The influence of liner geometry during injection process in gas chromatography

(semi-)volatile compounds. Although it is a common technique widely used in research and analysis laboratories, its weaknesses are not always known, which is essential to method optimization. In the case of quantitative analyses, the factors that need to be taken into consideration most are the injector and detector discrimination. Concerning the first type of discrimination, problems arise when vaporization injections are used, since the mechanisms involved in sample introduction into the injector and transfer from the injector into the column are complex. However, the most commonly used injection methods are those that involve vaporization of the sample (i.e., split and splitless). On the contrary, the on-column injection mode, which is not affected by injector discrimination as there is no vaporization step, is not widely used due to its lack of versatility (it can be used for trace compound analysis). One of the things to consider when performing split and splitless analyses is the choice of liner. This insert is designed to slow down the passage of the liquid sample between the syringe needle outlet and the column inlet, allowing for more complete and uniform evaporation of the sample. There are many types of liners with different characteristics (e.g., presence of packing, different geometries, presence of restrictions), and the choice is important to reduce injector discrimination. In this study, the effect of liner geometry on injector discrimination phenomena in split and splitless injection modes was evaluated. A standard solution containing 35 compounds with different chemical structures (linear alkanes, branched alkanes, cycloalkanes, alkenes, aromatics, N/O-containing compounds) was analyzed in GC-FID to evaluate recovery, discrimination among compounds with different volatilities, and reproducibility of the analysis for five liners with different characteristics in split mode and six in splitless mode. For split mode, five different split ratios were evaluated (i.e., 1:5, 1:20, 1:50, 1:100, and 1:200). Recoveries were calculated by normalizing the results of each analysis with those obtained from on-column analyses performed considering the same mass entering the column, in order to eliminate the detector discrimination. Discrimination between compounds with different volatilities was evaluated by calculating the CV% of the recoveries of the various compounds. While the reproducibility of the analyses was calculated by performing five repetitions for each type of analysis.