Classification of petroleum origin and integrity by FTIR

A method was developed for classifying petroleum into four types according to their origin and integrity: marine, non-marine, degraded and thermally transformed. The method was based on the relative proportions of functional groups measured by FTIR spectroscopy. Twenty-one petroleum samples were used for model calibration and six for validation, which were pre-classified. Aliphatic functional group absorption maxima were used at 2919, 2856, 1457 and 1376 cm−1, as well as at 1700  and 811 cm−1 for aromatic groups, and at 1152 and 1030 cm−1 for polar groups (sulphonyl and ether, respectively). Using hierarchical cluster and discriminate analyses, three canonic functions were obtained that classified the samples according to their origin and integrity, with a prediction efficiency of 100%, for both calibration and validation. The canonic functions derived from the discriminate analysis had correlation coefficients of 0.994, 0.900, and 0.867, among the variables studied and the classification factor. The method is efficient for the proposed classifications and can be useful for a range of applications.