Zhu Bin, Yi Ding, Wang Yuxi, Sun Hongyu, Sha Gang, Zheng Gong, Garnett Erik C, Tian Bozhi, Ding Feng, Zhu Jia
National Laboratory of Solid State Microstructures, College of Engineering and Applied Sciences, Nanjing University, Nanjing 210093, People's Republic of China.
Jiangsu Key Laboratory of Artificial Functional Materials, Nanjing University, Nanjing 210093, People's Republic of China.
Proc Natl Acad Sci U S A. 2021 Jan 26;118(4). doi: 10.1073/pnas.2010642118.
There has been a persistent effort to understand and control the incorporation of metal impurities in semiconductors at nanoscale, as it is important for semiconductor processing from growth, doping to making contact. Previously, the injection of metal atoms into nanoscaled semiconductor, with concentrations orders of magnitude higher than the equilibrium solid solubility, has been reported, which is often deemed to be detrimental. Here our theoretical exploration reveals that this colossal injection is because gold or aluminum atoms tend to substitute Si atoms and thus are not mobile in the lattice of Si. In contrast, the interstitial atoms in the Si lattice such as manganese (Mn) are expected to quickly diffuse out conveniently. Experimentally, we confirm the self-inhibition effect of Mn incorporation in nanoscaled silicon, as no metal atoms can be found in the body of silicon (below 10 atoms per cm) by careful three-dimensional atomic mappings using highly focused ultraviolet-laser-assisted atom-probe tomography. As a result of self-inhibition effect of metal incorporation, the corresponding field-effect devices demonstrate superior transport properties. This finding of self-inhibition effect provides a missing piece for understanding the metal incorporation in semiconductor at nanoscale, which is critical not only for growing nanoscale building blocks, but also for designing and processing metal-semiconductor structures and fine-tuning their properties at nanoscale.
人们一直在持续努力理解和控制纳米尺度下金属杂质在半导体中的掺入情况,因为这对于从生长、掺杂到制作接触的半导体加工过程至关重要。此前,已有报道将金属原子注入纳米级半导体,其浓度比平衡固溶度高出几个数量级,这种情况通常被认为是有害的。在此,我们的理论探索表明,这种大量注入是因为金或铝原子倾向于替代硅原子,因此在硅晶格中不具有移动性。相比之下,硅晶格中的间隙原子,如锰(Mn),预计会迅速方便地扩散出去。在实验中,我们通过使用高聚焦紫外激光辅助原子探针断层扫描进行仔细的三维原子映射,证实了纳米级硅中锰掺入的自抑制效应,因为在硅体中(每立方厘米低于10个原子)未发现金属原子。由于金属掺入的自抑制效应,相应的场效应器件表现出优异的传输特性。这种自抑制效应的发现为理解纳米尺度下半导体中的金属掺入提供了缺失的一环,这不仅对于生长纳米级构建块至关重要,而且对于设计和加工金属 - 半导体结构以及在纳米尺度上微调其性能也至关重要。