CDS&E: Numerical Investigation of Two-Particle Response Functions of Correlated Materials

非技术总结材料研究司和先进网络基础设施司共同资助这一奖项，该奖项支持旨在理解材料的实验测量和数值模拟之间的关系的理论和计算研究，在这些研究中，强电子关联效应是重要的。特别是，该团队将专注于获得模拟各种实验的计算数据，如核磁共振、中子光谱学和共振非弹性x射线散射。材料是由排列在晶格中的电子和离子组成的。在某些材料中，电子的运动强烈地依赖于或与许多其他电子的运动相关。因此，这些材料可能会表现出不寻常的行为，包括超导、磁性以及有趣的电学和光学性质。虽然这些性质使它们在技术应用中变得有用和有趣，但我们描述这种材料的标准理论工具是不够的，需要数值方法。该项目的主题是开发和应用准确和受控的数值方法，这些方法可以描述相关的电子系统并计算实验测量的数量。该项目将通过支持可持续开放源码社区软件库的开发和维护而产生更广泛的影响，这将加速未来代码的开发，并为科学界提供可靠和最先进的应用程序。目前由PI维护的软件库是少数几个已建立的用于强关联系统计算的开源库之一。作为该项目的一部分，研究生将接受现代理论技术和科学软件开发方面的培训。技术总结材料研究部和高级网络基础设施部共同资助该奖项，该奖项支持理论和计算研究，旨在了解在强相关性很重要的系统中，实验测量的两粒子响应函数和有效低能晶格模型的广义磁化率之间的关系。该项目将把新开发的数值方法与大规模计算相结合，以计算核磁共振(核磁共振)、中子光谱和共振非弹性X射线光谱(RIXS)的响应函数，并将调查观测信号可归因于电子相关效应的程度。极化率揭示了有关系统集体激发的重要信息，并且在实验中是可以直接测量的。然而，由核磁共振、RIXS或中子散射等双粒子探测器测量的广义极化率的强关联物理在理论上还没有被很好地理解。因此，对实验结果的解释存在几个矛盾之处。这个项目将建立在已被证明对在光电发射和扫描隧道显微镜中测量的单粒子激发是可靠的方法的基础上，并将它们扩展到研究两粒子关联函数。通过为费米子晶格模型的磁化率提供可靠的结果，并通过计算响应函数，这项工作将有助于将真正的电子关联物理从依赖于模型的人工产物中分离出来，这反过来将有助于解释实验。相关的电子材料对于信息技术、能源技术、材料科学和纳米科学等现代技术应用是必不可少的。通过澄清磁化率的基本行为及其与实验工作的关系，这个项目将有助于我们对相关材料及其表征的理解。该项目将通过支持可持续开放源码社区软件库的开发和维护来促进更广泛的影响，这将加快未来代码的开发，并向科学界提供可靠和最先进的应用程序。目前由PI维护的软件库是少数几个已建立的用于强关联系统计算的开源库之一。作为该项目的一部分，研究生将接受现代理论技术和科学软件开发方面的培训。

NONTECHNICAL SUMMARYThe Division of Materials Research and the Division of Advanced Cyberinfrastructure jointly fund this award, which supports theoretical and computational research aimed at understanding the relationship between experimental measurements and numerical simulations of materials in which strong electronic correlation effects are important. In particular, the team will focus on obtaining computational data simulating a variety of experiments such as nuclear magnetic resonance, neutron spectroscopy, and resonant inelastic x-ray scattering. Materials consist of electrons and ions arranged in a crystal lattice. In some materials, the motion of an electron is strongly interdependent on, or correlated with, the motion of many other electrons. As a consequence, these materials may exhibit unusual behavior including superconductivity, magnetism, and interesting electrical and optical properties. While these properties make them useful and interesting for technical applications, our standard theoretical tools for describing such materials are inadequate, and numerical methods are needed. The subject of this project is the development and application of accurate and controlled numerical methods that can describe correlated electron systems and compute experimentally measured quantities.The project will contribute to broader impacts by supporting the development and maintenance of sustainable open-source community software libraries, which will accelerate the development of future codes as well as provide reliable and state-of-the-art applications to the science community. The software libraries currently maintained by the PI are among the few established open-source libraries for calculations on strongly correlated systems. As part of the project, graduate students will be trained in modern theoretical techniques and in scientific software development.TECHNICAL SUMMARYThe Division of Materials Research and the Division of Advanced Cyberinfrastructure jointly fund this award, which supports theoretical and computational research aimed at understanding the relation between experimentally measured two-particle response functions and generalized susceptibilities of effective low-energy lattice models in systems where strong correlations are important. The project will combine newly developed numerical methods with large-scale calculations to compute the response functions of nuclear magnetic resonance (NMR), neutron spectroscopy, and resonant inelastic x-ray spectroscopy (RIXS) and will investigate the extent to which observed signals can be attributed to electron correlation effects. Susceptibilities reveal important information about collective excitations of a system and are directly measurable in experiment. However, the strong correlation physics of generalized susceptibilities measured by two-particle probes such as NMR, RIXS, or neutron scattering is theoretically not well understood. As a consequence, several contradictions in the interpretation of experimental results exist. This project will build on methods that have shown to be reliable for single-particle excitations that are measured in photoemission and scanning tunneling microscopy, and extend them to investigate two-particle correlation functions. By providing reliable results for the susceptibilities of fermionic lattice models and by computing response functions, this work will facilitate the separation of true electron correlation physics from model-dependent artifacts, which in turn will aid in the interpretation of experiments. Correlated electron materials are essential for modern technological applications such as information technology, energy technology, materials science, and nanoscience. By clarifying the basic behavior of susceptibilities and their relation to experimental work, this project will contribute to our understanding of correlated materials and their characterization. The project will contribute to broader impacts by supporting the development and maintenance of sustainable open-source community software libraries, which will accelerate the development of future codes as well as provide reliable and state-of-the-art applications to the science community. The software libraries currently maintained by the PI are among the few established open-source libraries for calculations on strongly correlated systems. As part of the project, graduate students will be trained in modern theoretical techniques and scientific software development.