10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Towards Establishing a Semi-Empirical Soot Formation Model for Strained, Non-Premixed, Oxygen-Enriched Flames
PHILLIP R. JOHNSON, Rajan K. Chakrabarty, Benjamin M. Kumfer, Washington University in St. Louis
Abstract Number: 1572 Working Group: Combustion-Generated Aerosols: the Desirable and Undesirable
Abstract The emission of carbonaceous soot particles to the atmosphere affects both the environment and human health. The presence of soot in industrial combustion systems also plays an important role for radiation heat transfer. For these reasons, prediction of soot fraction is often desirable in simulations of industrial boilers, furnaces and kilns using computational fluid dynamics (CFD). The complexity and size of these systems, in addition to limits on computational resources, often necessitates the use of reduced combustion models and simplified soot formation models. Many of the existing soot models are semi-empirical in nature and do not require large reaction mechanisms for fuel combustion because they are dependent only on the concentration of acetylene or the parent fuel. Existing models generally fall into two categories: one-step models in which total soot formation rate is calculated from a single Arrhenius expression, and two-step models in which nascent soot particle nucleation and surface growth are treated separately. However, most were developed for a specific application and/or validated under fuel-air combustion conditions only; their use in a different context, such as oxy-combustion, can lead to inaccurate predictions.
As reported in the literature, a soot-producing flame can become non-sooting (blue) with the combination of oxygen enrichment and fuel dilution, even while maintaining constant flame temperature. In addition, soot-producing flames may become blue with increasing strain rate. In this study, the accuracy of recognized and commonly used semi-empirical soot models is evaluated in response to oxygen enrichment and variable strain rate. The models are tested in a series of laminar counterflow flames which span the sooting-to-non-sooting transition. Flames are modeled using CHEMKIN with detailed fuel combustion chemistry. The gas-phase results are post-processed to calculate profiles of soot formation rate and volume fraction. Results indicate that existing models tend to overpredict soot formation in highly-strained flames and upon increasing oxygen enrichment, and do not adequately capture the sooting-to-non-sooting transition.
A new one-step semi-empirical soot formation model is proposed that aims to be applicable in a wide variety of flames. This model considers the effects due to changes in flame structure, caused by fuel dilution and oxygen enrichment, and also strain rate, on the soot formation rate. The model predictions are compared against experimental data from the literature for validation. Results demonstrate markedly improved accuracy for non-premixed, oxygen-enriched flames.