As I’ve mentioned before, there’s a lot of nanotechnology going on in my part of Upstate New York. So I was very pleased this weekend to finally make it to Community Day at the College of Nanoscale Science and Engineering (CNSE) of the University at Albany.
The event was part of the school’s month-long community and educational outreach initiative, which they’ve dubbed NANOvember. Although my son and I did not have time to explore all the family-friendly demonstrations and activities around the CNSE’s modernistic $14 billion campus, we did get to see the facility’s clean rooms and find out how a computer chip is made.
Our tour guide was Dr. Scott Tenenbaum, Associate Professor of Nanobioscience, who studies RNA biology using nano-based technology. (See the video below for a little more about his work.) Amazingly, the process as Dr. Tenenbaum described it was easy to understand, although the work involved is obviously incredibly complex.
Through a wall of observation windows, we could see banks of “tools” — actually, large machines lined up much like the room-sized computer banks — used in creating the microchips found in virtually every device imaginable. Stacked up on the other side of the window were carriers filled with stacks of silicon disks, about the size of a bowling ball bag (but much more precious, with each disk costing a few thousand dollars). And the clean rooms glowed with a mysterious bright yellow light, for reason we were soon to learn.
As Dr. Tenenbaum explained, the connections between the hundreds of millions of transistors housed on a tiny microchip — about the size of a flat Tic Tac — are much too small to use even the thinnest copper wire. So instead, microscopic “trenches” are etched into each chip and filled with molecules of copper.
The chips are cut from a disk of silicon — which today are about the size of an old vinyl 78, although the school, like the nanotech industry as a whole, is starting to work with much larger disks that could make chips even cheaper to produce than they are now. And like an old record, the disks are made to spin at super-high speeds for the steps that are to follow.
The disks are coated with a photoresist substance using the same technique as the Spin Art paintings kids can make at carnivals. A mask is laid over the hard coating which leaves the circuit design open, like a stencil. The disk is exposed to ultraviolet light, and where the photoresist is not covered by the mask, the energy of the light causes the coating to liquify. (The yellow-tinted lighting is used in the chip fab facility because it does not interfere with the ultraviolet light — kind of like the red light in an old-fashioned photographic darkroom.)
The softened coating is then washed away, leaving the trenches ready for the copper to be electroplated to the chip surface. This process is repeated over and over, creating layers of different metals — much the same way silk screen art is done, by layering on one color of ink at a time in different patterns to create the whole picture.
Some of the research being done at the CNSE includes finding ways to transport and work with the disks so they stay absolutely level during the fab process. That gets trickier as the disks get bigger, Tenenbaum explained. The school is also looking into the cost industry-wide of switching from copper to another element that might allow for even more transistors to be fit onto a chip. The effect, as Tenenbaum pointed out, would be so vast that it could impact the economy of emerging countries which now export the copper used to make chips today.
After years of reading about the business end of the nanotechnology going on around me, it was great to finally understand some of the science involved as well. But don’t worry if you missed the CSNE tour: here’s a slideshow from Intel (via apcmag.com) that illustrates the process in clear detail. And the CSNE also has a page of nano education resources for kids.