Advanced manufacturing of InN/Si nanowire tunnelling transistors for energy efficient electronics

自现代计算诞生以来，计算的能效一直是推动变革的动力。据估计，每年约有1500太瓦时或全球电力产量的10%消耗在电子产品上，主要是在云计算上。计算基础设施消耗的这些史无前例的能量所带来的影响越来越受到关注，这促使半导体社区寻找节能的替代方案。我们建议通过开发一种可扩展的低电压InN/Si隧道场效应晶体管(TFET)的先进制造工艺来解决电子产品的能效问题。采用基于先进纳米材料分子束外延(MBE)合成的TFET来降低晶体管工作电压是提高逻辑电路能效的最直接途径。TFET利用了电子通过势垒的量子隧道原理，因此可以在比传统晶体管更低的电压下工作。MBE生产最高质量的半导体，我们将把这项技术应用到Inn TFET的可扩展制造中。InN纳米线因其优异的电子质量和与硅的理想能带对齐而被确定为纳米材料的选择。TFET能效的提高将允许更快的器件操作，因为主要是散热限制了时钟速度。该项目结合了我们团队的协同互补专业知识，包括Szkopek教授的量子器件专业知识、米教授的外延纳米材料生长专业知识、工业合作伙伴Croslight的半导体器件建模专业知识以及工业合作伙伴Meaglow的氮化物纳米材料生长专业知识。通过参与这一学术/行业合作伙伴关系，加拿大公司Croslight和Meaglow将处于理想的地位，分别在技术计算机辅助设计(TCAD)和Inn Growth技术的特定领域，将TFET技术整合到年收入3000亿美元的半导体行业中。

Since the dawn of modern computing, the energy efficiency of computation has been a driving force forchange. It is estimated that approximately 1500 TWhr per year or 10% of global electricity production isconsumed on electronics, primarily on cloud computing. The implications of these unprecedented quantities of energy consumed by computing infrastructure are of growing concern, motivating the semiconductor community to seek out energy efficient alternatives.We propose to address the problem of energy efficiency in electronics by the development of a scalable advanced manufacturing process for low-voltage InN/Si tunnelling field effect transistors (TFETs). Reducing transistor operating voltage by adopting TFETs based on advanced nanomaterials synthesis via molecular beam epitaxy (MBE) is the most direct way of improving the energy efficiency of logic circuits. The TFET exploits the principle of quantum tunnelling of electrons through potential barriers, and can therefore operate at lower voltages than conventional transistors. MBE produces the highest quality semiconductors, and we will apply this technique to the scalable manufacture of InN TFETs. InN nanowires have been identified as the nanomaterial of choice because of their superior electronic quality and ideal band alignment with Si.The increased energy efficiency of a TFET would permit higher speed device operation, as it is primarily heat removal that limits clock speed. This project combines the synergetic complementary expertise of our team, including the quantum device expertise of Prof. Szkopek, the epitaxial nanomaterial growth expertise of the Prof. Mi, the semiconductor device modelling expertise of industrial partner Crosslight, and the nitride nanomaterial growth expertise of industrial partner Meaglow. By participating in this academic/industry partnership, Canadian companies Crosslight and Meaglow will be ideally placed to stake out a role in the integration of TFET technology into the $300 billion / annum semiconductor industry, in the specific domains of technology computer aided design (TCAD) and InN growth technology respectively.