Get access

Determination of bound 2,3-epoxy-1-propanol (glycidol) and bound monochloropropanediol (MCPD) in refined oils



A new method for the parallel determination of bound 2,3-epoxy-1-propanol (glycidol) and 2- and 3-chloropropanediol (2- and 3-MCPD) in oil matrices is presented. It is based on an improved alkaline catalysed release of MCPD and glycidol, followed by a transformation of glycidol to monobromopropanediol (MBPD), derivatisation with phenylboronic acid (PBA) and analysis by GC–MS. Quantification was performed using isotopic labelled standards. Method validation was carried out for the complete method procedure using glycidyl stearate and 3-chloropropanediol-1,2-bis-palmitoyl ester as reference compounds. Linearity was verified (r2>0.999) within the concentration range from 0.1 to 5.7 mg/kg. The limit of detection (LOD) for bound glycidol was 0.025 mg/kg in absence of 3-MCPD. Due to a lack of certified reference material a validation of bound 2-MCPD analysis was not feasible, but an indirect approach yielded a semi-quantitative estimation. The method was applied to the analysis of several different oils. None of the analytes was detected in virgin or crude oils. In contrast bound MCPD and bound glycidol was detected ubiquitously in refined oils. The determined concentrations of 3-MCPD and glycidol in the tested samples varied in 2 orders and 4 orders of magnitude, respectively. Bound 2-MCPD was estimated to occur commonly in refined oils.

Practical application: The presented application is advantageous to other methods that are either not suitable for the parallel determination of bound glycidol and bound 2- and 3-MCPD within on step or are based on an estimation of bound glycidol contents by calculation. In comparison to those methods based on the direct determination of the single mother compounds this approach to convert the eventually unclear assembly of different MCPD- and glycidyl derivatives into the basic analytes glycidol and 2- and 3-MCPD simplifies identification and quantification, enhances the analytical sensitivity and lowers the risk of underestimation. Furthermore, the investigation of basic transformations that may occur during sample preparation can be helpful for a better understanding of the complex chemistry in MCPD and glycidol analysis. Results from analysis of different oils and fats can be used for an estimation of the occurrence and distribution of bound MCPD and bound glycidol.