Improving flash memory: New molecular storage devices could bridge memory gap

November 20, 2014
An SEM image of the 5-nm Si nanowire channel with side control gate. Credit: Laia Vila Nadal, Felix Iglesias Escudero, Leroy Cronin, Cronin Group, School of Chemistry, University of Glasgow

(Phys.org) —New molecules could be the key to solving a looming problem with flash memory storage, researchers say in a new report published in the journal Nature today.

Flash memory is a popular form of commonly used in devices such as smartphones, cameras and memory sticks. However, there is a physical limit to the minimum size of the current design of data cells, which currently use metal-oxide-semiconductor (MOS) components. These are difficult to manufacture at a scale below 10 nanometers, effectively setting an upper limit on the amount of storage it is physically possible to fit onto conventional silicon chips

Scientists have previously suggested that using individual molecules to replace conventional data-storage components in flash memory could help create small devices capable of holding large amounts of data. However, attempts to design these molecules have faced significant practical barriers such as low thermal stability and high resistance, which has limited their use in pre-existing technologies.

A team from the University of Glasgow's Schools of Chemistry and Engineering and Rovira i Virgili University in Spain have developed a possible solution to these problems using metal-oxide clusters known as polyoxometalates (POMs).

Professor Lee Cronin, Regius Professor of Chemistry at the University of Glasgow, led the research team. He said: "Conventional flash memory uses transistors whose design allows them to 'remember' whether they have been turned on or off after they've been removed from a power source. Those transistors' positions correspond to binary, allowing data to be stored.

Nanoscale polyoxometalate clusters. Credit: Laia Vila Nadal, Felix Iglesias Escudero, Leroy Cronin, Cronin Group, School of Chemistry, University of Glasgow

"We've been able to design, synthesise and characterize POM molecules that can trap charge and act as flash ram, as well as dope of the inner core of the clusters with selenium to create a new type of memory we call 'write-once-erase'.

"The POM clusters provide a balance of structural stability and electronic activity and their electronic functionality is tunable, making them suitable as storage nodes for flash memory.

Nanoscale polyoxometalate clusters. Credit: Laia Vila Nadal, Felix Iglesias Escudero, Leroy Cronin, Cronin Group, School of Chemistry, University of Glasgow

"One major benefit of the POMs we've created is that it's possible to fabricate them with devices which are already widely-used in industry, so they can be adopted as new forms of without requiring production lines to be expensively overhauled."

The research team's paper, titled 'Design and fabrication of based on nanoscale polyoxometalate clusters', is published in Nature.

Explore further: Samsung starts mass production of industry's first 3-bit 3D V-NAND flash memory

More information: Design and fabrication of memory devices based on nanoscale polyoxometalate clusters, dx.doi.org/10.1038/nature13951

Related Stories

Advancing resistive memory to improve portable electronics

August 14, 2013

A team at the University of California, Riverside Bourns College of Engineering has developed a novel way to build what many see as the next generation memory storage devices for portable electronic devices including smart ...

The '50-50' chip: Memory device of the future?

September 13, 2013

A new, environmentally-friendly electronic alloy consisting of 50 aluminum atoms bound to 50 atoms of antimony may be promising for building next-generation "phase-change" memory devices, which may be the data-storage technology ...

Recommended for you

Probe for nanofibers has atom-scale sensitivity

January 20, 2017

Optical fibers are the backbone of modern communications, shuttling information from A to B through thin glass filaments as pulses of light. They are used extensively in telecommunications, allowing information to travel ...

Magnetic recording with light and no heat on garnet

January 19, 2017

A strong, short light pulse can record data on a magnetic layer of yttrium iron garnet doped with Co-ions. This was discovered by researchers from Radboud University in the Netherlands and Bialystok University in Poland. ...

7 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

alfie_null
4 / 5 (2) Nov 20, 2014
Yet another press release from a university, numbers expurgated. How much more dense? How fast? How reliable?
EyeNStein
5 / 5 (1) Nov 20, 2014
Great concept: Designer molecules can hold movable charges or change shape reversibly with electrical or optical write and read systems.
This could have a great future in memory or even rechargeable battery technology.
But this article shows how extremely 'early days' this research is at present. It certainly wont replace flash ram before more mature technologies (like phase change memory or memristors) do.
Eikka
3 / 5 (2) Nov 20, 2014
Yet another press release from a university, numbers expurgated. How much more dense? How fast? How reliable?


Well, you can go down to about 1 nanometer which is in the scale of individual molecules. A molecule can have inter-atom distances in the 100 picometers range, but since a molecule would have more than one atom, a practical lower bound of about 1 nm should be reasonable.

Of course this doesn't mean you get 10 times improvement, because while the individual memory elements get smaller, the wires leading up to them don't. They still have to be manufactured with traditional litography which has a limit in about 10-5 nm feature size.

If you look at a decapped flash memory chip on google, most of the surface area of the chip is filled by the interconnecting wires.
gkam
2.6 / 5 (5) Nov 20, 2014
We had flash memory in 1972, . . sort of. Our Teradyne J-283's, J-259's and others had ferrite core memories, probably 16k Bytes of 18 bits words in non-volatile memory. You could shut it off and turn it on and continue right where you left off. If you had the time for it to work.
gkam
2.6 / 5 (5) Nov 20, 2014
I'll bet the goobers who voted me down never worked with an 18-bit word.

The PDP-8s had volatile memory, but used punched paper tape, and a 16-bit word. We used the Teradynes to run all the tests for the IC wafer sorters and device handlers at the same time.
Eikka
4 / 5 (4) Nov 21, 2014
We had flash memory in 1972, . . sort of.


No you didn't, because the technology wasn't invented until 1980 and published in 1984.

Ferrite core memory is NOTHING like flash memory, which is based on the quantum tunneling of electric charge across the isolated gate of a MOS transistor to jam it open or shut. It's called flash memory because you need to zap it with high voltage to overcome the energy barrier to reset it, and the circuit to do that resembles the flash charging circuit of a camera.

I'll bet the goobers who voted me down never worked with an 18-bit word.


What's that got to do with anything? Why is that an achievement?

That's like saying "I bet you never worked with a seven foot tape measure". Well no, because I have a 12 foot tape measure.
Uncle Ira
3.7 / 5 (3) Nov 21, 2014
I'll bet the goobers who voted me down never worked with an 18-bit word.


I'll be you are right about that Cher. I haven't like you haven't either. What does that have to do with you making 60 or 50 postums a day about how you done more in life than Returnering-Skippy has done?

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.