Application of Clustering Methods for Interpretation of Petroleum Spectra from Negative-Mode ESI FT-ICR MS

Abstract

This study was performed to develop analytical methods to better understand the properties and reactivity of petroleum, which is a highly complex organic mixture, using high-resolution mass spectrometry and statistical analysis Ten crude oil samples were analyzed using negative-mode electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry (ESI FT-ICR MS) Clustering methods, including principle component analysis (PCA), hierarchical clustering analysis (HCA), and k-means clustering, were used to comparatively Interpret the spectra All the methods were consistent and showed that oxygen and sulfur-containing heteroatom species played important roles in clustering samples or peaks The oxygen-containing samples had higher acidity than the other samples, and the clustering results were linked to properties of the crude oils This study demonstrated that clustering methods provide a simple and effective way to interpret complex petroleomic data