Application of infrared spectroscopy and pattern recognition techniques in the characterization of additives of petroleum industry

Abstract

Chemical additives are widely used in the oil industry; these vary according to their functionality as well as according to the technological process in which they are involved. A physical-chemical characterization has to be carried out, in order to certify the required quality specifications. The test methods used are laborious, consuming time and an important volume of samples are needed. The aim of this work is to demonstrate the feasibility of combining medium range infrared spectroscopy data together with pattern recognizing methods to characterize two additives, an antistatic used during the loading and transportation of fuels and a corrosion inhibitor. The principal component analysis method was used to explore the data, with a view to visualizing the presence of groupings of similar behavior in the study samples. For the classification process the independent soft model was used by class analogies, the different test models were constructed from a training set of representative samples. Selectivity and sensitivity obtained in the training and validation steps were equal to 1, the models did not present classification errors that its suggested our model could be employed for the quality control of the additives reported in this study.