In 2001, a team of physicists from IBM and Stanford University in Silicon Valley revealed that they had built a remarkable computer capable of exploiting the strange rules of quantum mechanics to process information.
This quantum computer was designed to factorise numbers, a problem that conventional computers have particular trouble with. The team proudly demonstrated it by finding the two prime factors of the number 15 (3 and 5, in case you were wondering).
That was an impressive feat. It was possible because a quantum object can exist in two states at the same time, representing a 0 and 1 simultaneously. This kind of superposition allows one quantum object to compute with 2 bits simultaneously, two quantum objects to compute with four bits simultaneously, eight quantum objects to compute with 256 bits and so on.
The IBM/Stanford had just seven qubits at its disposal. But the promise from these kinds of devices is huge: a computer with just 30 qubits would be more powerful than any existing conventional computer.
But in the ten years since then, nobody has built a quantum computer much more powerful than this. How come?
The IBM/Stanford machine worked using a technique called magnetic resonance. The idea is to find a molecule that contains atomic nuclei that can be made to spin up or down at slightly different energies. This allows each nuclei to be addressed separately using the technique of magnetic resonance.
This involves placing them in a powerful magnetic field, zapping them with radio waves and then listening for the echo. (Anyone who has had an MRI scan will have had the same treatment.)
The technique works with all kinds of molecules such as acetone, caffeine and even alcohol, although the IBM/Stanford team used an exotic molecule known as a perfluorobutadienyl iron complex to get their seven qubits. And it also works at room temperature, which is handy.
But here’s the thing. The signal from a single molecule is too weak for this technique to pick up so you have to use a whole cupful of molecules to do the calculating. And that places severe limits on the scalability of the technique.
Using bigger molecules to increase the number of qubits dramatically reduces the signal you can pick up from each qubit. So the technique of magnetic resonance using a cupful of molecules just doesn’t work for many more than a handful of qubits.
That’s why physicists have been stuck for so long. Nobody knew how to increase the number of qubits, until now.
Today, Mike Grinolds and buddies at Harvard University say they’ve cracked the problem. And the way they’ve done it is to shrink the business end of a magnetic resonance machine to the size of a pinhead. (If you’ve ever seen a magnetic resonance machine, you’ll know what a feat that is.)
They’ve done it by placing a powerful magnet at the scanning tip of an atomic force microscope. In this way, they can create a powerful magnetic field gradient in a volume of space just a few nanometres across. That allows them to stimulate and control the magnetic resonance of single electrons.
They’ve tested their device on so-called nitrogen vacancies in diamond. These are created by burying single atoms of nitrogen in thin sheets of diamond. Quantum physicists are fascinated with these vacancies because they are well protected from the outside world and so stable, and are easy to see by the photons they emit.
These vacancies can also be placed close together so they can interact with each other, a crucial requirement for quantum computers because it allows the creation of quantum logic gates with more than one set of inputs and outputs.
But such gates will only work if the electrons in the vacancies can be manipulated in the right way.
That’s exactly what the new magnetic resonance technique allows: the manipulation of electrons in a way that could easily be adapted for quantum computation.
Grinold and co say this has “intriguing potential applications ranging from sensitive nanoscale magnetometers to scalable quantum information processors.”
That’s an exciting breakthrough. Nitrogen vacancies in diamond are well studied in many labs round the world and atomic force microscopes are fairly standard pieces of kit. Add that to the fact that the first large-scale quantum computer will almost certainly win its owner a decent prize and you’ve got all the ingredients for a humdinger of a race.
And that’s without mentioning the numerous other runners: ion traps, quantum cavities, superconducting qubits and optical logic gates and the like.
In the quantum computing steeplechase, it once seemed that magnetic resonance had fallen at the first. Now it’s back in the running again and chasing down lead.
Ref: arxiv.org/abs/1103.0546: Quantum Control Of Proximal Spins Using Nanoscale Magnetic Resonance Imaging
You can now follow The Physics arXiv Blog on Twitter