Researchers at the University of Toronto are at the forefront of a new computer revolution. Many computer engineers believe that advancements in computer chip technology, as we know it today, will cease by 2010, and that the next generation of computers will use optical technology-using light to carry information instead of electricity, like in fibre optic cables for telephones. Professor Geoffrey Ozin of U of T’s Department of Chemistry spoke at the Koffler Institute for Pharmacy Management this past Wednesday in a free public lecture entitled “Toward the Synthetic All-Optical Computer: Science Fiction or Reality?” He and his team are now conducting studies that focus on photonic crystals to discover new ways of building faster computers.
U of T’s own Professor Sajeev John of the Department of Physics developed the theory behind photonic crystals in 1987, but most scientists have only recognized their true importance in the last few years. These crystals forbid certain types of light to pass through them, making them ideal to build optical parts because they can control what type of light goes where and when.
Dr. Ozin and his team have successfully constructed microscopic pathways made of photonic crystals that carry light just like metal conductors carry electricity. The structure of these materials is a lot like opal stones. “Instead of opals on a ring, you now have opals as microchips,” claims Ozin.
They have also demonstrated that these pathways as well as other structures can be moulded into any shape or contour. This is absolutely essential for fabricating and mass-producing optical components.
This research comes at a crucial time. Currently, we try to create faster machines by miniaturizating electronic components-making them smaller and smaller. The laws of physics will soon thwart this effort.
To fully understand this limitation, we must consider transistors: the fundamental building blocks of modern computers. We can think of transistors as tiny switches made out of silicon, copper and other substances. Because silicon is a semiconductor, it can conduct electricity under certain conditions. We can allow electricity to flow by applying a small voltage to the middle of a transistor, like pressing a button to ring a doorbell. Instead of using our fingers, we use electricity.
A single transistor is not very useful. However, if we let transistors activate other transistors and interconnect them in a logical sequence, we can construct a device capable of making decisions and calculations. We can also store information by wiring together a network of transistors as well. These components collectively define a computer.
To increase the speed of computers, we try to reduce the physical size of transistors to shorten the travel time of electricity. The larger the physical size of a transistor, the longer it takes for electricity to travel through it. We are now at a point where we can mass-produce transistors that are only 60 nanometers across, or roughly 1/3,000th the thickness of a human hair.
Miniaturization of transistors cannot continue much further. Electricity is essentially made up of electrons, and quantum mechanics tells us that below a scale of about 20 nanometers, electrons are unpredictable. An electron may move uncontrollably from one side of a transistor to another. Transistors will act less like switches and more like leaky faucets.
This size limit inspired engineers and scientists to look at completely different alternatives, like optical computing. As optical computing would rely on light instead of electricity, it would overcome the unpredictability of electrons. Optical computing is also theoretically faster since basic physics says that light travels much faster than electricity.
An equivalent to the electronic transistor for light has been developed. This optical switch, which some researchers have dubbed the “transphasor,” is activated by light. It effectively becomes transparent to allow light to pass through it, and opaque to block light. Currently, transphasors are difficult to manufacture.
Dr. Ozin is working towards making the mass-production of transphasors feasible. Ozin declares, “It’s like the optical equivalent of silicon semiconductors. We have to take it seriously.”
