金沙官网网址大全

Nano-electronics boosted atom by atom

作者:竹又    发布时间:2019-02-28 08:12:07    

By Will Knight Nanoscale microprocessors could get a big performance boost from a technique that enables semiconducting materials to be doped with useful impurities one atom at a time. The impurities – or dopants – are added to semiconductors to fine-tune their electronic properties. Normally, a less conductive material, such as arsenic or phosphorus, is introduced to a semiconductor like silicon or germanium through diffusion or another chemical technique. The process is random on a molecular scale but uniform enough at the scale of current semiconductor components to produce a regular and predictable change in properties. However, as electronic components shrink ever smaller – enabling greater computing power to be packed into circuits – variations in the concentration of dopants can cause problematic variations in the material’s conductivity. Now Takahiro Shinada and colleagues at Waseda University in Tokyo, Japan, have found a solution to this problem – adding individual ions to semiconductors with nanoscale accuracy. The researchers used a technique known as single ion implantation (SII) to place a wide range of ions into semiconducting materials, including beryllium, boron, phosphorus, iron and cobalt. SII involves using a small aperture to extract single ions from a beam, which are then implanted into the target material. The ions were added 60 nanometres (billionths of a metre) apart from one another. Tests confirmed that, following targeted doping, the materials had more uniform electronic properties. The process is relatively slow – Shinada says it would take 11 days to process a complete computer chip. However, he says it could help researchers design specialised nanoscale electronics. “The ability to control both the number and position of single atoms will help understand nanoscale semiconductor physics and design single-atom devices like solid-state quantum computers,” Shinada told New Scientist Eventually, the researchers believe the process could provide a substantial speed injection to the computing world. “Our technique may enhance the prospects for extending Moore’s Law,” Shinada says. Journal reference: Nature (vol 437,

 

Copyright © 网站地图