 We are live Hello, everyone. Welcome to this 110 Latin American webinar for physics It's a pleasure to have today Dan Carney who will talk to us about Testing quantum gravity in the lab. Dr. Daniel Carney He studied his PhD at the University of Texas at Austin and then after that he worked as a postdoc at the University of British Columbia and then at Fermilab and the University of Maryland together in a joint position and today He has actually the title that many of us always want. He's a scientist at Berkeley lab and He will tell us how to measure quantum gravity today. Welcome Dan Yeah, thanks Walter and everybody for the invitation to give the talk I'm very flattered and honored to do this. It's eight in the morning. So I'll be drinking coffee A little bit at the beginning. I hope you don't mind So I should share my screen again. Yes, please. All right so Yeah, so So as Walter said, I'll talk about testing quantum gravity in the lab and I think the title often, you know gets people very Emotionally excited because it sounds crazy. So the first thing I want to do is go through and just say what exactly I mean It's sort of a very high level So my my origin in physics is actually the same place as Walter We did our PhD so they got there in Texas in a string theory group And so I'm really a high-energy physicist in my soul so If you want to think about quantum gravity, you can think about it in terms of the energy density you're interested in so string theory or loop quantum gravity or Asymptotic safe gravity or there's other possibilities, you know, these are models of the quantum behavior of gravity at Extremely high energy densities. So so here I actually gave a number just to point out how how insane these densities are So this is 10 to the 123 electron volt per centimeter cubed That's that's the scale at which things happen that are like black holes randomly start forming because there's too much matter around and You know, we have a breakdown of all the equations other than the ones of say string theory You know, and this is a very nasty problem These are very hard things and it's one of the deepest questions in physics What is the correct model of gravity at these high energies? and So people have been working on that for I don't know 100 years if you want to think about it that way Certainly intensely working for 50 years In this talk, I want to focus on a very Low energy question so much much lower than these kinds of energy scales and this is a question sort of underlies All these high level question the high energy questions in some way So at low energies, you could even ask this question. Is gravity quantum in the first place? And so sorry minimum there we go So is gravity quantum at all so so all these high energy models like string theory They all assume kind of from the beginning the fact that The gravitational field itself should be treated Quantum mechanically, but what's the reason we think that that's The case I mean, we don't actually have some piece of direct experimental evidence for that It's sort of the simplest model if you have certain, you know prior ideas like quantum field theory describes everything But you'd like to be a little more, you know, experimentally hard-nosed and say, okay Why, you know, can I do an experiment that would tell me for sure that gravity is itself quantum mechanical? If the answer is yes, then it means that all of these questions that high energies are, you know Continue to be worth asking you might though be surprised You never know gravity is a strange force and people have certainly thought that maybe Even at low energies quantum gravity could be, you know, something quite different than what you might think in terms of say gravitons Proturbative general relativity So I list some some options people have talked about there are options where gravity emerges from some kind of thermodynamic limit of some microscopic thing We don't know There are even crazier ideas like Roger Penrose has suggested that Gravity will cause wave functions to collapse in quantum mechanical systems even at low energy And so so broadly speaking what I want to talk about is testing some ideas like this at energies that we can actually achieve In the lab and this is a really just a tabletop lab energy This is not the LHC which would be in another 10 or 20 orders of magnitude higher But still way below this far off violet string scale So this slide is like my personal history into this question. So is gravity quantum? So this is the question I'm really gonna focus on and I started thinking this way at the end of my my graduate school career because Freeman Dyson here wrote this great paper and the title is is a graviton detectable and this is such a nice question This so can I even know that the graviton exists in a very direct sense? Just find one. Okay, and in Dyson's answer to this question is no, you cannot detect a graviton So you have to be careful about what exactly it says what exactly he says is He asked the question. Can I build a gravitational wave detector? You know like LIGO, but scale it and make it so sensitive that it would actually click when a single graviton of the appropriate energy goes through it and Dyson answer to this question is no the argument is that if you take sort of a LIGO type detector And you sort of rescale how big is it how much laser power and so forth It's always gonna have so much power so much energy density in a certain radius that that device itself will collapse to a black hole Or it'll have to be the size of the universe or some, you know horrible scale But just me it's not a this is not an answer about in practice. This is really an in principle. The answer looks like no That said it does rely on the LIGO type detector for a gravitational wave, but okay So I actually got quite depressed when I read this paper I thought this means that I'll never know if gravity is quantum mechanical and I can't test it But a couple of years later, you know I sort of started reading a little bit more about quantum information theory and This is a picture of John John Stewart Bell here who really, you know, I think made a lot of this stuff concrete And I realized you can ask a different question than can I detect a graviton you can ask can the gravitational interaction do something uniquely quantum This is what a bell test asks about bells bells test bells inequalities But they ask is can I have say two spin half particles and in some state and some property of that state can only be described Quantum it cannot be two coins, you know, classically sitting around getting flipped. It has to be something uniquely Weirdly quantum mechanical. So we'd like to ask the question. Can you do something like that with gravity? And the claim in this talk is going to be that yes You can ask this. Yes, you can do it such an experiment and I'll and then you know, yes, it will tell you something interesting about these sort of more high-energy questions about gravitons, for example So with that sort of intro the rest of the talk is set up as follows So I'll talk just by starting about some experimental scales. So this will look very different I'm going to show some pictures of actual things in labs I just want to show you kind of what scales to think about and how this stuff could be possible at all Then I'll move into a quick discussion of the basic theoretical ideas sort of riffing on the introduction Then I'll talk about two Experimental implementations that have been proposed. So one that was a few years ago now and one that we recently put Up on the archive And then finally I'll go back and I'll sort of put a connect what I've shown you to be sort of more fundamental questions If you want about graviton strings and so forth So I want to emphasize since I believe the audience does not say all gravitational physicists and such that the the really The amazing thing that's happening in the world right now that that is driving Me to even think about this kind of thing is is this emergence of what we call macroscopic quantum coherence in laboratories So this picture on top is the Google sycamore quantum computer chip The thing on the bottom is the mirror at LIGO. Okay, this thing has I don't know 50 qubits on it It's very small. Although it's you know, maybe centimeter on the table The mirror LIGO is 40 kilograms and what's quite remarkable is that both of these objects behave quantum mechanically And you can really you really have to worry about their quantum mechanical properties when you use them So that's obvious probably in the quantum computer in LIGO This the statement is that you measure Motions of the mirror at a size actually about a hundredth the scale of the vacuum fluctuations of the mirror So you really talking about quantum behavior in a very large systems And that's the kind of thing we're gonna need to be able to do a test of gravity because obviously gravity only works at sort of large masses So to give a sense of how large and what kind of scales we're gonna be thinking of this is a simple sort of figure of merit cartoon So, so let me show this is two masses. Let's take them to be the same I I separate them at some distance and what I'm drawing on the left with these two balls is you know Consider the mass this left mass being either in this one position or this other position separated by some small distance and I can ask so if I if I just took this thing put it in the Schrodinger equation and ran time evolution I can ask what's the the phase of the wave function picked up between these two masses? More importantly what I can ask is what's the difference in the phase between them this mass being on the one position or the mass being on the other position This is an easy calculation. You just plug the Newton potential into e to the IHT and calculate it And what you'll find is that the phase difference, you know, it has a simple-looking formula It's got a G Newton. It's got the mass squared. It's got D squared So, you know, there's the Newton fourth and then there's some factors of positions and times to make the thing into a phase The phase is sort of order one with Reasonably sized objects. So I put in nanogram masses. I put in micron distances and I put a second for the time that you run this So you can obviously move these numbers a little bit But the basic statement here should be at these sort of, you know, tiny but macroscopic scales nanogram second stuff like that That's when these types of low-energy quantum gravity effects, if you want, will become appreciable This is the scale at which I can actually measure the relative Newton potential between these two masses There's a nice conversion factor. Not everybody knows. I like to emphasize it that there's a M squared here and it's in units of G Newton So in high-energy physics, we would call that the Planck mass and the Planck mass is about the order of a hundredth of a milligram So people often think the Planck mass is some very super high-energy scale But in a way of looking at it, it's actually a very terrestrial scale This is like the mass of a typical salt grain and that sort of underlies why this could be, you know, doable So some experimental systems where this would be relevant Okay, so I just show some pictures to show that there's many options But typically what these are are mechanical systems So this is a mirror hanging on a string that can swing just like a mirror and LIGO This is a drum head that can vibrate. It's on a chip. This is a picture of a little dielectric bead that's levitated by an optical field And all these devices and a number of other devices are basically designed to operate as very sensitive force detection sensors So this one I showed the numbers just again emphasize these scales So this is an actual device and what it's designed to do is measure milligram scale objects through gravity at sort of millimeter distances And numerically just to give a sense that the Newtonian force at those scales is about 10 to the negative 17 Newtons And these kinds of devices pretty routinely now achieve sensitivities to forces at the level of 10 to the negative 21 Newtons Per root Hertz means it takes one second to resolve that So you guys should resolve a force much smaller than that now in these high level optimal mechanical systems Okay, so now back to the theory So I want to explain exactly what we want to test So this is probably the hardest part of the talk if you haven't really thought this way before it's a bit novel Okay, so so what do we want to test so I keep my picture of Bell here to emphasize who you know how we're thinking we're thinking like a hardcore quantum person So we asked the question is gravity quantum so a more concrete set of questions of the can the gravitational field be put into quantum superposition to make Is there a wave function for the gravitational field can I do all the usual things put it in superposition and so forth This is actually the same question under standard assumptions about quantum mechanics but you can think of them as different more generally Can the gravitational field be used to entangle to quantum systems. So I like to think of the second one and that's a sort of very concretely testable question and that's sort of what we'll focus on So what this would mean sort of pictorially is shown here. So what I imagine is I have some source mass this big ball and schematically I'm going to draw these, these wells so this means the metric field the gravitational field source by this massive object. So this is space time bending here a little bit because this masses around. And I prepare this thing and some definite location and I prepare a probe and definite location. And now the kind of experiment you want to do is, you take this mass and you prepared into a Schrodinger's half state where it's not in one definite position but in the superposition of say left and right. And so, under normal assumptions about how gravity would work you would think that then the metric the gravitational field itself would also be superpose. So in this state if you're familiar with quantum, you know, ket notation, where the masses on the left and the metric, the gravitational field is is on the left, plus the masses on the right and the gravitational field is on the right. So in some sense we're testing whether this is true, does this actually happen. But you actually have to think a little harder, which is how would I even know that this happened. Okay. Then I have to ask the question, how does this little test mass respond to the superpose gravitational field. And that's what's in this third cartoon. So if gravity behaves quantum mechanically. And I'll be a little more concrete later. What you would expect is actually that this test mess will become entangled, you know, with the with the source mass. On the left, the test mass will go to the left, if the source mass is on the right, the test mass will go to the right. And so the total evolution of the wave function of the universe in this little picture is that the, you know, the two masses branch, there's a branch where everything's on the left and branch everything's on the right. And this is a canonical entangled state in quantum mechanics. So what we're asking is the right description, that's one way of thinking about it. So why would I care about this. Okay. So just to give a sense of what this can test. So the answer to these questions is, if the graviton exist, the answer is yes. And this is maybe not the first thing you think but it is simple to work out. Okay. So this is a finding diagram where I have matter, and the matter is exchanging little gravitons. So there's a little space time ripples flying around. And the energies that looks like exactly like the thing I just showed you it looks like a Newton potential that talks that puts some interaction between the two masses in the picture. This this can generate entanglement and so you can go forth and look for a bell violation in an experiment with two masses. So just to give a simple counter example people have have long used equations like this so this is the metric this is the Einstein equation, where on the left you have space and curvature but on the right you'd have the quantum stress energy tensor of all the matter. But you put these little brackets which means you take the expectation value so now on the right side, this is a classical object this is just the average energy density in the universe. So it's easy to work out that if, if this is actually the equation for the universe, then, because this is an averaging operator, you can't actually see details like the superposition of the mass. So in fact in this, in this model you will not be able to entangle two masses through gravity. So this gives you a set up for the experiment, you know there's two clear options, if you want, and you will rule out one or the other. Okay so now I want to get a little bit more into how the experiments actually work with this sort of overall picture I'll talk a little more about these details models at the end. But how would you actually think about doing an experiment like this. The first thing you have to know is a little quantum information theory which is, okay I'm testing our two things entangled so how do you actually do that. So this is really where Bell comes in, so Bell really invented the way of doing this. You know building off of Einstein and Rosen and Podolsky but I would say they'll really hammer this down for everybody. So the classic way of thinking about this is the following so I'm given a state row. So this is the density matrix this is a little general way of writing it, you can think of row is just the wave function. And the question I want to ask is, is the wave function of the two masses, so the two masses are a and B is the wave function a product, meaning all the information in a, all the information in B has only quantum, sorry has only classical correlation that's what that would really mean. A non example of that would be like the singlet state of two spins that you learn about in quantum mechanics. So what Bell basically showed was that if you, for example take two qubits. There's some quantity w, which in this example is some product of the spin of the one guy in the spin of the other guy if you're doing two qubits. There's some operator whose expectation value will be will be positive. If the state is not entangled and it will be negative if you're in a specific entangled state. There's a lot to swallow. What, what this basically says is, if two systems are entangled, then there's going to be some correlation. So this expectation value this is a correlation function of these two operators. You might hear some scratching the back because my cat is randomly running around with the spring everywhere. What's going on here is in the, excuse me. So, if the state is classical it's not entangled, there will be a certain level of correlation allowed by classical mechanics. But in quantum mechanics you can have correlations between two objects which are actually stronger than anything allowed by classical mechanics, and Bell makes that very precise by local hidden variable models. And this is the essential statement. So, so when doing a belt test what you want to do is try to prepare your entangled state, measure some operator like this, and check if the correlations violate one of these classical correlation inequalities. That's basically exactly what we're going to do. So how would you do this in gravity. So this is a beautiful idea. So this, this proposal really was done by a set of bosses group in London. In, in, I guess, 2017 and time flies. There's some earlier work that this is closely related to by these people. This really goes back to page and gelker in the 80s. But these guys gave a nice like pretty concrete proposal. So, so what do they suggest so they say okay what do we use for the masses to make them little diamonds, and these diamonds have a nitrogen vacancy center so these exist, these are very useful for many things. But for the purpose of this talk what it is is it's just a piece of mass, and it's going to be pretty big like nanogram. And then mass has little spin half in it, and it's been half couples to the motion of the diamond. So, you can move the, you can move the center of mass of the diamond around by hitting the spin. So basically what that allows you to do is they make a superposition of the diamond by just shooting this thing through a Stern-Gerlach experiment because the magnetic field in the Stern-Gerlach will couple to the spin and therefore couple to the mass. And so what they propose is take a pair of these diamonds and take each one of these diamonds and shoot each one through a Stern-Gerlach apparatus. And then I have a diamond here superposed in two places. I have a diamond here superposed in two places. I let this thing free fall to my lab to get rid of noise. Okay, as it's falling, these things talk to each other gravitationally. And what you can work out is that the wave function of the two diamonds becomes entangled. So, mathematically it looks like I start them both in some initial state I call it left left here. And then as they go through, you become some entangled state where you have super positions of the, they're both on the left ones on the left ones on the right and so forth. And you can see that sort of the amount of entanglement is captured by some phase that shows up and this phase is given by exactly the same scaling I showed earlier. So it can be sort of order one meaning visible, you know, with nanograms micron seconds. Then what happens is, at the end, you have this entangled state of the two diamonds. And what you can do is now test this entanglement through literally a belt test, the belt test in the typical sense with two spins because you have spins in each diamond. So this is really a beautiful implementation. I mean, it's a very conceptually simple, in some sense, probably the simplest conceptually way of doing this. That's fantastic. Okay, the problem is the numbers you need to actually do this. Okay, so to actually do this. There's a number of difficulties one is that you need to measure both systems so a belt test even with spins was originally challenging. You have to measure two systems at exactly the right time and you have to do it fast enough, you know there's no faster than light signaling and so forth. In this system, the entanglement is extremely weak and the system is noisy so you're really trying to extract a tiny weak violation of the bell inequality, you know that is sitting around in some very noisy system. That's okay. What really makes this hard actually is more mundane experimental issues. One is that I need a huge magnetic field gradient and that's because I need a magnetic field gradient to operate the stern Gerlach to split the wave function. Maybe even, and by very large I mean you know a few orders of magnitude bigger than anything that exists currently. Even harder probably is you need a pressure, you need environmental isolation which is exceedingly good so a few orders of magnitude lower pressure than say interstellar space. So, I don't mean to say that you can't achieve these. People are for sure working on these things for many reasons, but this experiment is some years away from being done at best in because of these issues. So, it's a really nice concept and maybe it will be the way we learn that gravity is quantum mechanical. But I think at this point it was worth thinking through maybe there's there's other ways of doing this. So I want to briefly mention a way that we propose this is really the only slide on my own work here. This is an idea that Jake Taylor Holger Mueller who's an experimentalist at Berkeley jakes and theorist at Maryland. This is something we had, which we think we basically eliminate all of these technical difficulties. So it's got a fancy name we call the interactive information sensing the experiment in some ways is simple and the interpretation is hard. So the experiment, let me just describe what we would actually do. So the experiment proceeds like this cartoon. So what I do is I take a massive pendulum so this can be like those sort of milligram scale hurts frequency pendulums I showed you. I just hang this thing there and I take an atom and I prepare the atom in an atom interferometer so an atom interferometer is something where there's a laser field, and I can put the atom in different wells of the laser field. So in particular what I can do is, I can take two wells of the laser, and I can put the atom in a superposition of two positions. So this is just the shreddingers get this is exactly the same as the superposition of the diamonds. It's just now with a very tiny system and Adam, since with a very tiny system I can actually imagine this Adam sitting there for a long time and in Holger lab in Berkeley that he keeps them there for on the order of a few seconds. In this nice quantum superposition. Okay, so now what do you do so you so you take the pendulum you put it there say in the ground state, you put the atom, you prepare it in a superposition of two locations. And now what happens, well, well, if the Adam is on the right side, the interaction gravitationally between the pendulum and the Adam will be, you know to want to move the, the pendulum to the right, and if the Adam is on the left the pendulum to the left. And so that's what's shown in this picture. So there will be some extremely tiny, but finite motion of this pendulum, and the pendulum will swing just very slightly one direction or the other. And now what this really does is it displaces the equilibrium of the pendulum and so what happens is now the pendulum will sort of swing with the period set by the pendulum. And the, and the idea is to wait for different amounts of time and measure the state of the atom. So why what's happening is, as this evolution proceeds here in this middle picture I have an entangled state of the Adam and the pendulum right the information is shared and the whole wave function of the universe is in two branches pendulum goes to the right Adam on the right. The pendulum goes to the left and Adam on the left. Well these are entangled, if I look at only the state of the atoms, what I'll see then is a collapse of the wave function of the atoms, this is called decoherence. It's very well understood phenomenon by now, there's many experiments sort of like this just not with gravity, where you see exactly this, I generate some entanglement between two objects, I measure one object and I see that it's a wave function has decayed. Concretely, you would see a loss of contrast, if you do an interferometry experiment with this. But for the purpose of this talk the main point is you only have to measure the atom. You don't have to do some horribly difficult measurement on both things at the same time. The fact that you're only measuring the atom, and the fact that everything here is periodic in time with a slow frequency is what really basically enables this to be hopefully done. Okay, it really will eliminate a number of technical problems so for one thing. Okay, so let me. Good. So, so what's the real statement here is this really like a kind of a bell test. Basically the answer is yes, although you have to make a few caveats so I wouldn't try to read this theorem I just show this to say that we, we made it come precise. What I mean is that if you see in this atom, the collapse and revival and collapse and revival of its wave function, and you know that it's happening because of the pendulum in a precise sense. Then you can show mathematically that this is only possible if, if there's entanglement being generated and then destroyed and then generated and destroyed. And this has has to do with the monotonicity arguments and so forth but the output though is that what this means is that you can you can sort of do a bell test in the system. So instead of measuring two systems at one time I measure one system at multiple times. And I have to make some then assumptions about the interaction, but very weak assumptions. So essentially this is a bell test, it does essentially the same thing as the previous experiment. But it is substantially more robust and noise and the reason it's more robust and noise is because super positions of large systems quantum states of large systems die, they, they're killed by environmental noise is very rapidly. Here we make a very tiny superposition this mess it barely moves so actually this thing can barely be destroyed by the environment. So what we really need is a long Adam time along Adam coherence but this already exists. So, so we were sort of using a different technique and I'll just show. I'll flash this slide I think you should probably read the paper if you want to see sort of really the experimental details about why we think this could work. I'll just say the punchline is this looks like you should be able to do it. So what you need is an atom interferometer with, say 10 to the eight atoms so I draw with one atom but really you have a little cloud of, you know, nearly a billion atoms. But these exist all over the place now. You need the atoms to be around for about a minute. So this is maybe an order of magnitude longer than state of the art but nothing crazy to propose, you know, building. Yeah, these, these numbers all are just to say, at room temperature so you don't really need a lot of isolation, you can in on paper do the experiment so so Hogar's lab at Berkeley is starting to seriously think through doing this. But we're quite excited that maybe you could just experiment in, you know, the reasonably near future. Okay, so just to finish, I think I heard 40 to 45 minutes so I'll just last five minutes or so. My own natural state which is theory. You know, suppose we did this these experiments so maybe we do ours with the Adam and the oscillator. Maybe we do the bows at all with diamonds. There's been a number of other proposals that are related. Somebody's going to do one of these sooner or later. So successful. We can already see we can show quite robustly theoretically that what this demonstrates is that gravity is causing entanglement, gravity can generate entanglement. That's the concrete experimental state. So the question now is, okay, who cares, you know, suppose you knew that was true. What did we learn about gravity I mean okay and angle things very cool. Great. So did I learn anything about the gravitons I learned anything about string that I did I know of Penrose's thing about the wave function collapsing is true or not. And that's that's sort of I want to sketch what is known the partial answers to these questions in the last couple minutes of the talk. So I mentioned this before and I like to start with this. The first thing to say is that you can imagine ruling out some funny models. I talked about the semi classical gravity model, and let me just say it again. So many classical gravity would be the metric response to the quantum expectation value the average of the of the stress energy of the matter. And I made the argument before, which is that this this has no quantum information that this has deleted all the quantum information that's in fact why you take the average you don't want to deal with the if the universe really operated under this equation. You would not be able to see entanglement for gravity. So you would do the experiment. And what you would find is that the masses never get entangled. Now it's hard to say for sure that you've done that because if you never see them get entangled probably your experiment just wasn't quite good enough yet. Okay, but so you have to really convince yourself that was the answer. If you did see the entanglement. This model is just out you can. It doesn't make any sense. This model has theoretical problems also but you know, that's a whole other talk. There's a kind of more modern and I think interesting thing and this is an open question so I would encourage people young people, especially now that I'm getting old to think about this. Gravity is an emergent phenomenon. So this can actually mean a lot of things. And so let me give two sort of pictures here. So the first one is in string theory. There's the ADSDFT correspondence. And, you know, there's a whole lot of story about that but the short story is, okay, what if gravity emerges in that way gravity is some one wavelength version of string theory. In that model we know by construction that the answer is it goes as follows at high at high energies you have strings at lower energies you have gravitons and therefore by everything I said already, you'll see the entanglement the gravitons will just mediated Newton interaction and so forth. So, so if you do the experiment it doesn't mean string theory is true but if you do the experiment, and you don't see the entanglement I guess you would rule out string theory that would be one way of saying it. If you did see the entanglement it doesn't mean that you can go back up the chain though. So this is one way in gravity would emerge and basically give you the entanglement, you can imagine though that if gravity emerges if it's the long wavelength limit of a hydrodynamic model with lots of little degrees of freedom. So, if you took a typical model like that, you would actually expect that there would be no entanglement you have lots of little particles or whatever makes up the gravitational field. You know, hitting these objects. And so if you're trying to entangle them that would just destroy the entanglement that would be like gas, you picture like gas hitting your system gas destroys all the quantum information. So, you know, this is a very heuristic that there's no favor that really goes into this, you'd need a microscopic model, but I do like to suggest that, you know, maybe these entanglement probes are a way of sort of thinking about distinguishing ways that gravity could emerge, by by characterizing whether or not they can produce entanglement. But really, in my heart what I want to know about this gravitons you know my prior is the graviton exists and what I'm doing here is I'm proving that it exists that would be what I'd really like to claim. But that's a high stake claim and you have to be very careful if you're going to say such a thing. So let me show you to finish just what is the statement about graviton model in this experiment. So what is for sure true, and I've said it a million times now is that if the graviton exists, it will generate the entanglement. And it's easy to picture how it happens you have the two masses when you superpose one graviton fly around. Everything is quantum mechanical and beautiful works just like quantum I like for dynamics. And I get the entanglement. So you might think then if I saw the entanglement that this would mean that the graviton exists but this is wrong right that the implication is only from the left to the right not backwards. So you have to actually ask this question. If I see the entanglement do I necessarily know it's a graviton. And really what you want to say is under what assumptions would I know for sure that it's the graviton. Okay, so there's some partial results but this is an open question in my in my opinion. Here are some partial results. So, so if you see the entanglement in the experiment, there's this beautiful paper by Alessio Balencia and a bunch of people Bob walled included, which shows the following thing. Suppose you can do the experiment. What you know for sure then is that you can generate entanglement with the Newton potential the one over our law. So that's an a non local force. And so that would mean that you could actually potentially signal faster than the speed of light. So that would be a problem. So how do you resolve the problem. Well, you have to assume something beyond just entanglement through the Newton interaction. So in the paper that if you assume that there's quantized metric fluctuations roughly speaking, little graviton, you'll resolve all of these fascinating life problems. And indeed this is the way you don't sort of typically you think about resolving fascinating problems you just put quantum fluctuations. There's another related paper which is a similar statement this is much more recent now where they basically show that if you see the entanglement and you you know that the force cannot be mediated by a classical gravitational wave. So you would say rule out some weird model where there's just classical gravitational waves somehow doing all these interactions for you. But none of these are really nail on the coffin, you know you do the experiment, you see the gravity and you can imagine finagling with, you know, messing with these assumptions, you can imagine a model that would get around these kinds of the conjecture. And in fact this can. Okay, so the conjecture really kind of see me you want to make it in any field theory, where you get the Newton interaction at low energies. That that field theory definitely has a massless spin to state, meaning it has to grab a thumb. So you'd really like to show something like this and this is an open question. So certainly I work on this from time to time and I'm sure other people have thought about it it's not a simple problem. The answer is probably wrong. You know for the sort of connoisseurs in the audience that there's an there's an interaction you can imagine which would mimic all of this graviton stuff, which is a scalar graviton you can imagine a scalar field which couples to the trace of the stress tensor. And I think that would actually probably do everything else that's in this paper but that would reproduce the entanglement generation in the experiment. That's an interesting question, you know, do we know that the graviton exists if we get the entanglement. I mean, morally probably yes, but you know, formally, it's really not proven, I would say at this point. Nevertheless, this is sort of why I'm doing this. So, you know, can we, you know, it, you'd like to say that we've sort of come around have answered Dyson's question so can you detect the graviton. No, can you do an experiment where it would tell you that the graviton exists. Yes, it's kind of an interesting philosophical position. If you wind up in in that, in the end, you know, well we know that the graviton exists but we can't directly detect it. So does it exist I don't know I mean that's one for the philosophers. So good so I'm out of time and actually did this on time so so let me just conclude. So, again, if you know you zoned out through a lot of that the statement is in some way simple. You know, if gravity is quantum mechanical in the sense that it's a quantum field theory. If you want to be really precise and the statement is that space time perturbation metric perturbations are truly a quantum degree of freedom just like the electromagnetic field. You can generate entanglement between two masses to the gravitational interaction. And then the really the more bold claim in the talk is that you can actually test this question you can you can get a binary answer is this true or not. And you can do a tabletop experiments and the reasonable near future. And so I'll leave some open questions. The question is this one I just went through if we see if we did the experiment, we see the entanglement, we're sure it's from gravity by ruling out all the other possibilities and so forth. Do we know that graviton exists. And, you know, just again to hammer in that this is really an open question you know, for sort of more high energy people, you can think about gravity and two plus one dimensions as an interesting counter example, potentially right gravity and two plus one dimensions has no graviton and you can definitely entangle masses. So, you know, think about it. The question is, is there a smarter experiment to do. You know, I nobody's claiming that we have the optimum things at this point. And there's still time to make something before anybody actually achieves one of these things in a lab. So I put this reference here we wrote a long review paper a couple years ago, which goes into a lot more detail about the models, the various types of gravity that you could test and so forth. And I would recommend you to check that out if you're really interested in this. In the meantime, thanks very much for having me and thanks for listening to the stock. Thank you very much. Dan, now we're going to spend some minutes with questions. While we check if there are any questions on YouTube I do have a question that Roberto who had to leave early has for you. He asks, can the quantum behavior of gravity be nonlinear and in that case, can that feature make it harder to use the standard model of the standard quantum mechanics to build experiments. Yeah, this is a very interesting question. This is a very nice question. Yeah, thanks. So I think I even put a backup slide for this. The answer is baby. And yeah, this is interesting. So let me first actually show a, so there's these semi classical models right so this model actually is not well defined by just this equation. You might wonder, okay, well, this tells me how the metric evolves but how do I know what the matter does the matter there's a state of the matter and the state of the matter has to evolve under some equation and like okay what happened. So if you put this thing, if you assume the Schrodinger equation, this, this is a horribly broken system it has super luminal signaling and all kinds of stuff. So in fact the only way to make sense of this, and this is a recent sort of idea is you have to make quantum mechanics nonlinear. The quantum mechanical evolution has to be nonlinearized. So this is a slide and maybe I don't have time to go through everything on here, but this is a concrete proposal to do that. So now what you do is you say quantum mechanics, in addition to the Schrodinger equation. That's this first line has these extra terms. And what these extra terms mean these are nonlinear terms what they mean is basically that you think of gravity in this model is measuring the matter, doing a computation which tells, which is going to tell you gravity, the computer that is gravity, you know how to move objects around. And so these terms are basically a noise. So these are an inevitable source of noise in a model where gravity would be classical, coupled to quantum matter. So you can show that, you know, this makes sense theoretically it won't have super luminal signaling it won't have any, you know, obvious terrible problem. However, then it makes the test different as, as the question was, was hinting at. Now you have a nonlinear quantum mechanics, and we know things happen in nonlinear quantum mechanics. For example, these noise terms will cause heating in systems that you wouldn't expect. It'll cause some irreducible heating rate of anything. And so you can start asking like okay, you know, we have very sensitive systems they're cold for a long time, can we rule out this heating and therefore rule out these kinds of models and you know this is a whole thing and lots of people are writing papers about this right now actually so it's quite an interesting possibility I would say. Thank you. Any question from the audience here. I can, I can, I have to ill pose questions. Thank you, Dan, for that nice talk. The first one is like when you say this type of experiments might be performed in the near future. Could you please define what you mean by near future. Yeah, I mean it depends basically how optimistic I'm feeling on a given day. I would say that, you know, this experiment is sort of original proposal. One of the people in this paper is Andy Karachi, who would probably be actually doing a lot of the experiment. And he told me, you know, if it's done in his lifetime, he'll be quite happy. So I think they're thinking you know 10 plus years. I need these low backing pressures and stuff. So now, to put my optimist hat and, you know, obviously this is biased. This thing, you know, we think you could do it. Now, few years, you know, it depends on a lot of details that we don't understand about, you know, horrible things like the noise from the surface fluctuations and you know heating and Kessner forces and stuff. But at least order of magnitude level this looks like something we could do few years, let's say. Thank you. And then also regarding this slide. So sorry, I got lost a little bit but but but I was thinking like if you prepare this entangled state. I mean, when you measure you collapse it so as I was thinking like and actually this is, well, you already at the beginning. It's just, it's a microscopeical, let's say experiment but but it's still big in that sense. So my question is, could you just repeat again how this will work because I think like as soon as the pendulum starts swinging, then you just lost all the quantum mechanical behavior I would say, or I would think because the wave function will indeed collapse and then you just measure and then I don't know. Could you just go over this again. Sure yeah yeah this is a little subtle of an experiment so good so you prepare the atom in the superposition the pendulum starts moving. And I drew it this way because this is exaggerated but the pendulum really barely moves in one way or the other. So if you think about it as wave functions, the pendulum's wave function, you know, just barely doesn't overlap. Okay, okay. And so then this causes, if you now measure just the atom that will cause a tiny amount of decoherence in the I see so it doesn't collapse like okay. So that that yeah so I didn't explain this picture but that's what this graph means so this is saying sort of the atoms coherence, if you want. So that one in some units and then it goes to some small value and back up and some and this value is set by the Newton constant so it's a tiny you know this is a highly exaggerated graph. Okay. Oh, thank you. Sure. I have a question that is probably more philosophical so let's suppose that suppose that instead of having a gravity tone right we did the real theory of quantum gravity is really just that space and time are discretized and then we go into the realms of theories like loop quantum gravity or those things. So you wouldn't measure anything right. Yeah, it's a hard question I don't know if anybody has precisely stated anything about this. So, you know it would you'd have to actually calculate it, it would depend on the detailed microscopics and in a way I mean it's not the same model but you know it sort of gets into this statement. I can say that Carlo Rovelli wrote a couple papers and is very interested in these experiments, and he wrote a paper saying that you know his expectation is you would see the entanglement and loop gravity would predict that you would see it and so forth. So according to him at least I think that's the statement how exactly that works I'm not I'm not sure. If you imagine that that the space time is just sort of broken up in a little discrete chunks though. I mean, to me, that would suggest sort of might have some anomalous eating it seems very hard to view that as like a coherent quantum system but you know, it should be studied more detail probably. All right, let me check. Do we have, I have, I have one one question regarding your, your last discussion with Alejandro maybe you could go back to that slide. Right right so so so this is this, I find it very, very interesting I mean the conversation you just had with with Alejandro because initially when I saw this I thought it was some kind of like quantum. experiment right initially I thought that you would be measuring a like a displacement of the pendulum but the whole point is that you're not. Right, yeah, the whole point is that you, I mean I don't know if it's possible to measure that, but that you're not measuring the displays that the oscillation of the pendulum. So the main effect is what gives you this a. This the coherence and recoge right this kind of interaction with them. Okay, great. So, so, so then in that sense, what are your main sources of background when you're doing this experiment. So, this is a difficult question that, okay we're still thinking through and probably will require the experiment to be built to really know the answer. I can say things that we worried about and why we don't think they matter. So it's sort of obvious things if you, you know, in the lab you imagine this is a real piece of metal and an atom and lasers. Okay, so what what are the main problems you know you would have gas particles hitting everything. That's obviously a problem. I think more importantly than the gas is you would have black body photons everywhere. And both of those things you would normally worry would be a problem. So if if the pendulum was super pose say in a large superposition, that will be a huge problem that any time a gas particle hits it or a black body photon gets absorbed or something that will just destroy the experiment. Like this is kind of like the under the hood. The amazing thing about this is, you know, this the superposition in here that the distance is like 10 to the negative 18 meters or something. It's a very tiny position. And it's much smaller than the wavelength of any of that stuff. So, if a black body photon hits it the black body photon can't actually tell where the mass is if you want it can't cause any decoherence. And so in that sense, those noise sources don't actually in principle affect this. I think the more realistic problems are more annoying you know they're things like, I can't put the atoms too close before there are interactions between the surface of the pendulum, you know, which is some metal say, and the atom, things like that. So that you know really set some design parameters like you have to put the atoms at 100 micron at least and things like that. Right because you have the cool of interaction at the end. You know exactly. So, right so you know luckily atoms are remarkably neutral, it turns out. You have Van der Waals interactions, you know you have to really so these are pretty yeah there are things like that that depend on sort of the details of the realization. Because I was thinking about the previous experiment the one day like the Stern Gerlach. Yeah, and and there you can more or less like control the backgrounds in a way because you can send one diamond only. And then the other right so then then I can imagine okay you can reduce some background there right, but on the other in the other one I can, I wouldn't know how to remove the background. So one analogy with that is what you could do is you know you can just operate the atom without the pendulum and vice versa to get rid of some things. I but I do agree yeah in some way this is the background here are going to be harder. The thing though with this is, yeah you can you can understand the background, however, the signal itself will be destroyed in the experiment that this is what I mean by you need very low and this is related to this comment about black body photon. You know, as this proceeds, you're talking about nanogram objects separated at 100 microns or something. And that that is an extremely unstable quantum mechanical system. I mean a gas particle hits that and that's what position is gone and now you have a classical ensemble and you'll see nothing. So, so yeah, I just just to say. It's a really fundamental challenge in a setup like this. Right. No, thank you very much. Sure. I have another question that so when you are getting to this so sensitive experiment so you can measure those type of distances. Once you end up measuring maybe some other new physics or I don't know. Yeah, what a particle hitting your system and. Yeah, that's massive. That's right so actually this this system in Holger Mueller's lab is originally designed as a measurement of fifth forces you know new light degrees of freedom the couple amass to to the atom. So, so we're really piggybacking technologically off of that and this would just represent maybe an order of magnitude better sensitivity in some way. So you can definitely imagine using this to look for new things. So fifth forces are typical one you know we have done lots of work on looking for dark matter with this kind of stuff. Yeah. You know what this signal is kind of unique. This is a very, you know, human engineered signal if you want you really do a lot of controlling quantum mechanically and so it's hard to imagine something randomly doing this, which is an advantage. But the systems in general. Yeah, indeed they're so sensitive. Looking for new particles is obviously something I do in my spare time now with these systems. Definitely a lot of things to be looked at potentially could be seen. Excellent. Can I ask a question. Yes. Well, first of all, I apologize for coming a little bit later was lecturing and so I got, I would say least than a half of your talk done but still I find it fascinating thank you very much. So first a comment and then maybe a question regarding this last point of finding new physics. I can imagine a Holger also made maybe he thought about doing some like equivalence principles, principle check, because maybe you can you can arrange your atom, you have there in the probe and and try to build on the charge conjugate version of that and see if it couples in a different way gravitationally to the pendulum. I don't know. Maybe they've thought about that idea. But that's a crazy thinking that was my comment. And the question was regarding this experiment with the pendulum. All you care about the pendulum is its mass, right, you don't care very much about its constitution or what kind of muscle is that right or not. Yeah, that's exactly right. You need that it's periodic system, other than that, it doesn't matter. Okay, because I was wondering because one of the problems regarding the noise I think you mentioned it briefly was not only the cool of interaction I guess that those scales which are very tiny customer interaction might spoil and be very important there so maybe it could be a good idea trying to explore some kind of a pendulum that could somehow shield you from the customer interaction. Would that be a good idea. Do you think it would. Yeah, definitely. So I'd be curious to know if you have in mind a way of making the pendulum specifically to do that. But what people have done and what we assume we will do is put a shield. I wish I had a slide for this but there's a nice experiment in Vienna, last year which does exactly this so what you would do is in this picture, you literally put a superconducting sheet between these things and you can use that to screen the chasm your force to some level. You know the other thing you would do is of course you just put the atoms far enough away that you know the chasm your force goes down like one over our to the sixth or seven. And grabby's one over our for okay well. Okay, so but you know it was some much sharper scaling than the gravitational force so so there's some distance and okay in the dumb way of thinking in my way of thinking, you just put the atom far away enough you don't care. But yeah, in practice, you would shield it. I do wonder if you have some idea for what you can do to the pendulum itself. I think I might, but I would need to to read in more details I don't know if in your papers, you have information about this. I will look at it but I do have some idea but but still is it's, I don't know if it's feasible to construct a pendulum like this it's it's regarding to these new states of matter. There are some logical insulators in particular there are some proposals in which you could tune the customer force between a different materials actually either making it positive or attracting. So, that's a good interesting. I did not know that yeah I'd be happy to talk about that actually few. I will look into your papers and write send you an email about it. That's it. That those were my questions and comments. Thank you very much. Thanks. All right, so thank you everyone over YouTube for watching this seminar and we encourage you to subscribe to all of our social networks and follow us to be up to date for the next seminars and we once again thank Daniel for this amazing seminar, and we'll see you all soon. Thanks again.