The influence of solid solution on elastic wave velocity determination in (Mg,Fe)O using nuclear inelastic scattering

Elastic wave velocities of minerals are important for constraining the chemistry, structure and dynamics of the Earth’s mantle based on the comparison between laboratory-based measurements and seismic observations. As the second most abundant phase in the Earth’s lower mantle, (Mg,Fe)O ferropericlase has been the focus of numerous studies measuring the elastic wave velocities using various methods such as Brillouin spectroscopy and ultrasonic measurements. Recently, nuclear inelastic scattering (NIS) has been used to determine elastic wave velocities of iron-bearing phases. However, the elastic wave velocities of ferropericlase obtained using NIS are considerably lower than the velocities obtained by other methods, even at ambient conditions. One possible source of this discrepancy is the local nature of the NIS method. In order to test this hypothesis, we have investigated six ferropericlase samples with various iron contents using NIS. The Debye sound velocities calculated using the conventional method of NIS analysis are consistent with previous results obtained using NIS, yet the values are significantly lower than those obtained using ultrasonics and Brillouin spectroscopy. If the Debye sound velocities are re-calculated based on a mixture of different iron next-neighbour configurations with different compositions, the Debye sound velocities determined by NIS agree well with the results from other methods. Our new model was also successfully applied to high-pressure NIS data taken from the literature. Our results constitute an important step towards a better understanding of how to obtain reliable sound velocities of iron-bearing mantle minerals from NIS measurements.