Google recently announced a major breakthrough in quantum computing when its Willow processor solved an equation that would take a conventional computer practically forever. But PsiQuantum, the company planning to build in Chicago one of the world’s first commercially viable quantum computers, is taking a different path.
Instead of building a small-scale quantum computer designed primarily for testing and public demonstrations, PsiQuantum is shooting for the moon with a 1 million-quantum-bit machine capable of tackling practical, real-world applications.
It’s a future that even leading scientists aren’t sure is possible. But PsiQuantum co-founder Pete Shadbolt, in Chicago recently to share his plans, intends to prove that it is.
“It’s like breaking the sound barrier,” Shadbolt said, “… from having systems that are really toys to systems that are commercially valuable.”
Quantum computing attempts to harness the power of quantum mechanics to perform calculations. While a conventional computer operates with binary digits, or “bits,” with a value of 0 or 1, quantum bits, or qubits, use particles that can hold 0 and 1 values simultaneously, like a coin flipping in the air, exponentially increasing its computational power.
This summer, Illinois Gov. JB Pritzker announced that the state would provide $500 million to establish the Illinois Quantum and Microelectronics Park at the former U.S. Steel South Works plant, a site that has sat vacant since closing more than 30 years ago.
PsiQuantum is to be the anchor tenant, along with a joint state and U.S. Defense Advanced Research Projects Agency (DARPA) facility to test quantum technology prototypes. Earlier this month, IBM announced plans to build a National Quantum Algorithm Center on the site, with help from a $25 million state grant. IBM also has a $100 million joint venture with the University of Chicago and University of Tokyo to build a 100,000-qubit quantum-centric computer.
If successful, quantum computing is expected to revolutionize development of new medicines and materials and address problems like climate change. But huge engineering obstacles remain to be solved before any practical applications can be made.
How long that will take, and whether it’s even possible, remain unknown. The IBM project, for instance, is planned to take place over the next decade.
Like PsiQuantum, IBM and NVIDIA leaders have also made statements that eventually, they, too, will need 1 million qubits or more.
With that goal in mind, PsiQuantum hopes to break ground in early 2025 on its computer facility in Chicago. One of the first jobs would be for the state to fund, out of its $500 million commitment, a $200 million cryogenic plant to create liquid helium to cool the computer to nearly absolute zero, deep space conditions. PsiQuantum plans for a dedicated transmission line from ComEd to provide large quantities of carbon-free power for the site.
So far, PsiQuantum is testing one intermediate quantum system in England and a larger one at Stanford University. It also plans to build a larger system in Chicago and its first commercial device in Brisbane, Australia, followed by a similar utility-scale, fault-tolerant device in Chicago.
While building two sites would divide its resources, Shadbolt, his company’s chief science officer, believes it gives PsiQuantum advantage by being able to test variations in their systems and have the sites feed off each other.
To create its qubits, PsiQuantum plans to use silicon quantum chips and tiny units of light called photons.
One classic demonstration of quantum mechanics involves sending beams of light through two slits, creating an interference pattern, much like waves of water. Strangely, even when photons are sent through the slits one at a time, they seem to interfere with themselves, revealing their seemingly dual nature as waves and particles.
This phenomenon, known as superposition, is harnessed in quantum computers to perform multiple calculations simultaneously. Another key phenomenon, entanglement, enables quantum computers to link qubits in ways that dramatically enhance computational power and efficiency.
One of the greatest challenges facing any quantum computer are errors arising from qubits interacting with their environment. Significantly, Google’s recent announcement of Willow’s success with its 105 qubits included news that as the number of qubits increased, errors decreased.
Google conceded that its computer can’t yet solve real-world problems beyond the range of a conventional computer but called its creation “the most convincing prototype” and “a strong sign that useful, very large quantum computers can indeed be built.”
Google also issued a challenge to its competitors, saying that any team building a quantum computer should check first if it can beat classical computers on a standard industry benchmark called random circuit sampling. If it can’t, Google said, “there is strong reason for skepticism that it can tackle more complex quantum tasks.”
The acting director of the quantum park, Harley Johnson, a leading mechanical engineering professor from the University of Illinois Urbana-Champaign, previously issued a statement that spoke glowingly of plans for the site.
“The potential for our work in the park to change the world is drawing comparisons to historic tech initiatives like the Manhattan Project or the development of Silicon Valley,” Johnson said.
PsiQuantum’s reputation for secrecy has left some outside observers wondering how things are going. Aram Harrow, a professor at the MIT Center for Theoretical Physics, said it’s difficult to know.
“The big uncertainty about them is that we haven’t seen a lot of demonstrations, so it’s hard to know how far along they are,” Harrow said.
The photonic approach is faster and minimizes noise better, requires less cooling than other superconducting systems and uses some elements already developed by the telecom industry. It also poses difficulties in getting interaction among photons, Harrow said.
“You absolutely need to test with small steps,” he said. “Especially because this is a technology that hasn’t really been used by other groups. There’s certainly a lot of reasons why it’s promising, and why it’s hard.”
PsiQuantum shares its progress with investors, attracting $700 million in venture capital, and says it’s meeting some of its goals for components and system integration and will work until it meets myriad remaining tests and benchmarks.
One reason PsiQuantum chose Chicago is the rich academic and research environment of the Chicago Quantum Exchange, a consortium including area universities, Fermilab and Argonne National Laboratory, and their potential workforces.
Prof. Michael Wasielewski, director of the Institute for Quantum Information Research and Engineering (INQUIRE) at Northwestern University in Evanston, applauded PsiQuantum for taking a different approach than others and realizing the need to go big.
Northwestern will participate in the quantum park, hoping for a facility there that academics and startups could use to do research, to make the region the leader in the field.
Facing one of the hardest endeavors humans have attempted, Shadbolt said, PsiQuantum was attracted to Chicago by past local achievements, like the first nuclear reaction and reversing the flow of the Chicago River, and the academic and engineering firepower of the area.
“There’s an amazing attitude of… getting things done,” he said. That’s worth its weight in gold. The people, the culture, the technical background are all huge draws for us.”
The son of a wastewater engineer in England, as a boy Shadbolt wanted a Super Nintendo. When his parents instead made him learn computer programming to play video games, he said, he came to appreciate the value of working meticulously toward a practical goal.
“We’ll go as fast as humanly possible,” he said, “towards a machine that actually makes money and is actually useful.”