New Ultrathin Semiconductor Materials Found That Outdo Silicon

[Megalith]

Stanford engineers have identified two semiconductors that share or even exceed some of silicon’s desirable traits: hafnium diselenide and zirconium diselenide. Like silicon, both materials “rust” when exposed to oxygen (required for isolating circuitry) and offer the right energy to switch transistors on, but diselenides can be fashioned into circuits just three atoms thick. These ultrathin semiconductors could be made into transistors 10 times smaller than anything possible with silicon today.

“It’s a bit like rust, but a very desirable rust,” said Eric Pop, an associate professor of electrical engineering, who co-authored with post-doctoral scholar Michal Mleczko a paper that appears in the journal Science Advances. The new materials can also be shrunk to functional circuits just three atoms thick and they require less energy than silicon circuits. Although still experimental, the researchers said the materials could be a step toward the kinds of thinner, more energy-efficient chips demanded by devices of the future.

 

[sfsuphysics]

is it that time of the year again? We "discovered" some nifty material that's way better than silicon, and next year we're still using silicon in literally everything.

 

[Quartz-1]

I remember similar hope for Gallium Arsenide 30 years ago. But we'll see.

 

[ElGuapo242]

I think the biggest problem with these new materials, is the cost of switching over to them. Silicon has been around for decades now and all the foundries are geared for it. Switching up to a new material is probably going to be quite pricey. I guess we will see.

 

[M76]

is it that time of the year again? We "discovered" some nifty material that's way better than silicon, and next year we're still using silicon in literally everything.

Being better doesn't mean more practical in the short term. Practicality is also determined by a whole lot of outside factors, as others have already mentioned changing all manufacturing processes.

 

[Ultima99]

All the alternatives are super expensive, that's why you see zero movement. We'll milk silicon until there's no room left for improvements first.

 

[naib]

I remember similar hope for Gallium Arsenide 30 years ago. But we'll see.

GaAs was used, but it has some annoying limitation. GaN is coming into flavour for higher power semiconductors

The other thing to consider, almost the entire semiconductor supply chain is geared to Si. If something comes along that can use almost the same processes then bam... the same fabs are used, if not they become obscure footnotes in lecture materials

 

[Living Weapon]

Link in Stanford News is broken - use this one for full text of the paper: http://advances.sciencemag.org/content/3/8/e1700481.full

 

[Living Weapon]

Wow, OK. After reading a bit, these guys are pretty far down the path of checking fabrication techniques - either due to expedience for the sake of their own work, or possibly with an eye toward commercialization.

The wafers need to be made in high vacuum, but once passivated many of the handling steps can be done in nitrogen glove boxes (need 10^^-5 and 10^^-8 torr for a few process steps, but looks to tolerate a relatively impure atmosphere during electron beam lithography). There are some good notes throughout the paper about the levels of BS the process can tolerate and still produce useful transistors, and it looks surprisingly robust.

That's actually easier than the requirements for EUV lithography - so yeah, geeking out a little

 

[Poseur]

Silicon $
Zirconium $$
Hafnium $$$

 

[SecretStash]

I think the biggest problem with these new materials, is the cost of switching over to them. Silicon has been around for decades now and all the foundries are geared for it. Switching up to a new material is probably going to be quite pricey. I guess we will see.

Really? I would wonder how many nice things we miss out on because switching is cost prohibitive even if production costs would be the same or less after the initial ramp up period.

 

[mbelue]

Silicon $
Zirconium $$
Hafnium $$$

In the future there might be good reasons for $2000 processors...

 

[velusip]

Silicon $
Zirconium $$
Hafnium $$$

And with emphasis -- the approximate percentage found in Earth's crust:

Silicon, 27%
Zirconium, 0.013%
Hafnium, 0.00058%

 

[Hagrid]

I remember similar hope for Gallium Arsenide 30 years ago. But we'll see.

I worked with GaAs and that stuff is really brittle. This was a long while ago, so it may be different now. It had stress fractures just from the sawing process. Then the pick and place/die bond/wire bond added even more.(though I believe mostly because of the sawing, it just kept making it worse)

 

[Rob94hawk]

Silicon $
Zirconium $$
Hafnium $$$

Unobtanium $$$$
Adamantium $$$$$
Vibranium $$$$$$

 

[Zarathustra[H]]

I worked with GaAs and that stuff is really brittle. This was a long while ago, so it may be different now. It had stress fractures just from the sawing process. Then the pick and place/die bond/wire bond added even more.(though I believe mostly because of the sawing, it just kept making it worse)

Sounds like a chemical etching process after the sawing may be necessary, in order to eat away some of the micro-fractures.

 

[Deleted member 184142]

Really? I would wonder how many nice things we miss out on because switching is cost prohibitive even if production costs would be the same or less after the initial ramp up period.

If the cost over time leveled out in favor of the new material, they would use it. Companies don't just look at cost right now, they look at ROI, even if cost to switch over was very high, but ROI was fast or high enough, they would move in a heart beat to be the first ones to be in the market. The problem is that most of these are not only high cost to move/switch to, but the cost after the fact is also more expensive, as pointed out, the material it self is many fold the cost of Silicon. We have seen many material advancements, but most do not make it into consumer chips due to cost and so you see it only used in fringe cases where cost is not much of a factor.

No one is going to want to switch over to a more expensive process, with big up front investment and higher material cost, it would be business suicide.

 

[Khahhblaab]

If the cost over time leveled out in favor of the new material, they would use it. Companies don't just look at cost right now, they look at ROI, even if cost to switch over was very high, but ROI was fast or high enough, they would move in a heart beat to be the first ones to be in the market. The problem is that most of these are not only high cost to move/switch to, but the cost after the fact is also more expensive, as pointed out, the material it self is many fold the cost of Silicon. We have seen many material advancements, but most do not make it into consumer chips due to cost and so you see it only used in fringe cases where cost is not much of a factor.

No one is going to want to switch over to a more expensive process, with big up front investment and higher material cost, it would be business suicide.

Yeah. What needs to be done is show some 'proven" advantage for the new tech, then show how it could help the military. We forget that one of the greatest motivators for silicone today was the military and america getting behind integrated circuitry so america could win the space race. Without NASA's investment, IC chips and the necessary infrastructure invention would have been much slower. The need for lighter and more powerful electronics and computers, so NASA could lift off the launch pad, fast-tracked miniaturization.

The tools that make the nano scale chips today are made from a few companies. Novellus, AME, and ASML. A few more. They are already cash strapped to improve technologies based on silicone to make it work at even smaller dimensions. Imagine making machines based on a new material. The only way it would work is if it could be exchanged with silicone. If new materials means new machines then maybe a company like amazon would get into the chip making business.

It would take a bank account like amazons to create a new "nano scale better than silicone technology". I don't think current silicon base can be replaced easily. Even if it proves to be better.