Thermal transport properties of lowermost mantle minerals - Insights from atomic-scale simulations

地核-地幔边界（CMB）热流是与地磁场的产生和地核时间演化相关的中心过程之一。传递的热量由各自材料的热传递性质决定，而这些性质在CMB条件下没有得到充分的约束。在这个项目中，我们将使用原子尺度的数值模拟来预测地球最下层地幔中物质的导热性。在DeepDyn优先计划的第一个资助期，重点是（Mg,Fe）O铁长石、含铁和含铝的MgSiO3桥方石和后桥方石以及CaSiO3钙钛矿的晶格导热性。主要的研究目标是(1)量化下地幔条件下固溶体中质量无序效应导致的热导率降低；(2)更好地理解和减少由用于推导热导率的特定数值算法和模拟中使用的原子相互作用势引起的不确定性。不同的原子尺度模拟方法，如平衡和非平衡分子动力学模拟或玻尔兹曼输运方程方法，在这里系统地对材料和热力学条件进行基准测试。极化离子相互作用势的参数化为固溶体建模提供了一种新的方法。这些新的潜力有望做出比现有力场更现实的预测。与此同时，它们比计算密集的从头算模拟要有效得多，迄今为止，从头算模拟几乎只用于Mg端元，在这里主要用作新势的参考。模拟结果将进一步用于参数化下地幔矿物的热导率，作为压力、温度和化学成分的函数，并开发一个改进的通过CMB的热流模型。

Heat flow through the core-mantle boundary (CMB) is one of the central processes associated with the generation of the geomagnetic field and the temporal evolution of the Earth’s core. The amount of heat transferred is determined by the thermal transport properties of the respective materials, which have been insufficiently constrained at CMB conditions. In this project, we will use atomic-scale numerical simulations to predict the thermal conductivity of the materials in the Earth’s lowermost mantle. In the first funding period of the DeepDyn Priority Program, the focus is on the lattice thermal conductivity of (Mg,Fe)O ferropericlase, Fe- and Al-bearing MgSiO3 bridgmanite and post-bridgmanite, and CaSiO3 perovskite. The main research objectives are (1) to quantify the reduction of thermal conductivity due to the effect of mass disorder in the solid solutions at lower mantle conditions and (2) to better understand and reduce the uncertainties arising from the specific numerical algorithm used to derive the thermal conductivity and from the atomic interaction potential used in the simulations. Different atomic-scale simulation methods, such as equilibrium and non-equilibrium molecular dynamics simulations or the Boltzmann transport equation approach, are systematically benchmarked for materials and thermodynamic conditions of interest here. Parameterization of polarizable ionic interaction potentials provides an additional method for modeling solid solutions. These new potentials are expected to make more realistic predictions than existing force fields. At the same time, they are much more efficient than computationally intensive ab initio simulations, which have so far been applied almost exclusively for Mg end-members and serve here mainly as a reference for the new potentials. The results of the simulations will be further used to parameterize the thermal conductivity of the lower mantle minerals as a function of pressure, temperature, and chemical composition, and to develop an improved model of heat flow through the CMB.