Rapid determination of carbon isotope composition in carbonatites using isotope ratio mass spectrometry - Comparison of dual-inlet, elemental-analyzer and continuous-flow techniques

Rapid Commun Mass Spectrom. 2019 Aug 30;33(16):1355-1362. doi: 10.1002/rcm.8482.

Abstract

Rationale: Applications where stable C and O isotope compositions are useful require routine instrumental techniques with a fast sample throughput which should also produce accurate and precise results. We present a comparison of three different instrumental isotope ratio mass spectrometry (IRMS) approaches (Dual Inlet - DI; Elemental Analyzer - EA; Continuous Flow - CF) to determine the stable isotope composition of carbon in carbonate matrices, with a focus on evaluating the optimum approach for less complex instrumental techniques.

Methods: The DI-IRMS method is taken as an absolute method for obtaining accurate and precise 13 C/12 C ratios with internal errors usually < ±0.01‰ (2SD) and long-term reproducibility better than ±0.03‰ (2SD). The drawbacks of DI-IRMS are that it requires extensive offline sample preparation, rather large sample sizes (commonly >20 mg) and extended analysis times.

Results: EA-IRMS provides rapidity of analysis, relatively non-complex technique optimization and large sample throughput sufficient to distinguish natural trends although the larger internal errors and poorer reproducibility must be considered. The major disadvantage of EA-IRMS lies in a constant offset of the 13 C/12 C ratios against DI-IRMS, large internal errors (±0.2‰, 2SD) and the worst reproducibility (±0.3‰, 2SD) of all the explored methods. The results acquired using CF-IRMS are comparable with those obtained by employing DI-IRMS with an external reproducibility better than ±0.2‰ (2SD). Compared with EA-IRMS, however, this technique requires more elaborate sample preparation - more akin to DI-IRMS. None of these two latter techniques can provide C isotope results for coexisting phases such as calcite, dolomite and ankerite unless they are physically separated and analyzed independently.

Conclusions: All methods are appropriate for 13 C/12 C determinations with CF-IRMS and EA-IRMS less applicable to high-precision measurements but relevant for studies requiring high sample throughput. Periodical analysis of matrix-matched reference materials during the analytical sequence is warranted for both EA-IRMS and CF-IRMS.