Intel Granted Patent for Germanium Nanowire Transistors
Intel was granted a patent that covers the use of germanium as a material choice for compound semiconductors that promise faster processors and reduced power consumption.
Submitted as a patent application in April of 2010, the company first discussed a related invention at the 2010 International Electron Devices Meeting. Intel disclosed that it had developed P-channel transistors made from germanium, which the company said could be combined with complementary III-V N-channel transistors to form a suitable CMOS architecture. The focus on germanium is largely due to the fact that it is more mobile than silicon.
The patent itself reveals the use of a "germanium nanowire channel and the SiGe anchoring regions [that] are formed simultaneously through preferential Si oxidation of epitaxial Silicon Germanium epi layer." Intel leverages a silicon fin as a "template" to align germanium nanowires on a chip while silicon-germanium anchors are used to mount to a silicon substrate. Germanium is likely to become a much more critical material in chip manufacturing in the future, as such nanowires "provide better control of short channel effects such as sub threshold slop and drain induced barrier lowering," the patent states.
The extensive use of germanium has been discussed by the semiconductor industry for more than a decade, while first sophisticated germanium processors were predicted 15 years ago to arrive in the 2007 - 2008 time frame (similarly, the first graphene processors are now forecast to become available around 2020 - 2025). The history even goes back to 1959, when Jack Kilby built a microchip with germanium, but it was Intel-co-founder Robert Noyce who chose silicon and overcame. Back then, germanium was found to be limited by greater current leakage. Over the past ten years, the interest in germanium as a transistor material has increased significantly and has resulted in developments such as IBM's 500 GHz GPU.
- Intel,
- sige ,
- transistor ,
- germanium ,
- processor
- Send Your Kid for Free Coding Classes at the CoderDojo
- Hard Drive Prices to Remain Inflated Throughout 2012
- Epic Revealing Unreal Engine 4 Later This Year
- Download All the Torrents on The Pirate Bay in Only 90MB
- Google Paid $410,000 for Chrome Bugs
- Deals Feb 11: HP dm4t Beats Core i5 + 1TB HDD Laptop $749
- Blizzard and Valve at War Over DOTA Name
- Apple Now Worth More than Google and Microsoft Combined
- Gabe Newell Gives Update on Steam Hack of 2011
- Europeans Take to the Streets to Protest ACTA
- Origin PC Now Also Selling Intel X79-based Gaming Laptop
- Deals Feb 13: Dell 21.5" 1080p HDMI LCD Monitor $139.99
- Epic's "Samaritan" UE3 Demo Required 2.5 terraFLOPS
- Nvidia Kepler GK104: More Leaks, Rumors on Specifications
- There Will Be a Firefox Made Specifically for Windows 8 Metro
- Blizzard OKs WoW and StarCraft-based Board Games
- Internet Troll Gets Tracked Down, Confronted in Real Life
- Intel Haswell iGPU to support DirectX 11.1, OpenGL 3.2
At least there is some progress.
Now's the time to buy shares in germanium mining companies.
Hang on, wind that back...
500GHz CPU?
WANT ONE!!!
Hang on, wind that back...500GHz CPU?WANT ONE!!!
Germanium .... From Germany? Discovered by a German?
Iirc, germanium was pretty much abandoned in analog circuits because it was much more prone to "thermal runaway" than silicon. Has this issue been solved, or is it simply not relevant? What are the thermal limits on these germanium circuits?
Im all up for progress and anything that makes Ivy Bridge's tri-gate transistors better, but how were they granted a patent for something they didnt create and werent the first to use commercially?
Iirc, germanium was pretty much abandoned in analog circuits because it was much more prone to "thermal runaway" than silicon. Has this issue been solved, or is it simply not relevant? What are the thermal limits on these germanium circuits?
in nanotech, the length of the bridge in transistors is very small. So small that the currents flow in a straight line from source to drain as opposed to micro wires where the currents ricocheted towards the drains.
Progress?? How can anyone mention progress when a private company is granted patent on progress?
IBM's 500 Ghz GPU? It's over. AMD and NVIDIA are finished.
Im all up for progress and anything that makes Ivy Bridge's tri-gate transistors better, but how were they granted a patent for something they didnt create and werent the first to use commercially?
The germanium transistor was invented in 1948, but their invention is a germanium transistor in combination such that it is a suitable CMOS structure.
And that link is already 6 years old... dates back to 2006.... Man, I wasn't even a member then...
Epic reading fail
Duley noted, however...
STILL WANT ONE!!!!!!!!!!!!!
Given the lower threshold I knew we were going back to the future at some point.
Germanium .... From Germany? Discovered by a German?
Yes, it was discovered by a German chemist, in a mine in Germany.
http://en.wikipedia.org/wiki/Clemens_Winkler
It's not the only one like that either. Polonium, Francium, Scandium, Californium, Europium, Americium...
I don't know what you said, but you said it!
Yes, it was discovered by a German chemist http://en.wikipedia.org/wiki/Clemens_Winkler
By any chance, the Fonz's dad?
Iirc, germanium was pretty much abandoned in analog circuits because it was much more prone to "thermal runaway" than silicon. Has this issue been solved, or is it simply not relevant? What are the thermal limits on these germanium circuits?
i noticed my explanation was a little unfinished. what I tried to say is that in microtransistors(and similar sizes) the current ricochets getting from the source to the drain, generating lots of heat on the way and losing part of the energy with which it started. In Nanotransistors the distance between the drain and source is so small that the current travels in a straight line at the bridge very fast and not generating any heat(heat is generated at the source and drain connections to the bridge for the resistive requirements of the transistor) in the bridge and no loss of energy on the way. So the problem with germanium losing too much energy is no longer visible in todays nanotech.
I feel like I spoke chinese, but i still hope you understood.
Germanium .... From Germany? Discovered by a German?
Yes, by a German chemist.
http://en.wikipedia.org/wiki/Clemens_Winkler
Sorry... should have read thru the replies first before duplicating an answer already given in even more detail.
I understood, but not how it wouldn't apply; you explained away much of the source of the heat, but not the effect that heat has [on germanium]. Would it not matter though, in circuits that are either on or off, as long as they are actually still able to turn off?
I understood, but not how it wouldn't apply; you explained away much of the source of the heat, but not the effect that heat has [on germanium]. Would it not matter though, in circuits that are either on or off, as long as they are actually still able to turn off?
i watched this vid a few days ago and its pretty easy to understand if you know basic terminology of transistors:
http://www.youtube.com/watch?v=tW1-fSRiAdc
basically since the current travels in a straight line there is no heat on the wire(bridge), so the germanium element is generally unaffected by the heat.
Unless someone finds a new resource of Germanium, we will run out of it in a decade or so, because it is already used in the production of LCDs.I also think that the chips they want to produce are aimed at low power portable devices for further lowering the power consumption.
Ge was used in high-frequency electronics for decades. The innovation should be something more specific, although with the quality of USPTO work, you never know if even there was any real innovation.
Unfortunately, there is probably a patent troll out there that has a patent entitled "using a better material in a semiconductor to improve speed and efficiency".
Way ahead of the competition after this? Intel for the rest of my life for sure.
Imagine if a patent troll got hold of this patent, sat on it for a few years, and proceed to suing the living crap out of Intel, Samsung, and every other microchip die manufacturers.
I hope Intel doesn't use it to keep its competitors down.
Imagine if a patent troll got hold of this patent, sat on it for a few years, and proceed to suing the living crap out of Intel, Samsung, and every other microchip die manufacturers.I hope Intel doesn't use it to keep its competitors down.
Every patent Intel holds is up for licensing....how do you think TSMC operates? They license every manufacturing tech that Intel and IBM manage to perfect. Intel is in it for the money....and there's a shitload of money in patent licensing.
i noticed my explanation was a little unfinished. what I tried to say is that in microtransistors(and similar sizes) the current ricochets getting from the source to the drain, generating lots of heat on the way and losing part of the energy with which it started. In Nanotransistors the distance between the drain and source is so small that the current travels in a straight line at the bridge very fast and not generating any heat(heat is generated at the source and drain connections to the bridge for the resistive requirements of the transistor) in the bridge and no loss of energy on the way. So the problem with germanium losing too much energy is no longer visible in todays nanotech.I feel like I spoke chinese, but i still hope you understood.
its a bit hard to wrap your head around, but i get it.
Anyone else notice that the picture says SrGe? Last I checked the periodic table that would be Strontium Germanium, not Silicon Germanium. So...
The most important point of this article is that a patent was granted to a real invention.