Prediction and characterization of novel Earth-forming minerals using advanced ab initio simulations

人类在研究外层空间和原子物质的内部运作方面取得了很大的进步，但我们居住的星球仍然是一个谜。地球的内部深处隐藏着我们，无法直接探测，它定义了我们星球的面貌，并深刻地影响着我们的日常生活（通过地震，火山活动和从太阳风中拯救生命的磁场）。只有通过地震学、地球化学和矿物物理学的结合，才能在地球深层内部的研究中取得进展。计算矿物物理学最近取得了很大的进展，量子力学方法的发展使人们能够研究与地球相关的复杂化学，并对行星内部典型的高压和高温下发生的新化学现象进行无偏见的预测。该项目旨在借助先进的量子力学方法更好地了解地球的化学和物理。这种方法是相当普遍的，最近已经导致了新材料的计算设计的突破，这对我们的社会具有潜在的变革价值。这项研究跨越了地球科学、材料科学、物理学和化学等多个领域，并专注于激发人类好奇心的问题（“地球是由什么组成的”，“地球是如何形成的”）。如果没有年轻科学家的精力，这些研究是不可能的，他们将在这一多学科科学领域获得独特的培训，并将完全具备在任何这些领域为世界服务的能力。对地球内部深处的研究是现代研究中最令人着迷的领域之一。目前，在地球科学家几十年的持续工作基础上，终于建立了合理的地球深部矿物学模型。地球深部的热变化和成分变化可以通过地震层析成像和矿物物理学的数据得到很好的估计。矿物物理学提供有关地球形成矿物的物理性质及其对温度和成分的依赖性以及矿物系统的相图的信息。地球成因矿物的勘探是地球科学中最重要的问题之一。尽管实验高压技术取得了巨大的进步，但大多数对地球深部地幔和地核压力的物理测量都存在问题。最近，（Mg，Fe）SiO 3钙钛矿，长期以来被认为是稳定的在地球的深部下地幔，被发现分解成MgSiO 3钙钛矿和富铁的六方相（H相，其晶体结构仍然是有争议的）。PI的初步结果显示了令人惊讶的结果，即在Mg-Si-O系统中，钛铁矿MgSiO 3和钙钛矿MgSiO 3在30 GPa 0开尔文下都是化学不稳定的，并且在30 GPa下根本不存在其他三元稳定化合物，这表明可以观察到非常重要的意外现象，并且可能对神秘的地震特征产生重大影响。此外，除了410、520、660 km处的地震不连续外，800 km处的另一个地震不连续也出现在变化很大的区域，但迄今没有合理的解释，可能还有另一个未知的震相。地幔中的矿物可能比我们目前所知的更复杂。基于量子力学的理论模拟在地球科学中发挥着越来越重要的作用。因此，PI将应用最先进和最新开发的计算物理工具来解决与地幔，核和核幔边界区域相关的问题。PI将集中在三个问题上：（1）方法学发展，以预测三元甚至更高的多组分系统中具有可变化学计量比的材料。PI最近的工作表明，基于进化方法的二元系统预测是容易处理的。由于地球科学背景下的一些重要矿物具有令人困惑的化学计量学，剩下的挑战是在高P-T条件下预测稳定的三元，四元甚至更高的多组分体系。(2)Mg-Fe-Ca-Al-Si-O和Fe-Si-C-S-H-O体系在给定P-T条件下成土矿物的预测由于真实的地幔中的钙钛矿相和后钙钛矿相具有复杂的成分，额外的元素如Ca、Al和C也可以影响相的结构和稳定性场。因此，多组分体系的晶体结构预测将得到解决。计算地幔和地核中地球形成矿物的性质。在地幔和地核中，未解释的地震不连续面和H相的性质有望被揭示，一些新的矿物很有可能被发现。(3)地幔和地核之间的化学反应。这项研究应该大大加深我们对铁基合金的理解，将导致更好地了解地球核心的化学和矿物学，以及其特有的性质（如强烈的地震各向异性）。了解核幔边界区可能发生的反应可能有助于了解该区域的某些异常性质。简而言之，这个项目将使用先进的量子力学模拟来研究地球上最神秘的区域的材料。这项研究的核心是PI的进化算法USPEX，与第一原理电子结构计算相结合。这种方法在矿物学和物理学中产生了许多重要的预测，并被随后的实验成功证实。

Humans have greatly advanced in the study of the outer space and inner workings of atomic matter, but the very planet upon which we live remains a mystery. Hidden from us and impossible to probe directly, Earth's deep interior has defined the face our planet and profoundly influences our daily life (through earthquakes, volcanism, and the magnetic field that saves life from solar wind). Progress in studies of the Earth's deep interior can only be made through a combination of seismology, geochemistry, and mineral physics. Computational mineral physics has made great progress recently, with the development of quantum-mechanical methods allowing one to study complex chemistry pertinent to the Earth, and make unbiased predictions of novel chemical phenomena that take place under high pressures and temperatures typical of planetary interiors. This project aims toward a better understanding Earth's chemistry and physics with the help of advanced quantum-mechanical methods. Such methods are quite universal and recently have led to breakthroughs in the computational design of novel materials, which have a potentially transformative value for our society. This research bridges several the fields of geosciences, materials science, physics, and chemistry, and focuses on problems that excite curiosity of humans ("what is the Earth made of", "how was the Earth formed"). Such studies are impossible without the energy of young scientists who will get unique training in this multidisciplinary field of science and will be perfectly equipped to serve the world in any of these fields. The study of the Earth's deep interior is one of the most fascinating fields of modern research. Currently, based on several decades continuing works by geoscientists, reasonable mineralogical models of the deep Earth are finally constructed. Thermal and compositional variations in the deep Earth can be well estimated by data from seismic tomography and mineral physics. Mineral physics provides information about the physical properties of Earth-forming minerals together with their dependence on temperature and composition, and phase diagrams of mineral systems. Exploration of the Earth-forming minerals is one of the most critical issues in Earth sciences. Despite tremendous progress in experimental high-pressure techniques, most physical measurements at pressures of the Earth's deep mantle and core are problematic. Recently, (Mg,Fe)SiO3 perovskite, long believed to be stable in the Earth's deep lower mantle, was found to decompose into MgSiO3 perovksite and an Fe-rich hexagonal phase (H-phase, the crystal structure of which is still debatable). The PI's preliminary results show the surprising result that both ilmenite MgSiO3 and perovskite MgSiO3 are thermodynamically unstable at 30 GPa 0 Kelvin in the Mg-Si-O system, and no other ternary stable compounds exist at 30 GPa at all, which indicates a very important unexpected phenomenon can be observed and significant implications for enigmatic seismic features might be expected. Furthermore, besides the well-known seismic discontinuities at 410, 520, 660 km, another seismic discontinuity at 800 km is also seen in widely varying regions, but so far has no reasonable explanation, and another unknown phase might be expected. Minerals in the mantle are probably more complex than what we know currently. Theoretical simulations based on quantum mechanics have played an increasingly important role in Earth sciences. Therefore, the PI will apply the most advanced and recently developed tools in computational physics to address problems related to the mantle, core and core-mantle boundary region.The PI will focus on three problems: (1) Methodology development toward the prediction of materials with variable stoichiometry in ternary and even higher multi-component systems. The PI's recent work has shown that the prediction of binary systems based on the evolutionary approach is tractable. Since a number of important minerals in the context of earth sciences have puzzling stoichiometry, the remaining challenge is to predict stable ternary, quaternary and even higher multi-component systems at high P-T conditions. (2) Prediction of Earth-forming minerals in the Mg-Fe-Ca-Al-Si-O and Fe-Si-C-S-H-O systems under given P-T conditions. Since the perovskite and post-perovskite phases in the real mantle have complex compositions, additional elements, like Ca, Al and C, can also affect the structure and stability field of phases. Therefore, crystal structure predictions for multicomponent systems will be addressed. Properties of Earth-forming minerals in the mantle and core will be computed. The nature of unexplained seismic discontinuities and H-phase are expected to be uncovered and some novel minerals are highly likely to be discovered in the mantle and core.(3) Chemical reactions between the mantle and core. This study should greatly deepen our understanding of iron-based alloys, will lead to better understanding of chemistry and mineralogy of the Earth's core, as well as its peculiar properties (such as strong seismic anisotropy). Understanding of possible reactions at the core-mantle boundary region may shed light on some of the anomalous properties of that region. In short, this project will look at materials of the most enigmatic regions of the Earth using advanced quantum-mechanical simulations. Central to this study is the PI's evolutionary algorithm USPEX, interfaced with first-principles electronic structure calculations. This approach has produced many important predictions in mineralogy and physics, successfully confirmed by subsequent experiments.