Geochemical evaluation of crude oils recovered during water-alternating-gas injection using in silico replicas and untargeted analysis

- 346588
Poster presentation
Favoritar este trabalho
Como citar esse trabalho?
Resumo

Molecular characterization of crude oil is essential for monitoring enhanced oil recovery (EOR) processes. The chemical composition of petroleum undergoes significant changes when the reservoir is subjected to alternating injections of water and gas, resulting in different intermolecular interactions and, consequently, variations in its chemical composition. In addition to the intrinsic complexity of crude oil and the interactions within the reservoir, sampling petroleum presents significant challenges due to logistical and regulatory factors. The combination of advanced analytical techniques, a Orbitrap high-resolution mass spectrometry (ESI (±)-Orbitrap HRMS), with the generation of in-silico replicas for untargeted analysis, can provide an efficient approach for oil production monitoring. The objective of this study is to apply in-silico replica generation and untargeted data analysis to identify substances or molecular classes that differentiate crude oils water and gas injections. A set of eight crude oil samples was provided by Petrobras, consisting of one initial oil sample (t₀), three samples from the gas injection phase, and four samples from the water injection phase. The oils were collected over an eight-year monitoring period. The resulting data were processed using two different ways: homologous series analysis by Composer and chemical markers analysis using the combination of rIDIMS (R package for I ntermittent D irect I nfusion M ass S pectrometry (MS)) and multivariate analysis via MetaboAnalyst™. The constructel data matrix containing the sample groups was subjected to untargeted statistical analysis. Partial Least Squares Discriminant Analysis (PLS-DA) and heatmaps were employed to determine the compounds that differentiate samples from the water and gas injection phases, and the original oil (t₀). A comparison was performed between each injection period (gas or water) and the original oil, aiming to identify the most relevant molecular weight within each comparative group. Data processing indicated a high abundance of the N₁ class in both positive (ESI (+)) and negative (ESI (−)) ionization modes, with relative intensities exceeding 80%, along with the presence of N2 and NO classes in both modes. Overall, polar compounds showed higher abundance during the water injection phase, while their recovery decreased during the gas injection phase. Finally, it was possible to correlate statistically significant chemical markers. The integration of analytical approaches enabled the identification of marker ions capable of distinguishing different production phases during oil recovery. A clear separation was observed between samples, including replicas and non-replicas, according to the injection period.

 

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 Federal University of Rio de Janeiro
  • 2 Petrobras (Brazil)
  • 3 PETROBRAS - CENPES
  • 4 Universidade Federal do Rio de Janeiro
Eixo Temático
  • Data Mining & Artificial Intelligence in Organic Geochemistry
Palavras-chave
In-silico replicas
Orbitrap HRMS
Production monitoring
Statistically
Complex compounds