In this video excerpt from NOVA’s "Making Stuff: Smaller", host and New York Times technology columnist David Pogue visits IBM to learn about nanowires. Pogue and IBM scientist Frances Ross make a model out of pizza ingredients to symbolize how nanowires could be used to shrink transistors and power even smaller, more powerful computers. In a related demonstration, students discover how challenging it is to work on this small nanoscale. Using a Styrofoam® block and pipe cleaners, they simulate working with millions of wires and transistors that are placed onto tiny chips to produce smaller, more powerful electronic devices.
Making Stuff Smaller Demonstration (Document)
Yesterday’s room-sized supercomputers are today’s miniature microchips. Today’s smaller devices are more portable, cost less to make, consume less power, and have longer-lasting batteries. But how small can we go? How much power and performance can we squeeze out of ever-shrinking microchips?
A nanometer is one-billionth of a meter, which is thousands of times smaller than the width of a human hair. The nanochip does not yet exist, but materials scientists and engineers are steadily advancing toward nanocomputing. Creating materials at that scale raises unique challenges – electrons that jump their wires, circuits that overheat, and the need for super-tiny tools and innovative manufacturing techniques to create nanoscale objects.
The components of today’s electronics are measured on the macroscale (visible to the naked eye) and microscopic (visible using a microscope) scale. But to push the limits of “smaller”, materials scientists are investigating how to build and work on the nanoscale (one billionth of a meter). Nanoscale objects are billions of times smaller than everyday objects measured on the macroscale.
Nanotechnology involves engineering new materials out of individual atoms and molecules. When you build things on such a small scale, things act differently than our on scale. One area of nanotechnology is the development of nanowires, which are wires thinner than 100 nanometers in diameter. Nanowires are so thin compared to their length that they are considered to be one-dimensional objects.
They can be made from metals, such as titanium or molybdenum, or non-metals, such as silica, which is the most common component of sand. One way nanowires are made is by dragging a thicker wire through a very hot flame using the tip of a tiny probe to thin it. They can also be made “from scratch” by combining atoms using several common laboratory techniques.
Materials scientists are designing some nanowires that conduct light instead of electrons, which will eliminate the problems with overheating circuits. Replacing silicon with a newly discovered form of carbon called graphene, which is a single-atom thick layer of carbon, may make possible new computer chips that allow electrons to move 1,000 times faster, making computers even more powerful.
The promise of nanotechnology goes beyond smaller, more powerful computers. Picture a TV screen that’s a flexible film that you could roll up and put in your pocket. Or e-paper embedded with invisible nanowires that has the appearance of natural paper but is digital. Or new medical procedures to remove tumors without surgery or deliver medicine only where it is needed.
DAVID POGUE The transistors that power our stuff are about as small as they can get, unless scientists can come up with a new way of packing them ever-more tightly together.
To see one of those possible solutions, I've crossed the country to visit the IBM Research and Development..."kitchen?"
So this is Moore's Law of Italian cooking?
FRANCES ROSS (IBM Research Division) That's right. What we're going to do is explain why it's so important to get the transistors smaller and smaller.
DAVID POGUE Frances has a pretty appetizing way of visualizing this law and its limitations. Like pepperoni slices, the transistors on a silicon chip are flat.
Okay, so here's our...
FRANCES ROSS Silicon wafer.
DAVID POGUE ...silicon wafer.
FRANCES ROSS Now, these are the old-fashioned transistors. They're much larger, and you can see that you can't put that many onto each wafer.
DAVID POGUE So this would be a 1960 iPod?
FRANCES ROSS I think so, yes. This would be a '60s type of thing.
So, let's take off these old transistors and replace them with some new transistors.
DAVID POGUE Oh, these are much smaller!
FRANCES ROSS Yes, these new transistors are much smaller.
DAVID POGUE Technology has marched on.
FRANCES ROSS That's right. It's Moore's Law, in action.
DAVID POGUE So, in other words, all we have to do is make the transistors smaller every year, forever, and our gadgets will always be more powerful and more compact.
FRANCES ROSS That would be wonderful, but we can't make our pepperoni slices much smaller than this. And these transistors are now packed together about as close as we can get them.
DAVID POGUE The pizza party can't go on forever. There's a limit to how small you can shrink the transistors. If you reduce the surface area of a transistor too much and place it too close to its neighbor, electricity starts to leak, causing a short circuit. Not good.
FRANCES ROSS We've run out of area, so there's only one way to go, and that's upwards.
DAVID POGUE Slim Jims?
FRANCES ROSS That's right. This is a vertical transistor. Instead of having flatter, smaller transistors we go in the other direction.
DAVID POGUE Excuse me, vertical transistors?
FRANCES ROSS Vertical transistors.
DAVID POGUE With little toothpicks on the bottom?
FRANCES ROSS That's just for demonstration purposes.
DAVID POGUE Oh, okay.
By building vertical transistors, called nanowires, Frances can increase surface area, without bringing the transistors closer together, so no short circuit.
Ingenious. So this is what you're doing at IBM? You are making these?
FRANCES ROSS That's right. They're called nanowires, and the real thing is about a million times smaller than this.
DAVID POGUE A million times smaller?
FRANCES ROSS That's right.
DAVID POGUE Well, that would be hard to see!