Subscribe
Share
Share
Embed
In this episode of The Quantum Divide, Dan and quantum dot expert Poolad, CEO of Icarus Quantum, delve into the intricacies of quantum dots and their pivotal role in advancing quantum computing. From the fundamental principles of quantum mechanics to the latest in NIST's quantum research, we uncover the technical nuances and potential applications with quantum dots. Join us for a deep dive into photonic behaviour, quantum entanglement, and the innovative uses of quantum dots in today's tech landscape.
Key Discussions:
- Quantum dots & their applications
- Icarus quantum, University of Colorado, NIST and JILA
- Insights from NIST's latest quantum research
- The intersection of photons & quantum computing
- Poolad's perspective on quantum technology's future
Creators & Guests
Host
Guest
Poolad Imany
Quantum information expert and the founder and CEO of Icarus Quantum, a startup that develops and commercializes chip-scale integrated quantum photonic technologies. With over 10 years of experience in quantum networking, I have contributed to multiple publications and patents in the field, and worked at leading research institutions such as NIST and Purdue University. My core competencies include designing, modeling, fabricating, and testing quantum photonic devices and systems, as well as leading and managing interdisciplinary teams and projects. My mission is to leverage the power of quantum information to create innovative solutions for communication, computation, and sensing. I am passionate about advancing the frontiers of quantum science and engineering, and fostering a thriving culture in the quantum community.
What is The Quantum Divide?
The Quantum Divide is a podcast focussing on quantum technology, with a slant towards networking. Do you work in the IT industry, want to learn about quantum technologies, and are daunted by the extraordinarily high bar for deep comprehension? Yeah, me too. Join our podcast to hear interesting talks about quantum technology concepts, and fascinating interviews from individuals in the industry.
(Opinions expressed by Steve & Dan belong only to them, and not their employer)
Dan: Hello, and welcome
back to the quantum divide.
This is episode 22, actually.
Which is pretty exciting.
Thanks again for tuning in.
I'm sure you've missed me since last week.
I want to tell you about
my guest this week.
So I've got Poolad Imany who is joining us
from the U S he is CEO of Icarus Quantum.
Which is a spin-out from.
NIST national Institute of
standards and technology.
And, uh, university of Colorado.
So he has a PhD from Purdue university
in electrical and computer engineering.
, where he then went to work.
And here we are with this company
Icarus . Icarus is a spin out of the
quantum nanophotonics group at the NIST.
Icarus is developing on demand, single
and entangled photon generators for
scalable quantum networking applications.
So really what we're talking
about here is quantum dots.
But specifically a family of
quantum dots based on the gallium
arsenide, semiconductor platform.
Icarus had some interesting
collaborations with the US public sector.
So we're going to explore
that and dive into the.
Deep side of the technology,
as well as I always like to do.
Okay, welcome Poolad.
Welcome to the podcast.
Poolad: Thank you.
Dan: Thanks for joining us.
Yeah, great.
Let's let's start like I always
do with a little bit of an intro.
So I'm keen to know some of your
background, but also your path into
quantum would be super interesting.
If it's just through academia
or if there's any quirks in,
in, in the path that you took.
Poolad: Yeah.
So I did an undergrad in electrical
engineering and also went on to
do a PhD in the, same electrical
engineering focused on optics.
So for my undergrad, I
did that back in Iran.
I'm Iranian.
I grew up in Iran.
And after undergrad, I joined Purdue
University to work on optics, but
with one of the big names in optics.
Frequency combs and pulse shaping.
So an ultra fast optics.
And yeah, this, my advisor's name
was Professor Andrew Weiner who
was the inventor of pulse shaping.
And coincidentally, at the time
that I joined the group, he
got this project about, can you
actually shape quantum light?
Can you do pulse shaping
on single photons?
And it was right at the same time that I
joined the group that he got this project.
And it was very open about what
project do you want to work on?
I was like this looks amazing.
And so that's how I entered quantum.
And so from, through quantum
optics, and that was 10 years ago.
And yeah, fast forward, I'm
still doing quantum optics.
With a bit of a more scalable approach
now but the gist of it is the same.
Dan: And the rest is
history, as I, as they say,
Poolad: Yeah,
Dan: so you're, you've been at NIST.
Are you still at NIST?
Is that right?
Poolad: I'm still at NIST.
Yes.
I, after my PhD, I joined NIST
as a postdoc four years ago and
the quantum nanophotonics group
still working with quantum light
but with a more scalable platform.
So during my PhD, I was working with micro
resonators, And ways to generate quantum
light, let's say probabilistically.
So this was generating entanglement
with a process that you just
don't know when it happens.
It's very inefficient, but it gives
you, yeah, very cool and high quality.
Quantum light and entanglement I didn't
think that this was the way to the future.
So I, when I moved to NIST uh, here we are
it was a project uh, work on quantum dots
and to figure out how to actually connect
different pieces of quantum networks to
each other through quantum transduction.
The project that I was working on was
going from a superconducting qubit,
convert the quantum information from
there to surface acoustic waves,
so to vibrations on the surface,
so the same way that you can
quantify light and to photons, you
can quantify sound and to phonons.
And which
um,
Dan: I'm I'm smiling.
Poolad: episode.
Yeah, that I was listening to from QPhoX.
Yeah, which, yeah, Rob really
explained that beautifully.
But so, yeah, going from a superconducting
qubit phonons in our case, we were,
we had another intermediate quantum
system, which is a quantum dot.
And then the quantum dot would
generate your photons for you.
And so you would go all the way from a
microwave photon to an optical photon.
And you can do long range quantum
networking of quantum computers that way.
And so, yeah, that, that was the
project that I was working on at NIST.
And yeah, a couple of years into it,
we realized the quality of the quantum
dots that we're making are very unique.
And.
There is actually a demand on
the, industry side for them
and get a quantum networking
effort is taking off everywhere.
We started Icarus quantum
to focus on making quantum
light using our quantum dots.
Dan: I'm going to ask you about
some of the details around
quantum dots in a moment.
But first of all NIST, so National
Institute of Standards and Technology,
not being an American, I guess, I
don't fully understand how all the
different institutes are connected,
but I understood that NIST really
was there to develop standards.
Of course, everybody knows them for the
uh, post quantum cryptography standards.
But it sounds like you studied there.
We studied with them.
And is there a university as part of the
organization or is it a sponsorship thing?
Poolad: So definitely there is.
So here we're based in Boulder so
we're NIST Boulder, and NIST has two,
actually maybe three campuses, but two
giant campuses, one in Gaithersburg,
in Maryland, and the other one, which
is us, it's in Boulder Colorado,
and it has a very close relationship
to University of Colorado Boulder.
There is actually an entity that is
jointly owned by NIST and the university.
The university is called CU
Boulder not to be mistaken by,
with the University of California.
So they flipped C and U.
So it's called CU Boulder.
But yeah, there's this entity jointly
owned by CU and NIST called JILA.
And so there are professors there,
there are researchers there, and there's
really this amazing synergy between the.
The two entities and actually when I
was hired by NIST, because I was not
a citizen, I was technically hired
by the university, even though I was
just working at NIST, but I was a
NIST contractor, but a CU employee.
Dan: Got it.
Okay.
So JILA is the, is the astrophysics and
optics, physical science type center.
Is it, is that right?
It sounds a bit like a kind of
industry meets academia, but
specifically for public sector.
Do you say?
Poolad: Yes, it is kind of looking at
the looking at NIST and comparing it
with the university that there are, much
bigger projects that you can work on.
And there is, because people are not,
there's not this large turnaround
of people at a university, if
you're a PhD student, then you
eventually graduate here at NIST.
You can work on bigger projects and.
We're trying to make these
as collaborative as possible
to the university as well.
And yeah, we do in our group at NIST,
we do have graduate students who are
again, as a result of this relationship
between the university and NIST, but
yeah, they work at NIST facilities and.
Yeah it's amazing to actually be
exposed to all these, like the
clean room facilities and the
labs here as a graduate student.
Dan: Thank you for explaining
that to the daft Englishman.
Just trying to piece it together.
Hey, let's come on to your company.
And also Elevate Quantum, which is how I
found you or you find me, I should say.
Tell me a bit about your company,
how it's obviously it's a spin
off from the university or somehow
involved with NIST, through funding.
Give us a view of the overall scope
and scale of it at the moment.
And what your plan is, that
would be good to know as well.
Poolad: Yeah.
So we started Icarus
Quantum two years ago.
Just to reiterate commercialized
quantum dot quantum light sources.
And right now we have a very
close relationship with NIST.
We are working with CU as well.
So we have Right now, quantum dot growth,
nanofabrication, characterization,
everything's happening at NIST.
But with the university, we do
have an amazing scientific advisor
Professor Shuo Sun who joined
the course at the very beginning.
And we are doing a lot of collaborations
with him and his lab as well.
And he, his research is very
much focused on quantum emitters
which is a more general term.
So not only, limited to quantum dots,
but defect centers and diamond as well.
But we are working with him
to figure out ways to increase
the efficiency of our devices.
Yeah.
So.
Looking back, yeah, we started where
the unique value proposition that we
have is that we are doing it, making
quantum dots and actually putting
these types of quantum dots that
we're making into something is uh,
Very, I wouldn't say risky business,
but it needs a lot of iteration.
There are many things that are not
understood about these quantum dots.
So it's I would say it's very much
an art at this point until we figure
out everything piece by piece.
So the fact that we are doing everything
here at NIST facilities and that we can,
from the growth of the quantum dots to
nanofabrication to characterization,
and then giving feedback.
To growth that can take three, four days.
So that is very important that we can
do this and we can increase the quality
of our devices just step by step.
And that's what we've been
doing since we started Icarus.
Again, in a very close collaboration
with NIST NIST says Not only
funding us, but we have an official
collaboration going on through other
government contracts together as well.
And so just, yeah, making the
devices better and slowly growing
the team at Icarus as well.
Dan: That's great.
It sounds like it's almost
like an auto scaling machine.
Like you've got the power of NIST behind
you with all of access to their labs
and some of their resources and so on.
I expect, right?
Poolad: Yeah, it is.
It is an amazing facility.
Like looking at yeah, all the
resources there are here uh, for.
Of course, this is all happening as
basic research, and that's why NIST is
interested in moving forward with us.
But, quantum networking right now
is pretty much still basic research.
Dan: Definitely.
Poolad: Yeah, we are
very grateful to NIST.
Yeah, it's been an
amazing experience so far.
Dan: Have you got competition?
Are there multiple organizations like
you doing the similar kind of things?
Quantum networking, you've already alluded
to some of the different sources and the
different ways you need to move between
different wavelengths or whatever it is
you're doing with the quantum information.
What does that look like?
Poolad: In the quantum dot space,
there are a couple of amazing companies
in Europe who are working on that.
With the same platform, like
Quandela, SparrowQuantum, Aegiq and,
Dan: But in NIST, I
Poolad: uh,
what
Dan: within the NIST organization.
Poolad: oh, I see within NIST there
is a very unique dynamic here that
I haven't much seen in the, the
higher up academic institutions that
the collaboration level here, it
really, it trumps the competition.
So everybody wants to collaborate
with each other much more than, oh, I
want to beat this guy, this or that.
So I don't know if that's what you
meant by competition um, at this.
Dan: yeah, if they've selected you
to be the quantum dot manufacturer
and point of expertise within their
ecosystem or whether they're, whether
there's some competitive play, it's
quite often these public sector bodies
like to fund lots of the same thing to
see which one can come to the top, if
Poolad: yeah, so in our case,
that actually came about more
organically, I would say that
we we were doing this research.
And even though our group is a very
large group, it's a group of 40 people,
the people who are working on the
quantum dot project there's five of us.
And so we decided that
yeah, we should do this.
Start and commercialize this.
And so going about that, everybody's
been like from the students all
the way up to our group leaders and
division chief, everybody's been
very supportive about the efforts.
And it's not so much that we were picked
to be the only, quantum dot manufacturer.
It's just what, yeah, we're
the ones doing it right now.
Dan: Cool.
Thank you.
Yeah.
Just grew out of the ecosystem
then by the sounds of it.
Yeah.
Okay.
Let's jump onto some technology.
Shall we?
You give me a bit of a detailed
walkthrough of what quantum dots are?
Are there different types?
What are the pros and cons of them?
Which are the ones you use and why?
And what are the idiosyncrasies of
setting them up and managing them and
hosting them on a platform of some kind?
What's, give me a bit of a brain dump.
I think that'd be great for our
listeners as well to get that.
Poolad: Yes, of course.
Quantum dots are generally just
nanometer sized particles that can
show some type of, quantum effects.
And there are many different
kinds of quantum dots that you can
make by, in very different ways.
So you can actually make
TVs out of quantum dots.
You can make solar cells
out of quantum dots.
And you can also use them for quantum
technologies as quantum light sources.
So in our case the way that, uh, our
quantum dots are made by molecular beam
epitaxy, which is You just grow this
nano sized nanometer sized island of
one material on top of another one,
just atomic layer by atomic layer.
And whether it's due to strain, whether
it's due to you've already made a
hole in the other material and you're
just filling it These atoms are going
to sit right next to each other in a
crystalline lattice because the materials
that we are growing are crystalline.
So you can think of it as one
crystal and we're growing another
crystal right on top of it.
And in the case of, in one species of
the quantum dots that we're, we work
with, this is the, the islands, the
quantum dot is formed based on strain.
So we are growing.
Indium arsenide on top of gallium
arsenide and the crystal structure
of these two are very much similar.
It's just the lattice constant or the
size of the crystal for one of them
is slightly smaller than the other.
And so it's not different
enough that these two, these
crystals won't bond to each other.
So it's when we're growing this
other crystal on top of another
one, it is still they're bonding
and they become, one wafer.
But because of this tiny difference
between the crystal sizes, then
there's going to be a strain.
induced on top of the other material.
So it's going to pull it
until an island is formed.
And that's how we're making
the quantum, our quantum dots.
And it is really important if you want
to use these for quantum technologies,
that there are no defects around.
that everything is as
pure as you can make it.
And that is why we're using molecular
beam epitaxy and atomic layer by atomic
layer growth of these quantum dots.
Because even though these have
thousands of atoms inside of them,
we want to treat this quantum
dot as one giant artificial atom.
So we want it to have all the
quantum properties that an atom in
a very pristine condition that can
we want uh, we want the quantum dot, uh,
show the pristine conditions of an atom.
That, that is why we call
them artificial atoms.
Dan: Cool.
Thank you.
So first of all, I'm
thinking you're a crystal.
First of all, I imagined you with
a watering can, maybe a hoe, just,
but now it's atom by atom layer,
but almost sounds like 3d printing.
Is it a chemical reaction or some
kind, or is it a, how are you
planting the individual atoms there?
How does the process work?
Poolad: Yeah, so the machine that
we're using, which is a very expensive
machine for this, it works at ultra
high vacuum, and the process is at high
temperatures of 600 degrees Celsius.
And the way that it works is,
your wafer sits in this chamber,
and then there are cells.
In this chamber, cells of just
different materials, and we're just
opening these cells and bombarding the
surface with these different materials.
And so they come and sit.
And because, yeah, the crystals.
Match, more or less, with each
other, then they sit on the
wafer and they bond to the wafer.
After they touch the wafer,
they're part of it, from then on.
Dan: Got it.
Okay, cool.
So you end up with these quantum
dots and, You're talking about
them behaving as one atom.
Is that a bit like a Bose Einstein
condensate, in a way, where you
have multiple gaseous atoms in a
cloud which act like a, you know,
you can perform individual quantum
operations on a group of atoms?
Is it the same in quantum dots?
Am I understanding that correctly?
Poolad: More or less, I don't know much
about Bose Einstein condensates, but the,
the, from my limited knowledge what I
understand is that the main difference
between quantum dots and, a, Vapor cell
is that this is a solid state material.
Dan: Yeah.
Poolad: everything is in this
crystal structure and every, all
the atoms are sitting exactly where
they're supposed to be sitting.
So even though they're moving, they're
vibrating but the nice thing is you can
cool them down and they stop vibrating.
So in that sense.
There are with both Einstein
Condensates, there are quantum
properties that you can probe and use.
But there are some things that are out of
reach unless you trap individual atoms.
Whereas with quantum dots,
you can actually treat this
as a trapped individual atom.
Dan: Got it.
Okay.
Thank you for the
layman's response for me.
Next question then is, why are you
creating these, these quantum dots?
You said they're photonic
emitters, what's the use case?
Poolad: So the way that we're
using these quantum dots are as
deterministic quantum light sources.
And by quantum light, single photons,
entangled photons, a chain of photons
that are entangled with each other
in whatever ways that you want.
And quantum dots can generate all of
these and in principle deterministically.
And what I mean by deterministic
first of all, let me explain how we're
generating light from these quantum dots.
So the way that happens is.
With an atom or with a quantum
dot there are different energy
levels and they're quantized.
So if you are Exciting the quantum dot
somehow, you can do that electrically.
You can do that optically.
In our cases if you want the Property
of the quantum light that it emits
to be as high that you can use it
for quantum networking, then you
have to Excite them optically.
What we do is we go zap a laser onto
this quantum dot, and it goes from
the ground state to the excited state.
And it stays at the excited state
for some time, until it decays,
until it goes back to the ground
state, and it emits a photon.
And we can do this whole
process deterministically.
We can deterministically excite the
quantum dot and the quantum dot every
time that it emits every time that it
decays, it gives us a single photon.
And so the process, the
whole thing is deterministic.
And so this is the way that quantum dots
can generate single photons, and they've a
lot of groups are using these as the best,
the most efficient single photon sources,
and they've been breaking the record for
single photon efficiency year after year.
What we're more interested in
is to generate entangled photons
deterministically with them.
So instead of going level
up, we go two levels up.
So let's think of this as
exciting the quantum dot twice.
So it goes instead of going to the exciton
state, it goes to the biexciton state.
And then when it decays,
it gives you two photons.
But the cool thing is that two photons
can, are entangled in polarization.
So there are two different decay channels.
They can both be horizontally
polarized when they're emitted, or
they can both be vertically polarized.
And so that gives us an
entangled photon pair.
And this process can happen
deterministically as well.
Dan: I'm going to ask you a
whole bunch of questions now.
Poolad: Yes, please do.
Dan: So first of all, photons, I believe
when they're emitted from something,
which is changing energy levels,
they can be emitted in any direction.
Is that the case?
And if so, how do you catch them?
Do you have a little uh, little
wave guide or something that
you guide them into somewhere?
Poolad: Yes.
So that is with most of the quantum
emitters, the main challenge is, yes, how
do you catch the photons that are emitted?
Because yeah, they can emit
in all different directions.
The cool thing is if you put optical
cavities around the quantum emitters
it'll not only capture the light that
is emitted in that particular direction
that is in the quote unquote cavity mode
but also it forces the quantum emitter
to emit only into that cavity mode,
depending on how good the cavity is.
So if you have a cavity that is uh, and
you can quantify how good a cavity is
by its quality factor and think of this
as if I'm putting a, an optical cavity
is just two mirrors and we're putting it
a quantum emitter right in the middle.
And so how good a cavity is just
meaning how reflective the mirrors are.
And if the mirrors are
very reflective, then.
Most of the light and by most, I mean
more than 90, more than 95 percent
of the light can be emitted into only
this cavity mode, even though if the
cavity did not exist, the quantum dot
could or the quantum emitter could emit
in all different directions after the
cavity is placed around the emitter,
then it changes the properties of the
emitter in a way that It will most
probably it will more probably emit in
the cavity mode and in one direction
as opposed to all the other directions.
Dan: The only analogy I can think
of there is the wave function.
Is the wave function of the emitter
or the whole system affected when
you introduce the cavity then?
Is that, or am I just connecting two
things, which is totally different there?
Poolad: I don't know if we can think
about it that way but you might it
is, so the way that I like to think
about this is when you put the cavity
around an emitter, then it there's a
number which is called Purcell factor.
And the emission into the, into
that particular direction is
enhanced by that Purcell factor.
And when that Purcell factor is
higher and higher, Then yes if
you're looking at the emission of the
emitter, that is, that has changed.
By the cavity and it's it's influenced
by the, so definitely the, if you're
looking at the spatial wave function
or even in other in frequency or in
time, it actually changes the wave
function of the emitted photon.
So it can actually make the quantum
emitter to emit faster as well.
So not only in one
direction, but also faster.
So faster, meaning once you excited, then.
The decay is just going to
happen at a fraction of the time.
Dan: Okay.
And do you always go down to two
energy levels down to the ground state
every time or does it not matter if
it's already a level 50 or something?
Or do you just not, I
guess how discreet is it?
How specific is it in
terms of energy levels?
Poolad: With a, with an atom, it is
a rather easier, still a complicated
question, but with an atom, it's
easier to answer that if there are
two levels, then you get excited you
excite this atom, it takes some time,
so the excited states has a lifetime.
And meaning after this lifetime,
it is more probable than not
for the for you to find the atom
in the ground state after that.
And so it's just it's a probability game
that after a long time, then if you're
waiting for 10 lifetimes, then yeah,
most, most probably you're going to
find the atom and the ground state for
quantum dots, it's a little bit different
because there's many energy levels.
So that especially in a crystalline
structure because you have so many
atoms Instead of just two levels.
You actually have two bands two energy
bands Which is a result of atoms sitting
together in a crystal if you have all
these atoms like in a vapor cell, then
they have their individual, energy levels
and you can address them individually.
But in the case of the quantum dot,
you have this kind of collective energy
level which is now these two bands.
So we call the.
ground band, let's say we call
that the valence band, and then
there's another band above it,
which is called conduction band.
There are no allowable energies
in between the valence band and
conduction band but once you go below
the valence band, there are many
energy levels that you can occupy.
And once you go above the conduction
band, there are many energy levels.
So what we do by exciting the quantum
dot, we're going to go match the energy
of the laser that we're exciting it with
to the difference between the energy,
the valence and the conduction band.
And so that way.
If we were able to initialize the quantum
dot at the top of the valence band,
then it will jump to the only available
option that is in the conduction band,
which is the closest level up there.
And that way we know once it decays, it
goes hopefully back to the ground state,
to the ground state that we started from.
And it get more complicated than
that because it can go to other
ground that you don't want.
But there a cavity can help too.
So you can make a cavity
that only resonates with
the mode that you care about.
So once a quantum dot emits, you
know exactly which mode it emits at.
And this is only one way of exciting them.
You can actually excite the quantum
dots from different ground states
to different excited states.
And there are different methods.
It's just this is one particular method
which has shown to give you the most
let's say that the purest and the
most indistinguishable photons and
indistinguishability is really needed for
quantum technologies, because if you have
10 photons that you want to interfere with
each other, then you want all of them to
have the exact same frequency, the exact
same polarization you want And so on.
So if you're using this quantum dot
10 times, you want it to give you the
exact same photons every single time,
which is why we're using this scheme
of excitation that I mentioned.
Dan: Fascinating, thank you.
It's a complicated world pretty
quickly, isn't it, in photonics.
Poolad: Oh yeah.
So yeah, I
wish I could squiggle a little
bit on the whiteboard behind me.
Dan: yeah, go ahead, but it just wouldn't
translate to the podcast very well.
Poolad: Yeah, I understand.
Dan: Got a couple of more
topics following on from that.
First of all, the rate of omission.
The reason why I'm asking this is I
was reading about other entanglement
sources and the rate of omission that
they're saying that they can provide
first of all, with a single quantum
dot, what kind of rate do you get?
What rates are you targeting in an end
product, or maybe if you can't answer
that, what kind of rate do you think is
necessary for the use cases that your
customers and NIST are looking for?
Poolad: So with, with one quantum
dot even if it's not in a cavity the
rate that you can use it at, it just
very much depends on its lifetime.
And we're fortunate with quantum dots
because this lifetime is not that long.
So once you excite the quantum
dot, it decays really fast.
And by fast, and, but then.
A couple of nanoseconds.
That dictates how fast you can use
this quantum dot, you can re excite it.
And so in our case we can actually
use these at gigahertz levels.
So we can potentially make
billions of photons, generate
billions of photons a second.
With only one quantum dot, and that
is that is really in contrast with any
other types of quantum emitters which
in my knowledge are orders of magnitude
slower and it is the rates game.
When you get, uh, real world applications
of quantum networks, you want to have
reasonable rates for whatever application.
Once you distribute entanglement you
want to have You want to have enough
photons to do whatever protocol that
you want to do, whether it's generating
encryption keys, whether it's connecting
two quantum computers to each other.
So quantum dots are
really good in that way.
And now you can, you can only
enhance this number by putting
an optical cavity around them.
And so what we're targeting is that
not to go higher than 10 gigahertz.
But something in that range is a
very sweet of a spot that we can use
for really high rate entanglement
distribution over quantum networks.
Dan: So I'm an end to end systems guy.
I can't help but thinking about how all
the things are going to link together
when you've got a rate that's that quick.
And it's okay if you, if you haven't
thought about this yet, because I know
you're, you're, you're focusing on a very
nuanced technology at this point in time.
I guess with that rate of generation
across a network, the timing and the
synchronization between end nodes to be
able to, first of all, index, and then
select a particular entanglement entangled
pair of photons to interact with within a
very small window, the determinism levels.
And by that, I mean, the
network determinism levels go
through the roof, don't they?
They're extremely demanding.
Have you thought about that at all?
Poolad: actually coming from an ultra fast
optics group where I did my PhD, there
are the level of precision that you can
track, uh, fluctuations in optical fiber.
Is much, much better than, nanoseconds.
And so that's not a big, at least
right now, it's not the biggest hurdle
that faces us nanosecond stabilization
is, state of the art can do orders
and orders of magnitude better.
And it is like, even right now, the
entanglement sources that are used and,
The quantum network and test beds around
the world, they're based on probabilistic
processes, but they can they're excited
by these very narrow pulses, which are on
the order of Pico second, if not shorter.
And so Pico second is.
already three orders of magnitude
shorter than nanosecond.
And so for quantum network you
probably will need something on the
order of picosecond synchronization.
And when I mentioned that you can
actually get much, much better femtosecond
synchronization is also doable and
is also something that I personally
worked on just looking at generating
quantum states in different time bins.
And yeah, that, that is something
that people are working on
for quantum networks as well.
When the fibers get longer, it just
gets, more and more complicated.
So there's a question of as you get
to, hundreds of kilometers or even
more range of optical fiber, with
what precision you can stabilize them.
But yeah, nanosecond seems to be fine.
Dan: Yeah, that's good, so maybe
just my frame of reference, or my
level of expectation needs to adjust.
And you just reminded me about the,
of course, the Nobel Prize last
year in physics for the attosecond
science, of course, I'm not sure
whether that was relating, that
was relating to optics, wasn't it?
Quantum optics.
Poolad: The, it was related
to, I believe yeah generating
pulses, attosecond pulses light.
Dan: So nanosecond is slow,
basically, that's how I need to think.
Poolad: Now the second can be really
fast and some sense, but it can
be really slow in another sense.
And we're fortunate that it is.
Slow in the sense that we want it to be
slow and fast where we want it to be fast.
Dan: The determinism
then, I guess that's good.
A very useful thing when you're sending
that amount of entangled photons, because
you know, they're going to arrive within
a particular window, basically each
and every one is going to be reliably
produced at a particular time compared to
the one before it and the one after it.
So I guess this is one of
the main benefits, isn't it?
That the other systems in the network
can rely on that repeatability,
that reliability of the determinism.
Poolad: Yes.
The, this having a deterministic
process basically translates into
efficiency, what is, when I push a
button, what's the probability of
me getting an entangled photon pair.
And if that number is low maybe I can
wait, maybe I can use this device a bunch
of times to increase this efficiency.
That's fine.
Cool.
But in quantum networks when you bring in
quantum repeaters and when you interface
multiple of these sources to repeat
the quantum information to go longer
distances, then the sufficiency becomes
a much bigger problem because now all
of these sources have to fire together.
And so you're not just having one source.
And right now, the efficiency with
these, with the probabilistic sources
spontaneous parametric down conversion.
There's a, with them, there's a
trade off between entanglement
fidelity and efficiency.
So to have good entanglement
fidelities, you can't go to
efficiencies higher than 1%.
And so it taps out.
And so imagine if you have 10 of these
sources that you want to interface
with each other and they all have to
fire together what is the probability
of all of them firing together?
It's 1 percent to the power
of 10, so a very small number.
And people have done it.
It's amazing that people have done this,
but it becomes unscalable is the point.
And in our case the whole process
is deterministic fundamentally.
So then the question becomes, how
efficiently can you do all the steps?
How efficiently can you
excite the quantum dot?
Can you get the photons out
into optical fiber and so on?
But there's no fundamental
ceiling to the efficiency, at
least theoretically, anymore.
And that opens the door to
actually scaling quantum networks
to much longer distances.
Now, this is not the only piece
that a quantum repeater needs.
You will need quantum memories, for sure.
To ever increase the rate but
it is a very good first step.
We will be able to interface many of
these sources before needing a quantum
memory, so maybe even going, uh, city
scale to a city scale quantum network.
Dan: Cool, thanks.
So now I'm thinking about the user
stations, the connected devices.
It could be I mean, let's stick
with quantum computers for the time
being because that's, we see this
kind of collective direction towards
distributed quantum computing.
You mentioned transducers
at the beginning.
Is that how you envisage the quantum
dots being implemented as a in
the path between transduction?
I know I've said to others on previous
interviews, it depends, of course, on
the modality of the equipment and how
it carries its quantum information.
But.
I think that's how you described
it at the beginning, but I'm
just checking, is it different?
Have you got different of ways
that the quantum dots implemented
into the interface between
the computer and the network?
Poolad: Yeah.
So there are a couple of different
ways that we can go about this.
One is, okay, what is a quantum network
good for and, the first step of what a
quantum network is good for is that if
you distribute keys through a quantum
network, through an entanglement based
quantum network, then it is not breakable,
but by a large scale quantum computer.
In that sense, and yeah, based on all the
progress that QuEra or IBM have had in the
past few years we know that it's coming.
We know that a quantum, a large
scale quantum computer is coming,
not in the next couple of years,
but probably in the next 10 years.
So we have to move to some type of
quantum resistant encryption which
one way is post quantum cryptography
the, which is much easier to implement
it's however, not provably secure,
however, quantum networks are provably
secure against quantum computers, just
a different type of encryption than
what we're using right now with, two
lines of math, you can prove that.
A quantum computer or
a classical computer.
Doesn't matter how big it is.
It can't break this type of, the type
of encryption that is distributed.
Over quantum networks.
So security is the first application.
And if we're looking at the
security application of quantum
networks, then we use our quantum
dots as just quantum light sources.
So not to connect quantum computers.
This architecture does not need
quantum computers to operate.
It's just to distribute keys that are
unbreakable by a quantum computer.
Dan: Sure.
Poolad: So that would be
the, in my opinion, it's
more of a lower hanging fruit
Dan: Well, that's the, That's
the technology that's nearest
commercialization, right?
Where you're seeing more companies, have a
number of different universities and hubs
around the world that have been focusing
on this for the last five to 10 years.
Okay.
That's why we're at this point.
So yeah definitely.
I would expect that.
Poolad: yeah, absolutely.
So in, in that sense our quantum dots will
sit in nodes of generating entanglement,
and then you distribute them.
Of course, you will need single photon
detectors, you will need quantum
memories for this network to be
operational, and that's pretty much it.
Now, The holy grail of quantum networking
is, however, connecting quantum
computers to each other, to exponentially
enhancing their processing power, to
do distributed quantum computing, to do
modular quantum computing, the same way
that you do processing with GPUs, right?
And in that sense right now, I don't
think our quantum dots are to a level
that they can compete with other
platforms for quantum computing but
they can sit in transducers and they
can convert the quantum information
from light to superconducting
qubits to even the other platforms.
There are companies that are working on
quantum dots for quantum computing, and
there's been a lot of excitement there.
There are some types of quantum dots
that are actually not optically active.
And, but they're really good
for interactions between one
quantum dot to the other.
And in that sense that there's been
a lot of excitement and using those
quantum dots for quantum computing.
And there are a few startups who
are actively working on that.
And yeah, I believe Intel is working
on that as well, but I could be wrong.
Dan: In silicon, right?
Poolad: Yes.
And Silicon, yeah, that is the main one.
In Silicon, there are.
Think different platforms as well.
There were, I think they
started in gallium arsenide.
There are complications with gallium
arsenide with the quantum dots that
are in gallium arsenide for, and
using them for quantum computing.
And, yeah, I think that they've
been addressed by moving to
quantum dots and silicon.
Yeah.
Dan: Yeah, like a lot of these
different modalities, there's
all these pros and cons, right?
And you really need to know your, your
onions to know whether it's the right
thing for you to try and implement or not.
Um, you know, They're fixed.
They may have a good lifetime, but
they may have coherence issues.
There's all kinds of different
things going on, right?
It's such a sensitive system every time.
Yeah, so essentially I, the way I
see it is that Icarus is focusing
on entanglement distribution.
That's really the, or sources I should
say, entanglement sources, which is
deterministic, that's your sweet spot.
So tell me, what about a bit of a roadmap?
How far forward can you look?
Is there anything that you're dreaming
about on the horizon that you could share?
Poolad: Yes.
Right now it's a small startup,
so we're really focusing on
just building this device.
And we will be doing that
in the next couple of years.
The dream is to start building
our own quantum network.
Here in Colorado, there's a lot
of excitement around quantum
technologies in Colorado through
the yeah, we have this new entity
here in the state, elevate quantum
that you mentioned in the beginning.
So we're trying to figure
out how to work with.
Different entities to build a small
quantum network, let's say, and
figure out how to scale that to
different cities around the region.
And, building this quantum network of our
own, I think that would be a milestone.
And then, using our devices to show the
rates to show that It's operable and
you can use this for different use cases.
You can use this to connect data centers.
You can connect banks to each other
and to bring quantum networks from,
the test beds and transforming
them to the commercial side.
That they actually bring value to people.
That is something that we're
hoping to move towards.
Dan: Is the ecosystem in Colorado
complete in such a way that it, you
know, I'm aware that Infleqtion is part
of the group because of the location.
But then what about a QKD vendor that
uses entanglement and what about a quantum
computing company that could potentially
use your entangled sources at some point?
Have you got those types of
partners in Elevate Quantum as well?
Poolad: Yeah.
So the amazing thing in Colorado
and in Elevate is that we have
literally all the pieces needed.
For any type of quantum technology to be
implemented to go through, uh, the pieces
that you would need for quantum network.
Yeah, there are companies that
are working on quantum memories
on single photon detectors.
Our group at NIST makes the most efficient
single photon detectors in the world.
And the, there, there's this amazing
fiber infrastructure And in Boulder
and also in the region as well.
Now, moving a little bit farther beyond
quantum networking, there are these
amazing quantum computing companies which
yeah, Infleqtion on top of all the other
stuff that they're doing, they also, yeah,
very much focused on quantum computing.
There's Infleqtion, there's.
Atom computing that yeah, just
a few months ago announced a
thousand qubit quantum computer.
And there's Quantinuum here.
So it is very seamless for us to,
you know, go partner with these guys.
And yeah, figure out the logistics to
first build the quantum network, build a
very preliminary quantum network, then.
To scale that up.
And at the end of the day, most of these
technologies work at, low temperatures.
In our cases quantum dots should
be cooled down to about 4 Kelvin.
So 4 Kelvin compared to some of
the other platforms is very hot.
Um, like, Yeah, superconducting.
Yeah superconducting qubits have to
operate at a millikelvin temperature.
So they need dilution refrigerators.
But yeah, there are a couple of companies
that are making these fridges here.
Um, Colorado, like Mabel
Quantum and Donaher as well.
I hope I'm pronouncing that right.
But so it has, we have everything
from, the, bits and pieces.
And also the.
Types of technologies that kind
of go cross discipline and like
putting everything together.
So yeah, we're really excited to grow
the portfolio of, equipment and in
Colorado and yeah, bring everything
together in a network, pun intended.
Yeah, exactly.
Dan: It also means you've got no excuses.
You need to get on with it
Poolad: Yeah,
yeah.
Dan: done.
Oh, that's good to have that runway.
It really is.
Can we talk about customers a little bit?
You mentioned public sector bodies.
And.
On your website you're showing
free links, maybe optical and
satellite links, entanglement
sources, going down to the ground.
I know this is really bleeding
edge stuff, but are you working on
anything that kind of brings the
space and quantum domains together?
Poolad: Yeah, so we are working with the
Department of Defense very closely, and
one particular project that we have is to
use our quantum dots for satellite based
quantum networking, and this is the, there
are a lot of complications for the, like
the first step is For free space quantum
networking, you have to deal with all the
noise, all the light that lives around.
So you want to be in a very sweet
spot with the bandwidth of your
photons, and you want to be at a very
specific wavelength, and you want
to be able to travel for thousands
of miles to get to the satellite.
So we're trying to address these and we
think the deterministic nature of our
entanglement sources can play a big role.
And the fact that they magically work
at this gigahertz rate, which is, fast
enough that you can distribute these
high rate high rates of entanglement.
However, it's narrow enough
that you can it in frequency.
And you can filter out all the
noise and still be left with a very
good signal with very low noise.
We think that there's
a good use case there.
And so that is right now we're,
what we're pursuing with the Space
Force um, down in New Mexico.
Um, We are, however, working to
figure out how to use our sources
for fiber networks as well.
And we think that'll
be the, the first step.
We were not going to
go to satellites first.
We're going to use them in fiber networks.
And so for that, we're working
with NASA and Harvard University
to make our sources suitable just,
just for fiber optical networks.
And what I mean by that is that the.
Wavelength of the quantum dots is
uh, at least the ones that we're
using are not suitable or not the.
Telecom wavelength per se, and so
we're working on doing frequency
conversion from what we have.
To the telecom wavelength to do
just low loss to have, the lowest
losses when we go through the.
To optical fiber.
Dan: Yeah, I'm always fascinated
by this, the satellite based
stuff, just because it crosses two
domains that I'm interested in.
And it's a bit more futuristic, but
realistically, we would have come
back to, to the ground, haven't
we?
I mean,
Poolad: So,
So Colorado is a very big hub
for space applications as well.
There is a.
Big yeah, a big Space Force base in
Colorado Springs and a lot of space
companies around Ball Aerospace.
And yeah, once we're definitely not
ready to put our sources on a satellite.
But once we are then we have
a bunch of amazing partners
to reach out to here as well.
Dan: Ball Aerospace.
They've just been acquired by
British Aerospace, I believe BAE
Poolad: Yes,
Dan: Yeah.
So that's going to be interesting for BAE.
But yeah, no, I heard anecdotally about a.
A test of entanglement
distribution from a satellite.
I don't know if it was in China,
but they, I had the anecdote was
that they managed to achieve one
coincidence count per second.
, the success rate that they achieved was
one coincidence, if that's the right term.
One, one per
Poolad: One per second.
Dan: second.
Yeah.
Poolad: That is heroic.
Dan: you're sending, if
you're sending, what was it?
The billions a second or something.
Yeah.
Then.
You've only got to get more than one
through and receive it at both ends
and you've already beaten a record.
Poolad: Yeah, that is the hope.
Yeah, there was a, an experiment a
couple of months ago between that
with a satellite, they, I think,
distributed entanglement between China
and Russia was, yeah, very interesting.
I don't know the details there.
I don't think they,
Dan: I don't think much was
Poolad: Actually, yeah, I
don't think much was shared.
But
Dan: Yeah, I'm not sure if that
was the one or not, but but hey,
I'm looking forward to hearing
the news about the one in the U.
S.
that's going to come at some point.
Poolad: Yeah, there are, the
government is has noticed that,
we should move in that direction.
And both NASA and the Air Force
or Space Force, they're all
working towards it very actively.
Dan: it's really the only way that
long distance quantum communication
is going to work at the moment, right?
Long distance.
But there's so many things we should need.
The scheduling the swapping of states
the classical part of it, the timing let
alone the stuff that's on the payload.
Yeah, anyway, thanks for that
little detour to the, my kind of
fantasy world of satellite quantum.
Um, um, yeah,, okay.
Well, Thank, thanks very much, Poolad.
I'm going to wrap it up just
by saying thank you very much.
And thanks for.
Answering my barrage of
questions that I chucked at you.
Thanks very much.
And you, know, the
responses are brilliant.
So appreciate it.
Poolad: of course, this
is a pleasure, Dan.
Yeah, had a lot of fun.
Dan: Okay.
Bye for now.
Poolad: Bye.
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.