Advance could change modern electronics: High-performance 'metal-insulator-metal' diode created

Oct 29, 2010
This image of an asymmetric MIM diode reflects a major advance in materials science that could lead to less costly and higher speed electronic products. (Image courtesy of Oregon State University)

Researchers at Oregon State University have solved a quest in fundamental material science that has eluded scientists since the 1960s, and could form the basis of a new approach to electronics.

The discovery, just reported online in the professional journal Advanced Materials, outlines the creation for the first time of a high-performance "metal-insulator-metal" .

"Researchers have been trying to do this for decades, until now without success," said Douglas Keszler, a distinguished professor of chemistry at OSU and one of the nation's leading researchers. "Diodes made previously with other approaches always had poor yield and performance.

"This is a fundamental change in the way you could produce electronic products, at high speed on a huge scale at very low cost, even less than with conventional methods," Keszler said. "It's a basic way to eliminate the current speed limitations of electrons that have to move through materials."

A patent has been applied for on the new technology, university officials say. New companies, industries and high-tech jobs may ultimately emerge from this advance, they say.

The research was done in the Center for Green Materials Chemistry, and has been supported by the National Science Foundation, the Army Research Laboratory and the Oregon Nanoscience and Institute.

Conventional electronics made with silicon-based materials work with transistors that help control the flow of electrons. Although fast and comparatively inexpensive, this approach is still limited by the speed with which electrons can move through these materials. And with the advent of ever-faster computers and more sophisticated products such as liquid crystal displays, current technologies are nearing the limit of what they can do, experts say.

By contrast, a metal-insulator-metal, or MIM diode can be used to perform some of the same functions, but in a fundamentally different way. In this system, the device is like a sandwich, with the insulator in the middle and two layers of metal above and below it. In order to function, the electron doesn't so much move through the materials as it "tunnels" through the – almost instantaneously appearing on the other side.

"When they first started to develop more sophisticated materials for the display industry, they knew this type of MIM diode was what they needed, but they couldn't make it work," Keszler said. "Now we can, and it could probably be used with a range of metals that are inexpensive and easily available, like copper, nickel or aluminum. It's also much simpler, less costly and easier to fabricate."

The findings were made by researchers in the OSU Department of Chemistry; School of Electrical Engineering and Computer Science; and School of Mechanical, Industrial and Manufacturing Engineering.

In the new study, the OSU scientists and engineers describe use of an "amorphous metal contact" as a technology that solves problems that previously plagued MIM diodes. The OSU diodes were made at relatively low temperatures with techniques that would lend themselves to manufacture of devices on a variety of substrates over large areas.

OSU researchers have been leaders in a number of important material science advances in recent years, including the field of transparent electronics. University scientists will do some initial work with the new technology in electronic displays, but many applications are possible, they say.

High speed computers and electronics that don't depend on transistors are possibilities. Also on the horizon are "energy harvesting" technologies such as the nighttime capture of re-radiated solar energy, a way to produce energy from the Earth as it cools during the night.

"For a long time, everyone has wanted something that takes us beyond silicon," Keszler said. "This could be a way to simply print electronics on a huge size scale even less expensively than we can now. And when the products begin to emerge the increase in speed of operation could be enormous."

Explore further: A crystal wedding in the nanocosmos

Related Stories

Cheaper substrates made of oxide materials

Jul 27, 2010

Imagine building cheaper electronics on a variety of substrates -- materials like plastic, paper, or fabric. Researchers at Taiwan's National Chiao Tung University have made a discovery that opens this door, allowing them ...

Recommended for you

A crystal wedding in the nanocosmos

Jul 23, 2014

Researchers at the Helmholtz-Zentrum Dresden-Rossendorf (HZDR), the Vienna University of Technology and the Maria Curie-Skłodowska University Lublin have succeeded in embedding nearly perfect semiconductor ...

PPPL studies plasma's role in synthesizing nanoparticles

Jul 22, 2014

DOE's Princeton Plasma Physics Laboratory (PPPL) has received some $4.3 million of DOE Office of Science funding, over three years, to develop an increased understanding of the role of plasma in the synthesis ...

User comments : 22

Adjust slider to filter visible comments by rank

Display comments: newest first

solar2030
5 / 5 (1) Oct 29, 2010
This is huge !
dirk_bruere
4.5 / 5 (2) Oct 29, 2010
So how do you get a diode to replace a transistor? And how about some nubers eg speed, power, switching energy per bit etc?
Szkeptik
not rated yet Oct 29, 2010
I immediately thought when reading the title that they would do it with a tunneling effect, but the whole point of such a system is to actually control the electron flow. How the hell do they control tunneling electrons?
Skeptic_Heretic
3.1 / 5 (9) Oct 29, 2010
So how do you get a diode to replace a transistor? And how about some nubers eg speed, power, switching energy per bit etc?

Well diodes don't "switch". They're one way.

Vacuum tubes are diodes. Part of the reason why transistors were such a hugh innovation is that they could be made incredibly small when compared to diodes, however, with this breakthrough, diodes can be far smaller than they used to be. This is a huge breakthrough for electronics.
fmfbrestel
not rated yet Oct 29, 2010
Sounds like this might also be able to help advance nantennas. major limiting factor is cheap, easily produced MIM's.
jshloram
4.8 / 5 (4) Oct 29, 2010
Go to Wikipedia and read about Tunnel Diodes. You'll get an understanding about how a two terminal device can be used as switch or amplifier. These things have been around for at least forty years.
fmfbrestel
5 / 5 (3) Oct 29, 2010
right, the just haven't been easily produced. These are.
Dummy
not rated yet Oct 29, 2010
Wonder how long before it hits the commercial market, and I wonder how long before these guys are billionaires?
zslewis91
3 / 5 (2) Oct 29, 2010
These things have been around for at least forty years.


??do you know how to read??? maybe you need to go and re-read....jesus i hate these people>
stealthc
1.8 / 5 (5) Oct 29, 2010
I wonder if the consumer will even benefit from this, or haven't you noticed that people have been benefiting less and less by today's innovations?

I bet you this gets buried for years while current electronics manufacturers continue to rape us for every penny they have left and send it off to the banksters just like every big business nowadays does, sucking your country dry. This works great on cell phones, super facial scanning will make them into nice spy drones.
deatopmg
2.8 / 5 (9) Oct 29, 2010
@stealthc; the "sucking dry" is really done by the government. Those "big businesses" employ people and pay them from their profits on stuff that people buy, and just pass those gov't taxes on to you, the suckee.
Karl Marx is dead and stupid people keep resurrecting his fully discredited dead ideas so they can demonstrate again how they simply don't work.
Better go back on your meds.
MRBlizzard
5 / 5 (2) Oct 30, 2010
reply to dirk bruere
Begore there was TTL, there was DTL; before there was DTL, there was DRL, Diode Resistor Logic. It's power consuming, but it's nanosecond fast. The Oregon State University probably have something better than DRL in mind.
Pyle
3.7 / 5 (3) Oct 30, 2010
I am over simplifying this, but it sounds like, if this works out the way they think it will, we'll be printing circuits that are faster than anything we currently send through a fab. PRINTING, as in spewing metal ink on a board, laying the insulater, spewing more metal ink, then plug in and run.

If this pans out, I wonder what effect this could have to limit research/advances in quantum computing. Think of how cheap oil has killed other energy technologies.

This IS huge!
ECOnservative
5 / 5 (4) Oct 30, 2010
As to frequency, MIM diodes are pretty impressive, "Unlike Schottky diodes, MIM diodes are not affected by parasitic capacitances because they work on the basis of electron tunneling. Because of this, MIM diodes have been shown to operate effectively at frequencies around 150 THz" from http://en.wikiped...Nantenna .
Sebastien_Gagnon
not rated yet Oct 30, 2010
As to frequency, MIM diodes are pretty impressive, "Unlike Schottky diodes, MIM diodes are not affected by parasitic capacitances because they work on the basis of electron tunneling. Because of this, MIM diodes have been shown to operate effectively at frequencies around 150 THz" from http://en.wikiped...Nantenna .


True it's pretty impressive but most of RF electronics cannot support such frequencies anyway (by support i mean produce a coherent signal of course); On the other hand, this research opens the door for a new type of transistors unlike anything we've used before; if MIM diodes doesn't revolutionize electronics, those would surely do. In this optic, this IS a major breakthrought
Ravenrant
not rated yet Oct 30, 2010
This is great but the desire for profit and protecting intellectual property will probably put a huge damper its impact (remember Rambus?).
gopher65
4 / 5 (1) Oct 30, 2010
If you're referring to Rambus's ram (a competitor to DDR), then yes it was faster, but it also generated a lot of heat. It wasn't so much "better" than DDR, it just balanced the give and take of heat vs performance differently. Just like GDDR does.

GDDR5 is basically a slight redesign of DDR3, but it's rebalanced to give much higher performance at the cost of excessive power use and enormous amounts of waste heat. This is a large part of the reason why graphics cards are so bloody hot, and why they have such a high failure rate. That's fine for a graphics card, but not fine for normal system ram.
GSwift7
1 / 5 (1) Nov 01, 2010
On the good side, high frequency and cheap should be great for the people working on optical devices that will benefit from high frequency and low operating power. People like Cysco may be the early adopters that pioneer the new field. I would be willing to bet money that will be the case.

On the not-so-good side, I notice one little thing in this article that sets alarm bells ringing for me: "and it could probably be used with a range of metals that are inexpensive and easily available, like copper, nickel or aluminum" and this part too: "with techniques that would lend themselves to manufacture of devices on a variety of substrates"

Those two statements imply very early stage laboratory trial success, perhaps without enough application trials to zero in on the best combinations of metals and insulators/substrates. I'm always a cautious skeptic though, and maybe I'm being overly so, but again, I'd be willing to bet money that I'm right. It could still be years before it is used.
Pyle
4 / 5 (4) Nov 01, 2010
GSwift7
Beyond tinkering to the right combination of materials, there is the whole logic issue. We have spent decades optimizing circuit design to get to the currently inefficient state semiconductors are at. Throw in a whole new device type and we're still years away from matching current technologies with MIM, but if the production economies prove out like the scientists claim, I wager it will still be revolutionary when it happens.
bluehigh
1 / 5 (1) Nov 02, 2010
@Skeptic - Vacuum tubes are diodes.
No wrong. Maybe initially but tubes soon had a grid between the cathode and anode that allowed control of current flow just as a transistor does. The transistors advantage was of course a huge size reduction (as you say) plus ruggedness.

@Dirk - Transistors are often analysed as 2 diodes back to back because you get a 3 terminal analog that physically compares to the semiconductor structure of a transistor.

Field effect transistors and tunnel diodes both are understood to use the electron tunneling effect to function. So exactly what this method of device fabrication offers as an advantage is not clear in the article. Certainly its not groundbreaking material physics. Perhaps just a neat piece materials engineering looking for an application.
GSwift7
3 / 5 (2) Nov 02, 2010
Here's an article with a much better explaination of what these people did. Notice the part at the bottom, where it says that they plan to creat a three terminal tunneling transistor next.

http://www.eetime...nneling-

antialias_physorg
5 / 5 (1) Nov 06, 2010
Field effect transistors and tunnel diodes both are understood to use the electron tunneling effect to function. So exactly what this method of device fabrication offers as an advantage is not clear in the article.

The advantage is that FET (and tunnel effect transistors like SET) are structured in a much more complex way than the presented layer structure.

Therefore using this MIM structure you could reduce the size of transistors (and increase manufacturing quality) by quite a bit.