Beyond silicon: Transistors without semiconductors

Point out news stories, on the net or in mainstream media, related to polywell fusion.

Moderators: tonybarry, MSimon

Post Reply
Netmaker
Posts: 78
Joined: Sat Sep 11, 2010 8:17 pm

Beyond silicon: Transistors without semiconductors

Post by Netmaker »

BNNT + Gold QD = no heat/leakage, quantum tunneling of the electrons, each electron is confined to a single QD and it works at room temperature.

No leakage = the Ghz race back on, there will be less disparity between desktops/laptops/tablets/phones.

Main memory (DRAM) can be replaced with much faster SRAM. Will this also allow for non-volatile memory?

Can this be any use for the display portion for a monitor/tv/screen?

http://www.sciencedaily.com/releases/20 ... 121015.htm

Betruger
Posts: 2321
Joined: Tue May 06, 2008 11:54 am

Re: Beyond silicon: Transistors without semiconductors

Post by Betruger »

Is this also obviously a technology that simply requires a decade and mountains of $$$ to develop into mass market form? Or are there still enough unknowns for a showstopper somewhere between Yap & co's one-offs and desktop use?
You can do anything you want with laws except make Americans obey them. | What I want to do is to look up S. . . . I call him the Schadenfreudean Man.

JoeP
Posts: 523
Joined: Sat Jun 25, 2011 5:10 am

Re: Beyond silicon: Transistors without semiconductors

Post by JoeP »

Interesting development.
I wonder how stable the position the quantum dots are since they are only sitting on top of the insulator and do not appear to be anchored?

If there is very little energy lost to heat then a CPU based on this technology would run many times longer using currently available batteries.

Robthebob
Posts: 383
Joined: Mon Jun 23, 2008 11:12 pm
Location: Auburn, Alabama

Re: Beyond silicon: Transistors without semiconductors

Post by Robthebob »

not sure if yall heard, but there's this new technology which is the memristor.

Properties of memristor
-fundamental circuit element (like resister, capacitor, inductor)
-in fact http://en.wikipedia.org/wiki/File:Memristor.png
-thus the device can and should be smaller than transistors, which are combinations of fundamental circuit elements
-semi-non-volatile memory stage
-as in it can be volatile (it can "forget")
-or it can be non-volatile, depends on the design and what kind of desired applications
-based completely on physical electrochemical processes (the one I'm helping to work on literally builds a bridge between the gaps of the device)
-it's variable memory in the form of electrical resistance of the device
-it can be just 0 or 1
-or it can be between 0 and 1, depends on what kind of applications you want to use it
-potential massive applications in memory storage (obviously), logic, circuit design, etc
Throwing my life away for this whole Fusion mess.

Netmaker
Posts: 78
Joined: Sat Sep 11, 2010 8:17 pm

Re: Beyond silicon: Transistors without semiconductors

Post by Netmaker »

Robthebob wrote:not sure if yall heard, but there's this new technology which is the memristor.

Properties of memristor
-fundamental circuit element (like resister, capacitor, inductor)
-in fact http://en.wikipedia.org/wiki/File:Memristor.png
-thus the device can and should be smaller than transistors, which are combinations of fundamental circuit elements
-semi-non-volatile memory stage
-as in it can be volatile (it can "forget")
-or it can be non-volatile, depends on the design and what kind of desired applications
-based completely on physical electrochemical processes (the one I'm helping to work on literally builds a bridge between the gaps of the device)
-it's variable memory in the form of electrical resistance of the device
-it can be just 0 or 1
-or it can be between 0 and 1, depends on what kind of applications you want to use it
-potential massive applications in memory storage (obviously), logic, circuit design, etc

And what is your point :) ?

I am still waiting for HP and Hynix to come to up with were their flash killer memory. Maybe next year.

Or possibly be greet our new slime mold/memristor overlords:
http://www.extremetech.com/extreme/1589 ... lime-molds



Some additional background about BNNT-QD and Dr. Yap:

B-C-N Nanotubes and Related Nanostructures: 6 (Lecture Notes in Nanoscale Science and Technology)
http://www.amazon.com/B-C-N-Nanotubes-R ... 808&sr=1-1

Yap: Harnessing the Divas of the Nanoworld
http://www.mtu.edu/news/stories/2010/ja ... 21781.html

Room-Temperature Tunneling Behavior of Boron Nitride Nanotubes Functionalized with Gold Quantum Dots
http://onlinelibrary.wiley.com/doi/10.1 ... 9/abstract

Quantum-Tunneling Electrons Could Make Semiconductors Obsolete
http://slashdot.org/topic/datacenter/qu ... -obsolete/

Multiwalled Boron Nitride Nanotubes: Growth, Properties, and Application
http://www.buchhandel.de/WebApi1/GetMmo ... moType=PDF

Michigan Technological University - Search of MTU for Yap
http://www.mtu.edu/search/?cx=009927719 ... &gsc.q=yap

Yap's Research Lab publications
http://www.phy.mtu.edu/yap/publications.html

Robthebob
Posts: 383
Joined: Mon Jun 23, 2008 11:12 pm
Location: Auburn, Alabama

Re: Beyond silicon: Transistors without semiconductors

Post by Robthebob »

The benefits you listed for this new device arent as good as the benefits of memristors in many applications.

I dont know about heat or current leakage with respect to memristors, but I'm guessing they dont have those problems, but I could be wrong.

Memristor just trumps everything in terms of memory (memory in circuits, in large memory storage, etc). If you were excited about this innovation, I thought you would be really excited about memristors; just saying.

HP invested in the wrong technology. The one developed and are still being studied on by Arizona University, AFRL, etc are the real deal. HP's device isnt good. I'm doing some nice research simulations on effects of oxygen on performance and whatnot, it's looking really good.

PS: thanks for the article, will show this to my boss.
Throwing my life away for this whole Fusion mess.

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Re: Beyond silicon: Transistors without semiconductors

Post by happyjack27 »

difference between SRAM and SDRAM:
SRAM is essentially a pair of logic gates coupled in a feedback loop. (specifically, a pair of d-flops http://en.wikipedia.org/wiki/Flip-flop_(electronics) ) this makes it very fast. hence, it's used directly in the cpu.
DRAM memory is stored in rows of capacitors. this makes it very dense. however, it is slower to read and write, must be refreshed periodically (as the charge in the capacitors diminish), and can only be retrieved a row at a time.

so main benefit of SRAM is speed, main benefit of DRAM is density. (there's also a theoretical TRAM under development, which presumably would have benefits of both...)

memristors trump DRAM in density, and potentially SRAM in speed. they are theoretically not simply the best of both worlds - but better than both worlds.

so yeah, memristors. bam.

Robthebob
Posts: 383
Joined: Mon Jun 23, 2008 11:12 pm
Location: Auburn, Alabama

Re: Beyond silicon: Transistors without semiconductors

Post by Robthebob »

yeah. I think HP's device reads at femtosecond while the device we're working on works at nanosecond.

There's issue with the advance part of memristor. The basics, unless there's some hard physics that prevent us from using memristors effectively (im doing research on that right now, about what Oxygen is gonna do to this particular configuration), is pretty much settled.

We can do 0/1 memory storage just fine, we can make these things really really small just fine, the switch speed is really really fast, also the read speed is fast too, as mentioned. The kicker is variable memory (doing memory storage in between 0 and 1), that's been somewhat difficult, because of issues...

1. When you turn on a virgin device, the bridge you build in it (the metal ion from the top electrode flows from top to bottom due to the applied electric field, when it deposits on the bottom electrode, it starts to form a bridge) "digs" a tunnel through the electrolyte, this affects all sequential bridge building in the future...

2. It's pretty difficult to control how this first bridge is built, and if we dont have a way of standardizing the process (as in you apply x voltage for x amount of time, you get 0.5 as the memory, not 0.8, not 0.2, or any other number, but 0.5, then we have a winner) then the variable memory and all related applications will be impossible to achieve.

3. Right now I think we have it down to 2 orders of mag, which according to my boss is not too far from being acceptable. We'll see, if it's impossible to do variable memory, it may be impossible to have AI robots take over the world, which is a good thing tho...
Throwing my life away for this whole Fusion mess.

DeltaV
Posts: 2245
Joined: Mon Oct 12, 2009 5:05 am

Re: Beyond silicon: Transistors without semiconductors

Post by DeltaV »

EDIT - [Oops, link already posted. I'll leave the pic up anyway, it's cool.]

As Robthebob pointed out, memristors fill in the missing corner of the passive, two-terminal, electrical components map:

Image

Post Reply