Trimethylsilyl Derivatization/Gas Chromatography as a Method to Determine the Free Fatty Acid Content of Vegetable Oils

• Published in: Journal of the American Oil Chemists' Society Vol. 84; no. 8; pp. 701 - 708
• Main Authors: Wan, Peter J, Dowd, Michael K, Thomas, Audrey E, Butler, Boyce H
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Berlin/Heidelberg : Springer-Verlag 01.08.2007; Springer-Verlag; Springer; Springer Nature B.V
• Subjects: Acidity; Biological and medical sciences; canola oil; chemical analysis; Chemicals; Chromatography; Comparative studies; cottonseed oil; Extraction processes; Fat industries; Fatty acids; food analysis; Food industries; Food science; free fatty acids; Fundamental and applied biological sciences. Psychology; gas chromatography; lipid content; Measurement techniques; Molecular weight; organosilicon compounds; palm oils; peanut oil; rapeseed oil; rice bran oil; soybean oil; sunflower oil; trimethylsilyl; vegetable oil; Vegetable oils; Oils and fats industry; Gas chromatography; Vegetable oil; Free fatty acid; Method; Oil content
• ISSN: 0003-021X; 1558-9331
• DOI: 10.1007/s11746-007-1101-1

Thirty-one samples of vegetable oils were analyzed for free fatty acid (FFA) concentration by titration against sodium hydroxide and by trimethylsilyl (TMS) derivatization followed by gas chromatography (GC). In preliminary experiments, two silylation chemistries and three GC stationary phases were tested for TMS fatty acid ester formation and separation. No ideal combination of conditions was identified; however, hexamethyldisilazane with an acid catalyst and a non-polar J & W DB-5 column were chosen for comparison of the two approaches. Over the range of FFA values studied (0.04-12%), the results from the two methods were highly correlated (R > +0.996) and were generally in good agreement. However, values from the chromatographic method were slightly lower than values obtained by titration for the crude oil samples (which had higher FFA levels). For oils with >0.4% FFA, the GC approach was slightly less reproducible (average coefficient of variance of ~3%) compared with the titration approach (average coefficient of variance of ~1.4%). For oils with <0.4% FFA, the coefficients of variance were higher (8-9%) and comparable between the methods.