Skip to Content

Flash Memory That’ll Keep On Shrinking

Using atom-thick carbon instead of silicon could pack ever more data into portable electronics.
September 2, 2011

Researchers at the University of California, Los Angeles, and one of the largest manufacturers of computer memory, Samsung, have created a new kind of flash memory that uses graphene—atom-thick sheets of pure carbon—along with silicon to store information.

Incorporating graphene could help extend the viability of flash memory technology for years to come, and allow future portable electronics to store far more data.

Chipmakers pack increasing amounts of data in the same physical area by miniaturizing the memory cells used to store individual bits. Inside today’s flash drives, these cells are nanoscale “floating gate” transistors. Recent years have seen the rapid miniaturization of flash cells, enabling, for example, the iPhone 4 to store twice as much data as the iPhone 3. But below a certain cell size, silicon becomes less stable, and this has the potential to halt the march of miniaturization.

Graphene-based technology like that demonstrated the UCLA team and Samsung could let flash memory continue shrinking. The group’s prototypes devices are described online in the journal ACS Nano.

“We’re not totally replacing silicon but using graphene as the storage layer,” says Augustin Hong, who worked on the devices at UCLA and is now a research staff member at IBM’s Watson Research Center. “We’re using graphene to help extend the capabilities of the conventional technology.”

The graphene flash memory prototypes can be read and written to using less power than conventional flash memory, and they can store data more stably over time, even when miniaturized. The UCLA researchers have also demonstrated that they meet the industry standard of 10-year projected data retention—today’s flash memory does too, but future versions may not. Most important, the graphene memory cells don’t electrically interfere with one another—a problem with conventional flash cells as they are made smaller that can cause them to malfunction.

Other researchers are working on radical new kinds of computer memory that promise to hold more data. However, many of these alternatives require exotic materials and totally new manufacturing processes. Replacing silicon with graphene in flash memory cells could provide a simpler, more practical solution, at least in the short term.

Graphene flash memory cells perform better because of the material’s unusual chemical structure and electrical properties, says Kang Wang, professor of electrical engineering at UCLA, who led the work. Part of the problem with silicon-based flash is that as memory cells get smaller, the transistor gates have to be thicker relative to the rest of the circuit in order to store sufficient charge, and these thick-gated cells tend to interfere with their neighbors. Because gates made from graphene are ultrathin, says Wang, they do not interfere with one another. Graphene can also hold much more charge than silicon without its leaking out—another problem with conventional flash as the cells are miniaturized.

So far, the graphene flash memory cells the researchers have made are relatively large—on the order of ten micrometers. But graphene, unlike silicon, has no known physical properties that would cause a dip in performance as the devices are miniaturized. “Their simulation results suggest that graphene-made devices can be scaled down to about ten nanometers,” says Barbaros Özyilmaz, assistant professor of physics at the National University of Singapore, who was not involved with the research. Conventional flash is expected to become unstable below about 22 nanometers.

Wang says the researchers are now building smaller graphene cells to test. His group collaborated with researchers from Samsung on the project and is talking with Micron about commercialization.

“One question is when to get started with putting graphene on a commercial process line,” says Wang. Semiconductor manufacturing is an extremely well controlled process—defects at the scale of single atoms can turn a high-performance chip into trash—so introducing a new material takes much time and care.

Wang says that in theory, it should not be difficult to add graphene to chips, because the material is relatively stable and can be grown on wafers using processes that are already common in chip manufacturing plants.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.