 the program on the web page, just like with the lectures. Okay. It's yours. No, it's not mine. Anything up there? Hee hee hee hee hee. This doesn't look too good. We give it a try like this. Reduce it. This is the max it can do. Do it this way. So there should be something up there. Yeah, I can't somehow over there. I think something should go on there. Let's have a look. Great. So Ilya, please. Oh, I think we have to change this one. So here we are. So good evening, everyone. Hello. My name is Ilya Gerhard. I'm from Stuttgart. I'm from the big operation in Jörg Wachtrupp's group. So and I think I have to change gears a little bit. So I have to admit or I have to say to state that I'm an experimentalist. So I hope this doesn't bore you. I think this might be leading to some discussions what we can do. And today I'm talking about photons and particularly about single photons and a bunch of them. So if we think about single photons and how to generate them, a very common way where people feel comfortable with is to say, well, we just take a nonlinear crystal. We take some blue light. We fry the crystal with a lot of energy and with a very low probability there are two photons coming out which might be even entangled. So for sure these photons are correlated. But if you look careful, these are not really single photons but are very highly correlated photons. So in the moment I talk about single photons, I talk about single emitters which are ideally two level systems which you excite and then they can only emit a single photon one after the other. So single molecules were demonstrated for example in the 90s, in the 70s already there were ions and atoms. But if you talk about single kind of solid state systems single molecules were the first system which were implemented in the solid state. In our group we are very busy with single defect centers and diamonds. So these are also single photon emitters very popular are nowadays quantum dots and you maybe know this kind of turnstile single photon device and also what was found in our group was the detection of rare earth ions in a solid state matrix. And these are single photon source which give out photons one after the other but never two in the same time. So my like feel of expertise stems from chemistry and you have to realize if you kind of ask a chemist about a molecule he or she might say well this is something more complex than H2 plus or rubidium 2. So this is a molecule which is very known to organic chemists and this is something which might be known to you as a fluorescent marker for example. So you kind of shine in blue light and you get red light for example back. So this is a highly fluorescent molecule and we can single it out. So it forms something very close to kind of a simplified levels, a two level system here. We have a ground state. We have an excited state. Luckily we have some some vibrational state of the ground state. So if we pump it resonantly at cryogenic conditions there's a lot of kind of a relaxation going on into the manifold of the ground state. These are vibrational states and then it relaxes down to the ground state. So in the moment I have a single molecule I can shine in this light and the cycle goes on and on and on. So in these photons, the red shifted photons will be single photons one after the other. I can do something else. I can shine in a blue shifted manner. So I excite to the vibrational state of the excited state. It relaxes to this very narrow band transition. So this is as a 10 on a second lifetime and then it decays back to the ground state and emits exactly this typical light. So this is about 50-50. So you get 50% of the light going down exactly on this resonant transition and 50% of the light is going to live in this kind of red shifted levels. It's in a solid state matrix. So it's a low temperature matrix. Normally often some other organics. So in this case it's tetradicane. So very important. So unlike all the biologists we have a single molecule which is well behaved. It really behaves like an atom or an ion in a trap. So you can work with this. You can start in the beginning of the week and on Friday evening you warm up your cryostat and then everything is still there. There's no blinking or bleaching going on. We can modify this molecules. For example, if I cut it a little smaller it's going to tend kind of a blue shifted emission. So this is just the one we are just using. It will be extremely bright. For example, you have to realize although we have this 10 nanosecond lifetime here we detect only from this kind of orange arrow so to say. We detect more than 10 to the 6 clicks per second. So it's not corrected by any values. Everything is in there. Detection efficiency and so on. These, the line is from here is extremely narrow band. So we have a few megahertz liners like with an ion or with an atom. And it's a little bit detunable by the DC's dark effect. So although the molecule sits in a matrix and you have this inner neurogenius broadening that all the molecules in a matrix are a little bit different you can select one and shift it a little bit with some electric field. And the nice thing about this molecule is that this debents untimed train so this we call debat in the lab is resonant to atomic sodium. So we do now experiments where we combine the research on single molecules singled out in a solid state matrix with atomic vapors. So I don't know what this is. No, yeah, here like what all we have to remember for now is that we excited somehow kind of greenish and that we get some orange light out there which is then used in a follow up experiment. So let's do this. So we take a cryostat. So we take a kind of a so-called solid immersion lens to collect the light very efficiently and this is something you might know already. So we have these half spheres and they are made of cubic zirconia. Cubic zirconia is something which you might know as artificial diamond. So these are the gemstones Swarovski is selling. So it has a high index and majority of the light which is emitted by a molecule which is here in the focus is emitted towards our collection optics. And so I can be very efficient in detecting all these redshifted photons. I mentioned that these are resonant to sodium. So we have some interesting filter scheme here. So this is a so-called Faraday filter. So I take an optical polarizer which kind of is crossed out with another polarizer and the Faraday rotation in this atomic vapor is filtering only the resonant photons which are resonant to sodium from my molecule. So let's look here what's going on there. So here we have this is our kind of Faraday spectrum what we see by the atomic vapor and also we have a kind of single molecule kind of a very narrow emission line. As I mentioned so this is something like a few tens of megahertz. So this one the orange one is the filter function and this is a little bit tricky. I'm going to explain it in a second how this filter function comes about in this Faraday filter. So let's look if we really have single photons and it's very straightforward to measure that. So we declare this detector as a start. We declare this detector as a stop. So if this one clicks we start the clock and then if this one clicks we stop the clock. So it's clear that if the clock just started there won't be a second photon to stop the measurement. So this is a typical anti-bunching so we here go in the moment we fit it to negative values. So this is just the effect of shot noise. So we essentially do not do never get two photons at a time if we just detected one. So this shows the purity of our single photon state. And now we like to play some games with these photons. So and we have some kind of quantum gate here and I'm going to introduce this also in more detail. So a little bit to this atomic filter working with atomic sodium was very present I would say in the 70s and in the 80s last century but it's not very convenient I would say. So nowadays you like to work with rubidium or eventually with cesium which are light or kind of heavy atoms which are easily in the gas phase but for sodium you always need big large magnetic fields hot temperatures and it's a little tricky. So this is our fire filter and to explain how this works we kind of these are the parameters on this filter. I simply kind of introduce this little video here. So we have a polarizer here. We have a polarizer there and in the moment we have a B field and we change the color or the kind of the resonance frequency of our laser. For example, we get some fire derotation in the vapor. So you know if you have some absorption you intrinsically have some dispersion as well. So this is now the dispersion for example for our sodium vapor. In the moment I apply a magnetic field this is going to be shifted into two dispersion parts. One acts only at the sigma plus and the other one only at the sigma minus light. So now I can think about what does it mean if I supply a linear polarized light and then I get something which is the rotation. So at a certain point it rotates in one direction and at a certain point it rotates to the other direction exactly by pi half that it can pass the second polarizer. So this is the typical these are the side peaks so to say you see this is rotation to the other direction. This is the Zeeman effect. It's straightforward to calculate it for atomic vapors. You have to consider a lot of transitions. So if you want to play with these parameters for the atom of choice you can simply kind of go to this URL and play your kind of enter your magnetic field and vapor to calculate the spectrum which comes out. So now we want to do an experiment with single photons and so although we have a million photons per second we like to have it really stable. So I want to show this is our optical table when we were kind of preparing it so it's really heavy duty work to get this all together. So and now we have a kind of all optical gate there. So we supply kind of two single photon streams from here and here kind of integrate them on a beam splitter and then we analyze in a full tomography scheme what's coming out. So this is something which can stand there for hours and days and is not kind of moving by any means. So it's really heavy. We have single mode fiber couplers from A to B. The fiber through fiber coupling is really very well and we have a visibility of this gate is better than 99.9%. So this is really like we align it with the laser and we see really dim and bright laser light here in our fibers. So I think what we can do with this so how well are our photons and one option is to say well how well can they interfere here? So you have of course single photons can just interfere in such a mass-zine configuration. So what you will see when you move one of these mirrors you see well there's interference that if you monitor for example on this detector going up and down destructive and constructive interference and this will alternate with the other detector. So where is the light emitting our gate? So in reality something will eventually limit our coherence between these two sides. For example it might be jitter, it might be dark counts, it might be some technical or some other photons which are coming there which are broadband which cannot interfere. So we like to do an experiment where we characterize this visibility something we call coherence in a more sophisticated way. And the interesting point is now we introduce some delay line here such that the photons on this beam splitter here are essentially independent. So let's look at this configuration so we call this visibility coherence. You will see in a second what we can do with this. So the first experiment which comes to mind with this configuration is the so-called Hong-U-Mandel interference. So in principle you take two input photons on this beam splitter, you have two detectors and then simply you just write down all the options which can come out. Of course both can be transmitted, both can be reflected, one can be transmitted and the other reflected, this can be all the other way around so reflected and transmitted. So if you write down the mass you realize quickly, well this is our input state and interestingly these options they destructively interfere. So we have to realize in the moment I introduce this kind of reflection and reflection, I have a kind of eye phase shift such that this is going to be in total a zero. So I get out something which is called a noon state, noon because I have a number here and the zero, zero and a number, so N-O-O-N. So in this case it's a very trivial noon state, this two, zero, zero, two, this should come out. So let's have a look if this comes out. So this is now our input and output state. Here we supply two photons which are normally called as indistinguishable and this is our measurement. So we really in this CW experiment if this detector just clicked the other wound because essentially two photons were going just this into this direction. So in this, so and this is not a photon number resolving detector, it will just produce a single click. So what's going on exactly at this detector? So now we know that for this time here we should have essentially two photons. So one idea would be to say, well we do not look at the correlation of two output arms but we look what's the output here. So let's do that and so we change a little bit the configuration. So instead of having a detector here and a detector here, we look now on this two-photon state if you want to. So now we did a lot of calculations. We thought, hey what's going on there and in principle we should see like we are not theoreticians in this way that we could immediately predict what's coming out because if you think about two photons you have the feeling well it should go to 0.5 because it's one minus one or n. But if you think about two photons the other way you would say, well these are two photons and they should bunch. They should come at the same time. So is it a peak or a dip? I don't know what your intuition says. So our math in the end of the day was kind of coming to the end and we found out, okay this is to a good extent it's a dip. To the other extent it's a peak that we have this kind of two-photon coalescence and we found out okay it should give us a flat line. So if we just look at the Hamburger-Barland twist the single-photon characterization of one of these outward arms. And this interestingly is also what we measured. So we measure essentially a flat line for all delay times although we know that at 0.0 we have two photons and at point way later we just have a single photon to stop the measurement. So the interesting point is now it exactly cancels out that we have now the two-photon bunching so to say at time equals 0 and the 1 and 1 which comes at later stages. So this is something which is quite interesting. So we see this also at time equals 0 we can characterize it in terms of a Cauchy-Schwarz violation giving rather large values. So this is a clearly non-classical state. So now let's think about another experiment what is normally done when you talk about Hong-Gong Mandel interferences which is you supply kind of the input with two orthogonal states. This is well known and it's called the Shi-Ali configuration in the Hong-Gong Mandel case. Essentially you go in one arm like for example vertically and in the other arm you go horizontally. So in principle then you know you can write down the final state very interestingly that you have now all the four terms. This is kind of straightforward because you have all the four options and they don't cancel each other anymore. There is no destructive interference as for the Hong-Gong Mandel case going on. Interestingly also we have a five minus entangled state in here. So in the moment we detect something that's post-selected we can measure this entangled state. So this of course is straightforward. If you just supply these kind of the input state in an orthogonal way you get this 0.5 this one minus one over M state and in the moment we compare it for example to our Hong-Gong Mandel state we can realize that there is a visibility associated to this one of 93 percent which is pretty remarkable for a CW experiment where you just continuously generate photons. So normally these indistinguishability measurements are performed in pulsed excitation and then it's way, way easier to get very good values because you're not limited by dark counts and you're not limited by any jitter of your detector. So now let's think about another experiment here where we go in orthogonal but we do something where we analyze in another way. So as I said we go in for example vertically and horizontally into our central beam splitter here but now we supply kind of a polarizer at 45 degree angle. So let's say you put it in 45 degree you put it in minus 45 such that you can look what's going on. And of course we realize pretty quick that these two cases can cancel again. So now the photons although they are clearly distinguishable you can tell where they came from on the beam splitter but on the detector you can't anymore. So this is something which is known as a quantum eraser and you see that in two cases namely when we do it in a parallel fashion the analysis here with plus and plus 45 degree and minus 45 degree you again see the destructive interference of the Hong-Mandel dip. The interesting point is now if you look at the orthogonal state where plus 45 and minus 45 degrees are analyzed here and here we see that this phase which was kind of connected to lead to this destructive interference in the Hong-Mandel case now leads to a constructive interference. So this is now different to the state where we look at only one output arm in Hong-Mandel but we see that here we have a constructive interfering case. So it's exactly altering the phase in this Hong-Mandel measurement. So with this I'm already close to the end so I have shown you that a single molecule can act as a very clean and sodium-resonant single photon source. This atomic filtering allows us to suppress all other contributions, room light kind of other molecules, other dirt in the matrix which is very nice. I was showing you this kind of Hong-Mandel interference and we have very nice values for a CW experiment I believe maybe the best values which are out there and the self-correlation results in this exactly flat line which is an indication that the photons we generate are Fourier limited because the interference can only happen when they are kind of have this coherence among them. In the Chi-Ali interference when you go in orthogonally you see that this peak goes down to 0.5 and in the quantum eraser case we can revive the Hong-Mandel interference and also when we add all the kind of four acquired graphs, of course we get back this Chi-Ali 0.5 configuration. So let's have a look into the lab so I fear that this is barely visible so this is going to take a normal digital camera and take a picture into my cryostat so it's hard to see so but there this is essentially a single molecule you can just observe it by eye if you want so and if you now I don't know I haven't shown exact count rates but if you consider there are a million photons clicks per second generated in a single photon detector so this is very equivalent to if you're out at night and observe a very bright star so out there for example Vigar is a very bright star so the number of photons hitting your retina is about 1 million per second so this is about the same brightness next time you're in the night sky you can think about that that way that an ion or a molecule can be equivalently bright and you can see it just by your bare eye I have to thank my team so here this is Mohammed Reza who did all these experiments which I was just introducing Wilhelm was the master of the sodium filter kind of building this monster coil for the approximately a half Tesla field Kim was building up the gate I'm in the big big operation of Jörg Wachtrupp so it's a lot of fun so we are mostly doing NV center business quantum sensing but I'm a little bit responsible for the photonics part of all this I have to thank the Deutsche Forschungsgemeinschaft and so if you're very motivated I have a PhD position available and yeah thank you for your attention there's a workshop coming up on quantum networks next year in February in Bartonev I'm organizing and yeah thank you I hope this was kind of motivating and I hope this leads to a discussion thank you yeah I'm here