Automated Identification of Crude Oil Molecular Markers Combined with Expert Screening: A Compositional Change Evaluation after Enhanced Oil Recovery

- 344364
Flash Presentation
Favoritar este trabalho
Como citar esse trabalho?
Resumo

The application of comprehensive two-dimensional gas chromatography coupled with time-of-flight mass spectrometry (GC×GC-TOFMS) for characterization of light hydrocarbons in complex crude oil matrices generates extensive datasets that require meticulous, and time-consuming processing. Identifying key compounds to monitor crude oil compositional variations after enhanced oil recovery (EOR) processes depends on reliable and efficient data processing methods. Therefore, an expert manual processing approach and an automated target method were combined to optimize the identification of key substances and evaluate post-gas injection changes in oil composition. A total of 38 Brazilian crude oil samples from time-zero (t0: 25 samples from the well-test phase, prior to production) and time-one (t1: 13 samples from production phase) were analyzed by GC×GC-TOFMS. Substances across C5–C15 subclasses, including n-alkanes, branched alkanes, monocyclics, bicyclics, and aromatics were individually identified and semi-quantified using a peak-table approach. Subsequently, 128 features were implemented in an in-house automated target method to optimize the identification of key discriminant substances. Multivariate Principal Component Analysis (PCA) and Partial Least Squares Discriminant Analysis (PLS-DA) were performed to investigate the t0 versus t1 datasets. Samples collected at the same time point from identical or adjacent wells showed no significant differences, indicating a homogeneous reservoir composition. In the unsupervised approach, the PCA scores plot showed a clear discrimination between t0 and t1 samples, with the first two components retaining 63.3% of the total variance. PLS-DA further supported this distinction, explaining 62.6% of the variance within the first two components. The combination of expert curation and automated compound identification enabled the reliable determination of markers associated with EOR-induced compositional changes. Monocyclic C7–C11 alkanes, bicyclic C10–C11 alkanes, alongside C7–C9 n-alkanes and branched alkanes were identified as key indicators of gas-injection effects. The results support monitoring light hydrocarbons in the C7–C11 range to assess chemical enrichment in Brazilian crude oils following EOR processes. Additionally, miscible gas injection was associated with an enrichment in light and intermediate molecular weight hydrocarbons, consistent with recent findings.

Compartilhe suas ideias ou dúvidas com os autores!

Sabia que o maior estímulo no desenvolvimento científico e cultural é a curiosidade? Deixe seus questionamentos ou sugestões para o autor!

Faça login para interagir

Tem uma dúvida ou sugestão? Compartilhe seu feedback com os autores!

Instituições
  • 1 Universidade Federal do Rio de Janeiro
  • 2 Petrobras (Brazil)
  • 3 PETROBRAS - CENPES
  • 4 Federal University of Rio de Janeiro
Eixo Temático
  • Reservoir Geochemistry
Palavras-chave
Comprehensive two-dimensional gas chromatography
Light hydrocarbons
Enhanced oil recovery
Reservoir geochemistry
Crude oil composition