This work is aimed at understanding the inception phase of soot particles in diffusion flames at elevated pressures. In previous high-pressure studies, inception was invariably obscured by the excessive soot load. I here demonstrate that perfectly steady, nitrogen-diluted counterflow diffusion flames can be stabilized at pressures up to 2.5 MPa. In this configuration, increasing the global strain rate a, reduces the reactant residence time, yielding flame conditions from soot-free to heavily sooting at all pressures, with the option of highlighting the soot inception process.
I first describe the constraints that make the counterflow configuration suitable for fundamental kinetic studies, and schematize them in an envelope of allowed experimental conditions. Notably, flames can be made adiabatic, which eliminates the influence of heat losses to the burner on the evolution of soot, that are present in other studies [Gülder 2006]. Within this envelope, flames can be properly modeled with a one-dimensional computational code, at dramatically reduced cost. To satisfy these constraints in the range from 0.1 to 2.5 MPa, it was necessary to build two different combustors: a straight-duct combustor covering the range from 0.1 to 0.8 MPa, and a converging-nozzle combustor spanning the 0.8-2.5 MPa range. Temperature measurements with thin filament pyrometry in the visible range, confirmed theoretical and numerical findings that the thermal-layer thickness scales as (p · a )-1/2. Additional scaling of the computed velocity profiles proportionally to (p/a)-1/2 shows that the temperature-convective time history of a fluid parcel and of a soot particle is invariant within a set of experiments characterized by a change in pressure only. It is consequently possible to evaluate the net effect of pressure on the kinetics of aromatic and soot inception, and indirectly, of pressure-dependent transport.
Because of the link with soot nucleation, there is strong interest in improving current kinetic mechanisms for aromatics. To this end, gaseous species were quantified by microcapillary sampling followed by GC-MS analysis, in nitrogen-diluted, ethylene/oxygen flames in the range 0.1-2.5 MPa. Thin capillary probes and a novel point-to-point correction of the probe position relative to the flame, yielded sufficient spatial resolution even at the highest pressure. This result is confirmed by the excellent agreement with the one-dimensional kinetic model in [Narayanaswamy 2010], for mole-fraction profiles of major species and acetylene. We conclude that the model should capture the heat-release and fuel-destruction paths reliably. Notably, it shows that fuel destruction is more oxidative than pyrolytic in nature in the relatively low-temperature flames considered here. On the contrary, agreement for aromatic species was poor, suggesting the need to improve soot models.
Computational modeling showed an unexpectedly strong sensitivity of the aromatics to variations in the global strain rate. Two-dimensional cold simulations of the converging-nozzle flow revealed that the velocity profile at the nozzle exit deviates from the plug-flow profile typically assumed in the one-dimensional model. This effect is negligible for diameter-based Reynolds numbers above ~100, which applies above 0.8 MPa for the present choice of strain rate.
To probe the transition from gaseous species to soot nuclei, I retrieved soot particles with thermophoretic sampling using a minimally intrusive and chemically inert SiC wire of 13.5 µm diameter. TEM analysis revealed that soot particles have graphitic, "semi-graphitic" and amorphous structure. Small particles exhibit a "liquid-like" contact angle, possibly suggesting that the deformation occurs upon impact with the surface. SEM analysis confirmed the "liquid-like" appearance.
Sampling from a soot-forming flame with high soot load, captured the growth of aggregates, confirming that the wire has sufficient spatial resolution even in thin flames. Retrieval of primary particles from an incipiently-sooting flame yielded unexpected results. In view of the anticipated structure of a soot-forming flame, the observed trend was tentatively attributed to particle growth occurring on the wire during sampling. This effect would also provide an alternative explanation to particle liquidity. Sample collection at variable residence times showed that average size increased with sampling time, indicating that the wire temperature is too high to freeze surface growth of collected particles. Because of this growth artifact, thermophoretic data are inconclusive but indicate that successful sampling in thin high-pressure flames must satisfy two opposite constraints: a probe should be sufficiently small to minimize flame perturbation and particle mixing during the insertion phase, but sufficiently bulky to maintain a low surface temperature. In the second case, the role of condensation occurring on the probe as a potential growth artifact, which could explain "liquidity" observed by other research groups, should be considered.
The thesis is concluded with recommendations for future studies in the light of the present findings.
|School Location:||United States -- Connecticut|
|Source:||DAI-B 75/09(E), Dissertation Abstracts International|
|Subjects:||Chemical engineering, Mechanical engineering|
|Keywords:||Aronatics, Chemical Kinetics, Counterflow, Diffusion Flames, High Pressure, Soot|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be