Sebastian Hassinger: 00:30

Welcome to the new Quantum Era. I'm your host, Sebastian Hassinger, speaking to you from Paris this time where I've been attending the Q2B conference. All the Q2B conferences are great, and I really appreciate the effort that QCWARE has put into them all these years. But I really enjoy the Q2B Paris edition and not just because of the appeal of the city itself. It's great to spend time in the quantum community here in Europe and especially the vibrant community here in France.

Sebastian Hassinger: 00:56

The public sector efforts in France have been really impressive with great coordination across strong university research labs, the public sector labs, the federal government, and the private sector as well. There's a lot of really great work going on, in particular in hardware, and today, we're gonna be talking with a cofounder of one of those hardware companies. Pierre Desjardins is the cofounder and CEO of c twelve, a company that's working on a unique variety of silicon spin qubits. We previously learned about spin qubits when we talked to Andrew Dzirak, the CEO of Dzirak. But the ones that c twelve are working on are quite unique in a way that is foreshadowed by the name itself.

Sebastian Hassinger: 01:40

So while I was here for the conference, I took the opportunity with the help of Lydia Baril from c twelve to go to their beautiful new offices right by the Pantheon and record an interview. Pierre's story is quite unique. He's one of the few quantum hardware startup founders that doesn't have a PhD, an attribute I share with him. Though he did get a master's in physics from Columbia, so he's more of a scientist than I. He founded the company with his twin brother, Matthew, and he's building the hardware based on nanotubes made of carbon 12 atoms.

Sebastian Hassinger: 02:11

We'll dive into how c twelve is creating ultra pure, highly coherent qubits using these suspended carbon nanotubes and talk about how they're developing laboratory experiments into real world devices. Pierre shared with me the company's focus on material science, unique combination of microwave controls with electron spin qubits, and how their quantum bus may be the key to their scalable error corrected quantum computing. I learned a lot from our conversation, and I hope you will too. So let's jump right into my conversation with Pierre Desjardins. I would love to start by hearing a little bit about how you got started, how you got to where you are as a founder of a quantum computing company, know?

Pierre Desjardins: 03:25

So a rare cofounder of a quantum startup company without a PhD.

Sebastian Hassinger: 03:30

Yes. That's notable.

Pierre Desjardins: 03:33

And and another rare thing is that I co founded C12 with my twin brothers.

Sebastian Hassinger: 03:39

Right.

Pierre Desjardins: 03:39

So it's good. I mean, actually, to have, like, two twins at the head of a quantum startups, you know, I think. All these

Sebastian Hassinger: 03:47

the jokes have been very predictable, I'm sure.

Pierre Desjardins: 03:50

So we graduated from the same engineering schools and then Matthieu Dejardins, my my twin brother was a PhD student in the laboratory of physics at the Ecole Normale Superielle. Mhmm. And this is where they've made all this, like, incredible scientific experiment and discoveries about carbon nanotubes quantum electronics. Right.

Pierre Desjardins: 04:17

I think we will cover this Yeah. Of course later. And then when he was a postdoc in Takis Kontos lab, is this is when he had, like, yeah, we could say the vision Mhmm. And the ambition to, like, convert all this, like, scientific discovery into, like, one super cool application

Sebastian Hassinger: 04:39

which

Pierre Desjardins: 04:40

is like building quantum computer.

Sebastian Hassinger: 04:41

Right. Right.

Pierre Desjardins: 04:42

And then it was in 2020 that we decided to incorporate the the the company and really to supercharge what was developed in in the lab the years before. So we just celebrate the five years of C12. Congratulations. Yeah. When when you are in a in a deep tech and even quantum tech with such like, ambitious goals.

Pierre Desjardins: 05:12

And, like, when also, like, all the developments take a very long time because

Sebastian Hassinger: 05:17

Yes.

Pierre Desjardins: 05:17

We know our building a quantum computer is super hard. Yes. So being here five years later with such a great company, I mean, the team is just so I I love working with them. It's great. So it's it was a very good time to to to enjoy.

Sebastian Hassinger: 05:38

So did he was it hard for him to convince you to that this was a good idea to to go try to build this this company?

Pierre Desjardins: 05:45

The the good thing is that we were during COVID. I stopped working Nothing

Sebastian Hassinger: 05:53

better to do.

Pierre Desjardins: 05:54

Exactly. I stopped working on my previous job. It was like a consulting, and I had like a like day long of like just reading quantum computing. And so I did my kind of due diligence on like Right. How many two cubits and it's quite fascinating.

Pierre Desjardins: 06:13

The more you dig into this technology, the more you understand why this is just the way to build a a quantum computer. So then I was like very strongly convinced.

Sebastian Hassinger: 06:23

You were a convert. Yeah. Yeah. I think that's that is definitely critical to any founder of a quantum hardware company to be convinced that they have the way. Everybody needs to have that.

Pierre Desjardins: 06:34

I mean, you're going to dedicate your life to building this, so it better be easier.

Sebastian Hassinger: 06:38

And there's so many challenges and so many unknowns. You really have to have that conviction that you're on the right path. So you already mentioned carbon nanotubes, but what you're building is a variation on spin qubits. So there's a number of companies and a lot of research in silicon spin qubits, which are using the spin of electrons, individual or very small groups of electrons depending to capture the the two level system for the qubit. But you're doing it in a different way.

Sebastian Hassinger: 07:09

It's quite unique. So tell me about that.

Pierre Desjardins: 07:11

We are in this, like, indeed a big family of spin qubits where we manipulate, as you mentioned, the spin of a single electron. And it's it's it's a I mean, first, you have, like, different, qubit modalities, but even within the spin Mhmm. It's a whole, like, a different kind of zoology. So used to have, like, three, material, like gallium arsenide, but then people understood that it was, way too noisy. And so now you have like people building a spin qubits very close to the CMOS.

Pierre Desjardins: 07:52

So Right. Silicon spin qubits. You have Germanium now. So but then in the it was structural, some people are doing like Germanium, silicon Germanium, and other are doing silicon silicon Germanium. Right.

Pierre Desjardins: 08:06

You have HDSY, and then you have also like carbon nanotube. So carbon nanotube might looks like, wow, super strange. It is, of course. But it's just interesting to understand also like for the moment, like the the the there is like many flavors Right.

Sebastian Hassinger: 08:22

Like Yeah. Spinach. I guess what's interesting to me is you're you're fabricating these carbon nanotubes separately and then affixing them to the chip surface. Right? I mean, you're actually you're you're testing them for impurities and making sure they're perfectly formed and then attaching them to a chip surface.

Sebastian Hassinger: 08:42

So that seems like a really unique challenge.

Pierre Desjardins: 08:45

So the the goal and the vision for about using carbon nanotubes is to have the system which have the the the purest material to trap the Right. The the electron and something where with the lowest disorder Right. So that's why we are using a material that is really like what we call it. We call them single single wall carbon nanotube. Mhmm.

Pierre Desjardins: 09:14

So it's just like one carbon nanotube and the the the the wall of this like nanotube is just one layer of carbon atom.

Sebastian Hassinger: 09:24

Incredible.

Pierre Desjardins: 09:25

In term of, like, what you can it's it's much it looks like at the end, we trap the electron into, like, a macro molecules. And it's it's extremely, like, a low density of of material. And we do that exactly the way you are describing it. So we fabricate the carbon nanotube on one side. We fabricate the silicon chip on the other side.

Pierre Desjardins: 09:52

And at the end of the process, we integrate the carbon nanotube on the silicon chip. And because we are able to, like, do the the assembly as a last step. Indeed, at the end, the carbon nanotube is suspended over the the gate electron. So as Matthew described it, it's like the closest realization of a spin qubit in vacuum. This is really how we want to build, like, super high quality qubit.

Sebastian Hassinger: 10:26

And so how how many carbon atoms make up each ring of the tube?

Pierre Desjardins: 10:34

So then below the carbon nanotube, you have electrode. Mhmm. And like other spin qubit technologies, we we trap the electron to, like, double quantum

Sebastian Hassinger: 10:48

dots or Right.

Pierre Desjardins: 10:49

Quantum dots.

Sebastian Hassinger: 10:50

Right. Right.

Pierre Desjardins: 10:50

So then at the end, the, like, really active part or the the the the part that of the carbon nanotube that is used is only, like, a a few micron Mhmm. Like

Sebastian Hassinger: 11:01

And do do do you I mean, is there a specific number of of carbon atoms that that make up the the circumference of the tube?

Pierre Desjardins: 11:09

I I know where you you are coming. Like, why 12? Why is it 12? Yeah. That's what I was guessing.

Pierre Desjardins: 11:17

It doesn't come from the number of atoms that are forming the the the the the cecopherants of this tube. And and here, it's very interesting also because what you mentioned here is, like, carbon nanotubes, they have like each of them, they have a unique number of characterizing them with the scirality. And we are actually selecting the the the right carbon nanotube to make the the best tribute. But on this side, fortunately, we can have, like, a wide range of that that work. The way does it c the 12 from c 12 coming from is that we are using isotopically purified form of carbon.

Sebastian Hassinger: 12:02

Okay.

Pierre Desjardins: 12:02

So we only are using carbon 12.

Sebastian Hassinger: 12:04

Okay. Okay.

Pierre Desjardins: 12:05

Like silicon, it's only using Right. For example, silicon 28. So it's even number. Yeah. And so there is no magnetic spin.

Pierre Desjardins: 12:12

Right. No nuclear magnetic spin.

Sebastian Hassinger: 12:14

Right.

Pierre Desjardins: 12:14

So it's contribute to make the the material really And that's

Sebastian Hassinger: 12:19

I mean, that makes intuitive sense that that would make for a very noise free kind of environment for that single electron. And that I mean, it makes a lot of sense that, as you were saying, that the all of those explorations of different materials in spin qubits, the the the central challenge is getting an environment for that single electron or small group of electrons that is that's as as quiet as possible, basically. It's isolated but controllable. So is that that's that nanotube is providing that isolation, but it's also allowing the control of the electron within the nanotube?

Pierre Desjardins: 12:59

Yes.

Sebastian Hassinger: 13:00

Yeah.

Pierre Desjardins: 13:00

So when, for example, like, a number to share about the the carbon nanotube is for any kind of spin qubit, the worst, let's say, noise, the thing that will create the most perturbation is noise coming from charge. On our on our device, we are able to measure the level of charge noise that is like perceived seen by the qubit. Mhmm. And here we have the lowest charge noise if we compare to any kind of spin qubit. Right.

Pierre Desjardins: 13:52

And this is really because with this, like, suspended architecture Mhmm. We screen any kind of charge Right. That we cannot. And Right. This is why we have this profound conviction that the the the winning criteria for to build large scale quantum computer is is very fundamental and very at the heart of the design choice that you met at the beginning.

Pierre Desjardins: 14:20

For example, in the material that you choose for Mhmm. Encoding your qubit. And now you mentioned that Zen, indeed, we solve this like quantum paradox because when you have a super isolated qubit, if you Right. As you mentioned, cannot address it.

Sebastian Hassinger: 14:37

I mean, is the problem with the photonics. Right? I mean, they're they they love to say that they don't decohere, but they're it's because they don't react with anything.

Pierre Desjardins: 14:44

So yeah. So the the this how we actually address our our qubit. And here, we also are using a very, like, also unique approach at least compared to other spin qubit. Mhmm. Because in a way, we we drive our qubit the same way superconducting qubits are driven.

Sebastian Hassinger: 15:10

So microwave pulses.

Pierre Desjardins: 15:11

So we we use microwave pulse.

Sebastian Hassinger: 15:13

Interesting.

Pierre Desjardins: 15:13

And so what we introduced, and it was one of the core discovery of the laboratory of physics at the EcoNormal Superia is that you can use this, like, physics principle of spin photon coupling to, like, drive spin through a microwave photon. Okay. So this

Sebastian Hassinger: 15:37

So you are using photons?

Pierre Desjardins: 15:40

Microwave. Yeah. Yeah. They don't they don't suffer the same loss

Sebastian Hassinger: 15:44

You're right.

Pierre Desjardins: 15:46

Yes. Yeah. Yeah. And, you know, and the the the the first, like, experimental proof of spin photon coupling was done in the physics lab at Economics Bayern, and we really are using this principle to build our quantum computer. So what we introduced is the concept of a quantum bus.

Pierre Desjardins: 16:13

The idea of a quantum bus is like to use a superconducting element to make the interconnects between the different spin qubits. And this is where it's Interesting. Different from, like, other architecture is for other architecture, you connect the spin between each other because they are close to each other and, like, part of their Right. Wave function overlap. Yeah.

Pierre Desjardins: 16:39

And so you are constrained with nearest neighbor Right.

Sebastian Hassinger: 16:42

Or you have to do shuttling, which then brings in all of the Challenges.

Pierre Desjardins: 16:47

Yeah. Yeah. And the idea with the quantum bus is really to, like, be able to do any kind of interconnect between qubits.

Sebastian Hassinger: 16:59

Right.

Pierre Desjardins: 16:59

Even like if your qubits are not near stables. So it has like incredible benefits. First one is that for in in what we're is that it's really the only way to scale spin to bit because if you are stuck with the nearest neighbor architecture, you cannot actually wire all your cubits. Right. So it's what we call the fan out Yeah.

Pierre Desjardins: 17:30

Problem.

Sebastian Hassinger: 17:31

Yeah. And And then especially when you start looking at error correction, the the topology needs to match the the the code that you're gonna use, surface code or flocay or whatever and if you're stuck with just a very small number of nearest neighbor kind of connections, it's it's really really challenging.

Pierre Desjardins: 17:48

Yeah. You you made a very very extremely relevant point. It's why we think that all of this is important, like having high quality qubits, having like high connectivity between the qubits Right. Is at the end because then it's a platform that will be able to support much more sophisticated Right. Quantum error correction coding.

Pierre Desjardins: 18:09

When we say sophisticated, it's like with a much lower overhead.

Sebastian Hassinger: 18:13

Right.

Pierre Desjardins: 18:14

Right. And on this, we have like we it's it's interesting. We we started the company with this very materials oriented background. So we are all focused on and we are still, like, about, like, delivering the the the best kind of cubits, like mastering manufacturing and production processes. And we decided to launch some works about quantum error correction, like, quite late.

Pierre Desjardins: 18:42

Mhmm. Mhmm. Only started to work on this topic last year. But it's also good because the the field of quantum error correction Yes. So progressing so fast

Sebastian Hassinger: 18:56

Very fast.

Pierre Desjardins: 18:56

That probably like the the code that we are currently working on for our architecture did not exist Right.

Sebastian Hassinger: 19:03

Right. Right. Years ago. And before we move on to the air question, I'm still curious about the so how do you address the individual qubits on that bus? What's the addressability?

Sebastian Hassinger: 19:15

Is it different frequency resonances? Or or, like, how does that work?

Pierre Desjardins: 19:20

So we have one line that directly connects to the qubit Mhmm. And then can actually then drive the the qubit. And then it's also why with, like, our technology, it works to have a quantum bus. Because the concept of quantum bus was also, like, tried for also super connecting qubit because it's a lot of problem. You have, like, high connectivity.

Sebastian Hassinger: 19:49

Right. This is why I'm wondering because the you know, they've looked at either tunable connectors or or having the the being able to turn on and off the super conductivity so you can Exactly. Isolate the qubits from one one another when you don't want them to to interfere with one. Is is that

Pierre Desjardins: 20:05

And this is why it works in in our case is that we have a very good tunability and and like we can really tune the coupling between the the the the quantum bus and the spin qubit. So it's actually like very easy to understand how it works. So when the qubit is we can say on, so connected to the quantum bus And really, the qubit is really hopping between, the, two side of the, double quantum dots and then act like a very small antenna and connecting to the

Sebastian Hassinger: 20:47

Oh, okay.

Pierre Desjardins: 20:48

To the bus. But when you don't want to have, like, this coupling, you localize the the the electron to one side of the double quantum dot.

Sebastian Hassinger: 20:59

Okay.

Pierre Desjardins: 21:00

And so then you have, like, a off mode, memory mode where you can also, like, increase a lot of current time because electron is much more isolated.

Sebastian Hassinger: 21:11

Right.

Pierre Desjardins: 21:12

So this is how this, like, concept of quantum buses

Sebastian Hassinger: 21:16

Interesting. Is good. And is are the two dots in the double dot system, are they on the either end of the nanotube? Is it is it sort of physically separated, or is it is it a different sort of design?

Pierre Desjardins: 21:32

It's so you have, like, below the the the carbon nanotube, you have a set of electrodes, and this is what Okay. How they design how we, like, design the the

Sebastian Hassinger: 21:45

quantum dots. And quantum dots themselves, there's some challenges around other ways of achieving spin qubits around the lifetime of dots. There's sort of durability issues with some of the ways that you can fabricate dots don't have a tremendously long lifespan essentially as a material. Is that something that that No. We don't suffer Right.

Sebastian Hassinger: 22:09

From this. Yeah. I was assuming because, again, because of the purity of the the carbon that probably

Pierre Desjardins: 22:15

Yeah. Yeah. And, like, our chips travel and, like, we send some of our chip to to collaborators and, like, measure them, like, for years. And and I

Sebastian Hassinger: 22:28

That's great.

Pierre Desjardins: 22:28

It's actually something where where also where we like, a real difference is that in the stability of the the the qubit over time. So when we have, like, calibrate of cubits, then the the the drift is is very is very, very small. Wow. So it's not like like with superconducting qubits that's

Sebastian Hassinger: 22:57

Right.

Pierre Desjardins: 22:57

Like, each day you have to also, like, recalibrate

Sebastian Hassinger: 23:00

It sounds like you sort of got the best of both worlds with the the flexibility of the connectivity of superconducting superconducting qubits, but then the it's less of an analog system than a superconducting cube. It is where there's constantly calibration and drift and all sorts of

Pierre Desjardins: 23:18

Yeah. No. That's I think that we we have, like, really the speed Yeah. Of like, in term of gate gate duration. Right.

Pierre Desjardins: 23:26

That is very similar to what you have in, like, superconducting qubit, but we really leverage the, like, intrinsically stable spin quantity and also, like, of course, the addition of this very pure material to have, like, a very long current style.

Sebastian Hassinger: 23:47

That's fantastic. So okay. So that's amazing. It's great you get got to five years. As a last question, what do you see as the biggest challenges for the the year ahead?

Sebastian Hassinger: 23:58

Let's just look one year at a time.

Pierre Desjardins: 24:01

So last year oh, no. Two years ago and and no. Last year. Sorry. Okay.

Pierre Desjardins: 24:09

Last year, we realized the first, like, qubit operation on our qubits. In a way, we we we started with really like reimagine how quantum computer are built and we really started from like to first build the fundamental bricks of our technology. So and we presented our results in nature. So we really have like a long coherence time of spin qubit connected to the to the quantum bus on one qubit. And what we are currently like focusing is to realize quantum gates between two qubits.

Sebastian Hassinger: 25:00

Right.

Pierre Desjardins: 25:01

Two then distance qubits.

Sebastian Hassinger: 25:03

Right.

Pierre Desjardins: 25:03

So it's it's very different kind of gate than for other spin qubit because in in in this case, you really have like a spin spin interaction mediated by microwave photon. Yeah. And this is, I think, a very important milestone for c 12. Mhmm. I think also for industry because it really, like, opens the doors for, like, scaling this kind of and we know that also, like, IBM is working on this, like, long distance

Sebastian Hassinger: 25:37

Right.

Pierre Desjardins: 25:38

Qubit qubit, like, quantum gate.

Sebastian Hassinger: 25:41

Yeah.

Pierre Desjardins: 25:41

So what I wish is that we can announce it just before IBM.

Sebastian Hassinger: 25:47

Good. That's great. Well, thank you so much for your time today, Fiora. This has been really interesting. I really enjoyed it.

Pierre Desjardins: 25:54

Thank you, Sebastian. It was a pleasure listening. Thank you.

Sebastian Hassinger: 26:05

You for listening to another episode of the podcast, a production of the New Quantum Era hosted by me, Sebastian Hassinger, with theme music by OCH. You can find past episodes on www.newquantumera.com or on blue sky at newquantumera.com. If you enjoy the podcast, please subscribe and tell your quantum curious friends to give it a listen.