Interpretation of chromatographic and mass spectrometric data from analyses of fatty acid methyl esters Application of multivariate methods

Abstract

The purpose of this work has been to develop tools for fatty acid identification, with special emphasis on poly-unsaturated fatty acids and trans fatty acids. The work is based on application of gas chromatography with mass spectrometric detection and the fatty acids are analysed as fatty acid methyl esters (FAME). Various multivariate analytical techniques are applied as tools for interpretation of both chromatographic and spectral information. The ten papers are divided into five subsections with the following topics: Identification of FAME from shifts in retention indices: Papers I and II describe methods for partial identification of fatty acid structure from shifts in retention indices. These methods will basically provide information on the polarity (number of double bonds) and the chain lengths of the fatty acids. The procedure is not restricted to any particular class of fatty acids and may also be used to exclude compounds that are not fatty acids. Two-dimensional fatty acid retention indices (2D-FARI) are introduced in Paper II. Application to trans isomers: Papers III and IV are about trans isomers of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Paper III provides data about chromatographic properties of the trans isomers of EPA and DHA, which can be applied both for identification purposes and for optimisation of chromatographic parameters. A practical application of the data is shown in Paper IV. Prediction of equivalent chain lengths: Papers V and VI describe methods for prediction of equivalent chain lengths (ECL), which is the retention indices commonly applied for fatty acids. Paper V is restricted to polyunsaturated fatty acids with methylene-interrupted double bond systems. Paper VI is not restricted to a particular class of fatty acids, but 2D-FARI data must be available. Thus, the two procedures are complementary methods for ECL predictions. Identification of FAME from mass spectra: Paper VII and VIII are about multivariate analysis of mass spectra of FAME. In Paper VII it is shown that trans isomerism in certain positions has significant influence on the mass spectra of polyunsaturated fatty acids. In Paper VIIi it is shown that the number of double bonds in polyunsaturated fatty acids can be determined from selected ions in the mass spectra. Deconvolution of overlapping chromatographic peaks: Papers IX and X are about techniques for deconvolution of overlapping FAME peaks. Transformations for reduction of noise in the mass spectra were introduced in Paper IX. The transformations and the information from Paper VII about differences in the mass spectra caused by trans geometry were utilised for deconcvolution of overlapping chromatographic peaks of trans isomers in Paper X. Although the focus in this paper is basically on the quantification of the isomers, deconvolution of overlapping peaks has relevance also for the identification of the analytes because spectra and accurate retention times of the compounds are also provided.