A single analytical procedure for the simultaneous and confirmatory determination of melamine and related compounds in various food matrices by isotope dilution gas chromatography-mass spectrometry (ID-GC-MS)

Abstract

An analytical procedure for the simultaneous and confirmatory determination of melamine and related compounds, viz. ammeline, ammelide and cyanuric acid, in various food matrices (including milk, milk powder, cereal, biscuit and porcine muscle) using isotope dilution gas chromatography-mass spectrometry (ID-GC-MS) has been developed. In brief, the target analytes were extracted from a homogenised sample (which had been spiked quantitatively with the corresponding stable isotopically labelled analogues) by sonication using 0.01 M NaOH–acetonitrile (1 : 1). The extract then underwent solvent partitioning using diethyl ether followed by dichloromethane, further clean-up with mixed-mode ion exchange solid phase extraction (SPE) column chromatography, and derivatisation using N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) with 1% trimethylchlorosilane (TMCS) reagent in the presence of a base (pyridine) before being analysed by GC-MS in selected ion monitoring (SIM) mode. Repeatability studies showed that the established procedure for various food matrices was reliable and reproducible with a method precision within the range of 0.3–7.5% RSD. Mean recoveries of fortified samples in different food matrices were within the range of 83.7–108.3%. The reported method has been practically demonstrated to be applicable to concentration levels down to 0.2 mg kg⁻¹ (for melamine, ammeline and ammelide) and 0.5 mg kg⁻¹ (for cyanuric acid) for various food types, which can cope with the maximum tolerable limits set worldwide for melamine and related compounds. The limits of quantification (LOQ) for melamine, ammeline, ammelide and cyanuric acid of this method were found to be 0.10 mg kg⁻¹, 0.09 mg kg⁻¹, 0.10 mg kg⁻¹, and 0.25 mg kg⁻¹, respectively. The relative measurement uncertainties of the method for the target analytes with a coverage factor of 2 (at an approximately 95% level of confidence) were estimated to be in the range of 7.2–25%. This method can serve as an effective screening procedure and also as an alternative method for confirmation or for litigation purposes.