Investigating the Effects of Chemical Mechanism on Soot Formation Under High-Pressure Fuel Pyrolysis

We performed Computational Fluid Dynamics (CFD) simulations using a Reynolds-Averaged Navier-Stokes (RANS) turbulence model of high-pressure spray pyrolysis with a detailed chemical kinetic mechanism encompassing pyrolysis of n-dodecane and formation of polycyclic aromatic hydrocarbons. We compare the results using the detailed mechanism and those found using several different reduced chemical mechanisms to experiments carried out in an optically accessible, high-pressure, constant-volume combustion chamber. Three different soot models implemented in the CONVERGE CFD software are used: an empirical soot model, a method of moments, and a discrete sectional method. There is a large variation in the prediction of the soot between different combinations of chemical mechanisms and soot model. Furthermore, the amount of soot produced from all models is substantially less than experimental measurements. All of this indicates that there is still substantial work that needs to be done to arrive at simulations that can be relied on to accurately predict soot formation.

Validating Soot Models in LES of Turbulent Flames: The Contribution of Soot Subgrid Intermittency Model to The Prediction of Soot Production in an Aero-Engine Model Combustor

Predicting soot production in industrial systems using an LES approach represents a great challenge. Besides the complexity in modeling the multi-scale physicochemical soot processes and their interaction with turbulence, the validation of newly developed models is critical under turbulent conditions. This work illustrates the difficulties in evaluating model performances specific to soot prediction in turbulent flames by considering soot production in an aero-engine combustor. It is proven that soot production occurs only for scarce local gaseous conditions. Therefore, to obtain a statistical representation of such rare soot events, massive CPU resources would be required. For this reason, evaluating soot model performances based on parametric studies, i.e., multiple simulations, as classically done for purely gaseous flames, is CPU high-demanding for sooting flames. Then, a new strategy to investigate modeling impact on the solid phase is proposed. It is based on a unique simulation, where the set of equations describing the solid phase are duplicated. One set accounts for the reference model, while the other set is treated with the model under the scope. Assuming neglected solid phase retro-coupling on the gas phase, the soot scalars from both sets experience the same unique temporal and spatial gas phase evolution isolating the soot model effects from the uncertainties on gaseous models and numerical sensitivities. Finally, the strategy capability is proven by investigating the contribution of the soot subgrid intermittency model to the prediction of soot production in the DLR burner.

Comparison of soot models for reacting sprays in diesel engine-like conditions

Stringent emission legislations and growing health concerns have contributed to the evolution of soot modeling in diesel engines from simple empirical relations to methods involving detailed kinetics and complex aerosol dynamics. In this paper, four different soot models have been evaluated for the high temperature, high pressure combusting dodecane spray cases of engine combustion network (ECN) spray A which mimics engine-relevant conditions. The soot models considered include an empirical, a multistep, a method of moments based, and a discrete sectional method soot model. Two experimental cases with ambient oxygen volume of 21% and 15% have been modeled. A good agreement between simulations and experiments for vapor penetration and heat release rate has been obtained. Quasi-steady soot volume fraction contours for the four soot models have been compared with experiments. Contours of the species and source terms involved in soot modeling have also been compared for a better understanding of soot processes. The empirical soot model results in higher magnitude and spread of soot due to a lack of modeling framework for oxidation through OH species. Among the four models studied, the multistep soot model has been observed to provide the most promising agreement with the experimental data in terms of distribution of soot and location of peak soot volume fraction. Due to a two-way coupling of soot models, the detailed models predict an upstream location for soot as compared to the multi-step soot model which is one way coupled. A significant difference (of an order of magnitude) in the concentration of PAH (polycyclic aromatic hydrocarbons) precursor between multistep and detailed soot models has been observed because of precursor consumption due to the coupling of detailed soot models with chemical kinetics. It is recommended that kinetic schemes, especially those concerning PAH, be validated with experimental data with a kinetics-coupled soot model.

Soot and combustion models for direct-injection natural gas engines

This paper summarizes the validation of a modified multi-step phenomenological soot model and an enhanced combustion model used for direct-injection natural gas engines. In this study, a modified phenomenological soot model including the key steps for soot formation, such as particle inception and surface growth, was developed in KIVA-3V to replace the empirical model for use in a glow plug assisted natural gas direct-injection engine. The soot model was integrated with a CANTERA based kinetic model, which employs a recently developed low temperature natural gas mechanism to predict the reactions of some important gaseous species involved in the soot formation, such as acetylene and hydroxyl. The simulated in-cylinder flame propagation process induced by a glow plug was compared to the experimental optical images obtained in an engine-like environment. In addition, both the kinetic model and modified soot model were compared with the experimental emission data to validate their reliability for predicting natural gas engine emission characteristics. The engine combustion efficiencies obtained in simulations and experiments were compared as well. The matched results suggest that the computational models can well predict the natural gas combustion and emission characteristics, and will be suitable for investigating the direct-injection natural gas engine technologies.

Development and Application of a Reversible PAH Formation Model for Soot Prediction in Complex Fuel CFD Applications

An improved modelling approach for polycyclic aromatic hydrocarbons (PAHs) and soot formation in complex fuels is presented. The introduction of PAH radicals allows a reversible growth by hydrogen abstraction and carbon addition. Emphasis is placed on the model’s general validity with respect to fuel flexibility and operating condition using one set of model constants. A detailed gas phase mechanism describes the decomposition of fuel species as well as the formation and growth of PAHs and soot precursors. PAHs and PAH radicals are described by a sectional approach. Soot particle dynamics are modeled either by a two-equation model or alternatively by a sectional approach. All models take the processes of growth, collision, oxidation and agglomeration into account. The introduction of a temperature-dependent collision coefficient enhances the PAH and soot interaction. The differences between the two-equation model and the sectional approach are investigated. An extensive set of shock tube experiments is examined to verify the developed PAH and soot model over a wide range of temperatures, pressures, fuels and mixing-ratios. Thereby, the pyrolysis and oxidation of ethylene, benzene, kerosene and its major components are examined. In addition, ignition delay times and laminar diffusion flames are used for further validation. The overall agreement to experimental data demonstrates the applicability of the presented PAH and soot model even for complex fuels.