Grain boundary mobility of carbon in Earth's mantle: A possible carbon flux from the core

doi:10.1073/pnas.0710806105

Grain boundary mobility of carbon in Earth's mantle: A possible carbon flux from the core

Leslie A. Hayden* and
E. Bruce Watson

Department of Earth and Environmental Sciences, Rensselaer Polytechnic Institute, Troy, NY 12180

Edited by Albrecht W. Hofmann, Max Planck Institute for Chemistry, Mainz, Germany, and approved April 7, 2008 (received for review November 14, 2007)

Abstract

The importance of carbon in Earth's mantle greatly exceeds its modest abundance of ≈1,000–4,000 ppm. Carbon is a constituent of key terrestrial volatiles (CO, CO₂, CH₄), it forms diamonds, and it may also contribute to the bulk electrical properties of the silicate Earth. In contrast to that of the mantle, the carbon content of Earth's metallic core may be quite high (≈5 wt %), raising the possibility that the core has supplied carbon to the mantle over geologic time. The plausibility of this process depends in part upon the mobility of carbon atoms in the solid mantle. Grain boundaries of mantle minerals could represent fast pathways for transport as well as localized sites for enrichment and storage of carbon. Here, we report the results of an experimental study of grain-boundary diffusion of carbon through polycrystalline periclase (MgO) and olivine ([Mg,Fe]₂SiO₄) that were obtained by determining the extent of solid solution formation between a graphite source and a metal sink (Ni or Fe) separated by the polycrystalline materials. Experimental materials were annealed at 1,373–1,773 K and 1.5–2.5 GPa pressure. Calculated diffusivities, which range up to 10⁻¹¹ m²·s⁻¹, are fast enough to allow transport over geologically significant length scales (≈10 km) over the age of the Earth. Mobility and enrichment of carbon on grain boundaries may also explain the high electrical conductivity of upper mantle rocks, and could result in the formation of C-H-O volatiles through interactions of core-derived C with recycled H₂O in subduction zones.

Footnotes

*To whom correspondence should be addressed at the present address:
Institute of Geophysics and Planetary Physics, University of California, 595 Charles Young Drive East, Box 951567-1567, Los Angeles, CA 90095.
E-mail: lhayden{at}ucla.edu
Author contributions: L.A.H. and E.B.W. designed research; L.A.H. performed research; L.A.H. analyzed data; and L.A.H. and E.B.W. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
This article contains supporting information online at www.pnas.org/cgi/content/full/0710806105/DCSupplemental.
↵† For the equilibrium 2Fe + O₂ ↔ 2FeO, $\text{[math]}$
↵‡ For the equilibrium 2Ni + O₂ ↔ 2NiO, $\text{[math]}$
↵§ The measured C concentrations may exceed the known solubilities at 1 atm, and the high-pressure solubilities are unknown. The effect of limited C solubility in the metals may mean that the estimated fluxes are a lower bound.
↵¶ Determined by using the constant surface solution to Fick's second equation: , where C_(x_{, t}₎ is the carbon concentration in the foil, C_o is the source concentration of carbon, x is the distance from the source, t is the run duration, and D is the diffusivity.
↵‖ The C concentration at the source should be the solubility of C in Ni at run conditions, rather than in graphite (100%). This would increase the calculated diffusivities; therefore, the values presented in this article represent a lower bound.