Author's School

School of Engineering & Applied Science

Author's Department/Program

Energy, Environmental and Chemical Engineering


English (en)

Date of Award

January 2009

Degree Type


Degree Name

Doctor of Philosophy (PhD)

Chair and Committee

Richard Axelbaum


In the broadest sense, oxygen-enhanced combustion: OEC) refers to the use of oxygen to improve combustion and/or process characteristics. When a stream of oxygen is available, a wide range of flame configurations is possible. This work considers two specific configurations of OEC and is divided into two parts. In Part I, fundamental experimental and numerical flame studies explore the combustion of gaseous fuel/inert mixtures in oxygen-enriched air or pure oxygen under well-defined conditions. Part II targets a more practical application by considering the combustion of solid fuels in a variety of oxygen/carbon dioxide mixing scenarios. For gaseous non-premixed flames, combining fuel-dilution with oxygen-enrichment can dramatically alter the flame structure: i.e. the relationship between the local temperature and local species concentrations). The extent of fuel-dilution and oxygen-enrichment can be quantified by the stoichiometric mixture fraction, Zst, with fuel/air flames characterized by Zst values closer to zero and diluted-fuel/oxygen flames characterized by Zst values closer to unity. Changes in flame structure resulting in less fuel and more oxygen in the region of high temperature have been identified as the primary cause for reduced soot formation in high Zst flames. Local temperature-species relationships resulting in soot-free conditions have been shown to correlate with a single conserved scalar, the local atomic carbon-to-oxygen ratio: C/O). A simple model has been developed suggesting that for soot-free conditions to exist, the local C/O ratio and local temperature must be below critical values, i.e. C/O cr and T < Tcr. For high Zst flames, the local critical C/O ratio was associated with the increased presence of oxidizing species on the fuel side of the flame. This argument was supported by experimental and numerical results showing that for high Zst flames appreciable concentrations of molecular oxygen are observed at the location of maximum temperature: xTmax). Nevertheless, the significance of the local critical C/O ratio has not been fully explained and the role of oxidizing species on the fuel side of the flame in soot suppression has not been verified. Moreover, the mechanisms responsible for the presence of appreciable oxygen at the location of maximum temperature in high Zst flames have not been evaluated. These issues are addressed in Part I of this work. In Part I, coflow flame experiments were performed to compare and evaluate the influence of flame structure on soot formation when operating under normal and inverse flame conditions. Flame structure was shown to influence soot formation in a similar fashion for normal and inverse flames when the effects of residence time were removed. The simple model previously discussed was modified to account for finite-rate chemistry and residence time effects, and was correlated with experimental data leading to the determination of the critical local temperature and critical local C/O ratio for soot inception in ethylene flames. The presence of appreciable oxygen at the location of maximum temperature was investigated using a flame code with detailed chemistry. The mechanisms responsible for O2 at xTmax in high Zst flames were determined and explained. This phenomenon was attributed to a shifting of the location of maximum temperature relative to the location of oxygen depletion, and the temperature shift was explained by considering the variations in the heat release profile at high Zst. A second numerical investigation was also conducted to evaluate the significance of the local critical C/O ratio as a parameter describing soot-free conditions, the role of oxidizing species at this location, and changes that occur in the chemical pathway to the formation of soot precursors at high Zst. The critical local C/O ratio was shown to correspond to the edge of the radical pool for flames of any Zst, and oxidizing species did not appear to accelerate soot precursor oxidation at high Zst as previously thought. A reverse pathway analysis was used to determine the dominant chemical pathway leading to the formation of soot precursors. At high Zst, a key soot precursor formation step was observed to reverse leading to the destruction of propargyl: C3H3) to form acetylene: C2H2) as opposed to benzene: C6H6) and phenyl: C6H5). The existence of soot-free flames at long residence times was attributed to this phenomenon. In Part II of this work, a form of OEC currently being considered as an enabling technology for carbon dioxide capture from pulverized coal: PC) utility plants, termed oxy-fuel combustion, was considered. Oxy-fuel combustion utilizes oxygen and recycled flue gases: RFG) as the oxidizer instead of air, therefore the concentration of oxygen in the coal carrier stream, as well as any other concentric stream or quiescent environment, is a variable. The viability of oxy-fuel combustion can be enhanced by its ability to reduce capital and operational costs by, for example, lowering the emissions of nitrogen oxide species: NOx) in situ. Studies have demonstrated that oxy-fuel combustion can lower NOx emissions by as much as 70% when compared to conventional coal/air combustion, largely due to the reduction of recycled NOx to molecular nitrogen when interacting with hydrocarbon species in the flame. This work investigates the potential for reduced NOx emissions under oxy-fuel conditions through variations in the gas composition of the fuel carrier and concentric oxidizer streams. Nitric oxide: NO) emissions were measured during the combustion of PC and PC/sawdust mixtures under air-fired and oxy-fuel conditions. The effects of excess oxygen, secondary oxidizer swirl, carrier gas flow rate, and sawdust cofiring on NO emissions were investigated. Under oxy-fuel conditions, the effect of varying the compositions of the carrier gas and concentric oxidizer streams on NO emissions was also investigated. Under the optimal oxy-fuel conditions, NO emissions were reduced by 20% when compared to air-firing. Cofiring coal with sawdust that contained less fuel bound nitrogen did not reduce the NO emissions under air-fired or oxy-fuel conditions. Changing the adiabatic flame temperature by varying the oxygen concentration in the concentric oxidizer stream did not significantly influence NO emissions until the temperature was too low and flame instabilities were observed. When increasing the oxygen concentration in the coal carrier gas a critical local stoichiometric ratio was observed that led to increased NO emissions.


Permanent URL: