Skip to Main content Skip to Navigation
Theses

Improving the numerical simulation of soot aerosol formation in flames

Abstract : Soot particles have been identified as the second largest contributor to global warming (just after carbon dioxide). Also, these particles can penetrate deeply into the lungs and may be carcinogenic. In this context, one of the most important properties of soot particles is their morphology which is determined by the interplay between nucleation, surface reactions, and agglomeration in flames. Modeling this process by accurately considering the morphology of particles is currently a big challenge for most numerical codes in the literature. In this thesis, a new and robust approach to simulate the evolution of soot morphology in flames is introduced. It is a Monte Carlo Discrete Element Model, called MCAC, where the trajectories of individual soot particles are accurately solved in flames and detailed particle interactions and morphology are considered. MCAC is adapted to all soot formation mechanisms including nucleation, surface growth, oxidation (including fragmentation), and agglomeration. This code is coupled with continuous CFD simulations solving the flame chemistry, fluid dynamics, and soot aerosol mass transfer leading to realistic soot formation simulations in both premixed and di usion flames. Based on this approach, the detailed dynamics of soot formation are revealed. Soot aggregation takes place in the transition of both fluid flow and aggregation regimes. This simultaneous change of regimes considerably impacts soot particle size distribution, aggregation kinetics, and particle morphology. Indeed, fractal dimensions below the classical one derived by diffusion-limited approach are found for aggregates formed in the diffusive, and transition flow regime. Agglomeration still leads to self-preserving particle size distribution when this simultaneous change of regimes is considered. This distribution is found to follow a generalized Gamma function that may be expressed based on different equivalent diameters. Aggregation (sticking) of soot particles upon collision is only systematic for soot primary particles larger than 10 nm. New expressions to determine soot collision and sticking probabilities are introduced based on a energy approach based on a coarse-grained description of soot particles. Pure agglomeration leads to agglomerates in some resemblance with those observed experimentally however, soot aggregates are formed under the competition of aggregation and surface reactions. In this context, soot aggregates exhibit a complex morphology which is modeled here as an overlapping spheres approach. Equations take primary particle overlapping effect on aggregates morphology and projected area scaling-laws into account are proposed. Finally, coupling MCAC with CFD simulations revealed the detailed evolution of soot morphology along different trajectories in the flame including the centerline and the wings of a diffusion flame. Marked and robust morphological features of soot aggregates generated under different trajectories in the flame were observed.
Complete list of metadata

https://hal-normandie-univ.archives-ouvertes.fr/tel-03566608
Contributor : Jérôme Yon Connect in order to contact the contributor
Submitted on : Friday, February 11, 2022 - 4:17:07 PM
Last modification on : Wednesday, February 23, 2022 - 3:37:06 AM
Long-term archiving on: : Thursday, May 12, 2022 - 7:31:25 PM

Identifiers

  • HAL Id : tel-03566608, version 1

Citation

José Morán. Improving the numerical simulation of soot aerosol formation in flames. Engineering Sciences [physics]. COMUE NORMANDIE UNIVERSITE, 2021. English. ⟨tel-03566608⟩

Share

Metrics

Record views

44

Files downloads

38