SAN FRANCISCO—Claiming to offer the first phase change memory (PCM) in high-volume production, U.S. memory chip vendor Micron Technology Inc. Tuesday (July 17) announced the availability of a 45-nm multi-chip module for mobile devices featuring PCM.
Micron (Boise, Idaho) said its PCM solution for mobile devices features 1-gigabit (Gb) PCM plus 512-megabit (Mb) LPDDR2 in a multi-chip package.
The 1-Gb product is targeted at high-end feature phones, which can benefit from the speed of the read in PCM and simplified software development, according Philippe Berge, senior director of NOR, PCM, e.MMC business for Micron's Wireless Solutions Group. Berge said forthcoming Micron products would target smartphones and other mobile products.
"This is the first product using 45-nm technology, which is really validating high-volume manufacturing of this disruptive technology," Berge said. "Based on that, we are going to have some kind of proliferation in both directions to extend our coverage of the high-volume feature phone market and also later on to address the smartphone market."
PCM claims advantages over traditional memory devices in areas such as boot time, simplified software development, performance and overwrite capability. The technology is also said to be power efficient.
However, the technology has hit a number of obstacles on the road to deployment. detractors argue that it will never be cost competitive with traditional types of memory, and there are also lingering concerns about PCM's temperature sensitivity.
Berge said that improvements
in non-volatile memory technology over the past 20 years have been
evolutionary in nature, but that PCM is a truly disruptive technology.
natural that people need to touch it to really be convinced that it is
manufacturable and that it delivers according to expectations," Berge
said. "We are now at the stage that we can demonstrate that PCM delivers
according to our plans. And that will really be the difference."
Micron said the design-optimizing shared interface between LPDDR2 and PCM is fully compliant with Jedec industry standards.
Micron maintains that its PCM product line has established a foothold in the wireless industry, citing as eivdence relationships with global customers and enablers, engagement with device manufactures and cooperation with Intel Mobile Communications, which includes recent PCM qualification.
"Our commitment to innovation and continued development of advanced products to address the voracious demands of the wireless industry is clear and strong," said Tom Eby, vice president of the Wireless Solutions Group at Micron, in a statement. "We are determined to evolve and innovate by continuing to offer the best-tailored solutions for both today's and tomorrow's market requirements."
Yes Resistion the 1G-bit 45nm PCM-MCP appears to be history. Although the 128Mbit serial and parallel PCM are still posted on the newly designed Micron web site I could not find the 1Gbit under products or MCP. Might have something to do with the fact that the main (most likely the only) significant design win was with Nokia for the Asha phone, for which Micron claimed many millions of PCMs would be shipped, a number later modified downwards. The changes of fortune at Nokia may have something to do with the PCM removal. I suppose it is just possible that the target is now the HMC (Micron's Hybrid Memory Cube).
There is one paper at IEDM 2013 "Intrinsic Retention Statistics in Phase Change Memory (PCM) Arrays (Late News)" from Micron that covers the 45nm PCM . This work studies crystallization statistics in 1 Gb arrays of PCM devices. Suggesting that by tuning the programming conditions they can overcome problems of erratic retention due to crystallization variability. They claim to have developed a new retention model, which is capable of predicting cell to cell and cycle to cycle variability as a function of programming conditions. Perhaps the withdrawn 1Gb product does not have the capability to tune the programming to overcome the problems of "erratic retention"
Micron also suggested that their next target for PCM would be 20nm, if it takes as many iterations to get there, as it did for the 1Gb as well as including all new materials and operating conditions that papers at IEDM 2013 are now suggesting will solve PCM problems-don't hold your breath.
and that leads straight to anomalous Hall effects and rare earth doping.
Does anyone have insight or opinion as to whether Micron buying Elpida will provide Micron better access to the research at Riken?
Maybe EE Times can access the Nature article and offer some elaboration:
M. Nakano, K. Shibuya, D. Okuyama, T. Hatano, S. Ono, M. Kawasaki, Y. Iwasa and Y. Tokura. "Collective bulk carrier delocalization driven by electrostatic surface charge accumulation." Nature, 2012, DOI: 10.1038/nature11296
Here is another unusual development for EE Times to look into. RIKEN of Japan claims a new switching mechanism:
"...Described in a paper in Nature, the device uses an electric-double layer to tune the charge density on the surface of vanadium dioxide (VO2), a well-known classical strongly-correlated material. Thanks to the strong correlation of electrons and electron-lattice coupling in VO2, this surface charge in turn drives localized electrons within the bulk to delocalize, greatly magnifying the change of electronic phase. A potential of only 1 V, they show, is enough to switch the material from an insulator to a metal and trigger an astounding thousand-fold drop in resistance...snip...the research group analyzed the crystal structure of the VO2, showing that it, too, undergoes a transformation, from monoclinic to tetragonal structure..."