Boffins use silicon to push quantum computing toward reality
24 October 2014 | 0
in Australia have developed silicon-wrapped quantum technology that could solve problems that have held back the development of powerful quantum computers.
The scientists, working on similar but separate projects at the University of New South Wales (UNSW), used silicon as a protectant shell around the bits, also known as qubits, in a quantum machine.
By doing that, they’ve made the qubits, the building blocks for quantum computers, more accurate, increased the length of time they’ll hold information and possible made quantum computers easier to build.
With these two similar silicon-based breakthroughs, the researchers potentially have pushed quantum computing closer to becoming a reality.
“This has shown that quantum bits in silicon are a truly viable option for quantum computers,” said Andrea Morello, a quantum physicist and associate professor at UNSW. “What’s exceptional is how the perception of what we are doing has changed over the years. Ten years ago, the things we are doing now were really seen as borderline science fiction.”
Morello is one of two lead researchers at UNSW who have been working on new ways to create qubits.
Andrew Dzurak, a professor of electrical engineering at the university, and Morello both make quantum bits based on a single electron confined and localized in a silicon chip. Once the electron is in that confined state, it can be used to encode quantum mechanical information.
The difference in their research projects is that they use two separate methods to confine and localise the electron.
Dzurak’s method is to create a qubit with a device similar to the transistors used in laptops, smart phones and other electronics. Instead of using a transistor, Morello’s team uses a natural atom of phosphorous as a qubit.
The scientists simply took parallel paths to using silicon in a quantum machine.
“They both could work,” Dzurak told Computerworld, adding that both scientists are in the early stages of discussions with computer and manufacturing companies.
“We expect that the interest from the semiconductor industry will drive the progress even further,” Morello said. “There’s the technical aspect of what we’ve done, but there’s the perception aspect. The vibe in our community is extremely excited.”
Germano S Iannacchione, head of the Physics Department at Worcester Polytechnic Institute, called the research in Australia an exciting advance for quantum computing.
“From an inspirational point of view, this is the kind of advance that shows us we’re not just spinning our wheels,” Iannacchione said. “It’s a beautiful idea. This is bordering on enabling something that could actually be useful … That’s why this is so exciting. If you could scale this up, there are huge problems that we could tackle.”
Quantum computing is regarded as the holy grail of high-performance computing. It holds a great deal promise but remains elusive.
Classic versus quantum computing
Computer scientists and physicists say a quantum computer could surpass the top classic supercomputers in solving problems that involve analysing huge quantities of data. The hope is that quantum computers would find answers to problems so complex it would take supercomputers like IBM’s Blue Gene and Cray’s systems hundreds of years to solve them.
Other practical applications for quantum computers include cancer and Alzheimer’s research, advances in cryptography, and the hunt for distant Earth-like planets. They also could be used to simulate political and military situations, such as the unrest in the Middle East, enabling researchers or a government to test different options to see how they would affect the outcome.
Classic computers use bits — ones and zeroes — for processing instructions, and they work in a straightforward manner. Ask the computer a question, and it will move through the calculation in a linear, orderly way.
A quantum computer, however, combines computing with quantum mechanics, one of the most mysterious and complex branches of physics. Some of the world’s top physicists say they don’t understand how it works.
The field was created to explain physical phenomena, like the odd actions of subatomic particles, that classical physics fails to do. With quantum computing, it’s about the possibilities.
Unlike a classic computer, which uses ones and zeroes, a quantum machine uses quantum bits, or qubits, that can be both a one and a zero. It does not work in an orderly or linear manner. Instead, its qubits communicate with each other and calculate all the possibilities at the same time.
If a quantum machine has 200 qubits, it is calculating at 2 to the 200th power at the same time.
Quantum of doubt
In the computer science and the physics communities, there is contention over whether a quantum computer has actually been built.
D-Wave Systems Inc., a Burnaby, British Columbia-based quantum computer company, claims that it has built quantum computers, using its own quantum processor built with different metals, including niobium, a soft metal that becomes superconducting when cooled to extremely low temperatures.
One of the company’s machines, the D-Wave Two, is being tested by Google and NASA.
The disagreement involves whether the D-Wave machines are performing in full quantum states and if they provide any real speedup over traditional machines.
“The type of quantum machine that D-Wave has built is not what the broader quantum community would term a quantum computer,” said Dzurak, who focuses on nanoelectronics and quantum computing. “The broader research community is trying to develop a type of quantum computer in which one has greater control of the quantum bits. The D-Wave machine is designed to solve a particular class of problems, not the full class of problems that a real quantum computer could solve.”
Great control over the qubits is exactly what the Australian researchers are trying to provide. Both Morello and Dzurak said their methods of creating qubits have made calculations more accurate.
“You might be calculating 3+4, and you might get 8. That’s an error,” explained Morello. “The number encoded on the bit is not what it was intended to be because something went wrong in the operation. If the error is rare enough, you can correct it on the run. With greater accuracy, you can start to design larger quantum computers because you have the ability to correct the errors.”
Since using silicon in the qubits gives them a longer life, the machine has more time to correct any errors. The researchers have stretched the world record for the longest-lasting qubit from two seconds to 30.
“Two years ago, 50% of the time the calculation was wrong,” said Dzurak. “You weren’t going to go very far with that. The accuracy now has gone from 50% to 99%. This is a game changer. It goes from being impressive science to a serious manufacturing technology that can be taken forward.”
Using silicon and transistors that are similar to traditional transistors, also means that the qubits should be able to be manufactured in a traditional microprocessor plant without too much modification to the process.
Sharon Gaudin, IDG News Service