Molybdenite replacing silicon in ICs?

[thilanliyan]

http://www.engadget.com/2011/12/06/first-molybdenite-ic-delivers-silicon-crushing-chip-shrinking/

 

[PlasmaBomb]

Sounds interesting.

 

[pm]

Full article:
http://pubs.acs.org/doi/full/10.1021/nn203715c

Vgs to to Ids graph in the middle.
http://pubs.acs.org/appl/literatum/...15c/aop/images/large/nn-2011-03715c_0003.jpeg

A graph of using it as an inverter.... input voltages vary from -2V to 2V..... :|
http://pubs.acs.org/appl/literatum/...15c/aop/images/large/nn-2011-03715c_0004.jpeg

I love how Engadget always hypes these things like silicon is dead and this is the next big thing. From reading the article, it was hard enough to build two transistors. The idea that they could build hundreds of millions of chips with 2 billion transistors on each one of them is a long way away and I'm not even clear on how it would work... how do you do binary arithmetic with a cell that takes -2V to 2V in and outputs 0V to 2V out?

 

[SunnyD]

Full article:
http://pubs.acs.org/doi/full/10.1021/nn203715c

Vgs to to Ids graph in the middle.
http://pubs.acs.org/appl/literatum/...15c/aop/images/large/nn-2011-03715c_0003.jpeg

A graph of using it as an inverter.... input voltages vary from -2V to 2V..... :|
http://pubs.acs.org/appl/literatum/...15c/aop/images/large/nn-2011-03715c_0004.jpeg

I love how Engadget always hypes these things like silicon is dead and this is the next big thing. From reading the article, it was hard enough to build two transistors. The idea that they could build hundreds of millions of chips with 2 billion transistors on each one of them is a long way away and I'm not even clear on how it would work... how do you do binary arithmetic with a cell that takes -2V to 2V in and outputs 0V to 2V out?

60 years ago, I'm sure people were saying the same thing about germanium.

 

[pm]

60 years ago, I'm sure people were saying the same thing about germanium.

I doubt that Engadget said anything of the sort 60 years ago.

But, all of the early semiconductor research was on germanium. Like the first 10-20 years, it was all germanium. If anything, at the time silicon seemed like a bit of a long shot for IC's.

So, you disagree with what I wrote? Based on a fairly detailed reading of the paper, I'd still think that graphene is a better gamble. Molybdenite has a few advantages, but I'd still stay with graphene even with the bandgap and other issues.

 

[dajeepster]

silicon will never be dead because it's cheap and abundant

 

[Idontcare]

It's good to have options. R&D is all about false-starts and dead-ends.

What looks good now could be next year's 157nm litho catastrophe. Having plan B in the works is rarely a bad idea.

 

[Phynaz]

silicon will never be dead because it's cheap and abundant

640K ought to be enough for anybody.

The physical limits of silicon will be reached pretty soon.

 

[Denithor]

It's good to have options. R&D is all about false-starts and dead-ends.

What looks good now could be next year's 157nm litho catastrophe. Having plan B in the works is rarely a bad idea.

Tell me about it! I work as an R&D chemist in the industrial lubricants business, there are far, FAR more false starts that go nowhere than solid finds that lead to good/new/improved products.

 

[frostedflakes]

640K ought to be enough for anybody.

The physical limits of silicon will be reached pretty soon.

But he's right that it will probably continue to be used for a long time in circuits that don't require ridiculously high integration and cutting edge performance. Just like 0.25um and older nodes like that are still used today for many applications.

 

[CTho9305]

Full article:
http://pubs.acs.org/doi/full/10.1021/nn203715c

Vgs to to Ids graph in the middle.
http://pubs.acs.org/appl/literatum/...15c/aop/images/large/nn-2011-03715c_0003.jpeg

A graph of using it as an inverter.... input voltages vary from -2V to 2V..... :|
http://pubs.acs.org/appl/literatum/...15c/aop/images/large/nn-2011-03715c_0004.jpeg

I love how Engadget always hypes these things like silicon is dead and this is the next big thing. From reading the article, it was hard enough to build two transistors. The idea that they could build hundreds of millions of chips with 2 billion transistors on each one of them is a long way away and I'm not even clear on how it would work... how do you do binary arithmetic with a cell that takes -2V to 2V in and outputs 0V to 2V out?

Looks like you get a few orders of magnitude between 0V and 2V, so you could build NMOS- or PMOS-style logic (with static loads) rather than CMOS . Ignoring power implications of logic that has "on" and "less on" states, at 2V you're in trouble anyway because you're going to break down the dielectric between your metals if you try to build wires at the pitch achievable with modern silicon circuits.

 

[antisocialmunky]

Sounds too much like Molyneux to not end up overhyped.

 

[SunnyD]

I doubt that Engadget said anything of the sort 60 years ago.

But, all of the early semiconductor research was on germanium. Like the first 10-20 years, it was all germanium. If anything, at the time silicon seemed like a bit of a long shot for IC's.

So, you disagree with what I wrote? Based on a fairly detailed reading of the paper, I'd still think that graphene is a better gamble. Molybdenite has a few advantages, but I'd still stay with graphene even with the bandgap and other issues.

I don't disagree with you - at this time. What I'm saying is that like most technologies, this is another one to watch. There's no reason that processes won't evolve to make the issues that moly transistors currently have become less problematic in the future. Same goes for graphene or quantum dots, among other "exotic" materials.

After all, silicon is king of the hill, and we're just recently really starting to put effort into finding other materials to take its place. Only difference is history proves that it can be done (re: germanium-->silicon).

 

[Martimus]

Its always nice to see what the future MAY hold.

Lonbjerg, please stop trying to derail this thread with a completely pointless and irrelevant argument about something no-one would ever take seriously.

 

[ViRGE]

This thread has been cleansed of off-topic distractions and trolling.
-ViRGE

 

[cbn]

silicon will never be dead because it's cheap and abundant

But can designs built on it be sufficiently modified to help fix "heat density" issue I keep reading about.

Apparently with each die shrink the silicon area decreases by approximately 50%, but the amount of power required only decreases by 25%.

It would seem to me that if that trend continues "heat density" could at some point become a serious problem.

That leaves us the following question:

Which CPU company benefits the most by commercializing alternatives to silicon?

Intel? IBM? Or perhaps another company?

 

[pandemonium]

Molybdenite is a good interim material until Optical circuity takes over mainstream. I'll be waiting for that day to come around. Multiphasic, flexible, easier 3D implementation, and smaller interconnect possibilities. From there it's just a matter of time before we have nano-tech circuitry. ^^

 

[CTho9305]

But can designs built on it be sufficiently modified to help fix "heat density" issue I keep reading about.

Apparently with each die shrink the silicon area decreases by approximately 50%, but the amount of power required only decreases by 25%.

It would seem to me that if that trend continues "heat density" could at some point become a serious problem.

That leaves us the following question:

Which CPU company benefits the most by commercializing alternatives to silicon?

Intel? IBM? Or perhaps another company?

You may find this paper (PDF) interesting.

 

[Jeff7181]

Engaget did an article on synthetic diamond being used as a silicon replacement some years back, too. That prompted me to obtain diamondtransistor.com, however, I since let the registration expire and someone else immediately picked it up. I sorta wish I had kept it.

 

[Idontcare]

Diamond is already blasé, the next big thing will be compressed matter ala neutron star computing Its self-powered too!

 

[Daedalus685]

Diamond is already blasé, the next big thing will be compressed matter ala neutron star computing Its self-powered too!

We call it Computronium in the sci-fi world .

 

[cbn]

You may find this paper (PDF) interesting.

Thanks for posting that, even though it is completely over my head I still like to glance through a technical paper like that when I have the chance.

From the article:

V. SUMMARY AND IMPLICATIONS
The 30-year-long trend in microelectronics has been to increase
both speed and density by scaling of device components
(e.g., CMOS switch). However, this trend will end as
we approach the energy barrier due to limits of heat removal
capacity. For nanoelectronics, this result implies that an increase
in device density will require a sacrifice, due to power
consideration, in operational speed, and vice versa. Thus, it
appears that we are entering a regime where tradeoffs are required
between speed and density, quite in contrast to the
traditional simultaneous benefits in speed and density from
conventional scaling.

Yep, consumers (like me) have been spoiled by both the simultaneous speed and density increase of the past node progressions.

The conclusion of this paper written in 2003 warns of the upcoming compromises that may bring that trend to an end. So far Intel has avoided this with 22nm FinFET. (offering a 50% reduction in die size along with a 50% reduction in power required essentially make the node heat density neutral.)

But where we will go from here? What will happen with our CPUs?

Will Intel be able to maintain or even increase single threaded performance at the 16nm or 10nm nodes with x86? If so, how will they do this? Maybe some "dark silicon" workarounds inside the CPU core to increase single threaded performance while reducing heat density? But if this happens, what will be the cost of the overhead? When does the bubble burst with respect to chasing the classic metrics with an old ISA?

I've really been wondering where this is headed? I've also been wondering how companies like Apple see this upcoming heat density situation?

Should be interesting times (over the next 2-8 years) in the CPU world that is for certain. I can't wait to read the future Anandtech CPU articles as this competition for the single threaded performance metric continues.