We may be one step closer to quantum computing that's inherently protected from errors. Physicists at the University of Copenhagen have successfully created an exotic type of pseudo-particle that is immune to outside interference. The results are described in a new paper in Nature. Unlike a classical computer, with bits representing 0's and 1's, a quantum computer stores information in "qubits". Thanks to the weirdness of quantum mechanics, such a qubit can be in two states at once, both 0 and 1, just like Schroedinger's cat is simultaneously alive and dead until a measurement or observation is made. It's called a superposition of states.
There are many proposed methods for building a quantum computer, but most share a common challenge: the quantum information must be shielded from all external noise in the surrounding environment. The slightest bit of interference — a single photon bumping into the atom you've used to encode and store your information, for instance — will cause the entire system to "decohere", such that the all-important superposition that lets your qubit be both 0 and 1 at the same time is lost. That means errors in your calculations.
"It's almost like the environment bullies us into one of many possible states," Spyridon Michalakis, a physicist at Caltech's Institute for Quantum Information and Matter, told Gizmodo.
Physicists have devised ingenious methods in recent years for quantum error correction by exploiting the phenomenon of entanglement. This lets them check the data without making any actual measurement, thereby preserving the superposition. But what if you could have built-in protection from outside interference, making error correction unnecessary? That's the idea behind topological quantum computing, the focus of Charles Marcus and colleagues at the University of Copenhagen's Niels Bohr Institute, bolstered by a major investment from Microsoft.
Mathematically, topology is distinct from geometry. A ball is a very different shape from a cube, geometrically speaking, but from a topological standpoint, they are the same, because you can deform one into the other without having to make a hole or a cut.
A topological quantum computer would braid qubits into a kind of knot. Different kinds of braids would encode different computational tasks, and those structures would be topologically stable. You would be able to manipulate them without destroying the superposition, making such a computer inherently more robust.
Graduate student Sven Albrecht and Charles Marcus. Credit: Ola Jakup Joensen/NBI
"If you're using geometry in your quantum computer, you must rotate things very precisely to build a quantum gate," Michalakis said. Otherwise errors accumulate and throw off the entire calculation. But a topological qubit is so stable that you can rotate one around another, any way you like, provided you complete a full loop to create a simple knot or braid.
OK, but how do you build a topological qubit? First, you need to create special kinds of pseudo-particles called anyons. We usually consider electrons to be fundamental particles, that is, indivisible into smaller components. But things get weird when you get down to two dimensions. In this space, quantum mechanics allows an electron to split into two (or three) smaller components, each carrying a fraction of the charge. They're like bubbles that form in a quantum liquid. These bubbles are the anyons. "We just never get to see them unless we create very specific conditions," Michalakis said.
Such particles "don't exist on their own, but they can be created using a combination of materials involving superconductors and semiconductors," Marcus said in a statement. Superconductivity is the ability of electrons to flow through a material with zero resistance from friction, a state normally achieved at very low temperatures, on the order of liquid helium or liquid nitrogen. Marcus's group essentially created a new kind of topological superconductor.
As the video below explains, you can also think of anyons as needles, each with a string going back to the dawn of time. When another electron circles one of those anyons, the "threads" knot together; it's even possible to track how many times this happens.
This is important because of the nature of superconductivity. Much like entering the carpool lane on a highway, electrons can only jump into the superconducting "fast lane" in pairs. Anyons can split and be their own partners, however; they get a free pass without having to pay the usual "energy toll," according to Michalakis.
Caltech's Alexei Kitaev first proposed the idea of a topological quantum computer several years ago, and physicists at Delft University in the Netherlands found the first evidence for anyons in 2012. Kitaev envisioned a system comprised of a superconducting material with a thin semiconducting nanowire on top. Marcus and his Copenhagen colleagues inverted that design. They fabricated an ultra thin semiconducting nanowire and then squeezed a superconducting layer on top of it, like mustard being squeezed onto a hot dog.
The trick was to then grow the nanowires to sufficient lengths that there was no overlap between the two halves of the electron on either end (resulting from the dual wave nature of such particles). If the ends of the nanowire are too close together, the two halves will recombine into a whole electron. But get them sufficiently far apart, and electrons will split into two parts, one on each end of the nanowire.
C. Marcus et al./Nature
That's the result described in the new Nature paper: Marcus and his colleagues demonstrated that Kitaev's theoretical predictions were correct. There is an exponential improvement in making these anyons, the further you separate the two ends of the nanowire.
And the further that separation, the more robust the system is to outside interference. "To destroy its quantum state, you have to act on both ends at the same time, which is unlikely," co-author Sven Albrecht said in a statement.
The observation of these anyons on nanowires represents a critical first step toward a working topological quantum computer. According to Michalakis, the next step is to experimentally test whether a prototype quantum device will fuse, demonstrating that braiding should be possible. Then it's just a matter of figuring out how to braid the anyons to store quantum information.
Of course, the engineering challenges involved in building a working topological quantum computer are daunting. But being able to use existing semiconducting and superconducting materials will make the path to prototype that much simpler, because physicists should be able to leverage a lot of existing technology.
Top image: S.J. Lomonaco/University of Maryland, Baltimore County