Raman identification of adulteration in poly-alpha-olefin synthetic lubricant using principal component analysis and two-dimensional correlation spectroscopy

Poly-alpha-olefin (PAO) synthetic lubricating oil provides significant advantages over mineral oil. Highly discerning methods of PAO adulteration with inexpensive mineral oil is crucial to promote energy conservation and emission reduction and safeguard the rights of consumers. In this research, conventional principal compo-nent analysis (PCA), modified principal component analysis (MPCA) and two-dimensional correlation spec-troscopy (2D-COS) with temperature as perturbation factor were used for Raman identification of PAO adulteration with 5 %, 20 % and 50 % mineral oil (150 N). The influence of spectrum normalization on iden-tification was also studied. When using maximum intensity normalization, the modified principal component analysis is more efficient than the conventional PCA and can distinguish the adulterated oil from the authentic PAO even when the adulteration ratio decreases to 5 %. Comparably, the conventional PCA can distinguish the adulteration with 5 % mineral oil by using peak area normalization. These results suggest that the area normalization is prior to the maximum intensity normalization for PCA identification, as attributed to that the maximum intensity normalization might weaken the intensity difference of strongest peak among different samples. By either peak area or maximum intensity normalization, the difference between the authentic and the adulterated PAO can be detected by synchronous 2D-COS, in which both the autocorrelation peak at 2843 cm-1 and the negative cross correlation peak at (2843, 2935 cm-1) become weakened along with the increase of adulteration ratio. This work can provide important technical support to ensure the safe operation of equipment and crack down on the illegal trade of counterfeit synthetic lubricants.