澳门金沙最新官方网址

Gold dust may connect molecular computers to real world

作者:习溱烙    发布时间:2019-03-06 11:07:00    

By Will Knight A way of connecting molecular computer components to existing computers has been patented by Hewlett-Packard. The patent covers a theoretical way for linking the tiny molecular wires and switches with silicon wires, which are at least 100 times larger. Providing this interconnectivity is a crucial step towards making molecular computing a practical proposal. It would be virtually impossible to physically construct a system for controlling interaction between the molecular and silicon worlds, so the researchers have patented a way of randomly connecting the two and decoding the information afterwards to create a working device. The US patent was issued to HP scientists Phil Kuekes and Stan Williams. The researchers believe that the method may eventually provide greater memory and processing power than silicon by scaling down beyond the known limitations of the latter. “It’s absolutely the next step for molecular electronics,” says Cees Dekker, a physicist building molecular computer components at the Delft University of Technology in the Netherlands. “One of the things in our field is that a lot of people have done things with single molecules but haven’t addressed how you can control these and harness their computational power.” The Hewlett-Packard researchers envisage connecting silicon wires to molecular wires, with the latter in turn being connected to data storage molecules. The process would involve creating a mesh of a thousand molecular wires running horizontally and vertically, with the data storage molecules at each intersection. Hewlett-Packard’s research team was awarded a separate patent for this process in October 2000. Just 10 silicon wires would then be introduced to the mesh. The next step would be to sprinkle an electrically-conducting dust over the wires. These could be made of gold and would create effectively random connections between two types of wire. Analysis of these complex interconnections could then determine how to control the molecular storage device. The researchers estimate they will be able to use the technique to build a molecular circuit capable for storing 16 kilobits of information by 2005. However, significant obstacles still face molecular computing, according to Richard Martel, a molecular computing expert with IBM in New York. He says that researchers are still struggling to understand how to control electrons at the molecular level. “The field of molecular electronics is still in the experimental stage,

 

Copyright © 网站地图