Evaluation of the relative response factor among different gas chromatographic systems

When it comes to gas chromatographic analysis, we usually tend to consider MS as the reference detector to identify the analytes, while FID for their quantification. This is because the molecules do not respond in the same way to different detectors, and due to the analogue response of FID, the analytical signal is more linear to the concentration and the dynamic range maintains the linearity over a range of higher concentration. If the analysis goal is the chemical elucidation of a given sample, the response normalization of the single analytes on the total detected signal (area %), is an effective and common way to express the chemical distribution of within a given sample. This is reasonable and accurate with a detector as FID, whose response is fairly consistent between different compounds or at least predictable through the calculation of relative response factors (RRFs). Different approaches have been developed for their calculations, among which the effective carbon number (ECN)1. Applying total response normalization when using an MS detector is generating an error, especially if different chemical classes are considered: this is due to the different responses of molecules in relation to their different structure. In addition, predicting the RRFs is not yet demonstrated. In this contribution the relative response factors of 35 compounds with different chemical structure (nparaffins, i-paraffins, olefins, naphthenes, aromatics, and N/O-containing compounds) were calculated for GC coupled to different detectors: FID, qMS, TOFMS, and VUV. Using on-column injection, it was possible to understand the different responses of the various detectors for each class of compounds. In addition, it was studied the inlet splitting influence on RRFs, through the comparison of on-column and split injection analyses. References: [1] J.T. Scalon, D.E. Willis, Journal of Chromatographic Science, 1985, 23, 333-340.