CAREER: Finite Temperature Electronic Structure Methods for Predicting Material Phase Diagrams

布朗大学的Brenda Rubenstein是由化学系化学理论、模型和计算方法项目和促进竞争性研究的既定项目（EPSCoR）共同资助的，旨在开发新的方法来模拟相关材料的有限温度电子特性。从催化到天体物理学再到材料科学，分子和材料都可以承受高温。然而，相对较少的理论可以预测这些材料在这种温度下的表现，这阻碍了我们设计高tc超导体和下一代热响应技术等材料的能力。作为该项目的一部分，鲁宾斯坦集团将开发一套互补理论和模拟技术，使有限温度下复杂材料的高精度建模成为可能。除了这项研究之外，鲁宾斯坦集团还将加强其正在进行的努力，通过罗德岛美国化学学会项目SEED和倡导计划（由鲁宾斯坦博士领导）的科学博览会过程，积极指导罗德岛历史上代表性不足的高中生，并通过布朗大学化学系的STEM日计划，为该地区的高中生提供参与真实实验的机会。鲁宾斯坦博士还在开发一门新的加速化学发现课程和相关教科书，目的是让化学科学的本科生熟悉如何利用数据科学来解决化学中的日常问题。本提案的中心目标是开发一套新的有限温度电子结构技术来阐明相关材料的电子相图。在等离子体催化、天体物理学、材料科学和许多其他应用中，分子和材料都受到高温的影响，在高温下，它们的电子分布在广泛的能级上。然而，到目前为止，大多数电子结构方法都集中在基态上，而那些能解释温度的方法要么不能准确地解释电子相关性，要么绝大多数是用来研究晶格模型的，这些模型不能捕捉到真实材料的独特品质。为了应对这一挑战，鲁宾斯坦集团最近开发了一种新的、完全从头算的、完全相关的有限温度辅助场量子蒙特卡罗（FT-AFQMC）方法，该方法被证明能够在各种基准分子和简单固体上产生精确的结果。在这项CAREER提案中，鲁宾斯坦小组的目标是将这项技术扩展到更重的材料的研究中，包括氧化铁、氧化镍、氧化钒和三碘化铬，这些材料的有限温度电子相变尚未完全理解，同时还开发了一套新的支持平均场、选择CI（配置相互作用）和采样技术。这不仅可以提高FT-AFQMC的性能，而且还可以作为更便宜的独立方法。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Brenda Rubenstein of Brown University is jointly funded by the Chemical Theory, Models and Computational Methods program in the Division of Chemistry, and the Established Program to Stimulate Competitive Research (EPSCoR) to develop new methods for modeling the finite temperature electronic properties of correlated materials. In applications ranging from catalysis to astrophysics to material science, molecules and materials can be subjected to high temperatures. Nevertheless, relatively few theories exist to predict how these materials will behave at such temperatures,thwarting our ability to engineer materials such as high-Tc superconductors and next-generation thermally-responsive technologies. As part of this project, the Rubenstein Group will develop a suite of complementary theories and simulation techniques that will enable the high-accuracy modeling of complex materials at finite temperatures. In addition to this research, the Rubenstein Group will bolster its ongoing efforts to actively mentor historically-underrepresented high school students in Rhode Island through the science fair process via the Rhode Island American Chemical Society Project SEED and Advocate Programs (which Dr. Rubenstein leads) and grant area high school students opportunities to engage in real experiments through the Brown University Chemistry Department’s STEM Day Program. Dr. Rubenstein also is developing a new Accelerating Chemical Discovery course and related textbook with the aim of familiarizing undergraduates in the chemical sciences with how data science can be leveraged to solve everyday problems in chemistry.The central aim of this proposal is to develop a suite of new finite temperature electronic structure techniques to elucidate the electronic phase diagrams of correlated materials. In plasmonic catalysis, astrophysics, materials science, and many other applications, molecules and materials are subject to high temperatures at which their electrons populate a wide distribution of energy levels. Nevertheless, most electronic structure methods to date have focused on the ground state, and those methods that do account for temperature either do not accurately account for electron correlation or have overwhelmingly been developed to study lattice models that do not capture the unique qualities of real materials. To rise to this challenge, the Rubenstein Group recently developed a new, fully ab initio, fully correlated finite temperature Auxiliary Field Quantum Monte Carlo (FT-AFQMC) method that proved to be able to yield exact results on a wide variety of benchmark molecules and simple solids. In this CAREER proposal, the Rubenstein Group aims to extend this technique to the study of heavier materials, including iron oxide, nickel oxide, vanadium oxide, and chromium triiodide, whose finite temperature electronic phase transitions have yet to be fully understood, while also developing a set of novel supporting mean field, selected CI (configuration interaction), and sampling techniques. These will not only improve FT-AFQMC’s performance, but can also serve as cheaper, stand-alone methods in their own right.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.