Abstract

The melting points of fcc- and hcp-structured Fe_(0.9)Ni_(0.1) and Fe are measured up to 125 GPa using laser heated diamond anvil cells, synchrotron Mossbauer spectroscopy, and a recently developed fast temperature readout spectrometer. The onset of melting is detected by a characteristic drop in the time-integrated synchrotron Mossbauer signal which is sensitive to atomic motion. The thermal pressure experienced by the samples is constrained by X-ray diffraction measurements under high pressures and temperatures. The obtained best-fit melting curves of fcc-structured Fe and Fe_(0.9)Ni_(0.1) fall within the wide region bounded by previous studies. We are able to derive the γ–ϵ–l triple point of Fe and the quasi triple point of Fe_(0.9)Ni_(0.1) to be 110 ± 5GPa, 3345 ± 120K and 116 ± 5GPa, 3260 ± 120K, respectively. The measured melting temperatures of Fe at similar pressure are slightly higher than those of Fe_(0.9)Ni_(0.1) while their one sigma uncertainties overlap. Using previously measured phonon density of states of hcp-Fe, we calculate melting curves of hcp-structured Fe and Fe_(0.9)Ni_(0.1) using our (quasi) triple points as anchors. The extrapolated Fe_(0.9)Ni_(0.1) melting curve provides an estimate for the upper bound of Earth's inner core–outer core boundary temperature of 5500 ± 200K. The temperature within the liquid outer core is then approximated with an adiabatic model, which constrains the upper bound of the temperature at the core side of the core–mantle boundary to be 4000 ± 200K. We discuss a potential melting point depression caused by light elements and the implications of the presented core–mantle boundary temperature bounds on phase relations in the lowermost part of the mantle.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call