Harnessing the power of nature and digital technologies, researchers test new qubit-powered chips to pave the way for quantum computing’s future.
Even to the world’s top minds, quantum mechanics seems like magic.
Albert Einstein called entanglement and superposition — when the measurement of one tiny electron affects the state of another electron at a distance — “spooky.”
Richard Feynman, a Nobel Prize-winning physicist and one of the first people to envision applying quantum mechanics to computing, famously quipped, “I can safely say that no one understands quantum physics.”
That hasn’t stopped researchers and technology innovators from using the magical forces of quantum mechanics to build quantum computers that are exponentially more powerful than today’s highest-performing supercomputers.
Researchers at QuTech in the Netherlands and Intel are racing to engineer qubit crunching computer chips and specialized systems that use quantum mechanics. Focused on very small atoms and subatomic particles, this branch of physics may unlock the mysteries of nature, manage the complexities of digitally connected societies and maybe even solve some of the biggest problems facing humankind.
Quantum computing uses the strange ability of subatomic particles to exist in more than one state at any time. It has quickly evolved from a surprising scientific discovery to real-world applications, according to Lieven Vandersypen, co-director of the Kavli Institute of Nanoscience and lead scientist at QuTech, a research institute that has a $50 million research partnership with Intel.
“We are thinking in very practical ways about how to use superpositions for tasks that are otherwise simply impossible,” said Vandersypen.
He said quantum computing can lead to extremely fast computing, very sensitive detectors and the ability to create unbreakable encryption to protect important digital information.
“A quantum computer has a power that actually matches the complexity of the molecules or materials that we would like to understand,” he said.
How Quantum Computing Works
Today’s classical computers are based on bits, which encode information as a series of ones and zeros. Quantum computers use quantum bits (or qubits), each of which can represent multiple states: a one, a zero, and arbitrary combinations of one and zero.
“When you flip a coin, it lands either heads or tails. That’s binary,” explained James Clarke, director of Intel’s Quantum Hardware. “Now think of spinning that coin on its edge. Is it a one or a zero? It’s actually both at the same time.”
This is what separates classical from quantum computers, said Clarke. Today’s computers can break up several instructions and crunch them almost simultaneously, but quantum computers can process many more instructions in parallel.
Qubits’ ability to represent so many states creates a need for a fundamentally different kind of approach to computing, said Clarke.
Clarke’s team manufactured a 17-qubit superconducting chip, which is being tested at QuTech. Fitted inside a sub-zero chilled cylindrical computer dangling from the lab ceiling, the chip facilitates superposition and entanglement so the varied states of superconducting circuits can act on instructions much faster than transistors used in classical computers.
As more qubits are entangled within a quantum computer system, the number of states increases exponentially. For example, 50 entangled qubits would simultaneously represent two to the 50th power states, which is in the quadrillions, said Clarke.
“That’s a number so large that no supercomputer on Earth could represent it,” he said. “If you had 300 qubits, you could represent more states than there are atoms in the universe.”
Qubits are finicky little things, explained Vandersypen. Any noise or vibration can cause qubits to change states, thus leading to processing errors.
“The art is to create an environment that’s very well controlled so there are almost no fluctuations that can influence the qubit’s state,” he said.
The test chip is physically quite small — about 5 centimeters (almost 2 inches) — but the surrounding apparatus is quite large in order to keep qubits at the very low temperature of 20 millikelvin. That’s 250 times colder than deep space.
These so-called dilution refrigeration systems appear menacing at first sight.
“They look like pods, like what you would see in a science fiction movie, where they are hatching robots,” said Clarke.
The chip is located in the coldest part of the refrigerator and controlled via a circuit board with microwave connectors. To activate the qubits, researchers program high-level commands that are translated into microwave bursts. This causes the qubits to spin and interact with each other. The interaction produces the nearly limitless parallel processing of quantum computing.
Quantum Leap into the Real World
Working with Intel, Vandersypen and his team designed quantum computers that maintain perfect conditions needed to test and correct errors in the system. Their efforts could help pave the way for wider accessibility to quantum computing, so younger generations can experiment and build applications in university labs.
“They’ll be able to explore theories that they’ve never been able to fully explore before,” said Vandersypen. Quantum computing could allow students to better understand molecules and materials that may seem irrelevant today but could lead to new, valuable applications.
The representation of so many states makes quantum computing well suited for artificial intelligence (AI) and cryptography, according to Lawrence Freeborn, research manager at IDC Financial Insights. He said AI applications used in digital assistants like Amazon’s Alexa could benefit from quantum computing.
“Today, it’s very easy to trip up these voice-activated assistants because they are operating with limited data,” he said. Quantum computing could help digital assistants respond more like humans.
It could also lead to unbreakable encryption, vastly improving information security. Cryptography involves factoring large numbers into prime numbers, a computation in which quantum computers excel. In fact, Vandersypen is a renowned pioneer in quantum factoring, which is a big reason why Clarke sought him for the project.
Other possible applications include drug design and materials science (modeling complex structures in nature like proteins) and logistics optimization (finding the most efficient of any number of possible travel routes).
While QuTech researchers run Intel’s 17-qubit chip through a series of tests, Clarke said his team is refining its design and planning new chips made with more qubits. He said Intel is also exploring different types of qubits, including ones that are made from silicon and are similar to the transistors found in today’s Intel microprocessors.
The ultimate goal is to build a commercial quantum computer that governments and industries can use to tackle big data challenges. Clarke said that requires at least a million entangled qubits, plus hardware and software architecture to manage the processing.
By the time such computers hit the market, consumers may not be so baffled by quantum physics, said Vandersypen. Younger people tend to be more accepting and uninhibited when it comes to thinking big.
“My children will have less trouble understanding quantum than my 75-year-old father,” he said. “They might not be so surprised if you tell them that an electron is in two places at once. In fact, I said that to my children and they just nodded in agreement.”
Editor’s note: Just two months after delivery of the 17-qubit superconducting test chip, Intel unveiled Tangle Lake, a 49-qubit superconducting quantum test chip. Achieving a 49-qubit test chip is an important milestone because it allows researchers to assess and improve error correction techniques and simulate computational problems. Learn more about how Intel is advancing quantum computing technologies.