Oxford University researchers have, for the first time, generated a massive 10 billion entangled bits in silicon, taking an important step towards a real world quantum computer.
The researchers cooled a piece of phosphorus-doped silicon to within one degree of absolute zero and applied a magnetic field. This process lined up the spins of one electron per phosphorus atom. Then the scientists used carefully timed radio pulses to nudge the nuclei and electrons into an entangled state. Across the silicon crystal, this produced billions of entangled pairs.
Stephanie Simmons, researcher and lead author on the paper Entanglement in a solid-state spin ensemble — published in Nature, says that quantum computers really start to give classical computers a run for their money at a few dozen qubits, but her team is working to skip that stage altogether by going directly from a two-qubit system to one with 10 billion.
Each quantum bit — or qubit — is both a 1 and a 0 at the same time, so a single qubit can perform two calculations simultaneously. A two-qubit computer can perform four calculations: three gives you eight, four gives you 16 and so on, exponentially.
"Currently this is a set up with many, many copies of a fully functional two-qubit quantum computer. The major advantage of doing it this way is that computation is very fast, and there is no averaging to get the result out," Simmons told ZDNet UK. "But it is just a step on the way to our goal."
The next step will be to create a line of entanglement by using a burst of radiation to cause the entangled electrons from each site in the crystal to hop one place to the left.
"Think of it like line dancing," Simmons said.
Once the team has done this, they can test how quantum information survives this stage before moving on to hopping those same electrons one site forwards to create a two-dimensional array of entanglement, which would, in effect, create an instant 10-billion qubit quantum computer.
A quantum computer has the potential to vastly outperform a traditional computer because it takes advantage of a quantum phenomenon called superposition — the ability of quantum level particles to exist in a variety of states simultaneously — to make calculations in parallel.
Most famously, this could be put to use in cracking currently impenetrable cryptography where huge numbers need to be factored to their original primes. Brute forcing these calculations in sequence would not give a result inside a human lifetime, but apply the inherent parallelism of a quantum computer, and they would fall to pieces in minutes.
But getting the results of a calculation is very difficult, because to measure a quantum system is to destroy the superposition that makes it such a powerful computing force. This is where entanglement comes in, measuring the state of one of an entangled pair to learn about the other, without actually looking at it.
Once quantum computers are out there, who knows what people will come up with? No one thought we'd need many traditional computers.– Stephanie Simmons, Oxford University
Researchers have struggled to produce entangled pairs in silicon, the bedrock of traditional computing, but this breakthrough means the component of a quantum computer could be manufactured using existing fabrications.
"It doesn't matter about the temperature because no one is talking about adding a quantum computer to your laptop," Simmons said, referring to the extreme cold required to make the process work. "These would be servers which people could access via the internet. And once they are out there, who knows what people will come up with? No one thought we'd need many traditional computers, after all."
"Definitely the goal is still a long way off, but what's really encouraging is the recent pace of progress in silicon," John Morton, co-author of the paper and head of Simmons' group, said. A silicon-based quantum computer was first proposed in 1998, but there have been major breakthroughs in the past three years that have really given this approach a lot of momentum, and so it's something that people are becoming increasingly excited about.
"What I think we're seeing is most of individual elements of what's needed for a silicon-based quantum computer coming into place, and future research needs to focus on getting them working at the same time in the same chip," he said.