Determination of physicochemical properties and carbon-type analysis of base oils using mid-IR spectroscopy and partial least-squares regression analysis

It has been well recognized that the chemical composition of the base oil is primarily responsible for its physicochemical properties such as viscosity, viscosity index (VI) and pour point etc., which in turn influence the performance of the finished formulation. Hence, it is very essential to determine the chemical composition of the base oil to understand the physicochemical properties. The paper highlights an experimental and computational protocol for the simultaneous determination of physicochemical properties (viscosity index and pour point) and carbon type analysis (C-A, C-P, C-N, and isoparaffin carbon content, I-P) of base oils from their IR spectral features. IR spectra of 60 base oils of different origin and processing schemes are recorded using the horizontal attenuated total reflectance accessory with ZnSe windows. Partial least-squares (PLS) technique has been applied to construct the mathematical models which correlate the IR spectral features with the experimentally determined values (using NMR and standard ASTM tests). The regression coefficients are found to be in the range 0.80-0.95.