Once they are ready for commercial deployment, quantum computers are expected to bring about massive disruption and create enormous value across a broad range of industries. Quantum computing’s unique ability to simulate the chemistry underpinning all human activity means it could help achieve breakthrough innovations in carbon capture, new fuels, batteries, fertilizers, catalysts, and more. Jeremy O’Brien, CEO and cofounder of PsiQuantum, speaks with McKinsey’s Philipp Hillenbrand about his company’s approach to accelerating and scaling the technology and its bold vision to deploy it in the fight against climate change.
Key insight #1: Rapidly accelerating the development timeline for a commercially viable quantum computer requires a fundamentally different approach.
Philipp Hillenbrand: Could you elaborate on the different technologies underlying quantum computing and why PsiQuantum has placed its bets on a photonics-based system from the beginning?
Jeremy O’Brien: The first thing you need to understand is that all known useful applications of a quantum computer require error correction, and therefore something on the order of a million quantum bits, or qubits. Breakthroughs to date involve systems with up to around 100 qubits, so there’s a big gap. From the start, PsiQuantum has been exclusively focused on building a quantum computer capable of addressing commercially useful applications. My conviction for more than 20 years has been that for such a machine to become a reality in my lifetime, we would need to leverage the same advanced semiconductor manufacturing techniques that put a billion transistors in your cell phone.
The challenge is that semiconductor foundries are very constrained in the materials and devices that they can build. Qubits often require millikelvin temperatures, atomic-scale fabrication, or exotic materials that are not compatible with semiconductor manufacturing.
Our team has created an architecture based on photonic qubits, which avoids these more difficult requirements and allows us to use silicon photonics—a technology that has been developed over the past 25 years, principally by the telecom industry but increasingly for other applications. We can generate, manipulate, and measure qubits using standard components that already exist in commercial products. This approach massively accelerates our timeline to a million qubits.
Beyond the potential to manufacture large numbers of qubits, photons have significant advantages when it comes to scaling. Current quantum computing systems are hamstrung by four major challenges: cooling power, control electronics, connectivity, and testing. Our photonic architecture uniquely addresses all of these challenges and supports our ability to rapidly iterate toward a working machine.
These advantages are the basis for my long-held belief that photonics is really the only approach that can reach the necessary scale for fault-tolerant quantum computing on any practical level of time or money.
Key insight #2: Quantum computing applications, once thought to be decades away, could now happen within the next ten years.
Philipp Hillenbrand: What would you say to a skeptic who believes quantum computing will forever remain in the realm of science fiction and not become something of practical value?
Jeremy O’Brien: I would say that you’re right to be skeptical, given the immense challenge. If we hadn’t cracked the problem of using the production lines of a world-leading semiconductor foundry to manufacture the chips, I would still be telling everyone today that quantum computing is decades away. But whereas we used to build this stuff in a research lab, we are now building it in the production lines of the semiconductor foundry, shoulder to shoulder with the chips that are in your laptop and cell phone. It is rapidly becoming a mature technology. We have simulated the architecture in fine detail, and there are no fundamental technical obstacles. We have demonstrated all of the building blocks, such as entangling gates, small-scale algorithms, and so on.
In 1995, I first learned that quantum computing might bring about a revolution akin to the agricultural, industrial, and digital revolutions. Back then, it seemed far-fetched that quantum mechanics could be harnessed to such momentous effect. But today, PsiQuantum is seeing great interest coupled with a high level of sophistication from Fortune 500 companies that are working with us to understand how they will deploy quantum computing to drive major advancements across a wide range of applications and use cases. They are doing this now to ensure that they have access to this profoundly world-changing technology when it comes online, thereby enabling a first-mover advantage in what promises to be a winner-takes-all type of dynamic.
All this gives me tremendous confidence that we will be able to achieve useful applications within this decade.
Key insight #3: Quantum computing will revolutionize chemistry, enabling breakthrough innovations and advancements in low-carbon technologies.
Philipp Hillenbrand: Where do you see quantum computing being able to make a real difference in the field of sustainability?
Jeremy O’Brien: This is where things get really interesting. In short, quantum computing will revolutionize chemistry. And, if we can—as I am now convinced—build a million-qubit quantum computer in time, that’s great news for our climate.
Why? So many low-carbon technologies involve complex systems, particularly around chemistry and materials science, which nobody fully understands. Everyone is scrambling to find a new catalyst or electrolyte that will give us cheaper carbon capture or better electric batteries. Right now, we have to test thousands of molecular combinations, which means lengthy and hugely expensive trial-and-error lab experiments, with often disappointing, marginal improvements.
When it comes to materials and chemistry, powerful supercomputers work on the basis of approximation and will never be able to deal with the level of complexity. That is exactly where quantum computing will play such a critical role: in breaking through these scientific and technical barriers.
Let’s take an exciting example from the world today: ammonia, a molecule consisting of one nitrogen atom and three hydrogen atoms that is the basis of the fertilizer used to grow the food that feeds the world’s population. Today, we produce ammonia using hydrogen from natural gas and nitrogen from the air in a century-old industrial Haber-Bosch process that produces around 2 percent of all global CO2. There are ways to decarbonize this process—using carbon capture or green hydrogen as an alternate feedstock, for example. But both of those approaches add cost, delaying their use on a meaningful scale, and neither addresses the energy-intensive high temperature and pressure required for this industrial process.
And yet, we know that nature has a much better way of doing it. Microbes make ammonia at normal temperatures and pressures, drawing hydrogen from water and using a complex molecule known as an enzyme, which is really just a biological catalyst—a molecule that enables or speeds up chemical reactions but otherwise isn’t part of the reaction.
We know it should be possible to replicate the function of this enzyme using an artificial catalyst, but we simply can’t simulate the stability of the naturally occurring enzyme with normal computers. If we could, it would mean we could use water instead of natural gas as the source of hydrogen, as well as make ammonia at around 30 degrees instead of 400 degrees. We could then make green ammonia cheap enough that we might start using it as fuel in ships by 2030. But don’t take my word for it—Google’s CEO has said that Haber-Bosch is on the way out.1
Another interesting example is battery design. Nobody truly understands the electrochemistry of lithium-ion batteries, yet they are already ubiquitous, and better batteries will be important on the journey to net zero. If we could more accurately simulate electrolyte molecules, for instance, we could solve major challenges around energy density, safety, charge time, use of rare or conflict minerals, and more. With large efficiency gains and new materials that don’t face supply issues, battery prices could come down fast and accelerate the transition to electric vehicles—including in the more challenging sector of trucking, where we could get to cost parity many years sooner.
Key insight #4: Use cases in quantum computing could account for a substantial amount of emissions reductions needed to achieve a 1.5°C pathway.
Philipp Hillenbrand: What is the impact potential these use cases could contribute to slowing down global warming?
Jeremy O’Brien: The top use cases thus far include batteries with higher density (particularly for transportation and grid storage), perovskites for more efficient solar panels, new solvents for point-source carbon capture and adsorbents for direct-air capture, new zero-carbon cement clinkers, modeling membranes, catalysts and electrical currents in hydrogen production, and (as mentioned earlier) new ways of creating clean ammonia. Real cost reductions drive technology adoption rates, so this should speed up green technology use by five to ten years.
These top use cases could achieve large-scale decarbonization impact and help contribute toward the planet getting back on a 1.5°C pathway.
And this potential impact starts making a lot of sense if you think of quantum computing as an enabling technology solving computational bottlenecks that are holding up the design of new technologies. In turn, it accelerates the adoption of existing technologies by making them more efficient or economical, so people use them a decade earlier.
In fact, it is PsiQuantum’s vision that quantum computing needs to work on solving the critical challenges of our time, beginning with climate change. We believe this so strongly that we made an early commitment of our initial capacity to be dedicated to sustainability use cases, such as the ones I have described here. To that end, we launched Qlimate with a focus and mission entirely driven by climate impact.
Key insight #5: Technology is a critical element in the fight against climate change, but it can’t solve the crisis on its own.
Philipp Hillenbrand: The cynics could say that, as humans, institutions, and governments, we can continue to misbehave, since advancements in technology will help to prevent the worst anyway. What’s your answer?
Jeremy O’Brien: Absolutely not! My position is that we must do everything in parallel. We need to solve the climate emergency using every tool we have. We also need to figure out how to use quantum computing to build new tools. And we need to build that million-qubit quantum computing capability as quickly as possible.
For someone who has spent much of his life working on technology, I don’t subscribe to simplistic tech “solutionism”; net zero will come from a combination of policy, markets, consumer behavior, and technology. Given the dire situation we are in, we’d obviously be mad not to do everything in our power to develop the technology that could help us get to net zero and beyond. That’s why my colleagues and I at PsiQuantum are so dead set on supporting climate breakthroughs.
At COP26,2 I was encouraged that we already have a great deal of technology to make progress. Technology is obviously part of that market-policy-behavior nexus, as we’ve seen with solar power: better technology makes low-carbon solutions cheaper and more scalable, which affects markets and can support shifts in policy and consumer choice. As long as technology isn’t a distraction but an enabler, it will play a crucial role in fighting climate change. We are very focused on making sure that quantum computing serves a purpose, and, for us, that means addressing some of the most pressing issues of our time, starting with climate change.
Philipp Hillenbrand: If you look 30 years ahead, what role will quantum computing play in our daily lives? What will it have achieved by then in your personal best-case scenario?
Jeremy O’Brien: It is my conviction that the impact of quantum computing is going to be more profound than any technology to date, creating whole new industries and opening up solutions to problems that would otherwise forever be impossible to solve. The real question is whether we will make sufficiently powerful quantum computers in the time frame we need. I believe so. We will experience revolutions in our everyday lives: where our energy and food come from, the medicine we take, and, more broadly, how we manage complexity and risk in large systems such as financial markets. Those are just a few examples that spring to mind. But above all, I hope that 30 years from now we will look back in awe at how we managed to avert climate disaster and how quantum computing was deployed to support the tremendous collective efforts and human ingenuity that went into finding the right solutions.