BYR Achieve · 镜像论坛

在上一期热点文献推荐中，我们为您推荐了人工智能领域的最新发展前沿，包括使用AlphaCode生成竞争级别的代码、可穿戴式心脏超声成像仪、用于自动驾驶汽车安全验证的密集强化学习、用于深度学习建模的肌电数字孪生。本期我们为您选取了4篇文献，介绍半导体领域的最新发展前沿，包括用于太赫兹应用的电子元器件、用于深度学习加速器的CMOS兼容电化学突触晶体管阵列、用于晶圆级可拉伸电子器件的真空沉积聚合物电介质、使用粘附光刻的过渡金属二硫族化合物场效应晶体管的晶圆级集成。文献一用于太赫兹应用的电子元器件 Electronic metadevices for terahertz applications Nikoo, Mohammad Samizadeh, etc. NATURE, 2023, 614(7948): 451–455 The evolution of electronics has largely relied on downscaling to meet the continuous needs for faster and highly integrated devices. As the channel length is reduced, however, classic electronic devices face fundamental issues that hinder exploiting materials to their full potential and, ultimately, further miniaturization. For example, the carrier injection through tunnelling junctions dominates the channel resistance, whereas the high parasitic capacitances drastically limit the maximum operating frequency. In addition, these ultra-scaled devices can only hold a few volts due to the extremely high electric fields, which limits their maximum delivered power. Here we challenge such traditional limitations and propose the concept of electronic metadevices, in which the microscopic manipulation of radiofrequency fields results in extraordinary electronic properties. The devices operate on the basis of electrostatic control of collective electromagnetic interactions at deep subwavelength scales, as an alternative to controlling the flow of electrons in traditional devices, such as diodes and transistors. This enables a new class of electronic devices with cutoff frequency figure-of-merit well beyond ten terahertz, record high conductance values, extremely high breakdown voltages and picosecond switching speeds. This work sets the stage for the next generation of ultrafast semiconductor devices and presents a new paradigm that potentially bridges the gap between electronics and optics. 阅读原文：https://www.nature.com/articles/s41586-022-05595-z 文献二用于深度学习加速器的CMOS兼容电化学突触晶体管阵列 CMOS-compatible electrochemical synaptic transistor arrays for deep learning accelerators Cui, Jinsong, etc. NATURE ELECTRONICS, 2023, 6: 292–300 In-memory computing architectures based on memristive crossbar arrays could offer higher computing efficiency than traditional hardware in deep learning applications. However, the core memory devices must be capable of performing high-speed and symmetric analogue programming with small variability. They should also be compatible with silicon technology and scalable to nanometre-sized footprints. Here we report an electrochemical synaptic transistor that operates by shuffling protons between a hydrogenated tungsten oxide channel and gate through a zirconium dioxide protonic electrolyte. These devices offer multistate and symmetric programming of channel conductance via gate-voltage pulse control and small cycle-to-cycle variation. They can be programmed at frequencies approaching the megahertz range and exhibit endurances of over 100 million read-write cycles. They are also compatible with complementary metal-oxide-semiconductor technology and can be scaled to lateral dimensions of 150 x 150 nm2. Through monolithic integration with silicon transistors, we show that pseudo-crossbar arrays can be created for area- and energy-efficient deep learning accelerator applications. Oxide-based solid-state protonic electrochemical transistors that have symmetric operation and are compatible with CMOS technology can be used to create crossbar arrays for deep learning applications. 阅读原文：https://www.nature.com/articles/s41928-023-00939-7 文献三用于晶圆级可拉伸电子器件的真空沉积聚合物电介质 A vacuum-deposited polymer dielectric for wafer-scale stretchable electronics Koo, Ja Hoon, etc. NATURE ELECTRONICS, 2023, 6(2): 137–145 A thin and stretchable polymer layer can be fabricated over large areas with high uniformity using a vacuum-deposition method and used as the gate dielectric in stretchy carbon-nanotube-based transistors and circuits that can function at 40% strain. Despite recent advances in materials and fabrication technologies, the development of intrinsically stretchable electronic devices with large-area uniformity, low power consumption and performance comparable with conventional rigid devices remains challenging. A key limitation is the absence of an elastic dielectric material that can be thin and uniform over large areas as well as offer robust insulating properties and high mechanical and chemical stability. Here we show that a vacuum-deposited elastic polymer layer can be used as the gate dielectric in stretchy field-effect transistors with carbon nanotube channels and microcracked gold electrodes. The polymer dielectric layer has high insulation properties (at a thickness below 200 nm), high stability and large-area uniformity. An 8-inch wafer array of the stretchable carbon nanotube transistors exhibits good uniformity in their electrical performance and can maintain their performance after 1,000 stretching cycles at 40% strain. We use the transistors to construct stretchy inverters and logic gates that can function under applied strains of up to 40%. 阅读原文：https://www.nature.com/articles/s41928-023-00918-y 文献四使用粘附光刻的过渡金属二硫族化合物场效应晶体管的晶圆级集成 Wafer-scale integration of transition metal dichalcogenide field-effect transistors using adhesion lithography Nguyen, Van Luan, etc. NATURE ELECTRONICS, 2023, 6(2): 146–153 Field-effect transistors based on two-dimensional materials are a potential replacement for silicon-based devices in next-generation semiconductor chips. However, the weak interfacial adhesion energy between two-dimensional materials and substrates can lead to low yields and non-uniform transistors on the wafer scale. Furthermore, conventional photolithography processes-including photochemical reactions and chemical etching-can damage atomically thin materials. Here we show that the interfacial adhesion energy between two-dimensional materials and different substrates can be quantified using a four-point bending method. We find that a molybdenum disulfide/silicon dioxide interface has an interfacial adhesion energy of 0.2 J m-2, which can be modulated from 0 to 1.0 J m-2 by incorporating self-assembled monolayers with different end-termination chemistries. We use this to create an adhesion lithography method that is based on adhesion energy differences and physical etching processes. We use this approach to fabricate more than 10,000 molybdenum disulfide field-effect transistors on six-inch wafers with a yield of around 100%. 阅读原文：https://www.nature.com/articles/s41928-022-00890-z “前沿文献专题推荐服务”是图书馆情报服务部学科服务的重要内容。欢迎有相关需求的科研人员与我们联系，我们将为您提供更有针对性的定题服务。联系电话：62281933 62283502 地址：西土城校区图书馆125室情报服务部联系人：杨老师闫老师邮箱：yandong80@bupt.edu.cn

热点文献带您关注半导体领域的最新进展——图书馆前沿文献专题