Without the technological progress described by Moore’s law, our world would be a very different place. In 1965, Intel cofounder Gordon E. Moore predicted that the number of transistors that can fit onto a microchip would double roughly every two years. So far he’s been right, and his prediction has translated into decades of increasingly powerful and smaller electronics, from smartphones to laptops. But when scientists reach the ultimate physical limit for real estate on microchips, Moore’s law will inevitably grind to a halt.
David Bishop, head of the College of Engineering’s Division of Materials Science & Engineering, hopes that day won’t come for a while yet, and he’s working on a tool that may be able to help. He’s creating a kind of 3-D printer that can print using individual atoms.
“We’re developing a new technique for manufacturing devices at the nanoscale,” says Bishop, a professor of physics and electrical & computer engineering. “You don’t have to extrapolate Moore’s law for much more than a decade or two before you come to the point where devices will be composed of one atom, or at most a few atoms.”
Dubbed “atomic calligraphy,” Bishop’s new technique relies on his expertise in the field of microelectromechanical systems (MEMS), which are devices only micrometers in size. Fabricated by stacking thin layers of compounds in specific patterns on top of a silicon wafer, then washing away unwanted areas with dissolving acids, these tiny machines can be created in a dizzying number of varieties with many different functions.
Over the course of his 30-year career at Bell Labs—the research and development laboratory in Berkeley Heights, New Jersey, owned by Alcatel-Lucent, where he worked before coming to Boston University—Bishop had the opportunity to oversee many such projects, including the development of microphones smaller than a grain of sand and large arrays of microscopic mirrors for efficiently redirecting data sent through optical fibers.
To create a machine capable of writing with atoms, Bishop and his team fabricated a MEMS device with a flat circular plate roughly the diameter of a human egg cell. At the center of the plate lies an aperture only tens of nanometers in diameter, less than half the size of the HIV virus. When a beam of atoms is shot through the aperture hole, the atoms are deposited onto a surface below the disk at a controlled rate. Microscopic springs and programmable motors move the plate, causing the atoms to be deposited in precise patterns.
This allows for the printing of relatively flat structures like circuits. But the atoms can also be deposited on top of each other, meaning that complex nanoscale devices can be created in a similar fashion to a 3-D printer. For now, Bishop isn’t as concerned with what devices he can create with atomic calligraphy as he is with first making sure the technique works in a scalable and repeatable way.
“It’s less that I want to build a one-atom transistor—I know that an array of one-atom transistors will be interesting and important,” says Bishop. “The goal is to have a technique that lets us build devices at the nanoscale, with precision and control, and do so in a manufacturable way. When and if we discover some interesting new physics that creates a potentially important device opportunity, then we would know how to manufacture it as well.”
In addition to atomic calligraphy, Bishop and his team are also working on creating a micro-microscope. This MEMS device—also smaller than a grain of sand—consists of a stage that sits at the end of a long, thin optical fiber. There, the sample to be viewed is sprayed with a shower of photons. A sensor captures how those photons bounce off the sample, allowing the image to be reconstructed at a greater resolution than standard optical microscopes. “With atomic calligraphy and the micro-microscope we are creating all of the elements of a semiconductor fabrication factory on a single silicon wafer. We call it the ‘Fab on a Chip,’” says Bishop.
If it seems unusual that Bishop, who was most recently chief technology officer of Bell Labs’ federal research unit, would be developing these devices in an academic environment, he has a straightforward answer.
“As the recent twists and turns of the telecom industry happened,” says Bishop, “I started thinking it might be nice to train some students, which you don’t do at Bell Labs, generally, and really think about how I could give back to the field that’s been so wonderful and interesting to me.”
With the freedom to pursue his projects at BU, Bishop hopes to soon make a big impact with his small machines. For now, he says, “It’s been a wonderful experience living in Boston and working at the University. It’s everything I’ve hoped it would be. I consider myself fortunate to be here.”