Isolation of Fuel Property and Boundary Condition Effects on Low Load Gasoline Compression Ignition (GCI)

Roberts, J. A. Isolation of Fuel Property and Boundary Condition Effects on Low Load Gasoline Compression Ignition (GCI). University of Wisconsin-Madison, 2018.

Gasoline compression ignition (GCI) combustion is a promising solution to address increasingly stringent efficiency and emissions regulations imposed on the internal combustion engine. However, the high resistance to auto-ignition of modern market gasoline makes low load compression ignition operation difficult. The most comprehensive work focused on low load GCI operation has been performed on multi-cylinder research engines where it is difficult to decouple effects of the combustion event from air-handling and system level parameters (e.g., intake pressurization and exhaust gas recirculation (EGR)). Further, most research has focused on technology applications (e.g., use of variable valve actuation or supercharging) rather than fundamental effects, making identification of influential factors difficult. Accordingly, there is a need for detailed investigations focused on isolating the critical parameters that can be used to enable low load GCI operation. A full factorial parametric study was completed to isolate the effects of intake temperature, EGR rate, and fuel reactivity on low load performance. A minimum intake pressure metric was used to compare these parameters. This allowed combustion phasing and load to be held constant while isolating the experiment from fuel injection effects. The effort showed that increasing intake temperature yields a linear reduction in the minimum intake pressure required for stable operation. Adding a small amount of diesel fuel to gasoline improved combustion stability with minimal need for energy addition through intake pressurization. The minimum intake pressure requirement also showed very good correlation with the measured research octane number of the fuel. However, increasing the fuel reactivity with diesel fuel, caused NOx emissions to increase. Response model analysis was used to determine the intake conditions required to maintain NOx levels that may not require lean NOx after treatment. The combination of diesel fuel blending and EGR allowed NOx levels to be reduced to near zero values with the minimum intake pressurization required. A detailed investigation into the effects of EGR showed that, for a given fuel, there is a maximum EGR rate that allows for stable operation, which effectively constrains the minimum NOx prior to aftertreatment.

Accordingly, a method that enables the variation of the fuel reactivity on demand is an ideal solution to address low load stability issues. Metal engine experiments conducted on a single cylinder medium-duty research engine allowed for the investigation of this strategy. The fuels used for this study were 87 octane gasoline (primary fuel stream) and diesel fuel (reactivity enhancer). Initial tests demonstrated load extension down to idle conditions with only 20% diesel by mass, which reduced to 0% at loads above 3 bar indicated mean effective pressure (IMEPg). Engine performance over a mode weighted drive cycle was completed based on work by the Ad-Hoc fuels committee [1] to demonstrate the performance of various levels of fuel blending for five primary modes of operation encompassing low load to high load. Lastly, several simulated transient drive cycle were analyzed to investigate the consumption rate of the reactivity enhancer. A response model was fit to the experimental data and exercised over the load based drive cycle. Results showed that the diesel consumption could be reduced to additive levels over a 10k mile oil change interval, lower than typical diesel exhaust fluid (DEF) consumption levels, which presents a pathway to a full-time GCI engine.

Experimental efforts used a minimum intake pressure metric to evaluate the auto-ignition quality of seven fuels, including two pump fuels and five FACE gasolines in a GCI engine. The results showed that research octane number (RON) trends well with the intake pressure required to achieve a desired ignition delay at low-temperature conditions, which are representative of a boosted GCI engine. At higher temperature intake conditions poor correlation is observed between RON and intake pressure requirement. Effects of octane sensitivity were dominated by the general reactivity of fuel as characterized by RON. The Octane Index and K-factors were regressed for each operating condition, and good correlation was seen between the Octane Index and the intake pressure requirement. Main effects analysis of the impact of general properties of the fuel (RON, motor octane number (MON), and sensitivity (S)) on the intake pressure requirement showed that RON was the only statistically significant parameter. Analysis of the main effects of fuel composition on intake pressure requirement showed some trends, but none were statistically significant. This indicates that the auto-ignition quality of the fuel is not characterized by variations in any single species. Analysis of the stable start-of- injection (SOI) timing injection window showed that both RON and sensitivity describe stability at low temperatures. In general, a fuel with a higher RON will have a smaller stable SOI window than a lower RON fuel. Additionally, fuels with the same RON and different sensitivities will behave differently. Analysis showed that, for a given RON, a low sensitivity fuel would tend to have a wider operating window than a high sensitivity fuel. Analysis of the heat release for the experimental cases showed that this is due to the presence of low-temperature chemistry. Fuels that suppress low- temperature chemistry did not show low-temperature heat release (LTHR) and had a narrower stability window. At high temperatures, LTHR was suppressed for all fuels, as the temperature in the jet exceeded the ceiling temperature for low-temperature oxidation.