Technology News
03/01/2002
MRS Fall Meeting: Nanotubes, self assembly, and harnessing phages
As integrated circuit features shrink toward the molecular scale, intense effort is going toward moving beyond CMOS to the nanoworld. The wide range of this R&D was evident at the Fall 2001 Materials Research Society meeting in Boston in December. Work with carbon nanotubes, for electronic devices and sensors as well as displays, was extensively reported. There was also an update on next-generation lithography (NGL) progress. A few sessions probed the growing link between research in biotechnology and electronic devices and novel materials.
While there are many examples of self-assembly (the conference theme) observed in nature at the nano- or molecular level, finding repeatable production processes, and fabricating devices that could be wired together for practical systems is proving difficult. Carbon nanotubes are being extensively studied for a variety of applications, from integrated devices and nanowires to various types of electron tubes. Their use in a bright field-emission display (FED) was reported by researchers from DuPont Central R&D, Wilmington, DE. They are ideal field emitters, thermally and chemically stable with a large aspect ratio and low threshold voltage. They can be applied as a printable paste or photoimageable material, according to L.T. A. Cheng of DuPont, but a special surface treatment is needed to improve uniformity, current density and light emission. Experiments reported by O. Zhou et. al. of North Carolina U. with single-walled and multi-walled nanotubes for a number of potential devices showed that the longer MWNTs had lower threshold fields, which they demonstrated by building a working x-ray tube. J.M. Kim and his team from Samsung, Suwon, Korea, developed high brightness carbon nanotube FEDs with full-video with over 300cd/m2 brightness.
Many advances in microelectromechanical systems (MEMS) processing were reported. Helping to spur development is a projected multibillion business for MEMS automotive microsensors by 2005, according to G. Flik of Robert Bosch, Stuttgart, Germany. These would include mass flow and pressure sensors for engines and emission systems as well as passenger safety (airbags). Robert Ellis of Sandia reported further progress on amorphous diamond cantilever beams made by pulse laser deposition. These are 10,000¥ more wear resistant than silicon, while providing very low stiction (a problem with silicon cantilevers), and they are nearly hydrophobic.
Progress on the extreme ultraviolet (EUV) lithography engineering test stand (ETS) alpha tool that has been operating since last March was reported by Donald Sweeney of Lawrence Livermore Lab. A superior optical system, which has already demonstrated static imaging for 70nm features, will be installed this fall. One goal for EUV is optical coatings with greater than 70% reflectivity. There is also progress on cutting defects in masks, which are also reflective. A killer defect might be just 50nm wide and 2nm high, Sweeney said. With such short wavelengths (13-14nm), which are highly absorbed by all materials, photoresists must be only about 150nm thick, several times thinner than today's resist layers, observed Jonathan Cobb of Motorola. They must be able to efficiently use photons of fairly low (4-5mJ/cm2) doses with fluctuations that may be a significant percentage of the total dose. Line edge roughness will probably need to be less than 4nm, 3 sigma, and the resists must be stable in vacuum with minimal outgassing upon exposure, Cobb stated.
While there were many imaginative experiments reported on methods to fabricate nanoscale devices, one of the most fascinating projects was described by Angela Belcher of UT Austin. Biological methods are being explored for nanofabrication of crystals and other regular structures. By checking hundreds of thousands of bacteria using peptide combinatorial methods, proteins are identified that select for and bind particular semiconductor or magnetic particles. Some phages are found that bind to particular III-V or II-VI nanoparticles, such as gallium arsenide, zinc or cadmium sulfide, and these are multiplied in a culture. Some peptides bind so tightly to crystals that they cannot be removed, she said. Some were found that bind five zinc sulfide particles at one end of a peptide, for example, and these phages can be lined up to align the particles using their inherent crystal face specificity. Her group was able to make liquid crystals by what Belcher calls 'Phagemid Engineering.' Some experiments found bacteria that bound different metals at each end, and one grad student discovered that raising the temperature sometimes allowed more nanoparticles to be bound.
HP, UCLA get patent for molecular chips
Hewlett-Packard Co. and the University of California, Los Angeles have received a US patent for technology that could make it possible to build complex logic chips at the molecular scale.
The patent, issued to Philip J. Kuekes and R. Stanley Williams of HP Labs and James R. Heath of UCLA, builds on previous patents and scientific work by the company and university, working under a grant from DARPA, with matching funds from HP. The HP and UCLA invention proposes the use of a simple grid of wires — each just a few atoms wide — connected by electronic switches a single molecule thick. HP has demonstrated in the lab how some rare earth metals naturally form themselves into nanoscopic parallel wires when they react chemically with a silicon substrate. Two sets of facing parallel wires, oriented roughly perpendicular to each other, could then be made into a grid, like a street map.
In a related experiment, researchers from the collaboration crossed wires the size of those used in today's computer chips and sandwiched them around a one-molecule-thick layer of electrically switchable molecules called rotaxanes. Simple logic gates were then created electronically by downloading signals to molecules trapped between the crosswires.
"That work demonstrated for the first time that molecules could be used as electronic devices to perform computer logic," said Heath. The HP/UCLA collaboration has also patented a memory chip based on molecular switches.
ST fab using superconducting technology to guard power
Based on superconducting technology, an innovative solution that ensures power quality and reliability is helping ST Microelectronics in Agrate, Italy, cope with increasing potential for voltage variances from its local power grid.
 |
Power line disturbances are a leading cause of interruptions at semiconductor manufacturing sites. Even the briefest voltage sag can initiate a shutdown-cascade costing millions of dollars for each incident. Jeff Nestel-Patt, VP of corporate communications at American Superconductor Corp., Westborough, MA, says, "As the semiconductor industry strives to make the manufacturing process more efficient and productive, the cost of interruptions in production flows grows dramatically. Any lost production time is a mission-critical event that can cost millions of dollars."
The solution at ST is a power-quality industrial-voltage regulator (PQ-IVR, see illustration) from American Superconductor. The PQ-IVR has a superconducting coil of niobium-titanium wire that carries a large, continuously circulating direct current with zero electrical resistance. The magnet is contained within a vacuum-cooled, high-efficiency cryostat and is connected to the external environment through efficient, high-temperature superconducting current leads. In operation, a module detects voltage sags and, within half a cycle, injects real and reactive power from the magnet storage into the power grid to maintain voltage. The system can synthesize an essentially perfect waveform in the face of a three-phase disturbance because it can draw instantaneously on a large power reserve in the magnetic storage system.
While PQ-IVR is relatively new for semiconductor industry applications, power quality technology has been used for several years in other industries with sensitive processes, including the paper, plastics, chemicals, and aluminum-smelting industries, as well as data processing, military, and research laboratory applications.
At ST-Agrate — one of ST's largest operations with production capability down to 0.35μm, 180nm, and 130nm R&D, and a new 200mm facility for flash memories — power supplied from the local grid is subject to frequent voltage variances. Each variance could interrupt critical manufacturing processes. In August, two PQ-IVR units — provided by Edison, the largest private utility in Italy — went online. The two units protect one cleanroom operation representing 6.5MW; the protected load includes a 150mm-wafer line with a capacity of 13,000 wafer starts/week.
 |
Tiny bang theory
Chemists at the University of California, San Diego, have discovered that silicon wafers can easily be made into tiny explosives that may one day be used to chemically analyze samples in the field or serve as a power source for electronic sensors the size of a dust mote. The UCSD scientists knew that a silicon-based explosive would explode when mixed with potassium nitrate.
However, a post-doc researcher in Prof. Michael Sailor's lab discovered by accident while working with a porous silicon wafer that substituting potassium nitrate with gadolinium nitrate had the same effect.
"When he tried to cleave the wafer with a diamond scribe, it blew up in his face," recalled Sailor. "It was just a small explosion, but it really surprised us. The gadolinium produces a very clean-burning flame."
The absence of chemical impurities makes the explosive ideal for use in a device that could perform rapid chemical analysis of toxic metals and other elements in the field. A futuristic application might be to use the explosive as a propulsion source for MEMS devices. Another possibility may be to construct information-collecting devices that self-destruct.