LAUSR

Abstract Aviation jet fuels contain over a thousand different hydrocarbons, making the prediction of their properties from chemical composition difficult. The density of a jet fuel at 15 °C is necessary for its certification. We present here an analytical approach for the determination of the density of jet fuels based on the chemical composition of the fuel determined via comprehensive two-dimensional gas chromatography with flame ionization detector (GC × GC-FID). The analysis was carried out using two-dimensional gas chromatography with electron ionization high-resolution time-of-flight mass spectrometry detection (GC × GC-TOF/MS) and flame ionization detection (GC × GC-FID). A detailed chemical composition analysis was performed on 50 samples, including petroleum-based aviation fuels and all approved alternative aviation fuel blending components. Fuel constituents were classified into seven hydrocarbon classes (n-paraffins, isoparaffins, monocycloparaffins, di- and tricycloparaffins, alkylbenzenes, cycloaromatic compounds, and alkylnaphthalenes) with the number of carbons in the range of 7–20. Several correlation algorithms and approaches were explored, including partial least squares regression (PLS) and support vector machines method (SVM), which yielded the most accurate results with mean absolute percentage errors of 0.1740% and 0.0984%, respectively. All used methods were validated utilizing uncalibrated validation samples.