SI2-SSI: Collaborative Research: A Robust High-Throughput Ab Initio Computation and Analysis Software Framework for Interface Materials Science

一组被科学家称为“从头算”方法的技术，是从物理学的基本定律中衍生出来的，用最少的假设和近似，已经成为研究和设计材料的关键工具。随着计算的进步和软件的创新，高通量从头计算的自动化，特别是，预示着各种材料的计算数据的爆炸式增长。然而，这些高通量的努力仅限于特定的属性。相比之下，材料界面是材料科学与工程中发展最快的研究领域之一，在催化和电子等材料应用的许多领域中显示出越来越多的相关性。该项目将开发一个软件框架，使新型高通量界面材料的调查和设计成为可能。开发的软件平台将通过计算界面材料的界面特性来扩展材料基因组。这种基于社区的软件可以潜在地成为材料基因组计划的关键组成部分，不仅为大型和多样化的材料研究团体服务，也为物理和化学团体服务。除了主要研究人员在其所属院校所讲授的现有及计划中的课程，建议的架构亦有助其他院校的本科生及毕业生接受从头计算方法的训练。该项目还将开展公众外展活动，以提高对数据驱动界面材料科学可持续软件开发重要性的认识。该项目将开发必要的工作流程管理、纠错方案和系统分析工具，以支持热力学、动力学、扩散和界面材料（包括异质结构和晶界）电子特性的从头算研究。它的目标是对界面材料科学非常感兴趣的三个关键重点领域的发展努力：(i)从头算表面和界面热力学；ii)材料动力学和材料界面扩散的先进方法；iii)晶界结构构建及后期数据处理分析自动化算法。在此过程中，该项目将极大地扩展适合高通量从头开始处理的界面材料特性，为广泛技术应用中的材料研究和设计铺平道路，包括能源产生和存储、催化和电子。此外，通过与经典力学模拟代码的接口，该框架将弥合从头算和经典力场方法之间的差距，有望显著推进材料界面的高通量模拟。该奖项由高级网络基础设施部颁发，由美国国家科学基金会数学和物理科学理事会（材料研究部和多学科活动办公室）联合支持。

A set of techniques scientists call "ab initio" methods, which are derived from the fundamental laws of physics with minimal assumptions and approximations, has become a critical tool in the study and design of materials. With computing advances and software innovations, the automation of high-throughput ab initio calculations has, in particular, heralded an explosion of computed data for a large variety of materials. However, these high-throughput efforts are limited to specific properties. In contrast, materials interfaces, one of the fastest growing research areas in materials science and engineering, are showing an increasing relevance in many areas of materials applications such as catalysis and electronics. This project will develop a software framework that enables novel high-throughput interface materials investigations and design. The developed software platform will expand the genome of materials by including the computed interfacial properties of interface materials. This community-based software can potentially become a critical component of the Materials Genome Initiative and serve not just the large and diverse materials research community, but also the physics and chemistry communities. Besides featuring heavily in existing and planned courses taught by the Principal Investigators in their home institutions, the proposed framework will facilitate the training of undergraduates and graduates in the ab initio methodologies in other institutions as well. This project will also conduct public outreach activities to increase awareness of the importance of sustainable software development for data-driven interface materials science. The project will develop necessary workflow management, error correction schemes, and systematic analysis tools to support ab initio studies of thermodynamics, kinetics, diffusion, and electronic property of interface materials including hetero-structures and grain boundary. It targets developmental efforts on three key focus areas of great interest to interface materials science: (i) Ab initio thermodynamics of surfaces and interfaces; ii) Advanced methods for materials kinetics and diffusion at materials interfaces; and iii) Automated algorithms for structural construction of grain boundary and post data-processing and analysis. In doing so, this project will greatly expand the suite of interfacial materials properties that are amenable to a high-throughput ab initio treatment, paving the way for materials investigations and design in a broad spectrum of technological applications, including energy generation and storage, catalysis and electronics. In addition, by interfacing with classical-mechanics simulation codes, this framework will bridge the gap between the ab initio and classical force-field approach, which is expected to significantly advance the high-throughput simulations of materials interfaces. This award by the Advanced Cyberinfrastructure Division is jointly supported by the NSF Directorate for Mathematical and Physical Sciences (Division of Materials Research and Office of Multidisciplinary Activities).