SI2-SSE: Expanding the Scope of Materials Modeling with Electron Phonon Wannier (EPW) Software

引入有效的非经验计算方法来建模和预测先进的材料特性，是正在进行的加速理论指导材料发现的核心。开源软件电子-声子-万尼尔（EPW）代码提供了在量子力学水平上高精度计算特性的独特功能。特别是，EPW提供了对控制电子和原子振动之间相互作用的微观机制的见解。在这个项目中，现有的能力将扩展到具有复杂电子和磁性质的更广泛的材料模型。这些知识可以用来设计新一代的材料，用于收集太阳能和热能，从电子学到自旋电子学的过渡，或者实现物质的奇异状态。EPW将为物理学家、材料科学家、化学家和工程师提供广泛的电子结构社区服务，他们致力于热电、光伏、超导、自旋电子和其他应用的下一代材料的建模和设计。高性能材料的开发对于解决与能源和环境、交通、信息和通信技术相关的新兴社会挑战至关重要。开发的计算工具将在GNU通用公共许可证下发布，以确保科学界将直接和及时地从这项技术中受益。项目内提出的广泛的教育和推广活动将促进和普及不同社区的科学研究。计划在美国和欧洲举办的EPW实践研讨会将有助于创建一个强大的EPW社区，以进一步开发代码，并促进来自不同国家的参与者之间的新研究合作。针对纽约北部地区小学生的互动活动将有助于吸引新一代科学家从代表性不足的群体进入STEM学科。目前电子结构界的焦点是引入新的能力，使热电、光伏、超导、自旋电子和其他应用的新兴高性能材料的设计成为可能。众所周知，在这些应用中，函数定义属性很难用目前基于密度泛函理论的方法以期望的精度进行评估。该项目的目的是扩展开源软件电子-声子-万尼尔（EPW）在材料研究领域的功能和影响。EPW，现在作为Quantum ESPRESSO套件的一部分分发，已经成为一种独特的计算工具，提供标准电子结构软件包中不可用的功能。通过结合密度泛函微扰理论和最大局域万尼尔函数方法，EPW使得计算数百万电子-声子矩阵元素在计算上是可行的。拟议的工作将扩展EPW代码的当前能力，以模拟一类重要的自旋相关材料的特性。所建议的方法和面向用户的目标选择与SSE计划和DMR的重点方向保持一致，这些方向与针对大型用户群的健壮软件的创建、扩展和部署有关。特别是，自旋输运、自旋弛豫和自旋动力学的预测计算可以为原子尺度上的过程提供基本的见解，并为合理设计新材料和指导实验工作提供必要的基础。该项目还将为大规模计算基础设施上的日常科学实验提供方便的管理和执行。引入自动化、存储、管理和共享模拟的工作流程将促进数据透明度和通信，并将数据驱动的材料设计推进到新的领域。这些目标的成功实现将大大增强EPW代码的功能，并确保EPW用户社区的持续增长。开发的计算工具将在GNU通用公共许可证下发布，以确保科学界将直接和及时地从这项技术中受益。研究计划将与教育和推广活动紧密结合。它将使学生在先进的电子结构方法、计算材料科学和高性能计算方面的跨学科训练成为可能。其他工作将包括向小学生进行科学演示，开发一门特殊的材料建模课程，将计算机模拟纳入宾厄姆顿大学的本科和研究生课程，并组织研讨会，教授EPW代码的基本理论和最佳使用。该项目由计算机与信息科学与工程理事会的先进网络基础设施办公室和数学与物理科学理事会的材料研究部提供支持。

Introduction of efficient non-empirical computational methods for modeling and predicting advanced materials properties is at the heart of the on-going effort to accelerate theory-guided materials discovery. The open-source software Electron-Phonon-Wannier (EPW) code offers unique capabilities for high-accuracy calculations of properties at the quantum mechanical level. In particular, EPW provides insight into the microscopic mechanisms that govern the interaction between electrons and atomic vibrations. Within this project, the existing capabilities will be extended to model a wider range of materials with complex electronic and magnetic properties. The knowledge can be used to design new-generation materials for harvesting of solar and thermal energy, making transition from electronics to spintronics, or realizing exotic states of matter. EPW will serve the broad electronic structure community of physicists, materials scientists, chemists, and engineers who work on modeling and designing next-generation materials for thermoelectric, photovoltaic, superconducting, spintronic, and other applications. The development of high-performance materials is crucial for addressing emergent societal challenges related to energy and environment, transportation, and information and communication technologies. The developed computational tools will be released under the GNU General Public License to ensure that the scientific community will directly and timely benefit from this technology. A broad spectrum of educational and outreach activities proposed within the project will promote and popularize scientific research in diverse communities. Planned hands-on workshops on EPW in the US and Europe will help create a strong EPW community for further development of the code and foster new research collaborations among participants from different countries. Interactive events for elementary school students in the upstate New York area will help attract a new generation of scientists from underrepresented groups into the STEM disciplines.The current focus of the electronic structure community is to introduce new capabilities enabling the design of emerging high-performance materials for thermoelectric, photovoltaic, superconducting, spintronic, and other applications. Function-defining properties in these applications are notoriously difficult to evaluate with desired accuracy using present density functional theory-based methods. The aim of this project is to expand the functionalities and broaden the impact of the open-source software Electron-Phonon-Wannier (EPW) in the area of materials research. EPW, now distributed as part of the Quantum ESPRESSO suite, has emerged as a unique computational tool that offers functionalities not available in standard electronic structure packages. By combining density-functional perturbation theory and maximally-localized Wannier functions methods, EPW makes it computationally feasible to calculate millions of electron-phonon matrix elements. The proposed work will expand the current capabilities of the EPW code to modeling an important class of spin-dependent materials properties. The proposed methodological, and user-oriented objectives are chosen to align with the focal directions of the SSE program and the DMR pertaining to creation, expansion, and deployment of robust software targeting a large user base. In particular, predictive calculations of spin transport, spin relaxation, and spin dynamics can yield fundamental insights into processes at the atomic scale and provide the necessary foundation to rationally design new materials and guide experimental work. The project will also provide easy management and execution of day to day scientific experiments on large-scale computing infrastructures. The introduction of workflows for automating, storing, managing, and sharing simulations will facilitate data transparency and communication as well as advance data-driven materials design to new frontiers. Successful implementation of these objectives will substantially enhance the functionalities of the EPW code and ensure the continued growth of the EPW user community. The developed computational tools will be released under the GNU General Public License to ensure that the scientific community will directly and timely benefit from this technology. The research program will be tightly integrated with educational and outreach activities. It will enable interdisciplinary training of students in advanced electronic structure methods, computational material science, and high-performance computing. Other efforts will include science demonstrations to elementary school students, development of a special course in materials modeling to incorporate computer simulations in the Binghamton University's undergraduate and graduate curriculum, and organization of workshops to teach the underlying theory and optimal usage of the EPW code.This project is supported by the Office of Advanced Cyberinfrastructure in the Directorate for Computer & Information Science and Engineering and the Division of Materials Research in the Directorate of Mathematical and Physical Sciences.