backtop


Print 65 comment(s) - last by flipsu5.. on Dec 23 at 10:21 AM

Researchers have devised a way to build memory from graphene

Storage in today's computers is based on rotating magnetic platters or flash memory. Both of these mediums work well, provide large amounts of storage and have been around for a while now. Some scientists don’t believe that flash is the future for storage in computers and electronic devices. In fact, the future of storage could be based on something kids use every day at school -- pencil lead or graphite.

A team of researchers at Rice University has found a method of creating a new type of memory from a strip of graphite only 10 atoms thick. Graphite is the basic element in the new type of memory. The scientists describe in a paper published in the online journal Nature Materials a storage device that utilizes the conducting properties of graphene. A large clump of graphene is better known as graphite, something school kids doodle with everyday.

Rice professor James Tour says that graphene memory would increase the amount of storage in a two-dimensional array by about five times. He says that this massive improvement is due to the individual bits being able to be made smaller than 10 nanometers. By comparison, circuitry in your average flash memory chip today is 45nm. Another big benefit of graphene memory is that switches can be controlled by two terminals rather than the three terminals used in flash memory today.

The two-terminal capability is important because it provides the capability to make three-dimensional memory practical since the very thin graphene arrays can be stacked, multiplying the storage capacity of the array with each graphene layer.

Storage arrays using graphene will be mechanical devices at their core and as such, the chips will consume very little power. Much lost power in flash storage comes from leakage; graphene memory will need little power leading to less leakage while keeping data intact. Graphene memory has a massive improvement in on-off power ratio compared to current memory technologies.

Tour said in a statement, "It’s (power savings) huge — a million-to-one. Phase change memory, the other thing the industry is considering, runs at 10-to-1. That means the ‘off’ state holds, say, one-tenth the amount of electrical current than the ‘on’ state."

Tour explains that current tends to leak from an off that is holding a charge. He says, "That means in a 10-by-10 grid, 10 ‘offs’ would leak enough to look like they were ‘on.’ With our method, it would take a million ‘offs’ in a line to look like ‘on'. So this is big. It allows us to make a much larger array.”

Another benefit of graphene as a storage medium is that while it puts off little heat, it is able to operate in a very wide temperature range. The researchers have tested the system to minus 75 to over 200 degrees Celsius.

Performance of graphene-based systems is impressive, the researchers say that the new switches are faster than the lab's testing equipment can measure and they promise long life as well. "We’ve tested it in the lab 20,000 times with no degradation,” said Tour. “Its lifetime is going to be huge, much better than flash memory."

The processes uses graphene deposited on silicon via chemical vapor deposition making for easy construction that can be done in commercial volumes with methods already available says Tour.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

graphene nanocrystals
By flipsu5 on 12/19/2008 7:12:53 PM , Rating: 2
The cited paper abstract reads:

quote:
Transistors are the basis for electronic switching and memory devices as they exhibit extreme reliabilities with on/off ratios of 10^4–10^5, and billions of these three-terminal devices can be fabricated on single planar substrates. On the other hand, two-terminal devices coupled with a nonlinear current–voltage response can be considered as alternatives provided they have large and reliable on/off ratios and that they can be fabricated on a large scale using conventional or easily accessible methods. Here, we report that two-terminal devices consisting of discontinuous 5–10 nm thin films of graphitic sheets grown by chemical vapour deposition on either nanowires or atop planar silicon oxide exhibit enormous and sharp room-temperature bistable current–voltage behaviour possessing stable, rewritable, non-volatile and non-destructive read memories with on/off ratios of up to 10^7 and switching times of up to 1 microsecond (tested limit). A nanoelectromechanical mechanism is proposed for the unusually pronounced switching behaviour in the devices.


It looks like the graphene is in the form of nanocrystals (discontinuous films). How reliable is that? And 1 microsecond is a bit long for switching time. Maybe it is slower because it is mechanical. Most alternatives are faster than flash, phase change memory is on the order of 100 ns.

Lastly can the sense amplifier go to such high ratios (10^7); it looks like a signal-to-noise challenge.




RE: graphene nanocrystals
By Fritzr on 12/19/2008 9:05:14 PM , Rating: 2
The stated limitation on their testing equipment is that it cannot measure a switching time less than 1 microsecond.

They state upfront that their equipment is unable to distinguish between 10ns, 100ns or 600ns parts ... all are faster than what they can measure. Their part is also faster than what they can measure, so is no slower than 1000ns though probably faster.

Before they needed switching circuits to handle todays transistor speeds they didn't design them. Since there is now a faster switching part, one of the support circuits designed before release from R&D will be switching circuits that can either handle the max speed or will be one of the limiting parts on the final product. If the new switching circuit is slower than theoretical max then you can guarantee ongoing R&D to build one that is faster.

Pure graphene is one atom thick sheet. They state here that it will be built as a surface deposit of up to 10 layers on nanowire or SO substrate ... so strength will depend on the substrate and thermal expansion differences ... in other words it's an engineering design problem.


RE: graphene nanocrystals
By flipsu5 on 12/22/2008 8:32:36 AM , Rating: 2
Thanks for the clarification. I am a bit surprised they can publish without sub-microsecond data. Many journals don't cut that kind of slack. We have to reserve judgment on its speed then. But designers have told me the ratio of 10 million is excessive.


RE: graphene nanocrystals
By bridgeman on 12/22/2008 1:34:28 PM , Rating: 2
quote:
Lastly can the sense amplifier go to such high ratios (10^7); it looks like a signal-to-noise challenge.


The on-off ratio is the "signal-to-noise ratio" here. (I put that in quotes because it isn't really SNR, just something that looks similar from a design standpoint.) A higher on-off ratio means the sense amp spec is relaxed, not tightened.


RE: graphene nanocrystals
By flipsu5 on 12/23/2008 10:21:01 AM , Rating: 2
What is the read current of the selected OFF cell vs. the leakage for an unselected ON cell? If the former cannot dominate the latter, there is no OFF signal that is read above the din of the arbitrarily many unselected ON cells.


"We are going to continue to work with them to make sure they understand the reality of the Internet.  A lot of these people don't have Ph.Ds, and they don't have a degree in computer science." -- RIM co-CEO Michael Lazaridis

Related Articles













botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki