Extraction optimization and pixel-based chemometric analysis of semi-volatile organic compounds in groundwater

Semi-volatile organic compounds (semi-VOCs) are found in complex mixtures, and at low concentrations in groundwater. Chemical fingerprint analysis of groundwater is therefore challenging, as it is necessary to obtain high enrichment factors for compounds with a wide range of properties. In this study, we tested the combination of solid phase extraction (SPE) with dispersive liquid-liquid micro extraction (DLLME), or with stir bar sorptive extraction (SBSE), as an extraction method for semi-VOCs in groundwater. Combining SPE with DLLME or SBSE resulted in better separation of peaks in an unresolved complex mixture. SPE-DLLME was chosen as the preferred extraction method. SPE-DLLME covered a larger polarity range (log K-o/w 2.0-11.2), had higher extraction efficiency at log K-o/w 2.0-3.8 and 5.8-11.2, and was faster compared to SPE-SBSE. SPE-DLLME extraction combined with chemical analysis by gas chromatography-mass spectrometry (GC-MS) and pixel-based data analysis of summed extraction ion chromatograms (sEICs) was tested as a new method for chemical fingerprinting of semi-VOCs in 15 groundwater samples. The results demonstrate that SPE-DLLME-GC-MS provides an excellent compromise between compound coverage, enrichment, and selectivity for semi-VOCs. Particularly, the ratio between well separated peaks and the unresolved complex mixture was improved by the dual enrichment and cleanup step. Combined with pixel-based analysis based on sEICs, the SPE-DLLME-GC-MS method is a promising approach for chemical fingerprinting.