Soot Model Research Articles

Importance of detailed chemical kinetics on combustion and soot modelling of ventilated and under-ventilated fires in compartment

A novel in-house computation code based on large eddy simulations (LES) incorporating fully coupled subgrid-scale (SGS) turbulence, combustion, soot and radiation models for turbulent reacting flows in compartment fires has been developed. It uniquely embraces the detailed reaction mechanisms for the chemical processes involved during combustion, which provides a comprehensive description of the fuel oxidation processes. Furthermore, it gives a more complete representation of the generation of intermediate chemical by-products including toxic gases such as carbon monoxide and irritant gases such as soot. Since the modelling of hydrocarbons by-products are enabled when considering the full chemical profile, the formation of soot particles is correlated to the concentration of main soot incipient acetylene, which provides an appropriate representation of nucleation, surface growth processes. The LES code has been validated with experimental fire tests results for both ventilated and under-ventilated fires in compartment for confirmation of its robustness. The importance of incorporating the detailed reaction mechanisms in compartment fire simulations has been confirmed by comparing with experiments. For under-ventilated fires in compartment, the chemical kinetics become increasingly important since the combustion efficiency drops significantly involving generation of intermediate chemical species. It is discovered that species concentrations especially CO2 and CO are more accurately predicted by the detailed scheme comparing to the multi-step scheme, since the formation of hydrocarbons and nitrogen oxides are considered. In general, the simulation incorporating detailed kinetics showed an averaged improvement of 9.2% and 81.7% in the prediction of CO2/CO ratio and volume fraction of soot respectively. This also improves the replication of the flame structure as the fire is chemically-driven within the combustion zone.

Strain rate effect on sooting characteristics in laminar counterflow diffusion flames

The effects of strain rate, oxygen enrichment and fuel type on the sooting characteristics of counterflow diffusion flames were studied. The sooting structures and relative PAH concentrations were measured with laser diagnostics. Detailed soot modeling using recently developed PAH chemistry and surface reaction mechanism was performed and the results were compared with experimental data for ethylene flames, focusing on the effects of strain rates. The results showed that increase in strain rate reduced soot volume fraction, average size and peak number density. Increase in oxygen mole fraction increased soot loading and decreased its sensitivity on strain rate. The soot volume fractions of ethane, propene and propane flames were also measured as a function of global strain rate. The sensitivity of soot volume fraction to strain rate was observed to be fuel dependent at a fixed oxygen mole fraction, with the sensitivity being higher for more sooting fuels. However, when the soot loadings were matched at a reference strain rate for different fuels by adjusting oxygen mole fraction, the dependence of soot loading on strain rate became comparable among the tested fuels. PAH concentrations were shown to decrease with increase in strain rate and the dependence on strain rate is more pronounced for larger PAHs. Soot modeling was performed using detailed PAH growth chemistry with molecular growth up to coronene. A qualitative agreement was obtained between experimental and simulation results, which was then used to explain the experimentally observed strain rate effect on soot growth. However, quantitatively, the simulation result exhibits higher sensitivity to strain rate, especially for large PAHs and soot volume fractions.

The core–shell internal nanostructure of soot – A criterion to model soot maturity

A novel model called Surface Shell Formation (SSF) is developed which predicts soot maturity based on the equilibrium nanostructure of poly aromatic hydrocarbons (PAHs) inside soot primary particles. The characteristic isotropic core-graphitic shell internal nanostructure of soot primary particles is used to distinguish nascent from mature soot primary particles. A new sectional soot model is developed to track particle Hydrogen to Carbon (H/C) ratio and the growth in the molecular weight of PAHs inside soot primary particles. An independent Arrhenius term describes particle dehydrogenation/carbonization. It is shown that soot maturity depends on both particle size and H/C ratio. Graphitic shell formation in mature soot particles is related to the surface PAHs that change configuration from edge on surface for nascent soot to face on surface for mature soot particles. The new model is validated against experimental data for laminar premixed, partially premixed and diffusion flames. The SSF model addresses the two major limitations of the current soot modeling approaches. First, it predicts H/C ratio of soot particles by considering soot carbonization. Second, it distinguishes nascent soot from mature soot primary particles based on the internal nanostructure of soot primary particles and the presence of the graphitic shell.

PMMA壁面の上方向への燃え拡がりのLarge eddy simulation

A fully coupled fluid-solid approach for flame spread simulations has been developed using FireFOAM. Radiative heat transfer and soot treatment are fully coupled with the pyrolysis calculations. For gaseous combustion, the newly extended eddy dissipation concept (EDC) for the large eddy simulation (LES) is used. Due to low Reynolds number, the laminar combustion model based on viscous diffusion is presented and coupled with the EDC model. The soot model is based on the laminar smoke point concept recently extended to turbulent fires using the partially stirred reactor (PaSR) concept. For radiation, the finite volume discrete ordinate method (FVDOM) is used with the gaseous absorption coefficients evaluated using the polynomial coefficient of temperature. In the solid region, one dimensional diffusion equation for sensible enthalpy is solved. The effect of in-depth radiation is taken into account using the relatively simple formation according to Beer's law. The Arrhenius type pyrolysis model developed by FM Global is used along with their measurements of the PMMA physical properties as input data. The validation study has been conducted with the published experiment. The predictions are in very good agreement with the relevant experimental data, demonstrating that the present modelling approach can be used to predict upward flame spread over PMMA with reasonable accuracy. Such a fully coupled predictive tool can be used to aid fundamental studies of the flame spread phenomena such as investigating the effects of width, inclination angles and side walls on flame spread.

Methodology for the numerical prediction of soot formation in turbulent reactive flows and application to aircraft engine combustors

A strategy to simulate the soot formation in GT combustors is proposed: It is based on LES simulations using a tabulated chemistry and the soot model of Leung. The tabulated chemistry and the relative simplicity of the Leung model give rise to reasonable CPU times. In a first time, this strategy is applied to a burner working with gaseous fuel (ethylene) and experimentally studied by DLR. Good agreement between simulation and experiment is obtained for the temperature and the soot volume fraction. This strategy has been then applied to the TLC burner working with liquid kerosene and experimentally studied by ONERA and DLR. It was possible to recover by the calculation the shape of the soot volume fraction field obtained by the LII measurements. However, because of the temperature overestimate the quantitative level of soot given by the LES is too high compared to the experiment.

Combined optical and TEM investigations for a detailed characterization of soot aggregate properties in a laminar coflow diffusion flame

In order to develop soot models that include effects due to particle aggregation, data must be obtained from flames containing all of the relevant chemical and physical processes that affect the evolution of a soot particle. In this study, the effective radius of gyration of aggregated soot is experimentally determined in nitrogen-diluted ethylene coflow diffusion flames using 2-D multi-angle light scattering (2D-MALS). High spatial resolution is achieved by minimizing signal blur that results from imaging light scattering from the probe volume at oblique angles. The results are validated at six locations in the flame through comparison to an effective radius of gyration derived from transmission electron microscope (TEM) analysis of thermophoretically-sampled soot. Simulations are used to model the path taken by a soot particle to reach the TEM grid to ensure the sampling measurements have the spatial resolution necessary to differentiate between aggregate properties at different radial locations in the flame. A radial distance δ is determined from which the front face of the probe is offset from the desired sample location. The ability of the probe to separately capture wing and centerline soot morphology is confirmed by comparison to a previous time-resolved laser-induced incandescence (TiRe-LII) measurement of primary particle diameter. The observed increase in polydispersity and maximum diameter of primary particles along the flame wing compared to the centerline is thought to be a result of surface growth. Excellent agreement is found between the effective radius of gyration derived from 2D-MALS and TEM analysis. The TEM results not only confirm the optical measurement but also elucidate the effective radius of gyration, which depends on aggregate size and polydispersity.

Modelling soot formation from wall films in a gasoline direct injection engine using a detailed population balance model

In this study, soot formation in a Gasoline Direct Injection (GDI) engine is simulated using a Stochastic Reactor Model (SRM Engine Suite) which contains a detailed population balance soot model capable of describing particle morphology and chemical composition. In order to describe the soot formation originating from the wall film, the SRM Engine Suite is extended to include spray impingement and wall film evaporation models. The cylinder is divided into a wall and a bulk zone to resolve the equivalence ratio and temperature distributions of the mixture near the wall. The combustion chamber wall is assumed to exchange heat directly only with the wall zone. The turbulent mixing within each zone and between the two zones are simulated with different mixing models. The effects of key parameters on the temperature and equivalence ratio in the two zones are investigated. The mixing rate between the wall and bulk zone has a significant effect on the wall zone, whilst the mixing rate in the wall zone only has a negligible impact on the temperature and equivalence ratio below a certain threshold. Experimental data are obtained from a four-cylinder, gasoline-fuelled direct injection spark ignition engine operated stoichiometrically. An injection timing sweep, ranging from 120 CAD BTDC to 330 CAD BTDC, is conducted in order to investigate the effect of spray impingement on soot formation. The earliest injection case (330 CAD BTDC), which produces significantly higher levels of particle emissions than any other case, is simulated by the current model. It is found that the in-cylinder pressure and the heat release rate match well with the experimental data. The particle size distribution in the simulation has the same order of magnitude as the experimental one. By tracing the particles in an equivalence ratio-temperature diagram, it is demonstrated that the rich mixture near the wall becomes the source of the soot formation as a result of the wall film evaporation.

Sensitivity study of engine soot forming using detailed soot modelling oriented in soot surface growth dynamic

Development of detailed soot model independent of empirical expression is supposed to reasonably replicate the soot formation and oxidation dynamic kinetics over a wide range of diesel engine operations, thus facilitates exploring crucial factor of soot forming. Based on comprehending of varying mixture inhomogeneity on soot forming under engine operations, a new multi-step soot model oriented in surface growth dynamic had so far been updated in this study. By correlation analysis, the new soot model is able to display the influence of both temperature and mixture inhomogeneity on soot surface activity by revising the parameter of the fraction of active sites. Hence the dominant factor of soot forming is conducted by sensitivity study in aspect of three distinctive phases: fast soot inception, soot growth equilibrium and soot oxidation prevailing, under varying mixture inhomogeneity. The soot inception is more influenced by temperature under uniform mixture combustion, but the dominant factor is switched to mixture inhomogeneity in stratified mixture wherein the acetylene formation rate gets the maximum. Similarly, less oxygen helps soot abatement due to weaker chemistry reactive in soot surface growth caused by lower temperature in homogenous mixture. Nevertheless, stimulated soot surface growth by activated surface activity and vast precursor deposit under heterogeneous mixture conduces to soot forming. Simultaneously, impaired soot surface oxidation under nonuniform mixture accelerates the deteriorated engine-out soot in comparison with discharged soot in less homogenous mixture.

Dynamics of flow–soot interaction in wrinkled non-premixed ethylene–air flames

A two-dimensional simulation of a non-premixed ethylene–air flame was conducted by employing a detailed gas-phase reaction mechanism considering polycyclic aromatic hydrocarbons, an aerosol-dynamics-based soot model using a method of moments with interpolative closure, and a grey gas and soot radiation model using the discrete transfer method. Interaction of the sooting flame with a prescribed decaying random velocity field was investigated, with a primary interest in the effects of velocity fluctuations on the flame structure and the associated soot formation process for a fuel-strip configuration and a composition with mature soot growth. The temporally evolving simulation revealed a multi-layered soot formation process within the flame, at a level of detail not properly described by previous studies based on simplified soot models utilizing acetylene or naphthalene precursors for initial soot inception. The overall effect of the flame topology on the soot formation was found to be consistent with previous experimental studies, while a unique behaviour of localised strong oxidation was also noted. The imposed velocity fluctuations led to an increase of the scalar dissipation rate in the sooting zone, causing a net suppression in the soot production rate. Considering the complex structure of the soot formation layer, the effects of the imposed fluctuations vary depending on the individual soot reactions. For the conditions under study, the soot oxidation reaction was identified as the most sensitive to the fluctuations and was mainly responsible for the local suppression of the net soot production.

Experimental and numerical study of soot formation in laminar coflow diffusion flames of gasoline/ethanol blends

This paper reports the experimental and numerical results of soot formation in laminar coflow diffusion flames of vaporized gasoline/ethanol blends at atmospheric pressure to gain improved understanding of ethanol addition to gasoline on soot formation. Four gasoline/ethanol blends of different ethanol mole fractions in the fuel stream ranging from 0% up to 85%, E0, E20, E50, and E85, were investigated to quantify how soot loading varies with the amount of ethanol blending in the fuel. The laminar coflow diffusion flames were generated using a burner system designed for vaporized liquid fuels. The fuel stream was heavily diluted with nitrogen to lower the boiling points and to prevent the flames from smoking. The soot volume fraction distributions were measured using a 2D line-of-sight attenuation technique. These flames were also modeled numerically using the extensively validated CoFlame code and a recently developed reaction mechanism for gasoline surrogates including PAH formation. The results of experiment and numerical modeling agree quite well in terms of the levels of soot volume fraction and they both show a decrease in the soot loading as more ethanol is added in the fuel stream. The measured peak soot volume fraction occurs in the flame centerline region. The soot model used in this study is unable to capture this experimental feature and predicts the peak soot volume fraction in the co-annular region. Ethanol addition reduces the soot loading primarily by lowering soot inception and PAH condensation rates through decreasing the concentrations of aromatics. Additional measurements in gasoline surrogate/ethanol blend flames indicate that the gasoline surrogate model emulates well the sooting propensity of the real gasoline fuel.

Effects of soot absorption coefficient–Planck function correlation on radiative heat transfer in oxygen-enriched propane turbulent diffusion flame

The main objective of this article is to quantify the influence of the soot absorption coefficient–Planck function correlation on radiative loss and flame structure in an oxygen-enhanced propane turbulent diffusion flame. Calculations were run with and without accounting for this correlation by using a standard k–ε model and the steady laminar flamelet model (SLF) coupled to a joint Probability Density Function (PDF) of mixture fraction, enthalpy defect, scalar dissipation rate, and soot quantities. The PDF transport equation is solved by using a Stochastic Eulerian Field (SEF) method. The modeling of soot production is carried out by using a flamelet-based semi-empirical acetylene/benzene soot model. Radiative heat transfer is modeled by using a wide band correlated-k model and turbulent radiation interactions (TRI) are accounted for by using the Optically-Thin Fluctuation Approximation (OTFA). Predicted soot volume fraction, radiant wall heat flux distribution and radiant fraction are in good agreement with the available experimental data. Model results show that soot absorption coefficient and Planck function are negatively correlated in the region of intense soot emission. Neglecting this correlation is found to increase significantly the radiative loss leading to a substantial impact on flame structure in terms of mean and rms values of temperature. In addition mean and rms values of soot volume fraction are found to be less sensitive to the correlation than temperature since soot formation occurs mainly in a region where its influence is low.

Morphological effects on the radiative properties of soot aerosols in different internally mixing states with sulfate

The radiative properties of soot aerosols largely depend on their mixing state and morphology factors. In this paper, we generated soot aggregates in four mixing states with sulfate, including bare soot, partly coated soot, heavily coated soot and soot with inclusion. The number of monomers and fractal dimension of soot were varied in each mixing state while the radius of monomers was fixed at 0.025μm. Using the discrete dipole approximation method (DDA), we calculated optical parameters relevant for climate forcing simulation at mid-visible wavelength (0.55μm). Internal mixing results in enhanced absorption, scattering cross sections as well as the single scattering albedo. The enhancement ratio of the absorption is largest for heavily coated soot, which ranges from 1.5 to 1.65 with a soot volume fraction of 0.15 and is larger for soot with larger fractal dimension. The scattering cross section can be dramatically increased by factors larger than 10 when soot is heavily coated. The increasing of both the scattering cross section and the single scattering albedo is larger for soot aggregates with smaller number of monomers and fractal dimension. The asymmetry parameter is insensitive to the fractal dimension for heavily coated soot and soot with inclusion. Two simplified models including the homogeneous sphere model (HS) and the core shell sphere model (CS) were examined using the DDA results as references. The performance of the HS and CS model largely depends on the morphology factors and the mixing state of soot. For bare and partly coated soot, both the HS and CS model can introduce relative errors as large as several tens percent. For heavily coated soot, the HS model predicts the absorption with relative errors within 10%, while it overestimates the absorption with relative errors no larger than 20% for soot with inclusion. The HS model predicts the single scattering albedo and the asymmetry parameter with relative errors no larger than 10% for heavily coated soot and soot with inclusion, which is much better than the prediction by the CS model.

Numerical Investigation of Transient Soot Evolution Processes in an Aero-Engine Model Combustor

This article presents unsteady Reynolds averaged Navier–Stokes simulations (URANS) of a well-characterized aero-engine model combustor with finite-rate chemistry (FRC). The simulations give insight into the complex formation and destruction processes of soot at technically relevant conditions. It will be shown that a recently developed PAH (polycyclic aromatic hydrocarbons) and soot model is able to predict soot under complex combustion conditions at elevated pressure. Finite-rate chemistry is employed for the gas phase, a sectional approach for PAHs and a two-equation model for soot. Thus, feedback effects, such as the consumption of gaseous soot precursors by growth of soot and PAHs, are inherently captured accurately. In agreement with the experiment a precessing vortex core (PVC) is observed in the ethylene fueled combustor. This requires that the computational grid covers swirlers. The PVC intensifies mixing of fuel, primary air, and hot burned gas from the inner recirculation zone, thereby supporting flame stabilization and subsequently influencing soot. The numerical results (velocity components, temperature, and soot volume fraction) compare well with experimental data. Details of soot evolution and remaining differences to the experiment are analyzed.

Soot and nitric oxide modeling in reacting diesel jets with an unsteady flamelet progress variable model

In this work, computations of soot and nitric oxide in reacting diesel jets are carried out for a wide range of conditions by employing a Reynolds-averaged Navier–Stokes model in which an unsteady flamelet progress variable submodel is employed to represent turbulence–chemistry interactions. The computations are carried out in a constant-volume chamber. Soot kinetics are represented using a chemical mechanism that models the growth of soot precursors starting from a single aromatic ring by hydrogen abstraction and carbon (acetylene) addition, and nitric oxide is modeled using the kinetics from a submechanism of GRI-Mech 3.0. Tracer particles are used to track the residence time of the injected mass in the jet. For the soot and nitric oxide computations, this residence time is used to track the progression of the soot and nitric oxide reactions in time. The computational conditions selected reflect the changes in the injection pressure, the chamber temperature, the oxygen concentration and density, and the orifice diameter. Comparisons with the measured soot concentrations are shown when the measured results are available. Furthermore, the dependence of the soot and nitric oxide formed in the jet on the flame lift-off height is examined. Analysis of the entrained mass upstream of the lift-off height confirms that this correlation arises from the variation in the entrained oxygen.

Sectional soot model coupled to tabulated chemistry for Diesel RANS simulations

In future Euro norms, the soot volume fraction and the soot number density will be regulated. Car manufacturers need therefore accurate soot models for piston engine emissions prediction in order to develop future engine concepts. This paper addresses this question by coupling a sectional soot model with a tabulated combustion model for RANS simulations of Diesel engines. The sectional soot model, based on the work of Netzell et al. (2007), is implemented in the IFP-C3D RANS CFD code. At each time and location, transport equations are solved for several soot sections, including source terms for collisional and chemical processes. The soot model is coupled to a tabulated combustion model derived from the Engine Approximated Diffusion Flame one (EADF) (Michel and Colin, 2013). It allows to represent the minor species required by the soot model with a much lower computational cost than a kinetic solver. In order to evaluate the soot model coupled to the resulting combustion model called Variable Pressure Homogeneous Tabulated Chemistry (VPTHC), it is compared to the same soot model directly coupled to a complex chemistry solver. As this comparison can hardly be performed on a real Diesel engine case due to the very high CPU time involved by the chemical solver, it is performed on a variable volume and fuel/air ratio case which retains the essential features of a Diesel engine. Results show that the proposed coupling recovers with reasonable accuracy the evolution of the soot volume fraction and distribution. Finally, an experimental database of Diesel operating points is simulated. The database includes points with a commercial Diesel fuel and the computed surrogate (30% 1-Methylnaphthalene and 70% Decane in liquid volume) to validate the models against the experiments. Soot yields predictions from the model show an improvement against the current standards and reach industrial target of accuracy for most of the database while the model also provides a good estimation of the soot particles distributions in size.

Numerical Investigation on the Effect of Using Mixtures of Natural Gas and Aviation Fuel on Flame Characteristics of Swirl Gas Turbine Combustor

The present paper is directed to study aviation fuel combustion compared to that of natural gas in a gas turbine swirl Combustor. Finite-volume based CFD commercial package ANSYS Fluent 12 is used for the present investigation. The SST k-ω model is used to simulate turbulence. The non-premixed combustion model is used to simulate the reacting flow. Combustion of natural gas is sooty in nature. Thus, soot distribution in the combustor is investigated numerically using the two step soot model. Simple expressions for the soot formation and oxidation rates were employed. The radiation heat transfer equation was solved using the P-1 radiation model. The formation of thermal NO from molecular nitrogen was modeled and the prompt NO was also included in the computations. CFD modeling is validated against experimental temperature measurements for a swirl combustor when firing natural gas as referred to a previous paper. Using the SST-K-ω for predicting gas temperature distributions satisfy reasonable agreement with the experimental temperature measurements, except near the combustor upstream and in the vicinity of the combustor axis.

Temporally and spatially distributed combustion in low-octane gasoline multiple premixed compression ignition mode

An advanced combustion strategy namely “multiple premixed compression ignition” (MPCI) can exploit multiple combustion regimes. In this paper, the two-stage premixed combustion (i.e. low-temperature combustion and high-temperature combustion) using relatively low-reactivity and high volatility fuel is uncovered in MPCI mode. The mixture formation and combustion processes of low-octane gasoline MPCI mode is resolved by KIVA–CHEMKIN. The low-octane gasoline (G70H30) modeled in this study is a mixture of 70% gasoline and 30% n-heptane by volume, and it is represented by iso-octane and n-heptane oxidation mechanism. A discrete multi-component (DMC) vaporization model is used to simulate the evaporation process of G70H30, while a simplified 12-step GRI-Mech model and a two-step soot model are applied for the NOx and soot calculation, respectively. As a comparative study, partially premixed compression ignition (PPCI) mode of G70H30 is also simulated using the same numerical models. From the longitudinal and transverse sections of the combustion chamber, temporally and spatially distributed combustion events are found in the MPCI mode. That is to say, the first low temperature premixed compression ignition occurs outside of the piston bowl, and it occurs mainly in the squeeze zone above the piston triggered by the local fuel rich region. Then the second high temperature premixed combustion, which is controlled by the local turbulence resulting from the second fuel injection and chamber wall interaction, is located in the center of the combustion chamber. The temporally and spatially distributed combustion in MPCI mode decreases the local temperature, resulting in less NOx formation and lower heat transfer losses. However, there is no marked combustion distribution phenomenon in the G70H30 PPCI mode, and more NOx is generated due to higher local temperature. In addition, the early first injection of MPCI and PPCI causes spray-liner impingement, and becomes the main source of soot formation in the subsequent combustion processes.

Radiation effect on temperature distribution and NO formation in a diffusion flame under reduced gravity conditions

Combustion of hydrocarbon fuel is accompanied with the formation of nitric oxide (NO) amongst other harmful emissions. In this work, a numerical investigation has been made for understanding the effect of radiative heat transfer on temperature distribution and formation of thermal NO in a methane–air diffusion flame under different reduced gravity environments. Conservation equations of overall mass, species concentration, momentum and energy for the reactive flows have been numerically solved with the use of finite difference scheme. In addition to that a semi-empirical soot model and an optically thin radiation model have been incorporated in the simulation. Gravity level is varied by the changed values of acceleration due to gravity. A thermal NO model incorporated accounts for the NO formation process which is decoupled from the hydrocarbon combustion. The relevant conservation equations have been solved as a post combustion reaction process. The flame height drops marginally with the reduction of gravity. Temperature becomes more uniformly distributed at lower gravity. NO formation boosts up with the fall of gravity below normal level when no radiation effect is considered. However, when radiation is considered, NO formation declines marginally with the reduction of gravity levels. Also in this case, concentration values of NO compare substantially lower with those without radiation. The upsurge of NO formation due to decline in gravity; and on the other hand, a shrinkage in concentration values of NO due to radiation effect can be attributed mainly to the rise and fall of temperature respectively in the computational zone.

Evaluation and optimisation of phenomenological multi-step soot model for spray combustion under diesel engine-like operating conditions

In this work, a two-dimensional computational fluid dynamics study is reported of an n-heptane combustion event and the associated soot formation process in a constant volume combustion chamber. The key interest here is to evaluate the sensitivity of the chemical kinetics and submodels of a semi-empirical soot model in predicting the associated events. Numerical computation is performed using an open-source code and a chemistry coordinate mapping approach is used to expedite the calculation. A library consisting of various phenomenological multi-step soot models is constructed and integrated with the spray combustion solver. Prior to the soot modelling, combustion simulations are carried out. Numerical results show that the ignition delay times and lift-off lengths exhibit good agreement with the experimental measurements across a wide range of operating conditions, apart from those in the cases with ambient temperature lower than 850 K. The variation of the soot precursor production with respect to the change of ambient oxygen levels qualitatively agrees with that of the conceptual models when the skeletal n-heptane mechanism is integrated with a reduced pyrene chemistry. Subsequently, a comprehensive sensitivity analysis is carried out to appraise the existing soot formation and oxidation submodels. It is revealed that the soot formation is captured when the surface growth rate is calculated using a square root function of the soot specific surface area and when a pressure-dependent model constant is considered. An optimised soot model is then proposed based on the knowledge gained through this exercise. With the implementation of optimised model, the simulated soot onset and transport phenomena before reaching quasi-steady state agree reasonably well with the experimental observation. Also, variation of spatial soot distribution and soot mass produced at oxygen molar fractions ranging from 10.0 to 21.0% for both low and high density conditions are reproduced.

Modeling soot formation in premixed flames using an Extended Conditional Quadrature Method of Moments

The scope of this study is the application of the recently developed univariate moment method Extended Quadrature Method of Moments (EQMOM) (Yuan et al., 2012) to model soot formation in flames. Furthermore, it is combined with another advanced moment approach, called the Conditional Quadrature Method of Moments (CQMOM) (Yuan and Fox, 2011), and this extension leads to a bivariate model.Retaining the efficiency of a moment method, EQMOM enables the reconstruction of the number density function. CQMOM is a numerically robust multivariate moment method which allows a bivariate soot particle description in terms of particle volume and surface to take into account aggregation. The joint Extended Conditional Quadrature Method of Moments (ECQMOM) model combines the advantages of the two methods to arrive at a numerically efficient bivariate moment method which captures both the particle size distribution and the formation of aggregates.Both the EQMOM and the ECQMOM model are validated against experimental results for premixed burner-stabilized ethylene flames. Thereby, the gas phase is modeled using a modified version of a very detailed, well-established kinetic scheme, which is adapted to be consistent with the moment methods introduced. The results demonstrate the suitability of the applied models to describe both soot precursors and soot evolution in flames. Furthermore, the ability of the moment approaches to represent the statistical soot model accurately is evaluated comparing EQMOM and ECQMOM to other numerical approaches, which are based on the Monte Carlo method, the standard Gaussian Quadrature Method of Moments and the Gaussian-Radau Quadrature Method of Moments, respectively.