Free Republic
Browse · Search
News/Activism
Topics · Post Article

Skip to comments.

Refining Semiconductors, One Atom at a Time
NY Times ^ | April 8, 2004 | ANNE EISENBERG

Posted on 04/10/2004 12:28:29 PM PDT by neverdem

At the heart of semiconductor fabrication are crucial additives called dopants. These impurities change the electronic properties of silicon or other material to make the transistors and other components of a chip.

Currently these dopants are added in bulk, their exact location usually no more a problem than the exact location of grains of baking soda or raisins stirred into cake batter.

But as electronic devices shrink - and the hope is to get them down to the size of a molecule - serious problems with doping are expected. At that small a scale, the location of a needed doping atom must be known exactly: the atom has to be present and countable on the molecule that is acting as, say, a diode.

Although electronic devices are not yet that small, problems have already arisen in controlling the number and location of dopants. A difference as fine as one atom of dopant or two can, for example, change the voltage at which a device switches.

But a physicist has succeeded in controlling doping precisely at the atomic level. Michael F. Crommie, a professor of physics of the University of California at Berkeley, has attached single dopant atoms, one by one, to single molecules, thereby custom-tailoring their electronic properties.

The work is significant because of the tight control it offers, possibly in designing molecules that will in the future become single circuit elements.

Dr. Crommie and his group took the probe of a homemade scanning tunneling microscope and moved large carbon molecules across a chilled silver surface toward atoms of dopant. The dopant, potassium, hopped onto the molecule, immediately changing its charge.

"This is the first time someone has been able to actually control doping on an atom-by-atom basis," said Jun Nogami, an associate professor of chemical engineering and materials science at Michigan State University.

Researchers have managed to put dopant atoms both inside nanotubes and inside the molecule Dr. Crommie used, but not with the same level of control.

"Dr. Crommie can add and subtract the atoms," Dr. Nogami said. "Therefore, in principle, he can tune the electronic structure to produce a particular device characteristic."

The Berkeley group worked with a well-known carbon molecule, named buckminsterfullerene because its geodesic shape resembles the domes devised by the architect Buckminster Fuller. This fullerene, also known as a buckyball, contains 60 carbon atoms. Researchers in molecular electronics hope that one day molecules like these may form the backbone of electronic devices.

In the experiment, reported in the journal Science, all of the work was done in a vacuum chamber at an icy 7 degrees Kelvin - very close to absolute zero.

"Cold is the name of this game," said Dr. Nogami. "Otherwise the atoms move around too much."

When the single atom of potassium attached to the molecule, it changed the molecule's electronic properties, donating a charge in what is known as N-type doping. "In this way," Dr. Crommie said, "we scale the idea of changing properties right down to the molecular level by adding or removing single doping atoms."

The researchers could reverse the process, too, by dragging the buckyball across an impurity in the silver crystal and removing the doping atom.

Dr. Crommie was able to add up to seven atoms to the molecule, one by one. "The buckyballs were like molecular Pac-Men," he said, "gobbling up the dopant atoms."

The buckyball retained its basic shape during the change, Dr. Crommie said. "With four potassium atoms added, for example, it was about 9 percent wider and 3 percent shorter." Still, its diameter was only about one nanometer, or a billionth of a meter.

The electronic changes that the molecules underwent during potassium doping were measured with the scanning tunneling microscope.

"It's so nice that you can see the energy levels of the molecules shift as he traps the potassium at the surface of the molecule," Max G. Lagally, a physicist and professor in the engineering college at the University of Wisconsin at Madison, said of Dr. Crommie's work.

Each potassium atom gave about six-tenths of the charge of an electron. Some of the charge was probably left behind on the silver substrate, Dr. Crommie said.

In the future, the Berkeley team's work might help others design molecules that have exactly the properties needed for circuit elements. For example, if a device were made of buckyballs, the addition or subtraction of dopant atoms could control the voltage at which the device switched.

Dr. Lagally said that Moore's Law, which predicts that the number of transistors on a chip will double every 18 months, has had many challenges, one of them the limits of doping. "It's a real showstopper," he said, because of the problems of knowing exactly where the dopants are when the process is done in bulk.

But Dr. Crommie has gotten around that problem. "Here he's demonstrated that the atom is on the buckyball, and he knows it's there," Dr. Lagally said.

Many approaches for altering molecules' electronic properties are being tried, including chemical synthesis and the use of electrodes as gates, and much work lies ahead to create molecular circuit elements.

"But this is a starting point," Dr. Lagally said.

E-mail: Eisenberg@nytimes.com


TOPICS: Extended News; Foreign Affairs; Government; News/Current Events; Technical; US: District of Columbia
KEYWORDS: dopants; semiconductors; silicon; techindex; transistors

1 posted on 04/10/2004 12:28:30 PM PDT by neverdem
[ Post Reply | Private Reply | View Replies]

To: fourdeuce82d; Travis McGee; El Gato; JudyB1938; Ernest_at_the_Beach; Robert A. Cook, PE; lepton; ...
PING
2 posted on 04/10/2004 12:32:49 PM PDT by neverdem (Xin loi min oi)
[ Post Reply | Private Reply | To 1 | View Replies]

To: neverdem
So let me get this straight - a University of California Professor, who's a Crommie, has been working on his dope.
3 posted on 04/10/2004 12:40:50 PM PDT by Hoplite
[ Post Reply | Private Reply | To 2 | View Replies]

To: neverdem
Each potassium atom gave about six-tenths of the charge of an electron. Some of the charge was probably left behind on the silver substrate, Dr. Crommie said.

HUH???

4 posted on 04/10/2004 12:45:38 PM PDT by null and void (<----just the right blend of brains, nuts and sugar.)
[ Post Reply | Private Reply | To 1 | View Replies]

To: neverdem
I'll let my tagline speak for me...
5 posted on 04/10/2004 12:52:22 PM PDT by randog (Everything works great 'til the current flows.)
[ Post Reply | Private Reply | To 1 | View Replies]

To: neverdem
Full blown molecular nanotech is coming, folks. The estimates are now between 10 and 40 years to assemblers.
6 posted on 04/10/2004 12:58:41 PM PDT by marktwain
[ Post Reply | Private Reply | To 2 | View Replies]

To: null and void
Yeah, I know you're supposed to have conservation of energy, mass, charge, etc. and other trivial ideas. We shouldn't let these notions interfere with possible applications. Go ask Thomas Edison. We can probably generate better theories after the fact.
7 posted on 04/10/2004 1:05:02 PM PDT by neverdem (Xin loi min oi)
[ Post Reply | Private Reply | To 4 | View Replies]

To: null and void
Each potassium atom gave about six-tenths of the charge of an electron. Some of the charge was probably left behind on the silver substrate, Dr. Crommie said. HUH???

Sounds like either rounding, or dealing with up and down quarks.

8 posted on 04/10/2004 1:18:35 PM PDT by lepton
[ Post Reply | Private Reply | To 4 | View Replies]

To: neverdem; *tech_index
I'd like to see the machines that will assemble these into electronic devices. And then, how do you test and repair? Presumable, I will be out of this business by then. Right now the smallest board we assemble is 3/4 inch long and 1/4 of an inch wide. That's giant compared to what they are talking about.
9 posted on 04/10/2004 2:22:28 PM PDT by afraidfortherepublic (Re-elect Dubya)
[ Post Reply | Private Reply | To 1 | View Replies]

To: Hoplite
a University of California Professor, who's a Crommie, has been working on his dope.

LOL! Worse yet, he has had his evil probes deep inside minority carriers.

10 posted on 04/10/2004 2:42:14 PM PDT by snopercod (When the people are ready, a master will appear.)
[ Post Reply | Private Reply | To 3 | View Replies]

To: afraidfortherepublic
Connectors. How do you miniaturize interface connectors (power / signal / peripherals ) to match? It seems that's still a problem. When the connectors are bigger than the device being powered you've lost your advantage. You've got to miniaturize everything. That seems like a quantum leap from where we're at now. I'm sure someone's working on this.
11 posted on 04/10/2004 3:37:47 PM PDT by Sodbuster
[ Post Reply | Private Reply | To 9 | View Replies]

To: Sodbuster
What you do is put all the low-power part of the system on one chip, so the miniaturized connections are within the chip -- that's how it's done now. OTOH, these guys forget about the economics -- using a tunneling electron microscope to dope circuits is hardly cost-effective.
12 posted on 04/10/2004 7:53:32 PM PDT by expatpat
[ Post Reply | Private Reply | To 11 | View Replies]

To: neverdem
"Cold is the name of this game," said Dr. Nogami. "Otherwise the atoms move around too much."

Yes, those pesky little atoms just don't wanna sit still!

13 posted on 04/10/2004 9:42:33 PM PDT by Jorge
[ Post Reply | Private Reply | To 1 | View Replies]

To: neverdem
A difference as fine as one atom of dopant or two can, for example, change the voltage at which a device switches.

I wonder what change in switching voltage an extra atom could make in a semiconductor?

I mean if we're talking about a change in the switching threshold by microvolts or less, the tolerances engineers would have to factor into their designs would have to be incredibly tight.

14 posted on 04/10/2004 9:56:56 PM PDT by Jorge
[ Post Reply | Private Reply | To 1 | View Replies]

To: Hoplite
So let me get this straight - a University of California Professor, who's a Crommie, has been working on his dope. NY TIMES consider the source.
15 posted on 04/10/2004 10:27:32 PM PDT by SirTaurus
[ Post Reply | Private Reply | To 3 | View Replies]

To: lepton
Sounds like either rounding, or dealing with up and down quarks.

More likely still a "bulk properties" effect. That is 6 of 10 did not leave a charge on the silver substrate.

16 posted on 04/11/2004 11:15:40 AM PDT by El Gato (Federal Judges can twist the Constitution into anything.. Or so they think.)
[ Post Reply | Private Reply | To 8 | View Replies]

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.

Free Republic
Browse · Search
News/Activism
Topics · Post Article

FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson