Rapid determination and classification of crude oils by ATR-FTIR spectroscopy and chemometric methods

Mahsa Mohammadi; Mohammadreza Khanmohammadi Khorrami; Ali Vatani; Hossein Ghasemzadeh; Hamid Vatanparast; Alireza Bahramian; Afshin Fallah

doi:10.1016/j.saa.2020.118157

Rapid determination and classification of crude oils by ATR-FTIR spectroscopy and chemometric methods

Spectrochim Acta A Mol Biomol Spectrosc. 2020 May 5:232:118157. doi: 10.1016/j.saa.2020.118157. Epub 2020 Feb 16.

Authors

Mahsa Mohammadi¹, Mohammadreza Khanmohammadi Khorrami², Ali Vatani³, Hossein Ghasemzadeh¹, Hamid Vatanparast⁴, Alireza Bahramian⁵, Afshin Fallah¹

Affiliations

¹ Department of Chemistry, Faculty of Science, Imam Khomeini International University, Qazvin, Iran.
² Department of Chemistry, Faculty of Science, Imam Khomeini International University, Qazvin, Iran. Electronic address: m.khanmohammadi@SCI.ikiu.ac.ir.
³ Institute of Liquefied Natural Gas (I-LNG), School of Chemical Engineering, College of Engineering, University of Tehran, Tehran, Iran.
⁴ Petroleum Engineering Research Division, Research Institute of Petroleum Industry (RIPI), Tehran, Iran.
⁵ Institute of Petroleum Engineering, College of Engineering, University of Tehran, Tehran, Iran.

PMID: 32106028
DOI: 10.1016/j.saa.2020.118157

Abstract

Classification based on °API gravity is very important to estimate the parameters related to the extraction, purification, toxicity, and pricing of crude oils. Spectroscopy methods show some advantages over ASTM and API methods for crude oil analysis. The attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy coupled with chemometric methods has been applied as a quick and non-destructive method for crude oil analysis. In this work, a new analytical method using ATR-FTIR spectroscopy associated with chemometric methods were proposed for adressing regression and classification tasks for crude oils analysis based on °API gravity values. The designed methods are rapid, economic, and nondestructive ways in production process of oil industry. The spectral data were used for estimation of °API gravity using two approaches according to PLS-R and SVM-R algorithm, separately. The ATR-FTIR spectral data were also analyzed by classification method using the partial least squares-discriminant analysis (PLS-DA) for crude oil classification. The samples were classified into three classes based on their °API gravity values. The SVM-R model showed better results than PLS-R for °API gravity values using the F-test at 95% of confidence. The result of classification, showed about 100% accuracy and a zero classification error for calibration and prediction samples in PLS-DA algorithm.

Keywords: ATR-FTIR; Classification; Crude oil; PLS-DA; PLS-R; SVM-R; °API gravity.