Assessing the reliability of the forensic technique for the identification study of motor gasoline using gas-liquid chromatography

Abstract

A new approach to solving the problems of forensic examination of petroleum products, fuels and lubricants associated with establishing whether motor gasoline belongs to a common/different identification set is proposed. A methodology for the identification of gasoline using quantitative results of chromatographic determination of controlled parameters and subsequent evaluation of the results in pairwise compared samples according to the established criteria. The following parameters were selected as controlled indicators: values of research octane number (RON), concentrations of hydrocarbon groups (paraffins, isoparaffins, arenes, naphthenes, olefins) and oxygenates. Their determination was carried out using standardized methods. We used the hardware and software complex of a Chromatek-Crystal 5000 series, including the Chromatek-DHA data processing program. Estimated criteria or decision- making rules regarding the issues posed to an expert were determined to classify gasoline under one product name (generic set), one production technology (group set) or a common source of origin (one production batch, storage tank, etc.). The reliability of the method was assessed using a validation procedure consisted of three stages. We used a collection of AI-92 motor gasoline, selected at gas stations of four oil companies during six months of 2022 in various districts of Moscow. At the first stage, the performer analyzed 12 aliquots of each of four gasoline samples (samples that previously belonged to the common volume) at different times. It was established that the quality indicators of the proposed methodology (standard deviation of the repeatability and reproducibility, expanded uncertainty, limits of the repeatability and reproducibility) for each of the determined controlled indicators did not exceed the permissible errors established by standardized methods. At the second stage of the experiment, the performer combined 12 test samples into four groups from each manufacturer (one production technology) with three samples of different production times (different batches). By comparing the discrepancies in the measured values of the same controlled indicators between pairs of samples within groups and the discrepancy in average indicators between pairs of groups, the contractor identified gasolines manufactured using the same technology; gasolines produced by the same technology, but at different times (different batches) and, with a probability of 95%, gasolines having a common source of origin (previously belonging to a common volume). The conclusions of the validation study matched the original data on the samples, which confirmed the correctness of the developed comparison criteria. At the third stage (blind test), the performer examined seven gasoline samples of, information about the composition and properties of which was not provided to him. The results of the blind test were considered satisfactory. Thus, the validation results indicate the suitability of the methodology for solving forensic identification problems in relation to motor gasoline and the competence of the performer sufficient for implementation of the methodology.