First-principles Study Of Elasticity Of Carbonates And Aluminous Phases: Implications For The Mantle Material Circulation

The computing and designing of materials has developed into a new and fancy branch of research with the development of computer technologies.Progress in computational mineral physics based on first-principles calculations has also been dramatic in the last decades in conjunction with the rapid advancement of computer techniques.First-principles is one of important approaches that solve the fundamental equations of quantum mechanics.Density functional theory(DFT)is an exact theory to investigate the electronic structure(the ground state)and allows us to reduce the interacting many-electron problem to a single-electron problem.On the basis of the DFT,quantum mechanical hamiltonians of many-body electron systems can be efficiently and quantitatively evaluated.DFT is among the most popular and versatile methods available in condensed-matter physics,computational physics,computational chemistry,and even Earth science.For example,it allows us to calculate the equation of state(EoS),thermodynamic properties,elastic constants and other properties of minerals.The material circulation inside the Earth is a complex issue that has an important impact on the structure,property and evolution of the Earth.Obtaining the elastic properties of the important constituent minerals in the Earth under high temperature and high pressure is an important channel to interpret the internal structure and material composition of the Earth.In order to improve the understanding of the Earth's internal material circulation,in this thesis,we have studied the thermodynamics,elasticity and wave velocity of carbonates and aluminous minerals.Our research on carbonate minerals is originated from Prof.Weidong Sun's model.He proposed that diamond may be enriched at the base of the transition zone,because it is denser than peridotitic melt at depths of<660 km and becomes less dense at deeper depths,such that it should have accumulated near this density crossover in the Magma Ocean period.Such a diamond-rich layer would have been stabilised by the solidification of the Earth,and may have survived the effects of plate subduction and mantle plumes,but it may have been oxidised into a carbonated layer by Fe3+ released during upwelling of bridgmanite to the mantle transition zone.High-temperature and high-pressure experiments show that bridgmanite can have Fe3+/Total Fe ratios of 0.6 or higher.At least some subducted slabs sink into the lower mantle.In response,the lower-mantle materials,including bridgmanite are forced up into the transition zone through compensational background upwelling.Once in the transition zone,bridgmanite transforms to ringwoodite and majorite.Most of theferrous iron released from bridgmanite can be hosted in ringwoodite,whereas only a small portion of the ferric iron can be accommodated in majorite.After billions of years of plate subduction,the ferric iron released by the decomposition of bridgmanite gradually transforms the diamond layer into a carbonated peridotite layer.The solidus of carbonated peridotite and mantle adiabat show that carbonated peridotite is stable at the depths of-300-800 km.If a layer of carbonated peridotite is formed at the base of the transition zone,it will inevitably have a certain impact on the density and wave velocity at the bottom of the transition zone.To answer this question,we have studied carbonates which is the major carbon host phases in this thesis.Carbonates transport carbon to the deep Earth though the subducted slabs.Because more than 90%carbon is stored in the deep Earth,the deep carbon cycle is the key to understand the origin and process of surface carbon cycle.Deep carbon recycling influences the atmospheric CO2 budget and climate changes through geologic time.Magnesite(MgCO3)is a likely main host of carbonates in the mantle and is stable to-84 GPa without melting in the mantle PT conditions,it plays an important role in the transport and storage of carbon in the Earth's mantle.Therefore,the physical properties of magnesite at high pressure and high temperature(PT)are crucial for understanding the deep carbon cycle.Here we investigated thermodynamic and elastic properties of magnesite under the mantle PT conditions using first-principles calculations with local density approximation(LDA).Our results show that magnesite has the seismic velocities close to those of forsterite.The effect of the magnesite on seismic velocity of the carbonated peridotite and eclogite is subtle in the upper mantle.However,magnesite has much smaller seismic velocities and far larger elastic anisotropy than major minerals in the transition zone and lower mantle.We found that a carbonated layer at the boundary of the transition zone and lower mantle is able to explain the density and velocity gradient and anisotropy anomaly.The enrichment of magnesite in the transition zone and lower mantle will likely produce seismically detectable low-velocity zone and velocity anisotropy.In this thesis,we also studied the aluminous minerals in the mantle material circulation,which is mainly found in the Mid-Ocean Ridge Basalt(MORB).MORE is formed via seafloor spreading at mid-ocean-ridges.It is one of the most important rocks in the Earth's crust.Upon subduction and exposure to high pressures and temperatures(PT),the high Al2O3 content in MORB gives rise to a "new aluminous phase"(NAL)responsible for up to 25 wt%of its composition.This phase has hexagonal structure,21 atoms/cell,and is a complex solid solution with chemical formula XY2Z6O12,where X is a large monovalent or divalent cation,e.g.,Na+ or Ca2+,Y is a mid-sized cation,e.g.,Mg2+,and Z is Al3+ and/or Si4+.Chemical constraints on its composition may reduce the number of end-member compounds to less than twenty.Using first-principles calculations,we model the iron-free NAL phase by considering seven end-members.We perform Quasiharmonic(QHA)calculations to address composition dependent thermodynamic and thermoelastic properties of this phase.These results are essential to investigate the seismic signature of subducted MORB crust into the deep mantle.