Chris Monroe is a physicist at the University of Maryland, US, and the co-founder and chief scientist of IonQ, a start-up that is developing quantum computers using trapped ions as qubits. He recently spoke to Margaret Harris about the rise of quantum computing, and how his previous experiences – including stints in the labs of two physics Nobel laureates – fed into his decision to start the company in 2015.

How did you get interested in quantum computing? Because you really got into the field at the very beginning…

Yes, I’ve been in this field for more than 25 years, and I have to say it sort of landed in my lap. I did my PhD work on cold atomic gases at the University of Colorado, Boulder, in the group of Carl Wieman and Eric Cornell, who went on to share the physics Nobel prize in 2001 for making the first Bose–Einstein condensate. But I always knew that at some point I might want to get a “real job”, and atomic physics is good in that respect because it involves working with practical things like optics and lasers and photonics. There’s a lot of equipment involved, and I was attracted to the technical nature of the work.

After I got my PhD, I went on to do a postdoc. The postdoc system is a little stressful, because it’s a temporary job, and you’re in your late 20s, and everyone else you know is building their career. But postdocs are also wonderful opportunities to try something random, because there’s very little at stake if it doesn’t work out. And in my case, I didn’t have to move very far. I stayed in Boulder and went down the road to work with David Wineland at the National Institute of Standards and Technology (NIST).

In the early- to mid-1990s, Wineland’s lab was basically the atomic-clock division of the US government. He is an amazing researcher, and NIST allowed him to do academic-type research within this government lab. So instead of building the clocks that people use as a real time standard, we were doing research on how you might make better clocks. One of the crazy ideas we had was that by entangling multiple atoms or ions, we could make our clock run faster (and therefore more accurately), so we came up with a scheme to entangle two ions.

As it turns out, that meant we were building a quantum gate for a tiny quantum computer. But we didn’t know those terms at the time. I didn’t hear about quantum computing until the summer of 1994, when I learned of Peter Shor’s algorithm for factoring large numbers using a quantum computer.

When Wineland and I saw Shor’s article, it entirely changed our direction of research. We were still at NIST doing atomic clocks, but now we were also doing quantum computing, and government agencies got very interested in seeing what we needed to do to scale it up. Everything we did in that laboratory was ground-breaking, and Wineland went on to win the Nobel prize in 2012 largely based on his work in the 1990s. It was a pretty cool beginning to my career.

Several years passed from when you first heard about quantum computing to when you set up IonQ. What made you decide “This isn’t just a research topic anymore, I’m going to start a company”?

For the first 10 years or so, there was a lot of research to do. Picking out the best type of quantum gate. Deciding which atomic species to use. Working out how well the lasers perform and how big our quantum systems could be before they got killed by noise. It took a long time, not just for me, but for the whole community to do those experiments. And in terms of scaling things up from a handful of qubits or gates, really nothing happened for a long time apart from high-level proposals for scaling. Although we were starting to understand the limitations of the physics, we weren’t ready to do the engineering.

Beginning in 2010, though, we started to narrow things down and make decisions, and by 2014 or 2015 we had our first tiny quantum computer. And that was interesting, because after we initialized and calibrated the system in the morning, we stopped doing atomic physics in the afternoon. Once the system was seeded, we could stop tinkering with the lasers, go over to the PC that was controlling the experiment and run algorithms.

After that, a couple of things happened. I had a long-standing collaboration with a colleague from Duke University, Jungsang Kim. He’s an engineer, and we recognized that we kind of filled each other’s gaps. I’m a physicist, and I’d been in this field for a long time, but he has great experience in what’s called systems engineering, and he thinks differently about physical systems than I do. We realized that, together, we could do some amazing things.

Around that same time, in mid-2016, IBM built a five-qubit superconducting quantum computer and put it in the cloud so that people could use it. At first, that seemed a little goofy to us, because five qubits is really small – we’re not going to learn anything from that. But it was more than just a publicity stunt. I mean, it was a publicity stunt, but it also allowed anybody to use the system, which was huge – a genius move.

As it happened, the system we were building was also exactly five qubits, but in terms of performance it was much better than IBM’s. This is because atomic qubits are nearly perfect and exactly replicable; because we could connect a pair of the atom qubits with reconfigurable laser beams; and because we could run “deeper” circuits. So people started approaching us, saying that they’d tried to use the IBM cloud, but it didn’t work for what they wanted to do – could we help? Initially, it was more like a scientific collaboration: we started running applications and algorithms that other people would send us. But we realized that to go to the next step required such a serious dose of engineering that it probably couldn’t be done at a university. And that was the genesis of IonQ.

What do you know now that you wish you’d known when you started IonQ?

One thing I’ve learned has to do with the computer science aspect of our systems. Moving the operations around in our algorithms so they’re mapped to our system in an optimal way turns out to be incredibly powerful – much more powerful than I imagined. If I’d known that two or three years ago, I would have hired more computer-science theorists.

As an analogy, the first PC I ever used had four kilobytes of memory. Now we have hundreds of gigabytes, and that means we waste it – we take pictures that are way too high resolution and store them on our hard drive because memory is a commodity. It’s cheap, it’s easy. There’s no reason not to waste it. But at the early stages of any technology, including quantum computing, you have to squeeze out every ounce of efficiency you can, because it might mean the difference between running an application and not being able to. In 10 or 20 years, I hope that qubits and gates will be more of a commodity, and then we can be more wasteful with them. But to get there we have to extract as much efficiency as possible. That’s not really physics – it’s quantum computer science, and it’s a very rare skillset right now.

In 10 or 20 years, I hope that qubits and gates will be more of a commodity, and then we can be more wasteful with them

That leads nicely to my last question. Do you have any advice for today’s physics students?

With a physics degree, you can pretty much do anything. The challenge is that the doors are not open for you to the same extent as they are in some other fields. When you study engineering, for example, it’s almost like going to business school. You make connections, there are job fairs and the doors open for you to go work for these big engineering firms. You can still do that as a physicist. It just won’t come to you. You have to go find it.

So the advice I would give is to keep your options open. If you do a PhD, it may appear like you’re narrowing your options, because you’re working for several years on just one thing. But if you can solve a problem at the forefront of your field, even if it’s a narrow problem, you learn how to do that in any field. Going in-depth in physics will help you no matter what you want to do, even if it’s something unrelated, such as finance.

We all learn quantum physics as physics students, but in recent years this field has taken on a whole new life. It’s not an esoteric theory anymore, something that only describes tiny effects in extreme forms of matter. It’s going to form the basis for a whole new type of technology. So I think that, because physicists have a bit of a leg up in this area, they should go all-in.