Abstract
In this work, the classification of three fuel types (gasoline, diesel, and kerosene) using H-1 nuclear magnetic resonance (NMR) spectroscopy and principal component analysis (PCA), and cluster analysis was achieved successfully. Nine samples collected over 9 months period of each of the above fuels were collected and studied; the total number of all neat samples of fuels was 27 samples. The resulted PCA score model was used as a calibration model for detecting changes in gasoline composition. To imitate adulterated gasoline with diesel, kerosene gasoline was mixed with less expensive fuels, such as diesel and kerosene, with different volume fractions (5, 10, and 20 %). The same protocol used above for the classification has also been used successfully for the qualitative determination of 18 adulterated gasoline samples. The qualitative detection of diesel adulteration with kerosene of the same previous volumetric ratios was also achieved for nine adulterated diesel samples using the H-1 NMR and PCA.