[Zurück]

E. Papaefstratiou, E. Rosenberg, N. Thomaidis:
"Quantitative comprehensive two-dimensional gas chromatography: Taking a closer look";
Poster: IMA 2013 - Instrumental Methods of Analysis, Thessaloniki, Greece; 15.09.2013 - 19.09.2013.

Comprehensive two-dimensional gas chromatography (2D-GC) is still mostly seen as a qualitative technique, allowing the separation of complex mixtures with a resolving power unparalleled by even the most advanced one-dimensional separations.
This neglects the fact that this technique also offers excellent quantitation capabilities. In theory, comprehensive 2D-GC should even offer improved detection limits over the one-dimensional technique, as there is a re-focusing of the analytes before being injected to the 2nd dimension column, leading to very narrow and proportionally higher peaks. On the other hand, a single 1st dimension peak is usually split, as an effect of the applied modulation, into two or more 2nd dimension peaks which reduces sensitivity again. Although it is clear that the quantitation ability of comprehensive 2D-GC will to a great extent depend on the modulation parameters, this topic has only recently been the subject of more detailed investigations [1,2]. Before that, the common understanding was that the quantitative response of 2D-GC should be the same as for 1D-GC, following the principle that "what goes in must come out" [3].
We demonstrate in this presentation that this latter assumption is an over-simplification of the actual situation. While it is of course true that there is no loss of mass when moving from one- to two-dimensional operation, a varying fraction of the true peak area is lost in the integration, depending on the separation, modulation and integration parameters. This illustrated by recording and evaluating the quantitative response of an alyklbenzene standard mixture which was chromatographed and integrated under various conditions. While 2D-GC shows excellent quantitative response for this set of test analytes, the actual integration results are never identical to the ones obtained in 1D-GC. Their variability as function of modulation and integration parameters furthermore supports the assumption that the bias of the quantitative response in one- and two-dimensional chromatography is an effect induced by the data evaluation, and not related to the separation. It is discussed under which conditions this effect becomes significant, and also which of the two popular software packages (Chromaleont´s ChromSquare and GC Image) produces better quantitative results.


References
[1] A.P. de la Mata, K.D. Nizio, J.J. Harynuk, J. Chromatogr. A, 1255 (2012) 190-195.
[2] J. Krupčík P. Májek, R. Gorovenko, I. Spánik, P. Sandra, D.W. Armstrong, J. Chromatogr. A 1286 (2013) 235-240.
[3] M. Kallio, T. Hyötyläinen, J. Chromatogr. A, 1148 (2007) 228-235.