Research on the combustion of alcohol-gasoline blends

Problem statement

Our current energy supply, based on fossil fuels, is becoming untenable. Their use on a massive scale has led to an intolerable stress on the local (air quality) and global environment (greenhouse effect). An important and increasing share in the energy consumption goes to the transportation sector. The presently used powerplant for most transport, the internal combustion engine (ICE), is in itself a sustainable technology, as the ICE is made from abundantly available and moreover recyclable materials. However, the presently used ICE fuels clearly are not. Alternative energy carriers for transport applications are needed.

Liquid alcohols, methanol and ethanol in particular, are attractive alternatives compared to other alternatives such as hydrogen or battery electric vehicles. They are carbon-neutral when produced from biomass or synthesized using renewable energy (see Figure below); they are liquid and thus are largely compatible with the current fuel infrastructure; they can be used in the current vehicle powertrains (ICE's) with little or no changes; they are miscible with gasoline; and enable high efficiencies and power outputs when used with advanced engine technology (highly downsized engines).


There are several ways of using alcohols as fuels. They can be used in pure form, e.g. 100% ethanol (E100) or methanol (M100), or blended with gasoline. This can be done at low concentrations, for instance in Europe up to 5% by volume of ethanol can be added to normal gasoline (E5), and up to 3% of methanol (M3). This is done both for meeting the biofuels directive of the European Union (2003/30/EC), and for increasing the knock resistance of gasoline, with the alcohol serving as octane booster. High concentration blends are also available, with E85 being the most common one.

Several million vehicles are already on the market that are able to run on E85, pure gasoline, or any mixture of both, these are termed flexible fuel vehicles (FFV's). The production of ethanol from biomass can be quite energy-intensive, especially if anhydrous ethanol is required (as for the current E85). This is ethanol containing less than 0.5 vol% water. As this requires additional processes to remove the water content of the azeotropic mixture that is obtained from distillation (about 95.5% ethanol, 4.5% water), this has an important impact on the well-to-wheel energy balance of ethanol fuel. Thus, where possible, it is beneficial if hydrous ethanol ("wet ethanol") can be used which, for instance, is the case in Brazil. Moreover, because of the polar nature of light alcohols, they attract water and some water absorption is almost unavoidable when stored.

There is a lot of practical experience with alcohol blends as fuel and a large body of literature exists on experimental work on alcohol-(blend-)fueled ICE's. However, there are only a few vehicles that actually exploit the characteristics of alcohols for increased efficiency and performance. When using "downsized" engines (engines with a smaller capacity but with the same peak power as a larger capacity engine, through the use of turbocharging) for increased part load efficiency, it is possible to take advantage of the high heat of vaporization and high knock resistance of alcohol fuels and obtain higher efficiencies and power outputs than with pure gasoline.

Phd goal

Since advanced engines incorporate a host of technologies (direct injection, variable valve timing and/or lift, exhaust gas recirculation,etc.) and thus many degrees of freedom for engine optimization, engine cycle models have become indispensable development tools.

The goal of my doctoral research is the development of an engine cycle model intended for parametric studies and optimization of state-of-the-art engines running on alcohol-gasoline-water-blends. The model should provide accurate predictions of the performance, efficiency, emission levels and knock occurrence in these engines.

Within the research group a two-zone thermodynamic model was constructed for the closed part of the engine cycle (beginning of compression till end of expansion), the GUEST code (Ghent University Engine Simulation Tool). This code was validated for spark-ignition engines running on hydrogen.

The GUEST code is composed of a number of sub-models, such as for the fuel injection and mixture formation, ignition, turbulent flame propagation, emission formation etc. Some of these sub-models are fuel-specific or need fuel-specific data. While many different models have been proposed for gasoline, and an extended property database is available, this is much less the case for alcohols and almost completely lacking for blends. From the above, it is clear that many different blends are used in practice, with varying content of ethanol, methanol, gasoline and water. Thus, there is a need for submodels valid for these fuel blends as well as fundamental property data. As determining this data for all possible blends is not feasible, and some blend properties are not easily determined from the blend constituents' properties, there is also a need for " blending laws ". My aim is to develop a model for the combustion of GEM blends in spark-ignition engines and introduce it in the GUEST code.