ABSTRACT
Reversed-phase liquid chromatography and detection with atmospheric pressure chemical ionisation tandem mass spectrometry was used for the determination of kava extracts in herbal mixtures. One percent of kava extract can be detected, corresponding to approximately 0.05-0.2 mg/g of the individual kava lactones kavain, dihydrokavain, yangonin, desmethoxyyangonin, methysticin and dihydromethysticin. Reliable quantification is obtained from concentrations of 0.25-1 mg/g, depending on the compound. At these concentration levels, the relative standard deviations were 10-14%. Validation showed good linearity and recoveries for all the kava lactones with the exception of yangonin. During method development, degradation of yangonin was observed. The degradation product was identified by nuclear magnetic resonance (NMR) as cis-yangonin. The method was applied to the analysis of commercial herbal products available in the Dutch market before and after market restrictions of kava-containing preparations. The results showed that even though 'old' products contained kava extract, the new formulations were negative on kava lactones. cis-Yangonin was also present in the herbal products.
Subject(s)
Chromatography, Liquid/methods , Food Analysis , Kava/chemistry , Mass Spectrometry/methods , Atmospheric Pressure , Calibration , Magnetic Resonance Spectroscopy , Reference Standards , Reproducibility of ResultsABSTRACT
A gas chromatographic-electron capture detection (GC-ECD) method for the analysis of deoxynivalenol (DON) in cereals was investigated. The sample was extracted with a mixture of acetonitrile-water and purified with a MycoSep #225 column. The silylation was performed with Tri-Sil-TBT reagent, followed by dilution with hexane and a washing step with buffer. By using Tri-Sil-TBT reagent no double peaks were observed for DON in the gas chromatograms, in comparison with two other silylation reagents TMSI and Tri-Sil-Z. The use of trichothecolone (TRI) as an internal standard for DON was studied in order to indicate possible problems in the derivatisation reaction. TRI proved to be a relatively good internal standard for DON in cereal samples, as well as 1,1-bis-(4-chlorophenyl)-2,2-dichloroethylene (DDE), which was used as a GC standard for ensuring the function of GC-ECD. During the study, a matrix effect was clearly observed between the cereal matrix-assisted calibration curve and the calibration curve prepared without cereal matrix. The results of spiked and reference material samples, quantified with the calibration curve prepared without and with matrix, demonstrated that the matrix affects the results. However, after recovery correction the results were comparable. The validation results demonstrated that the GC-ECD method for DON analysis in cereals is sufficiently reliable.
