Andrei Dragomir, CEO of Aquark Technologies, is interviewed by Yuval. His company, emerging from the University of Southampton, focuses on miniaturizing quantum devices, particularly for sensing applications, aiming to transform room-sized quantum systems into mobile phone-sized units. Aquark is also part of NATO Diana accelerator and is pioneering in enhancing the practicality and robustness of quantum sensors. Andrei highlights their recent airborne experiment, describes what keeps him up at night, and much more.
Full Transcript
Yuval Boger: Hello, Andrei, and thank you for joining me today.
Andrei Dragomir: Thank you for having me. It’s great to be here.
Yuval: So, who are you, and what do you do?
Andrei: I’m Andrei Dragomir, I’m the co-founder and CEO of Aquark Technologies. We are a small startup from the University of Southampton, a spin-off from the University of Southampton in the south of the UK. We focus on improving the size, weight, power, and cost of quantum devices based on the cold metal platform or neutral atoms platform in order to accelerate their development and overall market adoption. And cold atoms or neutral atoms can be used for computing, can be used for sensing, they can be used for a range of quantum technologies.
Yuval: Do you focus on one particular area?
Andrei: That’s a great question. What we’re building is a platform, so our focus is on the cold atom platform itself in terms of improving size, weight, power and cost. We see our technology eventually reaching computing and communications as well. But we feel like sensing is here. We feel like computing potentially has other problems than size, weight, power, and cost to solve before that sort of thing becomes relevant.
Yuval: And you’re part of an accelerator, right? Maybe the DIANA accelerator, is that correct?
Andrei: Yes, it’s one of our latest achievements, latest announcements. I’m guessing that now, with the geopolitical situation in the world, NATO is focusing more on innovative technologies. And as part of that focus, they put together the NATO DIANA accelerator. And we’ve recently announced that we’ve been selected to be part of the first cohort of this accelerator. And yeah, it’s a huge honor and recognition of what our company is trying to achieve.
Yuval: Why do these systems need to be smaller and cheaper? I mean, if I put a quantum computer, if the room needs to be just a little bit bigger, so what? Why do we need to make them smaller and cheaper?
Andrei: Well, particularly in terms of sensing, the technology, the cold atom or neutral atom technology has been around for quite a long time. I think 30-plus years at this stage. And we know it works very well in terms of using atoms as a ruler to measure different environmental parameters, such as time, gravity, rotation, and acceleration. And they work very, very well in a laboratory environment. And even in today’s world, atomic clocks, for example, are a very good application of neutral atom and cold atom technology. They underline our use of GPS and GNSS. So, our Google Maps or Apple Maps kind of rely on quantum technology already. But it’s hidden away. It’s hidden away in the lab, and it is distributed wirelessly throughout the world. And the question is now, how can we make that even more beneficial? What if we can bring that technology outside of the hidden room and take it out in the world and actually be able to use it? And one big obstacle to overcome is the current size, weight, and power that these devices require, together with the time intensive investment needed to comprehend and manage those systems, to be able to use them, to maintain them.
So, we believe it is the next step in the technology phase to step outside of the lab. And our goal as a company is to hide the magic going on behind the button. Mostly, the people that are going to benefit from this technology don’t really need to know that it’s quantum-based. They just need to benefit from what it can offer. So, removing quantum from quantum is what we’re trying to achieve, and miniaturizing this technology is the first big step towards that. And in terms of computing as well, in the future, there will be a lot of cloud-based platforms, but there will also be requirements for quantum computers locally. So, when computing is ready to reach those types of markets, then hopefully, we’re going to be ready with the miniaturization solutions as well to make that happen.
Yuval: So, let’s talk about sensing for a second. How small is small? So, what is the goal if I have, say, an airborne or a vehicle-carried sensor? How small would you like it to be? How much power would it require?
Andrei: Very, very good question. And it depends a lot on the sensor itself and on the application. But today’s systems generally occupy entire rooms. The level of miniaturization we’re seeing starting to emerge in different commercial applications is still quite large; systems the size of a washing machine or fridge. We’re trying to, as a first step, push this forward a little, and what we’ve created now is a minimal viable product, or prototype, the size of a shoebox. It’s more or less the size of a 19-inch network rack that contains everything required to generate the cold atom cloud. And that requires about 50 watts of power at this point in time, if I’m not mistaken, and weighs just less than 10 kilograms. So, that in itself already could be suitable for a range of applications. So, if you want to install an atomic clock on a boat or lorry or something similar, that would be perfectly suitable.
But that’s not where we see the technology really shining. We really want to push this to the size of a mobile phone today or a hard drive. And that will allow us to not only expand the applications that this technology can reach, but also enable mass production. And I think that is the key word that this technology needs because with mass production, you’re also, as a company, going to be able to reduce the cost of it, and reducing the cost is going to be a huge driver in terms of accessibility and taking this technology to the wider world and larger market.
Yuval: You mentioned applications such as gravimeters or acceleration meters or atomic clocks. Putting the atomic clock aside, there are non-quantum sensors that do this, and I’m guessing they’re much, much smaller. So, how much better are the quantum sensors going to be to justify that additional cost, space, and power consumption?
Andrei: It’s interesting because performance is not the only metric that surrounds this technology. The key to neutral atoms and cold atoms is that they’re always the same no matter where you are in the universe. So, devices built on this type of platform will not necessarily require calibration. And overall, beyond that, they will be able to perform a continuous measurement whilst not drifting too much from their initial status. So, the quality of the data that we get through is going to be much higher than what existing systems allow today. With gravimeters, for example, there are some technologies out there that in terms of just brute performance in a single snapshot, would appear very close to what quantum technologies can and will offer. But in terms of drift and creating a long-term valuable solution, cold atoms will end up being a clear winner. If we want to enable long-term monitoring, for example, current technology is not the way to do that. We need something new. And that’s where quantum will step in.
Yuval: I read somewhere that you had an airborne experiment recently. Did I read correctly? And if so, could you tell us a little bit about that?
Andrei: Absolutely. It’s one of the most exciting developments we’ve had in the past couple of years since we started the company. First and foremost, we wanted to show that the platform we’re working on is suitable for being taken outside of the lab. As I said, we created our first system of this shoebox size. And we like to learn by breaking things. So we thought of the most exciting way to break the system to learn from it so we can build it better again. We had the opportunity to install this system on a prototype drone together with a defense company and to just simply fly it around. And this happened in November 2022 in the fields of Somerset here in the UK, it was at 0 degrees Celsius and over 80% humidity.
I’m happy to say that it was a fun day for the entire team, spending that day in the cold field. But to our surprise and to our happiness, our system performed absolutely admirably. It effectively did not care that it was attached to a tiny helicopter made out of wood and 3D-printed parts. So it wasn’t a fancy drone by any extent. But just good enough to carry the cold atom system and perform some quite aggressive maneuvers in that sense. And it just simply worked. We couldn’t find any fault.
So, we decided to take it to the next level. In the end, we concluded our trial experiments for the day by dropping the atoms out of the sky. We took the drone up with the cold atom system at about 120 meters altitude, and we cut the engine off. And we let it fall. And the first three times we did this, it worked brilliantly. The drone recovered, and the cold atom system was still active after that entire fall. The last time we did it, the drone did not successfully recover and the entire system crashed into the ground, suffering over 17 G of an impact according to the data we collected there. We’re very happy to say that there was almost no damage inside of the system. Which, when you start from working with a system in a lab where it is sometimes enough to sneeze in the wrong direction and you lose your experiment, being able to take it in the field and fly it around and drop it from the sky and it’s still working, that’s quite a big achievement. It now allows us to move to the next phase of the company, which is applying this platform to different sensing applications.
Yuval: How heavy was that system, and what did it measure in flight?
Andrei: The system was about 9.6 kilograms. In the UK, there are weight restrictions for drones, and any drone weighing 20 kilograms or more requires a license for operation. And with the drone weighing about 10 kilograms in itself, we couldn’t have our system weigh anything more than that. So the entire box was 9.6 kilograms, out of which 3.5 is just the box itself. So the internals weigh much, much less. And we effectively monitored the consistency of the whole atom cloud and how it behaved in flight. Together with the atom number that the cloud produces, which is an important factor in creating sensing applications from it. And we also had the conventional gyroscope, a whole probe for measuring magnetic fields. Similar to conventional sensors, we mapped out the environment and then linked our cold atom cloud behavior to that environment to see what the weakest point of it is.
Yuval: So, did the cold atom environment also perform measurements, or was it more internal telemetry and seeing how it behaves in various conditions?
Andrei: It was the internal telemetry in itself. That was a huge challenge — maintaining a cold atom cloud, and the generation of cold matter in flight in such a harsh environment. So, we concluded our system is challenged enough at that point and we didn’t perform any measurements directly with it. But despite that, looking at the data, it served as a very good magnetic sensor, even just monitoring the atom number and the fluorescence that comes out of the cloud. You can directly map that to the Earth’s magnetic field and measure it and its position relative to the flight pattern of the drone. And in that system, particularly, we didn’t have any magnetic shielding. We didn’t protect it in any way. We simply want to learn how it behaves as basic as possible so we can learn from that.
Yuval: I think cold atom systems require a vacuum. And if you are on a drone, you may not have constant power supply, unlike an office. How long does it take for the system to be ready to perform in terms of vacuum or other aspects?
Andrei: The vacuum angle is a very interesting one for us because it goes back to our core technical advantages. The research group from which we spun out was the Integrated Atom Chief Group at Southampton which had as a main goal the miniaturization of vacuum systems. We have over 12 years of experience now that has been imported into the company to allow us to build very robust, very small vacuum systems. And that goes a long way in terms of having a portable cold atom generation platform. The vacuum is one of the essential pieces of that puzzle.
We adopted some of the technology that has been developed at CERN in the 90s. While CERN developed what they call thin-film Getters or Getter-Film technologies to maintain an ultra-high vacuum environment in the Large Hadron Collider, we started from there and then redressed the technology to suit the other spectrum of vacuum, which is very, very small systems. And we’ve got pretty good at it. So, actually, our vacuum is constantly on — almost to the level of being passive, which is one of our goals as a company, to create a completely passive vacuum system that will not require any type of active pumping behind it. And it requires, because we can make them so efficiently, very little power. Running on a drone battery, the vacuum system can survive for days. That is not the main problem. The biggest problem came out from the conventional sensors and computers required to interface with it. So, in the box, we actually have a small Windows 10 computer. And that is our biggest power consumer. So, based on just a drone battery that was installed in the system, the system can survive for up to three, four hours at this point in time. And you can be ready to go in about 10 minutes from having nothing on to generating cold matter.
Yuval: If you open that shoebox, and I think you mentioned 19-inch size, so it’s pretty large shoes. But if you open that shoebox, what is the biggest part? What needs to be really miniaturized so you get more into the hard drive size sensor?
Andrei: If we were to open the box, the first thing you will notice is that it’s mostly empty. What determined the size of the box was really end-user requirements first and foremost, and then our way of just putting together the puzzle, so to speak. So we’ve only used available off the shelf components for this particular system. So there’s a lot of electronics inside that otherwise will not be needed if we are to focus on a custom electronic solution, for example. So just by doing that, the system size is reduced to at least half. The next after that would be the vacuum component. And for that, we’re using some medium-sized systems, not necessarily the smallest we can build, because we focus on robustness more than anything else. Having a slightly larger system simply allowed us to look inside and better understand what is happening with the cold atom cloud during the flight. And then it’s the light delivery system. So, for us, addressing these three components will essentially bring us to the size we want to be at, to a hard drive size. So making some custom solutions for the electronics; the vacuum system; and then the light delivery for either the adoption of photonic integrated circuits or micro tips assembly. Then, that will bring us to the hard drive size we’re envisioning.
Yuval: That sounds like a scary proposition to take a prototype, put it on a drone, cut the engine off, and cross the fingers that it survives. But congratulations on pulling that off. Tell me a little bit about the company. How large are you? How are you funded? Anything that you can share?
Andrei: Absolutely. And thank you. It was an interesting endeavor. But as I mentioned before, we like breaking things. So we’re going to continue doing this type of stuff. The company was created in 2021. On the 16th of May, we received our first investment. We are supported now by an incredible group of angel investors who are also acting as our mentors.
Both founders, both myself and Alex, my co-founder, come from a scientific background. So we have to change our core and our way of viewing the world to take on a leadership position in the company and actually make it happen. And since then, we had been running for a long time with a team of four. So, the flight experiment and all that was done with a very small team. And just recently, in the last couple of months, we’ve now increased to a team of 10 full-time employees and a much larger family in terms of part-time engagements and other contractors that are helping deliver our vision overall. So it’s growing fast. It’s quite an adventure. And we’re still learning as well in the process ourselves as founders.
Yuval: From the DIANA Accelerator, is it just money or do you feel that you’re getting additional value?
Andrei: Actually, I think the money is the least interesting part. It does come with a 100,000 euro grant to support our development. But the network and the DIANA team have been absolutely incredible, together with the mentors that we get through the program and the exposure to the defense world, which obviously is a big early adopter of this type of technology. All this is going to benefit us a great deal in the long term. I think the benefits for us coming out of this accelerator are going to be absolutely incredible.
Yuval: As the leader of the company, professionally speaking, what keeps you up at night?
Andrei: Well, besides the huge amount of work, it’s a lot, actually. It’s difficult to pick. But generally, making sure that our team and the company has a chance to try and innovate as much as possible and deliver this system to a first market is probably the biggest concern. Seeing this technology out of the lab is the big win that I personally want to get out of it. And that is generally what keeps me up at night.
Yuval: And lastly, hypothetical, if you could have dinner with one of the quantum greats, dead or alive, who would that be?
Andrei: That’s a tough one. Now, in terms of people alive, the community is quite small, so we actually get to meet almost everyone. So I’m going to turn it around a bit and say, I would really want to sit down and have dinner with not a quantum great, but potentially someone that has been involved with, for example, NASA as part of the Apollo missions on the ground and witnessing that happen. I think that story is quite inspirational and I’ve been taking as much as I can from it in trying to articulate our vision. And just for example, in terms of computing, the trend up until we tried to put the man on the moon was who has the biggest computer. But to actually deliver that feat of getting to the moon was then a question of how do we make the computer small enough to fit in a spaceship. So that created a turning point for how technology was viewed and it eventually resulted with us walking around with smartphones. So taking that analogy forward to what we’re trying to do today is definitely not landing a man on the moon, but we’re very excited about what this technology can offer in terms of applications that haven’t yet been thought of. So we can use it as sensors, we can use it for computing, but really in 10, 15 years time, who knows what this technology can offer to us. Miniaturizing it and putting it out in the world is what’s going to bring us there. So we’re very excited to see that. And I think having dinner with someone that lived that, that had that experience in those days, I think would be extremely beneficial to us.
Yuval: I do love the lunar lander analogy because I think their computer had 70 kilobytes of program memory. And when we speak today about small quantum computers and what you could do with them, we say, well, you know, a man landed on the moon and got back with 70 kilobytes. So you can do a lot with a little if you’re motivated enough. Andrei, thank you so much for being here today and sharing your story.
Andrei: Thank you very much for having me. It was absolutely amazing. Thank you.