July 28, 2025 • 23 mins
Researchers and industry are coming together to develop computer systems that can take advantage of quantum mechanics. Christopher Monroe, a professor at Duke University and co-founder of IonQ, discusses quantum computing, advances in the field and IonQ's journey from startup to being the first publicly traded quantum computing company.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:03):
This is the Discovery Files podcast
from the U..S National Science Foundation.
At the quantum scales of molecules, atomsand electrons.
The laws of physics become peculiar.
NSF supported researchers are exploringhow to control the behavior
of these quantum systems to translatequantum knowledge into next generation
(00:23):
technologies, such as quantum computers,that can tackle complex problems
that are beyond the capabilities of eventhe most powerful supercomputers of today.
We're joined today by Christopher Monroe,Gilhuly Family Presidential Distinguished
Professor of Electricaland Computer engineering and physics
at DukeUniversity, and co-founder of IonQ,
the first publicly traded quantumcomputing company.
(00:46):
Professor Monroe,thank you so much for joining me today.
It's a pleasure to be here.
I want to start with a general question.
People kind of get lost a little bit,maybe glaze over when you mention the term
quantum computing.
What is quantum computing?
Yeah. The question.
Well, let me start withmaybe the bad news.
Quantum computing.
We don't yet know exactly howthese devices will be used in the future.
(01:07):
The very good news.
And to me it's not hype, It'sa fact, quantum
computers are able to process and storedata in ways
that are impossible using standardcomputing machinery, and that's why
they hold great promise for mostly unknownapplications in the future.
The reason I can say all this isthat they're based on totally new physics.
(01:32):
Really physics that is not partof our everyday existence.
There's no explaining certain thingsusing everyday intuition.
This is interesting.
And the downside tothat is that people aren't used to them.
There's some very weird ideas behind itand some people don't even believe in it.
I think we're maybe, fortunately, beyondthat second statement that, you know,
(01:53):
people are starting to understandthat the industrial play in quantum
these days, it's largelydriving the field now.
It's a very interesting time.
But if I had to go to 101quantum computing, I would say
the bits that we're used toin classical computing,
zeros or ones,they are called quantum bits.
(02:14):
That's one way to processquantum information.
Quantum bitscan be in a superposition of zero and one.
Zero and one at the same time.
And when you add more and more bitstogether,
the possibilities grow exponentially.
And that's really the bonus.
We can store exponential amountsof information way more,
even with hundreds or thousands of qubits.
(02:35):
That's way more informationthan we could store in classical machines.
So it's very sci fi like in a way.
There's a lot of hype behind it.
We know they're going to be usefulfor something in the future.
The question is when.
It's a wonderful confluenceof fundamental physics
and very applied engineeringand real devices of the future.
In your devices,you're using a trapped ion approach.
(02:57):
Can you talk a little bit about whatthat difference is?
Right.
One of the chief and most difficultrequirements in a quantum computer
is that you can maintaina superposition of states.
You can have a zeroand one at the same time.
And the trick is the wayI think about it is,
the superpositions only existas long as you're not looking,
(03:17):
as long as it's not observed.
So you demand an extreme levelof isolation of these systems
to do the computation.
The computation happens without lookingand that's very tricky to do.
It's hard to control somethingwhen you can't even look at it.
So we indeed use individual atomsas the bits.
There are many flavors.
They don't have to be atomic ions.
(03:38):
They don't have to be chargedatoms. As we use.
They can be neutral atoms.
They can be single electrons,single photons.
But they should be something quantum,something that obeys
these strange lawsof existing in superposition.
The trapped ion approach, these areindividual atoms that are charged.
And because they’re charged,we can control them so well.
(03:59):
We can apply electric fields and actually
levitate them with electromagnetic fields.
And they're in a vacuum chamber.
Because not only cannobody look at the system,
the system can't be disturbedby the environment, for instance, by air.
So we get rid of all the air, they’rein a vacuum chamber.
And they're floating.
They're not part of the surface.They're not part of a solid.
(04:22):
They're just sitting thereall by themselves.
They see each other.
Trapped ions are acknowledgedto be sort of the purest
form of isolation of qubits.
And there are lots of engineeringchallenges to scale this platform up.
But yes, individual atoms,these are sort of like the transistors
of our quantum computers.
(04:42):
I want to build
towards where you are today a little bit,and ask you about your earlier work.
How did your early work with future Nobel
laureates, you were working with lasercooled atoms and entangled atoms,
how did this set you on the pathfor working with quantum computing?
Yeah,that's a very easy thing for me to answer.
I worked for many years in Boulder,Colorado with David
(05:03):
Wineland at the National Instituteof Standards and Technology.
We were basically the atomic clockdivision of the U.S.
government, but more than that,we weren't really making primary
frequency standards we were conductingresearch on to make them better
and what the future clock pieceswould look like.
And we, in fact, figured out thatby entangling, this is a buzzword.
(05:26):
I'll explain in a minute.
By entangling multiple atoms,we could make the clock run
with better signal to noise.
So this is largely academic at the timebecause the improvement was marginal.
If you entangle two atoms,
you had a factor of roughlytwo better in signal to noise, which is,
you know, factor two is nice,but it's a lot of work to entangle them.
(05:47):
So we were playing around with the physicsof entanglement, being able
to do all kinds of foundational aspectsof studying the entangled superposition.
So an entangled superposition is onethat has more than one qubit.
And the reason it's weird, this is firstpointed out by Einstein in the 1930s.
He thought he had a way to provethat quantum mechanics was wrong,
(06:10):
and he came up with an example,and it's called entanglement.
Now, if I give you two qubits,I think of them as coins.
Sometimes heads and tails, zeros and ones,whatever you want.
If I produce two qubitsand they're in a superposition
of both zero and both one,that's a very interesting superposition,
because when you measurejust one of those two qubits,
(06:31):
what happens is theand I didn't say this earlier.
It's importantyou have to look at qubits in the end.
I mean, what good is a computerthat you never look at?
When you measure a superposition, itrandomly pops into one of its many states.
So there's sort of noisecoming out of the ether.
In a way.
This superposition that's entangled,I mentioned zero zero and one one.
(06:53):
When you look atjust one of the two qubits,
you know it's state,you measure it's state 0 or 1,
but then the other one gets measuredalso implicitly at the same time
because they're correlated.
You prepared themin this entangled superposition.
And what Einstein realized is thatthat's very strange
because it appears that there was actionat a distance by
(07:16):
these qubitscould have been really far apart.
And by measuring one of them,
somehow we learn thatthe other one is in the same state.
And that's very strange.
It doesn't violate any laws of physicsof course, it doesn't allow you
to communicate informationfaster than light,
because even though you knowthat state of the other qubit, faster
than light could have made the distance,if you do it over and over again,
(07:39):
it's totally random, and there'sno information content in random bits.
That's sort of the close of that paradox.
But Einstein didn't know about informationtheory at the time in the 30s, I guess.
So we were making entangled statesfor these clocks for academic reasons,
indeed to do foundational quantum studies.
And then this is 1993, 1994,
(08:01):
and then Peter Shorpublished his algorithm that showed that
if we had a quantum computer,we could factor numbers into their primes.
We could therefore break codes,break public key encryption systems.
And this was a big deal
because it seemed to have changedthe complexity class of a certain problem.
In other words,it put quantum computing on the map.
(08:22):
It showed that yes,there is an application.
It's not. As I said earlier,we don't know what's going to be used for.
We do know there's one exception.
We know it will be used to factor numbers,
which sounds kind of boring in a way,but it's very important for cryptography.
So we sort of backed into the fieldof quantum computing by making clocks.
And I guess I lived through that.
(08:42):
And I'd been told that you dobasic research, fundamental research,
you never know where it's going to go.
And most of the time it fails.
It doesn't go into a cool direction.
But this is one example,is kind of a spectacular one where
studying one thing pivots into somethingelse.
And, you know, that was 30 years ago andit's been a remarkable growth since then.
So early on playing aroundwith these atoms to make better clocks.
(09:04):
And this followed up on my workgoing toward
Bose condensation with cold atomsin Carl Wieman's group, also in Boulder.
That gave me the backgroundin atomic physics
to start this path of quantum computing.
As you said, it's been a number of yearssince then, and things have progressed.
At what point did you knowit was time to start a company?
(09:25):
Well, literally,I suppose it was when a particular venture
capitalist visited my officeand said, it's time to start a company.
I certainly wasn't going to go out.
I didn't know how to start a company.
I wasn't going to go out and try unless,you know, unless the bar was low
and there was a good business planand it wasn't that simple.
But there was a particular venturecapitalist, really nice guy
(09:47):
name of Harry Weller.
He visited my office in College Park,Maryland, 2014.
So a little more than ten years ago.
He had a physics undergrad degreeso he could read some of this stuff.
And my colleagueand friend junks saying, come.
So I was at Maryland at the timeand Jungsang Kim.
At Duke.
We started collaborating about ten yearsbefore that, building bigger iron traps,
(10:09):
thinking about how to scale up, applyinglots of engineering approaches,
making it cheaper, using lasersthat were much easier to use.
And so it wasn't a business exactly,but we were sort of plotting the path
so we didn't have to worry about all theatomic physics and the control of atoms.
With optics,
we could start thinking about programingsmall algorithms and our machines.
(10:30):
We were actually doing that,and we wrote a paper,
it was an architecture paper on howto scale a large scale quantum computer
based on photonic interconnects
and so forth with these trapped ions,although it's more general than that.
And this venture capitalistapparently read Physical Review a
he read that paperand he waved it in my face in my options.
You know, this reads like a business plan.
(10:53):
And so I sort of laughed it off.
But over the next few years, he convincedour universities about IP agreements.
Just the other day, I was telling somebodythat IP is a nuisance,
but you have to get intellectual property,
wrap it up,get it licensed to have a startup.
And so it was easy to do that because wehad a lot of ideas that could be patented.
And the universities at Marylandand Duke really bought into the idea.
(11:14):
And there we were.
We started off and again,even to this day, that's ten years ago,
industry and quantum computing,they're still doing a lot of research.
And so it wasn't likewe were building cars all of a sudden,
and we had to roll off10,000 cars a week off the assembly line.
We still had to execute a plan that waswe didn't know exactly how we'd play out.
(11:36):
We would need to invent new technologies.
It would be very expensive.
And that's why
the business model made sense,because we could have budgets that were,
you know, we didn't have to worryso much about duct tape and gluing
things together with epoxy,like we do in a physics lab.
So how has NSF supportimpacted your work at IonQ?
Very much so.
(11:57):
I think the formative days uptill we formed IonQ.
I can give you one example.
I was at the University of Michiganfor many years.
They are and they were well knownfor their work in pulsed
lasers, high energyultrafast lasers to study
fast dynamics and nonlinear optics,and so forth.
I had never used a pulse laser in my life.
When I went to Michigan in 2000
(12:18):
and I got to workingwith some of my colleagues there
Phil Bucksbaum, Gerard Mourou, and otherswho use these lasers.
And I got less afraid of these lasers.
Now, the reason I bring that up
is that when the NSF PhysicsFrontier Center started.
That's right.
When I arrived at Michiganand we went after one of the physics
frontier centers, and we were awardedthe focus center at Michigan
(12:40):
that would use ultra, study ultrafastscience in many different facets.
And I had not done anything everin ultra fast.
But I started toand the reason this is important
is that the laser that we usedto control these atoms without looking,
it turned outthe best lasers were pulsed lasers.
The reason they were the best lasersis that they were made for
(13:01):
lithography of silicon chips,for a totally different reason.
They were very reliable.
I used to say this laser has one knoband it's the on key.
You turn it on and it works.
It works for years.
And that allowed us
to not have to pay attentionto tuning up a laser for four hours
every single day, which I'd been used to,and we could move on to other things.
(13:21):
And so in Michigan,
I learned how to use lasersthat were more mature, better engineered.
And, you know, that was a direct resultof a big NSF center there.
And when I went to Maryland,we also had a physics frontier center
on controlling quantum systemsand so forth.
And throughout the years, ever sinceI left NIST in 2000, I like to say
that the NSF grants were the bestcolor of money because I had more freedom.
(13:45):
Of course, there are milestones.
You have to write a proposaland it's very competitive.
But once the NSF awards you the grant,they're not going to check on you
every six months. You say, well,how did you do?
You promised you do that?You promised to do that?
And there was in a sense, trust.
And that's what you have to do when, when
you're on a path of fundamental physicsor fundamental science
that you're not surewhere it's going to go.
(14:06):
And of course, if you want to get fundedagain, you better show something.
Even if it pivots, even if it's somethingtotally different, it's valuable.
And those grants throughout the years,there were many programs in AMO and
physics at the informationfrontier, and NSF, see really
wanted to capture quantum in waysthat the other agencies really couldn't.
And they, you know,they largely succeeded in that.
(14:26):
I will say, in the last several years,one of my heroes at NSF, Bogdan Mihaila,
he realized that we were not justdoing atomic physics,
but we were studying algorithms.
We were designingalgorithms to match with our atoms.
And and that we call ita vertical research institute.
And that that spawned what we call STAQ.
S-T-A-Q.
(14:46):
Basically, applying quantum
software to a very mature quantum systemthat we can engineer.
We it's almost like a user facility.
The way we run this place hereat Duke, at the Duke Quantum Center,
we have so many collaborationswhere people run their ideas
through our machines.
It's almost like a company could do that,but we do it for science.
We're not interested in optimizinga traveling salesman problem for Amazon.
(15:09):
I mean, we love to do that,but we're also interested
in studying phase transitions,simulating correlated matter.
And again, I think NSF reallybecause of their science basis,
it was a super important part of the USposture in quantum.
What has been the biggest challengetranslating
researchinto more industrial applications.
(15:30):
IonQ by now is a very big company.
And they're a public company.
And with that comes severe challengesof talking to the street.
I've never thought that job number oneis to
make sure the stock is buoyant,but I understand that it needs to be.
Otherwise the company doesn't exist.
And so to me, the central challenge ishow do you balance?
(15:51):
I'll sound a cliche, you know, how do youbalance future research and development
with current revenueand sales and interest in the company?
I’ll never claim to be a businessmanor a,
you know, even though I'm involvedwith a company, but you have to have
a little bit of both, and it'syou really shouldn't be all of one
(16:11):
and none of the other.
You can't do 100% research at a company.
People tend not to want to invest in that.
If you have a long term investor, it'sprobably not the public that's investing.
If you're only doing, you know,sales and revenue and you're not looking
in the out years, that's an equal recipefor disaster in this field.
So to me,that was the most difficult calculation.
(16:32):
How much to sell and how much to do R&D.
I will add a little short story on that.
I was CEO of the company for about a year,when we were just starting to produce
our first few systems,and we started working with AWS,
Amazon Web Servicesto put one of our machines on the cloud.
I mean, the great thing about any computeris that especially a quantum computer,
(16:54):
it doesn't need to be in front of you.
You can use it on the cloud.
And that exercise was tricky becauseit meant that we had to run a machine
that would be up 24-7,and therefore it's not getting any better.
It's just running.
We're not improving things,and that's not what I'd been used
to, you know, always improving things,looking at the next thing.
And I was actually sort of againstdoing it, putting it in the cloud.
(17:17):
It would also get revenue and that's fine.
The revenue was good.
But I learned later on that it actuallywas really important
to build a system that would work 24-7the engineering behind that.
And, you know, engineeringis something that I don't come natural to.
It's very hard to teach engineering,systems engineering, but making something
work 24-7 is an exercisethat you will learn the next.
(17:40):
The research systems will benefitfrom that.
That was, an interesting lessonI learned.
So maybe I erred too much on the researchside, I'm not sure, but we ended up
putting them on the cloud,all of the clouds.
And I think, IonQ and other companiesthat do that are much better off
for doing that.
The other kind of business questionI want to ask you
was, can you tell us a little bitabout your experience
(18:00):
making that jump from being a startupcompany into a public traded company?
We were sort of lucky in
the timing of this, andit actually happened right out of Covid
when as a private company,we'd just raised a bunch of funding
that would give us,I don't know, maybe a 3 or 4 year
runway, as they say, because we weren'treally bringing in much money.
(18:23):
We were spending it.
And the investors understood this.
This is a long term play.
And when Covid hit our board,which was largely
our investors, they said, we don't knowhow bad this Covid thing is going to be.
You need to really think about that.
3 or 4 years to be 7 or 8 years, is itmaybe there's a recession and so forth.
And the one risk for the company backthen was raising capital.
(18:45):
Really, we weren't too worriedabout our technical plan.
It was very solid, not easy, butwe weren't fundamentally worried about it.
We were worriedabout having to raise money again.
And that'sthat's the bane of startup companies.
With venture capital, you have tokeep raising and you're always out there.
And so when the opportunity came, a so-called blank check company
also called SPAC,Special Purpose Acquisition Company,
(19:08):
they have a spot on the stock exchange,in this case, the New York Stock Exchange.
They have investors with lots of cash,and they want to plow it
into a particular companythat will become public.
So it's a shortcut to an IPO.
We were sort of lucky that these folkswere very interested in the quantum field,
in our company in particular.
And I was told by the folksthat ran the SPAC
(19:30):
that they wanted a companythat had excellent foundational people
fundamentals, and that was Junsang Kimand myself in the in the ion trap basics,
you know, largelyriding on decades of mature research
that was ready to be engineered,
but then also the CEO we hadand that was Peter Chapman.
And he he is an excellent salesman.
(19:51):
So we had sort of both partswhat you need to have a company
and our SPACs kind ofI call them benefactors.
It's not the right word, but our sponsors,they wanted to make a play in quantum,
knowing how risky it was.
But they had a good foundation,an eye toward revenue and sales as well.
You know, how do you balance that?That's the tricky thing.
So that was the transition.
And, you know, we started getting moreprofessional, getting a professional
(20:14):
finance team, legal team, all that stuffthat comes with the public company,
the company was differentafter going public for sure.
And I think the pressure you know,their quarterly stock report earnings
where we talkdirectly to investors in the street.
It's still strange those interactions.
But I give the analysts and the investors,the people on Wall Street
(20:35):
a lot of credit.
They seem to read about this field.
They think quantum will will enhanceAI in the future.
It's a question of time,I guess, in terms of prosecution
of our technical planthat just went into hyperdrive
with the public funding,we were really able to just go crazy
and go much faster,make more systems, hire a lot of people.
(20:56):
So, you know,I have to say, you know, it's good timing.
I think it's a little harder to do thatnow, but, yeah, it worked out well.
Very cool.
So for my last question,you said there's a lot of question
for what quantum computingwill ultimately be used for,
but what impacts do you think it willhave on science and society?
(21:17):
Again, taking the long view,
I'm quite sure that we will have quantumcomputers decades from now.
For sure.
That will be deployed for certain problemsand the tricky thing there is that the
public largely doesn't know or understandor care about these new laws of physics.
They really are fundamentallydifferent laws of physics.
(21:39):
And I think for people to fully appreciatethe power of quantum computers
and to accelerate their usein different fields, people should be
learning the fundamentals of quantumeven at a young age.
And I do believemaybe it's more of a hope, but in 50 years
people will have an intuitionfor some of these weird things,
like “spooky action at a distance”,the Einstein quote about entanglement.
(22:02):
They'll have a feel for it.
They'll they'll know that small,very simple systems, when they're cold
and they're isolated, they,they behave according to these laws.
So I think that's a total shift in the waywe think about fundamental physics,
for sure.
And maybe it'll be infectiousand hit other fields
from chemistry to more complexthings like in biology.
(22:24):
So I think quantum is not readyto tackle biological problems.
They’re still way too complicated,
but I am quite sure it will eventually,and I can say that safely because.
I probably won't be alivewhen it hits that level, but
quantum is, in termsof the entire community of physics.
We all eat and breathe quantum physics,but it's not so useful
(22:46):
out in the real world.
This is a way to change that, and I thinkit also will propel physics overall
as something that is absolutely requiredif you want to understand how to compute.
I mean, I guess physics is behindthe transistor, we all know that.
But the model of a transistoris very easy.
It's like turning on a valve,
a water hose with a valve or somethingelectronically controlled valve.
(23:07):
You can make models that go far.
You don't have to learn all the physics.
But quantum,unfortunately, is not like that.
We have nothing. We can turn to.
So I think quantum computingcan be much bigger than just devices.
It can really bring physics to the masses
and get many more people studying physicsand then going on to other things.
Special thanks to ChristopherMonroe and Alexander Cronin.
(23:28):
For the Discovery Files, I'm Nate Pottker.Watch video versions
of these conversations on our @NSFscienceYouTube channel.
Please subscribe wherever you get podcastsand if you like our program, share
with a friendand consider leaving a review.
Discover how the U.S.
National Science Foundationis advancing research at NSF.gov.
This is the Discovery Files podcast
from the U..S National Science Foundation.
At the quantum scales of molecules, atomsand electrons.
The laws of physics become peculiar.
NSF supported researchers are exploringhow to control the behavior
of these quantum systems to translatequantum knowledge into next generation
(00:23):
technologies, such as quantum computers,that can tackle complex problems
that are beyond the capabilities of eventhe most powerful supercomputers of today.
We're joined today by Christopher Monroe,Gilhuly Family Presidential Distinguished
Professor of Electricaland Computer engineering and physics
at DukeUniversity, and co-founder of IonQ,
the first publicly traded quantumcomputing company.
(00:46):
Professor Monroe,thank you so much for joining me today.
It's a pleasure to be here.
I want to start with a general question.
People kind of get lost a little bit,maybe glaze over when you mention the term
quantum computing.
What is quantum computing?
Yeah. The question.
Well, let me start withmaybe the bad news.
Quantum computing.
We don't yet know exactly howthese devices will be used in the future.
(01:07):
The very good news.
And to me it's not hype, It'sa fact, quantum
computers are able to process and storedata in ways
that are impossible using standardcomputing machinery, and that's why
they hold great promise for mostly unknownapplications in the future.
The reason I can say all this isthat they're based on totally new physics.
(01:32):
Really physics that is not partof our everyday existence.
There's no explaining certain thingsusing everyday intuition.
This is interesting.
And the downside tothat is that people aren't used to them.
There's some very weird ideas behind itand some people don't even believe in it.
I think we're maybe, fortunately, beyondthat second statement that, you know,
(01:53):
people are starting to understandthat the industrial play in quantum
these days, it's largelydriving the field now.
It's a very interesting time.
But if I had to go to 101quantum computing, I would say
the bits that we're used toin classical computing,
zeros or ones,they are called quantum bits.
(02:14):
That's one way to processquantum information.
Quantum bitscan be in a superposition of zero and one.
Zero and one at the same time.
And when you add more and more bitstogether,
the possibilities grow exponentially.
And that's really the bonus.
We can store exponential amountsof information way more,
even with hundreds or thousands of qubits.
(02:35):
That's way more informationthan we could store in classical machines.
So it's very sci fi like in a way.
There's a lot of hype behind it.
We know they're going to be usefulfor something in the future.
The question is when.
It's a wonderful confluenceof fundamental physics
and very applied engineeringand real devices of the future.
In your devices,you're using a trapped ion approach.
(02:57):
Can you talk a little bit about whatthat difference is?
Right.
One of the chief and most difficultrequirements in a quantum computer
is that you can maintaina superposition of states.
You can have a zeroand one at the same time.
And the trick is the wayI think about it is,
the superpositions only existas long as you're not looking,
(03:17):
as long as it's not observed.
So you demand an extreme levelof isolation of these systems
to do the computation.
The computation happens without lookingand that's very tricky to do.
It's hard to control somethingwhen you can't even look at it.
So we indeed use individual atomsas the bits.
There are many flavors.
They don't have to be atomic ions.
(03:38):
They don't have to be chargedatoms. As we use.
They can be neutral atoms.
They can be single electrons,single photons.
But they should be something quantum,something that obeys
these strange lawsof existing in superposition.
The trapped ion approach, these areindividual atoms that are charged.
And because they’re charged,we can control them so well.
(03:59):
We can apply electric fields and actually
levitate them with electromagnetic fields.
And they're in a vacuum chamber.
Because not only cannobody look at the system,
the system can't be disturbedby the environment, for instance, by air.
So we get rid of all the air, they’rein a vacuum chamber.
And they're floating.
They're not part of the surface.They're not part of a solid.
(04:22):
They're just sitting thereall by themselves.
They see each other.
Trapped ions are acknowledgedto be sort of the purest
form of isolation of qubits.
And there are lots of engineeringchallenges to scale this platform up.
But yes, individual atoms,these are sort of like the transistors
of our quantum computers.
(04:42):
I want to build
towards where you are today a little bit,and ask you about your earlier work.
How did your early work with future Nobel
laureates, you were working with lasercooled atoms and entangled atoms,
how did this set you on the pathfor working with quantum computing?
Yeah,that's a very easy thing for me to answer.
I worked for many years in Boulder,Colorado with David
(05:03):
Wineland at the National Instituteof Standards and Technology.
We were basically the atomic clockdivision of the U.S.
government, but more than that,we weren't really making primary
frequency standards we were conductingresearch on to make them better
and what the future clock pieceswould look like.
And we, in fact, figured out thatby entangling, this is a buzzword.
(05:26):
I'll explain in a minute.
By entangling multiple atoms,we could make the clock run
with better signal to noise.
So this is largely academic at the timebecause the improvement was marginal.
If you entangle two atoms,
you had a factor of roughlytwo better in signal to noise, which is,
you know, factor two is nice,but it's a lot of work to entangle them.
(05:47):
So we were playing around with the physicsof entanglement, being able
to do all kinds of foundational aspectsof studying the entangled superposition.
So an entangled superposition is onethat has more than one qubit.
And the reason it's weird, this is firstpointed out by Einstein in the 1930s.
He thought he had a way to provethat quantum mechanics was wrong,
(06:10):
and he came up with an example,and it's called entanglement.
Now, if I give you two qubits,I think of them as coins.
Sometimes heads and tails, zeros and ones,whatever you want.
If I produce two qubitsand they're in a superposition
of both zero and both one,that's a very interesting superposition,
because when you measurejust one of those two qubits,
(06:31):
what happens is theand I didn't say this earlier.
It's importantyou have to look at qubits in the end.
I mean, what good is a computerthat you never look at?
When you measure a superposition, itrandomly pops into one of its many states.
So there's sort of noisecoming out of the ether.
In a way.
This superposition that's entangled,I mentioned zero zero and one one.
(06:53):
When you look atjust one of the two qubits,
you know it's state,you measure it's state 0 or 1,
but then the other one gets measuredalso implicitly at the same time
because they're correlated.
You prepared themin this entangled superposition.
And what Einstein realized is thatthat's very strange
because it appears that there was actionat a distance by
(07:16):
these qubitscould have been really far apart.
And by measuring one of them,
somehow we learn thatthe other one is in the same state.
And that's very strange.
It doesn't violate any laws of physicsof course, it doesn't allow you
to communicate informationfaster than light,
because even though you knowthat state of the other qubit, faster
than light could have made the distance,if you do it over and over again,
(07:39):
it's totally random, and there'sno information content in random bits.
That's sort of the close of that paradox.
But Einstein didn't know about informationtheory at the time in the 30s, I guess.
So we were making entangled statesfor these clocks for academic reasons,
indeed to do foundational quantum studies.
And then this is 1993, 1994,
(08:01):
and then Peter Shorpublished his algorithm that showed that
if we had a quantum computer,we could factor numbers into their primes.
We could therefore break codes,break public key encryption systems.
And this was a big deal
because it seemed to have changedthe complexity class of a certain problem.
In other words,it put quantum computing on the map.
(08:22):
It showed that yes,there is an application.
It's not. As I said earlier,we don't know what's going to be used for.
We do know there's one exception.
We know it will be used to factor numbers,
which sounds kind of boring in a way,but it's very important for cryptography.
So we sort of backed into the fieldof quantum computing by making clocks.
And I guess I lived through that.
(08:42):
And I'd been told that you dobasic research, fundamental research,
you never know where it's going to go.
And most of the time it fails.
It doesn't go into a cool direction.
But this is one example,is kind of a spectacular one where
studying one thing pivots into somethingelse.
And, you know, that was 30 years ago andit's been a remarkable growth since then.
So early on playing aroundwith these atoms to make better clocks.
(09:04):
And this followed up on my workgoing toward
Bose condensation with cold atomsin Carl Wieman's group, also in Boulder.
That gave me the backgroundin atomic physics
to start this path of quantum computing.
As you said, it's been a number of yearssince then, and things have progressed.
At what point did you knowit was time to start a company?
(09:25):
Well, literally,I suppose it was when a particular venture
capitalist visited my officeand said, it's time to start a company.
I certainly wasn't going to go out.
I didn't know how to start a company.
I wasn't going to go out and try unless,you know, unless the bar was low
and there was a good business planand it wasn't that simple.
But there was a particular venturecapitalist, really nice guy
(09:47):
name of Harry Weller.
He visited my office in College Park,Maryland, 2014.
So a little more than ten years ago.
He had a physics undergrad degreeso he could read some of this stuff.
And my colleagueand friend junks saying, come.
So I was at Maryland at the timeand Jungsang Kim.
At Duke.
We started collaborating about ten yearsbefore that, building bigger iron traps,
(10:09):
thinking about how to scale up, applyinglots of engineering approaches,
making it cheaper, using lasersthat were much easier to use.
And so it wasn't a business exactly,but we were sort of plotting the path
so we didn't have to worry about all theatomic physics and the control of atoms.
With optics,
we could start thinking about programingsmall algorithms and our machines.
(10:30):
We were actually doing that,and we wrote a paper,
it was an architecture paper on howto scale a large scale quantum computer
based on photonic interconnects
and so forth with these trapped ions,although it's more general than that.
And this venture capitalistapparently read Physical Review a
he read that paperand he waved it in my face in my options.
You know, this reads like a business plan.
(10:53):
And so I sort of laughed it off.
But over the next few years, he convincedour universities about IP agreements.
Just the other day, I was telling somebodythat IP is a nuisance,
but you have to get intellectual property,
wrap it up,get it licensed to have a startup.
And so it was easy to do that because wehad a lot of ideas that could be patented.
And the universities at Marylandand Duke really bought into the idea.
(11:14):
And there we were.
We started off and again,even to this day, that's ten years ago,
industry and quantum computing,they're still doing a lot of research.
And so it wasn't likewe were building cars all of a sudden,
and we had to roll off10,000 cars a week off the assembly line.
We still had to execute a plan that waswe didn't know exactly how we'd play out.
(11:36):
We would need to invent new technologies.
It would be very expensive.
And that's why
the business model made sense,because we could have budgets that were,
you know, we didn't have to worryso much about duct tape and gluing
things together with epoxy,like we do in a physics lab.
So how has NSF supportimpacted your work at IonQ?
Very much so.
(11:57):
I think the formative days uptill we formed IonQ.
I can give you one example.
I was at the University of Michiganfor many years.
They are and they were well knownfor their work in pulsed
lasers, high energyultrafast lasers to study
fast dynamics and nonlinear optics,and so forth.
I had never used a pulse laser in my life.
When I went to Michigan in 2000
(12:18):
and I got to workingwith some of my colleagues there
Phil Bucksbaum, Gerard Mourou, and otherswho use these lasers.
And I got less afraid of these lasers.
Now, the reason I bring that up
is that when the NSF PhysicsFrontier Center started.
That's right.
When I arrived at Michiganand we went after one of the physics
frontier centers, and we were awardedthe focus center at Michigan
(12:40):
that would use ultra, study ultrafastscience in many different facets.
And I had not done anything everin ultra fast.
But I started toand the reason this is important
is that the laser that we usedto control these atoms without looking,
it turned outthe best lasers were pulsed lasers.
The reason they were the best lasersis that they were made for
(13:01):
lithography of silicon chips,for a totally different reason.
They were very reliable.
I used to say this laser has one knoband it's the on key.
You turn it on and it works.
It works for years.
And that allowed us
to not have to pay attentionto tuning up a laser for four hours
every single day, which I'd been used to,and we could move on to other things.
(13:21):
And so in Michigan,
I learned how to use lasersthat were more mature, better engineered.
And, you know, that was a direct resultof a big NSF center there.
And when I went to Maryland,we also had a physics frontier center
on controlling quantum systemsand so forth.
And throughout the years, ever sinceI left NIST in 2000, I like to say
that the NSF grants were the bestcolor of money because I had more freedom.
(13:45):
Of course, there are milestones.
You have to write a proposaland it's very competitive.
But once the NSF awards you the grant,they're not going to check on you
every six months. You say, well,how did you do?
You promised you do that?You promised to do that?
And there was in a sense, trust.
And that's what you have to do when, when
you're on a path of fundamental physicsor fundamental science
that you're not surewhere it's going to go.
(14:06):
And of course, if you want to get fundedagain, you better show something.
Even if it pivots, even if it's somethingtotally different, it's valuable.
And those grants throughout the years,there were many programs in AMO and
physics at the informationfrontier, and NSF, see really
wanted to capture quantum in waysthat the other agencies really couldn't.
And they, you know,they largely succeeded in that.
(14:26):
I will say, in the last several years,one of my heroes at NSF, Bogdan Mihaila,
he realized that we were not justdoing atomic physics,
but we were studying algorithms.
We were designingalgorithms to match with our atoms.
And and that we call ita vertical research institute.
And that that spawned what we call STAQ.
S-T-A-Q.
(14:46):
Basically, applying quantum
software to a very mature quantum systemthat we can engineer.
We it's almost like a user facility.
The way we run this place hereat Duke, at the Duke Quantum Center,
we have so many collaborationswhere people run their ideas
through our machines.
It's almost like a company could do that,but we do it for science.
We're not interested in optimizinga traveling salesman problem for Amazon.
(15:09):
I mean, we love to do that,but we're also interested
in studying phase transitions,simulating correlated matter.
And again, I think NSF reallybecause of their science basis,
it was a super important part of the USposture in quantum.
What has been the biggest challengetranslating
researchinto more industrial applications.
(15:30):
IonQ by now is a very big company.
And they're a public company.
And with that comes severe challengesof talking to the street.
I've never thought that job number oneis to
make sure the stock is buoyant,but I understand that it needs to be.
Otherwise the company doesn't exist.
And so to me, the central challenge ishow do you balance?
(15:51):
I'll sound a cliche, you know, how do youbalance future research and development
with current revenueand sales and interest in the company?
I’ll never claim to be a businessmanor a,
you know, even though I'm involvedwith a company, but you have to have
a little bit of both, and it'syou really shouldn't be all of one
(16:11):
and none of the other.
You can't do 100% research at a company.
People tend not to want to invest in that.
If you have a long term investor, it'sprobably not the public that's investing.
If you're only doing, you know,sales and revenue and you're not looking
in the out years, that's an equal recipefor disaster in this field.
So to me,that was the most difficult calculation.
(16:32):
How much to sell and how much to do R&D.
I will add a little short story on that.
I was CEO of the company for about a year,when we were just starting to produce
our first few systems,and we started working with AWS,
Amazon Web Servicesto put one of our machines on the cloud.
I mean, the great thing about any computeris that especially a quantum computer,
(16:54):
it doesn't need to be in front of you.
You can use it on the cloud.
And that exercise was tricky becauseit meant that we had to run a machine
that would be up 24-7,and therefore it's not getting any better.
It's just running.
We're not improving things,and that's not what I'd been used
to, you know, always improving things,looking at the next thing.
And I was actually sort of againstdoing it, putting it in the cloud.
(17:17):
It would also get revenue and that's fine.
The revenue was good.
But I learned later on that it actuallywas really important
to build a system that would work 24-7the engineering behind that.
And, you know, engineeringis something that I don't come natural to.
It's very hard to teach engineering,systems engineering, but making something
work 24-7 is an exercisethat you will learn the next.
(17:40):
The research systems will benefitfrom that.
That was, an interesting lessonI learned.
So maybe I erred too much on the researchside, I'm not sure, but we ended up
putting them on the cloud,all of the clouds.
And I think, IonQ and other companiesthat do that are much better off
for doing that.
The other kind of business questionI want to ask you
was, can you tell us a little bitabout your experience
(18:00):
making that jump from being a startupcompany into a public traded company?
We were sort of lucky in
the timing of this, andit actually happened right out of Covid
when as a private company,we'd just raised a bunch of funding
that would give us,I don't know, maybe a 3 or 4 year
runway, as they say, because we weren'treally bringing in much money.
(18:23):
We were spending it.
And the investors understood this.
This is a long term play.
And when Covid hit our board,which was largely
our investors, they said, we don't knowhow bad this Covid thing is going to be.
You need to really think about that.
3 or 4 years to be 7 or 8 years, is itmaybe there's a recession and so forth.
And the one risk for the company backthen was raising capital.
(18:45):
Really, we weren't too worriedabout our technical plan.
It was very solid, not easy, butwe weren't fundamentally worried about it.
We were worriedabout having to raise money again.
And that'sthat's the bane of startup companies.
With venture capital, you have tokeep raising and you're always out there.
And so when the opportunity came, a so-called blank check company
also called SPAC,Special Purpose Acquisition Company,
(19:08):
they have a spot on the stock exchange,in this case, the New York Stock Exchange.
They have investors with lots of cash,and they want to plow it
into a particular companythat will become public.
So it's a shortcut to an IPO.
We were sort of lucky that these folkswere very interested in the quantum field,
in our company in particular.
And I was told by the folksthat ran the SPAC
(19:30):
that they wanted a companythat had excellent foundational people
fundamentals, and that was Junsang Kimand myself in the in the ion trap basics,
you know, largelyriding on decades of mature research
that was ready to be engineered,
but then also the CEO we hadand that was Peter Chapman.
And he he is an excellent salesman.
(19:51):
So we had sort of both partswhat you need to have a company
and our SPACs kind ofI call them benefactors.
It's not the right word, but our sponsors,they wanted to make a play in quantum,
knowing how risky it was.
But they had a good foundation,an eye toward revenue and sales as well.
You know, how do you balance that?That's the tricky thing.
So that was the transition.
And, you know, we started getting moreprofessional, getting a professional
(20:14):
finance team, legal team, all that stuffthat comes with the public company,
the company was differentafter going public for sure.
And I think the pressure you know,their quarterly stock report earnings
where we talkdirectly to investors in the street.
It's still strange those interactions.
But I give the analysts and the investors,the people on Wall Street
(20:35):
a lot of credit.
They seem to read about this field.
They think quantum will will enhanceAI in the future.
It's a question of time,I guess, in terms of prosecution
of our technical planthat just went into hyperdrive
with the public funding,we were really able to just go crazy
and go much faster,make more systems, hire a lot of people.
(20:56):
So, you know,I have to say, you know, it's good timing.
I think it's a little harder to do thatnow, but, yeah, it worked out well.
Very cool.
So for my last question,you said there's a lot of question
for what quantum computingwill ultimately be used for,
but what impacts do you think it willhave on science and society?
(21:17):
Again, taking the long view,
I'm quite sure that we will have quantumcomputers decades from now.
For sure.
That will be deployed for certain problemsand the tricky thing there is that the
public largely doesn't know or understandor care about these new laws of physics.
They really are fundamentallydifferent laws of physics.
(21:39):
And I think for people to fully appreciatethe power of quantum computers
and to accelerate their usein different fields, people should be
learning the fundamentals of quantumeven at a young age.
And I do believemaybe it's more of a hope, but in 50 years
people will have an intuitionfor some of these weird things,
like “spooky action at a distance”,the Einstein quote about entanglement.
(22:02):
They'll have a feel for it.
They'll they'll know that small,very simple systems, when they're cold
and they're isolated, they,they behave according to these laws.
So I think that's a total shift in the waywe think about fundamental physics,
for sure.
And maybe it'll be infectiousand hit other fields
from chemistry to more complexthings like in biology.
(22:24):
So I think quantum is not readyto tackle biological problems.
They’re still way too complicated,
but I am quite sure it will eventually,and I can say that safely because.
I probably won't be alivewhen it hits that level, but
quantum is, in termsof the entire community of physics.
We all eat and breathe quantum physics,but it's not so useful
(22:46):
out in the real world.
This is a way to change that, and I thinkit also will propel physics overall
as something that is absolutely requiredif you want to understand how to compute.
I mean, I guess physics is behindthe transistor, we all know that.
But the model of a transistoris very easy.
It's like turning on a valve,
a water hose with a valve or somethingelectronically controlled valve.
(23:07):
You can make models that go far.
You don't have to learn all the physics.
But quantum,unfortunately, is not like that.
We have nothing. We can turn to.
So I think quantum computingcan be much bigger than just devices.
It can really bring physics to the masses
and get many more people studying physicsand then going on to other things.
Special thanks to ChristopherMonroe and Alexander Cronin.
(23:28):
For the Discovery Files, I'm Nate Pottker.Watch video versions
of these conversations on our @NSFscienceYouTube channel.
Please subscribe wherever you get podcastsand if you like our program, share
with a friendand consider leaving a review.
Discover how the U.S.
National Science Foundationis advancing research at NSF.gov.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Special Summer Offer: Exclusively on Apple Podcasts, try our Dateline Premium subscription completely free for one month! With Dateline Premium, you get every episode ad-free plus exclusive bonus content.