Sensitivity and noise in GC-MS: Achieving low limits of detection for difficult analytes

Abstract Gas chromatography–mass spectrometry (GC–MS) instrument limit of detection (LOD) is typically listed by major vendors as that of octafluoronaphthalene (OFN). Most current GC–MS instruments can achieve LODs in the low femtogram range. However, GC–MS LODs for realistic analytes in actual samples are often a few orders of magnitude higher than OFN's. Users seldom encounter 1 pg LOD in the single ion monitoring mode in their applications. We define this detectability difference as the “OFN gap.” In this paper, we demonstrate and discuss how the OFN gap can be significantly reduced by the use of GC–MS with supersonic molecular beams (SMB). Experimental results were obtained with a recently developed GC–MS with SMB named 1200-SMB, that is based on the conversion of the Varian 1200 system into a GC–MS–MS with SMB. With this 1200-SMB system, the LOD of all types of analytes, including OFN, in real samples is significantly improved through the combination of: (a) enhanced molecular ion; (b) elimination of vacuum background noise; (c) elimination of mass independent noise; (d) elimination of ion source peak tailing and degradation; (e) significantly increased range of thermally labile and low volatility compounds that are amenable for analysis through lower sample elution temperatures; (f) reduced column bleed and ghost peaks through sample elution at lower temperatures; (g) improved compatibility with large volume injections; and (h) reduced matrix interferences through the combination of enhanced molecular ion and MS–MS. As a result, the 1200-SMB LODs of common and/or difficult compounds are much closer to its OFN LOD, even in complex matrices. We crossed the 400 for methomyl, and >2000 for C 32 H 66 . In general, the harder the compound analysis, the greater is the gain in sample detectability using the 1200-SMB versus traditional GC–MS. Thus, the 1200-SMB lowers LOD, particularly for difficult analytes that are normally sacrificed in methods, and the detectability gains can amount to a few orders of magnitude over traditional GC–MS in real-world applications.