Low Power Logic Operations through Pure Spin Currents

基于自旋的电子学的概念已经达到了一个阶段，它正在先进的科学和高效技术方面做出变革性的变化，如磁电阻场传感器、硬盘驱动器和磁性随机存取存储器。自旋是每个电子与生俱来的量子属性，它已经在每个电子设备中流动，但很少被收获。通过将自旋自由度结合到现有的电子设备中，如中央处理器(CPU)，它们可以获得与当前基于电荷的技术相比的新功能。此外，现代半导体电子已经达到了时钟频率的瓶颈，这主要是因为只能实现有限的散热。例如，Intel CPU在十年前达到了3 GHz的速度，但由于散热限制在4 GHz左右，此后一直停滞不前(同时使用多核解决方案)。这被称为“电源墙”，也就是现代CPU处理能力的定义极限。相比之下，自旋电子器件天生就不受这个问题的影响，也就是说，它们不会从扩散性质产生焦耳加热。因此，用纯自旋电流驱动的设备比传统电子设备的功耗要低得多；然而，自旋电子器件还带来了其他技术挑战。最重要的是通道内和通道之间有限的自旋转移效率，而对通过各种介质和界面的自旋传输过程的基本了解对于进一步开发这些器件至关重要。*拟议的研究计划将在更广泛的原子工程复杂材料和设备中开展开创性的自旋输运研究，预计这些材料和设备将显示出先进的功能，从而导致基于自旋的电子产品更多功能、更节能。由于自旋已经在每个电子设备中流动，从经典电子学到自旋电子学的技术转移只需要对世界上现有的生产线进行最小的改变。因此，该计划开发的技术与现代半导体技术完全兼容，可以很容易地被大公司采用。预计要寻求的行业(潜在合作以及总部职位)包括那些已经启动战略自旋电子开发的信息巨头，如英特尔、IBM、日立、Global Foundries。对加拿大来说，这意味着更多的就业机会和在世界上更多的知名度。归根结底，基于SPIN的处理器并不是要完全取代现代的基于硅的处理器，而是要在现有优势的基础上，只改进非常关键的处理组件。这进一步确保了快速的技术转让。它们将使信息处理更环保、更强大，使摩尔定律回到应有的轨道上来。

The concept of spin-based electronics has reached a stage where it is making transformational changes in advanced science and high-efficiency technology, such as magnetoresistance field sensors, hard disk drive and magnetic random access memory. A spin is an innate quantum property of every electron, and it is already flowing within every electronic device but rarely harvested. By incorporating the spin degree of freedom into existing electronic devices such as central processing units (CPUs), they can gain novel functionalities compared to the present charge-based technologies. Furthermore, modern semiconductor electronics have reached a bottleneck in clock frequency, mainly because only limited heat dissipation is achievable. For example, Intel CPUs reached 3GHz speed a decade ago, yet has been stalled at this level ever since (while using multicore solutions) due to the thermal dissipation limits at about 4GHz. This is termed the “power wall” the defining limit to modern CPU processing power. In contrast, spintronic devices are inherently immune to this issue, i.e., they do not create Joule heating from the diffusive nature. Therefore, devices driven with pure spin currents offer far reduced power consumption than conventional electronic devices; however, spintronic devices present other technological challenges. The limited spin transfer efficiency into and across the channels is the most important, and a fundamental understanding of the spin transport processes through various media and interfaces is critical for further developing these devices. ***The proposed research plan will carry out pioneering spin transport studies in a wider range of atomically engineered complex materials and devices, which are expected to show advanced functionality, thereby leading to spin-based electronics that are more versatile and more energy efficient. Because spins already flow inside every electronic device, the technology transfer from classical electronics to spin electronics only requires minimal changes to existing production lines in the world. The technologies developed in this program are thus fully compatible with modern semiconductor technologies and can be readily adopted by major companies. The projected industries to pursue (for potential collaborations as well as HQP positions) include those information giants such as Intel, IBM, Hitachi, Global Foundries, who already have strategic spintronics development initiated. For Canada, this means more employment and more visibility in the world. Ultimately, spin-based processors are not to fully replace modern silicon based processors, but to build on their existing strengths and improve only very critical processing components. This further ensures fast technology transfer. Together, they will make the information processing greener yet more powerful, putting the Moore's law back on its due track.