Universal Transistor Could Enable Much Smaller Circuits
German researchers claim to have developed the first transistor that can configure itself to either being a n-type or p-type field-effect transistors (FETs). Current CMOS transistors are locked into a static state of being a p- or n-type transistor.
According to Physorg.com, the research "Synthetic nanowires are used to realize the proof-of-principle" was recently published in an issue of the journal Nano Letters and could lead to smaller semiconductors that require far fewer transistors than today's devices. While the research is in its nascent stages, the scientists claim that their dynamic transistor revealed encouraging electrical characteristics, "including a record on/off ratio and reduced leakage current compared to conventional nanowire FETs," Physorg wrote. Future work will focus on improving the transistor's performance.
According to the scientists, the reconfigurable transistor is "enabled by employing an axial nanowire heterostructure (metal/intrinsic-silicon/metal) with independent gating of the Schottky junctions." Other than traditional field-effect transistors, their invention uses "selective and sensitive control" of charge carrier injections at each Schottky junction to monitor "charge carrier polarity and concentration". This feature would allow the FETs to be reconfigured during their operation and execute virtually any Boolean logic computations and provide more design flexibility for hardware engineers.
- Super Talent Intros Storage POD Mini USB 3.0 External SSD
- Asus Introduces 10-Point Multi-Touch AIO PC
- TI Sued Over Patent Infringement in OMAP 3 and 4 Series
- Mozilla Renews Firefox Search Royalty Deal With Google
- Deals Dec 20: Panasonic Ultra-fast Blu-ray Player $65 FS
- Nvidia Announces Carma: Cuda ARM Development Kit
- Oracle Releases Linux 6.2 With Driver Upgrades
- AMD Unleashes 2nd-Gen A-Series APUs for Laptops, Desktops
- Firefox 9.0 Download Released; From 3.6 to 9.0 in 2011
- Acer Developing a $699 15" Ultrabook, Says Supply Chain
- Deals Dec 21: 13.3" Dell V131 Core i3 2.2GHz $583 FS
- AMD CMO Nigel Dessau Leaving Company
- MSI's GT780DX Gaming Notebook Arrives in Time for Xmas
- SW:TOR Hit With Code Errors, Long Server Queues
- Intel Sampling Prototype Medfield Smartphones, Tablets
- New Remotely Exploitable Vulnerability Found in 64-bit Win7
- CES 2012 Will Be Microsoft's Last CES Keynote, Showing
- Mozilla May be Aiming For a Firefox Games Platform





Cool
Beam me up Schottky!
can someone translate this? whats the possible effect on humans life?
geeze, smaller FET = smaller embedded system, lower current leak, such as smaller replacement organs, monitors so on
That's pretty awesome. Means that circuits can re-configure at a hardware level. Multi-use chips for all your needs.
Now, THAT"S something that is worth a patent!
Okay, I studied ASIC design in college. It was actually my specialization, though I'll admit I was never able to land a job in ASIC design. But still, I took those classes through the 500 level, most of the way to a masters and created functioning silicon. I know every word mentioned in there. I'm familiar with every concept mentioned. And yet, it all sounds like gibberish. n-type and p-type are determined by their substrate. n-type, negative charge. p-type, positive charge. The difference between n-type and p-type is if you apply a positive charge to the gate to turn it on, or apply a negative charge to the gate to turn it on. But historically, p-types are extremely large on silicon, and as such, takes up a lot of real estate, so because of that though CMOS is ideal for leakage speed, nobody uses them, because they're too expensive to produce. Everybody strictly uses n-type transistors. How are you changing the transistor type without fundamentally changing the charge of the substrate. I mean, hell, it'd be a big enough of an advancement just to make a p-type that's the same size as an n-type, and to hell with them switching.
can someone translate this? whats the possible effect on humans life?
I'm not 100% sure on this, but the basics are that if this works, any electronic device that relies on transistors could be manufactured cheaper and smaller becuase of not needing to include current switching technology. In other words, future devices like various PC components can be made even smaller and more efficient. How much? That I do not know.
can someone translate this? whats the possible effect on humans life?
Basically, better power use and less lose of power in electric components.
can someone translate this? whats the possible effect on humans life?
The satisfying knowledge of a job well done.
The article tells you what you need to know: "could lead to smaller semiconductors that require far fewer transistors than today's devices"
Be a proactive reader, not passive, no matter how bland the material is.....
Could this lead to much higher levels of storage or will it really only affect efficiency and speed?
Did anyone else read the title of the article as, "Universal Translator Could Enable Much Smaller Circuits."
I was a bit confused, but it makes sense now. lol
ββββββββββββββββββββββββ
βββββββββββββββββββββββ
βββββββββββββββββββββ
ββββββββββββββββββββ
Did anyone else read the title of the article as, "Universal Translator Could Enable Much Smaller Circuits."I was a bit confused, but it makes sense now. lol
I did.
What a Universal Translator? I don't have to learn Klingon now?
What a Universal Translator? I don't have to learn Klingon now?
SoH DIchDaq ghoj tlhIngan joq Hegh!
The satisfying knowledge of a job well done.
I'm not sure why, but I laughed. Thank you sir.
though CMOS is ideal for leakage speed, nobody uses them, because they're too expensive to produce. Everybody strictly uses n-type transistors.
I'm not sure where you studied, but this is false. In modern digital IC's, CMOS is absolutely used. The industry abandoned pure NMOS and pseduo-NMOS back in the early 80's due to power concerns and switching characteristics. In general, yes, you get better drive strength, etc. with NMOS vs. PMOS due to the differences in electron vs. hole mobility, thus leading to the sizing disparity you cited. But that difference is a fact that digital IC designer have lived with for 30+ years. To say that no one uses p-type transistors is crazy.
How are you changing the transistor type without fundamentally changing the charge of the substrate.
You need to read the article closer. The novel advance is not relying on dopant concentrations to determine p-type vs. n-type. Instead, they are relying on a mechansim to modulate the charge carrier injection directly at the Schottky junction. In other words, they have a structure which allows the device to change whether electrons or holes are the majority carriers directly at the junction, thus determing whether it is configured as n-type or p-type.
Did anyone else read the title of the article as, "Universal Translator Could Enable Much Smaller Circuits."I was a bit confused, but it makes sense now. lol
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
Yep, I saw that too...
SoH DIchDaq ghoj tlhIngan joq Hegh!
I just fed that into Google translate - it didn't recognize it.
Google, a Klingon translator would make a great easter egg.
"Future work will focus on improving the transistor's performance."
Why don't they work on massproducing what they have, if it's already superior to existing technology? I keep hearing about supposedly revolutionary things being developed, but nothing ever gets released. If it is too expensive to manufacture such radically different technology, then they should research improved manufacturing processes.
Okay, I studied ASIC design in college.
you studied current technology which is silicone based where band gap is created by doping of adjoint areas. current research is exploring other technologies such as use of nanowire, with goal to eventually move from silicon to carbon.
you may have proper education but you either don't know how to read or your claims are inaccurate and you have no idea what "intrinsic" semiconductor is. read closer - this transistor is still made from silicon but uses nanowire and there is NO DOPING, they use INTRINSIC silicone. no distinct areas of different material are persent. this 'p' and 'n' is NOT p and n, it only ACHIEVES CHARACTERISTICS of p or n...
http://www.sinano.eu/data/document [...] -nnano.pdf
SoH DIchDaq ghoj tlhIngan joq Hegh!
What? My mother was a saint! Get out!
a few questions come to mind, will it be cost effective? will they be able to make it perform as well as current transistors. heat, speed, efficiency?
Kindly note the article title specified "could enable" which has a different meaning from "enables". There was no speculation whether this can actually be used for a circuit or that it is commercially viable. If folks are wild to fund something promising, just let me know. I have some really nice pixie dust and for the right funding I will promise you anything. As a heads up, pixie dust works best with government money. The use of other people's money fosters a broader vision.
It seems like this stuff could get much wilder than just having smaller circuits. Sounds like they could reprogram the logic the gates are using so that they're suddenly doing different computations. Maybe in your Arithmetic Logic Unit you could have 6 general units that can be configured to compute either floating point or integer operations rather than having 3 fixed FPUs and 3 fixed IUs. And graphics cards have some programmable things like shaders or something, but this could take it to another level.
Anyone have any idea how the program gates work? Do they use positive & negative voltages these days rather than positive and ground? Man, I need to restudy the old college books & learn lots of new things!
Klingon translator :-)
http://uta.mrklingon.org/uta
"SoH DIchDaq ghoj tlhIngan joq Hegh!" => "you shall learn Klingon or death"
Happy xmas to ya'll!
But historically, p-types are extremely large on silicon, and as such, takes up a lot of real estate, so because of that though CMOS is ideal for leakage speed, nobody uses them, because they're too expensive to produce. Everybody strictly uses n-type transistors.
Where have you been during the last 30 years? CMOS has been the technology used for microprocessors.
Okay, I studied ASIC design in college.....
no you didn't...
no offense but your post is plain ridiculous and i don't think i've read anything like that in a long time. if that's sort of thing you would say on an interview, no wonder you could not land a job. replies from 'eilersr', 'ik242' and 'aldaia' are right on the money. you can't read, you don't understand article, you are not familiar with concepts and if you really graduated somewhere, there must have been some serious error.
- there is no P and N in technology described in article, they used I type (intrinsic)
- CMOS is not some obscure technology, it is used (a lot... cameras, logic, discrete transistors)
- P-type may be used less than N-type, but to say that 'nobody uses it' is a complete nonsense
... or companies making or selling following must be throwing money out of window;
http://ca.mouser.com/Search/Refine.aspx?Keyword=p-mos
Klingon translator :-)
http://uta.mrklingon.org/uta
Thanks. I needed that!
"replies from 'eilersr', 'ik242' and 'aldaia' are right on the money"
) of the article, but i did find it fascinating. Also thanks to the comments by the people named in the above quote i understand quite a bit more. I honestly felt i was trying to understand something Daniel Jackson was trying to tell me
I must admit, i didn't understand most (if any
"The most beautiful thing we can experience is the mysterious. It is the source of all true art and all science. He to whom this emotion is a stranger, who can no longer pause to wonder and stand rapt in awe, is as good as dead: his eyes are closed." -Albert Einstein