The interpretability of stable hydrogen isotopes in modern herbivore tooth enamel

• Published in: Geochimica et cosmochimica acta Vol. 270; pp. 84 - 94
• Main Authors: Drewicz, Amanda E., Trayler, Robin B., Holloway, Mariah E., Harrigan, Claire O., Kohn, Matthew J.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.02.2020
• Subjects: Apatite; Biogenic; Hydrogen isotopes; Labile hydrogen; Tooth enamel; Tooth enamel; Apatite; Biogenic; Labile hydrogen; Hydrogen isotopes
• ISSN: 0016-7037; 1872-9533
• DOI: 10.1016/j.gca.2019.11.013

Stable hydrogen isotopes (δD values) in animal organic tissues such as collagen and keratin typically correlate with local meteoric water, but little is known about δD in tooth enamel apatite. In this study, we analyzed comminuted tooth enamel for oxygen isotopes (δ18O) of the CO3 component and for δD of bulk enamel. We find positive correlations between enamel δD and δ18O (R2 = 0.70) and between enamel δD and local precipitation δD (R2 = 0.53). However, the slopes of these relationships are much shallower (less variation in tooth enamel δD) than expected from studies of other tissues. Based on mass spectrometric peak areas, H2 contents for enamel are 2–5 times higher than expected from chemical compositions, and we interpret as much as 50–90% of measured hydrogen from tooth enamel may be adsorbed water derived from laboratory water vapor. We tested this hypothesis by equilibrating tooth enamel with very high and very low δD values of water vapor, then exposing to laboratory air for different periods of time ranging from minutes to 8 hours. These experiments show that the apparent δD value of enamel converges to a nearly constant δD value in 1 to 2 hours. The large amount of adsorbed water and rapid approach to equilibrium will make it difficult to infer provenance from δD measurements alone, or to reproduce measured tooth enamel δD values among laboratories with different water vapor compositions. Heating at 70 °C for 48 hours in air does not remove adsorbed hydrogen, but does reduce δD values by c. 10‰ compared to unheated samples. Differences in δD values for heated vs. unheated enamel may reflect either a different, temperature-dependent partition coefficient between adsorbed water (on apatite) and water vapor, exchange of structural H at elevated temperatures, or subtle changes to crystal structure, such as loss of structural H.