Comparison of Enabling Wafer Bonding Techniques for TSV Integration

Author:

Kim Bioh1,Matthias Thorsten2,Wimplinger Markus2,Kettner Paul2,Lindner Paul2

Affiliation:

1. EV Group, Inc., Tempe, AZ

2. EV Group, Florian/Inn, Austria

Abstract

In this study are compared the technical merits and demerits of three bonding methods suitable for manufacturing 3D-ICs. Patterned metal thermo-compression bonding facilitates fine-pitch, high-density TSV stacking with lower electrical resistance and higher mechanical strength. Direct Cu-Cu bonding is preferred over transient liquid phase bonding with Sn or Sn alloys, but reliable Cu-Cu bonds result only from high process temperature and long process time. Both bonding temperature and post-bond annealing temperature have the most significant influence on Cu-Cu bond properties. The pre-bonding of silicon oxide bonds occurs at room temperature and thus does not induce any run-out errors in wafer alignment, resulting in higher post-bond alignment accuracy. Subsequent heating to high temperatures is necessary to achieve covalent bonds, but modifying the surface chemistry by plasma activation allows the formation of strong chemical bonds at significantly lower annealing temperatures (200–400°C). Adhesive bonding has such advantages as low bonding temperature and process time compared to metal bonding, the tolerance to wafer topography and surface conditions, and the ability to join any wafer materials. However, the material reflow imposes some challenges for maintaining the alignment accuracy and another major concern is the reliability of polymer adhesives during the post-bond processes.

Publisher

ASMEDC

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

1. Recent Advances and Trends in Cu–Cu Hybrid Bonding;IEEE Transactions on Components, Packaging and Manufacturing Technology;2023-03

2. Cu-Cu Hybrid Bonding;Chiplet Design and Heterogeneous Integration Packaging;2023

3. Hybrid Bonding;Semiconductor Advanced Packaging;2021

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