Abstract :
[en] From a Separation Science standpoint, Life Sciences represent a broad source of complex compound mixtures that require powerful analytical strategies to be considered when characterization is aimed. The depiction of the complex volatile organic compound (VOC) component of these mixtures is an important part of such characterization. The aim of this doctoral work was to develop a versatile analytical approach to resolve such VOC samples.
For the characterization of complex VOC mixtures, comprehensive two-dimensional gas chromatography (GC×GC) is the method of choice. The combination of two GC separation columns offers higher peak capacity than single dimension GC (1D GC). Moreover, the hyphenation of GC×GC to a mass spectrometer provides an extra dimension of identification, especially when high-resolution (HR) time-of-flight mass spectrometers (TOFMS) is used. However, the optimization of the GC×GC conditions is not straightforward. Currently, there are no defined strategies to obtain the best capabilities these instruments, not only in terms of the optimization of the separation itself, but also mainly in terms of data processing efficiency.
The first section of this work was devoted to the evaluation of theoretical aspects and the definition of a specific optimization strategy for GC×GC separation. The first part dealt with the heavily debated concept of ‘orthogonality’ was specifically investigated by means of a metric calculation, the Orthogonality Index (OI), linked to a new nomenclature describing the separation space usage. The second part was focused on the development of an optimization strategy for the sampling and the chromatographic separation of VOCs, as well as the processing of large sets of GC×GC data. This was conducted on complex beer aroma headspace replicates for sampling method selection, peak dispersion optimization, and robust multivariate statistical treatment. The third part of this theoretical investigation relied on the implementation of fast-GC conditions for GC×GC separation. The idea was to evaluate the effect of combining short columns, fast modulation, high temperature ramping, and high flow on GC×GC efficiency for explosive headspace analyses.
In the second section, the focus was directed to challenging applications in the area of thanatochemistry, i.e. the chemistry of death, especially considering the volatile fraction of cadaveric decomposition, a complex chemical process releasing numerous VOCs. In this context, soil surrounding decomposing remains, headspace of cadavers, and more “specific” matrices (i.e. synthetic solutions and internal gas from decomposition) were analyzed by GC×GC TOFMS. Major attention was dedicated to the development of an analytical and data mining procedures for these complex VOC samples with emphasis on the study of the impact of different parameters on the cadaveric decomposition process.
