Search Results

Computational fluid dynamic (CFD) simulations that include realistic combustion/emissions chemistry hold the promise of significantly shortening the development time for advanced high-efficiency, low-emission engines. However, significant challenges must be overcome to realize this potential. This paper discusses these challenges in the context of diesel combustion and outlines a technical program based on the use of surrogate fuels that sufficiently emulate the chemical complexity inherent in conventional diesel fuel.

The development of surrogate mixtures that represent gasoline combustion behavior is reviewed. Combustion chemistry behavioral targets that a surrogate should accurately reproduce, particularly for emulating homogeneous charge compression ignition (HCCI) operation, are carefully identified. Both short and long term research needs to support development of more robust surrogate fuel compositions are described. Candidate component species are identified and the status of present chemical kinetic models for these components and their interactions are discussed. Recommendations are made for the initial components to be included in gasoline surrogates for near term development. Components that can be added to refine predictions and to include additional behavioral targets are identified as well. Thermodynamic, thermochemical and transport properties that require further investigation are discussed.

This paper presents experimental results of studies investigating the effect of nitric oxide (NO) on the autoignition chemistry of a primary reference fuel blend with an octane rating of 87 in a motored engine. The experiments were conducted over a range of operating conditions in a single cylinder research engine at compression ratios of 5.2 and 8.2. The inlet manifold was heated and supercharged to pre-stress the fuel-air mixture in order to produce in-cylinder pressure and temperature histories similar to practical engines. The exhaust gas carbon monoxide concentration was monitored and used as a measure of overall reactivity. In-cylinder pressure histories were also recorded and processed to calculate in-cylinder temperature histories. Results showed that at low manifold temperatures, below that necessary to produce negative temperature coefficient behavior, up to 100 ppm of NO promoted reactivity, whereas higher concentrations retarded the reactivity.

Butane is the simplest alkane fuel for which more than a single structural isomer is possible. In the present study, n-butane and isobutane are used in a test engine to examine the importance of molecular structure in determining knock tendency, and the experimental results are interpreted using a detailed chemical kinetic model. A sampling valve was used to extract reacting gases from the combustion chamber of the engine. Samples were withdrawn at different times during the engine cycle, providing concentration histories of a wide variety of reactant, olefin, carbonyl, and other intermediate and product species. The chemical kinetic model predicted the formation of all the intermediate species measured in the experiments. The agreement between the measured and predicted values is mixed and is discussed. Calculations show that RO2 isomerization reactions are more important contributors to chain branching in the oxidation of n-butane than in isobutane.

Because diesel engines operate on an efficient thermodynamic cycle at a low economic cost, their particulate pollution has been tolerated. However, the Environmental Protection Agency is now setting more stringent particulate emission standards to be met by all diesel-powered vehicles. One candidate system to meet the new EPA requirements is a continuously regenerative porous metal trap oxidizer device. A test program was initiated to evaluate the design and performance of this diesel particulate filter (DPF) on a single-cylinder diesel engine. This DPF is comprised of a rotating porous metal filter element driven by a variable speed electric motor, a heater element for filter regeneration, a nozzle for controlling the particulate deposit pattern on the filter element, a scraping device to insure proper filter regeneration, and a stainless steel housing. A 20-micrometer powder metal element and a 21-micrometer fiber metal element were tested.

Experiments were conducted in a modified single cylinder research engine to study the correlation between heat release and knock. A skip-fire strategy was used to isolate the heat release due to chemical preignition reactions from the exothermicity from the propagating flame front. n-Butane, isobutane and mixtures of these two were used as the fuels in the study. Pressure profiles of n-butane and isobutane mixtures indicate that there is a decrease in the heat release occurring during the second skip cycle, upon increasing the percentage of isobutane in the mixture. Analysis of chemical species indicate that there are two competing chemical pathways occurring during the heat release. The role that these chemical pathways play during autoignition is discussed.

Experiments were conducted using a single cylinder, spark ignition research engine to examine the effect of the branched chain alkane isobutane (RON 102) on the autoignition chemistry of straight chain n-butane (RON 94), Isobutane was added in increasing quantities from 10 to 48% by volume. Measurement of cylinder pressure and the analysis of end gas species were used as diagnostics to determine the knock point and ascertain the end gas reactivity. Increasing the amount of isobutane suppressed the chemical reactivity, decreased the knock intensity, and delayed the time to autoignition. Examination of the chemical species in the end gas suggests that (i) low temperature chemical reactions are occurring; and (ii) increasing the amounts of isobutane in the fuel mixture leads to higher yields of propene and chain terminating methyl radicals.

We have studied the chemical aspects of the compression ignition of n-butane experimentally in a spark ignition engine and theoretically using computer simulations with a detailed chemical kinetic mechanism. The results of these studies demonstrate the effect of initial charge composition on autoignition. Experimentally, when the initial charge consisted of 80% fresh charge and 20% recycled products of combustion, we observed that autoignition was inhibited. On the other hand, charging with 80% fresh charge and 20% partial oxidation products from the previous motored cycle resulted in enhanced low-temperature chemistry (with the associated heat release and temperature increase) and autoignition. We assessed how well the detailed kinetic model could predict the autoignition and modified the model to better simulate the experimental observations. We also assessed how chemical preconditioning of the fuel-air charge affected the autoignition process.

Ceramic monolith particulate filters were evaluated for reduction of odor and irritant species in diesel exhaust. Three types of diesel particulate filters (DPF's) were tested: high efficiency catalyzed and uncatalyzed, and mid efficiency uncatalyzed. Testing was done with a single cylinder CFR diesel test engine run under steady-state conditions at low, mid and high equivalence ratios. Exhaust was sampled immediately upstream and downstream of the DPF's and analyzed on liquid chromatographs. The odorant species analyzed were oxygenated hydrocarbons (oxygenates), measured using the DOAS methodology. The irritant species analyzed were Cl to C5 aldehydes, measured using a DNPH reagent method. The high efficiency catalyzed DPF reduced exhaust oxygenate concentrations about 30% at low and mid equivalence ratios, which was the only substantial oxygenate reduction seen.

A laboratory burner has been operated with paraffinic mixtures, aromatic mixtures, n-paraffins, cetane standard fuel mixtures, and diesel No. 2 to measure fuel effects upon the production of diesel odor. Of the variables studied which included aromatic content, volatility, cetane number and specific gravity, only aromatic content was found to have a significant effect upon measured odor intensity. Normal paraffinic fuels were found to produce comparable exhaust odor intensities as a function of stoichiometry, irrespective of wide variations in their properties. At lean stoichiometries, branched paraffinic mixtures were found to produce low LCO and high LCA concentrations compared to n-paraffins.

A single cylinder diesel engine test facility and an air-aspirated spray burner facility have been modified and instrumented for studying diesel odor formation. Odor sampling and analysis techniques, based on the Arthur D. Little, Inc. Diesel Odor Analysis System (DOAS), have been developed and refined. Emissions mapping of the burner and the engine operating at steady state conditions, as well as transient engine conditions, are presented and discussed. The engine mappings presented indicate that operating parameters, such as injection angle, exceptionally high or low compression ratio, and exceptionally rich or lean air-fuel ratio, can have a significant effect on odor production. The burner results presented indicate that stoichiometry as well as combustion aerodynamics can strongly influence odor formation.

Diesel Odor sampling and analysis techniques and procedures using the Arthur D. Little, Inc. Diesel Odor Analysis System (DOAS) have been evaluated. Reproducibility of ± 0.1 TIA unit at the 2.0 TIA level and ± 0.2 TIA unit at the 1.5 TIA level are achievable if a consistent, well defined sampling procedure is used. Significant odor sample trap breakthrough and sample volume effects have been isolated. This study indicates that care must be given to defining a standard odor sampling configuration and procedure.