Multi-scale Modeling of High-temperature Chemistry and Soot Formation of Bio-fuels

Open Access
- Author:
- Kwon, Hyunguk
- Graduate Program:
- Chemical Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 02, 2021
- Committee Members:
- Kristen Fichthorn, Major Field Member
Phillip Savage, Major Field Member
Randy Vander Wal, Outside Unit & Field Member
Yuan Xuan, Co-Chair & Dissertation Advisor
Adri van Duin, Chair & Dissertation Advisor
Seong Kim, Program Head/Chair - Keywords:
- Soot
Combustion
Pyrolysis
Biofuels
Computational fluid dynamics
ReaxFF molecular dynamics
Multi-scale modeling - Abstract:
- Soot refers to carbonaceous particles that have negative impacts on the environment and human health. To intelligently manage ongoing changes in fuel composition, there is an expanding interest in quantifying chemical propensity to form soot from different fuel compounds, ranging from traditional fuels to more sustainable alternative fuels. The overarching objective of this thesis is thus to develop multi-scale modeling combining computational fluid dynamics (CFD) and ReaxFF reactive force field based molecular dynamics (MD) that can determine the yield-based sooting tendency of a fuel and identify the chemical reactions leading to soot formation. The Yield Sooting Index (YSI) measured in a fuel-doped methane/air coflow diffusion flame is chosen as the specific sooting tendency metric in this thesis. For fuels with well-known combustion chemistry, CFD simulation combined with a kinetic model is performed to complement the YSI methodology. To calculate YSIs efficiently, a 1D flamelet-based YSI simulation approach is employed. The CFD of reacting flows specifically deals with two research topics. First, the pressure-dependence of YSI is investigated to identify the applicability of the YSI methodology at elevated pressures. Second, the YSIs of a large number of biofuels with complex chemistry are predicted using 1D flamlet-based YSI simulation combined with a large kinetic model. Detailed 2D CFD simulations are difficult to achieve this due to their very high computational cost. A new sensitivity analysis developed in this thesis is applied to quantify the impact of kinetic parameter uncertainties on YSI predictions. For advanced biofuels with poorly-known chemical kinetics and no associated existing kinetic models, the kinetic-based CFD simulation is not applicable. Therefore, we develop a ReaxFF reactive force field based MD simulation framework to study sooting tendencies of biofuels both quantitatively and qualitatively. For aromatic fuels, we develop a unique ReaxFF MD simulation framework that can quantitatively predict yield-based sooting tendencies, and this framework is applied to toluene and phenol as a proof-of-concept. For non-aromatic fuels, this thesis presents the methodology to study the sooting tendencies qualitatively. Polycyclic alkanes and alkyl-substituted 1,3-dioxolanes recently synthesized as potential jet-fuels and biodiesels, respectively, are specifically studied, since very little effort has been made on their combustion chemistry and sooting tendency. The findings and methodologies provided in this thesis will help to accelerate the introduction of low soot emitting advanced combustion fuels.