Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > Memory/Storage

Intel exec considers what comes after flash

Posted: 12 Mar 2004 ?? ?Print Version ?Bookmark and Share

Keywords:flash memory? eprom? etox? nand? intel?

Flash memory, in its EPROM tunnel oxide (ETOX) and NAND variants, still has more than five years as the key mainstream nonvolatile memory, but the race is on to find the memory technology that can scale to sub-45nm manufacturing processes, according to Stefan Lai, VP of the technology and manufacturing group at Intel Corp.

Lai recently presented to a nanotechnology forum his views on why phase-change memory and ferroelectric polymer memory devices have a chance to become established as next-generation nonvolatile memories while magnetic and ferroelectric RAMs may be less likely.

But Lai also indicated that what he calls seek-and-scan memory, based on atomic resolution probes, could ultimately outperform all lithographically defined memory including polymer and phase-change versions.

Lai joined Intel in 1982 he co-invented the EPROM tunnel oxide (ETOX) flash memory cell and has managed almost every generation of flash memory development at Intel. Lai said there was good visibility on how to get flash memory to 65nm and 45nm manufacturing process technologies.

Lai said Intel was now into its eighth generation of ETOX flash memory and has succeeded in achieving a 50 percent cell size reduction per generation. But he also acknowledged that NOR-style flash memory could not compete with NAND on cell area. "Two-bit per cell NAND will be the lowest cost NV memory for the rest of this decade," Lai said in his presentation.

Lai then turned to novel memory technologies including the chalcogenide material phase-change memory and the polymer ferroelectric memory, two approaches in which Intel has an interest.

Energy Conversion Devices Inc., founded and led by Stanford Ovshinsky, has researched phase-change material for memory and storage applications for more than 30 years.

Intel's interest dates back as far 1970 when company co-founder Gordon Moore co-authored a paper on the subject. Thirty years later, Intel renewed its relationship with Energy Conversion Devices Inc. through the latter's Ovonyx Inc. joint venture subsidiary. At the same time, Intel is reported to be pursuing polymer ferroelectric memory through a research contract with Norwegian company Opticom ASA and its subsidiary Thin Film Electronics AB.

In his presentation, Lai spoke against conventional perovskites material-based ferroelectric memory and against magnetic RAM. Although magnetic RAM has close to SRAM speed, like ferroelectric memory and flash memory, it suffers from needing transistors for storage.

Transistors are big compared with diodes Lai pointed out.

"Memories requiring a transistor switch will be limited in scaling," said Lai, pointing to MRAM and FRAM. "Diode switch scales better."

- Peter Clarke

Silicon Strategies

Article Comments - Intel exec considers what comes afte...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top