Elements: Cyberinfrastructure for spin and charge transport calculation of partially disordered alloys

金属合金在高科技、工业甚至我们的家庭中无处不在。当两种或多种化学物质结合形成单一金属相时形成的合金，提供了改善纯金属元素性能的机会。通过合金化可以改变的一些性质包括机械强度、磁性、熔化温度和抗氧化性。该项目的重点是导电性，通过开发计算机代码来评估电子从原子的量子力学散射。为了准确预测散射，我们必须知道原子的位置，以及在合金中的位置，这意味着要了解不同的化学物质在空间中的排列方式。通常这些排列是随机的，但即使元素是随机分布的，由于化学键的偏好，某些物种相对于其他物种的位置也会有相关性。该代码将包含使其能够预测这些相关性并揭示相关性如何影响电导率的功能。除了开发计算机代码外，该项目还将建立一个有兴趣并能够运行代码并为其进一步开发作出贡献的科学家用户群。对高中学生及其教师的宣传将加强未来科学家的渠道。在PI教授的大学和研究生课程中包含散射理论将为物理和材料科学学生理解和应用代码做好准备。在科学协会会议上的介绍将向现有社区介绍这些能力，而讲习班和网络研讨会将为活跃用户提供具体培训。电子密度泛函理论（DFT）作为计算能量、力和电子结构的实用工具而蓬勃发展;它现在在基础科学和工程中的应用都很广泛。电荷和自旋输运计算是一种尚未达到更广泛用户群体的能力，部分原因是包含这些效应的代码并不广泛可用，部分原因是这些特性对晶体有序度高度敏感。通常缺乏关于有序度的基本知识，因为它可能与温度有关，并且大多数DFT代码无法捕获热效应。为了满足这一需求，将开发一种易于使用（能够在台式计算机上运行）的代码，该代码可以预测作为温度函数的化学有序或无序程度，并且可以计算所得的电荷和自旋电导率。这将通过建立在电子结构计算的创新基础上，再加上统计力学方法来解决热无序问题来实现。具体来说，我们将修改相干势近似（CPA），将短程有序的影响，通过统一的总能量与簇变分法（CVM）预测温度依赖的障碍。修改后的CPA将总能量表示为原子间相关函数的函数，而CVM将以相同的术语表示熵，允许确定平衡能量与熵的相关性。这种密度泛函理论的方法采用了在公共领域代码MuST中实现的多重散射。这种方法确定了电子绿色函数，因此它自然地与Kubo和Greenwood电荷和自旋电导率公式相结合。这种内部一致的近似组合将实现高精度和高性能。该项目由计算机和信息科学与工程局高级网络基础设施办公室资助，数学和物理科学局材料研究部也提供资金。该奖项反映了NSF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。

Metallic alloys are ubiquitous in high technology, in industry, and even in our households. Alloys, which form when two or more chemical species are combined to create a single metallic phase, offer the chance to improve upon the properties of pure metallic elements. Some of the properties that can be altered through alloying include mechanical strength, magnetism, melting temperature, and oxidation resistance. This project focuses on the property of electrical conductivity, by developing computer codes to evaluate the quantum mechanical scattering of electrons off of atoms. To accurately predict the scattering we must know where the atoms are located, and in an alloy that means understanding how the different chemical species arrange in space. Often these arrangements are random, but even if the elements are randomly distributed, there will be correlations in the positions of certain species relative to others as a result of chemical bonding preferences. The code will contain features that enable it to predict these correlations and reveal how the correlations influence the conductivity. In addition to developing computer codes, this project will develop a user base of scientists interested and able to run the code and to contribute to its further development. Outreach to high school students and their teachers will enhance the pipeline of prospective scientists. Inclusion of scattering theory in college and graduate level courses taught by the PIs will prepare Physics and Materials Science students to understand and apply the codes. Presentations at scientific society conferences will inform the existing community of the capabilities, while workshops and webinars and webinars will provide specific training for active users. Electronic density functional theory (DFT) has flourished as a practical tool for calculating energies, forces and electronic structure; its use is now widespread both in basic science and in engineering. Charge and spin transport calculation is a capability that has not yet reached the broader user community, partly because codes that incorporate these effects are not widely available and partly because these properties are highly sensitive to the degree of crystalline order. Basic knowledge of the degree of order is often lacking, as it can be temperature dependent, and thermal effects are not captured by most DFT codes. To address this need, a code will be developed that is easy to use (capable of running on a desktop computer), that can predict the degree of chemical order or disorder as a function of temperature, and that can calculate the resulting charge and spin conductivities. This will be achieved by building upon innovations in electronic structure calculation, coupled with methods of statistical mechanics to address thermal disorder. Specifically, we will modify the Coherent Potential Approximation (CPA) to incorporate the effects of short range order by unifying the resulting total energies with the Cluster Variation Method (CVM) to predict temperature dependent disorder. The modified CPA will express the total energy as a function of interatomic correlation functions, while the CVM will express the entropy in the same terms, allowing the determination of correlations that balance the energy against the entropy. This approach to density functional theory employs multiple scattering as implemented in the public domain code MuST. This method determines the electronic Green’s functions, and consequently it integrates naturally with the Kubo and Greenwood formulas for charge and spin conductivity. This internally consistent combination of approximations will achieve both high accuracy and high performance. This project is funded by the Office of Advanced Cyberinfrastructure in the Directorate for Computer and Information Science and Engineering, with the Division of Materials Research in the Directorate for Mathematical and Physical Sciences also contributing funds.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.