Dan: Hello.

Welcome back to the quantum divide.

This is going to be an exciting episode.

We're going to be talking to a
company who are tackling the art

of quantum networking head-on.

Maybe a dark art.

I think that's up for debate.

I'm joined today by Claire Le Gall from
Nu Quantum, a startup out of the UK.

They're, from my observations, fairly
central to the UK strategy due to the.

membership and leadership in UK
Quantum, and also links to academia.

The company was founded to
commercialise research generated

over the last decade at the Cavendish
laboratory, which is in Cambridge.

So the path to valuable quantum
computing requires novel and

high-performance quantum networks.

Nu Quantum are focusing on the
networking for quantum computers.

It's essentially looking to
deliver a flexible platform, that's

adaptable to all qubit modalities.

Delivering order of magnitude
improvements in rate and fidelity

over the current state of the art.

No mean feat.

Okay.

Enjoy.

thank you Claire for joining us.

So why don't we start with a
bit of your background and where

you come from, your path through
quantum and what it is you do.

Claire: Hi.

Nice to meet you.

I'm Claire Le Gall.

I'm VP of technology here at NuQuantum.

NuQuantum is.

a company that was founded in 2019, and
is a company that I joined about nine

months ago as, so as VP of technology.

I come from an experimental physics
background I'm someone who's um,

played with uh, qubits , since her
PhD , optically active qubits, and I've

been implementing elementary quantum
networks in an academic setting in

particular demonstrating the highest
entanglement rate between distant

qubits that has been achieved to date.

And yeah that's me in a nutshell.

Dan: Yeah, it's fantastic
to have you on the podcast.

Uh such an accomplished
physicist and scientist, what

about your path into quantum?

I guess you studied through
the traditional route.

I think you're at
Cambridge, is that right?

Claire: uh, So , I'm, I've been
passionate about physics from a young age.

I've also been passionate about
maths and philosophy and there's

really like different points in
my life where I thought, am I

going to continue with physics?

But I, I did stick uh, with, with physics.

I then did some undergraduate
studies in France.

And then did my PhD in,, in Grenoble,
which besides Paris was the, you

know, a fantastic place to do
sort of, um, advanced solid state

experiments and quantum experiments.

So during my PhD I worked with a not
so established qubit platform and for my

postdoc I wanted to work with something a
bit more established where we could really

do proper quantum protocols, if you like.

And so I moved to Cambridge
in the group of Mete Atatüre.

Who is within the larger group of
AMOP, Atomic Molecular and Optical

Physics, and I've worked in Cambridge
for 10 years as, yeah as a postdoc

and then a Royal Society fellow.

We've, I've brought in my expertise in
spins and, soon enough we demonstrated

the first qubit control, then spin
entanglement and then, maybe something

we'll cover a little about, I've been
interested in developing the functionality

of this platform and interfacing this.

It's, qubit that we had in, in in our
quantum node with other uh, qubits to

really allow the system to be scalable.

Dan: Great, yeah, we'll
come on to that, definitely.

So first of all let's Talk about
the current day Nu Quantum, you

know, what is it you're doing there?

Tell us what you can about what
you're developing if possible.

What's the ecosystem landscape,
which partners are you working with?

Any interactions with external bodies,
UK government, that type of thing.

Claire: Okay.

So Nu Quantum is a busy working
place with lots of things going on

but they're all organized around.

the idea that we want to develop quantum
networking and quantum networking as a

way to allow to scale quantum computers.

So there's Lots of physics that we have
to unpack to explain, this, but yeah this

is our central mission to provide quantum
networking to scale quantum computers.

And so what am I doing at at NuQuantum?

So I joined the company NuQuantum at
a stage where NuQuantum was engage

in early quantum networking projects.

These started about a year ago
um, to, prove and develop very

sort of elementary blocks that are
required for quantum networking.

So in particular, one thing that you
really need, if for quantum networking

to be fast enough and thereby useful
to scaling, is that you need to

collect photons really efficiently.

And so, NuQuantum has this microcavity
technology which allows to collect photons

very efficiently and has these two
projects with two different end users

One which is to use cavities to interface
with rubidium atoms, and this is a

collaboration we have with Infleqtion,
a US based quantum computing company.

And the other collaboration we have
is Intercom which uses our cavity

technology to interface it with ions.

And this is a collaboration with the
University of Oxford and the NQCC,

National Quantum Computing Centre.

The, then there's like a third project,
which is a third building block

actually of the quantum networking
system that we want to build.

Which is focusing on developing
the optical hardware that

will allow to carry bell state
measurements and high connectivity

between a small number of nodes.

For this we use a integrated
photonics platform.

And we're seeking, massive improvements
in terms of losses and switching

speeds, which are, key metrics again
to achieve quantum networking with

high fidelity and with high rates.

So these are the three key projects, and
then when I joined, I joined just before

NuQuantum started system level programs.

So in these uh, system level
programs, we're looking at using

those elementary uh, building
blocks that we have confidence in.

Of course, they're going to continue
to evolve and improve in terms of

performance, but we're also starting to
look at capturing qubit requirements

for certain certain types of qubits
and building a useful quantum network.

And we have a partnership with the
NQCC which is a four year program to

stand up a four node quantum network.

And yeah, one thing I should say
is, compared to quantum computing

people are already working with
hundreds of qubits in a computer.

And with quantum networking, we're
talking about interconnecting four nodes.

That's just reflecting that quantum
networking is in its infancy.

It's still small scale but
we have to start from there.

And the biggest quantum networks to
date, which has been proven with with

solid state qubit nodes in Diamond and
by the Delft group is, is three nodes.

That's the biggest we can do.

And if you talk to these people, they'll
tell you, we can't scale to four nodes.

It's too hard.

So that's really what
we're trying to do here.

Provide the means to build
this in a scalable way.

And then there's, there's another system
level development that we are pursuing.

Um, This one with Cisco as the
end user, which is to build a

deployable quantum networking unit.

such that Cisco, as the end user
who is building a quantum networking

testbed, can arrive with with qubit
nodes and essentially plug them into

our networking unit and build a small
scale quantum network, multi node again.

Steve: One question I had when
you mentioned the, the nodes,

what consists of that node?

What is that node?

Is that a networking device?

Is it some kind of solid state qubit
or what do you mean when you say node?

Claire: What I mean by node
is either a single qubit, or

multiple qubits of a certain type.

If you take a solid state instantiation
of this is usually um, an atomic

defect that can emit light.

So you can think of this as your
communication qubit which spits out

photons which are entangled with
its internal state, and that allows

you to create distant entanglement.

I'll go into that in a bit, but then
this communication qubit is also

like, um, can be deterministically
entangled with nuclear spins.

Usually that's how it
works in the solid state.

If you look at atomic systems it
can be, a certain ion coupled to a

different type of ions, for example.

And co trapped, and again, so the
key ingredient at a high level is one

qubit that can emit a flying qubit,
a photon, if you like, that you can

route towards a measurement apparatus
to create distant entanglement.

And Other data qubits and, theory tells
you, you need a minimum of five qubits

to actually have the, that's your minimum
quantum register size to, build a quantum

computer in a modular way from there on.

Steve: So in my mind, I'm
picturing something like, okay,

we have a quantum network.

The quantum network is used for
connecting quantum computers.

There's a bit of a far out vision.

So how would you see that particular
node set up in order to do something

like fault tolerant quantum computing
or something like maybe a, how can

it connect to potentially a dilution
fridge with the superconducting

qubits or another technology?

Is it sitting in front or is it inside
the hardware or how do you envision?

Future distributed
quantum computing setup.

Claire: How do we, yeah, how do we
envision the full stack you're asking me?

Steve: Not the stack.

We're just just where does that note sit?

Is it a switch?

, Claire: So the, the, what we
mean by the node, the nodes can

be at least five qubits to all
the way to to a full computer.

With hundreds of qubits but you want
to be able to connect a similar system.

So your node is your elementary quantum
computing module, your elementary quantum

processing unit, you could also call
it your elementary quantum computer.

Steve: okay?

Claire: And then you need
to be able to network that.

To, to a system that is of, that is
exactly a replica of it if you're

thinking about modular construction.

And this networking has to
be fundamentally quantum.

If you just send classical signal
between these two units you're not

really gaining in computational
power from, in the quantum world.

Dan: If I may before we go on to the
sending of quantum state and perhaps some

of the detail on individual components.

I'm really keen to understand your
perspective on why you would network.

quantum systems.

Just because, for me connecting
quantum computers for distributed

algorithms, okay, potentially
connecting sensors of some kind.

But I'm yeah, really keen to
hear your perspective on that.

Claire: For me the profound advantage
of quantum networking is that

it offers a path to build large
system sizes, but in a modular way.

And if you talk to quantum computing
companies even with photonic qubits

you, you can scale up to a certain
system size, and then it just

becomes incredibly complex to just
build the fridge that will host.

This large system size, and also it
becomes increasingly difficult to, to

measure and test to test your system.

So the idea is, we don't
know if the final node.

will be optimum with, if it
contains hundreds of qubit, if

it contains thousands of qubit,
if it contains just five qubit.

These will all be architecture questions
that people will answer and we'll be

part of the people answering these
as time goes on and as we develop our

understanding, but for sure there is
value in the ability of saying We're

going to build, we're going to have
the ability to build the system in

small chunks and then connect them.

That's the first advantage
of quantum networking.

Then there's also a second key
advantage which is, which is more

recent but it's the emergence of
quantum error correction codes that

make use of the long range connectivity.

given by quantum networking
to massively alleviate quantum

error correction overheads.

So just to, to sort of, um, ground it in a
few numbers, we know that with a thousand

logical qubits and, millions of gates,
we can have useful quantum computing.

The issue is to have these logical qubits
with error correcting codes that are

suitable for short range interactions,
then you need millions of qubits.

If you can drop this number to needing,
requiring only 100, 000, you've also

you've also won a lot in terms of how
easy it is to scale the quantum computer.

So these are the key things,
like building quantum computers

in a modular way and having less
quantum error correction overheads.

Dan: Yeah, I'm hearing a lot these days
about quantum error correction, but

then also getting better quality qubits
in the first place so that You don't

necessarily need as much error correction.

And I think all the different modalities
of tackling this in different ways.

Yeah, it's fascinating.

Claire: I think what you mentioned,
the quality of the qubits and how you

really want to optimize that qubit
quality to need to lower your overhead

is also a very important point.

It's a very important point when you
look at quantum networking as well.

because the detractors of quantum
networking, or say, let's say the

main downside of long distance
quantum networking is the fidelity

of the long distance links.

And this is, something we can talk about.

And as as a company is really something
we have, we keep a close eye is like, how

are we going to achieve high fidelity?

networking.

As soon as you lose in fidelity,
you have to think about it as you're

going to have to pay the price.

You're going to have to
distill the entanglement.

And at the end of the day is
the same thing as having high

fidelity, but very low rate.

Quantum networking.

Yeah.

Fidelity, super, super important.

Quali quality.

Quality of the entanglement.

Steve: there's a kind of an interesting
thought that I don't know if other

people had, but I had this thought,
we have a NISQ quantum computers,

intermediate scale quantum computer.

Do you think there's an equivalent
to quantum network, like a NISQ with,

instead of a computer, you have a network?

It's I don't know how you would pronounce
that acronym, but like an intermediate

scale quantum network that does something
meaningful before we go to like full

scale distributed quantum computing.

Is there like an intermediate step or
do we have to go from zero to a hundred?

Claire: It's an interesting question.

First of all, I would really ask
the proof that NISQ is useful.

I think it's still unclear.

It's still an open question.

But for sure, there are NISQ algorithms
which allow quantum computing companies

to benchmark how good their quantum
computer is and to really understand,

fundamentally are errors correlated,
not correlated, et cetera, et cetera.

I think the same works
for quantum networks.

If you're not there yet in
terms of unlocking scalability

uh, there are still metrics.

then elementary quantum networking
protocols and all of this, which are

very analogous to NISQ quantum computing,
that you can use to, to understand the

system and improve its performance.

Steve: Yeah.

Makes sense.

There's a lot of protocols for
quantum networks, but I think they're

all quite challenging to implement.

I agree with that as well.

For example, okay, we have QKD, but that
can use current technology, but then the

question of how useful QKD is, is open.

But I, I believe in the distributed
quantum computing direction, but I think

it's, as I said, it's very challenging
and there's a lot of things that

have to happen before we get there.

I'm just in my mind thinking what's,
could we do something in between

that's easier, but also useful.

It's hard to answer that, I think.

Claire: I think, so they like
to, to enter into, like to give.

examples of things which, need to be
demonstrated and will be, but will not

necessarily unlock scalability just
yet, is, you can perform deterministic

quantum teleportation, teleported
gates and all these are like elementary

quantum networking operation that you
really want to demonstrate and improve

on until you reach certain metrics
that, the below 1 percent error, and

then you're like, okay, now I can
apply error correction on top of this.

Steve: So quantum networking means
different things to different

companies and different people.

For example, people develop quantum
key distribution networks that used

to be called the quantum network.

Now, I think we changed the name of that
to a quantum key distribution network.

Now, when we say quantum
networks, generally people

think about entanglement based.

Quantum networks where you distribute
entanglement and use teleportation

to transmit the qubit state.

What does quantum networking mean in the
context of Nu Quantum and potentially your

own opinion of what a quantum network is?

Claire: So in, in our own opinion, like
quantum networking for quantum computing

is the production of entanglement, distant
entanglement between quantum nodes.

And this entanglement is then, formally
equivalent to having, these two nodes

next to each other and allowing you
to have a bigger quantum computer.

And for this.

For this to hold true, if you want
this distant entanglement to really

allow you to do as if you had a bigger
computer, you need the entanglement

to be high fidelity and high rate.

The challenge here for if you target
quantum computing as an application,

The challenge is if your entanglement,
if your distant entanglement.

is far slower or far worse in quality
and fidelity than your local interactions

within the quantum processing unit.

It means that quantum networking is
helping you scale, but not enough,

not as efficient as it could.

So it's really trying to
build something which is.

providing you fast link, it will
never be as fast as local operation,

local, entanglement with with the two
qubits located next to each other.

But it's, it's really having
this as as an objective.

Steve: So the differentiating factor is.

Do you think it's potentially that it's
a bit near in terms of distance so that

you're looking at, data center style.

Quantum network, or is this
still a project in terms of long

range over kilometers of fiber?

Claire: Oh, so we're that's
a very good question.

Like practically we're completely
focused on data center scale networking.

So if you want to scale the
quantum computer you're going

to build these processing units.

You're going to try to build them in a
robust way that maybe You know, as big as

a room, hopefully smaller, and then you
want the rooms as close to each other and

as fastly interconnected as possible to
then scale that computing power of the

quantum computer functioning as a cluster.

So you want multiple quantum
networking units that allow you

to, to network on a short scale.

I personally don't believe that we'll
ever be able to scale quantum computers

and grow their power using long distance.

Computing long distance, meaning a
kilometer, like we're going to build

a computer, no tens of kilometers.

We're going to build a quantum computer
in this city, in this other city, and

we're going to link them to make them
more powerful that will not work for

the simple reason that it takes time
for photons to go from A to B and that

time is time lost to your computation.

So you want to keep things.

Nice and close to each other, but, there's
this intrinsic size to quantum computers.

They can't be as, they can't be
all with, on, on the same chip.

Dan: So I've got a question
leading on from that.

That's a question from the IT
guy, basically, on the call.

I ask the stupid questions when
it comes to quantum physics.

But what I'd like to understand, let's
go to this small local data center

environment with multiple quantum nodes.

What are the physical devices
or things that need to sit

between node A and node B?

And what are the processes
that have to happen?

Imagine a, the life of a photon
or the information encoded in

a photon going from A to B.

That would really help me
understand what's there.

You know, You mentioned um, different,
you mentioned your photon cavity product.

I guess that fits in there somewhere.

You mentioned different you mentioned
ion traps and rubidium atoms.

I guess they're different quantum
computing modalities potentially.

Maybe that you're just using
them as systems at the moment.

Yeah, if you could try and talk through
that, that chain of events, those chain

of devices, that would be really helpful.

Claire: So that, that chain has
this optically active qubit at the

beginning that emits a photon a
potentially isotropically anywhere

in space and but you want to
be able to collect this photon.

You want that photon eventually
to go into a detector.

So that's where our cavity comes in.

We put a cavity around either an
atom, or an ion, or a, an actively

active solid state qubit to collect
the photons emitted by this system.

And To give you numbers, if you put a lens
you typically collect maximum 10 percent

of the emission of your quantum system.

If you put a cavity, you can get up to
70 percent of emission inside the cavity.

Then that photon that you've collected
with the cavity, essentially the name

of the game is then to propagate it
all the way to your detectors without

losing it, which is just done by
carefully designed optics hardware.

And if you just have two nodes in
the story, it's as simple as sending

photon from one node node one and
node two, towards that beam splitter,

erasing the which path information,
and there are certain detection

events, so behind the beam splitter,
you put two detectors, and if your

two detectors, as we talked earlier,
detects blue photon and red photon,

that's it, you have the signal, that
means you've just created entanglement.

But your detectors, and that's
quite an important thing to say

as well, your detectors might
as well detect red and red.

In terms of photons, and
that's not an entangled state.

And so that means that with quantum
networking there's an intrinsic

thing, which is, you have to
react to an event, the event

that you want, or you don't want.

You have to do different things based
on what is the event that you detect.

That's very similar to what we need
in quantum error correction, or what

we need in one way quantum computing.

as well.

You need to react to an event, the
control system needs to interpret

that event, and then needs to follow
an appropriate course of, control

gates at the qubit nodes, which is
either to start attempt entanglement

again, Or use this entanglement.

Dan: So this isn't necessarily
a point to point process, is it?

There is something in the middle that is
enabling the entanglement between A and B.

So this is your control node, as
you mentioned it, that's ultimately

affecting the entanglement between
the solid state or the, whatever

type of qubits are in the node team.

Claire: So it's like there's a control
hardware layer that is interpreting

signals from photon detection, and
there's a software layer that is then

deciding, and all this has to happen
really fast, but yeah, that is then

deciding this this is what you do if you
detect this event, if that makes sense.

Steve: I think sometimes it calls
like a heralding station or something.

I don't know if that's the

Claire: Yeah it's, heralding
Station to me sounds more QKD.

Steve: Okay.

Yeah.

Claire: But it's a very, as
I said it's a very similar.

Modality, the difference in, in quantum,
because yeah the key thing conceptually

is you need to be able to react to
this probabilistic event that distant

entanglement has been successful.

And having having the ability to
be sure that entanglement happened.

Makes the probabilistic nature of
photon mediated entanglement okay.

If you didn't have the Herald you
wouldn't be able to scale, you

wouldn't be able to make anything
useful out of this entanglement.

But the Herald means that then
you can put some classical control

stuff layer above, and then use
The, the entanglement usefully.

Steve: for the heralding station, it's
like the one that just that piece below

what you do right below the control just
says, okay, we got the event continue.

Claire: oh, okay.

Yeah.

Steve: But yeah, that's, it's very
interesting to hear about Yeah.

The stack and what's necessary.

Claire: If you want to
scale to a few nodes.

Then there's a very simple thing that
happens with all quantum systems, is

that, systems, modalities, is that
usually it takes way more time to prepare

your system in a way that is ready to
for entanglement than to entangle it.

So that means that actually what's really
interesting to do from a hardware point

of view is is to have a switch in front
of your beam splitter and detectors.

So the beam splitter and the detectors is
often called like a bell state analyzer.

And you want some kind of switch
architecture, which which allows

you to, first of all, use these this
Bell State Analyzer in a resource

efficient way and a bunch of switches
allows you to select, okay, I want to

create pairwise entanglement between
node 1 and 2, between node 1 and 4.

And so on and so forth.

Yeah.

So that's the other technology
block that, that we're developing.

It can be made out of, off
the shelf optical components.

But then you're very limited in
terms of the switching speeds

that, that you can achieve.

And so that's why we're also developing
an integrated photonic solution

to network on this small scale.

And then the scaling to a thousand
nodes which is which is, a very

interesting problem because.

Integrated photonics you incur a loss
if you want to increase, if you want

all to all connectivity with integrated
photonics for thousands of inputs, you're

going to incur a massive photon loss.

Which you don't want, because
then that's low rate entanglement.

And so it means you need to do something
more clever above and decide on which

connectivity you want to go for.

So this is also something that we're
looking at, a lot at NuQuantum more

from a kind of top down approach,
where we're understanding for, this.

quantum error correction protocol
what kind of connectivity would

actually support it efficiently?

Steve: I guess one benefit of
this all to all connectivities,

you don't need networking per se.

In a traditional sense, you don't need
routing, you don't need these protocols,

but because everything is point to
point in a sense what about, yeah,

I guess that could be the questions.

Leads up to the next part is
what about topologies that

are not all to all connected?

Then you need some kind of network control
or something that can do switching and

routing based on the state of this the
network For example, do you see that being

a critical factor or is that something
that probably won't apply very quickly?

And it's the same answer probably
to things like classical data

center networks How did they scale?

Probably a very similar approach,
but in a quantum setting, it's

obviously much more complicated.

Claire: so it's a very good question.

I think it's, you know, at the hardware
level it's not like we're worrying about

it just yet and that we're trying to
implement it just yet, I should say.

But We want to think about it now,

That's it's very important to think
about the end point where you are

actually operating efficiently
at the quantum error corrected

Steve: Yeah, so many questions to answer.

It's such a fresh topic.

I think there's so many
open problems to consider.

And it's great to think about it, yeah.

You have to think about it at some point.

It's going to come faster

Dan: Steve there's so many
layers to it as well, right?

You've got the optical
integrated optical kind of stuff.

You then got the software for control,
which has to happen and it's faced

with hardware, but then you've got even
further levels of abstraction, right?

In terms of automating processes.

Claire: Yes.

And yeah, and eventually for, we know
there's like a whole lot of work that

has to go into, taking a useful a useful
quantum algorithm And then breaking it

down into something that can be operated.

Yeah, on the hardware, you're right.

The stack is huge, and
of course, we're not.

We're not developing the
whole stack here at NuQuantum.

Where we see ourselves is really as
the quantum network provider giving

the ability for quantum computing
companies to have an efficient an

efficient optical link to create
distant entanglement between, between

their quantum computing units that
they know how to create so well.

Dan: Hey, one thing that you
mentioned earlier on around

Fidelity of the, the quality of the.

signals of photons going
across the network.

What are the main roadblocks in technology
preventing a perfect photonic fidelity?

Because you have photonic sources, you
have the the cavities you mentioned,

the fibre um, transducers in some cases.

That's one question I want to come
on to is where do they fit in?

But yeah, first of all, the kind
of the quality of the photonic

information, what's standing in
the way for getting that right?

Claire: I'm going to give like a
numbered answer to just give you a

sense of where the challenges are.

I think that the challenges
are the following.

You have roughly, say, 1
percent of error at the qubit.

I'll, I can explain a bit
more what types of error.

You have roughly 0.

1 percent in the classical
hardware, and you have 0.

01 percent in electrical
and control hardware.

, so the difficulty still is With the
qubit control whether it's a solid

state system and it, it couples to
stuff you don't want, it couples

to its solid state environment,
so it's not like a perfect qubit.

Or it's a trapped atom and similarly
it will couple to vibrational modes

to the environment and, not give
you like perfect perfect behavior.

that's, That's still where
there's a lot of work.

When you bring in cavities into the
picture, it's it's really interesting.

There's uh, there's the potential
depending on the regime in which you

operate, that cavities add Contribute
to the optical hardware noise.

So you have to be careful
that this does not happen.

And Also, the the sort of photon
generation part is, requires

a cavity tailored protocol.

We're also, yeah, we're also
thinking about, entanglement schemes.

and things like this once, once we
focus on a certain qubit modality.

Dan: So on the modalities I'm just
leading through the questions here as I'm

thinking you mentioned rubidium atoms.

Is that the same as a neutral
atom type concept, right?

So are these rubidium gas
atoms or are these solid state?

Yeah,

Claire: These are, yes, neutral
atoms in a vacuum chamber.

I'm not going to say similar operation to
trapped ions because they're trapped in a

different way, but you know, kind of the
equivalent where we just haven't taken

out that electron away from the atom.

So yeah, a neutral atom modality.

And the reason for this is that,
neutral atoms and trapped ions, they're

the best in class qubits these days.

They're the ones with the
lowest qubit gate errors.

So they're a fantastic platform
to benchmark the capability of.

Of your networking link.

Dan: that's why I asked is it, are there
any, which from a networking standpoint

looking, are looking better for releasing
photons in such, in a more reliable way

and allow the interfacing between the
flying qubit side to the fixed qubit side.

Claire: There's a lot to be.

Said for solid state systems.

Obviously, I know them very well,
having worked for, 15 years with them.

Um, With solid state systems you
don't have to worry about trapping

like Using lots of lasers building
an ion trap cooling the system.

There's there's a lot of
simplicity that comes with it.

And you can potentially build the system,
you can really build devices that are

then left in a cryostat and operate almost
autonomously, whereas with atomic systems.

It's tough, but some people doing
it though, like with, building iron

trap quantum computers that operate
autonomously, et cetera, et cetera.

But there's definitely, it's
definitely easier to get there

with a solid state system.

That's what I mean.

The other advantage of solid state
system, generally speaking, is is

that the gates are faster they emit
photons faster, they also have shorter

coherence time, nothing comes for free
but overall they're in an operation

regime which is, every, everything is
fast which means that The cycle time of

your future quantum computer is faster.

And that's also quite an important
thing from an application perspective.

So I, I, you know, I do believe
that, solid state systems are

ultimately where we want to go.

But right now there's no
perfect solid state system.

There's this.

This funny thing where either solid
state systems have, really good

qubit properties, long coherence
time, blah blah blah, but crap

optical properties, or the opposite.

Great optical properties quantum dots, the
system I was working with was an example

of that, like great optical properties
but the spin coherence was short and it's

like an issue to scale with this modality.

There's no perfect system, but people
are looking so, you know, you have

to keep um, you have to keep posted.

But the fact that a three node quantum
network has been realized with NV

centers in Diamond is, illustrates
that, solid state systems are out there.

You know also really interesting
as quantum networking as a

modality for quantum networking.

Dan: The more I hear you talk, the
more delicate I feel the whole setup

is with so many different factors at
play and levers you've got to, you've

got to play with to try and optimize.

Uh, you mentioned the, the quantum
dot uh, semiconductor quantum dot

platform that you're working on,
looking at some of your academic work.

There's lots of papers on spin
and spin dynamics and so on.

Did you want to elaborate a bit more on
the work you've been doing there and take

a bit of a detour from the current topic?

Claire: So quantum networks when I you
know, when I started working with them,

I mean they were already very well
established as a fantastic source of

photons photon indistinguishability
is quantifying, your ability to do a

quantum interference between photons,
so ultimately, It's the thing that

feeds into the distant entanglement
fidelity, if you see what I mean.

and so quantum dots, you can generate
these beautiful photons, which will

give you high high entanglement quality.

Which is super cool.

But the problem at least with,
the types of uh, the qubits that I

was working with is that the spin
coherence was limited to a microsecond.

And there was none of the sort
of quantum control that you can

do to extend the coherence time
of the qubit that would work.

And so So, I've been working
first on fixing that, that

this was like a material issue.

We tweaked the material of this platform
and we demonstrated, nearly a hundred

fold improvement in the coherence
time of the spin qubit in this system.

So that's like one, one first
thing that, that we demonstrated.

And then the other thing Is that for
scalability, you want a quantum networking

node, which has multiple qubits.

And we, in the system, we
only had, we only had one,

one qubit, the electron spin.

And so we've been working towards
Using the nuclear ensemble and the

collective states of the nuclear
ensemble as as a qubit register.

So this is work in progress but
yeah, follow closely because there's

really things moving on this front.

And this is a game changer because
once you have two qubits and the

potential for more, it's completely
different having just one qubit.

One qubit, you create
entanglement, and now what?

If you have a second storage node,
you can then store the entanglement

on that storage node and entangle with
a third node and create a, a three

node GHZ state, or that sort of thing.

Or you can do deterministic
quantum teleportation.

So these are more like applications geared
towards long distance quantum networking.

As opposed to what we do on NuQuantum,
but it's, yeah, it's very key to

have multiple qubits and that's,
that's what we've been working on.

Dan: Fantastic, thank you.

A lot to get my head around.

Thanks very much.

And all working at such a small
scale, that's what kind of

blows my mind a lot of the time.

It's just, it's stuff that
you can't really see, can you?

Without some very clever imaging.

Yeah.

Claire: It's really interesting like, most
of these um, emitters, you can actually

see the photons they emit on a camera, or
sometimes with your bare eye, you can see

the fluorescence from an eye and you can
see the fluorescence from a quantum dot.

Yeah.

But you can't see the little
structure unless you use a scanning

electron microscopy, for example.

So yeah, it's,

Dan: Yeah.

Claire: They're tiny.

. Dan: And what you were talking about
there was rather than using an individual

electron, which is something released
from an atom or Still within an atom.

You're talking about the whole
atomic structure itself and

using that as coding a qubit,

Claire: in, in uh, Yeah in, in
semiconductor systems generally what you

can do is you simply, very simply, you
can dope the semiconductor to create the

free electrons, and then you can also
create a blob of a different semiconductor

material, which acts as a quantum well
and will capture one of these electrons.

And then you have A system which
is behaving as if it was a single

electron and, by the way, people have
also done this with ions instead of

having an ion, they're trapping a
single electron, that, That is a thing

as well, but here you're doing it in
the solid state and the trapping is

realized by a semiconductor material
which acts as a potential well to trap

that single particle, single electron.

And then you just play with atomic
transition in, in your solid state

system, which are exactly the same
as in, in a, an atom, which is why

they're called artificial atoms.

But yeah, they're, it's all the
same thing to a certain extent,

Dan: Listen, that's fantastic.

When it starts to wrap up a little
bit uh, yeah, I mean, like I said,

so much for me to think about now and
hopefully that's been a riveting thought

inspiring conversation for our listeners.

I guess just finally going back, coming
back to the UK as a whole I know Nu

Quantum's fairly involved with uh,
UK Quantum, which is the organization

representing the UK quantum industry.

Is there anything anything you'd like
to comment on that membership or perhaps

the activities that are planned for 2024?

Claire: so yeah, the, maybe I should
start with what UK quantum is.

And when it started, which was about
three, three years ago, unofficially

so the UK government asked a group of
six companies to talk to each other

and agree on what was important to do.

So the six uh, UK companies in the quantum
industry are, NuQuantum, Riverlane, BT,

BAE Systems Orca, and Oxford Instruments.

And so the, these companies have
been talking and then advising the

government on, on, on a quantum strategy.

And then after, one or two years
of preparation, they launched

officially In November 2022.

And the first year was super successful.

There's over 50 members that joined.

There were many events,
organized webinars, opportunities

for members, et cetera.

And then they've set up six, six
working groups across government

strategy, international strategy,
supply chain, which is Like, we haven't

had the time to cover everything,
but there's so many things to think

about when you're doing quantum.

Supply chain is an issue.

Market use cases if you're
aiming for quantum computing,

what does that look like?

Resourcing Is also worth thinking
about in, in, in the quantum sphere.

Ultimately they've advised the government.

And the quantum strategy was
published in February this year.

And the quantum mission published in.

And so Carmen, our CEO is one of the
four directors of UK Quantum and is

involved in steering this this committee.

And yeah, It's a really important
initiative it's really how we can

bring cohesion into into a progress
towards, end user applications, I think

because with quantum, I guess you're,

you're at risk of going in all
directions and so it's really good

that there is there is a, this
unified collaborative thinking

about what it is that we should do.

Dan: Yeah, very good answer.

Thank you for that.

, and I'm looking forward to seeing, the
next year or so with the increased funding

from the UK government, the more stimulus
going into Innovate UK the change at

the end of this year with the with the
different quantum hubs that's impending.

So yeah, it's an exciting time.

Claire: It is hugely exciting.

The pace at which the quantum
industry is progressing is.

I'm sure some people are like, Oh,
there's so much hype, blah, blah, blah.

But and there is as well, but
there is also so much real tangible

progress and it's, yeah, it's
an exciting field to work in.

Dan: Superb.

Okay, well I'm going to close there.

Thank you very much, Claire.

Really appreciate your time today.

Claire: You're welcome.

We haven't talked about
superconducting and transduction.

I apologize for that

Dan: yeah.

I think that sounds like an
excuse for a second episode

at some point in the future.

Heh heh

Claire: Please, let's uh,
let's, let's, let's not have it.

I haven't thought about the issue yet,
but all I can say is for us, it's somebody

else's problem for now, because there's
a lot for us to think about already.

But, that's We want to talk to, we
want to talk to a company like QPhox

for instance, who's working on the
transduction problem, because at the end

of the day it is all about understanding
what the others are doing and how we

can interface and help them scale.

But the capturing this, this these
requirements is a many month effort.

That's why we had to pick and
choose at NuQuantum and we have to

be like, okay, for now we're just
going to focus on atomic modalities.

They're quite mature, they're
quite ready for quantum networking.

Let's do it.

Dan: it's a system, isn't it?

And you're doing the right thing, right?

But staying focused, I think it's
very easy to get lost in swathe, like

all the different technologies, the
different ways of implementing things.

And when it comes to working
in the lab, I'm sure that

makes it even more complicated.

Yeah, it's great to see that.

It's great to see that.

It's great to see it in the UK.

Claire: Nice.

Dan: Thanks for coming.

And we'll talk to you soon.

Claire: Thank you.

Dan: I'd like to take this moment to
thank you for listening to the podcast.

Quantum networking is such a broad domain
especially considering the breadth of

quantum physics and quantum computing all
as an undercurrent easily to get sucked

into So much is still in the research
realm which can make it really tough for

a curious IT guy to know where to start.

So hit subscribe or follow me on your
podcast platform and I'll do my best

to bring you more prevalent topics
in the world of quantum networking.

Spread the word.

It would really help us out.