Analysis of Biodiesel Feedstock Using GCMS and Unsupervised Chemometric Methods

Various biodiesel feedstocks were evaluated using gas chromatography–mass spectrometry data combined with unsupervised chemometric methods of analysis. Peak areas of the fatty acid methyl esters (FAMEs) present in the biodiesel feedstocks (soybean oil, canola oil, waste grease, animal tallow, etc.) were utilized. The importance of chromatographic parameters, such as temperature program and column polarity, was examined with respect to the clustering that was observed using principal component analysis (PCA) and hierarchical cluster analysis (HCA). Biodiesels in this study clustered based on feedstock type regardless of temperature program or column type, as long as FAME isomers were resolved from one another. As such, the number and type of FAME components required to observe this clustering was investigated further. In general, the minor components in the sample did not provide improved clustering and thus did not need to be included. In addition, data from various temperature programs or column types were combined to yield similar clustering, showing potential versatility in analyzing similar samples across laboratories using different columns and column properties. Overall, we determined that (1) minor FAME components are non-essential for feedstock identification and (2) PCA and HCA clustering is based on feedstock, regardless of column selection, so long as resolution of FAME isomers is achieved.