Carbon nanotube electronics - Materials, devices, circuits, design, modeling, and performance projection

Electron Devices Meeting(2011)

引用 46|浏览61
暂无评分
摘要
This paper summarizes a multi-disciplinary effort to realize the high expectations of the use of carbon nanotube (CNT) as a material for highly energy-efficient future digital systems. Today CNTs are grown on full wafers, highly aligned in one direction, and subsequently transferred to a target substrate multiple times for increased CNT density. Device level performance of carbon nanotube transistor (CNFET) rivals the best silicon transistor. Inherent CNT imperfections such as mis-positioned and metallic CNTs have been overcome through a combination of processing and imperfection-immune design techniques. As a result, CNFET-based digital arithmetic and storage circuit are fabricated using conventional optical lithography steppers and fabrication tools, some of them at full wafer scale. CNT interconnects with sub-ns delays have been measured using ring oscillators. Physics-based analytical and compact models have been developed to enable circuit design, circuit-level analysis of CNT-specific variations, and performance projection at the device, circuit, and system level. It is projected that CNFET circuits can enable 5x speed-up at the same power consumption over Si CMOS(PDSOI) at the 11 nm technology node for a four-core processor with 1.5M logic gates and 5MB SRAM per core. CNFET is the only FET that is projected to outperform the 11 nm node ITRS target.
更多
查看译文
关键词
sram chips,carbon nanotube field effect transistors,carbon nanotubes,digital arithmetic,logic gates,oscillators,photolithography,semiconductor device models,c,cmos pdsoi,cnfet,cnfet-based digital arithmetic,cnfet-based storage circuit,cnt density,cnt electronic,cnt material,cnt subns delay interconnection,sram,carbon nanotube electronic,carbon nanotube material,carbon nanotube transistor,circuit design,circuit-level analysis,four-core processor,imperfection-immune design technique,logic gate,memory size 5 mbyte,metallic cnt,optical lithography stepper,power consumption,ring oscillator,size 11 nm,carbon nanotube
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要