partial least squares-discriminant analysis (pls-da) for classification of crude oil samplesbased on sulfur content using atr-ftir spectroscopy

Crude oil is a mixture of hydrocarbons, predominantly hydrogen and carbon, with varying amounts of nitrogen, sulfur, and oxygen. sulfur, one of the main contaminants in crude oil, is determined for several reasons. importantly, sulfur compounds are present in different types of chemical structures, such as thiophenes, sulfides, benzo thiophenes, and dibenzothiophene. crude oils are classified based on sulfur content into sweet and sour oil. in this study, a new and simple approach for classifying the sulfur content present in crude oils is proposed, using attenuated total reflection fourier transform infrared (atr-ftir) spectroscopy associated with chemometric methods. the feasibility of atr-ftir spectroscopy associated with chemometric models was evaluated for sulfur classification in crude oil samples. the sulfur content was determined using x-ray spectroscopy in the crude oil samples, and the resulting values were used as input data to model the atr-ftir spectroscopy. additionally, for classifying crude oil samples into sweet and sour categories based on sulfur content, the application of atr-ftir spectroscopy in combination with the partial least squares-discriminant analysis (pls-da) and support vector machine-discriminant analysis (svm-da) algorithms was successfully performed. two sets of samples, 70 and 30, were considered for the calibration and prediction sets, respectively, in the classification models. the samples were classified into two classes, sweet and sour crude oil, according to sulfur content. the classification results showed an accuracy of 96% and a classification error of 0.0384 for the prediction set in the pls-da algorithm. these results indicate that atr-ftir spectroscopy associated with classification models is a rapid and reliable approach for quantifying the sulfur content in crude oils. the interest in classifying sulfur content in crude oil could serve as a model for the development of crude oil analysis in the oil industry.