One-molecule-thick material has big advantages: Researchers produce complex circuits from molybdenum disulfide

Aug 23, 2012 by David L. Chandler
Diagram shows the flat-sheet structure of the material used by the MIT team, molybdenum disulfide. Molybdenum atoms are shown in teal, and sulfur atoms in yellow. Image courtesy of Wang et al.

The discovery of graphene, a material just one atom thick and possessing exceptional strength and other novel properties, started an avalanche of research around its use for everything from electronics to optics to structural materials. But new research suggests that was just the beginning: A whole family of two-dimensional materials may open up even broader possibilities for applications that could change many aspects of modern life.

The latest "new" material, molybdenum disulfide (MoS2) — which has actually been used for decades, but not in its 2-D form — was first described just a year ago by researchers in Switzerland. But in that year, researchers at MIT — who struggled for several years to build electronic circuits out of graphene with very limited results (except for radio-frequency applications) — have already succeeded in making a variety of from MoS2. They say the material could help usher in radically new products, from whole walls that glow to clothing with embedded electronics to glasses with built-in display screens.

A report on the production of complex electronic circuits from the new material was published online this month in the journal ; the paper is authored by Han Wang and Lili Yu, graduate students in the Department of and Computer Science (EECS); Tomás Palacios, the Emmanuel E. Landsman Associate Professor of EECS; and others at MIT and elsewhere.

Palacios says he thinks graphene and MoS2 are just the beginning of a new realm of research on two-dimensional . "It's the most exciting time for electronics in the last 20 or 30 years," he says. "It's opening up the door to a completely new domain of electronic materials and devices."

Like graphene, itself a 2-D form of , molybdenum disulfide has been used for many years as an industrial lubricant. But it had never been seen as a 2-D platform for until last year, when scientists at the Swiss university EPFL produced a transistor on the material.

MIT researchers quickly swung into action: Yi-Hsien Lee, a postdoc in associate professor Jing Kong's group in EECS, found a good way to make large sheets of the material using a chemical vapor deposition process. Lee came up with this method while working with Lain-Jong Li at Academia Sinica in Taiwan and improved it after coming to MIT. Palacios, Wang and Yu then set to producing building blocks of on the sheets made by Lee, as well as on MoS2 flakes produced by a mechanical method, which were used for the work described in the new paper.

One-molecule-thick material has big advantages
An optical-microscope image shows a complex integrated circuit, called a JK flip-flop circuit, a basic logic device, made on a piece of molybdenum disulfide by the MIT team. Image courtesy of Wang et al.

Wang had been struggling to build circuits on graphene for his doctoral thesis research, but found it much easier to do with the new material. There was a "hefty bottleneck" to making progress with graphene, he explains, because that material lacks a bandgap — the key property that makes it possible to create transistors, the basic component of logic and memory circuits. While graphene needs to be modified in exacting ways in order to create a bandgap, MoS2 just naturally comes with one.

The lack of a bandgap, Wang explains, means that with a switch made of graphene, "you can turn it on, but you can't turn it off. That means you can't do digital logic." So people have for years been searching for a material that shares some of graphene's extraordinary properties, but also has this missing quality — as molybdenum disulfide does.

Because it already is widely produced as a lubricant, and thanks to ongoing work at MIT and other labs on making it into large sheets, scaling up production of the material for practical uses should be much easier than with other new materials, Wang and Palacios say.

Wang and Palacios were able to fabricate a variety of basic electronic devices on the material: an inverter, which switches an input voltage to its opposite; a NAND gate, a basic logic element that can be combined to carry out almost any kind of logic operation; a memory device, one of the key components of all computational devices; and a more complex circuit called a ring oscillator, made up of 12 interconnected transistors, which can produce a precisely tuned wave output.

Palacios says one potential application of the new material is large-screen displays such as television sets and computer monitors, where a separate transistor controls each pixel of the display. Because the material is just one molecule thick — unlike the highly purified silicon that is used for conventional transistors and must be millions of atoms thick — even a very large display would use only an infinitesimal quantity of the raw materials. This could potentially reduce cost and weight and improve energy efficiency.

In the future, it could also enable entirely new kinds of devices. The material could be used, in combination with other 2-D materials, to make light-emitting devices. Instead of producing a point source of light from one bulb, an entire wall could be made to glow, producing softer, less glaring light. Similarly, the antenna and other circuitry of a cellphone might be woven into fabric, providing a much more sensitive antenna that needs less power and could be incorporated into clothing, Palacios says.

The material is so thin that it's completely transparent, and it can be deposited on virtually any other material. For example, MoS2 could be applied to glass, producing displays built into a pair of eyeglasses or the window of a house or office.

Ali Javey, an associate professor of electrical engineering and at the University of California at Berkeley, who was not involved in this research, says layered materials such as MoS2 are "a promising class of materials for future electronics," but cautions that "the future looks bright for layered semiconductors, but still work needs to be done to better understand their performance limits and large-scale manufacturing."

Overall, Javey says, the MIT team's research is "elegant" work that "takes an important step forward in advancing the field of layered semiconductors."

In addition to Palacios, Kong, Wang, Yu and Lee, the work was carried out by graduate student Allen Hsu and MIT affiliate Yumeng Shi, with U.S. Army Research Laboratory researchers Matthew Chin and Madan Dubey, and Lain-Jong Li of Academia Sinica in Taiwan. The work was funded by the U.S. Office of Naval Research, the Microelectronics Advanced Research Corporation Focus Center for Materials, the National Science Foundation and the Army Research Laboratory.

Explore further: Simpler process to grow germanium nanowires could improve lithium ion batteries

More information: pubs.acs.org/doi/abs/10.1021/nl302015v

Related Stories

Graphene could lead to faster chips

Mar 19, 2009

(PhysOrg.com) -- New research findings at MIT could lead to microchips that operate at much higher speeds than is possible with today's standard silicon chips, leading to cell phones and other communications ...

First molybdenite microchip

Dec 05, 2011

(PhysOrg.com) -- Molybdenite, a new and very promising material, can surpass the physical limits of silicon. EPFL scientists have proven this by making the first molybdenite microchip, with smaller and more ...

Atom-thick sheets unlock future technologies

Feb 03, 2011

(PhysOrg.com) -- A new way of splitting layered materials, similar to graphite, into sheets of material just one atom thick could lead to revolutionary new electronic and energy storage technologies.

Recommended for you

A single molecule device for mobile phones

15 hours ago

Researchers from the Delft University of Technology, Groningen University and the FOM Foundation have designed a single molecule which can act as a useful building block in nanometer-size circuits. They found ...

User comments : 11

Adjust slider to filter visible comments by rank

Display comments: newest first

antialias_physorg
3.3 / 5 (3) Aug 23, 2012
even a very large display would use only an infinitesimal quantity of the raw materials. This could potentially reduce cost and weight and improve energy efficiency.

Not to mention that the transistor part of a display is usually not transparent . A part of your backlighting is therefore lost.

With thransistors just one molecule thick that effect is negligible - which could mean the switching electronics can be as large as the pixel if need be (i.e. you can go to vanishingly small pixel sizes without reducing luminosity/efficiency)

an entire wall could be made to glow,

I'd rather have entire wall displays - which shouldn't be too much harder to produce than a uniform glow with this. Imagine having any kind of scene play on your wall (sunrises, movies (finally get rid of your TV/beamer), great vistas, trippy mood lighting, ...)
Jeffhans1
1.8 / 5 (6) Aug 23, 2012
Why bother with a wall when you could put the display right on your cornea and include a see through camera at the same time. Once you shrink to this level of IC you can put the entire capture and display system directly in your line of site and make the thing overlay and enhance your regular vision.
antialias_physorg
2.7 / 5 (3) Aug 23, 2012
Why bother with a wall when you could put the display right on your cornea and include a see through camera at the same time.

How would you power it, though?
You could transfer the energy via terahertz radiation - but the eye is notoriously bad at dissipating any kind of heat. Also getting an upgrade would require an operation.

A see through camera is a contradiction in terms. Cameras need to capture photons to get information about what they see. Any photon captured by a device would not go through (i.e. not be available to your eye).
Tektrix
Aug 23, 2012
This comment has been removed by a moderator.
tpb
4 / 5 (1) Aug 23, 2012
I see that the ring oscillator was only running at 1.6 MHz versus silicon that runs at tens to hundreds of GHz.
I wonder if one molecule thick layers implies very high resistances and inherently low speeds.
SoylentGrin
2 / 5 (1) Aug 23, 2012
Make it as a contact lens, powered by blinking.
(http://phys.org/n...#ajTabs)
Shunting 25% of the photons hitting your eye to a recording device wouldn't be that noticable, and is something the brain would adapt to pretty quickly.
Tektrix
2 / 5 (1) Aug 23, 2012
Power them with salty (ionic) tears: http://electronic...tery.htm
Tektrix
4 / 5 (1) Aug 23, 2012
@ tpb: "I wonder if one molecule thick layers implies very high resistances and inherently low speeds."

Well, the electron mobility in graphene is about 100 times better than copper and resistance is about 35% less than copper- and this is at room temperature. So monomolecularity alone does not imply the performance hits you are wondering about.

http://news.softp...34.shtml
tpb
2.5 / 5 (2) Aug 23, 2012
@Tektrix
This is not graphene, but MoS2 and I haven't seen anything about electron mobility being unusual in MoS2.
Also extremely thin layers implies higher capacitance at the same time as higher resistance.
It would be interesting to see some numbers on what is theoretically possible with MoS2.
FastEddy
1 / 5 (5) Aug 27, 2012
@Tektrix - This is not graphene, but MoS2 and I haven't seen anything about electron mobility being unusual in MoS2.


Mono-layered MoS2 must have some kind of band gap, otherwise it would have the same, same problems as Graphene ("The lack of a bandgap, Wang explains, means that with a switch made of graphene, "you can turn it on, but you can't turn it off....")

Also extremely thin layers implies higher capacitance at the same time as higher resistance.
It would be interesting to see some numbers on what is theoretically possible with MoS2.


Yes, this "stray capacitance", resistance (and/or inductance = C R L = impedance) question is a tough one and is what puts upper limits on the bandwidth (speed). That's the way God planned it? Or maybe some other mono-layered molecule will do this trick with reduced impedance.

All in all a very good start ... I still think that Graphene may prove to be that nearly "perfect" conductor that will work on the macro scale = the grid.
FastEddy
1 / 5 (5) Aug 27, 2012
There is a whole other world out there other than the TV typewriter (and eyewear monitors) ... Graphene and now MoS2, et al, will find vast untaped venues for improving on Mother Nature.
FastEddy
1 / 5 (5) Aug 27, 2012
... With thransistors just one molecule thick that effect is negligible - which could mean the switching electronics can be as large as the pixel if need be ... an entire wall could be made to glow ... I'd rather have entire wall displays - which shouldn't be too much harder to produce than a uniform glow with this. Imagine having any kind of scene play on your wall (sunrises, movies (finally get rid of your TV/beamer), great vistas, trippy mood lighting, ...)


Further: a wall of "glass" that serves as window, display and room lighting ... on either side?