(768i) Property Prediction for Diesel Fuels Based Upon Surrogates Composed of Real Components

State-of-the-art correlations for prediction of fuel properties like density, viscosity or cloud point work well for atmospheric conditions when applied to non-polar fossil fuels. However, addition of biogenic energy carriers like fatty acid methyl esters (FAME) or ethanol will increase in future. For a successful simulation of such fuels and their mixtures. It is, therefore, necessary to establish reliable property prediction methods for mixtures of fossil and biogenic components over a wide range of pressure and temperature.

Surrogate fuels are a promising concept for fuel description. In this paper, a fuel is emulated by a surrogate composed of only few (up to 30) real components. The generation of such a tailor-made surrogate is based on the simultaneous fitting of a selection of significant fuel properties like boiling-curve, cetane number, liquid density or C/H-ratio, yielding the composition of the supplied components in the surrogate.

As the surrogate is exactly characterized in terms of its composition, rigorous thermodynamics like g^E-models or equations of state can be applied for property-data prediction. Properties predicted include Cetane number, PNA-analysis, flash point, heating value, different types of boiling-curves, cloud point, the mixture vapour pressure as well as pressure and temperature dependent properties like the liquid density and viscosity. Since the surrogate consists of real components these predictions can be compared to actual measurements.

Especially in the field of engine simulations reliable property prediction over a wide range of pressure and temperature is essential. One important property is the liquid density for which the Volume-Translated-Peng-Robinson (VTPR) equation of state is evaluated. In addition to the property prediction of surrogates containing only hydrocarbons, special attention is paid to blends of fossil fuel components and FAME representing Biodiesel.