A Novel Impurity-Profiling Workflow with the Combination of Flash-Chromatography, UHPLC-MS, and Multivariate Data...

In this paper we present a new, versatile workflow for a synthesis impurity profiling concept, using the combination of flash chromatography (F-LC), liquid chromatography coupled to mass spectrometry (LC-MS), and multivariate data analysis. For three highly pure, structurally different synthetic cannabinoids, we demonstrate that via F-LC more than 99% of the main component (API) can be removed from a sample to enrich present impurities and yield combined fractions of targeted synthesis impurities with reproducible chromatographic signatures via LC-MS. The maximum overall relative standard deviation (RSD) of the complete experimental procedure for isolation and measurement of the impurity profiles (FL-C + LC-MS) was found to be 13.8% on average. The impurity signatures of 40 1 kg samples of MDMB-CHMICA (methyl ( S)-2-(1-(cyclohexylmethyl)-1 H-indole-3-carboxamido)-3,3-dimethylbutanoate) from one large seizure by Luxembourg customs were assessed via UHPLC-MS and compared via principle component analysis (PCA) to possibly discriminate between individual synthesis pathways or production batches and to deduce batch sizes. Three of these 40 samples could be identified as outliers, i. a., as a result of a highly abundant impurity with m/ z 498, isolated via F-LC and identified as methyl 2-(2-(1-(cyclohexylmethyl)-1 H-indole-3-carboxamido)-3,3-dimethylbutanamido)-3,3-dimethylbutanoate, most probably manufactured with a varying synthesis pathway. The remaining 37 samples were subdivided via PCA and hierarchical cluster analysis into five clusters between five and ten samples, representing a maximum possible batch size of 10 kg of pure MDMB-CHMICA. Furthermore, the profiling concept was successfully applied to self-produced and seized "spice-products" to extract impurity profiles of MDMB-CHMICA without any ion suppression or chemical interference.