Fully automated determination of the sterol composition and total content in edible oils and fats by online liquid chromatography–gas chromatography–flame ionization detection

Abstract

Sterol analysis of edible oils and fats is important in authenticity control. The gas chromatographic determination of the sterol distribution and total content is described by ISO norm 12228. Extraction, purification, and detection of the sterols are time-consuming and error-prone. Collaborative trials prove this regularly. Purification by thin-layer chromatography (TLC) and robust GC determination of all mentioned sterols is not straightforward. Therefore, a fully automated LC-GC-FID method was developed to facilitate the determination of sterols. The only manual step left was to weigh the sample into an autosampler vial. Saponification and extraction were performed by an autosampler while purification, separation, and detection were accomplished by online coupled normal-phase LC-GC-FID. Interlacing of sample preparation and analysis allowed an average sample throughput of one sample per hour. The obtained quantitative results were fully comparable with the ISO method with one apparent exception. In the case of sunflower oils, an additional unknown sterol was detected generally missed by ISO 12228. The reason was found in the omission of sterol silylation before subjection to GC-FID. The derivatization reaction changed the retention time and hid this compound behind a major sterol. The compound could be identified as 14-methyl fecosterol. Its structure was elucidated by GC–MS and ensured by HPLC and GC retention times. Finally, validation of the designed method confirmed its suitability for routine environments.