Density, magnetic properties and sound velocities of iron-rich materials at high temperature and high pressure

Understanding the composition of Earth’s inner core is crucial for revealing the mechanisms of core formation and the evolution of Earth. The presence of light elements in the Earth’s inner core has been indicated in recent studies, based on the mismatch between the observed density of the inner core and the density of pure iron at relevant conditions. The nature and abundance of light element(s) are under debate, yet they are fundamental in understanding the formation and evolution of the Earth’s core. Carbon has been considered a possible major light element candidate, besides hydrogen, oxygen, silicon and sulphur. In particular, Fe3C has been proposed to be the major component in the Earth’s inner core in a previous thermodynamics study. However, the possibility of Fe3C being a major inner core component has been under debate in recent studies, largely due to our limited knowledge of the properties of Fe3C at extreme pressure and temperature (P-T) conditions. In this thesis work, I investigated the possibility of carbon as a principal light element in the inner core in the form of Fe3C. Considering the lack of direct accessibility to the inner core, the only way to test a carbon-rich inner core model is to compare the properties of iron-carbon compounds, including the density and sound velocities, with the observed values of inner core, e.g., the values in preliminary reference Earth model (PREM) determined using normal mode data and seismic travel time data. In this work, I studied the density, elasticity, sound velocity and magnetism of Fe3C using a series of experimental methods, including X-ray diffraction (XRD), nuclear resonant inelastic X-ray scattering (NRIXS), synchrotron Mössbauer spectroscopy (SMS) and conventional Mössbauer spectroscopy (CMS). The starting materials of (57Fe-enriched) Fe3C samples were synthesized using large-volume presses. The composition and purity of the samples were confirmed using high-resolution XRD and CMS methods. A magnetic transition in Fe3C from the low-pressure ferromagnetic phase to a high-pressure non-magnetic phase was reported in literature; however, the transition pressure has been controversial, ranging from 9 GPa to 25 GPa. The effect of this transition on compressibility is not well understood. In this study, I carried out SMS and CMS experiments in an attempt to resolve the controversy. The results from both methods show that the transition pressure is ~ 6 GPa (Chapter 4-Part II). A discontinuity around this pressure was also observed in sound velocity versus density data (Chapter 4-Par I), as well as the compression curves from the XRD data (Chapter 5). In addition, the SMS data in this work indicate an electronic transition between 50 GPa and 73 GPa (Chapter 5). The nature of this electronic transition remains to be further investigated. The compression curves from the XRD data also indicate a discontinuity around this pressure (Chapter 5). To study the sound velocities of Fe3C, I performed NRIXS experiments on a few-crystal sample (composed of one or a few single crystals) up to 50 GPa at 300 K (Chapter 4 - Part I) and on powder Fe3C samples up to 45 GPa and 1450 K (Chapter 4 - Part III). Compressional velocities VP and shear velocities VS were derived combining an existing equation of state, estimated thermal expansion parameters and the phonon (vibration) densities of state extracted from NRIXS spectra. The derived VP and VS follow Birch’s law - a linear relationship between sound velocity and density. The extrapolated values of VP and VS at the inner core P-T conditions are higher than those of the inner core. This is consistent with one of the criteria for a light element candidate – the light element should raise the VP of iron, as the sound velocities of pure iron are suggested to be too low for the inner core from previous studies (e.g., Mao et al., 2005a). The effect of temperature on sound velocity is not well understood, partially due to a lack of data. Among the existing data, it has been controversial whether or not the high-temperature sound velocities deviate from Birch’s law. To shed light on the temperature effect on sound velocities, I carried out NRIXS experiments on powder Fe3C samples up to 45 GPa and 1450 K (Chapter 4 - Part III). The results at high temperatures suggest temperature induced shear velocity decrease, and also indicate that the temperature effect increases as temperature increases and decreases as pressure increases. The temperature needed to reconcile the sound velocity mismatch between Fe3C and the inner core at 300 K is within the expected values for the inner core, supporting Fe3C as a possible candidate material for the inner core. In Chapter 4 - Part IV, a recent experimental capability of simultaneous nuclear resonant scattering and XRD measurements using synchrotron radiation at beamline 3-ID of the Advanced Photon Source is discussed. Here the application of this method to determine the sound velocities of compressed Fe3C is shown. The XRD measurements allow detection of microscale impurities, phase transitions and chemical reactions upon compression or heating. They also provide information on sample pressure, grain size distribution and unit cell volume. By combining the Debye velocity extracted from the NRIXS measurements and the structure, density and elasticity data from the XRD measurements simultaneously obtained, more accurate sound velocity data can be derived. In this chapter, I also reported the anisotropy in Fe3C at ambient conditions, inferred from the difference in sound velocities between the few-crystal sample and a powder sample Chapter 4-Part III, IV; Gao et al., 2009). To study the density and elastic properties of Fe3C, I carried out single crystal XRD measurements to 200 GPa at 300 K (Chapter 5). Elastic constants of bulk modulus and pressure derivative of bulk modulus are derived through equation-of-state fitting to these density versus pressure data. The extrapolated densities of Fe3C at inner core P-T conditions are close to PREM values. These results suggest that pure Fe3C or Fe3C mixed with a small amount of iron can match the density of the inner core, supporting carbon as a major light element candidate in the Earth’s inner core