Evaluation of saturated and aromatic hydrocarbons oil-oil maturity correlation parameters (SE Pannonian Basin, Serbia)

Abstract

Twenty three crude oils from the Serbian part of the Pannonian Basin (14 from the Vojvodina Province and 9 from the Drmno Depression) were investigated, aimed at an evaluation of oil-oil maturity correlation parameters based on the distribution and abundance of saturated biomarkers and alkylarene constituents. Factor and cluster analyses were used for this purpose. Factor analyses using varimax rotation were first run separately, i.e., of maturity parameters based on the abundance of (a) n-alkanes and isoprenoids, (b) steranes and triterpanes, (c) alkylnaphthalenes, and (d) alkylphenanthrenes. These analyses yielded 9 important "maturity factors". Eight of them, showing higher than 30 % of variance, were further involved in another factor analysis, as well as in cluster analysis using the Ward method. In this way, all maturity parameters based on saturated biomarkers and alkylarenes were evaluated and ranged, considering the fact that the observed factors represented their linear combinations. The results showed that in the correlation of crude oils from the Serbian part of the Pannonian Basin, the most important were maturity parameters based on isomerization reactions involving one methyl group in thermodynamically less stable ?-methylnaphthalenes, ethylnaphthalenes, dimethylnaphthalenes and methylphenanthrenes, and their change into more stable isomers with the methyl group in the ?-position in the aromatic ring. Processes constituting high loadings factor 2 and factor 3 parameters were also defined. Hierarchy between the "factors" and parameters were controlled, and approved, by cluster analysis using the Ward method. Finally, the investigated crude oils were correlated by factor and cluster analyses, using all the important "maturity factors". Differences in maturity were observed between the Vojvodina and Drmno Depression crude oils, as well as between oils originating from South Banat, North Banat and the Velebit oil field (Vojvodina locality).

Application of Rotated PCA Models to Facilitate Interpretation of Metabolite Profiles: Commercial Preparations of St. John’s Wort

“…Thirteen preparations were formulated as tablets (preparations 1 -4, 11, 12, 14, 16, 17, 21 -24), and the remaining as capsules (preparations 5 -10, 13, 15, 18 -20). Preparations 1 -10, 23, and 24 originated from Europe, preparations 11 and 17 from Asia, preparations 12,13,15,16,18, and 20 -22 from North America and preparations 14 and 19 from Africa. For one of the brands, two different batches were obtained (preparations 7 and 8).…”

“…Rotating the PCA model towards simplicity of scores rather than simplicity of loadings is equally feasible, as follows from the symmetry of the PCA model. However, most studies published so far have used rotation of the PCA model for obtaining simpler loadings [15], [16], [17]. Rotation of scores can be particularly useful when a certain clustering is expected among the samples, as shown in the following paragraphs.…”

“…If γ = 0 the equation becomes the quartimax criterion and if γ = 1 it becomes the varimax criterion. The varimax criterion is by far the most often applied method among the orthogonal rotations [13], [14], [15], [16]. Maximizing the varimax criterion provides a solution where the variance of the squared loading elements is maximized.…”

“…The use of rotations in multivariate data analysis is not a new approach, and it has been used for decades in some areas, e. g., in psychometrics [12]. However, in natural sciences in general and metabolomics and metabonomics in particular the use of rotations of PCA models is far more limited, although a few recent examples can be found [13], [14], [15], [16]. Traditionally, the use of rotations in PCA modelling has not been strictly needed, because multivariate modelling has usually been performed on fairly simple data.…”

Discriminating cereal and pseudocereal species using a binary system of GC/MS data: A pattern recognition approach

“…With respect to food fraud, unsupervised multivariate chemometrics approaches may be used as a powerful data-reduction tool used qualitatively for grouping or classifying unknown samples with similar characteristics. 53,54 Although systems based on GC-MS combined with various multivariate classification and pattern recognition methods have been extensively used to measure lipids and chemical composition in general (e.g., protein, moisture, oil) of different cereal and pseudocereal species, to the best of the authors' knowledge, no studies have been conducted reflecting the use of this method for varietal discrimination and traceability of cereals and pseudocereals.…”