From hot silicon-dot qubits to ion traps, scientists are working on innovative ways to build more complex and powerful quantum computers that could potentially deepen our understanding of complex molecules, crack encryption algorithms, make capital markets more efficient, accelerate the development of better batteries, and even realize the promise of strong artificial intelligence (AI).
Have we achieved quantum computing supremacy? Google thinks so. On October 23, Sundar Pinchai, chief executive officer (CEO) of Google, an American multinational technology company with headquarters in Mountain View, California, published a blog post trumpeting the triumph of the company’s researchers in building a quantum computer that “performed a test computation in just 200 seconds that would have taken the best known algorithms in the most powerful supercomputers thousands of years to accomplish.”
In a separate post on the Google AI Blog, John Martinis, chief scientist of Quantum Hardware, and Sergio Boixo, chief scientist of Quantum Computing Theory, Google AI Quantum, said their goal is to build a fault-tolerant quantum computer as quickly as possible. They envision such a quantum device as being capable of spurring advances in materials design that could lead to new lightweight batteries for cars and planes, more effective medicines, and better catalysts for producing fertilizer more efficiently with fewer carbon emissions.
Details of Google’s computational feat prompted immediate pushback from other heavyweights in the quantum computing field such as IBM, an international technology company headquartered in Armonk, New York, which suggested that an ideal simulation of the same computational task on a classical system could be accomplished in just a couple of days and with much greater fidelity. IBM also critiqued Google’s use of the word “supremacy” and reiterated its vision of quantum computers and classical computers working together in a complementary way.
In any case, quantum information science (QIS) finds itself in the spotlight once again. As David Awschalom, physicist and quantum engineer at the University of Chicago, in Illinois, puts it, “Quantum information science uses the properties of nature at the smallest scales to create a meaningful technology.” The U.S. government has recognized the value of QIS with a National Quantum Initiative Act, which aims to accelerate quantum research and development for both economic and national security purposes.
Quantum may seem esoteric to many casual observers, but Jacob Taylor, assistant director for quantum information science (QIS) at the White House Office of Science and Technology Policy, notes that quantum-based technology has been in use for decades. “The atomic clocks that underpin global positioning systems (GPSs) are based on quantum theory,” says Taylor. “In medicine, we use quantum technology to power magnetic resonance imaging (MRI) machines, probing the aggregate properties of nuclei spins inside the body to find, for example, where blood is giving up oxygen.”
Researchers in academia and industry are now pushing ahead with efforts to develop more advanced technologies based on QIS. These efforts can be grouped into three main areas–computing, communications, and sensing.
Of these, Awschalom believes some of the immediate QIS applications will occur in the fields of sensing and communication. “The very fragility of quantum states is what makes them the basis of powerful sensing technologies,” says Awschalom. “Within a decade, it’s possible that advances in quantum sensing will allow us to push MRI resolution down to the level of single molecules and place sensors inside living cells to watch cellular mechanisms at work.”
The power of fragility
The same quantum state fragility that makes for a good sensor creates challenges when trying to build a quantum computer.
“The unique capabilities of quantum computers stem from two special features available in the quantum bits, or 'qubits,' that are the basic units of quantum information,” says Jungsang Kim, an electrical and computer engineer at Duke University, Durham, North Carolina, and co-founder of IonQ, a quantum computing startup based in College Park, Maryland. “One feature is the superposition principle, where the quantum bit can exist in both the 0 and the 1 state at the same time, with controllable weight and relative phase between them, until the qubit is measured. The other feature is entanglement, where correlation among several qubits is present even if the state of each qubit is not fully determined to be in 0 or 1.”
Quantum technology typically involves exotic materials and conditions in order to protect the superpositions stored in the quantum particles, explains Chris Monroe, a physicist at the University of Maryland who is also co-founder and chief scientist at IonQ.
“One of the fundamental quantum rules is that superposition only exists when you don’t look at it,” emphasizes Monroe. “This means that quantum works best in simple systems such as isolated atoms that are not part of solids or surfaces and levitated in a vacuum chamber, or exotic solid-state devices that are refrigerated to nearly absolute zero temperature.”
If those conditions are met, Monroe says that such devices can form quantum computers that have the potential to solve problems that regular, classical computers will never be able to resolve.
He gives the example of chemical modeling of complex molecules. “Consider a molecule like caffeine that has over 100 electrons,” suggests Monroe. “How do those electrons figure out where to go and what their energy levels should be? Currently, we cannot calculate the binding energy of these electrons, which means it’s hard to know how a specific molecule will interact with other molecules. Quantum computing may have huge applications for optimizing those sorts of simulations.”
Moving to a solid state
Could quantum computers ever be household or even enterprise machines? Awschalom believes that the mass production of quantum computers depends on finding ways to make qubits work in less exotic circumstances. “It’s impressive to make qubits using an atom in a vacuum or a superconductor, but we’re focused on creating qubits using solid-state materials,” he says. “There’s a trillion dollars of existing infrastructure around electronics manufacturing. If we could develop scalable qubits using semiconductors, industry could produce billions of them.”
Awschalom’s group found they could take commercial silicon carbide diodes, create defects to trap electrons, and build what he calls “surprisingly good” quantum states based on the electron spin with long coherence times and tunable quantum energies. “We’re at the proof-of-concept phase right now, but these initial results suggest that we have a pathway for scalability,” he says.
Right on the dot
If there’s any company that knows about large-scale electronics manufacturing, it’s Intel, a multinational technology company headquartered in Santa Clara, California. The company has a workforce of more than 100,000 people and builds over 10 billion transistors every second (or over 300 quadrillion transistors per year).
Anne Matsuura, director of Quantum Applications & Architecture at Intel, says that her company is leveraging its special capabilities around silicon fabrication and packaging to develop silicon quantum-dot qubits that she compares to one-electron transistors.
Intel has invested USD 50 million in a 10-year collaboration with Delft University of Technology and TNO, the Dutch Organisation for Applied Research, in order to advance quantum computing. One aspect of the collaboration involves experiments with “hot qubits”—research into whether silicon dot qubits can function at elevated temperatures. “So far, our partners have promising experimental results at 1-degree Kelvin,” says Matsuura. “Obviously, that’s still quite cold, but it’s orders of magnitude warmer than the cryogenic temperatures that are necessary today. Our hope is we can continue to increase the temperature range in which the qubits can operate.”
Intel is taking other steps to create a 300-mm, high-volume fabrication and test line for semiconductor spin qubits. Partnering with Bluefors and Afore, Intel has developed a Cryogenic Wafer Prober—or Cryoprober—that can test qubits at temperatures of a few kelvins. Intel anticipates that the Cryoprober will allow it to automate and accelerate testing on sources of quantum noise and the quality of quantum dots from weeks to just minutes.
Using Mother Nature’s qubits
While Intel is looking for ways to accelerate qubit error testing, IonQ aims to sidestep challenges in the qubit manufacturing process by not making qubits at all.
“We use ionized atoms—Mother Nature’s qubits,” says IonQ president and CEO Peter Chapman. “We don’t make atoms, so each of the qubits is perfect without any discrepancies. Floating in a vacuum, they have a fundamental degradation timescale of 10,000 years and a decoherence time that can be extended to years.”
Using electromagnetic fields, IonQ deploys and traps atomic qubits on a silicon chip within ultra-high vacuum chambers to create what the company calls the world’s first commercial trapped-ion quantum computer.
“Our technology is based on the same principles as atomic clocks,” says Chapman. “If you look back at an atomic clock from the 1950s, it would take up an entire room, but now it fits on a single chip. Our approach works at room temperature and can be built for the most part with off-the-shelf components. It’s reasonable to expect that our technology will follow the same path as atomic clocks and every other piece of electronics and will shrink over time. I would not be surprised if people were placing orders for quantum laptops in 10 years.”
And what might those quantum laptops be able to do? Chapman believes that only quantum computers could handle the explosively combinatorial requirements of natural language processing necessary for strong artificial intelligence (AI).
“If you look at the meaning of a single word in a sentence, it is bound to the context of all the other words in the sentence, the paragraph, the document, and your previous experience,” says Chapman. “Language contains a combinatorial explosion of possible meanings that quantum computers are really good at. The real world is naturally quantum. Would it really be a surprise if our intelligence also springs from the same place?”
IBM turns up the volume
While many quantum computing researchers are focused on increasing numbers of qubits, IBM is taking a somewhat different approach. In 2017, IBM announced a metric called Quantum Volume that determines a quantum computer’s performance based not just on qubit numbers, but also on a holistic assessment of various factors such as qubit coherence time (the amount of time they stay in superposition) or measurement errors that might impact the precision and accuracy of a quantum processor’s operations.
“If there is a weak point in a quantum computer system, then it doesn’t matter how wonderful your individual qubits are,” says Bob Sutor, vice president of IBM Q Ecosystem Development at IBM Research, in Yorktown Heights, New York. “Software is also important. We must be able to take user programs and transform them so that they run in an optimized way on real quantum hardware.”
IBM has put its quantum computers online through an open program and the commercial IBM Q Network that gives enthusiasts, academics, and industry researchers a chance to learn and experiment with actual quantum machines. Since the IBM Q launched in 2016, some 175,000 people have registered to use the system. IBM quantum scientists have used the Q Network to collaborate with industry partners such as Barclays, JPMorgan Chase, and Mitsubishi Chemical to try to improve the efficiency of securities settlements in capital markets, achieve a quadratic speedup in options pricing, and simulate the initial steps of the reaction mechanism between lithium and oxygen in Li-air batteries.
Codemaker and codebreaker
Using quantum computers to design better batteries seems like a noble goal, but what if bad actors try to use the processing capabilities of a powerful quantum computer to crack encryption algorithms?
Fortunately, the U.S National Institute of Standards and Technology (NIST) is on the case. “It will be a long time until quantum computers can threaten [classical] cryptographic methods by factoring large numbers, but NIST has already been working to certify post-quantum cryptography systems,” says Taylor. “We’re on track to achieve that certification by 2022, which means that chief technology officers should plan on making the transition to post-quantum cryptography protection within the next 10 years.”
Meanwhile, advances in QIS could lead to the development of encryption methods that are truly impossible to hack—at least without alerting the recipient that a message had been intercepted.
“In the quantum world, the act of looking at something changes it. That’s an advantage from a security standpoint,” explains Awschalom. “We can send information down fiber-optic networks using entangled pairs of photons. If someone tries to intercept the message and view the contents, they won’t be able to ‘put it back’ in the same state. The message will arrive scrambled and the recipient will know that someone tried to eavesdrop on the communications in transit.”
Only time will tell
Jun Ye, a physicist at the National Institute of Standards and Technology and the University of Colorado in Boulder, is advancing the measurement of time by using quantum mechanics. Using laser beams and evaporation, he loads and traps atoms one by one in a crystal lattice made of light. These new atomic clocks are about 100 times more accurate than traditional atomic clocks. Based on this new clock, in October, Ye and colleagues published a paper showing that they had developed a new time scale in the optical domain with performance 10 times better than the current generation of time scales used to define the world time.
Ye says this improvement could have direct benefits for communications and navigation, with satellites sending information to each other and to Earth using laser beams instead of microwaves. “You could send precise instructions and navigation coordinates to an interplanetary spaceship so that you can synchronize its automatic landing on Mars,” he says. “Every time we increase clock technology, it usually leads to multiple layers of technological advances elsewhere in society. In fact, the technology powering advances in both clocks and quantum computers is emerging from the same quantum revolution happening today.”
Newly offered instrumentation, apparatus, and laboratory materials of interest to researchers in all disciplines in academic, industrial, and governmental organizations are featured in this space. Emphasis is given to purpose, chief characteristics, and availability of products and materials. Endorsement by Science or AAAS of any products or materials mentioned is not implied. Additional information may be obtained from the manufacturer or supplier.