minor species. In the flow field, the break-down of the potential core, the vortex structure and the mixing intensity are well captured. In the combustion field, mass fractions of major species (H2O, CO2, CO) are well predicted quantitatively. Minor species are well predicted qualitatively. In the present study, the simulations conducted on the Cartesian and cylindrical grids with approximately 6.6× 10⁵ nodes are compared.     

Error analysis of large-eddy simulation of the turbulent non-premixed sydney bluff-body flame

A computational error analysis is applied to the large-eddy simulation of the turbulent non-premixed Sydney bluff-body flame, where the error is defined with respect to experimental data. The error-landscape approach is extended to heterogeneous compressible turbulence, which is coupled to combustion as described by a flamelet model. The Smagorinsky model formulation is used to model the unknown turbulent stresses. We introduce several measures to quantify the total simulation error and observe a striking ‘valley-structure’ in the error that arises as function of the spatial resolution and the Smagorinsky length parameter. The optimal refinement strategy that can be extracted from this error-landscape is reminiscent of that for non-reacting turbulent flow.     

Laminar burning velocities and flame stability analysis of syngas mixtures at sub-atmospheric pressures

The effects of low pressure on the laminar burning velocity and flame stability of H₂/CO mixtures and equimolar H₂/CO mixtures diluted with N₂ and CO₂ were studied experimentally and theoretically. Experiments were conducted at real sub-atmospheric conditions in three places located at high altitudes 500 m.a.s.l. (0.947 atm), 1550 m.a.s.l. (0.838 atm), and 2300 m.a.s.l. (0.767 atm). Flames were generated using contoured slot-type nozzle burners and Schlieren images were used to determine the laminar burning velocity with the angle method. The behavior of the laminar burning velocity at low pressures depends on the equivalence ratio considered; it decreases at lean and very rich equivalence ratios when pressure is increased. However, a contrary behavior was obtained at equivalence ratios corresponding to the highest values of the laminar burning velocity, where it increases as pressure increases. Numerical calculations were also conducted using a detailed reaction mechanism, and these do not reproduce the behavior obtained experimentally; a sensitivity analysis was carried out to examine the differences found. At lean equivalence ratios, flame instabilities were observed for all the syngas mixtures. The range of equivalence ratios where flames are stable increases at lower pressures. This behavior is due to the increase of the flame thickness, which considerably reduces the hydrodynamic instabilities in the flame front.     

Implementation of the eddy dissipation model of turbulent non-premixed combustion in OpenFOAM

This work discusses the implementation of eddy dissipation model in OpenFOAM CFD toolbox. The code was validated in modeling of confined non-premixed Methane jet flame. The model predictions were extensively compared against published experimental results as well as ANSYS Fluent® predictions. The differences between the implemented model in OpenFOAM and Fluent were demonstrated.     

Oxidizer effects on ammonia combustion using a generated non-premixed burner

In the present study, the pure ammonia combustion in a model combustor is performed to seek ammonia-fueled applications. To this aim, effects of the oxygen enrichment with an oxygen concentration of 100% in the oxidizer on flame characteristics, temperature profiles and NO profiles during the ammonia combustion were evaluated in terms of excess air/oxygen coefficients. Furthermore, in order to better understand the effect of an oxygen of 100% usage under the oxy-ammonia combustion conditions, the air-ammonia combustion has been studied as well and their results are compared and discussed each other. According to the results predicted, the oxidizer with an oxygen content of %100 provides better flame stability in the case of pure ammonia combustion. The most stable flame for oxy-ammonia combustion can be achieved when the excess oxygen coefficient is 1.0 or 1.2. Furthermore, the minimum NO levels emerge under the fuel-rich condition. Temperature and NO emissions decrease considerably under the air-ammonia combustion. However, except the fuel-rich conditions, flame stabilities are not satisfactory due to ammonia's flame speed under the air-ammonia combustion. Moreover, the air-ammonia combustion under the fuel-rich condition seems as a good option for obtaining the lowest NO levels. On the other hand, the oxy-enrichment condition is thought as a promising method for pure ammonia combustion provided that NO emissions should be optimized by using NO reduction methods.     

MILD combustion for hydrogen and syngas at elevated pressures

As gas recirculation constitutes a fundamental condition for the realization of MILD combustion, it is necessary to determine gas recirculation ratio before designing MILD combustor. MILD combustion model with gas recir- culation was used in this simulation work to evaluate the effect of fuel type and pressure on threshold gas recir- culation ratio of MILD mode. Ignition delay time is also an important design parameter for gas turbine combustor, this parameter is kinetically studied to analyze the effect of pressure on MILD mixture ignition. Threshold gas re- circulation ratio of hydrogen MILD combustion changes slightly and is nearly equal to that of 10 MJ/Nm3 syngas in the pressure range of 1-19 atm, under the conditions of 298 K fresh reactant temperature and 1373 K exhaust gas temperature, indicating that MILD regime is fuel flexible. Ignition delay calculation results show that pres- sure has a negative effect on ignition delay time of 10 MJ/Nm3 syngas MILD mixture, because OH mole fraction in MILD mixture drops down as pressure increases, resulting in the delay of the oxidation process.     

A level set formulation for premixed combustion LES considering the turbulent flame structure

In this paper, a consistent and rigorous formulation is developed for the coupling of the G-equation model to an LES flow solver that describes the interactions of the scales of the flame, the turbulence, and the filtering procedure from the resolved turbulence regime to the broadened preheat regions regime. A progress variable equation is introduced to describe the filtered flame structure. The models provided for the sub-filter diffusivity and the filtered reaction term appearing in this equation are consistent with the solution of the G-equation model. The solution of the progress variable equation ensures that the resolved part of the turbulent mixing in the preheat region can be described. However, the C-field is underresolved if the sub-filter Damköhler number is not much smaller than unity, and hence the solution of the C-equation cannot be expected to produce the correct flame propagation speed. The coupling with the G-equation ensures that the flame front described by the filtered reaction progress variable moves with the correct propagation velocity, independent of numerical diffusion caused by an underresolution of the flame. Formulations both for low-Mach number flow solvers and for fully compressible solvers are presented. To validate the formulation, the model is applied in compressible LES of two turbulent flames anchored by a triangular flame-holder. For the statistically stationary case, the mean and RMS progress variable are in very good agreement with experimental data, demonstrating that the model correctly reproduces the flame anchoring and the flame-turbulence interactions in the recirculation zone. For the acoustically pulsed case, the LES fields show the same large scale fluctuations that are present in the experimental data.     

Nonlinear combustion instability analysis based on the flame describing function applied to turbulent premixed swirling flames

Instability analysis of swirling flames is of importance in the design of advanced combustor concepts for aircraft propulsion and powerplant for electricity production. Thermoacoustic instabilities are analyzed here by making use of a nonlinear representation of flame dynamics based on a describing function. In this framework, the flame response is determined as a function of frequency and amplitude of perturbations impinging on the combustion region. This model is adapted to the case of confined swirling flames comprising an upstream manifold, an injection unit equipped with a swirler and a cylindrical flame tube. The flame describing function is experimentally determined and is combined with an acoustic transfer matrix representation of the system to provide growth rates and oscillation frequencies as a function of perturbation amplitude. These data can be used to determine regions of instability, frequency shifts with respect to the acoustic eigenfrequencies and they also yield amplitude levels when self-sustained oscillations of the system have reached a limit cycle. This equilibrium is obtained when the amplitude dependent growth rate equals the damping rate in the system. This requires an independent determination of this last quantity which is here based on measurements of the combustor resonance response curve, together with numerical estimates of the flame contribution to the system response. The geometrical parameters of the upstream manifold and flame tube are varied and the corresponding operating regimes are compared with those predicted with the FDF framework. The present demonstration of the FDF framework in a generic configuration indicates that this can be used in more general situations of technological interest.     

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

A numerical study of hydrogen turbulent diffusion flame structure is carried out in the pressure range of 1-10 atm with a special emphasis on mixing. The investigation is conducted under constant volumetric fuel and air flows. Mixing is characterized by mixture fraction, its variance and the scalar dissipation rate. The flow field and the chemistry are coupled by the flamelet assumption. Mixture fraction and its variance are transported by computational fluid dynamic (CFD). Computational predictions are analysed at two radial stations (the first one represent the near-field region and the second one the far-field region). The computational results indicate a deterioration of mixing with pressure rise. As a result, flame reaction zone becomes thicker. In addition, mixing and flame structure sensitivity to pressure are found to be high in the first location. Further analysis revealed that the gas becomes increasingly heavy with pressure rise, which hampered its ability to mix.     

Rarefied flow effects on stabilization and extinction of rotating-disk flame at low pressures

An activation energy asymptotic analysis with one-step overall reaction was performed for the stabilization and extinction of a premixed flame over a rotating disk at sufficiently low pressures, for its relevance in low-pressure CVD (chemical vapor deposition) operations in which the flow is weakly rarefied. Extinction criteria based on the critical Damköhler number were obtained through the S-curve concept, parametrically demonstrating the influence of the CVD operating conditions, such as the spin rate and temperature of the disk, on flame extinction. It is further shown that, while decreasing pressure and hence the reactivity of the mixture tends to extinguish the flame, the trend can be substantially weakened by taking into account of the influence of the Knudsen layer, which reduces the heat loss to the disk as well as the flow stretch rate at the flame.     

Laminar flame speeds of cyclohexane and mono-alkylated cyclohexanes at elevated pressures

The propagation speeds of expanding spherical flames of cyclohexane, methylcyclohexane and ethylcyclohexane in mixtures of oxygen/inert were measured in a heated, dual-chamber vessel, with the corresponding laminar flame speeds extracted from them through nonlinear extrapolation. Measurements were conducted at atmospheric and elevated pressures up to 20 atm. Computational simulations were conducted using the JetSurF 2.0 mechanism, yielding satisfactory agreement with the present measurements at all pressures, with a slight over-prediction at 1 atm. Measurements reveal the following trend for the flame speeds: cyclohexane > n-hexane > methylcyclohexane ≈ ethylcyclohexane at all pressures, with the maximum difference being approximately 5% at 1 atm and 13% at 10 atm. Examination of the computed flame structure shows that owing to its symmetric ring structure, decomposition of cyclohexane produces more chain-branching 1,3-butadiene and less chain-terminating propene. On the contrary, a more balanced distribution of intermediates is present in the flames of methylcyclohexane and ethylcyclohexane due to substitution of the alkyl group for H.     

Characteristics of non-premixed oxygen-enhanced combustion: II. Flame structure effects on soot precursor kinetics resulting in soot-free flames

A detailed computational study was performed to understand the effects of the flame structure on the formation and destruction of soot precursors during ethylene combustion. Using the USC Mech Version II mechanism the contributions of different pathways to the formation of benzene and phenyl were determined in a wide domain of Z_st values via a reverse-pathway analysis. It was shown that for conventional ethylene-air flames two sequential reversible reactions play primary roles in the propargyl (C₃H₃) chemistry, namely

(1)     

Lift-off of jet diffusion flame in sub-atmospheric pressures: An experimental investigation and interpretation based on laminar flame speed

This paper reveals lift-off behavior of jet diffusion flames in sub-atmospheric pressures less than 100 kPa, in view of that the current knowledge on this topic is limited for normal pressure conditions. Physically, the variation of ambient pressure may have significant influence on the lift-off behavior of jet diffusion flames due to the change of some critical parameters such as laminar flame speed. In this work, experiments are conducted in a large pressure-controllable chamber of 3 m (width) × 2 m (length) × 2 m (height) at different sub-atmospheric pressures of 60 kPa, 70 kPa, 80 kPa, 90 kPa as well as at normal pressure of 100 kPa. Axisymmetric turbulent jet diffusion flames are produced by nozzles with diameters of 4 mm, 5 mm and 6 mm using propane as fuel. It is revealed that the lift-off height increases as the pressure decreases and being much higher than that in normal pressure condition. The laminar flame speed with its dependency on pressure is introduced to interpret such behavior based on classic Kalghatgi model. It is found theoretically that the lift-off height has a power law dependency on pressure by P¹⁻ⁿ, where n is overall reaction order of the fuel which is usually larger than 1 indicating a negative power law function with pressure (for example p^−0.75 for propane as n = 1.75) as well verified by the experimental correlation. Finally, a global model is proposed by including such pressure dependency function into the Kalghatgi model, which is shown to well collapse the experimental results of lift-off heights of different sub-atmospheric pressures.     

Experimental study on velocity characteristics of recirculation zone in humid air non-premixed flame

To examine the effect of the flow field within the recirculation zone on flame structure, the characteristic velocity fields of methane/humid air flame in non-premixed combustion behind a disc bluff-body burner were experimentally studied by particle image velocimeter (PIV).The results show that two stagnation points exist on the centerline in the recirculation zone flame. However, the distance of the two stagnation points in humid air combustion shortens, and the minimal dimensionless velocity increases compared with the conventional non-humid air combustion. In addition, the positional curves of the minimal velocities can be partitioned into three phases representing three different flame patterns. The analysis of axial minimal velocities on the centerline and their positions under different co-flow air velocity conditions reveals that fuel-to-air velocity ratio is the crucial parameter that governs humid air combustion flame characteristics.