Monitoring of biodiesel content and adulterant presence in methyl and ethyl biodiesels of jatropha in blends with mineral diesel using MIR spectrometry and multivariate control charts

Abstract

In this work, fast and efficient analytical methodologies were developed for quality monitoring of the content of jatropha methyl and ethyl biodiesels in blends with mineral diesel, and to monitor the presence of adulterants using mid-infrared (MIR) spectrometry allied to the method of multivariate control charts based on net analyte signal (NAS). The methodology of multivariate control charts consists of dividing the spectral signal of calibration samples into three vectors, namely NAS, interference and residual, generating three charts. The NAS chart monitors the quality of the analyte of interest in blends (biodiesel content, in this case), while the interference chart monitors the diesel quality, i.e., monitors the presence of adulterants or contaminants in the binary mixtures. The residual chart monitors the information or variations not modeled by the NAS and interference vectors. The validation of control charts in both models was done in two ways: (i) in relation to the biodiesel content in the biodiesel/diesel blends and (ii) in relation to the presence of contaminants in blends. In relation to the biodiesel content, sets of samples with biodiesel content in and out of control of the quality specifications were prepared for validation. In relation to the presence of adulterants, validation was performed by applying other sets of samples adulterated by the partial substitution of biodiesel in the blends by soybean oil and used frying oil (UFO) in the range of 0–30%, and by partial substitution of diesel by residual automotive lubricant oil (RALO) and gasoline, also in the range of 0–30%. The results of the application of multivariate control charts demonstrated satisfactory monitoring for the identification of samples of B7 blends in or out of quality specifications and identification of the presence of adulterants in blends. That is, the proposed methodologies correctly identified samples as in- or out-of-control. Therefore, the methodologies were shown to be effective for use by the supervisory board of such blends of these biofuels.