February 20, 2008

IBM using DNA to make carbon nanotube grid computer chips

Left is previous work where DNA was wrapped around carbon nanotubes.

Scientists at IBM are conducting research into arranging carbon nanotubes--strands of carbon atoms that can conduct electricity--into arrays with DNA molecules.
Once the nanotube array is meticulously constructed, the laboratory-generated DNA molecules could be removed, leaving an orderly grid of nanotubes. The nanotube grid, conceivably, could function as a data storage device or perform calculations.

Previous work to create grids using DNA

Potentially, DNA could address, or recognize, features as small as two nanometers. Cutting-edge chips today have features that average 45 nanometers.

"These are DNA nanostructures that are self-assembled into discrete shapes. Our goal is to use these structures as bread boards on which to assemble carbon nanotubes, silicon nanowires, quantum dots," said Greg Wallraff, an IBM scientist and a lithography and materials expert working on the project. "What we are really making are tiny DNA circuit boards that will be used to assemble other components."

UPDATE: The attachment sites on DNA, which is where the nanowires and transistors would attach on the template, can be made much closer together than with traditional pattern manufacturing techniques.
With DNA, the attachment sites are 4nm to 6nm apart. Normally, they're about 45nm apart.

"Think of it as tiling a floor. These DNA pieces are like tiles," explained Gordon. "Each tile has some array of electronic components. Those tiles are placed on a chip in a larger array so there are thousands or millions on a chip. The second step, which we don't know how to do yet, would be to wire them all together. We've got sizes well below conventional lithography."

Wallraff said the next steps will be connect all the tiles together and check the defect levels during assembly.

Actually using this pattern technique is probably 10 to 20 years away, he noted.

Other work to enable self assembly of electronics:
A simple surface treatment technique demonstrated by a collaboration between researchers at the National Institute of Standards and Technology (NIST), Penn State and the University of Kentucky potentially offers a low-cost way to mass produce large arrays of organic electronic transistors on polymer sheets for a wide range of applications including flexible displays, “intelligent paper” and flexible sheets of biosensor arrays for field diagnostics.

The researchers found that by applying a specially tailored pretreatment compound to the contacts before applying the organic semiconductor solution, they could induce the molecules in solution to self-assemble into well-ordered crystals at the contact sites. These structures grow outwards to join across the FET channel in a way that provides good electrical properties at the FET site, but further away from the treated contacts the molecules dry in a more random, helter-skelter arrangement that has dramatically poorer properties—effectively providing the needed electrical isolation for each device without any additional processing steps. The work is an example of the merging of device structure and function that may enable low cost manufacturing, and an area where organic materials have important advantages.

In creating chip arrays, DNA assembly might work as follows: scientists would first create scaffolds of designer DNA manipulated into specific shapes. Rothemund has made DNA structures in the shapes of circles, stars, and happy faces.

A pattern would then be etched into a photo-resistant surface with e-beam lithography and the combination of several interacting thin films. A solution of the designer DNA would then be poured on the patterned surface and the DNA would space themselves out according to the patterns on the substrate and the chemical/physical forces between the molecules.

The nanotubes would then be poured in. Interactions between the nanotubes and the DNA would occur until they formed the desired pattern. Single strand DNA, along with origami, could be used in concert.

Another key part in the system revolves around peptides that can bind to the DNA and a nonbiologically inspired molecule like a nanotube.

With DNA, chipmakers could phase out multibillion fabrication facilities stocked with lithography systems, which cost tens of millions of dollars, and the other "top-down" style equipment.

Potentially, DNA techniques could allow manufacturers to produce features that are smaller than patterns that could be achieved even with the most advanced lithography systems, predicted Wallraff. E-beam lithography, which is extremely difficult to use in mass manufacturing, goes down to 10 nanometers.

"Of course, the devil is in the details," said Wallraff. "These are self-assembly procedures and error rates--missing features could be the downfall."

Carbon nanotube transistor work at IBM

UPDATE: Combine this work to get to 2 nanometer feature sizes with configurable more customized processors and we can go from affordable exaflop computers to zettaflop supercomputers. Exaflop is 1000 petaflops, 1 million teraflops. Zettaflops is 1 million petaflops and 1 billion teraflops.

UPDATE: John E. Kelly, IBM's Director of Research, is focusing on four top research priorities.
Each of the projects will get $100 million over the next two to three years, in hopes of generating at least $1 billion, each, in new revenue. The projects: inventing a successor to today's semiconductor, designing computers that process data much more efficiently, using math to solve complex business problems, and building massive clusters of computers that operate like a single machine—an approach called "cloud" computing.

Kelly foresees creating dozens of new joint ventures for research, which he calls "collaboratories," with countries, companies, and independent research outfits. One venture with Saudi Arabia, focusing on nanotechnology, was unveiled on Feb. 26. The two sides plan to develop technologies for solar energy and water desalinization.

In 2003, Kelly took a gamble and set up research alliances with a handful of partners, including Sony Electronics (SNE) and Advanced Micro Devices (AMD), to share expenses and brainpower. The approach eventually paid off, as IBM's chip business returned to profitability and remained on the cutting edge of technology.