Read more about China in our special issue
THIS is where the revolution might begin. One night five years ago, Jian-Wei Pan had a vivid dream. Sitting in front of him was the world’s first quantum computer. It was a strange mix of translucent solid and swirling fluid with a piercing beam of blue light emanating from within. He approached it, trying to figure out how it worked, but it was too fuzzy to examine. Nor could he imagine where it had come from. Did it have a “Made in China” sticker on it? “I have no idea,” Pan laughs.
In his lab at the (USTC) in Hefei, Pan is now striving to make his dream a reality. When I visit, it is buzzing with activity. In an office thick with cigarette smoke, young professors argue over their latest experiments, while in the next room students scurry around a vast optics bench, adjusting lasers, lenses and detectors. If all goes to plan, the first quantum computer worth shouting about will be built right here.
Advertisement
Hefei is in some ways an improbable birthplace for such a revolutionary device. Hours away from the tech powerhouses of Shanghai and Guangzhou, this quiet town in south-east China has till recently been better known for its traditional tofu and sesame cakes than cutting-edge physics. Pan and his colleagues are changing that, and have already placed the USTC – and the country as a whole – firmly on the quantum computing map.
“Hefei was best known for its tofu and sesame cakes. Now the town is on the quantum computing map”
If a working quantum computer can be devised, it promises to easily eclipse the power of today’s fastest supercomputers. With that glittering goal in mind, a plethora of groups around the world are racing to build one. Pan’s approach is different from the rest. What makes it particularly promising is that it combines a quantum memory with a new architecture known as cluster states.
Pan believes his team’s method could be scaled up to perform useful calculations more reliably and easily than any other scheme devised so far. The competition is tremendous, but researchers who have worked with Pan wouldn’t bet against him producing the goods. “His group is one of the leading groups worldwide in developing quantum technology,” says physicist Caslav Brukner at the University of Vienna, Austria.
As an idea, quantum computing has been around since the early 1980s. It envisages harnessing the weird properties of quantum mechanics to perform tasks that classical computers cannot do in a reasonable time, such as searching large stores of data, or factoring the huge numbers that underlie today’s most secure encryption schemes. The power of a quantum computer comes from the fact that a quantum particle can exist in more than one state at a time. So unlike a data bit in an ordinary computer, which can have the value of either 0 or 1, a quantum bit (or qubit) can simultaneously have the value 0, 1 or any “superposition” of the two. So perform a calculation using qubits and you get a huge number of calculations for the price of one.
Foreign origins
Things get even more interesting when you add the quantum phenomenon known as entanglement, which can link the properties of several qubits. With only a few hundred entangled qubits, it is in principle possible to represent more numbers than there are atoms in the universe. A quantum computer that uses entangled qubits could perform different sets of calculations on a huge number of inputs all at once. At least that’s the theory. So far, quantum computers’ practical mathematical abilities are no better than those of the average 10-year-old.
Pan began learning about the practical side of quantum computing in 1994, as a novice graduate student with Anton Zeilinger, who was then at the University of Innsbruck in Austria. At the time, virtually no experimental quantum research was happening in China. Not long after Pan arrived in Innsbruck, Zeilinger asked him: “What’s your future plan?” Pan says he didn’t hesitate: “I want to have a lab in China like yours.”
Soon afterwards, he began building ties with researchers in his homeland. Starting in 1997, he returned to China once a year to give lectures and meet students and faculty at various universities, including USTC. In 2001, with three Nature papers under his belt, he won a $250,000 grant from the Chinese Academy of Sciences to spend three months a year establishing quantum information research in China – the first grant of its kind. By 2003, he was splitting time equally between his labs at the University of Heidelberg in Germany and at USTC.
In the meantime dozens of groups around the world, including Zeilinger’s in Austria, had been pursuing various approaches to quantum computing using photons, trapped ions, superconductors or quantum dots as their qubits. Pan’s experience of using photons for quantum communication has convinced him these are the way forward.
Each approach has its merits, but there is one problem that they all face: the entangled states crucial for quantum computation are horribly fragile. The hardest part of manipulating qubits is keeping the entanglements intact and forming new ones as needed during the course of the calculation.
Cluster states are an attempt to solve this problem. Instead of performing multiple operations over time on a given set of qubits, each step of the calculation has its own set. This has the advantage that you don’t need to manipulate the entanglements during the calculations and thereby risk destroying them. Instead, the entanglements are prepared at the start and then left alone (Â鶹´«Ă˝, 25 March 2006, p 42). But this technique has drawbacks too. Many more qubits have to be entangled before the calculation starts, and these entangelments have to be preserved for longer than is needed with other approaches.
Pan and his team are now working to overcome these problems, and in May they reported the using only two photons. It’s a promising start: other groups have created four-qubit cluster states but these tend to use more photons or atoms, which could be far trickier to handle in large numbers. What’s more, Pan’s method seems to create cluster states more reliably.
Scaling up nevertheless presents a daunting technical challenge, as it involves keeping track of so many photons. “It’s like trying to kill six birds with one stone,” says Zeng-Bing Chen, a colleague of Pan’s at USTC.
That’s where quantum memory comes in. Pan knew from his work on quantum communication that you can use an isolated group of cooled atoms to absorb the quantum state of a photon that hits them. The atoms can collectively store this state for a few microseconds before transferring it to another photon. In June Pan’s group reported and they hope it could eventually provide a simple way to entangle a network of qubits.
Their ultimate goal is to use this type of storage to set up entanglements of 100 or more qubits, which would be enough to start doing practical quantum computing. The quantum memory would help keep track of the photons’ states until the exact moment they are needed.
Don’t expect this to happen tomorrow, though. Pan warns that it may take more than 10 years to achieve a useful result. But fast forward a decade or two and the shiny new computer you are unpacking may look unlike any other PC you’ve ever owned before. Except for one little thing: the “Made in China” label on the box.
Quantum World – Learn more about a weird world in our comprehensive special report.
Back to their roots
Jian-Wei Pan is emblematic of a generation of young Chinese researchers who are returning to their homeland after establishing themselves in the west. Pan did his postgraduate training in Anton Zeilinger’s prestigious group in Austria, and then set up a lab at the University of Heidelberg in Germany (see main story). He returned to China full-time earlier this year.
Attracting star Chinese researchers back from abroad is part of the government’s strategy for building up the nation’s science and technology. Computer scientist Andrew Yao and , who is now in his 80s, have both been wooed back in the past three years. Yao specialises in the theory of cryptography and communication and won , computing’s equivalent of a Nobel prize, in 2000. Both now have labs at
“Ten years ago it was very difficult to attract people with permanent positions in the US or Europe to China,” says Hou Jianguo, executive vice-president of the University of Science and Technology of China (USTC), where Pan now has his lab. “Gradually that has changed.”
In 1996, the Chinese Academy of Sciences began its “” programme, offering lucrative research grants to lure top professors in all areas of science back to China. Having achieved its initial goal of bringing back 100 top researchers by the year 2000, the programme has gone on to boost the number of elite returnees to more than 1000.
According to the Ministry of Science and Technology, a total of more than 20,000 researchers have been attracted back to China in the past 10 years, via more than 100 funded channels. “It is getting more competitive here,” says Tieniu Tan, the academy’s deputy secretary general. “Salaries are still not as good as abroad, but many new PhDs are returning.”
A returning faculty member can expect to earn anywhere between half and two-thirds as much as they would in a comparable position in the US or Europe, but because the cost of living is generally much lower than in the west, this is enough to achieve a standard of living which is, if anything, higher.
Scientists setting up their own research groups can steal a march on competitors in the west because of the much lower wages expected by junior staff in China. Pan says graduate students at USTC earn less than $200 a month – only 12 to 15 per cent of what their European or American counterparts would be paid. This is one of the factors that is allowing relatively new Chinese research labs to catch up with established ones elsewhere. “The whole level is going up fast,” Pan says.
