Nano memory gets big chip backer
Major semiconductor maker signs deal to use nanotube-based memory technology from Nantero, just weeks after LSI Logic licensed it.
Nantero CEO Greg Schmergel would not identify the licensee but said on Thursday that an announcement would come in a couple of weeks. Other deals are in progress, he said. In June, LSI Logic said it had licensed the memory technology. Ideally, LSI will incorporate Nantero's technology in products by the end of next year, Schmergel said.
"It (the new licensee) is a very large company. It is a company with a fab," or semiconductor fabrication facility, he said. "There are several other ones in process, but some of these deals won't be announced."
Nantero has come up with a way to make transistors, the on-off switches inside chips, with carbon nanotubes, which are hollow tubes of carbon atoms that exhibit a number of remarkable properties.
The Massachusetts company's technology exploits two of these properties: the flexible/bendable nature of nanotubes and the strong attraction carbon atoms exhibit with each other through van der Waals forces.
In Nantero's memory design, ribbons of carbon nanotubes are suspended over a carbon substrate. In the "off" state, the ribbon of nanotubes does not touch the substrate, and electricity does not flow between the interconnect suspending the ribbon. In the "on" state, the nanotube bends downward and then adheres to the substrate through van der Waals forces. Electricity flows, and the memory cell registers as a "1," in data terms.
The on-off states can be switched back and forth by applying different charges to the ribbon.
Nantero's memory is faster than static random access memory (SRAM), an embedded form of memory used for caches on processors, Schmergel said. LSI will use Nantero's memory as an SRAM replacement, as will the unnamed licensee. Samsung, STMicroelectronics, Hynix and Cypress Semiconductor, among others, manufacture SRAM, said Betsy Van Hees, an analyst at iSuppli.
The nanotube-based memory is also nonvolatile, which gives it an advantage over conventional computer memory, or DRAM, and is faster than many forms of flash memory, Schmergel said.
Nantero's success in licensing its technology lies in a radical shift in architecture that was implemented in the past year. Initially, Nantero proposed making memory cells in which a single nanotube would connect or detach from a perpendicular nanotube below it.
While this would allow for incredibly dense memory chips, most analysts and scientists doubted that the company could come up with a way to erect millions of uniform, microscopic crossbars on a sliver of silicon measuring less than a few square centimeters. Moreover, the nanotubes would have to be nearly identical--a goal that has eluded researchers--and all face the same way.
In the new architecture of the company's chips, a layer of nanotubes is spread on to a substrate. Engineers then use conventional lithography to "draw" electrical contacts that are connected to each other by the thick ribbons of nanotube material and the substrate. It does not matter what direction they face or whether they are identical. The overriding factor is whether the ribbons exhibit the promised mechanical properties.
Nantero's chief scientific officer, Thomas Rueckes, came up with the idea for the ribbon a few years ago, but the patents only recently came though, Schmergel said.
In the end, though, chips based on Nantero's design are fairly large, rely on the expensive semiconductor manufacturing processes that nanotechnology promoters hope to one day eliminate and will hold about the same amount of data as regular memory.
"We won't be achieving thousandfold density improvement just yet," Schmergel said.