 Gersh Bloomberg from Rutgers University and he's going to tell us about Istanbul, Nero, Nero and the Xitonik and Sveta. Okay, thank you for inviting me to this scenic place. So this research was done in collaboration, it's a collaboration between Rutgers group and Technion group and mainly carried on by my student, my year, who is now in Karsur and a postdoc by the work of who accepted faculty position in Yukon. All the results I show today are published in these three papers, so in this quick talk I will not going to give any details. Okay, doesn't work, still doesn't work. Okay, so I will skip the motivation here because in some sense it was given. In Nicolai's talk, we learned that all our troubles are coming from familiar statistics and if we deal with problems, all problems are done, so therefore let's go straight to to Pazonic excitations and in this case I will talk about excitonic excitations. So this subject has a very long history. People try to create excitonic condensate for a long time, the most successful attempts were in devices to couple to the electron gases in quantum well structures, but more recently, yeah of course the story goes back about 60 years where Korn, Keldish, Halterin, Reiss, etc. proposed that there is an instability and metal can turn metal with electron-hole interaction, can turn in an insulator. So this was perhaps first time realized in the Sorbana paper in Pietro Bamantes group, where in TISC2 the claim was made that they saw something that looks like charged density wave, but the claim is that it's excitonic insulator and the reason for that is that they did electron-load spectroscopy yield measurement at finite q and they saw softening a plasmon mode also with that softening of the phonon mode and the argument in the science paper is that in trivial Pyre's transition you wouldn't expect softening of a plasmon but but because they observed that it's an excitonic insulator. So I will go straight to discussion of this system and it's fairly old compound that was known for about 60-70 years. It was known that there is a structural phase transition just above room temperature but the interesting thing is that in Hida Takagi's group by doing corpus they noticed that the top band is not parabolic but has this flat top as function of temperature and that let Hida, a collaborator, make a claim that this is a good candidate for an excitonic insulator. Yes, I will get to that. Yes, yes, I'll get to that. Yes, so okay so the compound it's a 21D compound contains tantanomenical chains and well this is Hida's walk where they show that there is a second or the phase transition at just above room temperature and what happens in that structural phase transition there is an orthorhombic to monoclinic transition basically 90 degree angle becomes larger than 90 degree angle and those are the distances between the tantanum, tantanums are shown here at the browse. So this happens at 328 degrees. Now the order parameter that does that it's quadruple order parameter so it's b2g in b2h group so it's a gerade so this kind of probe is not accessible by optics, some other methods need to be used. Hida went further and he realized that doing substitution on selenium site by Isovile and Sulphur allows to change the condition of this material and this material the parent material you need to look for the dotted line it's a compensated semi-metal and by doing this substitution due to some chemical pressure a gap opens up it becomes small gap insulator so there's a transition from one to another and then conveniently if this is an excitonic insulator then we can probe different regimes bcs regime all the way to bc regime. Okay so the physics case trivial the idea that very old idea can put it forward I guess first that that if you have a semiconductor and there is excitonic state due to electron hole excitations with some binding energy b and if this binding energy is larger than their gap then this system is unstable and there is renormalization of the bands and the symmetry is going to be lower. Now in a regime of semi-metal this gap is negative and this is like in bcs the the system is always ready for the instability instability here the hybridization of the bands in high symmetry this hybridization is forbidden because they belong to different symmetries symmetry and transportation but when you break the symmetry it becomes allowed and then the gap opens up and this transition is basically the coherence feaks controlled by coherence factor essentially bcs like type story and what in that key this initial paper was observed is essentially is essentially this band so this this this was the idea there and while accompanying with structural change because as soon as we break the symmetry and this symmetry breaks the mirrors also there is a lattice distortions that come with it and then the question is what drives the transition related to what you're asking me so the transition could be driven by electronic part excitons condensation of excitons it could be by softening of optical phonon or it could be strain field uh basically ferroelastic transition yes they belong in high temperature phase they belong to different representation that's right yeah yeah yeah these points the crossing points are close to high symmetry point in in practice well let's discuss that i don't have time to go to that let's let's get the big picture so the big picture is now how do we how do we find out what drives the transition and the way to do it so that's the that's the new method let's put it if to look for critical dynamics so you basically measure the response function as function of frequency and see what happens with that response function so what you expect in second order phase transition is a soft mode and if this mode is of excitonic origin electronic origin then it's driven by by indeed excitons that would be the hallmark that would be the uh method to see that if it if it's softening of the phonons then it's phonon driven or it could be also acoustics basically basically the strain field um now the method to look for that because it's uh in in even representation the excitations is the Raman scattering and in high temperature phase these two bands these excitations belong to different representations and when the symmetry is broken mirror plates are gone so they become become in the same representation that's essentially opens up the mixing uh so main points um we can show that we show that there indeed there is a condensation of electronic uh so so the the virgin susceptibility comes from the electronic response and to do that uh this is now finally some data come quite complex plot this is energy in semi-lock this is the interband transition appearing here uh the sharp lines of phonons uh this is the electronic continuum that this is temperature on some non-linear scale just to emphasize the region around tc linear plot of the same data and you can basically see how the spectral rate uh moves to the low energy and then below the transition gap opens up so that's the blue sky with sharp sharp excitations yeah i'll show you i'll show you everything just one second so um i'm going to skip the technical plot i mean the details are here in paper but the data looks like these points and in order to get to that question that andrei correctly asked is that this need to be decomposed and we decompose it using using the final model leon it spoke about that okay very good uh in in two parts so there is this electronic part which is blue and then then there is a phononic part which has this green sharp apex and if we turn on interaction which we also can determine spectroscopically then we can compose this this red line so that that repeats the data right and this blue line is the one that you're asking for and that can be described by ua damp response function exotonic response function it's semi-metal so it's everything is in in ua damp regime and then if i calculate if i basically do the pits and find what the frequency of that peak does then above the transition i see that this mode gets goes up before that gets interacted at the transition so if i continue this line i get to some temperature which is higher lower than the real transition but i see that transition now phonons don't soften the phonons actually harder at the transition so these are phonons uh this mode softens above the transition this is tc this mode softens this goes down yeah okay but that's yes remember remember we break discrete symmetry but in principle quasi-goldstone yes this if there is but yeah but this is this is discrete symmetry break okay so we can extract susceptibility and well that's the uh lando siri here and now i plot inverse susceptibility that comes from that same data so one over chi would diverge uh well chi would diverge at this temperature and this is curie y sunday you say there's not enough points well that's that's what we do it's quite linear here uh and uh phonon by itself doesn't do that that's accessibility from the phononic part that does nothing clear now if i turn on the coupling between between this uh electronic and phononic response then analogously to what york did 15 years ago for pneutides you can see that you can enhance transition from this point to this point your susceptibility will look like this and if you include also the strain field then you can basically get the divergence at the right place okay so now um the point two is that we can actually directly probe this transition and that's directly seen in the spectra that will be the spectra of different channel uh so we we essentially see this coherent interbar transition peak and that gives us even even the magnitude of the of the gap which is well some large numbers set in ktc susceptibility is of course in the in the b2g channel if the channel where you expect the instability right and uh uh uh so yeah so that's that's what i just said and the last point is that if we dope the system and go away and drive the system to um drive the system to semi-metal excuse me from from semi-metal to to uh a semiconductor i'm going to skip the details here then the interesting thing is that we actually start seeing this exciton in the gap as a very sharp peak so this exciton is in b2g channel it's a dark exciton it's a quadrupole exciton which by itself is interesting you can't see it in optical conductivity it doesn't give you light back but it's a quadrupole quadrupole bound state uh okay so uh let me get to the phase diagram so this is now the doping uh from semi-metal to small gap semiconductor this would be there based on the all the data and analysis this would be the excitonic transition if you wouldn't if it wouldn't couple to lattice if you don't have lattice and so we would have the critical point here uh if we add optical phonons the line shifts up if you uh add just a constant strain field the line goes here and if we actually use some uh well account for the softening of the shear modulus then we recover the measured points of the phase transition so uh so take away message is that uh um the hidden phase transition the low temperature phase transition uh is lifted by interaction with the lattice but the only channel that sustainability diverges is the really excitonic channel so it's it's quite clear to call it excitonic insulator because no other responses do uh do diverge in the system so that's and that's where the details are and I'll I'll be happy to talk further with you right okay thank you Gersh we have time for one quick question or maybe two we already have quite a few questions can you start setting up uh no question do we have questions online okay so let's move on to the next speaker