Headspace gas chromatography based methodology for the analysis of aromatic substituted quaternary ammonium salts.

The analysis of quaternary ammonium salts (QAS) using GC is often performed by "in injector" pyrolysis to create volatile degradation products for quantification purposes. Besides the risk of severe system contamination, the application of this approach on aqueous samples is problematic. In this work, the sample is treated in a vial with 2,2-dimethoxypropane (DMP) under acidic catalysis. In addition to the removal of water and sample enrichment, the QAS are decomposed. As HS transfers only volatile compounds to the GC system, contamination is avoided. It was found that depending on the presence of benzyl, phenyl or methyl groups on the quaternary nitrogen; benzyl chloride, N,N-dimethylaniline or chloromethane are formed respectively in the sealed vial. All these can be used as an analytical target. A calibration curve for benzyl chloride could be derived from the pure compound. Chloromethane was generated from pure benzyldimethyldecylammonium chloride (BEDIDE), a pure QAS with benzyl and methyl groups, to construct a secondary calibration curve using a back analysis approach. It has been proven that by quantifying the formed analytical targets, the mass balance for the QAS under investigation was close to 100%. The presented procedure allows the quantification of any aromatic substituted QAS without the need for a matching reference, which is a major advantage over existing CE and LC methods The proposed methodology was validated for mouth sprays containing benzethonium chloride (BZTCl) or benzoxonium chloride (BZOCl) and for denatonium benzoate (DB) in ethylene glycol (EG) based cooling liquids. Results showed that the approach provided excellent linearity (R2≥0.999) and limits of detection around 0.01μg/vial for benzyl chloride. It was found that the reaction product of DMP and glycerol which was also present in the mouthspray and some cooling liquids, caused chromatographic interference with benzyl chloride. Treating those samples in the vial with N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) after the enrichment step removes the interference and leaves a possible pathway for the simultaneous determination of glycerol in those samples.