Georgia Institute of Technology
Modeling of Soot Formation and Its Sensitivities in Reacting Flows
Abstract
dc:description.abstractStringent government regulations on non-volatile particulate matter (nVPM) emissions have necessitated the need for predictive models of soot formation in practical combustion systems. Accurately modeling soot remains a significant challenge because it involves strongly coupled physical and chemical processes, including nucleation, surface growth, oxidation, condensation, coagulation, and particle evolution, that interact across a wide range of spatial and temporal scales with turbulence, gas-phase chemistry, differential diffusion, and radiative heat transfer. This dissertation develops a systematic computational framework for modeling soot formation and quantifying its sensitivity across a hierarchy of reacting flows, ranging from canonical flames to laboratory-scale combustors. The work begins with the development and assessment of soot modeling strategies in zero- and one-dimensional premixed and non-premixed laminar flames, identifying the dominant sources of uncertainty associated with gas-phase chemistry, soot nucleation, surface growth, and particle evolution. A reduced monodisperse soot model is then integrated into the in-house large-eddy simulation (LES) solver LESLIE for multidimensional reacting flows. Two-dimensional mixing layers are first used to assess semi-empirical and PAH-based soot models. A novel reaction-rate closure for soot transport relying on a multiscale subgrid Linear Eddy Model (LEM) is proposed and assessed in LES of canonical turbulent premixed and non-premixed ethylene flames. LES-soot methodology is further applied to investigate soot formation in a laboratory-scale Rich-Quench-Lean (RQL) combustor under two operating conditions. Finally, a computationally efficient two-stage methodology combining LES-time averaged data with a Chemical Reactor Network (CRN) is developed to examine the influence of detailed chemical kinetics and soot subprocess uncertainties on its predictions. Together, these studies establish a comprehensive computational framework and provide new insights into predictive soot modeling for practical combustion systems.
Degree
thesis:*- Name thesis:degree_name
- Aerospace Engineering, PhD
- Department thesis:degree_department
- Aerospace Engineering
- Grantor
- Georgia Institute of Technology
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Karpe, Shubham Basavaraj
- Advisor dc:contributor.advisor
-
- Menon, Suresh
- Committee members dc:contributor.committeemember
-
- Oefelein, Joseph
- Sun, Wenting
- Steinberg, Adam
- Mueller, Michael
Rights
- Language dc:language.iso
- English
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/82192