Hello, and welcome to the Physics World weekly podcast. This episode is part of our ongoing celebration of the International Year of Quantum Science and Technology, and it features the Nobel Laureate William Phillips in conversation with Physics World's Margaret Harris. This episode is supported by Atlas Technologies. Atlas custom aluminum and titanium vacuum chambers and hermetically sealed, bonded bimetal components are used in quantum applications, physics labs, and semiconductor fabs.
And they are built in a fully integrated facility with on-site design, development, and manufacturing capabilities. Let Atlas help you solve your next engineering challenge. Learn more at atlasuhv.com. Here's that conversation with Bill Phillips. He talks about quantum entanglement, atomic clocks, the future of quantum technologies, and much more. It's the International Year of Quantum Science and Technology. I'm speaking with a real pioneer in
this field. Bill Phillips shared the 1997 Nobel Prize in Physics for his contributions to laser cooling, which uses light from precisely tuned laser beams to slow atoms down and reduce their temperature to just above absolute zero. Hello, Bill. Welcome to the podcast. Glad to be here. So how did you first become interested in quantum physics? Well, I guess that, as an undergraduate, I was invited by, one of the, the professors at the college I went to, a small college.
There was only four physics majors who graduated with me. But I was invited to participate in research that he was doing on what we called electron spin resonance. In other words, what we were doing was using the, flipping of of, unpaired spins in a in this case, in a solid, sample as a way of investigating the the structure behavior of, of a particular, compound. And this is a fundamentally quantum mechanical thing. The the direction in which the spins can point is a quantized, quantity.
It, unlike a spinning top, which in principle, you could have the axis pointing any direction you want. Spins, electron spins, or any other, spins, proton spins can only have certain directions in which they can spin. A spin, what we call spin one half. That is something that has the the smallest amount of angular momentum allowed, by by quantum mechanics. And that kind of a spin only has two different, possible orientations. Itself, a pretty weird thing. Why why is
this? It's it's one of the things that I found really fascinating, at the beginning of my scientific career, the this what we called space quantization, the fact that, that you could only have two possible directions. So I was beginning to to be part of the quantum adventure as as an undergraduate. And in a certain sense, everything is quantum mechanical. I mean, we wouldn't be here. We wouldn't be able to have this conversation.
There wouldn't even be rocks around if it wasn't for quantum mechanics. Matter as we know it could not exist without quantum mechanics. So to to say, you know, how did I get started in quantum mechanics? Well,
that's all there is. But interestingly then, I I did a, a semester study at, Argonne National Laboratory, outside of Chicago and working on the same kind of thing, with, scientists who became my mentors, two, physicists from Argentina who became my my mentors for that that semester, also working on on electron spin resonance essentially full time. You know, this is my first introduction to people who are doing research,
full time. And then I was invited by the the man who became my thesis adviser at MIT, Dan Kleppner, an amazing physicist. He taught me so much, really taught me how to think like a physicist. And, again, doing resonance, in different ways in different, systems, but using this quantum mechanical, idea of, of quantum mechanical spins in a different way.
And it was in his laboratory that I first encountered tunable lasers, another wonderful tool for using the the quantum properties of matter to to explore what's what's going on at the atomic level. That was the thing that that led me to laser cooling. Now you mentioned that the idea that spin is quantized is not like a clock, that it can only have, you know, spin up, spin down. You mentioned that being sort of counterintuitive.
A lot of people come across quantum mechanics and think, oh, this is this is so weird. This is so strange. Did you did you find it strange? Well, probably not at the beginning. That is the my my first introductions to quantum mechanics I mean, certainly the the the spin quantization is strange, but probably the things that are most weird about quantum mechanics, superposition and entanglement were not the first things that I learned
about, quantum mechanics. Now the fact of the matter is I'm the kind of person that is just enthralled by everything. My my colleagues, sometimes joke that I'm I'm I'm just so entranced by everything that happens in the natural world. You know, drop something. Oh, wow. It's accelerating. Yeah. So so I guess that that when I first was introduced to quantum mechanics as a boy reading,
books about about quantum mechanics. I was I was enthralled with with the ideas, but I don't think it was until it wasn't until I was in graduate school that I understood the deeply weird nature of, of quantum mechanics. And I don't think even in graduate school did I understand how strange, entanglement is. Only when I started to study, what John Bell had done. Okay. Maybe I should back up
a little bit. Nineteen thirty five, Einstein with two of his friends, Podolsky and Rosen, write a paper in which they claim quantum mechanics just can't be the whole story. Because if it were, then things would be just so weird it couldn't possibly you couldn't possibly believe that the world was this weird. And what they were talking about was they they said, here's here's a a system in which two quantum objects are are entangled. What do we mean by entangled?
Okay. When you measure something quantum mechanic, let's say it's a spin. It could be either up or down. But the wonderful and amazing thing about quantum mechanics is it can be in a superposition state of being up and down. Now how I describe that is very difficult to do because there's no classical analog. Sometimes I say it's both up and down at the same time. I have colleagues who say the right way to describe it is to say it's neither up nor down.
But but in any case, somehow or other, it's not up or down. It has the potential for being either when I measure it. And when I measure it, it will be one or the other. And there's no way that I could predict beforehand which one it'll be if I initially prepare it in this state we call a superposition, which is hard to describe because it's quantum mechanical, and we don't have an analog in our ordinary everyday classical world.
But the but the cool thing about entanglement is that you can have two particles that are in a state such that when measured, if you measure one of the particles to be up, the other particle will necessarily be down. But before you do it, you have no idea whether that first particle is gonna be
up or down. But once you've measured that one, you know what the other one's gonna be even though beforehand, you'd have no way of predicting what that other one would be and even though there's no connection between those two particles, they could be so far apart that no signal traveling at the speed of light could get from one particle to the other to tell that second particle, you'd better be down now, because the other the first one was measured to be up.
So Einstein and Podolsky and Rosen had a not exactly that situation, but a similar situation and showed, in in their paper that this is just too weird. How could nature be like this? Bohr wrote a response that everybody agrees was totally inadequate, saying no. This is just the way things are. And what Bell did was he said, look. Let's make some very simple assumptions. Let's make the assumption that the property of an object exists before you measure it. Seems like a reasonable assumption.
You know, permanence of objects that we learn as babies, you know, that, that things exist even though we're not looking at them. And then let's also assume that I only need to understand what happens in the immediate vicinity of something in order to know everything I need to know about that. In other words, if I'm gonna be making some measurements in the next microsecond, I don't need to know what's going on more than a microsecond away traveling at the,
the speed of light. Those two things sound like perfectly reasonable ideas. And what Bell showed was that if you made those assumptions, you would predict results that were in contradiction to quantum mechanics. So now it was basically what Einstein had already done, but in a very clear way that could capture the attention of every physicist. The trouble with Einstein's nineteen thirty five paper, what we call the EPR paper for Einstein Podolsky Rosen, was that most people thought, this
is some technical thing. It's it's a very subtle point. We don't need to worry about this. Well, Bell said, look, if you believe in quantum mechanics, then that means you can't believe in these completely reasonable ideas. And I think that captured people's attention in a way that that Einstein's paper had not because people said we don't have to worry about this. This is a technicality. It's not
gonna affect us. But the amazing thing was that nobody had yet done an experiment, to show whether quantum mechanics was right in this situation or whether what we call local realism was right. That what I've described you is basically a way of of of describing what we call local realism. There's a perfectly reasonable idea that quantum mechanics says isn't true.
And now we have the experiments done by first John Clauser and then by Alan Aspey and then all sorts of extra things by, Anton Zeilinger. And I mentioned those three because the three of them got the Nobel Prize in, what, 2022, for their experiments basically proved that Einstein was wrong. Nature, yes, is indeed that weird. I didn't understand that even as a graduate student. How how deliciously weird nature is because of quantum mechanics.
Would you say that entangled then was the most challenging concept in quantum mechanics you've come across in your work? Most challenging? Let's say the most deliciously weird, because it's it's easy enough to describe. I can write down an entangled state. It's pretty easy. But to understand how weird that is, that's the thing that's, that's really, I think, so impressive about quantum mechanics is that it's so deliciously
weird. So it's not like it's that hard to it's not that hard to understand in a formal sense, but it's hard to get your mind wrapped around it because it is so weird and so distinct from the kinds of things that we, experience on a day to day basis. And the the thing that it violates, local realism, is so reasonable. You know, Einstein said, are you telling me the moon doesn't exist when I when I
don't look at it? You know, it What quantum principle would you say has had the biggest impact on your own turning from your your university experiences onto the work you've done and the work you continue to do now? For me, look, the things that I've done with laser cooling are pretty garden variety compared to, entanglement and, and and superposition. So laser cooling, I would say it's just, you
know, the quantization of energy levels. That's the thing that that has driven, the idea of laser cooling, but it has enabled laser cooling has enabled the creation of, atomic clocks of incredible precision. And these atomic clocks fundamentally use superposition. It's the bread and butter of atomic clocks. The process known as, Ramsey spectroscopy uses entanglements. It's one of the early practical, applications of entanglement. So you take, an atomic system that has two energy levels.
And the Ramsey method, what it does is it you put in a pulse of, say, radio frequency or microwaves or laser light, in more the more modern way of doing it and put the atom into a superposition of these two states. What that means is that the the atomic system will evolve in a certain sense. That is something is pulsating or rotating in the atomic system if you like, at a frequency that is equal to the
frequency difference between those two states. And then after a certain length of time, the longer the better, you give it another pulse. And depending upon the phase of of that pulse, be it a, you know, radio frequency or light or whatever, it's some oscillating thing. The phase of that thing compared to the phase of the atomic system, you'll either promote, the atom to the state different from the one you started in or put it back in the
state that you started with. And the longer you wait, the more precisely you'll know how much the phase was different. So this is the principle behind Ramsey's method for, making an atomic clock, and that's the way almost all atomic clocks work today. And my work and the work of my my group and other groups around the world who have contributed to this whole laser cooling enterprise has made atomic clocks of just incredible precision possible.
And so I feel a a real kinship to the idea of of superposition because the major application of laser cooling in which I've been so so privileged to to to be a part of is, is using this, superposition principle. So let me tell you a story. Mhmm. Sure. When I first came to NIST, where I am now, the the National Institute of Standards and Technology, at that time, it was called the National Bureau of Standards.
And, when I first came to to to our metrology institute in, 1978, The very best clock in the world was in our laboratories in Boulder, Colorado. Now Boulder, Colorado is about a kilometer and a half above sea level, and here I am in Washington close to sea level. Because of Einstein's theory of general relativity, he showed that clocks would run slower if they're deeper in a gravitational potential. The effect is not very big.
At a kilometer and a half, it would be about one and a half parts in 10 to the 13. The very best clock operating in Boulder at an at a an altitude of one and a half kilometers was good to a part in 10 to the 13. So what that means is if you had two such clocks that were at the very best level, which we didn't have two such clocks, only had one. But if you had two such clocks, one at sea level and one at, the height of our Boulder labs, you would just barely not be able to resolve the difference.
Today, we can resolve a difference of less than a millimeter with the clocks that exist today due in part to laser cooling and in part due to any number of other developments that have happened along the way. I just find that so amazing. Turning now to the present, actually. What are you working on now? Are you still involved in atomic clocks or in laser cooling or Well, not so much
involved in atomic clocks. I I think of of our laboratory as having been a generator of ideas and techniques that could be used by people who make atomic clocks and that we're continuing to do that that that kind of thing to doing some of the fundamental work that enables other people to do some amazing things that that, that they're doing. So one of those people is, Junyi at in our laboratories out in Boulder making these clocks that are good to better than a part in 10 to the 18.
Okay? But the same kind of work is going on at the National Physical Laboratory. So there's atomic clock groups there where they're doing microwave clocks and optical clocks and, just amazing, amazing things. Again, using laser cooling techniques. But but we're doing all kinds of, adventures using ultra cold atoms. So ultra cold atoms is the main theme of what our research group is doing, which came from laser cooling, but other tricks as well.
So after you laser cool it, the coldest temperatures we get using laser cooling is a mere one millionth of a degree above absolute zero. So I know that sounds pretty cold, but by today's standards, it's not that cold. It's where the atomic clocks, that define what we mean by a second today are operating at that temperature. But those clocks, which are based on a microwave transition in cesium, They're good to about a part in 10 of the 16. I mean, you might say, well, part in 10 of the 16.
Who needs to do better than that? And the answer is we are never satisfied. We always wanna do better. And so people are making clocks at a part in 10 of the 18, but not with cesium, with, atoms like strontium or ytterbium, depending on what, lab you're in, aluminum ions.
So people all over the world are making atomic clocks using laser cooling techniques, using other techniques that have developed in our laboratories and in other laboratories that are that are good to on the order of a part in 10 to the 18. So in other words, two orders of magnitude better than the the cesium clocks, but cesium is the definition of the second. That is the formal internationally agreed upon definition of the second is a certain number of oscillations of a cesium atom.
That's the definition of the second. And that definition is two orders of magnitude worse than what we can do with a ton of clocks. So what that means is we're gonna redefine what we mean by a second. And I am serving on a committee along with people from from all over the world to decide or to recommend to the international bodies that make the final decisions what should be the new definition of the second. And
there's no question. The new definition of the second is going to be the oscillations of some atom, maybe some set of atoms at optical frequencies, not at microwave frequencies, not at nine gigahertz, but at something on the order of 10 to the 15 hertz. Now why? I mean, when it's going so much more rapidly, it means that a lot of errors are much smaller as a fraction. So that's one of the reasons why we wanna do,
optical quarks as as we say. And, and so instead of microwave shining onto atoms, we're gonna be shining lasers onto atoms. Lasers that have a stability that is so amazing compared to what was possible when I first got into the laser business. These lasers are are stable to a tiny fraction of a cycle per second when the light is oscillating at 10 to the 15 cycles per second. It's just it's just amazing. And so we have contributed
to the laser cooling part of that. We've contributed to some of the ways in which the atoms are trapped because you gotta hold on to these atoms. If, if you let go of the atoms, they'll just fall in the gravitational field. Now we actually use that in the cesium clock. We toss the atoms up and they fall back down after about a second. But we want something longer than that,
so we hold them in traps. And these traps have to be very carefully arranged so that they don't themselves distort the the ticking frequency of the atoms. But we're doing other things like, quantum information. So this is one of the big deals in current day, investigations having to do with quantum mechanics, is using single atoms or other single quantum entities as what we call qubits. Quantum bits, ordinary digital information is stored and processed using what we call
bits. And those bits represent a mathematical zero or one that in binary, mathematics is the thing that allows you to store and and process information. And now we're gonna make those zeros or ones be represented by, say, for example, a spin that points up or down. The two spin states that we talked about earlier fit very nicely into the idea of binary
logic. But the beauty is or the the amazingly difficult thing is that they can be zero or one, but they could be in this superposition state, which is neither zero or one or both zero or one depending on how you wanna describe it, which we don't really know how to do in classical terms.
Sounds like a disaster because one of the great strengths of binary information is that, the things that store those those bits are things like a transistor that is turned on or off like a switch, a spot on a, a DVD that's burned or not burned, a patch on a magnetic disc that is magnetized in one direction or another. These are the things that store classical information, and they're so good that you can be really, really confident that this thing is one thing or the other. No
uncertainty. The the the uncertainties the the the errors where you might get something that was, you know, only half burned or or some so you wouldn't be able to tell what it was. That's so tiny. It's just it's just ridiculous. And that's a wonderful strength of binary information. And now I'm telling you, we wanna make it so so nobody could know what the state of this thing is before you measure. That sounds like a disaster, but it's not.
If you do the right kinds of things for certain kinds of problems, the ability to put the bits, now quantum bits, into superpositions means that you can do the problem in a lot fewer operations than would be needed to do that same problem if you used a classical computer with ordinary bits. So one of the first examples of that was, done by a guy named Peter Shor,
nineteen ninety five. He came up with an algorithm that showed it was possible to factor numbers using a quantum mechanical approach in a time that would be much shorter or at least the number of operations that would be much smaller than the number of operations needed by an ordinary classical computer. Factoring is what is technically known as a hard problem. What that means is that the number of operations required to do that problem, to solve that problem grows exponentially
with the size of the number. So if I want to factor a hundred digit number, that the time it takes to do that grows exponentially with how long how many digits that number has. But if you do it quantum mechanically, it doesn't grow exponentially, so it becomes an easy problem. This is absolutely amazing that changing the hardware on which you do the calculation has changed what we call the complexity class
of a problem. Now I have to be careful because nobody's ever proved that a classical calculation of, factoring is exponentially hard. It might be that some clever mathematician will find an algorithm that makes it easy. Nobody thinks that's gonna happen. Everybody thinks that this problem is gonna remain exponentially hard, and that's one of the reasons why certain kinds of encryption are used with a lot of confidence. So let's
back up. Why was it so important that Shor came up with this algorithm? And the reason is that something called public key encryption, which you may not know what that is, but you use it every day. Public key encryption depends upon the fact that it's hard to factor numbers but easy to multiply them. So this process, multiplication, is the inverse of factoring. And the fact that it's asymmetric leads to the possibility of what we call public key encryption.
So let's say you buy something online. Now your credit card has to go to the company that's selling it to you. So your computer actually, that company sends your computer a number, a a big integer, and they know the factors and nobody else does because they don't know how to factor big numbers because it would take too long. It would take years or centuries to factor this big number, so nobody
knows the factor. But they know the factors because they took two smaller numbers and multiplied them together, and that was easy. That number is then used to encrypt your credit card number. And the encryption is easy for the person who knows what the factors are and essentially impossible for the person who doesn't. So that means somebody trying to intercept the, transmission between your computer and the the computer at the ecommerce company can't get any useful information.
Now what if that evil doer had a quantum computer? Then that evil doer could factor the the number and figure out what your credit card number is and then steal it and use it to buy, you know, TVs or whatever the, you know, evil doers buy. Now the fact of the matter is that probably you don't have enough money in your bank account to make it worthwhile for some criminal who made a quantum computer. We don't have quantum computers that can do
this yet. Okay? It's not easy to make a quantum computer, even one that can do very simple problems, let alone one that can factor big numbers. But if somebody did that, then they could decrypt secret messages that really do matter. So, for example, the diplomatic communications between, say, diplomatic offices of some country often contain secrets that people don't want revealed for decades, that it would be it it would it would matter to the security of of of a country if if these diplomatic
secrets were revealed even decades from now. Military secrets, intelligence secrets, you know, who is the spy and who have they contacted? These these could be devastatingly dangerous things if they were were revealed even decades from the time that they were first happened. And so people worry that sometime in the future, you're gonna have a quantum computer that could reveal secrets because people could wiretap communications, but it wouldn't do them any good
because it's encrypted. Okay? But if you had a quantum computer, you could decrypt things. So what are you gonna do? Well, quantum mechanics comes to the rescue. You can have quantum communications that cannot be eavesdropped upon because of something called the no cloning theorem. An eavesdropper could not intercept the message, make a duplicate of the message, send it on so that the person receiving the message doesn't realize that the message has been,
intercepted and then use that intercepted message. That's not allowed by the laws of physics. So we're very confident that these quantum, communication methods would work, and people are actually using them. There are places where, I'm not sure they need to, but they can say, look, our bank communicates with other banks using quantum, cryptography. So we know that our our transactions are, are secure.
So that's one way. Another way is a classical approach where you come up with new encryption algorithms called post quantum encryption that are not dependent upon this asymmetry of factoring and, and and multiplying. And then these will not be able to be attacked by a quantum computer as far as we know. No one has proved that these new algorithms can't be attacked by some quantum computer. It's just that nobody has figured out how.
And very clever people have been working on it to try to figure out ways of doing this and have failed. And so we think these things are pretty secure. But with so many like, with so many things, we're not sure. So this is places where quantum mechanics puts us in danger and quantum mechanics comes to the rescue. And so in my lab, we're not making quantum computers, but we're doing some of the things that will enable new kinds of quantum computers to be, to be to be made. Mhmm.
So let's talk about quantum computers. You need you mentioned that you can make a quantum computer or a quantum processor with cold atoms. You can make it there's other types of qubits out there, trapped ions, super connecting circuits, that sort of thing. Do you have a view about which of these platforms is likely to succeed or which Yeah. Well, this this this has been a an ongoing question. What's the what's the winning platform?
And my attitude from the beginning and still today is that it's too early to settle on one particular platform. And it may very well be that the final quantum computer that does things like factor numbers, it's not even clear that that's the most important problem that you want to do, is likely to be a hybrid where computations are done on one platform and storage
is done on another platform. So superconducting quantum computers are really nice and fast, but they can't store information for very long, whereas atoms and ions can store information for a really long time. They're very robust, very isolated from the environment, but they're kinda slow in the in the computation process. So it might very well be that you wanna use the best features of different platforms in different in in in different parts of your of your quantum computer.
But, you know, what do I know? It's, I think we're a long way, and this is not a majority view or at least is not a universal view. We're a long way from having quantum computers that can do interesting problems that, a lot faster than what, classical computers can do. You've probably heard that somebody used a quantum computer to do a a problem that would take a classical computer, you know, a septillion years to do. Maybe so, but it's a little misleading. They chose a problem.
It was easy for a quantum computer to do and hard for a classical computer to do, and it's a problem nobody cares about. Nobody's gonna make any money solving that problem. People definitely gonna make money factoring numbers. People probably are gonna make money by, doing quantum chemistry, learning how some molecule that either you haven't made yet or you haven't tested yet, how it would behave.
That's the sort of thing that could very well lead to to things that people care about that that, are gonna make a difference in our lives. That hasn't happened yet, but we may not be that far from it. Just to give you an idea of how far away we are from this, I proposed, a bet with a colleague of mine. His name is Carl Williams. He's well known in the quantum business.
And the bet is, at a certain future date, will we or will we not have a quantum computer that can factor numbers that a classical computer of that future time cannot factor? Okay? And he proposed the the date of 2045, which is only about twenty years from now. Right? Why 2045? It's about fifty years after Shor and after the first demonstration of a quantum gate by Dave Wineland and his group at our NIST laboratories in Boulder. And and it's a well defined problem and not an easy one.
And, I'm taking the point of view that, no, we will not have a computer that can do that, And he's taking the point of view that, yes, we will. And we'll put up some money and it'll go into some fun for a scholarship or a prize depending upon who wins. And I expect to be dead by then. But, what I hope is that this bet will encourage people to work hard on solving the problems that need to be solved in order to,
to make this work like error correction. We haven't talked about error correction, but that's the big thing because these quantum processes are prone to errors much more so than the classical computers. So you've got to be doing something to fix that, and that's called quantum error correction. And people are starting to do that, and we need more of that. And so that's one of the reasons for having a bet
like this. When I talk to people, other scientists working in the field, and ask them which side of the bet they would like to be on, it's about fifty fifty. So I think this is a great bet that that we could expect that in about twenty years, we're gonna have or, you know, it'll be of that order of time where we'll have really competent quantum computers that can do that kind of thing. Now as I said, I don't think factoring is the most important thing.
What I want is a quantum computer that can do quantum mechanics. This, this exponential growth that I described for factoring is also true of many problems in quantum mechanics. Let's say that you've got a model that is trying to describe magnetism. Magnetism is one of our favorite subjects in, in quantum mechanics. So let's imagine it's a chain, a one dimensional chain of spin. So on each link of the chain, there's
a spin, a quantum mechanical spin. And magnetism is all about how do these spins interact to produce certain kinds of states, like all the spins pointing in the same direction. That's what we call ferromagnetism. Every other spin pointing in opposite directions. That's what we call antiferromagnetism.
And quantum magnetism is, at least in some circumstances, a hard problem because the number of possible states of all these spins pointing up and down is let's say that the spin can only point up or down and you've got n spins. There's two to the power n number of ways you could do this on a chain. And if you add one more spin to the chain, you've doubled the number. That's what we mean by exponential growth.
So that means that any reasonable size chain is impossible for a classical computer to do. Once you've got a few tens of spins, a classical computer cannot do a brute force calculation, but a quantum computer could. And so having a quantum computer that can, reliably calculate tens of of of qubits that represent the spins, this will be a big deal. Now you've probably heard that these quantum computers already have that many qubits. Yeah. But they're not what we call logical qubits.
You see, the thing is that because of these errors, you're not gonna get a reliable answer from the kinds of of quantum computers we have today. You have to do error correction. The way you do error correction is you assemble a number of physical qubits into what
we call a logical qubit. And then by making measurements on the logical qubit that are not determining what the state of the logical qubit is, but determining what the relationship is between some of the qubits without learning what the state because you can't learn what the state is because that would destroy the superposition state. But you can determine whether an error has happened and and fix
that error. That's what quantum what quantum error correction is about, and people are just starting to do this. And this is what one of the things that's just so exciting right now is that people are making computers that have logical qubits and doing error correction. It's not good enough yet, but it's getting there. So would you say that's one development that our listeners, this podcast, should look out for in this field? You know? Who are the
people to watch in this area? What are the things we should be seeing happen in the between now and 2045, let's say Yeah. I think when the bet's gonna happen? Right. So I think the thing you should be looking for is things like how many logical qubits you've got that can be entangled with each other. And this is one of the things that some recent news, has shown that this number is starting to become larger than it was in the past.
And how long or for how many operations can these logical qubits maintain their coherence? That is their their quantum ness, without, having some sort of an error with and that those errors occur because they're connected weekly, we hope, to the environment. The environment comes in and and and messes things up, and that's what we have to correct by quantum error correction. First, we have to prevent it.
That's one of the things that using cold atoms allows you to do is to prevent the connection to the environment, but it's never perfect, and correct it when when the environment does mess you up. And it's that correction. How long is the coherence of this, of this state lasting? What I often say the thing we need is an immortal qubit, something that is not killed by the environment that lasts throughout the whole calculation and and for a calculation that's long enough
to do interesting things. So those are the the things you wanna be, looking for is how many operations can you do and still maintain the the quantumness of of that qubit, and how many, logical qubits can you have. Those are are some of the key things that are gonna determine whether you really have a, a competent quantum computer. And what about in the other areas we talked about in quantum information and, optical clocks? Yeah. So so optical clocks are already wonderful
and getting better all the time. One of the things that recently came up was a nuclear clock. So now, I mean, it's optical if you call, sort of far ultraviolet optics. Okay? I mean, it's not so far. Okay. So let me back up and say why nuclear clocks are so wonderful. There's one nucleus, thorium, that I forget which isotope it is, that has an isomer state, an excited state of the nucleus that is only a little bit above the ground state.
It's low enough in energy. I think it's about eight electron volts, whereas, like, optical things like two electron volts. So it's not not not not so far above, which means that we can use well known techniques to produce what is effectively laser light to shine on the nucleus and have it make a transition. Most nuclear states are kilovolts or more above the ground state, and we just don't have any way of making laser
like light at those energies. But, you know, eight electron volts, yeah, we can do it and have done it. That's the thing. So in the just in the past year, starting at PTB, they they used a broadband laser to excite, the state for the first time. Now starting to really pin down what the energy was because people didn't even know what the energy was. This is amazing thing. So a guy at PTB, I don't know, two decades ago said this looks like a really great idea.
They didn't even know what the energy was to within a factor of two. And now they know what it is to to some really fine, level. I forget what, but, you know, many, many digits. So so this has been a huge improvement over the years. And just this year, laser excitation at PTB and then a couple of other groups, one of them being, our NIST laboratories in Boulder, have done it with with really narrow band lasers. So this is the beginning of the possibility
of a nuclear clock. It's still years, in the future, but the beauty is really high frequency. So that means, again, various perturbations are gonna be a smaller fraction, but a nucleus. The nucleus is much less susceptible to outside influences than the atom where its outer electrons are the ones that are doing the work. And those outer electrons are basically exposed to all kinds of bad things that might
happen. Whereas the nucleus is protected. It's sitting sitting in deep inside the atom with lots of electrons around it sort of protecting it. And, and the hope is that you'd be able to make a clock that is much less susceptible to the kinds of things that that mess up our our atomic clock, but probably not for some decades. Is it gonna be good enough that it's really gonna compete with the part in 10 of the 18 atoms that are going on?
So one of the things you're gonna look for is is the international community gonna come to a decision about what is gonna be the new, the new definition of the second? What atom is it gonna be? We haven't even decided whether it's gonna be a single atom or whether it's gonna be a bunch of atoms. I'm very much in favor of the single atom approach because Simpler. Simpler. Everybody knows what you mean when you say this is what a second is. So,
but we'll see. You know, there's there's, arguments to be made, on, on both sides of that. And, so that's another thing to watch, and those things are getting better all the time. So here's one of the the really wonderful things that's being done with these incredibly accurate clocks. One of the the really fundamental questions we might ask ourselves about the way nature works is, are the fundamental constants of nature, in fact, constant?
Now the things for which this is important are not things like, say, the charge of the electron or Planck's constant, because the value of these constants depends upon what we choose for our, unit system of units. In fact, right now, the charge in the electron and Planck's constant cannot change because it would be illegal. Those constants have been set by international agreement. The things that matter are things like the fine structure constant. What is that?
It's a combination of constants like the charge electron Planck's constant and the, what's called the, electric, permittivity of of the vacuum, and the speed of light all combined together to create a constant that is dimensionless. It is a dimensionless way of describing how strong electric interactions or electromagnetic interactions are. And we have similar dimensionless constants that describe how strong nuclear,
interactions are, strong force, weak force. Well, anyway, the the the fine structure constant describes how, how strong electrical interactions are, and that means it's basically the thing that sets the scale for almost everything that's part of our daily lives. Chemistry, which is, you know, basically what makes our bodies work, is defined with the kinds of things that are possible defined by the fine structure constant.
Some people pointed out that if fine structure constant were different by 10%, we couldn't exist. Our body chemistry wouldn't work in the way that it does. So, you know, it's kind of a either a happy accident or, you know, some divine intervention that makes our, our body chemistry work. Or we won the cosmic lottery and there's 10 of the 500 other universes where the flying structure comes in something different and there's nobody interesting there.
But one of the questions we could ask ourselves is, does it change with time? Well, now because of these atomic clocks, we could have two atomic clocks that operate on different atoms or maybe on different transitions even within the same atom that depend upon the fine structure constant in different ways. So depending on what the transition is, it could depend very strongly on the fine structure constant or very weakly on the
fine structure constant. So if you've got two such clocks and you just let them, you know, compare their frequencies year after year, you can determine whether from one year to the next the fine structure constant is changing. And this has allowed us to say that we know the fine structure constant is not changing by about a part in 10 to the eighteenth per year. And as the clocks get better, we'll be able to put a finer, determination on that, or you might find that
indeed it does change. That would change everything. We would have to change our our view of of the way the world works. Another thing which is really exciting is Einstein's theory of general relativity tells us that if we have two clocks and that are of different sorts and we move them together in a gravitational field. And these are clocks that don't depend on gravity, so a pendulum clock wouldn't be the right the right thing, but an atomic
clock would be. If we move them into different, places in a gravitational potential, those two clocks better stay in the same ratio. If they don't, it means that the most fundamental principle of general relativity is wrong, what we call the equivalence principle. So putting two such clocks, maybe one in a satellite, one on the earth, maybe two of them together in a satellite with an eccentric orbit, this would really
be be exciting. I mean, we've done this to some extent, but but with these new clocks, we could do this in in a way that has been unparalleled in the past. And so we could test this feature of general relativity, at a level that's never been possible before. This is really important because we don't know how to have a unified theory of quantum mechanics in general relativity.
Well, if we could find out that the equivalence principle fails in this particular way, it might give us some insight into how we can we can solve this great unsolved problem, this great challenge of our current time of how do we reconcile quantum mechanics with, with gravity. There's a lot of big things there. Taking it back to your own career just for a moment, reflecting on your career so far, what are you most proud of?
Well, back in around 1988, we accidentally discovered that the temperature to which you could laser cool atoms was lower than what everybody said was possible based on the theory of laser cooling at that time. And it was an accidental discovery. I mean, we were not looking for this. We were just fooling around in the lab trying to to see whether laser cooling was working the way it was supposed to, and the first indications was it looks every like,
everything's great. And then we started to see some problems, like, everything wasn't great. And we started to pound on that and eventually learned that the temperature was was too low. And that was the thing that made people pay attention. There's other little things, things weren't working out. You know, there's some detail. We'll figure it out. But when the temperature started to be lower, you see I mean, the whole idea of laser cooling is to make the temperature as low
as you can. And we had apparently made it lower than you can. So that got people excited. And then people came up with explanations. Our our our friends in Paris at the Ecole Normale came up with, with explanations for what was what was going on. Steve Chu, who was at that point at Stanford, was also working on on understanding the the the theory behind it, and that really changed things in an important way. It made possible
laser cooled atomic clocks. So all the laser cooled atomic clocks that use cesium today use, that feature that the temperature is lower than what the original theory of laser cooling said it was. You know, even these optical clocks can be, made to work pretty well with that, but there's other techniques. And it made those other techniques like, I haven't even talked about Bose Einstein condensation.
An amazing process that happens because of a purely quantum mechanical feature that makes atoms of the same kind fundamentally indistinguishable. And this is a feature that that quantum mechanics gave us, this idea of indistinguishability, which was not something that that was understood before
before quantum mechanics. In fact, in 2024, a hundred years ago, Satyendra Bose came up with the idea that photons were indistinguishable and therefore the statistical mechanics of photons would be different from the usual statistical mechanics that Boltzmann or Maxwell and and and such people came up with because of the fact that the particles are indistinguishable.
And the way you count the number of possible states is different because, if you interchange the particles, that's not a new state because, because of the indistinguishability. And that indistinguishability has led to this new what some people call a new state of matter called a Bose Einstein condensate where all of the atoms almost all the atoms are
in the same quantum state. Well, that was facilitated by this discovery that the, that the temperature could be so much lower because it meant in order to get this state of Bose Einstein condensation, you've got to cool the atoms to a very low temperature. You've got to have a high density of atoms, and you start off in a better place if if the atoms are colder. And and our accidental discovery allowed that that to happen. So that's probably the thing that I'm most
proud of. But we also accidentally discovered what are called optical losses. Now I shouldn't say that we accidentally discovered it because people already had the idea. In fact, one of the earliest ideas of laser trapped atoms was was in fact 1968. So this is laser cooling. The idea wasn't even came out until '75, and it wasn't even demonstrated until
'78, '10 years later. But in 1968, a Russian, physicist named, Vladland Litokov came up with the idea of trapping atoms in a standing wave of light. And in that way, eliminating the Doppler shift because the atoms would be trapped in over such a small distance that effectively the thing called, there's a thing called Dicke narrowing that, that that gets rid of the Doppler shift. And so this would be great.
Ten years before there was even laser cooling that would allow you to put the atoms in such a thing, that just wasn't possible, but he had the idea. Well, that's exactly what Junyi does in his lab. He puts the atoms in an optical lattice and holds them in place and then uses that as an atomic clock. Well, okay. So everybody knew this was a was was a thing, but we saw that process of the atoms being trapped in the optical lattice in our laser cooled samples when we weren't looking for it.
We were trying to measure the temperature of the atoms in the, the laser cooling configuration Instead of turning lasers off and just letting the atoms expand and measuring the temperature from the distribution of velocities, instead, we wanted to measure what the temperature was while they were being laser cooled. And the idea we came up with was, let's look at the Doppler shift
induced by the scattered light. So light comes in, and if it bounces off an atom that's moving, there'll be a Doppler shift, and we can measure that Doppler shift and see what the distribution of velocities was. So we did that, and the distribution of velocities just floored us. It was so odd. Instead of being a nice smooth broad distribution showing the distribution of of velocities, it was that with a big sharp peak right in the middle. So what is this peak?
And, and and and we fought for a very short period of time. Did we accidentally make a Bose condensate? Because this is before peep anybody made a Bose condensate, but people were talking about it. Then we realized, no. No. That's not what we did at all. What we're doing is we're trapping the atoms so the Doppler shift goes away. So this thing that we were trying to learn about, the Doppler shift, was partly being canceled because the atoms were being held in what we now call an optical lattice.
And then people started to do it everywhere, and, and now it's led to these wonderful clocks that have the atoms in the optical lattice. So that was another case. It wasn't nearly as astounding as the sub Doppler laser cooling because it was expected. It's just that we weren't looking for it at the time that we saw it, but that theme of learning about things accidentally has seems like it's been a recurring theme in our laboratory.
And I think it's an important thing for people to understand about the way that science is done. Often, science is done not because people are looking for a particular goal and working towards that, but it happens because they're fooling around and see something that wasn't expected. And we have to have a lot of that kind of activity. If all of our science activity is directed toward specific goals, we're gonna miss a lot of really important stuff that allows us to get to those goals.
And this idea of what is sometimes called curiosity driven research just fooling around, is so important to the development of science. And and without it, we're just not gonna get to where we need to go. So final question, maybe the most nebulous one of all. What does quantum mean to you?
Yeah. So interestingly, I was on a panel a couple of years ago that was run out of Oxford, in which they had gathered a handful of Nobel laureates together in physics to answer the question, what was the most important discovery of twentieth century physics? And most of us said quantum mechanics. And so when it was my turn, I said quantum mechanics. You know, what is quantum mechanics? Just the I said, it's it's the fact that particles behave like waves, which we haven't even talked about here,
and waves behave like particles. This wave particle duality is at the heart of quantum mechanics. And that understanding, which a lot of people would say was sort of the ordinary part of quantum mechanics, that understanding led to a technological revolution that completely changed our daily lives. We all walk around with mobile phones that wouldn't exist were it not for quantum mechanics. Semiconductor electronics depends upon the quantum mechanical nature of
electrons. And I'm just looking around at the kind of, electronic devices that you're using for for for this, and and it's it's all quantum mechanics. So for me, quantum mechanics is this idea that waves are particles and particles are waves and that this has led to a technology that has changed our daily lives in a fundamental way. Steve Weinberg was also on this panel. And when it came time for him to say to answer this question, what is quantum mechanics?
He says, well, quantum mechanics is the departure from classical mechanics where we thought of particles as having a position and a momentum, and we described the state of something by saying what the position and momentum was of all the particles in the system. In quantum mechanics, it's a vector in Hilbert space. And I thought this is the difference between an experimentalist and a theorist. And, of course, Hilbert space behaves differently from from
regular space, and so everything behaves differently. But but the perspective you see, both of which are, of course, are true and both of which I think embody the the the wonder of of quantum mechanics. But but we had very different points of view about what quantum mechanics was. Bill Phillips, thank you so much. It's been really wonderful to talk to you. It's been a pleasure. That was the Nobel laureate William Phillips in conversation with Physics World's Margaret Harris.
I'm afraid that's all the time we have for this week's podcast. Thanks to Bill and Margaret for a fascinating discussion, and also to our producer, Fred Isles. And a very special thank you to Atlas Technologies for their generous support of this episode. We'll be back again next week, but in the meantime, do check out the latest episode of the Physics World Stories podcast.
Host Andrew Glester is joined by three expert guests to explore the impact of artificial intelligence on discovery, research, and the future of physics. That episode is called AI and the future of physics, and you can find it on the Physics World website or at your favorite podcast provider. Atlas Technologies is happy to support this episode and the exciting work being done in quantum science. Atlas helps solve engineering challenges everywhere.
Particle colliders, space missions, quantum, cryogenics, and more. Custom vacuum chambers and bimetal flanges and fittings are built in Atlas' fully integrated facility with on-site design, development, and manufacturing capabilities. Learn more at atlasuhv.com.