Publications

Abstract not provided.

Abstract not provided.

Based on a phenomenological model of diesel combustion and pollutant-formation processes, a number of fuel additives that could potentially reduce in-cylinder soot formation by altering combustion chemistry have been identified. These fuel additives, or ''combustion modifiers'', included ethanol and ethylene glycol dimethyl ether, polyethylene glycol dinitrate (a cetane improver), succinimide (a dispersant), as well as nitromethane and another nitro-compound mixture. To better understand the chemical and physical mechanisms by which these combustion modifiers may affect soot formation in diesel engines, in-cylinder soot and diffusion flame lift-off were measured, using an optically-accessible, heavy-duty, direct-injection diesel engine. A line-of-sight laser extinction diagnostic was employed to measure the relative soot concentration within the diesel jets (''jetsoot'') as well as the rates of deposition of soot on the piston bowl-rim (''wall-soot''). An OH chemiluminescence imaging technique was utilized to measure the lift-off lengths of the diesel diffusion flames so that fresh oxygen entrainment rates could be compared among the fuels. Measurements were obtained at two operating conditions, using blends of a base commercial diesel fuel with various combinations of the fuel additives. The ethanol additive, at 10% by mass, reduced jet-soot by up to 15%, and reduced wall-soot by 30-40%. The other fuel additives also affected in-cylinder soot, but unlike the ethanol blends, changes in in-cylinder soot could be attributed solely to differences in the ignition delay. No statistically-significant differences in the diesel flame lift-off lengths were observed among any of the fuel additive formulations at the operating conditions examined in this study. Accordingly, the observed differences in in-cylinder soot among the fuel formulations cannot be attributed to differences in fresh oxygen entrainment upstream of the soot-formation zones after ignition.

It is generally accepted that thermal (Zeldo'vich) chemical kinetics dominate NO formation in diesel engines, so control of temperature is critical for reducing exhaust NOx emissions. Recent optical engine data revealed that when the start of injection (SOI) was retarded to very late timings, combustion luminosity decreased while exhaust NOx emissions increased, causing a "NOx bump." This data suggested that changes in radiative heat transfer from soot may affect in-cylinder temperatures and subsequent NOx formation. In this study, soot thermometry measurements of in-cylinder temperature and radiative heat transfer were correlated with exhaust NOx to quantify the role of radiative heat transfer on in-cylinder temperatures and NOx formation. The engine was operated at low-load conditions, for which the premixed burn was a significant fraction of the total heat release. Soot thermometry showed that radiative cooling reduced flame temperatures by 25-50 K for early SOI, reducing exhaust NOx by 12-25%. Near the NOx bump, radiative cooling was essentially absent, and the NOx reduction was thus removed. Radiative cooling alone, however, was found to be insufficient to be solely responsible for the observed changes in NOx emissions. Thermodynamic analysis showed that in the slow-mixing limit, compression-heating of burned gases for the large premixed-burn conditions near the NOx bump can also increase temperatures by about 25 K, for a further 12% increase in NOx. The data also shows that these two factors can contribute to other exhaust NO observations, such as the intake temperature optimum for minimizing NOx emissions. Finally, although it was not directly measured, the rate of post-flame mixing was identified and examined as a third primary factor affecting in-cylinder temperatures and NOx formation in diesel engines.Copyright © 2005 SAE International.

Laser-extinction diagnostics can provide spatially and temporally resolved measurements of attenuation from combustion-generated soot within the path of the beam. When laser-extinction techniques are utilized in high-pressure combustion environments, however, a number of complications may be encountered that are not present in low-pressure environments. Several of these experimental difficulties were investigated in diesel engine environments, and solutions that facilitated acquisition of reliable laser-extinction data were demonstrated. Beam steering due to refractive index gradients within the combusting gases was observed, and a full-angle beam divergence of over 100 mrad was measured. A spatial-filtering scheme was employed to reduce the collection of forward-scattered light and background combustion luminosity while ensuring full collection of the steered beam. To further reject combustion luminosity, a narrow-bandpass laser-line filter was employed, after diffusing the transmitted light sufficiently to avoid the effects of significant spatial non-uniformities of the filter. As the windows were subjected to thermal and mechanical stresses, dynamic etaloning effects due to the photoelastic properties of synthetic fused silica were observed. Dynamic changes in the polarization of the exit beam were also observed, as stress-induced birefringence in the windows caused dynamic phase retardation of the transmitted beam. Although these photoelastic effects could not be eliminated, they were mitigated by introducing curvature to the wavefronts in the laser-extinction beam and using polarization-insensitive elements in the detection optics. Soot deposits on window surfaces were removed ablatively using a coaxial, high-energy, pulsed Nd:YAG laser beam.