Isotopic analogs as internal standards for quantitative GC/MS analysis--molecular abundance and retention time difference as interference factors.

The following analyte/isotope-labeled internal standard (IS) systems are adapted to further study the interference phenomenon previously reported from our laboratory-the intensity ratio of the ion-pair designated for a specific analyte/ 2 H-analog system increases as the solvent used to reconstitute the extraction/derivatization residue is increased: (a) Three analyte/ H-analog pairs with 2 H-atoms positioned at allylic sites (butalbital, secobarbital, methohexital); (b) Two analyte/ 2 H-analog pairs without these structural features (pentobarbital, phenobarbital); and (c) Two analyte/ 1 3 C-analog pairs (butalbital, secobarbital). Major experimental parameters adapted in this study include: (a) Varying reconstitution solvent volume while keeping a constant analyte/IS concentration ratio; (b) Varying analyte/IS concentration ratio; (c) Varying gas chromatograph (GC) injection port temperature; and (d) Varying GC column temperature programming conditions, rendering difference in the degree of overlap of the peaks derived from the analyte and the 2 H-analog. This study results in the following observations: (a) Changes in the intensity ratio of the ion-pair designated for a specific analyte/ 2 H-analog system depend on molecular abundance, regardless of whether the 2 H-atoms are positioned at active allylic positions or not-thus, ruling out hydrogen/deuterium exchange as the cause of the observed interference phenomenon; (b) Variations in GC injection port temperature do not alter the observed interference phenomenon-thus, ruling out chemical reactions at the injection port as the underlying cause; (c) Variations in peak-overlapping between the analyte and the 2 H-analog, facilitated by changing GC column programming conditions, alter the observed interference phenomenon. Abundance of the analyte and the 2 H-analog and their overlapping characteristic in the mass spectrometer ion source are believed to be the underlying cause of the observed interference phenomenon. The interference phenomenon observed for a specific analyte/ 2 H-analog system has significant consequences on the linearity of the thereby generated calibration curves. Nonlinear approaches can better describe the calibration data and are needed more in comparison to systems in which 1 3 C-analogs are used as the ISs.