Separation of instrumental and chemical errors in the analysis of oils by gas chromatography—A collaborative evaluation

Thirty-five analysts studied the concept that, in the gas chromatographic (GC) analysis of fatty acid composition, errors can be separated into those caused by poor chromatograph optimization and those related to inefficient conversion of triacylglycerols (TAG) to fatty acid methyl esters (FAME). A primary standard mixture of FAME was used to determine how well the participants had optimized their chromatographs. A primary standard of the equivalent TAG was used to determine total error of analysis. “Chemistry error” was calculated as the difference between the absolute errors found for the FAME and the TAG standards. Grades of analysis were computed for the FAME and TAG results and for the chemistry errors calculated from these analyses. Only four analysts achieved grades of analysis for the FAME standard that can be considered excellent or good. These four analysts used different injector/column configurations, indicating that, when properly optimized, a GC with a flame ionization detector is an extremely accurate instrument. Conversely, it is evident that there is the potential for most analysts to improve their instrumental optimization. In agreement with published information, AOCS method Ce 2-66 and AOAC method 969.33 gave low chemistry grades, but a number of analysts used modifications of these methods, and some achieved much better grades. It would appear that many of the standard methods that are in common use are capable of producing improved results, but that critical parameters need to be better specified to ensure minimization of error. The concept of separating errors into those of instrument origin and those caused by the chemical component of the total method would appear to be a useful concept for the validation of analytical methods.