CAREER: Dielectric Screening - From First Principles to Mesoscale

该职业奖支持使用计算机模拟设计和发现新型光学材料的研究和教育。原子和电子的量子力学性质在微观长度和时间尺度上支配着材料的行为，但在大尺度上，完全原子化的方法是难以解决的。跨越不同长度尺度的计算机模拟有望使基础研究更接近制造业：新材料和技术应用可以更快、更便宜地开发出来。对于光学材料来说尤其如此，例如，光电设备和能源生产、显示器以及用于生物和医疗应用的新型高效传感器都需要光学材料。在这个项目中，研究小组试图开发一种跨多个长度尺度的光学材料的计算设计和发现方法。研究人员将改进用于实现光学吸收的量子力学描述的近似方法。反过来，这将使团队能够成功地描述电子和离子之间具有强相互作用的材料，这是当前非常感兴趣的主题。基于这些预测模拟的结果，该团队将放弃模型的原子分辨率，以模拟纳米尺度结构的光学材料。该团队将把最近开发的公共在线数据库中的数据整合到计算框架中，旨在发现新的候选材料，并研究它们对现代光学应用的适用性。投入和产出的研究数据集将公开提供，以供进一步核查和确认，并可能用于不可预见的用途，例如在数据挖掘方面。虽然超级计算机的广泛使用改变了现代研究的方式，但熟练的跨学科研究人员正在成为必不可少的组成部分。因此，该研究项目与教育活动紧密结合，以培养材料和计算机科学联系的下一代劳动力。PI将把计算机模拟纳入伊利诺伊大学厄巴纳-香槟分校的本科课程，并开发可直接用于课堂的计算学习模块。PI还将建立一个由本科生研究人员组成的跨学科团队，开发基于虚拟现实的教育技术，并将其推广到包括高中生和本科生在内的广大受众。这将通过使用智能手机进行三维仿真结果的交互式可视化的新技术来实现。所有为研究开发的代码和实现、用于教育的计算模块以及用于高中教育和推广的虚拟现实应用程序都将被记录在案，并供更广泛的研究和教育界使用。技术概述：该职业奖支持计算材料科学框架发展的研究和教育，具有跨多个长度尺度的预测能力。该框架将使研究小组能够准确地描述光学特性，并允许计算机辅助设计和发现新型光学材料。研究人员将扩展目前使用的第一原理量子力学技术，以克服导致光学吸收光谱的巨大不确定性的限制，特别是极性材料中的激子效应。他们将通过求解麦克斯韦方程来实现纳米结构材料的多尺度预测，并且他们将开发方法来筛选激发态特性的在线数据库。这个框架将允许完全基于计算机模拟的新型光学材料的设计和发现。为了实现该项目的目标，研究小组将基于无参数量子力学模拟，详细了解复杂的电子-电子和电子-离子相互作用。他们将通过考虑自由载流子和晶格的贡献来改进介电屏蔽的描述，克服新型极性材料的不确定性。在基于麦克斯韦方程的纳米结构材料建模中使用这些结果消除了对实验输入的依赖，并将原子第一原理模拟扩展到纳米尺度。该团队将利用基于密度泛函理论的大型在线数据库中的现有数据来了解激发态特性，并促进在数万个可用数据集中发现新的光学材料。该项目将开发普遍适用的技术，并利用它们来了解具有理想特性的极性材料，例如用于等离子体应用。虽然超级计算机的广泛使用改变了现代研究的方式，但熟练的跨学科研究人员正在成为必不可少的组成部分。因此，该研究项目与教育活动紧密结合，以培养材料和计算机科学联系的下一代劳动力。PI将把计算机模拟纳入伊利诺伊大学厄巴纳-香槟分校的本科课程，并开发可直接用于课堂的计算学习模块。PI还将建立一个由本科生研究人员组成的跨学科团队，开发基于虚拟现实的教育技术，并将其推广到包括高中生和本科生在内的广大受众。这将通过使用智能手机进行三维仿真结果的交互式可视化的新技术来实现。所有为研究开发的代码和实现、用于教育的计算模块以及用于高中教育和推广的虚拟现实应用程序都将被记录在案，并供更广泛的研究和教育界使用。

NONTECHNICAL SUMMARYThis CAREER award supports research and education in designing and discovering novel optical materials using computer simulations. The quantum mechanical nature of atoms and electrons governs the behavior of materials at microscopic length- and time scales, but at large scales a fully atomistic approach is intractable. Computer simulations that bridge different length scales carry the promise of bringing fundamental research closer to manufacturing: New materials and technological applications could be developed faster and cheaper. That is particularly true for optical materials that are, for instance, needed for photovoltaic devices and energy production, displays, and novel efficient sensors for biological and medical applications.In this project, the research team seeks to develop an approach for computational design and discovery of optical materials across multiple length scales. The researchers will improve approximations that are used to achieve a quantum mechanical description of optical absorption. This in turn will enable the team to successfully describe materials with strong interactions between electrons and ions, a subject of intense current interest. Based on the results of these predictive simulations, the team will abandon the atomistic resolution of the model in order to simulate optical materials that are structured at nanoscale dimensions. The team will incorporate data from recently developed public online databases into the computational framework aiming to discover new candidate materials and investigate their suitability for modern optical applications. The input and output research data sets will be made publicly available for advanced verification and validation, and possibly unforeseen uses, e.g. in data-mining.While the broader availability of supercomputers transforms how modern research is done, skilled, interdisciplinary researchers are becoming an essential component. For that reason, the research project is tightly integrated with educational activities to train the next-generation workforce at the nexus of materials and computer science. The PI will incorporate computer simulations into the undergraduate curriculum at the University of Illinois at Urbana-Champaign, and develop computational learning modules that can be directly used in the classroom. The PI will also build an interdisciplinary team of undergraduate researchers to develop virtual-reality-based techniques for education and exciting outreach to a large audience, which includes high school and undergraduate students. This will be achieved by new techniques for interactive visualization of simulation results in three dimensions using a smartphone. All codes and implementations developed for research, the computational modules for education, and the virtual-reality apps for education and outreach in high schools will be documented and made accessible to the broader research and education community.TECHNICAL SUMMARYThis CAREER award supports research and education in the development of a computational materials science framework with predictive power across multiple length scales. The framework will enable the research team to accurately describe optical properties and allow for computer-aided design and discovery of novel optical materials. The researchers will extend presently used first-principles quantum-mechanical techniques to overcome limitations that lead to large uncertainties for optical absorption spectra, and especially for excitonic effects in polar materials. They will achieve multiscale predictions for nanostructured materials by solving Maxwell equations, and they will develop approaches to screen online data repositories for excited-state properties. This framework will allow the design and discovery of novel optical materials, entirely based on computer simulations.To achieve the goals of this project, the research team will develop detailed understanding of complicated electron-electron and electron-ion interactions based on parameter-free quantum-mechanical simulations. They will improve the description of dielectric screening by accounting for contributions from free carriers and the lattice, overcoming uncertainties for novel polar materials. Using these results in Maxwell-equation-based modeling of nanostructured materials eliminates the dependence on input from experiment, and extends the atomistic first-principles simulations into the nanoscale regime. The team will use existing data from large density-functional theory based online databases to learn about excited-state properties, and to facilitate the discovery of new optical materials within tens of thousands of available datasets. This project will develop generally applicable techniques and use them to understand polar materials with desirable characteristics, e.g. for plasmonic applications.While the broader availability of supercomputers transforms how modern research is done, skilled, interdisciplinary researchers are becoming an essential component. For that reason, the research project is tightly integrated with educational activities to train the next-generation workforce at the nexus of materials and computer science. The PI will incorporate computer simulations into the undergraduate curriculum at the University of Illinois at Urbana-Champaign, and develop computational learning modules that can be directly used in the classroom. The PI will also build an interdisciplinary team of undergraduate researchers to develop virtual-reality-based techniques for education and exciting outreach to a large audience, which includes high school and undergraduate students. This will be achieved by new techniques for interactive visualization of simulation results in three dimensions using a smartphone. All codes and implementations developed for research, the computational modules for education, and the virtual-reality apps for education and outreach in high schools will be documented and made accessible to the broader research and education community.