Gasoline Quality Assessment Using Fast Gas Chromatography and Partial Least-Squares Regression for the Detection of Adulterated Gasoline

摘要

To find out if gasoline is adulterated, it is essential to analyze the sample's information efficiently; however, current official test methods are time consuming and costly. Much research has been conducted to supplement these difficulties using multivariable analysis and instruments such as spectrophotometers or gas chromatography (GC). However, spectrophotometers are unable to determine the chemical components, and conventional GC systems take more than 30 min to obtain sufficient data from gasoline samples. In this work, fast GC and a partial least-squares regression (PLSR) were used as analytical methods to determine research octane number (RON), aromatic compounds, methanol, and other oxygenates in under 6 min. The samples of gasoline unadulterated and adulterated with benzene, toluene, xylenes, and methanol were predicted using PLSR, which showed a good correlation between the reference values greater than 0.97. The methodology was validated, estimating specific figures of merit for quantitative multivariate analysis, showing a good value of the root-mean square error of prediction (RMSEP) and good relative error of prediction (REP %) in the range 0.0-13.0%. Therefore, these results indicate that the fast GC and PLSR model could be an alternative analytical method to effectively manage the gasoline quality.