The limited depth range of a metallic-Fe-bearing layer in the lower mantle and its implications for partial melting

Jennifer Girard, Yale University, Geology and Geophysics, New Haven, CT, United States and Shun-ichiro Karato, Yale University, Earth and Planetary Sciences, New Haven, CT, United States
Abstract:
Partial melting in (most of) the lower mantle occurs only by the presence of volatile elements such as hydrogen (and/or carbon). The experimental studies by [Kawamoto, 2004] showed that partial melting is possible even at temperature of 1673 K in the shallow lower mantle if there is sufficient water. However, if metallic Fe is present in the lower mantle as suggested by [Frost et al., 2004]). most of hydrogen will be dissolved in metallic Fe, and thus melting will be prevented, making it difficult to explain a seismic velocity drop in the shallow lower mantle [Schmandt et al., 2014].

In this study we conducted high pressure experiments using the Rotational Drickamer Apparatus (RDA), on bridgmanite (Mg,Fe)SiO3 + (Mg,Fe)O mixture at pressure up to 23-29 GPa and temperature of about 2000-2200K. Using the advantage of the new RDA cell design which provide a larger pressure gradient (~6 GPa across the sample), we report experimental observations showing that metallic Fe is formed only in the low pressure conditions, 24-26.5 GPa (corresponding to the depth range of 660-730 km), leaving the shallow lower mantle minerals “dry”. Our results are also consistent with the published results by [Irifune et al., 2010; Sinmyo and Hirose, 2013] where they did not find any metallic Fe above 27 GPa. Therefore we conclude that metallic Fe is present only in the narrow depth range in the lower mantle. In such a case partial melt would be impossible and only occur at depth greater than 730 km. Our results explain why a velocity drop is observed at ~730 km not at 660 km [Schmandt et al., 2014]. The present results also have important implications for other geochemical issues including the behaviors of siderohpile elements during core formation.

Frost, D. J., et al., (2004), Nature, 428, 409-412.

Irifune, T., et al., (2010), Science, 327, 193-195.

Kawamoto, T. (2004), Physics of Earth and Planetary Interiors, 143/144, 387-395.

Schmandt, B., et al., (2014), Science, 344(6189), 1265-1268.

Sinmyo, R., and K. Hirose (2013), Physics and Chemistry of Minerals, 40, 107-113.

Back to: Physics and Chemistry of the Deep Earth III