Effects of Fuel Composition on Combustion Stability and NOX Emissions for Traditional and Alternative Jet Fuels

Thesis (Ph.D.)--University of Washington, 2014Synthetic jet fuels are studied to help understand their viability as alternatives to traditionally derived jet fuel. Two combustion parameters - flame stability and NOX emissions - are used to compare these fuels through experiments and models. At its core, this is a fuels study comparing how chemical makeup and behavior relate. Six `real', complex fuels are studied in this work - four are synthetic from alternative sources and two are traditional from petroleum sources. Two of the synthetic fuels are derived from natural gas and coal via the Fischer Tropsch catalytic process. The other two are derived from Camelina oil and tallow via hydroprocessing. The traditional military jet fuel, JP8, is used as a baseline as it is derived from petroleum. The sixth fuel is derived from petroleum and is used to study the effects of aromatic content on the synthetic fuels. The synthetic fuels lack aromatic compounds, which are an important class of hydrocarbons necessary for fuel handling systems to function properly. Several single-component fuels are studied (through models and/or experiments) to facilitate interpretation and understanding. Methane is used for detailed modeling as it has a relatively small and well-understood chemical kinetic mechanism. Toluene, iso-octane, n-octane, propylcyclohexane, and 1,3,5-trimethylbenzene are included as they are all potential surrogates for jet fuel components. The flame stability study first compares all the `real', complex fuels for blowout. A toroidal stirred reactor is used to try and isolate temperature and chemical effects. The reactor has a volume of 250 mL and a residence time of approximately 8.0 ms. The air flow rate is held constant such that the inlet jets are sonic and turbulent mixing is present throughout the reactor. The fuel flow rate (hence equivalence ratio) is slowly lowered until the flame cannot sustain itself and it extinguishes. The results show that there is very little variation in blowout temperature and equivalence ratio for the synthetic fuels when compared to JP8 with low levels (0, 10, and 20%) of the aromatic additive. However, the 100% aromatic fuel behaved significantly differently and showed a lower resistance to blowout (i.e., it blew out at a higher temperature and equivalence ratio). The modeling study of blowout in the toroidal reactor is the key to understanding any fuel-based differences in blowout behavior. A detailed, reacting CFD model of methane is used to understand how the reactor stabilizes the flame and how that changes as the reactor approaches blowout. A 22 species reduced form of GRI 3.0 is used to model methane chemistry. The model shows that the reactor is quite homogenous at high temperatures, far away from blowout, and the transport of chain-initiating and chain-branching radical species is responsible for stabilizing the flame. Particularly, OH radical is recirculated around the reactor with enough concentration and at a high enough rate such that the radicals interact with the incoming fuel/air and initiate fuel decomposition. However, as equivalence ratio decreases, the reactor begins to behave in a more zonal nature and the radical concentration/location is no longer sufficient to initiate or sustain combustion. The knowledge of the radical species role is utilized to investigate the differences between a highly aliphatic fuel (surrogated by iso-octane) and a highly aromatic fuel (surrogated by toluene). A perfectly stirred reactor model is used to study the chemical kinetic pathways for these fuels near blowout. The differences in flame stabilization can be attributed to the rate at which these fuels are attacked and destroyed by radical species. The slow disintegration of the aromatic rings reduces the radical pool available for chain-initiating and chain-branching, which ultimately leads to an earlier blowout. The NOX study compares JP8, the aromatic additive, the synthetic fuels with and without an aromatic additive, and an aromatic surrogate (1,3,5-trimethylbenzene). A jet stirred reactor is used to try and isolate temperature and chemical effects. The reactor has a volume of 15.8 mL and a residence time of approximately 2.5 ms. The fuel flow rate (hence equivalence ratio) is adjusted to achieve nominally consistent temperatures of 1800, 1850, and 1900K. Small oscillations in fuel flow rate cause the data to appear in bands, which facilitated Arrhenius-type NOX-temperature correlations for direct comparison between fuels. The fuel comparisons are somewhat inconsistent, especially when the aromatic fuel is blended into the synthetic fuels. In general, the aromatic surrogate (1,3,5-trimethylbenzene) produces the most NOX, followed by JP8. The synthetic fuels (without aromatic additive) are always in the same ranking order for NOX production (HP Camelina > FT Coal > FT Natural Gas > HP Tallow). The aromatic additive ranks differently based on the temperature, which appears to indicate that some of the differences in NOX formation are due to the Zeldovich NOX formation pathway. The aromatic additive increases NOX for the HP Tallow and decreases NOX for the FT Coal. The aromatic additive causes increased NOX at low temperatures but decreases NOX at high temperatures for the HP Camelina and FT Natural Gas. A single perfectly stirred reactor model is used with several chemical kinetic mechanisms to study the effects of fuel (and fuel class) on NOX formation. The 27 unique NOX formation reactions from GRI 3.0 are added to published mechanisms for jet fuel surrogates. The investigation first looked at iso-octane and toluene and found that toluene produces more NOX because of a larger pool of O radical. The O radical concentration was lower for iso-octane because of an increased concentration of methyl (CH3) radical that consumes O radical readily. Several surrogate fuels (iso-octane, toluene, propylcyclohexane, n-octane, and 1,3,5-trimethylbenzene) are modeled to look for differences in NOX production. The trend (increased CH3 → decreased O → decreased NOX) is consistently true for all surrogate fuels with multiple kinetic mechanisms. It appears that the manner in which the fuel disintegrates and creates methyl radical is an extremely important aspect of how much NOX a fuel will produce