Organic impurities, stable isotopes, or both : a comparison of instrumental and pattern recognition techniques for the profiling of 3,4-methylenedioxymethamphetamine

Buchanan, Hilary A. S. and Meier-Augenstein, Wolfram and Daeid, Niamh Nic and Kerr, William (2011) Organic impurities, stable isotopes, or both : a comparison of instrumental and pattern recognition techniques for the profiling of 3,4-methylenedioxymethamphetamine. Analytical Methods, 3 (10). pp. 2279-2288. ISSN 1759-9660 (https://doi.org/10.1039/c1ay05088e)

Full text not available in this repository.Request a copy

Abstract

In this study, we precisely synthesised 61 3,4-methylenedioxymethamphetamine hydrochloride (MDMA.HCl) samples. The synthetic route, reaction conditions, and batch of starting material used were carefully controlled in order to facilitate the assessment of the sample linkage abilities of: (1) GCMS organic impurity profiling using different sets of target impurities recommended from the published literature, and (2) stable isotope ratio mass spectrometry (IRMS) for delta C-13, delta N-15, delta H-2, and delta O-18-values. For GCMS analysis, we utilised the extraction parameters, instrumental conditions, and data analysis techniques recently published. In addition to this, we analysed all MDMA.HCl samples by IRMS for their C-13, N-15, H-2, and O-18 isotopic composition. The resulting data sets were subjected to hierarchical cluster analysis, principal component analysis, and discriminant analysis to identify which type of measurement (i.e. GCMS, IRMS, or both), which set of target impurities for GCMS, and which data pre-treatment method offers meaningful discrimination of the samples according to batch of starting material used, synthetic route used, or both. For our data set, discriminant analysis using a combination of the IRMS data and GCMS data ('Van Deursen' impurities pre-treated with the fourth root method) provided the most accurate discrimination of the MDMA. HCl samples. Principal component analysis had the second highest success rate, and hierarchical cluster analysis had only limited success at producing meaningful discrimination of the samples into their known groupings.