Superconducting Qubits and (Maybe) Quantum Computing
Robert McDermott | University of Wisconsin
For a decade or so, physicists have made headlines with talk of a new, powerful quantum computer, one that could solve problems with undreamed of complexity. Reactions have ranged from optimistic to skeptical.
For Robert McDermott, an assistant professor of physics at the University of Wisconsin, that goal is at best a point on a distant horizon, but giant steps have been accomplished.
"We are a long way from realizing a computer large enough to do something practical," McDermott said. "We have moved toward that goal, but there's still no guarantee that we will get there. Meanwhile, however, we're learning much about the quantum mechanics of nanoscale systems."
McDermott is running experiments that are part of a global effort to guide the eventual construction of such a computer, based on nanoscale units called quantum bits, or qubits. The curious qubit has the freedom to assume, in computer language, both a zero and a one and some other states, or superpositions, as well, simultaneously. If that is not strange enough, quantum mechanisms also have a property called "entanglement," by which a change in an atom appears instantly in a separate atom some distance away.
"We can take advantage of quantum superpositions and entanglement," McDermott said. "I am an experimentalist, and my interest is in how to build a computer that operates quantum mechanically."
McDermott's approach makes use of superconducting circuits, somewhat like the circuits on a silicon chip in a conventional computer. The superconducting circuits incorporate special metals, in which electricity flows without resistance, along with layers of dielectric materials -- ones that cannot conduct electricity. The circuits exploit a phenomenon called the Josephson Effect, the result when two superconductors, divided by an insulating barrier, form tunnels for electrons to move across that barrier. To function properly, these circuits must be maintained at a very low temperatures: around 20 millikelvin, about 100 times colder than outer space. To achieve these low temperatures, researchers use a machine called a dilution refrigerator, which cools with helium isotopes. McDermott and his students assembled such a cooling system from parts fabricated in the physics department's machine shop.
A key problem central to McDermott's work is decoherence: the degradation of the quantum state of the qubit. While the bits in an ordinary computer stay put, at zero or one, and retain your data, in a quantum computer that won't happen. "In microseconds all the quantum information is lost," McDermott said. "We need to find a way to extend coherence times by improving the materials of the qubit circuit."
McDermott hopes to help explain the deeper fundamental physics of such decoherence. "It would be great to reduce decoherence to a level where we could build a quantum computer on a large scale," he said. While some projects have achieved a scale of two or three qubits, a useful computer would require maybe 1,000 qubits or more to solve the problems we cannot solve with today's computers.
As for applications, quantum computers would excel at factorization of large numbers, McDermott said, which has implications for cryptography. Potential spin-offs of this work include the realization of improved detectors based on superconductors. For example, McDermott's research may help improve a type of ultrasensitive detector known as a SQUID, for superconducting quantum interference device. SQUIDs have been applied to such varied tasks as measurement of the magnetic fields generated by the brain, and astrophysical searches for subtle gravity waves from outer space.
Meanwhile, just getting a quantum mechanics computer up to the needed qubit levels is one of his major goals. And, says McDermott, "That's still a long way off."
Robert McDermott's Teaching Plans
McDermott aims to expand the Advanced Physics Laboratory course at the University of Wisconsin-Madison. "We have a great course, but I will work to expand it by adding new experiments," he said. "It's important to get undergraduates involved, doing hands-on modern physics, and letting them figure things out on their own. That will expose them to the type of work they will do in graduate school and as future scientists."
McDermott also will collaborate with Wisconsin's teacher enhancement program, an outreach effort working with teachers in local high schools. He plans to set up an inexpensive receiver station, for radio frequency (RF) signals broadcast to earth by National Oceanic and Atmospheric Administration (NOAA) weather satellites. Students will download the data from satellites in geosynchronous orbit, matching the Earth's rotation. "They can dump the data into their laptops and display the images in real time," McDermott said. "They can see Earth, the cloud cover, and depending on where it's passing over, they can see to Florida, or north into Canada." They will use infrared (IR) data, showing temperatures and other measures. "It is kind of fun to see these images," McDermott said. "This will stimulate the interest of high-school students learning what it takes to build a data acquisition system. They will think about climate, satellite orbits, classical mechanics of orbits, electromagnetism. There's a lot of physics in it."
