Abstract
Fluorescence spectrometry, infrared spectrometry, liquid chromatography with ultraviolet detection, and gas chromatography with flame ionization (GC/FID) or mass spectrometric (GC/MS) detection are investigated for possible use in oil-spill identification. The discriminating power of these five procedures is determined for 23 oils from four geographic regions by application of the ANOVA, Wilcoxin-Mann-Whitney, chi-square and SIMCA methods. Data analysis commenced with 270 variables from the chemical procedures for each oil; 190 of these were from GC/MS. The multivariate SIMCA method is shown to increase the discriminating power of the data compared to the univariate methods. The ranking indicates that fluorescence spectrometry, selected peak-area ratios from GC/FID, sulphur, nickel and vanadium contents, and selected GC/MS peak ratios are useful for classification of oils.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
