The GLC analysis of three component fat/oil mixtures in the determination of component proportions : a thesis presented in partial fulfilment of the requirements for Master in Technology degree at Massey University

Abstract

By making use of the fatty acid composition of mixture and the three-fat(s)/oil(s) mixture, obtained from the gas-liquid

chromatographic analysis, the relationship between mixture components and mixture was expressed by the general mixture equation:

(Ai - Ci) X1 +(Bi-Ci) X2 = (Mi - Ci)

where Ai is the fatty acid percent value of the first mixture component; Bi is the fatty acid percent value of the second mixture component; Ci is the fatty acid percent value of the third mixture component; Mi is the fatty acid percent value of the mixture; X1 is the fraction of the first component in the mixture; and X2 is the fraction of the second component in the mixture.

The fraction of first two mixture component can then be estimated by solving simultaneous mixture equations. This can be done either by employing the Cramer's Rule method (McCracken and Dorn, 1964) or by the Gauss-Jordan Elimination method (Dodges, 1978). The third mixture component proportion (i.e. X3) can be obtained by substituting X1 and X2 into the following equation:

X3 = 1 - X1 - X2

The F77 FREQC program, using the Gauss-Jordan Elimination Method to estimate the proportion of each mixture component and the Frequency Method to calculate the weighted mean proportion of each mixture component, gave the best results with almost all of the calculated weighted mean mixture component proportions to be within two calculated standard deviations.

The accuracy of the calculated estimations was dependent upon the

accuracy of the experimental and literature component data. In most cases, literature data did not produce as good results as those calculated from experimental data.