The researchers, led by John Martinis built a superconducting quantum circuit that consists of nine qubits. Each qubit is a small circuit consisting of a capacitor and a non-linear inductor (a so-called Josephson junction), which acts as an artificial atom. The materials making up the ensemble are an aluminium film evaporated onto a sapphire substrate.

"Our nine-qubit system can protect itself from bit errors that unavoidably arise from noise and fluctuations from the environment in which the qubits are embedded," explains team member Julian Kelly. "We also show that 'more is better': nine qubits protect the system better than five qubits, a critical requirement when going to more qubits in a real quantum computer of the future."

Quantum states are fragile

Quantum computers will work on the principle that a quantum particle can be in a superposition of two states at the same time – "spin up" or "spin down" in the case of an electron, for example. The two states represent a logical "1" or a "0", so N such particles – the qubits – could be combined, or "entangled", to represent 2N values simultaneously. This would lead to the parallel processing of information on a massive scale not possible with conventional computers.

Such quantum machines would perform much better on certain tasks, like machine learning or simulating complex molecules, for example, because their processing speeds should increase exponentially with the number of qubits of information involved. In practice, however, physicists have struggled to create even the simplest quantum computer because the fragile nature of these quantum states means that they are easily destroyed and are difficult to control.

Measuring parity

"In quantum mechanics, we cannot measure a qubit without destroying the superposition and entanglement that makes quantum mechanics work," says team member Rami Barends, "but we can measure something called parity – which forms the basis of quantum error correction."

The researchers exploited this fact and repetitively measured the parity between adjacent "data" qubits by making use of "measurement" qubits. "Each cycle, these measurement qubits interact with their surrounding data qubits using quantum logic gates and we can then measure them," Kelly tells "When an error occurs, the parity changes accordingly and the measurement qubit reports a different outcome. By tracking these outcomes, we can figure out when and where a bit error has occurred and correct for that."

Identifying and correcting for errors

A larger number of qubits give us more information to identify and correct for errors, adds team member Austin Fowler. "Errors can occur at any time and in all types of qubits: data qubits, measurement qubits, during gate operation and even during measurement. We found that a five-qubit device is robust to any type of bit error occurring anywhere during an algorithm, but a nine-qubit device is better because it is robust to any combination of two bit errors."

Although still a long way off from real-world applications, the researchers say that a "self-correcting" device like theirs could be a great platform for testing some of the ideas behind error correction – such as protecting a quantum state against so called phase-flip errors. "We are also now busy improving the quality of our qubits and the materials we used to make them," says Kelly.

The research is detailed in Nature doi:10.1038/nature14270.