Rapid Commun Mass Spectrom 2016 08;30(16):1857-64
Department of Human Evolutionary Biology, Harvard University, 11 Divinity Avenue, Cambridge, MA, 02138, USA.
Rationale: The δ(2) H values of some nitrogen-containing organic compounds measured by High-Temperature Conversion (HTC) with a glassy carbon reactor have been shown to be inaccurate. A probable explanation for these analytical inaccuracies is the formation of HCN, allowing some hydrogen atoms to escape isotope ratio measurement. We assess this isotopic effect in sample types commonly used for (paleo)ecological, environmental, archaeological, and forensic investigations.
Methods: The δ(2) HVSMOW-SLAP values and mass fraction H using a factory-recommended glassy carbon HTC reactor packing were compared with those obtained from using two Cr-containing reactor packings for a variety of N-containing substances, including amino acids, collagen, hair, and silk.
Results: δ(2) HVSMOW-SLAP values and mass fraction H differed by reactor packing for most, but not all, N-containing samples. The δ(2) HVSMOW-SLAP difference was 10-11 ‰ for modern collagen and 12-14 ‰ for hair, demonstrating that reactor configuration is important for these proteins, and that the use of a chromium-packed reactor may be desirable. In contrast, Bombyx mori cocoon (silk) δ(2) HVSMOW-SLAP values did not differ with reactor type. In general, δ(2) HVSMOW-SLAP and mass fraction H differences by reactor packing increased with mass fraction nitrogen in the sample. With the Cr-packed reactor hydrogen mass fractions were at theoretically expected values, while the glassy carbon reactor produced lower yields of hydrogen.
Conclusions: The protein and amino acid δ(2) HVSMOW-SLAP values measured by factory-recommended online HTC methods differ from those from Cr-containing reactor packing. The magnitude of the differences is variable with sample type; the molecular structure and diagenetic history of each sample may be important. Careful attention to this effect is therefore recommended for the δ(2) H measurement for all nitrogen-containing analytes. Copyright © 2016 John Wiley & Sons, Ltd.