Characterization of non-ideal blending in infrared spectra of gasoline surrogates

Infrared spectroscopy is a popular tool for fuel characterization. This study sheds light on the non-ideal blending of typical fuel surrogate components. We measure Attenuated Total Reflection Fourier-transform Infrared (ATR-FTIR) spectra of 230 liquid blends with varying compositions of n-heptane, iso-octane, ethanol, toluene, methyl-cyclohexane, 1-hexene, and cyclopentane. We report frequency shifts and excess absorbance for Primary Reference Fuels (PRFs) and their blends with ethanol, toluene, and multi-component mixtures. We showcase a variety of quantitative methods, starting with the simple Beer–Lambert linear fits, for the prediction of blend composition. We also provide calibration curves that account for non-linear blending effects, and use these to correct simulated spectra. Lastly, we discuss chemometric methods and their accuracy and robustness to interference and noise. Beyond concentrations, we predict functional groups, molecular weight, and branching index, which transform the mixture spectra to a unified basis.