That was the conclusion of a day-long conference sponsored by market watcher Semico Research Corp. here Thursday (Nov. 4). "We think nanotech represents the next phase of Moore's Law, and it is creating an interdisciplinary renaissance," said Steve Jurvetson, managing director at Draper, Fisher Jurvetson (Menlo Park, Calif.). Since 2000, the venture capital firm has invested about $80 million in some 20 companies broadly in the field.

"It is hard to think of an industry that won't be revolutionized by nanotech in the next 50-100 years, including everything from agriculture on out," he said in a keynote here.

Jurvetson defined nanotech as anything working at 100 nm and below in the area where life sciences, materials science and electronics come together. "This is at the nexus of the sciences," he added.

Nanotech startups range from developers of basic materials and tools to suppliers of novel thermal management technologies and optoelectronics components. Several players are queuing up to deliver novel nanotech memories by 2007.

"In electronics, memory is the first place nanotech plays out in a big way," said Jurvetson.

LSI Logic is developing an embedded memory for its ASICs using the carbon nanotube based technology from startup Nantero Inc. (Woburn, Mass). LSI believes it could use the technology to embed more than 30-Mbits fast memory on an integrated cellphone processor, the current embedded memory limit due to power leakage and yield problems with conventional memories, said Norm Armour, vice president of LSI's fab operations in Gresham, Ore.

"It's a couple years out bit it looks very promising. The promises of single-wall nanotubes are almost too good to be true," said Armour.

Bob Merritt, Semico's vice president of emerging technologies, said by 2007 several companies will deliver nanotech-based nonvolatile memories with SRAM-like speeds and DRAM-like densities. The parts — especially when combined on-chip with high performance logic — could offer dramatic new capabilities to wireless devices, microprocessors, ASICs and a range of fabless semiconductor makers.

"We foresee a new paradigm of performance well beyond what could be forecasted by Moore's Law," said Merritt.

Indeed, current memories were architected with PCs in mind, while today's high volume cellphones stack DRAM, SRAM and flash die in an effort to meet their memory needs, said Avo Kanadjian, vice president of marketing for Nantero that plans to sample Mbit-class memories in 2005.

"The performance of ASICs and SOCs are definitely limited by the performance of DRAM and flash," he said.

For its part, Axon Technologies Corp. (Tempe, Ariz.) said it will deliver a very low power nonvolatile DRAM as a discrete chip by early 2007. "If we don't have a new memory out I that time frame we will miss the bus and be doomed," said Michael Kozicki, chief technology officer and founder of Axon.

Freescale Semiconductor Inc. is sampling its magnetic RAM in Mbit densities now. It expects users to qualify discrete parts starting next year for applications replacing battery-backed SRAM and flash/SRAM combinations, said Saied Tehrani, Freescale's director of MRAM.

"The need for a new memory architecture is very real. This is something all semiconductor manufacturers are currently looking at," said Salvatore Coffa, a director of research at STMicroelectronics focused on optoelectronic and post-silicon technologies.

ST is working on a variety of nanotech-based components including opto-couplers, DNA microarrays and LED replacements that could offer more tightly focused light beams and lower costs than today's products. The company is also working with a European consortium on polymer electronics based on nanotech that could deliver KHz range devices at very low costs.

Other areas where nanotech will start to have a significant impact in the next three to five years include solar cells and novel battery types such as fuel cells, said Jurvetson.

By: DocMemory
Copyright © 2023 CST, Inc. All Rights Reserved