Organosulfide Inhibitor Instigated Passivation of Multiple Substrates for Area-Selective Atomic Layer Deposition of HfO2

CHEMISTRY OF MATERIALS(2024)

引用 0|浏览0
暂无评分
摘要
With recent advancements in semiconductor technology, continuous efforts are being made to meet the requirements for further reductions in the feature sizes of electronic interconnects in semiconductor devices. Efforts to improve area-selective deposition (ASD) processes have led to researchers manipulating deposition surfaces using surface inhibitors as tools for area-selective atomic layer deposition (AS-ALD). In this study, organosulfide small-molecule inhibitors (SMIs) were utilized for AS-ALD on metal, oxide, and nitride surfaces such as Cu, SiO2, and TiN, respectively. Upon hightemperature exposure, the organosulfide SMI decomposes to assist the adsorption of its fragmentation products on the Cu and SiO2 substrates, thereby simultaneously adsorbing and passivating the two surfaces upon SMI exposure. The surface chemistry and reactivity were explained by calculations using density functional theory with the slab approach and Monte Carlo simulations. Furthermore, the blocking potential of the SMIs was evaluated using atomic layer deposition (ALD) of HfO2. The SMIcovered Cu substrate showed inhibition against ALD growth of HfO2 with a selectivity of approximately 98% over 25 growth cycles compared to the uncovered Cu substrate successfully blocking approximately 3 nm of HfO2 ALD. The SMI-covered SiO2 substrate showed a lowered selectivity compared to the SMI covered Cu substrate but still, a substantial selectivity was present compared to bare SiO2 and TiN substrates where no blocking was observed. These results agree with the theoretical findings. This possibility to block two important surfaces in semiconductor manufacturing (Cu and SiO2) while leaving a third one (TiN) unblocked for ALD growth is an important step for the future application of ASD in the production of ever smaller semiconductor devices.
更多
查看译文
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