CDS&E: Fast Computational Methods for Quantum Simulation of 2D Spintronic and Electronic Devices

非技术性：人工智能和物联网等新技术推动了对计算机处理和数据存储的需求不断增长。因此，迫切需要开发基于新的物理机制和材料的半导体存储器和计算设备，以提高性能和功率效率。自旋电子学使用电子自旋而不是电荷的物理性质，这对数据存储和传输的效率有影响。原子薄的二维（2D）半导体是具有独特物理性质的有前途的材料。这些新现象和材料为开发新器件和电路提供了希望。计算机辅助仿真和设计在实现现代电子产品方面发挥了重要和关键的作用。快速计算方法和计算机辅助设计工具的新类别的二维自旋电子和电子器件，但是，仍然在很大程度上未开发。这阻碍了它们在未来计算和数据存储技术中的采用。为了解决这个问题，新的模拟方法和新的计算技术将在这个项目中开发。这将使快速量子模拟和2D自旋电子和电子器件的设计成为可能。这项研究的结果将对先进设备设计、高性能计算、低功耗电子设备、新型存储设备和柔性电子设备等研究领域产生直接影响。该项目的目标是开发快速的计算方法和近似方法，以显著降低计算复杂性和提高计算效率，用于二维器件的量子模拟，并探索其在二维自旋电子器件和电子器件的模拟和设计中的应用和局限性。拟议的研究活动包括：（i）开发用于2D器件模拟的解开技术，其将非平衡格林函数模拟中的耦合矩阵方程转变为非耦合随机波矢量方程，（ii）通过利用大规模并行通用图形处理单元计算平台来开发量子输运方程的可扩展的高性能解决方案，（iii）通过利用两种近似方法有效地处理2D自旋电子器件和电子器件的横向尺寸，（iv）开发相关函数混合方法以实现快速和稳定的收敛，和（v）应用上述计算方法来开发新的模拟能力和工具，用于过渡金属二硫属化物的测试套件和基于拓扑绝缘体（TI）。该项目开发了量子器件模拟计算方法的基本知识基础，并为基于二维半导体和拓扑绝缘体独特物理特性的器件的高效计算模拟工具铺平了道路。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

Nontechnical:New technologies such as artificial intelligence and the Internet of Things have driven ever increasing demands for computer processing and data storage. There is accordingly an imperative need to develop semiconductor memory and computing devices based on new physical mechanisms and materials for improved performance and power efficiency. Spintronics uses the physical property of electron spin rather than charge, with implications in the efficiency of data storage and transfer. Atomically thin two-dimensional (2D) semiconductors are promising materials with unique physical properties. These new phenomena and materials hold promise for developing new devices and circuits. Computer-aided simulation and design have played an essential and critical role in enabling modern electronics. Fast computational methods and computer-aided design tools for the new classes of 2D spintronic and electronic devices, however, have remain largely undeveloped. This hinders their adoption in future computing and data storage technologies. To address this deficiency, new simulation methods and novel computational techniques will be developed in this project. These will enable fast quantum simulations and design of 2D spintronic and electronic devices. The findings from the proposed effort will have direct impact on research areas such as advanced device design, high-performance computing, low-power electronics, new memory devices, and flexible electronics. Simulation codes and tools will be developed and shared with the research and education community, and students from high school to graduate levels will be engaged in this project.Technical:The goals of the project are to develop fast computational methods and approximations to significantly reduce the computational complexity and improve the computational efficiency for quantum simulation of 2D devices and to explore their applications and limitations in simulation and design of 2D spintronic and electronic devices. The proposed research activities include: (i) develop an unraveling technique for 2D device simulations, which turns the coupled matrix equations in the non-equilibrium Greens function simulations to uncoupled stochastic wave vector equations, (ii) develop scalable, high-performance solutions to the quantum transport equation by taking advantage of the massively parallel general purpose graphics processing unit computational platform, (iii) efficiently treat the transverse dimension of the 2D spintronic and electronic devices by exploiting two approximate methods, (iv) develop a correlation function mixing method to achieve fast and stable convergence, and (v) apply the above computational methods to develop new simulation capabilities and tools for a test suite of transition metal dichalcogenide and devices based on topological insulators (TIs). The project develops the essential knowledge base for computational methods in quantum device simulations, and paves the way toward computationally efficient simulation tools for devices based on the physical properties unique to 2D semiconductors and topological insulators.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.