CAREER: Understanding and Modeling of Cryogenic Semiconductor Device Physics down to 4.2K

低温半导体器件和互补金属氧化物半导体（cryo-CMOS）在低至4.2K的温度下工作，对于某些新兴技术的成功至关重要。例如，低温CMOS是量子计算机（QC）以及深空探测的合适外围电路候选者。然而，cryo-CMOS器件物理的关键要素仍然不清楚，它们的设计仍然主要基于分析方程。此外，低温电子教育并不特别存在于大多数电气工程项目中。因此，该项目的目标是缩小cryo-CMOS器件物理学的关键知识差距，开发新的，高效的和有效的模型，以促进cryo-CMOS技术计算机辅助设计（TCAD）的创新，几十年来一直是CMOS发展的动力，并教育多样化的低温电子工作人员。成功实施将加强圣何塞州立大学的低温实验，这是加州州立大学系统中代表性不足的少数民族（URM）服务机构，并将在最先进的低温测量和模拟方面培训学生。由URM本科生组成的多元化研究团队将进行测量、TCAD模拟和代码开发。一个新的cryo-CMOS和QC会议将在硅谷妇女工程（WiE）会议上创建。学生们将在电子器件协会的研讨会上向硅谷的工程师们共同介绍研究结果。一个免费的暑期课程，介绍cryo-CMOS和QC将被介绍给当地社会和经济上处于不利地位的高中生，并建立一个未来的学生在量子计算的管道，以创造一个多元化的劳动力，并成为一个经济驱动力的脆弱社区。新课程和拓展课程的评估结果将公布，与教育界分享。为了实现这一目标，具有创新测试结构的65纳米测试芯片将被测量到4.2K。TCAD和从头计算模拟将用于验证各种低温理论。一个新的自动编码器框架将开发自动校准。将使用已知制造条件的附加芯片来验证和改进研究结果。这项工作将解决该领域长期存在的争议，即异常亚阈值摆动（SS）的起源，从而在低温CMOS器件物理方面取得重要突破并产生新的知识。将开发一套准确、完整和实用的TCAD模型，以促进用于量子计算机等新兴技术的低温CMOS的开发和优化。异常SS起源的发现将导致界面和能带结构工程的新研究，可能开辟一个全新的研究领域。准确和强大的模拟模型的迁移率和场依赖的电离将推导出这将有利于新的低温电子器件的发展。所提出的整套模型和强大的设置将使TCAD能够加速高性能和可靠的低温集成电路设计自动化工具的开发。该项目的成功将为低温CMOS的亚4.2K建模铺平道路，以实现量子计算机和CMOS的最终共集成。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。

Cryogenic semiconductor devices and Complementary Metal-Oxide-Semiconductors (cryo-CMOS) that operate at temperatures down to 4.2K are crucial to the success of certain emerging technologies. For example, cryo-CMOS are suitable peripheral circuits candidates for quantum computers (QC) as well as for deep space exploration. However, critical elements of cryo-CMOS device physics remain unclear and their design is still mostly based on analytical equations. Moreover, cryogenic electronics education is not particularly present in most Electrical Engineering programs. The objectives of this project are, therefore, to close the critical knowledge gaps in cryo-CMOS device physics, develop new, efficient and effective models to facilitate cryo-CMOS innovation in Technology Computer-Aided Design (TCAD) that has been the powerhouse of CMOS development for decades, and to educate a diverse cryogenic electronics workforce. Successful implementation will enhance cryogenic experiments at San Jose State University, an underrepresented minority (URM)-serving institution in the California State University (CSU) system and will train students in state-of-the-art cryogenic measurements and simulations. A diverse research team of URM undergraduate students will perform the measurement, TCAD simulations, and code development. A new cryo-CMOS and QC session will be created in the Silicon Valley Women in Engineering (WiE) Conference. Students will co-present the findings at the seminar of the Electron Device Society to the engineers in Silicon Valley. A free summer course to introduce cryo-CMOS and QC will be introduced to socially and economically disadvantaged high school students in the local area, and build a pipeline of future students in quantum computing to create a diverse workforce and become an economic driver for vulnerable communities. The new classes and outreach course evaluation results will be published to share with education communities.To achieve the goals, 65nm test chips with innovative test structures will be measured down to 4.2K. TCAD and ab initio simulations will be used to verify various cryogenic theories. A new autoencoder framework will be developed for automatic calibration. Additional chip with known fabrication conditions will be used to verify and improve the research results. This work will settle a long-standing controversy in the field, namely the origin of abnormal subthreshold swing (SS), resulting in important breakthroughs and generating new knowledge in cryo-CMOS device physics. An accurate, complete, and practical set of TCAD models will be developed to facilitate the development and optimization of cryo-CMOS for emerging technologies such as quantum computers. The finding of the origin of abnormal SS will lead to new research in interface and band structure engineering, potentially opening up an entirely new research field. Accurate and robust simulation models for mobility and field-dependent ionization will be derived which will facilitate the development of novel cryogenic electronic devices. The proposed complete set of models and robust settings will enable TCAD to accelerate the development of automation tools for high performance and reliable cryogenic integrated circuit designs. The success of this project will pave the path to sub-4.2K modeling of cryo-CMOS for the ultimate co-integration of quantum computers and CMOS.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.