This study was conducted to investigate on the effect of different sampling regions of palm-refined oils and fats on the 2- and 3-monochloropropanediol fatty acid esters (MCPDE) and glycidol fatty acid esters (GE) levels. The American Oil Chemists' Society (AOCS) Official Method Cd 29a-13 on the determination of MCPDE and GE in edible oils and fats by acid transesterification was successfully verified and optimised, with slight modification using 7890A Agilent GC system equipped with 5975C quadrupole detector. The determined limits of detection (LOD) for MCPDE were 0.02 mg kg-1 and 0.05 mg kg-1 for GE. The method performance has showed good recovery between 80% and 120% for all pertinent compounds with seven replicates assayed in three separate days. Round robin test with two European laboratories, i.e. Eurofins and SGS, has shown compliance results with those of the present study. Among the sampling regions, only one refinery located in the central region of Malaysia showed a significant increment of the MCPDE and GE levels after refining process. The GE level averaging at 2.5 mg kg-1 was slightly higher than that of 3-MCPDE averaging at 1.3 mg kg-1. Both esters were preferentially partitioned into the liquid phase rather than the solid phase after fractionation. However, the overall results exhibited no direct correlation between the esters content and the different sampling locations of the palm oil products in Malaysia. Analysis of total chlorine content also displayed significant variations between sampling locations which clearly show its effect on the chlorine content in the CPO samples.
The study aimed to establish the detection method for bound 3-, 2-MCPD, and glycidol using accelerated solvent extraction (ASE) and gas chromatography mass spectrometry (GC-MS). The ASE was modified for reduced solvent volume and process time to extract lipid from the chocolate spread, infant formula, potato chips, and sweetened creamer. The solvent selected for ASE was a mixture of iso-hexane and acetone at 100°C with the lipid and analyte recovery ranging from 96.9% to 98.6% and 84.1% to 107.5%, respectively. The derivatisation of analytes was adopted from the AOCS method Cd29a-13 for GC-MS analysis. The results showed that the coefficient of determination (R2) of all analytes was >0.99. The limit of detection (LOD) was 0.1 mg kg-1 expressed in lipid basis for both bound 3- and 2-MCPD and 0.2 mg kg-1 expressed in lipid basis for bound glycidol. The limit of quantitation (LOQ) was 0.3 mg kg-1 expressed in lipid basis for both bound 3- and 2-MCPD and 0.6 mg kg-1 expressed in lipid basis for bound glycidol. A blank spiked with 3-monochloropropanediols fatty acid esters (MCPDE) and 2-MCPDE (0.3, 2.1, and 7.2 mg kg-1) and glycidol esters (0.6, 4.7, and 16.6 mg kg-1) were chosen for accuracy and precision tests. The recoveries were 91.7% to 105.9%. Both repeatability and within-laboratory reproducibility of the analysis were within the acceptable level of precision ranging from 1.7% to 16%. This is the first time that a full validation procedure extending to both accuracy and precision tests has been carried out for sweetened creamer and chocolate spread. Overall, the combined protocol of ASE and AOCS Cd29a-13 was successfully validated for both solid and liquid food samples with lipid content from 10% to 30%.