Center for NanoTechnology to push lithography below 35 nanometers
Using a one nanometer X-ray wavelength and masks fabricated onto diamond membranes, the Center for NanoTechnology and Mitsubishi Electric will develop lithography technology to take the semiconductor industry four generations into the future.
McFarland-Bascom Professor of Electrical and Computer Engineering Franco Cerrina says a three-year, $4.2 million dollar grant from the Defense Advanced Research Projects Agency (DARPA) will help the Center for NanoTechnology acquire new equipment, modify existing tools and develop new processes to extend lithography below 35 nanometers.
Currently, the semiconductor industry uses an ultraviolet light wavelength of 248 and 193 nanometers to pattern transistors as small as 130 to 150 nanometers, says Cerrina. But to push fabrication below 100 nanometers, researchers need new technical solutions. For example, at shorter wavelengths lenses intended to focus the circuit pattern on the silicon absorb much of the light. Overcoming these obstacles dramatically increases the complexity and costs of nanometer-scale device fabrication systems.
Working with X-rays, Center for NanoTechnology researchers developed much simpler solutions using lithographic techniques on the order of 70 to 80 nanometers. The DARPA grant will help the center push the limits of lithography to 50, 35 and ultimately 20 nanometers.
"The region past 35 nanometers is well beyond the technology roadmap of the semiconductor industry," says Cerrina. "We call the move below 50 nanometers 'X-ray phase two' because we're using a new set of mask materials and systems. Previously we used a silicon-carbide base material. Now we will develop new photochemistry and will use two to five-micron-thick diamond membranes. The new diamond masks are much more rigid and will help to eliminate some of the problems that plagued earlier versions of X-ray lithography."
Mitsubishi Electric will supply the diamond membranes and develop the masks. Cerrina's team will modify an existing X-ray beamline and develop the exposure system together with Suss Advanced Lithography (SAL). "Our facility has one of the only synchrotrons with the proper characteristics to carry out this research," says Cerrina. "It's an excellent collaboration."
Cerrina says project funding will provide important upgrades to the center's equipment including a new high-resolution scanning electron microscope and a new atomic force microscope.
DNA chip technology to identify viruses and other long genetic sequences
Biotechnology Center Director Michael Sussman and Center for NanoTechnology Director Franco Cerrina will adapt their DNA Micro Array Synthesizer (MAS) for the study of long genetic sequences of up to 10,000 base pairs. A three-year, $2.7 million dollar grant from the U.S. Navy will help the team develop new exposure tools and photochemistry needed to achieve their goal.
"Suppose you have a very long strand of DNA. Currently, you can select maybe 20 or 25 short strands that are unique to the DNA and then you look for those. The ability to quickly detect long genetic sequences could lead to a portable tool that could identify viruses and other biological agents in real time," says Cerrina. "But even more important will be the diagnostic capabilities that will be associated with these applications. This will be useful in medicine and research from stem cells to corn hybridization."
The Micro Array Synthesizer will allow virtually any research scientist to make customized DNA chips that would otherwise take months to create. Current technology depends on mask-based photolithography, a process that requires shining ultraviolet light through a series of stencil-like masks onto a glass chip resulting in the synthesis of tens of thousands of DNA molecules of interest.
MAS technology sidesteps the need for delicate and expensive masks by relying on an array of 480,000 tiny aluminum mirrors arranged on a computer chip. By manipulating the mirrors, the team found they could synthesize DNA by shining light in very specific patterns.
Cerrina says the programmability of the mirror arrays will be key to successfully developing a tool for longer sequences.