Quantitative gas chromatographic analysis of lipids: Comparison of gas density balance and flame ionization detector

Abstract

Accuracy of available detectors for gas chromatography is a subject of continuing research in analytical chemistry. The quantitative deficiency of the flame ionization detector, as well as of other detectors, has been widely recognized, and empirical correction factors have been required. By contrast, the gas density balance, the forgotten ideal detector, should not require calibration. A gas density balance, now available in a commercial chromatograph, and a flame ionization detector were compared for quantitative analyses of lipids. Wt percents of known methyl ester mixtures were determined, as well as mole percents of aldehyde fragments from certain ozonized octadecenoate isomers. Percentages were calculated from area response without correction factors for the gas density balance and with correction factors, based upon the number of ionizable carbon-atoms, for the flame ionization detector. Accuracy, as measured by percentage deviation from either known or theoretical values, was better for gas density balance data than for flame ionization detector data. Aldehyde and aldehydic ester fragments formed by reductive ozonolysis of octadecenoate isomers from partially hydrogenated methyl linolenate also were determined with each detector. Theoretically, ozonolysis of these monoenes should yield an aldehyde and an aldehydic ester in equal mole percents. Experimentally, the average of the ratios of aldehyde to aldehydic ester from each of the Δ5-Δ13 monoenes was 1.29 for the FID data (corrected) and 1.01 for GDB data (uncorrected). This difference in averages approaches significance at the 95% confidence level. For the Δ14 and Δ15 monoenes from which C₄ and C₃ aldehydes are formed, ionizable carbon-atom corrections proved even less adequate.