Print

Shinebarger, Steven

Chemistry

2004

MS

Isotope ratio mass spectrometry (IRMS) is a high precision technique that is capable of measuring very fine differences in ¹³C, ²H, and ¹⁸O isotopic content of simple gases such as CO₂, N₂, and H₂. Because JRMS measures simple gases, analyte material must be combusted or pyrolyzed offline prior to IRMS measurement. This means that compound specific measurement requires the analyte of interest be separated from the matrix it resides prior to combustion or pyrolysis. Alternatively, compound specific measurements by IRMS can be performed using gaschromatography-combustion-isotope ratio mass spectrometry (GC-C-IRMS) or gas-chromatography-pyrolysis-isotope ratio mass spectrometry (GC-P-IRMS). GC-C-IRMS has become a commonly used technique for the online separation of biological material by GC, followed by online combustion of GC effluent using a flow-through oxidative furnace. Because the analyte must be volatile enough to pass through the GC, derivatization reactions are performed to increase analyte volatility. The addition of a tertbutyldimethylsilyl (TBDMS) group is a common derivatization approach. However, preliminary measurements of ¹³C labeled amino acids derivatized as the TBDMS derivative did not produce the expected ¹³C tracer enrichment. To investigate this finding, measurements were made using leucine and glutamate standards of known ¹³C enrichment derivatized both with and without silicon. The measured enrichments using derivatives without silicon were as expected. However, when silyl derivatives were used, results indicate that the silicon retains one C per silicon. It is postulated that this is due to the formation of silicon carbide in the oxidation furnace. Because IRMS instrumentation is mature, the predominant limitation on accuracy and precision of GC-C-IRMS and GC-P-IRMS instruments is our ability to integrate GC peaks and subtract background. Two different integration algorithms were therefore explored in detail using simulated peak data: 1) standard summation integration and 2) integration by regression. Integration by regression plots one ion-current against another and fits a linear function to the data where the slope of the line is the isotope ratio. The effects of background and isotopic fractionation on the calculation of isotope ratios were determined for the two integration methods. It was found that integration by regression degrades in accuracy much faster than standard integration when peaks are separated in time. However, fractionation can be corrected by moving the minor beam peak toward the major beam peak until the coefficient of determination for the regression of the minor beam versus major beam peak data is near 1. When background effects were considered, integration by regression was independent of background subtraction error, while standard integration was not. Because integration by regression is independent of background and isotopic fractionation can be completely corrected, integration by regression is a more robust technique for peak integration. Therefore, this integration approach was implemented in a Visual Basic program called Psi-Cal that can calculate isotope ratios for data obtained on two commercially available instruments (Finnigan delta plus GC-C-IRMS and PDZ Europa 2020 GC-P-IRMS). The results for ¹³C/ ¹²C and ²H/¹H isotope ratios calculated by Psi-Cal indicate that integration by regression offers promising advantage over standard integration approaches when background subtraction becomes problematic. In summary, it was shown that silyl containing derivatives retain carbon, presumably in the form of SiC. If silyl derivatives are to be used, a correction scheme has to be used that will return the tracer enrichments back to expected levels. Two different integration approaches were investigated that had both advantages and disadvantages. Because isotopic fractionation can be corrected, integration by regression should be a more robust integration technique than standard integration. Finally, a program called Psi-Cal was written in Visual Basic to calculate isotope ratios. Psi-Cal contains the integration algorithms discussed above and is cross-platform for two commercially available IRMS instruments.