Contamination-Free Cu/SiCN Hybrid Bonding Process Development for Sub-m Pitch Devices with Enhanced Bonding Characteristics

2023 IEEE 73RD ELECTRONIC COMPONENTS AND TECHNOLOGY CONFERENCE, ECTC(2023)

引用 0|浏览0
暂无评分
摘要
Hybrid bonding has emerged as a promising 3D integration technology for the next generation stacking devices with advantages of higher performance and smaller form factor. As for the future devices with fine-pitch (sub-mu m level) copper (Cu) interconnects, however, there are critical issues to consider for a successful application of hybrid bonding processes; (1) Cu diffusion with scaled Cu pads and (2) Cu re-sputtering and chamber contamination during plasma surface activation step that lead to fatal device failure (e.g. electrical shorts) even with a small amount. Here, we demonstrated robust connections for 0.4 to 0.7 mu m-sized Cu pads of fine-pitch with wafer-to-wafer (W2W) Cu/SiCN hybrid bonding scheme through N-2 plasma development and present comprehensive analyses for surface activation process. A computational work carried out by reactive molecular dynamics (MD) simulations was employed to propose an optimal N-2 plasma condition that can enhance SiCN-SiCN bonding characteristics. The optimized N-2 plasma condition reproduces SiCN-SiCN bonding strength of above 2.0 J/m(2) while suppressing oxidation of Cu pads below 2 nm. Cu/SiCN hybrid bonding characteristics using N-2 plasma were confirmed to be void-free, well Cu-Cu interdiffused, and mechanically well aligned from the assessment involving SAT, TEM/EDS and FIB-SEM/EBSD analyses. Moreover, our optimal plasma condition meets the contamination-free criteria indicating that it is a preferable surface activation scheme without contaminating the plasma chamber regardless of continuous operations. The methodologies demonstrated in this work suggest that hybrid bonding can be applied to future fine-pitch device applications for successful 3D integration.
更多
查看译文
关键词
Cu/SiCN hybrid bonding, wafer-to-wafer bonding, plasma surface activation, Cu contamination control, molecular dynamics simulation
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