|
|
|
| Current Issue |
|
|
| Resources |
|
|
|
Quantum Electronics
| Article
# : |
17759 |
|
|
Section : |
NATURAL SCIENCE
|
| Issue
Date : |
6 / 1990 |
2,609 Words |
| Author
: |
Udi Meirav and Mordehai Heiblum
Udi Meirav is at the department of physics at the
Massachusetts Institute of Technology. Mordehai Heiblum is a
research scientist at the IBM Thomas J. Watson Research Center
in Yorktown Heights, New York. |
If you have ever used a pocket calculator, you probably could not help but be amazed how something so small can be so smart - and so fast. In fact, the smallness is even more impressive when looking beyond what meets the eye. The actual brains of these calculators, integrated circuits, are made of silicon transistors too small to be seen with the naked eye. But try to program your calculator to perform somewhat more complex calculations and you won't fail to notice a slight delay before the answer is displayed. This perceptible delay is the major reason driving many scientists to seek new ways to make transistors faster yet.
As the tasks expected from computers get more numerous and complex, the number of electronic steps - and consequently the time - required to complete each task increases. The brief pause of the pocket calculator becomes a significant delay in big computing machines. One of the keys to quickening the pace of these machines is to have each element of the electronic circuit - transistors, of which there are millions - carry out its own little function faster. Faster means passing electrons, the carriers of electric current, from one end to another in less time.
Looking at this simply, for a given electron speed, the smaller the device, the less time it takes it to perform its function. Since electrons can easily travel at speeds of 200,000 miles per hour, or more, and transistor can be 10 times smaller than a human blood cell, a single operation may take well under a billionth of a second. Still, for complex computer tasks involving trillions of operations, these delays rapidly add up to precious seconds, minutes, and hours, leading to the desire for even faster device operations and further size reduction.
Underlying this simplistic reasoning is the assumption that as we reduce the size of the electronic components, they behave the same way - only faster. Is this assumption valid? This question heralds a new field of research sometimes referred to as mesoscopic physics - the realm of very small scale electronics.
The breakdown of scaling
Much of the intrigue of mesoscopic physics arises because scientists have discovered that as electronic devices shrink in size beyond certain limits, new and unfamiliar phenomena occur; the simple rule of size-scaling no longer work.
Why this breakdown of scaling? There are several
...
Read Full Article
Look for this article in Ask.com
|
|
