![Sample our Physical Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110819/TOC-Subject-Sample-2021-Physical-Sciences.jpg"})
Abstract
A methodology was developed to monitor the adulteration of the B10 blend of diesel and crambe biodiesel using proton nuclear magnetic resonance (1H NMR) spectroscopy combined with data driven soft independent modeling of class analogy (DD-SIMCA) model. The training was performed only with samples of the target class (B10) while the validation was performed with a test set consisting of new samples of the target class (B10) and samples of B10 adulterated with crambe oil, used frying oil, and residual automotive lubricating oil. The efficiency of this methodology was characterized based on the sensitivity parameters for the training set and specificity for the test set, in which a value of 1 was obtained for both parameters. This sensitivity value for the training set indicates that no target class samples were classified as extreme or outliers. The specificity for the test set shows that all samples in the test set were correctly classified into their respective classes, demonstrating the high efficiency of the DD-SIMCA model in monitoring adulterants in B10 mixture of diesel and crambe biodiesel. The DD-SIMCA model is simpler to construct than the multivariate control chart and the partial least squares discriminant analysis (PLS-DA) because its development does not require prior information about the adulterants. The excellent obtained results in the application of this model suggest that this analytical methodology is efficient, feasible and suitable for use by inspection agencies to characterize the quality of this fuel.