Tim Spiller, director of the Quantum Communications Hub in the UK, is interviewed by Yuval Boger. Tim describes the Hub’s work on quantum key distribution (QKD) and quantum networking. He highlights advancements in short- and long-range quantum communications, including a CubeSat-based QKD system. Tim elaborates on the transition from research to commercialization within the UK National Quantum Technologies Program, touching on intellectual property transfer, industry partnerships, and ethical considerations in secure communications. He also shares insights on global quantum initiatives in China, Europe, and the US, future quantum networking goals in the UK, and much more.
Transcript
Yuval Boger: Hello Tim, thank you so much for joining me today.
Tim Spiller: Oh, it’s very nice to be here. I look forward to our discussion.
Yuval: So who are you and what do you do?
Tim: Okay, well I’m Tim Spiller, I am professor of quantum information technologies at York, but I think more for the purposes of this discussion, it’s also relevant that I am the director of the Quantum Communications Hub. Now that is one of the current parts of the National Quantum Technologies Program in the UK, which is leading, I guess, the low technology readiness aspects of quantum technology development. There are four hubs in the UK, I lead the one on communications, and each of those hubs is a collaboration of roughly 10 universities plus industry partners, plus national laboratories and institutions. So they’re big spread out things across the UK and we collaborate together to push forward various quantum technologies. In my case, it’s quantum communications.
Yuval: And when you say low technology readiness, I mean, sometimes communications come up as being closer to prime time than say computing. What do you mean by that?
Tim: Well, what I really mean is that it’s understood that the quantum science research that underpins all quantum technologies is done in university labs, clearly elsewhere in the world, but for the UK program within the UK, and we build on top of that to take that quantum science and push it towards having demonstrators of new technologies in quantum communications. We don’t turn it all the way into a product. If we are successful in our development stages, then what usually happens is that our work will transfer into another project, which might be funded in a different part of the UK national program, and it may well be led by industry. That project might develop it further. Then, if it all goes well, either an existing company or perhaps a startup that comes out of the work will then take it even further and raise, you know, venture or whatever funding to fully commercialize it. So we do the lower few stages of that and then pass on the work right through the national program so it can be commercialized in the future.
Yuval: And I believe that you guys are now in phase two of the hub, right? Phase one was maybe until 2019 or something like that. Is that correct? And what are the goals of phase two?
Tim: Yes. So the national program started in 2014, the funding was found in 2013. The program started in 2014, had phase one for five years, has phase two for five years, and there will be a phase three starting shortly. Although our particular hub has a little overlap into that, which I could talk about later. But basically in phase one, we pushed forward various new technologies in quantum key distribution, which was already established as something that could work and be deployed around. In particular, in phase two, we have focused on what we’ve called quantum communications at all length scales. We’ve looked at short-range free-space handheld quantum key distribution devices. We’ve looked at further developing what’s been done in optical fiber networking, so QKD in optical fibers around parts of the UK. And we’ve also looked at the longest distances by building both a quantum transmitter and a quantum receiver. The transmitter will go on a small CubeSat that is planned to be launched next year. We have a quantum receiver that will go on a ground station that can receive those signals. So we want to demonstrate quantum key distribution from a satellite to the ground.
Yuval: A lot of the QKD publications, at least commercial publications, have been around use cases of, say, you know, a large bank needs to securely make transactions and thus QKD is important. I’m curious about the short-range applications that you mentioned. Why would QKD be useful in short-range communications?
Tim: Well, this would be for consumers or individuals. The idea might be that if you want to do future transactions with your bank or indeed with your employer, possibly the health service, or the government, then if your handheld device is used to share a key with any of those institutions, you then carry around one half of the symmetric key and those institutions have the other half. You can then use that key to subsequently secure either a financial transaction, some communication about your health, or some communication with your employer. Or possibly even to get through the door at your employer or whatever. Shared symmetric keys can be used for quite a number of things. Encryption and decryption is one, but it could get you through the door or something like that. So the idea is that it would be a handheld device that is yours, and you can exchange key in a kind of top-up way when you go to your bank or your employer or something. You then carry around your copy of the symmetric key in that device, and you can secure transactions or whatever in the future that could be done at some other place that’s connected to that institution. So it’s really for individuals and consumers.
Yuval: You mentioned that phase two is coming to an end in 2024, although there may be some overlap. What does phase three look like?
Tim: Well, phase three was officially announced in August, I think. But the process ran for a year or more. It’s been announced by the UK National Programme, and in particular, the Engineering and Physical Sciences Research Council funds the hubs and these low TRL aspects of the work. There will be five new hubs that all start on the 1st of December. So there’s very much continuity from phase two into phase three, but there are genuinely new hubs starting up. In particular, one of them is focusing on medical and bio applications of quantum technology, so that is the new one. The other four are an evolution of work that’s been going on in phase two into phase three. I did mention there was one caveat: the planned launch of our satellite is going to be next year, 2025. Because we still have some funding left, with Covid and everything, people generally developed a bit of underspend in their hubs. Our phase two hub has been given a one-year extension just on the space mission part so that we can actually deliver all the way through to the launch of our satellite. After that, once the satellite is up there and operational, the phase three activities relevant for quantum communications can take over and deliver further work on that. But we’ve got a one-year extension just for that part of our hub so we can deliver our space mission successfully.
Yuval: How does it work on an IP level? I mean, you mentioned that there’s sort of fundamental research in technologies, and then the hope is that these technologies get transferred into the market through startups or maybe program participants that decide to start companies. Do they now have to license the IP from the developing university or from the hub? How does one start a company this way?
Tim: Well, it depends how it works. So within something large like a hub, there are many smaller collaborative projects focused on different aspects of quantum communications. In general, if there is something that focuses on, let’s say, a technology development where you would expect IP to come out, there will be a sub-agreement for that smaller project between the relevant partners. Now, I mentioned that there are industry partners involved in the hubs, so it may well be that existing companies are involved in that particular project. In that case, they might already be partial inventors along with university people of the relevant IP, in which case there will already be an IP agreement between some of the university inventors and the company that could be doing the commercialization. If it’s purely a university thing or if universities and a company don’t yet exist, there are mechanisms for creating a new startup at a university. In general, there will be some agreement between the university or relevant universities about transferring or licensing the IP into the company so it can be commercialized that way.
It’s also possible that we do some level of R&D up to a certain point and then pass it on to another project. That project may well be funded by Innovate UK rather than EPSRC, and that would operate at a higher TRL. It would be industry-led, but would still have university partners. That would have a completely separate IP agreement, and it could be that new IP is developed within that project to carry the work further towards commercialization. So in general, it’s done on a case-by-case and specific project or technology basis. I think it would have been extremely hard for the whole UK national program to have an overall IP management strategy across the board, because companies would have been reluctant to buy into that when you don’t know where you’re going. It’s much easier to have agreements that are done case by case, and then you can choose to commercialize that particular technology based on the IP in that area.
Yuval: I believe I read that there were some commercial demonstrations in the UK. I think BT and HSBC perhaps demonstrated QKD. To the extent you can share, where is the gap between what you guys are working on and what the commercial enterprises are demonstrating?
Tim: I would quote that particular example as actually a hub success. There was preliminary work on QKD, both in companies and in universities, before the UK national program started. But during the first part of the UK national program, and carrying on into phase two as well, we built the UK quantum network. This involved a small metropolitan network around Bristol, a small metropolitan network around Cambridge, and we borrowed parts of the National Dark Fiber Facility, which is an R&D fiber network funded by EPSRC for all sorts of networking R&D. We used parts of that to connect across between the Bristol network and the Cambridge network. We also set up, funded on a separate project by EPSRC, a link from the Cambridge network to BT’s R&D center at Adastral Park, near Ipswich. All of that was established, and we showed the principle with industry partners involved.
Those industry partners then took on further work in Innovate UK-funded projects to develop it further, and they did additional demonstrations. That’s progressed now to BT and Toshiba running a small QKD commercial demonstration network in London, with HSBC and EY publicly mentioned as customers. So, we’ve seen the evolution of QKD work go up the TRL stack from research at the bottom to near-commercial technology, which is great.
Since then, the hub has been working on further refinements to QKD. We’ve done work on the dynamic reconfiguration of quantum networking, known as software-defined networking, allowing you to reconfigure the network in software. We’ve also collaborated with industry partners on that. Another focus has been the distribution of entanglement. At the moment, all the commercial QKD demonstrated is prepare-and-measure type QKD, where you send single photons or single photon light pulses, or possibly coherent pulses. But the commercial stuff doesn’t yet function on entanglement. That will be a focus in phase three of the UK national program, and it has already been a focus of some hub work. We expect that to move up the TRL over the coming years. But it’s nice to see the basic QKD work progressing all the way to commercial demonstration.
Yuval: How do you see the focus of the UK programme as similar or different to other countries in Europe or the US? What’s the difference between your focus and what you see in other countries?
Tim: I’m not sure there’s a huge difference in focus. I think we’ve put a lot of effort into getting QKD through to the demo level with BT, Toshiba, and other industry partners. Our focus will now shift significantly onto entanglement-based quantum networking. In Europe, it’s approached similarly. Within the EU, which we are now not a part of, they are developing the EuroQCI, the European Quantum Communications Infrastructure. At the minute, it’s focused on setting up trusted node QKD links across Europe. That parallels much of what we’ve done in the UK. Also, in Europe, they are working on entanglement-based connections between qubits in different laboratories. In the UK, our focus has been more on distributing entanglement and using it immediately. But in phase three, you’ll see us focusing more on getting distributed entanglement to interact with matter qubits.
There are also parallels with the US. Although QKD hasn’t been widely adopted at the government level in either the US or the UK, there have been trusted node demonstration networks in the US. The US is now also emphasizing quantum networking involving entanglement distribution and matter qubits, which parallels what we expect to see in the next few years in the UK and the EU.
Yuval: How about China?
Tim: I would say they’re doing the same. The one thing to note is that there’s been very significant government investment there. The Chinese launched a substantial satellite with quantum sources on it quite a few years ago, in 2016, I believe. They’ve demonstrated quantum communications from space already, showing the feasibility of such communications. They’ve also built a substantial trusted node QKD network from Beijing to Shanghai. So what they are doing is similar to what’s being done elsewhere, but at a larger scale and with greater investment. That’s probably because of the stronger backing from the Chinese government to push QKD alongside other forms of future secure communications that will be proof against quantum computer attacks.
Yuval: We were talking about communications, long-range and short-range, but one area that seems to intersect with quantum computing is actually connecting quantum computers with some kind of network, whether to scale up their capabilities in the case of superconducting or trapped ions or others. Does your hub also deal with that application of networking?
Tim: Our current hub doesn’t. However, the evolved version of our hub in the next phase of the UK national program certainly will. In the five new hubs, there will be one that focuses on quantum computing and one that focuses on quantum networking. Both of them are very much aware of and addressing this particular problem.
I’ll pull apart a technical difference in just a minute, but it’s worth noting that in addition to the new hubs, the UK national program has framework entities called missions—there are five of those as well. There is a mission on quantum computing and a mission on quantum networking. These missions are making sure that the quantum networking aspect you just alluded to will be covered and addressed.
Now, I think there are two different aspects worth separating. If you want to connect quantum processors together, say in different fridges if they’re superconducting processors, you might be able to connect them using microwave links or possibly by transducing from microwaves to some other frequency, then using short-range quantum communication to connect the processors. The focus there is to connect coherently over a short distance.
If, however, you want to connect two quantum computers that are 50 kilometers apart, you’re almost certainly going to have to use fiber and telecom-wavelength photons, as those are the types of photons that travel down fiber. That may require frequency conversion, because the frequencies that quantum computing qubits operate at are not the frequencies that can propagate down telecom fibers. So there will need to be some clever frequency conversion happening at both ends if you want one computer to talk to another over a long distance.
To summarize, in the next phase of the national program, the shorter-range networking between adjacent fridges will likely be handled by one hub, while longer-distance networking involving quantum memories or frequency conversion will be managed by the other hub. But I think it’s fair to say the UK has got it covered, and both aspects of networking will be addressed in the next phase.
Yuval: You’ve been working on quantum networking for a number of years, both in the context of a hub and much longer as a physics professor. What have you learned about quantum networking in the last, say, 12 months that you didn’t know before?
Tim: That’s quite a tricky question. I think what I’ve learned is that it’s going to be quite hard work. The more you dig into the challenges, the more you realize that there’s still quite a lot to be done. Naively, you might think you can just take photons out of a qubit, send them straight down a fiber, and put them into a qubit at the other end, but it’s going to be much trickier than that at the moment. There are numerous engineering challenges, and the more you investigate them, the harder they seem. But, you know, we’ve overcome a lot of challenges in the past, so I’m optimistic. I think it’s going to take quite a lot of work, though. In short, while we’ve achieved a lot over the last 10 years, there’s still a great deal to do in the next 10 years.
Yuval: As we get close to the end of our conversation, I wanted to ask you about ethics in computing. There’s a big discussion about ethics, as quantum computers could be used for both wonderfully good purposes and for potentially harmful purposes. Does the ethics conversation also permeate quantum networking?
Tim: Well, it certainly has from the privacy aspects of quantum communication. But I don’t think this is just a quantum ethics issue; it’s a general secure communications ethics issue. The UK has to decide whether it wants secure communications to be accessible and interceptable in cases where something suspicious might be going on, or whether it wants to offer true end-to-end security for individuals.
If we offer genuine end-to-end security to individuals, we can’t see what they are communicating with each other. People want to have their cake and eat it too—they say, “Well, I want my end-to-end security, but I don’t want criminals to have it. I want their communications to be interceptable.” But either we have a system where communications can be intercepted in the middle, or we don’t.
The trusted node approach with QKD kind of gives you this. You could ask the service provider to provide access to what went through that node. But if we have quantum repeaters in the future, we might be able to offer genuine end-to-end quantum security. So we are aware of the technology we’re developing, and ultimately the UK will need to decide what kind of secure communications it wants. Does it want the right to lawfully intercept anyone’s communications, or does it want end-to-end security for individuals? That’s a decision for the government, not for us as technologists. The technology can be built to accommodate either scenario, but a decision needs to be made about which path to follow.
Yuval: And finally, a hypothetical: If you could have dinner with one of the quantum or communications greats, dead or alive, who would it be?
Tim: Well, it’s not just quantum communication for me. If I go all the way back, my heroes are Einstein, Dirac, and Feynman. It’s hard to pick just one, but those three have always been my favorite in terms of their contributions to quantum theory. Einstein usually comes out on top for me, so I’d probably pick him. But if I had the option, I’d choose all three. Although I understand Dirac might not have said very much at dinner, as he was rather quiet and reserved.
Yuval: Wonderful. Tim, thank you so much for joining me today.
Tim: Thank you.