地幔中主要矿物晶格热导率的第一性原理计算

矿物在地球深处高温高压条件下的热导率是开展地幔对流数值模拟所需的核心参数之一，对理解地球内部的热量交换具有重要意义。目前第一性原理矿物热导率计算大都基于声子气体模型。这些计算结果与室温下的实验数据符合较好，但在高温下出现系统性偏差。初步分析认为是由于声子气体模型只考虑了体系真实热流的二阶(简谐)部分，在高温下，高阶(非谐)热流对热导率的贡献不能被忽略。本项目将从第一性原理出发，发展一种能定量计算高阶热流对热导率贡献的新方法。该方法将超越声子气体模型，预计能显著提高高温高压下晶格热导率计算的准确性。本项目将利用这一新方法确定MgSiO3钙钛矿，后钙钛矿,MgO的晶格热导率，给出核幔边界处热流大小的更准确估计；系统地计算主要矿物在不同温度和压强下的热导率，并据此建立一个较为完整的地幔矿物热导率模型，为地幔对流等地球动力学问题的数值模拟提供有力支持。

Thermal conductivities of minerals at the pressures and temperatures of the Earth's deep interior are the key parameters for mantle convection simulations. They are also crucial for understanding the energy exchange inside the Earth. At present most first principles calculations of lattice thermal conductivities of minerals are based on phonon gas model (PGM). The predicted thermal conductivities using PGM are in good agreement with experiments at room temperature. However systematic deviations emerge at high temperature. This is because PGM incorporates only the second order(harmonic) part of total heat current. At high T, contributions from higher oder(anharmonic) heat currents are no longer negligible. In this project we will develop a new first-principles method to determine the contributions of anharmonic heat currents to thermal conductivity. The new method goes beyond PGM and should significantly improve the precison of the calculated thermal conductivity at high temperature. We plan to use the new method to determine the lattice thermal conductivities of MgSiO3 perovskite and post-perovskite, MgO，etc., and to provide a better estimate of the heat flux at the core-mantle boundary. With a systematic exploration the thermal conductivities of typical minerals in a wide temperature and pressure range, a comprehensive model of thermal conductivities of mantle minerals will be established. This model will be useful for mantle convection and other geodynamic simulations.