Computational approaches to quantum materials

我们建议使用计算技术来研究量子材料。量子材料是指强关联或量子力学效应发挥突出作用的材料，在某些情况下会产生新的相和有趣的新物理，在某些情况下，例如石墨烯，具有巨大的商业潜力。我们所说的强关联是指材料中的电子不是独立行为，而是共同运动。从技术上讲，这被称为多体效应，对理论上的理解提出了巨大的挑战，分析结果往往是不可能获得的。幸运的是，可以使用计算方法获得关于这些材料的非常详细的知识。这通常是非常具有挑战性的，但最近在允许使用微观模型模拟这些材料的行为的算法方面的进展使这成为可能。现在已经有可能从宏观上研究大量的粒子，并使用这种技术来完整地描述它们的行为。为了实现这一点，一个重要的计算能力是必要的，而且经常有数百台计算机并行使用，每台计算机都在解决部分问题。就其本身而言，开发这些算法是一个理论特征，而在并行计算机上有效地使用它们是一个技术挑战。这个建议的目的是开发这样的计算工具，并将它们应用于量子材料的研究。

We propose to study quantum materials using computational techniques.  Quantummaterials are materials where strong correlations or quantum mechanical effectsplay a prominent role often giving rise to new phases and interesting newphysics, in some cases, such as graphene,  with a large commerical potential.By strong correlation we mean that the electrons in the material do not behaveindependently but instead all move collectively. Technically this is referredto as a many body effect and presents a tremendous challenge to understandtheoretically and analytical results are often impossible to obtain.Fortunately, a very detailed knowledge of these materials can be obtained usingcomputational methods. This is often very challenging but recent advances inalgorithms that allow for the simulation of the behavior of these materialsusing microscopic models has made this possible. It has now become possible tostudy a macroscopiccally large number of particles and to completelycharacterize their behavior using such techniques. In order to achieve this asignificant computational power is necessary and often several hundredcomputers are being used in parallel, each one of them solving part of theproblem.  In it self the development of these algorithms is a theoretical featand the efficient use of them on parallel computers is a technical challenge.The obejctive of this proposal is to develop such computational tools and toapply them to the study of quantum materials.