Monolithic three‐dimensional integration of aligned carbon nanotube transistors for high‐performance integrated circuits

Author:

Fan Chenwei1ORCID,Cheng Xiaohan1,Xu Lin1,Zhu Maguang1,Ding Sujuan2,Jin Chuanhong2,Xie Yunong1,Peng Lian‐Mao1,Zhang Zhiyong1ORCID

Affiliation:

1. Key Laboratory for the Physics and Chemistry of Nanodevices and Center for Carbon‐based Electronics, School of Electronics Peking University Beijing the People's Republic of China

2. State Key Laboratory of Silicon Materials, School of Materials Science and Engineering Zhejiang University Hangzhou the People's Republic of China

Abstract

AbstractCarbon nanotube field‐effect transistors (CNT FETs) have been demonstrated to exhibit high performance only through low‐temperature fabrication process and require a low thermal budget to construct monolithic three‐dimensional (M3D) integrated circuits (ICs), which have been considered a promising technology to meet the demands of high‐bandwidth computing and fully functional integration. However, the lack of high‐quality CNT materials at the upper layer and a low‐parasitic interlayer dielectric (ILD) makes the reported M3D CNT FETs and ICs unable to provide the predicted high performance. In this work, we demonstrate a multilayer stackable process for M3D integration of high‐performance aligned carbon nanotube (A‐CNT) transistors and ICs. A low‐κ (~3) interlayer SiO2 layer is prepared from spin‐on‐glass (SOG) through processes with a highest temperature of 220°C, presenting low parasitic capacitance between two transistor layers and excellent planarization to offer an ideal surface for the A‐CNT and device fabrication process. A high‐quality A‐CNT film with a carrier mobility of 650 cm2 V–1 s–1 is prepared on the ILD layer through a clean transfer process, enabling the upper CNT FETs fabricated with a low‐temperature process to exhibit high on‐state current (1 mA μm–1) and peak transconductance (0.98 mS μm–1). The bottom A‐CNT FETs maintain pristine high performance after undergoing the ILD growth and upper FET fabrication. As a result, 5‐stage ring oscillators utilizing the M3D architecture show a gate propagation delay of 17 ps and an active region of approximately 100 μm2, representing the fastest and the most compact M3D ICs to date.image

Funder

National Natural Science Foundation of China

Beijing Municipal Science and Technology Commission

Publisher

Wiley

Subject

Materials Chemistry,Surfaces, Coatings and Films,Materials Science (miscellaneous),Electronic, Optical and Magnetic Materials

Reference73 articles.

1. JeddelohJ KeethB.Paper presented at: Symposium on VLSI Technology (VLSIT) 12–14 June 2012 Session 10 – TAPA 3 Technology / Circuits Joint Focus Session – Memory.2012:87‐88.

2. LeeDU KimKW KimKW et al.Paper presented at: IEEE International Solid‐State Circuits Conference Digest of Technical Papers (ISSCC) 9–13 February 2014 Session 25: High‐Bandwidth Low‐Power DRAM and I/O.2014.

3. HuangPK LuCY WeiWH et al.Paper presented at: IEEE 71st Electronic Components and Technology Conference (ECTC) Session 3: Advanced Heterogenous Chiplet and Integration for HPC.2021.

4. LeeC‐C HungC CheungC et al.Paper presented at: IEEE 66th Electronic Components and Technology Conference (ECTC) Session 31: 3D Applications and Wafer Processing.2016.

5. MahajanR SankmanR PatelN et al.Paper presented at: IEEE 66th Electronic Components and Technology Conference (ECTC) Session 13: Substrate Embedding and Advanced Flip‐Chip Packaging.2016.

Cited by 3 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3