As microchips get progressively smaller and denser, each decrease in size brings its own set of problems. Currently, the fine metallic lines can be as narrow as a few dozen nanometers, so edgewise irregularities or edge roughness of just a few nanometers can cause major problems, such as current leakage or voltage fluctuation in a microprocessor. These tiny blemishes on the nano lines arise from the random fluctuation of electrons or photons during the process of nanofabrication.
Electrical engineer Stephen Chou, who runs Princeton University’s NanoStructures Laboratory, has developed an inexpensive way to reduce edge roughness and smooth the sides of nanostructures. Eventually, the technique could be used to make more-precise chip features. The idea is fairly simple: instead of trying to fine-tune the fabrication process itself, which is already highly refined, Chou’s strategy is to correct intrinsic defects after fabrication.
Chou’s method, called self-perfection through liquefaction (SPEL), involves using an ultraviolet laser to melt away the defects. Various types of SPEL, applied to polymers, metals, and semiconductors, are outlined by Chou and his former graduate student Qiangfei Xia in the May 4 online Nature Nanotechnology.
“When you make things very small, eventually, you’re going to be limited by the noise of the manufacturing process itself,” Chou says. “Any attempt to try to improve them becomes a fruitless effort.” So Chou took a different tack, smoothing out defects after the structures were produced. Previously, Chou pioneered several fabrication methods for making nanostructure and optical devices.
Chou’s SPEL technique utilizes an excimer laser–the same kind of laser commonly used in eye surgery–that heats only the top surface layer of its target object. Chou says that he could have used extreme heat to reshape the microchip components, but at an extreme cost. “If you use the ordinary heating process for the hard material like silicon or metal, you’re going to not only melt the structure you want to fix, but also everything else,” Chou says. The result would be a mess–“That’s not going to work.”
Instead, a 20-nanosecond laser pulse melts only the superficial rough spots in the structure and leaves the rest intact. Then the liquid flows or is guided into the correct shape before solidifying. Chou has previously used the technique on polymers, but this research represents the first application to metals.
“What is nice about the method is that it takes advantage of self-assembly,” says George M. Whitesides, a professor of chemistry at Harvard University and a pioneer in nanofabrication. “You start with a structure that isn’t the shape you want, and let it fold itself into the shape you want.”
Of the three ways to use SPEL described by Chou and Xia, who is now at the Information and Quantum Systems Laboratory at Hewlett Packard Labs, Chou says that the one called “guided SPEL” is the most exciting. During that process, a quartz plate is placed above the flawed material, with a small gap in between. The laser passes through the plate and melts the material, which then rises to meet the plate. (Chou says that he does not completely understand the electrostatic interaction that explains why the material rises and narrows.)
This process produces smooth sides and a flat top to the line of metallic material. In addition, it makes the material taller and narrower, which means that microchip manufacturers will be able to make a denser chip.
The process has some limitations. When nanostructures get so small that they are the same size as the defects, for example, SPEL won’t be able to help. Nonetheless, Chou says that the small, regular structures that emerge from SPEL have already caught the attention of major chip manufacturers.
“The semiconductor manufacturing people do an incredibly good job at this right now,” says Karl Berggren, an associate professor of electrical engineering at MIT, “but they may be able to take advantage of this new trick, to push things even further.”
The big new idea for making self-driving cars that can go anywhere
The mainstream approach to driverless cars is slow and difficult. These startups think going all-in on AI will get there faster.
Inside Charm Industrial’s big bet on corn stalks for carbon removal
The startup used plant matter and bio-oil to sequester thousands of tons of carbon. The question now is how reliable, scalable, and economical this approach will prove.
The hype around DeepMind’s new AI model misses what’s actually cool about it
Some worry that the chatter about these tools is doing the whole field a disservice.
The dark secret behind those cute AI-generated animal images
Google Brain has revealed its own image-making AI, called Imagen. But don't expect to see anything that isn't wholesome.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.