Researchers Build First Molybdenite Microchip

An anonymous reader writes "A Swiss team may have found an alternative to silicon microchips which could result in smaller, more flexible and less energy hungry processors. The Swiss team's chip does not use silicon, but molybdenite (MoS2) a dark-colored, naturally occurring mineral that is able to be used in much thinner layers (paywall)."

So soon.

[jellomizer]

The iPhone 10.
No glass no hard plastic. Built in a classy silicone (you know the rubbery stuff) case. That will not break when you drop it. And new sensors that detect bending for more new apps....

On the other hand, I don't think I want to see these apps.

Re:

[MisterMidi]

You mean like Nokia does [youtube.com]?

Re:So soon.

[jellomizer]

No no. This is an Apple product. It is better, and innovative.

Re:

[rwise2112]

But also come several years later, and still be the first!

Re:

[RockDoctor]

[drum roll]
"Get out! We don't serve superluminal particles in here!" said the bar man.
A neutrino walks into a bar.
[cymbal crash]

That would be an Apple-flavoured neutrino then, not an electron-, muon- or tau-flavoured neutrino?

Silicone?

[Pope]

Is there a chance the case could bend?

Re:

[Yvan256]

Not on your life, my Slashdot friend!

Re:

[tragedy]

The tab came off my pudding can!

But

[bhcompy]

But is it as common and cheap as silicon?

Re:

[bhcompy]

They'd be even less cheap if silicon wasn't so common

Re:

[fnj]

Actually the processing costs so far outweigh the materials cost that it is not possible to say that, unequivocally.

Re:But

[doconnor]

I don't think the cost of the raw material is significant compared to the cost to design and manufacture computer chips.

Re:But

[betterunixthanunix]

Wikipedia says Molybdenum is the 54th most abundant element on Earth. This is less abundant than silicon, but nowhere near as rare as other commonly used elements in semiconductors; Indium is far more rare.

Re:But

[Sique]

But not as common as Silicon, which you can literally shovel in your backyard - the upper layer of the Earth crust is called Sial because of the two most abundant metals, Silicon and Aluminium. Iron comes in as a strong third.

There is a reason, why the three commerically most used metals are also the three most abundant. Molybdenum is often found in the compounds iron ore consists of, but it takes quite some energy to extract the Molybdenum from the iron ore.

Re:

[EdZ]

Comparing garden-variety sand to the highly purified, monocrystalline, carefully doped silicon used in chip fabs is like grabbing a lump of charcoal and claiming graphene is just lying around.

Re:

[RockDoctor]

On the other hand, quite pure (95%+) deposits of silicon dioxide, from which you can start your purification processing, is a lot more common on the surface of the Earth than diamond. Or reasonably pure graphite, for that matter.

Re:But

[MagusSlurpy]

It takes less energy to get moly out of ferrous ore than it does to reduce silicon oxide to silicon. It's also a matter of availability - even though there's more silicon than molybdenum, molybdenum is often much more pure, and found in distinct deposits, and easy to mine, versus silicon being mixed in with all kinds of other crap and distributed wildly all throughout the crust. It's the same thing with rare earths - for example, indium isn't actually rare, it's just that it's EVERYWHERE in very small amounts, instead of convenient little deposits like gold, which is in the crust at about 1/500th the amount.

Re:

[Kamiza Ikioi]

But is it better than Adamantium?

Re:

[RockDoctor]

Is molybdenite better than adamantium? Very broad question.

But for certain, the first has a very much lower coefficient of fiction.

Re:

[ChrisMaple]

Molybdenum has a thermal coefficient of expansion quite close to silicon. Because of that and other reasons, it is frequently used as a thermal conductor between the silicon die and a copper or aluminum heat spreader. My point is, it's already used with semiconductor devices and using it in the semiconductor itself isn't going to greatly increase demand.

Re:

[MagusSlurpy]

As far as raw material goes, yes. And the purification methods are vastly different, but it seems likely that when scaled up they will be priced similarly to those of silicon. They're certainly less complicated.

Also, this is the same stuff [wikipedia.org] that lubricant companies like Valvoline put in high-mileage oil to provide friction protection when all your oil leaks out of your crumbly, busted-ass gaskets, as well as a billion other uses. It's really cool stuff.

Re:

[RockDoctor]

The cost of silicon chips is almost entirely in the processing, not in the raw materials. Clean sand with 95%+ SiO2 costs almost exactly the same as the fuel to move it from the mine to the first processing plant.

At that point, if you've got an industrial source of reasonably pure molybdenum (which there is - it's used in considerable tonnages in the steel industry), you tap into that. And start purifying it.

Fabrication

[P-niiice]

Also, will we have to retool the universe to produce these chips? I don't want to be too old to enjoy them!

Wait, what?

[Wonko the Sane]

On top of that, they also use numerous precious metals including gold, platinum and silicon.

Gold, sure; platinum, no problem; silicon, WTF?

Currently used as a strengthening agent in plastic, molybdenite is abundant in the natural word. Speaking to the BBC, Prof Andras Kis, the director of the Laboratory of Nanoscale Electronics and Structures (LANES) in Lausanne, said that his team’s research could transform the world of computing. ”[Currently] there is something like 19 million metric tonnes around,” Prof Kis said in an interview. ”You can just go on some websites on the internet and buy a 1cm by 1cm crystal for around $100 [£64].”

Abundant as compared to what? Silicon is the third most abundant element on earth and makes up 15% of its mass. Molybdenum is a rare earth element. Also, you can't use the current price of some element based on it not being used to make microprocessors and expect that the cost won't change if you increase the demand for it by many orders of magnitude. There might be good reasons for building microprocessors from molybdenite but replacing scarce silicon with abundant molybdenite is not one of them.

Re:Wait, what?

[ardor]

I tend to agree, however, keep in mind:
Silicon is abundant. Highly pure silicon is not. You need the latter for microchips.

Re:

[steelfood]

So it's a question of whether the purification process of silicon is more expensive than the mining and purification process of molybdenum. Unless the molybdenum doesn't need to be purified to the same degree.

Molybdenite probably won't replace silicon in all cases. Probably, it will be used in select components and chips, where speed and power usage are both equally important factors.

Re:

[aaron552]

where speed and power usage are both equally important factors.

Like transistors?

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:

[idontusenumbers]

The WTF on silicon was because silicon was listed as a precious metal despite it being abundant.

Re:Wait, what?

[compro01]

Well, silicon is a metaloid, and by price, ultra-high-purity silicon is certainly precious.

A 300mm IC-grade wafer costs about $150. Weights about 1.6 grams.

That's about $93,000/kilo. Gold is about $55,000/kilo.

Re:

[compro01]

Dammit, screwed up my numbers. I multiplied by 2 rather than squaring.

The wafers should weigh about 120 grams and that comes to about $1200/kilo, though that's still a bit more expensive than silver.

Re:

[Algae_94]

That price drops dramatically if you drop the wafer and break it.

Re:

[Algae_94]

Although you did mention 300mm wafers. I don't think there are many (if any) fabs that have those handled by hand so breakage is not a big issue compared to yield loss from defects on the wafer.

Re:

[tragedy]

That's a very good point. A 300mm IC-grade wafer isn't a measured amount of precious material, it's more like an expensive product. If it's broken, or simply flawed to begin with, most of the value is gone and it's back to being highly refined silicon to be melted and formed into a crystal and cut again. A more realistic way to look at the value of the silicon is how much they pay for the refined material they make the big crystals out of.

Re:Wait, what?

[Anonymous Coward]

Molybdenum is not a rare earth element (lanthanoid), it's a transition metal.
And rare earth elements are not neccessarily rare.

Re:

[vikingpower]

Molybden(i)um most frigging certainly is NOT a "rare earth element". It belongs to the iron group, along with e.g. Vanadium, Chrome, Nickel and, well, yes, Iron. Molybden(i)um occurs in extractable concentrations on almost all locations where iron ore is abundant.

Re:

[khallow]

Molybdenum is a rare earth element.

Not in any sense of the phrase. It's not rare nor a rare earth element [wikipedia.org].

There might be good reasons for building microprocessors from molybdenite but replacing scarce silicon with abundant molybdenite is not one of them.

The real question along these lines is how much energy does it take to make ICs from each material. And frankly, ICs are generally high enough value that there's not much point in considering material or energy input costs.

Duplicate. Old news. Nothing to see here...

[Anonymous Coward]

This has already been reported: http://science.slashdot.org/story/11/01/31/021258/molybdenite-as-an-alternative-to-silicon [slashdot.org]

And yes, they're the same. They link to the same Nature Nano article...

MOS-COS-MOS

[regular_guy]

Could this then be considered a MoS-COS-MOS?

And a lubercant too. We can use it in mice :)

[AbrasiveCat]

The properties also include being a good high temperature lubricant, so the gamers who are pushing their mice to the limit would like this in those mice. Maybe a bit messy. I wonder if the MoS2 flakes and would be hard to clean up.

Molybdenite occurs with radioactive Rhenium-187

[G3ckoG33k]

Molybdenite occurs together with Rhenium-187 which is radioactive. http://en.wikipedia.org/wiki/Molybdenite [wikipedia.org]

Is that a serious issue here?

Re:

[NEDHead]

Perhaps it can become a self powered chip?

Re:

[kiehlster]

I venture they use the radioactivity as a clock chip. Sounds like it has a pretty good half-life.

Re:

[NEDHead]

Perhaps it can also be adapted for an on board random number generator

Re:

[GodfatherofSoul]

No, we'll just let 3rd world peasants mine it for us.

More info

[Geoffrey.landis]

Moly disulfide is mostly known as a lubricant.

Interesting, but I wouldn't sell my stocks in silicon electronics yet. Silicon is way down the learning curve. I wouldn't bet a new semiconductor against it.

Some of what the article says is a little dubious, like the fact that silicon "The surface likes to oxidise - it likes to bind with oxygen... and that makes its electrical properties degrade when you want to make a very thin film." Yes, it forms oxide easily. No, that doesn't "degrade" the electrical properties-- in fact, this is exactly why silicon is so incredibly useful in electronics. Oxide, and the fact that silicon oxide passivates the surface to prevent electron-hole recombination at the dangling bonds, is what makes silicon electronics possible. I note that the moly disulfide transistors use hafnium oxide for a gate. That's a high dielectric constant material that is indeed also used in silicon, but the silicon oxide is still the critical interface.

By the way, I think there's slightly better info from eetimes http://www.eetimes.com/electronics-news/4212757/New-material-for-semis-said-to-beat-silicon [eetimes.com] or physicsworld http://physicsworld.com/cws/article/news/45056 [physicsworld.com]

Molybdenum element 42..

[Anonymous Coward]

What I can't believe no one has pointed out yet is that one of the primary elements in Molybdenite is Molybdenum, which is the 42nd element in the periodic table. 42 = The answer to the ultimate question of life the universe and everything. Therefore this has to be the best answer to any microchip issues QED

IN case anyone has trouble...

[HaeMaker]

It's Moe-lib-den-ite.

Re:

[Ultracrepidarian]

This may prove to be a problem for some conservatives.

PVs?

[surd1618]

I am wondering, this stuff is really far away for microchips, but what about PVs? I am trying to find info on how they got the conductivity to be better, and if it is being used for junctions in photovoltaics.

I am well aware that molybdenum is being used for backing contacts. Actually that is part of why I am posting this, because it is making searching more complicated.

Oh great

[TheSkepticalOptimist]

Here comes the parade of the Moly vs Silicon fanboys. Personally I am a graphene fanboy.

Re:

[vandamme]

What about Gallium arsenide? Silicon carbide? Indium phosphide? Gallium nitride? All were going to replace silicon.