Effect of sampling modes on the discrimination of commercial gasolines using direct analysis in real time mass spectrometry

Abstract

The identification and classification of gasolines are crucial in a forensic context because direct evidences are provided on gasoline adulteration or other related offenses. However, it is attractive but challenging to develop a facile and efficient strategy to discriminate different brands of commercial gasolines. Herein we propose a means to realize this goal by combining direct analysis in real time mass spectrometry (DART-MS) and principle component analysis (PCA). In this process, different experimental parameters, including the ratio between gasoline and methanol, heating temperature of helium, sample moving speed and helium gas pressure, were found playing vital roles to determine the analysis sensitivity of DART-MS. By evaluating the performance with three characteristic peaks (C8H12N+ at m/z 122, C17H17+ at m/z 221 and C16H24N+ at m/z 230) from gasolines, abundant fingerprinting spectra of gasolines were achieved for later favorable brand classification. Furthermore, sampling mode was found crucial to the distinguishing ability of different brands of gasolines for the first time. Two frequently used sampling modes (sample card sampling and glass rod sampling) were compared to discriminate 6 brands of standard gasolines and 28 brands of the Research Octane Number (RON) 92 and 95 (namely 92# and 95#) commercial gasolines. In contrast to glass rod sampling, a full separation of different brands of gasolines was more favorable to be generated by using sample card sampling. Though limited gasoline samples were performed, results generated from in this study indicated that it was feasible and promising to identify and classify different brands of commercial gasolines using DART-MS and PCA analysis.