Absolute molecular configuration strategies using preparative gas chromatography and multidimensional gas chromatography with spectroscopy

This thesis is about method development for the structural elucidation of selected known and unknown organic compounds using prep-GC and prep-MDGC combined with NMR spectroscopic analysis. Quantification of a model compound using prep-GC combined with NMR spectroscopy, and the introduction of two novel prep-MDGC approaches for complex samples with NMR spectroscopy, were successfully achieved. Prep-GC techniques have been introduced especially for structural characterisation of unknown or trace compounds which cannot be absolutely identified using MS data alone, or in the absence of appropriate reference compounds. For example, differentiating structural isomers using GC-MS can lead to uncertainty in assignment and often further structural elucidation is also required. Therefore prep-GC techniques make it possible to isolate individual pure compounds of interest out of complex matrices that can then be subjected to other spectroscopic studies such as NMR, FTIR and Raman spectroscopy, AMS and X-ray crystallography in a wide range of research applications. Method development was initiated with the quantitative determination of caffeine using a prep-GC system combined with off-line 1H and 13C NMR spectroscopic analysis to demonstrate the efficiency of a purpose-designed xTA. This study was also designed to understand how prep-GC and prep-MDGC techniques could be readily applied to routine analysis. Therefore, caffeine as a model compound was collected with the different number of GC injections, and with an internal standard added to the collected material; these solutions were analysed by both GC and 600 MHz NMR spectroscopy. Both techniques demonstrated good correlation between the number of collected injections and caffeine-to-IS response areas obtained by each technique. This quantitative study provides basic understanding to inform the performance of prep-GC and prep-MDGC studies. For it is important to recognise the minimum amount of analyte that must be collected - which increases with multiple injections - and to ensure that the analyte obtained was of sufficient purity and collected in a quantity that allowed for structural characterisation. In the next phase of the investigation, two new prep-MDGC systems were developed to achieve better resolution – ideally to generate completely resolved compounds – and study the possibility for absolute characterisation of target compounds out of complex mixtures. MDGC systems and in particular those based on capillary GC formats, have not been commonly used as a preparative isolation approach. For validation of these two systems the preparative-scale isolation of DMN isomers (from a standard mixture) was achieved. Using prep-H/C-GC×GC system retrofitted with a Deans switch (DS) and xTA, two out of the most abundant peaks were separately successfully collected into the xTA. The NMR data obtained for the most abundant peak (using 50 replicate injections) showed that the peak was in fact an unresolved peak that contained approximately the same amount of two isomers, 1,3-DMN and 1,6-DMN. The NMR data obtained from the least abundant peak out of 3 major peaks (again from 50 injections) demonstrated that the isomer was 1,7-DMN. A further innovation involved prep-H/C MDGC but now with dual DSs, and was set up with the best column set (BPX90 as 1D and VF-200ms as 2D) as selected for the DMN isomers separation. The NMR data obtained from one out of three major peaks (38 replicate injections) collected demonstrated that the two isomers, 2,6-DMN and 2,7-DMN, were unresolved. One of the methods developed in the previous study (prep-H/C MDGC-MS) was validated for profiling of crude oil samples. To characterise some target compounds (i.e. biomarkers) out of complex samples using prep-GC, enhanced resolution is a prerequisite. Therefore the “total” sample analysis of a crude oil as a model sample was performed using a technique for which the term ‘incremented sequential H/C analysis’ has been coined. This approach demonstrated substantially increased resolution for structural identification of biomarkers, and an interesting grouping of different chemical classes for this complex crude oil sample. Even though this method required relatively long analysis time, the enhanced resolution that was accomplished served to support additional compound identification. As a model compound, one of the most important biomarkers in crude oil samples, pristane (2,6,10,14-tetramethylpentadecane), was separated from a C17 linear hydrocarbon (heptadecane) compound with which it often co-elutes in a conventional single column GC system. This method will allow preparative-scale isolation of other biomarkers, permitting both GC-MS and NMR analysis identification, depending on the total mass of sample that can be collected. Subsequent to the quantitative determination of caffeine, some illicit drugs, known as legal highs provided by Racing Analytical Services Ltd (RASL) and a synthesised legal high were used to develop the prep-GC method combined with 1H NMR spectroscopy. The aim of this research was to validate the prep-GC system for the identification of pharmacologically active compounds from the complex mixture using the system. Legal highs require authorities to quickly act to identify the occurrence of new substances, and to support this with adequate characterisation of compounds. It was found to be deleterious to health, and then steps for prevention of proliferation may be required since they have recently caused serious social issues. The NMR data obtained from one of the legal highs collected after derivatisation showed that the methyl groups severely decomposed. One of the legal highs, 4-methylcathinone was synthesised to confirm the stability of these drugs. The NMR spectrum obtained from the synthesised compound, collected following the prep-GC procedure, showed that the methyl groups decomposed as well.