Could Molybdenite Replace Silicon?

[sanman]

Here's an interesting article about molybdenite having superior semiconductor properties in comparison to silicon:

http://www.eurekalert.org/pub_releases/2011-01/epfd-nta012811.php

What's the verdict? Is this stuff the better material? Could be it a viable successor for future applications?

 

[caffenta]

I don't know anything about MoS₂, but having superior properties compared to Si is nothing to write home about. Si is a very poor semiconductor.

 

[sanman]

And yet our modern informatics technologies are overwhelmingly based on it. Perhaps because of its abundance?

Molybdenite is supposed to be fairly abundant too, although probably not as much as silicon. I can't wait to see someone come out with the first molybdenite chip. I wonder where we'd be today if this material had been selected over silicon from the very beginning?

 

[caffenta]

I wonder where we'd be today if this material had been selected over silicon from the very beginning?

The major advantage of Si is really the Si/SiO₂ interface, and the real question is where would we be without it. I wonder if we may have called it quits on the MOSFET if it had been as hard as with other semiconductors like III-Vs (there is still no equivalent MIS type device in III-Vs, but we don't really need it). Without MOSFETs, you would not have VLSI. And ironically, the low electron mobility in Si is what makes CMOS possible. The success of Si was a driving force for III-Vs since we knew that the concept worked. But Si is still a poor semiconductor. There are plenty of others in use right now for applications that cannot be accomplished with Si. Your cell phone is full of III-V devices.

Don't hold your breath on MoS₂ devices. Just because the place that made them says it will revolutionize the world doesn't mean it's true. They just want funding. I've heard it all before. But it doesn't necessarily mean they won't work either. Just take it with a grain of salt.

 

[sanman]

Well, what would be the counterpart to the Si/SiO2 interface? Couldn't MoS2 be similarly modified through dopants as well? I think the next step is for someone to create some practical working devices from MoS2, and show their performance advantages.

Currently there are mixed lattice substrates which mate one silicon to other materials. Maybe this could be done with MoS2 as well.

 

[PAllen]

And yet our modern informatics technologies are overwhelmingly based on it. Perhaps because of its abundance?

Molybdenite is supposed to be fairly abundant too, although probably not as much as silicon. I can't wait to see someone come out with the first molybdenite chip. I wonder where we'd be today if this material had been selected over silicon from the very beginning?

As a former mineral collector, just want to pass the two comments:

1) Silicon is the most common solid element in the Earth's crust by far. However, extracting elemental silicon is extremely energy intensive, and it never occurs in elemental form in nature.

2) Molybdenite is a reasonably common mineral and occurs as is. Only purification needed to use it as a semiconductor.

 

[Araucaria]

The beauty of Si as a semiconductor was that you could oxidise the surface then deposit an electrode on top of it to form a MOSFET. Now, with deposited dilectrics like HfO2, that advantage is unimportant and the opportunity exists for new substrates. MoS2 looks interesting but a viable deposition technology shall be required. That probably requires volatile reactive precursors for MOCVD or ALD. There's a lot more to it than only purification to use it as a semiconductor!

 

[Gokul43201]

Due to scaling issues with Si, the ongoing search for replacements is extremely widespread. One need only check out the "Beyond CMOS" session at one of the anuual conferences of the Materials Research Society. From CNT based FET's to complex oxide based Mott Transistors to (now) Molybdenite, there have been thousands of materials that have been studied for potential to replace Silicon, and many dozens have graced the pages of Science, Nature and Nanoletters, as suitable candidates. Yet, here we are, still with Si-based technology.

I'm not saying that Silicon won't be replaced - I believe that to some extent it almost definitely has to be, if we hope to keep scaling channel sizes downward ... but in the mean time, one has to not be tempted to think that the next material that is expected to potentially replace Silicon (especially in the Press), will likely do so. There's many a slip between the lab and the lip of the science reporter.