 OK, please take your seats so we can start. Very good. OK, so welcome to everybody. This is a great occasion. It's great to see also the room totally full, which is wonderful. And this is one of my favorite events. We're celebrating this year's the ICTP Prize and which was established in 1982 for young scientists from developing countries. Young means under the age of 40. So it has been given every year. And for me it's particularly special because I was on the other side many, many, many years ago. So this year we have two great scientists who were given the prize. And we have them here with us. This is Ruiz Fua and Hon Jun-shan. Ruiz Fua is from the University of Chile in Santiago, even though he is from Argentina. And Hon Jun-shan from the Fudan University in Shanghai in China. So I would say more about them later on. But the prize is always every year is given in honor of a big figure in science, in physics in particular. And this year the award is given in honor of Professor Wan, Kun Wan, who passed away 14 years ago. But people consider him the father of condensed matter physics in China. And so it's good to honor also him by mentioning him as part of this prize. And to recollect a little bit of memories about him and his relationship to STP, we have us, Erio Tosati here who knows, knew him very well and who knows all the details about how condensed matter has been developing in China to say a few words. So please, Erio. OK, my name is Erio Tosati. I apologize for cold. So the ICTP decided to dedicate this year's ICTP prize to the memory and figure in the work of Professor Wan Kun. He was actually born 100 years ago. And I understand this is the year in which there will be celebrations in China about him and his contribution. He started off in Beijing, his home place, and then moved on to Europe. Edinburgh first got a PhD in Bristol. Those of you who know Mike Berry would immediately recognize that everything important took place in Bristol, one time or another. And this would fit into his theory. But then he moved on to Liverpool, and this is where he did the things that made him more famous, collaborating with famous Max Born, later on Nobel Prize winner, with whom they did several important things that carry also his name, and also wrote this book, which some of us who are diversely young used to keep on their bookshelf for comfort about theory of phonons. He then moved back to China in 1951, became professor and eventually deputy dean and various other things at Beida, Beijing University. Then he became a leading figure in the Chinese Academy of Sciences and founded the Institute of Semiconductors, which he directed until the end of his career. Among the things that carry his name apart from the book, there is a Born-Wang approximation, which is a variant of the Born-Oppenheimer. Wang scattering from impurities, the Wang-Greeze factor when you make an electronic transition that changes the lattice coordinates, the Wang-Pekal theory. And several other things. The other things that he did besides science are establishing condensed matter in China or helping doing that very powerfully, raising generations of Chinese leading scientists and leading contacts with the West. His wife was a Scottish, and he had a profound knowledge of both worlds at the West and China and that allowed him to play a very, very important role. To me, in particular, this thing comes as a very impressive memory because I saw him leading the first generation, the first delegation ever to this place. And for that, I, somebody discovered a fossil find here, which I, with a bit of shame, I'm going to show you because I think it illustrates some of the things now. The shame, shame later, I hope. That's the Italian, that's the Italian TV, 1979, Italian TV, National Broadcast. So here he is. Very typically moving very fast. Sorry about the Italian. This is probably going to be permanent in the future and we are very happy with this fact that it has never been verified before in the past. What kind of contacts have been established by the Chinese delegation with the Chinese, not only Italian, but also with those of other countries that are present here? Very good contacts, very cordial. Here the center of theoretical physics has been established by the whole world, in particular from the developing countries. And the Chinese scientists have had a great contact with the scambi, seminars and discussions and also at the level of their own human. Regarding the Italians, of course, the contact has been very strong here in the center, but the Chinese have also gone to visit some Italian universities on our advice, in particular the Institute of Physics of the University of Padova and the Institute of Physics of Milan and other places, with a reciprocal satisfaction, I think. I think apart from the movie, this illustrates the action of ICTP as a can opener. There have been over the history, over the relatively long time we've been around here, closed communities, closed countries, countries where people could not get out and in. And for some very good reasons, ICTP could act as an opening instrument, so we could get people that could not go other places. And I think this is a rather important action and something of which we're all around here, very, very, very proud. And we look forward to doing more of this can opening in the future. Now, thank you very much. Thank you very much, Erio. I can see Erio hasn't changed at all in the last 40 years. The smile is the same. Very good. So let's continue with the next part of the ceremony. So the procedures that I will just first give the award to Luis and then to Honjun. I forgot to tell you that we have to do it there because of the photograph. But it's not yet, I will... I guess that's it. So for this year, the prize recognizes the independent contributions to the theoretical advancement of contest matter physics of modern solid state materials, including low-dimensional and nanoscale systems. So for Luis, the theory work by Luis Foa Torres contributed importantly to our understanding of topological insulators, graphene, and two-dimensional materials and nanotubes, including quantum transport and optoelectronics. So for this, I think he received the award. So let's congratulate Luis. So talking to Luis before he told me that he has been very familiar with ICTP throughout his career. He spent some time here as a participant in conferences. Then he was a postdoc for some time, and then he was an associate. So it's a symbol of what connections we have with many scientists in the world. And Luis is a great example that I think we are very proud that he has been attached to ICTP for many years. And so now Luis will give his talk. The title is Using Light as a Topological Switch. So thank you very much. Good afternoon to everybody. I feel really honored to be here and thankful for this prize. And also very thankful to ICTP for the support over all this year throughout my career since I was a PhD student. At first conference I came in Europe, was here at ICTP. And I'm very happy to be here back to give this talk that it's going to be on Using Light as a Topological Switch. I've chosen this topic because I think it's the research line that best reflects what we have been doing over the last few years. But towards the end of the talk, I will also mention other research lines that we are pursuing in Santiago de Chile. So this talk has at its center light matter interaction. And so to introduce this topic, I would like to use some images from a movie. It's a famous movie. It's called Wings of Desire. It's directed by Vin Benders. It's got already a couple of decades. Interestingly, this movie is based on the writings of Reiner Maria Rilke, or at least it's inspired, let's say, at least partially by the writings of Reiner Maria Rilke, who wrote the Duino Elegy while he was living very, very close to here in Duino. Duino in summer. The movie is about angels, and these angels live above the sky in Berlin. And these are immortal beings. They are able to see everything that's going on down on Earth. They are able to hear the people's thoughts, their feelings. They know everything that's going on down on Earth, but they cannot interact with people. And so, this thinking about how we use many times light matter interaction, this reminds me a lot of many of the best techniques we have for characterizing materials for observing phenomena, are really using light to observe the phenomena, to characterize the material without changing it. And there are many examples. For me, one of the examples that I prefer the most is Raman spectroscopy that's impressively useful, especially in graphene-based materials. And this is an example from our own research, a paper we published a few years ago with Victoria, experimental work. This is all experiments done with Victoria in Córdoba, a few years ago in Argentina. So, this is an example of the kind of techniques where we use light matter interaction to observe phenomena, not to change the material. But as it happens also in this movie, at some point this angel that is characterized by Bruno Ganz decides that he wants something more, so has the desire to come down to Earth and interact with people. So for this, he risks everything, he loses his immortality, and this is right the moment when he's down on Earth. The movie was released in 1987, so the wall was still there and you can see it there in the background. And you can see the moment when he starts to feel everything, the movie starts to run in color from this point on, up to this point it was black and white. And so he's down there on Earth. So our idea would be what I'm going to convey you today. It's a way of using light, but now to change a property of the material we would like to use light matter interaction for active purposes. And in particular, we would like to use light as a topological switch to be able to switch from a topologically trivial material to a non-trivial one. So as a very first step in this talk I would like to acknowledge my collaborators. There's a long list, this list is not complete, I apologize for this, but for the topics of this talk I would like to thank especially Pablo, who was a PhD student with us and is now in Delft, Virginia, who was also a PhD student with us, Victoria that I mentioned before, Hernán Calvo, who is now a researcher and a professor in Argentina, Stefan Roche, Gonzalo Usach and Carlos Balcedo, Bariloche, Eric in Valparaiso. So thanks to all of them, to a big list of people that I cannot mention all of them, and also to the institutions that made our work possible, starting with the University of Chile that is a place where I worked since 2016 from the seed that is our main funding agency, my alma mater, the University of Córdoba, where I did my PhD and I also worked for six years before moving to Chile. Connie said, and of course ICTP who supported me from my PhD very early years. So the outline of the talk is as follows. I will start with a very brief introduction to topological insulators for those of you who are outside the field, a couple of slides. Then I will turn to how we use light to change the topological properties of a material. I will give a very brief overview of theoretical framework, a rough idea, and then I will go to an example that will be the case of laser illuminated graphene. And there I will discuss different things like the appearance of chiral edge states, the whole response associated to these light-induced states, and then I will give some conclusions and I will briefly mention towards the end a couple of other research lines that we are pursuing in Santiago. So let's start with this. For me as a condensed matter physicist, the full story of topological states of matter and topological insulators starts with the integer quantum Hall effect. This gentleman, Klaus von Klitzing, discovered in 1980 what we call today the integer quantum Hall effect. At that time he managed to make some measurements in very carefully prepared samples of 2D electron gases. Today you can just think about the graphene, but at that point it was much more difficult. So you had 2D electron gas with a strong magnetic field perpendicular to it connected to a source, a drain, where you pass the current through, and perpendicular to this current you could have other contacts to measure a whole voltage. With this whole voltage and the current, one can determine the whole conductance and in contrast to the classical case what turned out to happen is that you have these plateaus that are almost perfect, where these plateaus correspond to a quantized Hall response that is essentially an integer multiple of phi square over H. Up to a part in a billion that is kind of a metrological type of precision, and indeed it is used today for metrological purposes. What happens essentially, this was very surprising at that moment, extreme regime of high magnetic field, high temperatures, what happens essentially is that under this strong magnetic field the bulk of the sample becomes insulating and at the edges you get states that move through them as if they were one-way streets, like with these arrows here. So if you want to produce backscattering essentially what you would need is to be able to connect one edge to the other, but this to tunnel through it's exponentially forbidden process because the bulk is essentially insulating. And so that's why this quantization doubtless came with a beautiful argument to be able to connect this Hall response directly with the topological invariant. And this was really the birth of, for me at least, of these topological arguments. And many years later, 27 years later, this was found now in materials, similar physics, but now with some differences is what we call today topological insulators. Now the role of the magnetic field is played by spin orbit coupling and typical plots you have is like this. You have energy dispersion as a function of quasi-momentum. You have in gray the bands that correspond to the bulk states in color the ones that correspond to the edges. So you see that there are these two states bridging the gap and they correspond to states that are counter-propagating and spin polarized. So spin up goes in one direction, spin down goes in the opposite. And these states turn out to be very robust and they happen not only in 2D systems, but also in three-dimensional systems. Three-dimensional systems is always a bit more difficult because you have less order for these surface states, but they are there. This wonderful playground for studying new physics, but the materials are a bit exotic and so one would like to expand the prospects for this type of topological states to a broader set of systems and materials. This is the point when we ask if we can make of any material become topological. And doing, for example, a graphene with such properties would be immediately very useful to bring all these physics to a material that is relatively easier to handle and so this would be very, very useful. So there's a full line of research around this that gained momentum over the last few years that generically people speak about flocate topological insulators. I will come a bit later to why this flocane I'm here. Essentially the idea is now to use a laser or more generally a time periodic potential to imitate these properties and to develop states that will be topological. So here this field for me at least was inspired mostly by this paper by Oka and Naoki. It's a PRB of 2009. There was also a paper by Lindner in 2011 where they coined this term flocate topological insulators and here there is a very personal selection of papers in this field that expand beyond materials to ultra matter and also photonic crystals. So the ideas I said would be to use light as a topological switch. From now on I will concentrate mostly on graphing. So in the case of graphing the problem is that you don't even have a bulk gap to start with. So one would need first to create a bulk up and then to determine if this gap is topologically trivial or not. So there's a lot to do. So the name this name flocate topological insulators comes because flocate is the, flocate theory is the prevalent theory for dealing with time periodic potentials. And I will give a brief two slide overview. It's very simple. What I will do is resort to what you already know and essentially what I will do is resort to what you already know about Bloch's theorem. So here is Mr. Bloch that his theorem applies to space periodic potentials and in the case of flocate I tried to make it as symmetric as possible so you see I left a white square there because I couldn't find any picture of messy flocate because it's actually an older theorem and now we use it mostly in the context of time periodic potentials and so the idea is that now we have in the case of Bloch we had a factorization in terms of plane wave times space periodic function in the case of flocate when the Hamiltonian is time periodic we factorize this in terms of a complex exponential that has the quasi energy instead of quasi momentum and now times what we call the flocate state. Flocate state is now periodic in time with the same period as the Hamiltonian and the interesting thing about this simple factorization is that it allows to separate dynamical scales essentially everything that happens in long time scales goes into this complex exponential and everything that is fast goes into the flocate states so taking advantage of this factorization when you plug this back into the time-dependent Schrodinger equation you get an effective you get this equation for the flocate states that effectively looks like a time-independent Schrodinger equation but now with this Hamiltonian that is the flocate Hamiltonian is a modified operator and now this flocate states that depend on two coordinates space and time so essentially it's an eigenvalue problem now in a space that is the direct product of the usual space and the space of time periodic functions. These functions immediately are essentially Fourier space so they are developed by an index let's say n so immediately if you start with for example the problem of graphing you imagine something like this if you think of the same problem now in terms of this flocate picture you have an additional variable this n that makes you have replicas for each state of excitation of the time-dependent part of your system the photon part so you can have the system with zero photons plus one photon minus one photon or so on so based on a reference you have this replica picture space this is very simple but it's very useful to think and with this you can there's already a lot of theory for computing topological invariance for doing transport in these driven systems I will talk a bit more about this later so how it comes that we can get a gap in graphing using light it's very simple think of for example this is energy dispersion of graphing energy projected over one direction in momentum space and you have this cone so if I now put on this full replica picture what you have in flocate space is now one of these cones for each replica they are shifted in h bar omega and so immediately you see that this becomes a bit richer and there are these crossing points between different replica states this is for example crossing between zero and one and so at these red points if you have a proper polarization of your laser you would introduce a matrix element and then you would lift the degeneracies and open a gap this was already proposed in the original paper by Oka and Naoki we got back to this later on because we wanted to dig farther and understand if this could fit any lab at all if this could be something that could be feasible in a laboratory so we were dealing for a while with this kind of things and this is work we did with Erman Kalvo who is there in this picture this is for example the one type of calculations we did this is a time average density of states as a function of the energy in gray you see what happens with pristine graphene in blue what happens when you turn on the laser in red when you put the laser that is now even more intense and so you can see that there are these gaps developing here at right the photon energy over two that is exactly where you would see this construction because by symmetry these red points are half the way between one dirac cone and the other that are shifted in the photon energy there is also a small gap developing at the dirac point that comes because of virtual process of photon emission reabsorption so here I would like to mention that this range this scale is not any scale it's a very precise scale after dealing with this problem there was a sweet spot where you could achieve a sizable gap without at the same time having a power that was too high and this corresponds to a photon energy on the order of 120 millilectron volts this lies exactly in the mid infrared and so the best tradeoff for experimental realization of this is something that we propose was in the mid infrared this is very simple physics a lot of stuff could happen here and ruin this simple picture was observed experimentally a couple of years later not in graphing but at the surface of a three-dimensional topological insulator if you think of a three-dimensional topological insulator the surface states are very much like have a dispersion very much like in graphing it's a cone like dispersion and what they manage in MIT the group of New Gedeck is to do a special type of arpes where they were actually observing these floccare bands these lines that are superimposed here in orange correspond to the different floccare replicas so impressively you could see these floccare replicas in an experiment and depending on the polarization you could see the opening or closing of a gap as well there this was really impressive for us there are more newer experiments that are even better than by the same group and so for the for us this was a boost because it made us think we were exploring in the right direction but then the next question was are these gaps topologically trivial or not so experimentally they were observed in theory we were expecting them but what happens with topology of these gaps this is something that we studied in more detail in this paper and here you can see again energy dispersion as a function of quasi momentum and here you can see a gap appearing at half the photon energy so the direct point is down here and you can see these two lines bridging the gap these are H states that are chiral they move like with the arrows that you see there in this diagram and so effectively this is kind of smoking gun that this gap is topologically non-trivial we did farther tests as well but immediately we could see that there were a lot of opportunities for tuning for example you could change the chirality of these states by changing the laser polarization you could change the speed the group velocity of these states by changing the laser power so these states that are flocac chiral states essentially do two things they decay towards the bulk of the sample and they move along the edge for the decay towards the sample that is this C here we found that interestingly it turns out to be independent of the material parameters it's just the photon energy over the amplitude of the electric field this comes because of cancellation of the Fermi velocity factor that also appears in the gap we also got some expressions for the group velocity how it is modified now for these states this eta is the dimensionless electron photon coupling strength so it controls also the group velocity of the states we got as well the first analytical expressions for these kind of states and this is work with Gonzalo you can see there in the in the blackboard he's happy doing this irradiated graphing things and Pablo and so we explored this farther we tested numerically the robustness of these states they turn out to be robust to different types of defects and perturbations very much like in the integer quantum hole effect so in this sense a question that one could immediately ask in the case of the quantum hole effect the hole conductance is given directly by essentially the number of edge states time e square over h and one could ask is it the same in this case with driving with the laser and this puts us immediately in a bigger problem because to be able to do a transport calculation we need to set occupations and this is a system that is out of equilibrium so it's tough in principle to do it one would like ideally to have the states occupied up to some point but this is something that we cannot guarantee and so in the context of these problems there are two very different and extreme limits one that I would call the coherent limit where you think that the laser is all over the place and so you will need to take care of this patient unavoidably in the opposite regime that I would call more coherent like regime you think that you have a spot that is finite where the laser is there and then you have your leads that are non-irradiated and so there you can use thermal occupations and if you further assume that this patient is not taking place inside the central part of the of the illuminated sample then you can do a very simple scattering picture for this a flocase scattering picture there was there's a lot of papers on this on this topic with both times both regimes but even in the coherent limit that is the one that seems simpler to deal with there was a lot of controversy when we decided to face this problem there were some predictions that said well the whole conductance will be quantized with the number that it's essentially given by the number of age states and some other people said no we don't see any quantization we do some numerics and we see an anomalous suppression so looking a bit farther we realized that in one of the cases the calculation was not too terminal was multi-terminal was not multi-terminal but two terminal sorry and while in the other it was an analytical calculation for very high frequency so the photon energy was actually larger than the bandwidth of the system and so this is very far from the regime where we are interested in and so the question was really open so we decided to do full numerical calculation for a system that is multi-terminal that has a region that is being illuminated using Floquet scattering theory and to make a long story short what we found is that actually there are plateaus developing here in the in the whole conductance or resistance they are rough they are not as nice as the plateaus as in the integer quantum call effect and but most interesting we can now try to check what happens with this number of channels associated to this plateau we can see here that there is only one channel active but if I look at the dispersion relation for the same set of parameters I see that indeed I have more edge states there's not only one and looking more carefully numerically we could see that effectively there were one of these edge states that remained essentially silent only one was contributing fully the other one was silent and the reason is that in this Floquet picture you always have a lot of Floquet states that could come from the hybridization between higher order replicas and these states although they are there and you can see them numerically you can see them analytically they can be very difficult to excite in a mattering situation so essentially what you see is that you cannot get into these states and you have only a subset that is contributing more importantly at least in the weak electron photon coupling limit interestingly this was this kind of things was this problem was addressed very recently in experiments that appear in the archive in November it's a group in Hamburg, Andrea Cavallieri experiments in a four terminal configuration using a laser in the mid infrared regime once more and and they managed to see this light induced whole effect and everything seems to to be reasonable in regards of our predictions so this was great for us very nice to see that this is advancing forward with experiments that are very much needed so our message from this is that you could see a rough plateau but in principle there are differences with the case of the integer quantum whole effect there may be many channels that essentially remain silent and we also found that there is indeed a hierarchy of these edge states that is very complex if I had to start over again with this I would probably not start with the case of graphing I would start with the simplest system that is the simplest case of a laser of one-dimensional topological insulator the Sushi-Frenheger model if you start driving this system you could see a lot of topological phase transitions that happen this is work we did with Virginia some time ago so here is the reference if you are interested in getting into this field interestingly this was observed experimentally not in ultra cold matter as we thought at the beginning but in photonic crystals by this group in China is a collaboration between people in Beijing and the other institute so very very nice experiments so with this I will come to my conclusions I try to draw a picture of this field that is not as a modern map it's more like one of these old age maps it's rough and with many many warnings around so I hope that I managed to convey the idea that we can use light to really change the properties of a material to change the topological properties of it I talk a bit about the hole conductance and there is still a lot to do especially regarding interactions and dissipation it's a problem where there is a lot of stuff to do still and so I would like to take these are some references of the works we did in this research line over the last few years so if you want to take a look at them so I will just mention briefly two other research lines and I close one of the other research lines that we are carrying on in Santiago has to do with achieving one way transport of charge, veil, spin and of course as well it could be energy and for this what we are doing is trying to achieve this one way transport by crafting more reciprocal band structure in the system so imagine you could have a band structure where you have states that only move in one direction in a given range and not in the opposite so this could be very useful for many things and for this we are using kind of multi-layer configurations we like to talk about layer tronics and where the states move only in one direction in each layer and so this is I think a very promising idea although not many people found it that interesting yet where we were more lucky was with this other research line that has to do with the search for topological states but now in systems that have a non-armitant term due to either a life to describe a lifetime due to interactions or for example because it's an effective description and you want to take into account gains and losses as happens in photonic systems or mechanical systems and we are looking for topological edge states, topological states in these non-armitant systems and we found something else we found that non-armiticity in some systems could produce that in a lattice that is ordered it becomes the void of extended states that's something that it's really crazy it shuffles what we know from solid state physics and some people more recently call this the non-armitant skin effect and it's very interesting thing we wrote a short review that appeared recently on this there are a couple of papers more papers that are appearing this work with Victor Manuel and Eduardo Matias in Santiago and so with this I would like to thank you again for all this thank you very much please very nice talk someone may have some questions please very nice talk thank you my question is about roughly quantized you know so usually in topology it's either quantized or it isn't quantized so you find something that's almost at the quantum what determines the order of magnitude of the deviation from the quantum and is it it looked random but I presume it's not really random is it? it's a very very good question actually what happens here is that in this configuration that we are studying you have a spot a laser spot so you have a frontier between the irradiated and the non-irradiated region immediately this puts you some problems in the contacts because it's like intrinsically even if it is the same material you start to have problems because of these contacts and also you start to having problems because in the weak coupling regime where when the intensity of the laser is small then the picture looks kind of pristine but as soon as you start increasing a little bit the intensity typically you need to do in numerics to be able to fit this in a reasonable system size you start to have contributions from higher order processes so this start to put some dirt all over in your nice plateau and so it's it's very complicated I would say more questions? thanks very nice talk normally when we think about light irradiation we tend to think as a reference which is the non-excited reference and then build this excited status excitations from it how would you see this kind of situation that is clearly beyond that picture how can you reconcile it with that kind of a picture? okay if I get correctly you are essentially asking about the difference between the really quantum version of the problem and this Floquet version in the really quantum version you get a lower always the spectrum is bounded from below in this case not because we are using a kind of semi-classical approximation where we replace really the field by a term in the Hamiltonian that has a well-defined amplitude in the face and so in this sense it's at the end both pictures match in the limit where you have a very strongly excited system so you are in the limit where you have any photos in your system other questions? perhaps I did not understand when you have the intense light shining on the graphene you have this one way edges what determines the way? great great very very good question it's determined essentially by the laser polarization because all the time here it's crucial in order to have this chirality you need to break time reversal symmetry to use a circularly polarized light and this the fact this polarization determines the chirality of the states there was a circularly polarized light so then does that make it like this turn insulators that have you know broken time reversal symmetry? very good question as well here for graphene it's necessary to break time reversal to have non-trivial properties but I could go to another class and there I could have different phloketopological phases that preserve time reversal symmetry with time reversal symmetry that is preserved and there you go to different class it's wonderful physics as well very good more questions? as you have something from an outsider do you see any potential applications of this? well it's difficult to tell for the moment I would say that we are very excited to see that this finally got inside the lab many people start to speak about tailoring the properties of a system on demand now using the radiation the truth is that at this point now it's difficult still because it's very very tough setup few laboratories in the world can do it but I think that we are in a very interesting way in a way of using light now for very active purpose and who knows maybe one day we get applications out of this physics so let's thank this again let's go to the final part of the ceremony we are going to call Honjun he developed first principles based computational methods addressing advanced variety of problems including low dimensional materials and multi ferroicis where his approach has become standard in the field so I'm very pleased to give the award to Honjun and then Honjun will give a presentation the title will be theoretical stories on new mechanisms of ferroelectricity and multi ferroicity first I would like to thank the ACEP for the selection committee for the support and also I would like to thank the organizer of the total energy workshop for the invitation today I will talk about theoretical studies on new mechanisms of ferroelectricity and multi ferroicity before that I would like to clarify a long term collaboration with professor and also my former postdoc advisor and also my PhD advisor from USTC this work I will talk about also in collaboration with some other people including my postdoc my PhD students and also professor Lorna Blesch and Dr. Yang from University of Arkansas and professor Jorge Yenegas from NIST and Dr. Huang and professor Erjun Kan from Lanzhin University of Science and Technology and also professor Picasso from CNR Spin and also the founding and supporting agencies has the outline of my talk first I will give a brief introduction to ferroelectricity and multi ferroxity and then I will describe several unusual mechanisms of ferroelectricity first I will describe the unified process model for speed order induced ferroelectricity we find that the ferroelectricity can also be enhanced by the rotation I will also discuss ferroelectricity due to the coexistence of object order and charge order and I will discuss totemation of ferroelectricity systems in a brief summary to start with let's first take a look at frame magnetism I think almost everyone is familiar with frame magnetism so in a frame magnetical system it has a spontaneous magnetization and also it can be controlled by the external magnetic field similarly in a free electrical system it has a spontaneous polarization and the polarization can be controlled by the external electrical field in a multi ferroelectric system it has not only magnetization but also polarization besides if there is the magneto-electric coupling then we can realize the control or magnetization not only by the magnet field but also by the electric field and also we can also realize the control or polarization by the magnet field Felix is already has many applications for example this kind of capacity and also the sensors and also some other applications for multifarcal systems it may also has many promising applications because the presence of magneto-electric coupling it can be used to make an electrical rat a magnetic ray of memory in such kind of device of this material is the multifarcal so if the wattage switch the electric position and it also switch the magnetization because of the magneto-electric coupling the switch of the magnetization we also switch the magnetization of this part and then it will affect the resistance of this so we can read this state by reading the resistance but we can write the information by applying an electrical wattage so in this way we don't need any magneto-field so we can realize low power, high density high speed memories so of both favoritics and multifarcal they display favoritity so let us take a look at the mechanism of favoriticity so in the most typical favoritic system for example a barrier titanium oxide the mechanism is due to the covalent bond formation in such case at heart temperature in the peri-electric case the titanium ion is at the emission center when the temperature is lowered it will move toward one of the near bi-oxygen in this way the titanium ion can form a bond with the near bi-oxygen it will result in low energy bonding state and high energy anti-bonding state because of the titanium D states are empty then this way the electron will occupy the low energy and the bonding state in this way the total energy will be lowered so that's the usual mechanism for the typical favoritics this kind of mechanism usually require the empty D states or long pairs for example abysmous 3 plus however currently all known multifarx are not that useful because the performance is not that good so in order to develop or design high performance multifarx we need to discover new favoritical mechanisms one of the mechanisms which can induce the favoritical is the spin order this is because the spin order is the inversion symmetry of the system we can this example for example this is a one-dimensional case so if this is a one-dimensional magnetic ion is a formal chain so the magnetic ion is the inversion center however when the spin order the spiral spin order is formed so let's say that the inversion center it is longer than the inverse center this is because the spin order breaks the inversion because if you apply inversion to this spin structure you will get another spin structure this spin structure is not the same as this spin structure so that's why the spin order breaks the inversion symmetry so in this way at this moment we can understand the magnetism might give rise to a favoritical proposition however this does not give us a microscopic mechanism of spin order induced favoriticity because the proposition is due to a charged degree of freedom of electron where the spin is mostly due to the a degree of freedom of the electron so they are not related to each other so the mechanism of the spin order induced favoriticity are communicated and not clear at that time in recent years we have proposed a unified proposition model for spin order induced favoriticity so let us consider a spin pair so we have two spins and then we can express the electron proposition as a function of the spin of two sets so one set is only consider this the same the same set term because the interest set term and the other is the idea they are different so this will equal the the same set and this will equal the interest set interest set term so if we consider from the spin order interaction this term is related to the same entropy of the spin interaction and this pair term is related to the other three terms of the spin interactions and this is the the same exchange and also the the same exchange and also there is another similar and also tropical exchange terms we can also understand what spin order will give us to position by considering the coupling between the spin or the degree of freedom so if we consider the spin charge degree of freedom it means if the structure itself crystal structure itself is fixed the structure itself has a motion symmetry however if the spin order does not have a motion symmetry the spin order will deform the charge density so the charge density will have no motion symmetry so this part where we call the pure electronic contribution to the position and also there is another term the spin order will also deform the crystal structure the structure might deform so the arm might move and the movement of the arm might also induce the proposition so we call this as an onical displacement contribution to the proposition and we found there is another term to the proposition so if because the spin order might also induce a strain to the system so if the system itself is a partial electrical then the strain will also give rise to another contribution to the proposition so we call this as lattice deformation contribution so all these three kind of contributions are related to calculus by the parameters for example this kind of spin charge coupling and this is the spin form the coupling terms and spin lattice coupling interaction terms so in order to make our model useful so we should be able to calculate this kind of interaction parameters so to this end we developed a four state method let us consider the purely electronic part these coefficients for example if we want to calculate the p1 2x1 coefficients of this so we can consider these four states so in these four states the spin of s1 and s2 are different and the other spin other spin can be chosen to be the same and then we can derive that the coefficient of this can be written as the polarization of these four states and then divided by four so in this way we can actually calculate all the parameters needed in the unified polarization model and the the first state method can also be used to the other interaction parameters for example spin interaction j and also the anti-symmetrical exchange geomagnetic mirror interaction terms and also the spin lattice interactions and all methods was also adopted by other groups including Albert Fett and Steven Newey and Spartan with all model and all method we can now understand the mechanism of spin order induced failureity for example in this system Magnet's add-in-to system experimentally it found that such kind of spin order will give rise to a polarization however this cannot be explained by any of the present models we found that actually all model can explain the experimentary laws and we found this is due to a generous spin-corner interactions and in this another system this system is a scammy system but on the magnet field the system were ordered in a failure electrical failure magnetic system failure magnetic state this failure magnetic state will give rise to a polarization however this model cannot be explained failure magnetic state will give rise to a polarization and we found that actually this is because of the special sincestate terms they can contribute to the polarization and in the most famous modified system iron oxide system it was known that as a polarization mostly it is due to abysmose this is the big one actually we found that the spin order due to the iron will also contribute to the polarization actually the spin polarization and due to the the electrical polarization due to the spin order is in the opposite direction to the anonical contributions previously we believed that this spin contribution to the electric polarization is due to the spin current terms we found that it should be due to the exchange friction which has nothing to do with the spin coupling this is a current term if you require spin coupling and we also found that actually this system has a very large lattice deformation contribution let us turn to the second mechanism so we found that the fallacy can be enhanced by the rotation usually in a classical system there are several different kind of instabilities one is the fallacy so it means all the cotton along the same direction and another is the it means the rotation it was usually believed that the rotation compete with the fallacy as demonstrated in one bit of 95 papers however in 2008 it was found that it was discovered this kind of material think tin oxide it has a large rotation along the one-one direction and also a very large electrical position also along the one-one direction so this is very unusual because this kind of system has no empty D state and it has no long pair and it has no hydrogen bond and as I mentioned this system has very strong rotation and after that actually people also found some other systems for example Lyssen-Oxmin-Oxat others it has a similar similar behavior so it also has fallacy and also strong rotations so we are trying to understand where the system can have very large rotation and very strong fallacy so we first calculated the frequency of the fallacy as a function of our rotation amplitude so I calculated a number of systems so if we look at this curve for the tin oxide so we can see that when the rotation is increased so the frequency of the fallacy mode so immediately it is an immediate frequency so the magnitude of the fallacy is increased this means this means the increase of rotation will suppress the fallacy because if the magnitude of the fallacy mode is large it means it has a larger instability this is the usual behavior however if we look at this side when the rotation is large so we can see in this case the increase of the rotation will increase the magnitude of the fallacy mode so this means that the fallacy disability gets enhanced when the rotation we also consider the coupling between the fallacy mode and the rotation in another way so we can directly calculate the fallacy amplitude as a function of the rotation amplitude so similarly to the previous results so when the rotation is increased initially the fallacy mode decrease when the rotation is very large and then at this time the rotation increases it means the fallacy electric amplitude will be increased so we are trying to understand why there is a strange coupling between the fallacy mode and the rotation mode we use the longer like a model so we consider a coupling between the rotation and also the fallacy mode for more calculation we found that the coupling between the this this term is used square terms this term the coefficient before this term is positive it means that when the rotation is small this kind of coupling will suppress the fallacy however we found another term this we call the u u square and other fourth term and we found the coefficient before this term is negative so it means that when the rotation is large then this kind of term will enhance the strategic instability in order to lower the fair energy so this is from the longer model point of view so let us consider this problem from the atomic point of view so if we consider the rotation the structure the A-set is a zinc if there is no rotation around the each year zinc arm we have 12 nearby oxygen all the zinc oxygen has the same boundary length when there is a rotation around the one-on-one direction the distance between zinc and oxygen is split into different groups in the plane so one group the zinc oxygen distance is decreased for example to 2.28 oxygen and the outer plane oxygen distance between the outer plane and oxygen and zinc does not change very much we know that the usual zinc oxygen boundary length is about 2.1 so in this case oxygen boundary length is still larger than the usual boundary length it means that the interaction between the zinc and this oxygen is mostly attractive so this attractive interaction will prevent the movement of the zinc out of the plane so we will pour the zinc in the plane so this way we can explain why the presence of small rotation will suppress the failure to go into the bead however when the rotation is large so in this case if the zinc is still in the same plane as the oxygen so the boundary length between zinc and oxygen is very small is 1.7 then it will be much smaller than the usual boundary length so this means that the interaction between the oxygen and the zinc it becomes repulsive due to the polyviportion in order to reduce the polyviportion the zinc will either move up or move down in either case if we induce the local dark moment and just the reason why a large rotation will enhance the failureity we can see that this kind of mechanism actually is a kind of a steric effect so it is compatible with magnetism in fact we predict a wrong temperature frame magnetic and multiviral system with this kind of mechanism this kind of mechanism also may display unusual pressure effects usually but the pressure will suppress the failureity either the Tc or the position however we find that if the failureity is due to the rotation then in this case the failureity might be enhanced by the pressure might increase the rotation and then increase the enhanced failureity some after our prediction the explainer groups in Hong Kong they confirmed our prediction and they found this in another similar systems in recent Ottoman systems in the third part I will talk about failureity induced by the combined charge order and orbit order it was known that either charge order or orbit order itself can induce failureity however we found that in some case the single presence of charge order or orbit order cannot give rise to failureity however when charge order and orbit order coexist then it will give rise to multifroxity this can be understood in this system if you if you consider the holicom graphon lattice so the system has a inversion symmetry here and also it has a three-fold rotation and this is the two-fold rotation and if the system as a charge order for example it becomes a boundary matrix then the system has lost the inversion symmetry but still the system is still non-polar although it doesn't have inversion symmetry however the presence of orbit order for example the orbit order like this then the orbit order will lead to the system resulting in a polar symmetry so this kind of structure will give rise to failure electricity actually we found such kind of mechanism it can be indeed occur in realistic systems so we found that if we dope electron to chromium bromium and then for example with lithium in this case before electron doping the chromium is a three-plus and the three-electron occupied T2G orbiters however when there is some doping so this chromium comes to a two-plus and this becomes a three-plus so the two-plus set has four D-electron then this has a jontal visibility so we have some charge order and also have some this charge order and this set has an orbit order so in this case we have this not only charge order but also orbit order and actually this kind of combination will give rise to a failure electric or failure electric and multifloric systems so basically we consider the combination of orbit order and charge order might give rise to a multifloric city but also the combination of some other orders may also give rise to multifloric cities finally I will talk about two-dimensional failure electricity so in order to make the device smaller and smaller because a two-dimensional this smaller device will consume less energy and also it has a larger storage abilities so that's why people are very interested in making a failure electric or same famous however there are some difficulties first well known difficulties that there is the depletion field so if you consider the same thing with the auto plane electric proposition in this case the same thing there will be a charge at the surface this kind of charge will give rise to the depletion field this field will destabilize this so we call this kind of field as the depletion field this usually makes the auto plane failure electric is not stable in same famous and also there is not an argument even for in plane failure cities to failure city in same famous usually also decrease in same famous this is because when the charge is reduced then the interaction is cut off and also there are some other such effect will also decrease the failure efficiency we consider the prospect same famous because currently most commercial failure city bus are made of plastic oxides to continue let us introduce the total factor so total factor you define as this so it means if the A set is large then we have a large total factor if B set is large then it has a small total factor so let us discuss the large total factor case first then I will discuss the small total factor case later for example if we consider the barrier total oxide same film so the barrier is very large so it has a very large total factor so barrier total oxide itself it is failure electric and we find that in the same film actually we find besides the usual the failureity due to the the usual failure mechanism this is because the bulk itself it is failure electrical but we also find another state this state is due to the surface it is due to the surface titanium and we find that this surface failureity as on usual film and sickness dependence so this is the usual failure electrical mechanism so we can say that when the sickness decreased the failureity it is decreased as usually believed however for the surface failure electricity we find that when the same film sickness decreased the surface failureity actually the surface failureity may also in some other systems the bulk system itself it is not failure electrical but the same film case can be failure electrical this is because of the surface failure electrical mechanism so we can find that the surface failure mechanism it has nothing to do with whether the B-SAT has an empty D state or not so we can use this kind of mechanism to design a modifier system let us now consider the small torrent factor case so if the A-SAT is small it means the torrent factor is small then in this case there will be a rotation of the octahedral rotation in this way the A-SAT can interact more with the labor oxygens the most calm rotation of the project is the so-called A- A-C+, rotation as shown in this case so for example in this Cassian-T-Oxac case it has such kind of rotation so when we started a same-fame case we found that because of the rotation so A-SAT will have movement so A-SAT can move along this way and the laboring A-SAT will move along the opposite way so if the same-fame case has all the number of A-SAT then it means that these layers will cancel but they still have some remaining position so in this sense we can call this system as a very very electrical system so in this way we can understand so when the same-fame is decreased and if the number of A-SAT of the A-layer is odd then the position will be large when the signal is small and this kind of this kind of phylogical mechanism actually is well known in the literature in the three-dimensional case they call the three-dimensional hybrid improper phylogity and this is because of the coupling and also the rotation the two-rotation mode therefore in the project case we find two new phylogical mechanisms one is due to the surface effect another is due to the two-dimensional hybrid improper phylogical mechanism and in both cases we find the position might be enhanced this is different from the usual dependence and we also started another two-dimensional phylogical system so this is the Tintera the same-fame so in 2016 the group in Qinhua University they grow the Tintera the same-fame on graphene and they found interesting observation so they found that the T-C the phylogical T-C of the same-fame can be even higher than the bulk T-C especially the two-unicell or three-unicell or four-unicell case they have very high T-C which is even higher than the wrong temperature in their paper they explain their experimental results in terms of defects so it's kind of extrinsic mechanism and they they suggest that in the same-fame case they have less defects than bulk and if the decision has a way a lot of defects the T-C will be decreased because of the defects we are screened a double-dap interaction and that's why the T-C can be decreased this is their explanation where also this question so how the T-C depends on the same-fame if the if the system has no defects both in the same-fame case and in the bulk case first we calculate the energy barrier between the phylogic state and the periodic state of the same-fame as a function of the same-fame case we find an interesting dependence on the same-fame case first for the wrong-unicell case the barrier is small rather small however it is not monotonic if first increase then decrease because it is well known that the barrier any barrier is related to the phylogic T-C so this T-C might also have the same dependence but this will be very different from what are usually believed in the literature so we are trying to understand why such kind of system has such kind of usually dependence on the barrier on the same-fame case so we calculate the first constant of the T-terra same-fames this is the result for different same-fames and also for different T-atoms for example we can calculate the first constant of this T-atom and also this T-atom and also these T-atoms so if we look at this curve so this means that this is the surface T-atom and this is for the T-unicell same-fame case and this for the surface T-atom and the soft surface and also the inner part we have some observations so the surface T-atom has a very small first constant this means that the surface T-unicell will contribute not to the phylogic instability compared to the inner part and this why in this region the barrier will get increased when the frame thickness is decreased that because when same-fame is decreased the surface contribution is getting increased because the ratio of the surface of the surface-atom is increased however then it is not this way and then without what decrease at some point this is because actually when the same-fame get way same we find that the first constant get increased especially when the same-fame is way same for example two-unicell and one-unicell the first constant increase a lot so this is because the actually when the same-fame is way same and then the system has very large band gap and the stabilization becomes very weak so this can explain this part this is the previous this is the resource on the inner barrier and we also estimate the TC by using the fact that actually this was originally developed by the one-bit group and we constructed the Hamiltonian from the derivative calculations from all Monte Carlo simulation we found that the TC has a similar dependence as the energy barrier so we found that especially for the for UC case it has a TC which is much higher than the bulk case so always suggest that the defect-free same-fame case might have stronger ferricity than three-dimensional defect-free bulk so this is intrinsic effects and this might be important because maybe we can make the same thing and on the one hand we can reduce the size of the device on the other hand we can enhance the ferricity and thus to improve the performance of the device this is the summary of my talk basically I described some new mechanism for ferricity and also some of these mechanisms can be used to design motor forks thank you for your attention thank you very much Hong-Jin are there any questions? people think let me ask myself when you discuss the increase for electricity by rotations you say that your model has been tested experimentally but you pass a little bit fast that part because how quantitative it was this test yes so it's very quantitative for instance you can compute the minimum and then how big the rotation is when you start increasing and these kind of things you can be actually we also computed calculate the phallic position on the pressure and this is a schematic draw but all religious calculation is all similar to this so in this case we find that actually the pressure will increase the rotation because the rotation will enhance the ferricity so in this way the ferricity also gets enhanced by the pressure and especially they measure the Tc or lesion and on the pressure and has a similar behavior very good very impressive more questions? ok so before we finish there are some refreshments outside that we can all share to celebrate this award so everybody are welcome to join us outside except for the winners because we ask them to stay here for some photo session and then you are welcome to join so let's thank and congratulate both of the winners