Assessing secondary alteration processes of Brazilian crude oils using gas chromatography and chemometrics

Biodegradation and oil mixing are common processes in Brazilian crude oils and can occur simultaneously. In general, the geochemical investigation of these processes are traditionally assessed through a detailed study of biomarkers analyzed by chromatographic analyses. In some cases, contrasting geochemical parameters may suggest concomitant alterations, although more in-deep evaluation of biodegradation extent is necessary. In this study, 31 crude oil samples (Cretaceous Brazilian Basins, Campos, Espírito Santo, and Sergipe-Alagoas Basins) were evaluated using gas chromatography with flame ionization detection (GC–FID) and gas chromatography-mass spectrometry (GC–MS). Whole oil analysis (n = 31) and saturate fractions (n = 93) were analyzed by GC using internal standard (C₂₄D₅₀). Total n-alkane, C₂₉ 17α,21β(H)-25-norhopane (25NH), C₃₀ 17α,21β(H)-hopane (H30) concentrations were calculated, and diagnostic ratios, including peak/hump (P/H) and 25NH/H30 were determined. Additionally, hierarchical cluster analysis (HCA) and principal component analysis (PCA) were applied to the full chromatograms acquired by GC–FID and GC–MS analyses. Total n-alkane concentrations were high for light crude oil (9.24×10⁴ ng/mg for GC–FID and 1.16×10⁵ ng/mg for GC–MS) and the intermediate oils (2.02×10⁴ to 4.23×10⁴ ng/mg for GC–FID and 1.31×10⁴ to 5.63×10⁴ ng/mg for GC–MS), whereas biodegraded oils exhibited lower concentrations (15.3 to 1.79×10³ ng/mg for GC–FID and 1.42 to 2.00×10³ ng/mg for GC–MS). In contrast, C₃₀ 17α,21β(H)-hopane concentration was not effective in differentiating the oil groups, showing fluctuations in values ​​for both biodegraded oils (6.24 to 483 ng/mg for GC–FID and 25.7 to 661 ng/mg for GC–MS) and intermediate oils (106 to 359 ng/mg for GC–FID and 99.5 to 466 ng/mg for GC–MS). The geochemical data (P/H and 25NH/H30 ratios) were used for the discrimination between groups of non-biodegraded, intermediate (including mixtures), and biodegraded oils, showing significant differences among the groups (p and adjusted p < 0.05). Multivariate analyses using HCA and PCA separated the biodegraded oils from the other samples, and highlighted the compositional differences between the groups, possibly associated with oil mixing processes. The distribution of samples in the PCA score (PC1 versus PC2, and PC1 versus PC3) plots from GC–FID (85.97% of total variance explained) and GC–MS (83.80 of total variance explained) data showed a direct relationship with the weight of n-alkanes and the 17α(H),21β(H)-25,30-bisnorhopane (25,30BNH). The combination of analytical tools enabled the characterization of the oils and the identification of groups that had undergone secondary alteration processes, such as biodegradation and mixing.