backtop


Print 16 comment(s) - last by YashBudini.. on Feb 3 at 7:36 PM


Molybdenite, a common mineral, may produce better transistors than silicon.  (Source: Wikimedia Commons)

The key to the success of the material is it's 2D structure, which limits lost current when the transistor is in standby. The material also has a band gap, making it easily applicable to the creation of transistors.  (Source: Nature Nanotechnology/EPFL)

Researchers at École Polytechnique Fédérale de Lausanne, Europe's top engineering college, have created a MoS4 transistor, which they say could hold the key to future electronic devices.  (Source: EPFL)
Molybdenite is commonly occurring in nature, could it spark a future era of supercomputing

As scientists ponder on how to increase computing power past the theoretical limitations of optical die shrinks, one focus is on developing new materials that are more efficient than traditional silicon-based circuit semiconductor blends.  The 2004 discovery of graphene -- a highly conductive 2D carbon sheet -- earned its discoverers, University of Manchester physicists André Geim and Konstantin Novoselov, a 2010 Nobel Prize in physics.  Now, even as scientists are working to incorporate graphene into transistors, others are pondering on what's the next super-efficient semiconducting material.

Researchers at the Europe's top engineering college, École polytechnique fédérale de Lausanne(EPFL), claim to have that question answered.

Located in Switzerland, the EPFL's Laboratory of Nanoscale Electronics and Structures (LANES) has published preliminary results on circuits composed of molybdenite (MoS2), a commonly occurring mineral that has thus far received little attention as a semiconductor.

Molybdenite looks similar to mica and is often found with quartz deposits.  Its current uses include steel alloy blends and industrial lubricants.

What the EPFL team discovered was that molybdenite is also an excellent semiconductor with a 1.8 electron-volt gap.  The highly 2D naturally occurring structure of the material makes it more efficient electrically than silicon, this could result in a reduction in standby mode losses by as much as a 10^5.  

States [press release] EPFL Professor Andras Kis [profile], "It's a two-dimensional material, very thin and easy to use in nanotechnology. It has real potential in the fabrication of very small transistors, light-emitting diodes (LEDs) and solar cells. In a 0.65-nanometer-thick sheet of MoS2, the electrons can move around as easily as in a 2-nanometer-thick sheet of silicon."

He continues, "But it's not currently possible to fabricate a sheet of silicon as thin as a monolayer sheet of MoS2."

Molybdenite's properties and advantages largely fall in line with those of graphene.  However, graphene does not occur naturally and still relies on relatively expensive production processes.

Graphene, additionally, doesn't have a band gap, as it is a semi-metal.  Graphene transistors have been created by carefully lining up atoms into a graphene sheet/row, but these transistors would likely be less efficient than band-gap designs.

Three other EPFL researchers -- B. Radisavljevic [profile], Professor Radenovic [profile], and M. Brivio  -- assisted on the project.  The work has been published [abstract] in the top peer-reviewed nanotechnology-specific journal, Nature Nanotechnology.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Band gap is higher
By Shadowself on 1/31/2011 5:02:06 PM , Rating: 2
The band gap is more than 50% higher than that of silicon. Thus the switching power will be more than double that of silicon for active components. However, they claim the quiescent state will require *much* less power.

Maybe it will all balance out. Maybe not.

It will be interesting to find out.




RE: Band gap is higher
By Mint on 2/1/2011 10:44:08 AM , Rating: 2
The voltage needed to charge/discharge the gate can be tuned with doping and biasing of the substrate. That's not a big issue, AFAIK.

I think the key advantage here is that MoS2 has a higher mobility than silicon, but I assume it's just electron mobility. As another poster mentioned, though, you need hole mobility as well to have a good CMOS process, and I don't know about that here.


"I'd be pissed too, but you didn't have to go all Minority Report on his ass!" -- Jon Stewart on police raiding Gizmodo editor Jason Chen's home














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki