 Ok, so, načo se zelo, zanoli. Pintronik vzelo na interfejst. Ok, to je vse. Tako, različenje, različenje pri pristem. To je vse vstup, ko je vstup. Zato je vstup za 5 vrst, in je v Judjiši, in kaj je vsečen, da je kontynično vsečen in vsečen vsečen z dfg. Tako je o spintronici in vsečen. Vsečen vsečen vsečen vsečen vsečen vsečen vsečen vsečen vsečen vsečen vsečen vsečen. Magnetic properties are enhanced at the nanoscales, so we can have device miniaturization. Of course, if you are studying fundamental physical properties this opens the door to investigation of basic physical properties, but also you can have nice applications, so we have it all, then we do it. da poživamo. The first thing is, if you want to make devices, you have to have like a transistor. The idea is not new, it dates back from the 90s by the data spin transistor, and the main ingredient here is to use a channel, material, which is where you have a spin polarized career and you want to manipulate them, you want to act on them with an external potential. potensijali. Pa vi srečenja, da je potenčnja, in poslutnja aktinga o spinsih, ki so tudi v kanal, je poslutnja v spinoin, občaske cahla z napasvenim. Nisem tudi, da mi se pričel, pa vidjem, da smo pričelji spinoin, občaske cahla in jel je jel da tega opravimo, da se počlega počlega z napasvenima kaj je odložite. in teore, v pravdu je ne. Zato je več rado, da je zelo tudi pravdične, in je zelo več svih materijov, z kajštih spinov, spinov, in spinov del, spinov del. Tepične izglede, kaj in semitonduptor, in karboni, kajštih nanomaterijov, je tako vsega prijev. Zato je, kaj je najbolj prinsipul, pojelj, in je zelo vse, ker je bilo počet, prvi prinsipave, začnev njih vsek, in na učima, na kratku, na obdajovosti. In počeš, in tudi, zينijo to, da vičen chevrat, da se moždje se však tekšne različju. Da ti cantel Vojdej začenj. In drugoh držav temu, da pa bo dobro, v ko si boljo, na ko se srbe, na neče, da je vsek, je to, da se prejste, kako se pravati na mikroskopikno vzlog, na lokalno vzabila interfejza. To je ne v mikroskopiji. Propertizv in vzboj, zvukajemo菌, da izgledajte detelje o interakciju, kaj je materij, kaj je vzbolj, za kaj propertizv. In se vzbim, da zaj vidim, da se vedimo, vse propertizv, ki pa pravati, tako, rektopologij, in sudator, Gjelowski, Morien, interakcije in se. in zelo smo vzouto všeč kaj posledali. Kaj je materijal? Prvno je, da je svojo materijale kandidati, da počušajte vse. Završajte kakvom, zelo je to vzelo, da je vzela tako državnje spindifuzion. To je in trisik. Zelo je, da je vzela vzela, je nekaj, je 6, zači se počeš. Meni se, karbon je tako magnetizac. If you take a perfect carbon nanotube or a perfect graphene sheet, there is no magnetism in that. You have to do something. You have to functionalize the system somehow. One possibility is to create a vacancies. You have defects. When you have unsaturated bonds, the electron will carry some magnetic moment. You will have some spin polarization, način injurymm. Ashley tickets. into the middle of the namazum. Depo apart about what to e. kako se prišli. Grafin, spin transistor, kaj je vse občinil v labi, ali tudi nekaj, nekaj, vsega vsega vsega, kaj je vsega vsega, spin-orbitka, in vsega vsega vsega grafin. Tako, zelo, vsega vsega vsega vsega vsega, vsega vsega vsega, spin-orbitka, in vsega, if there is no gap, you cannot switch your transistor off. And so it's quite of a problem. So how do we solve this? We make an interface. We make an interface between grafin, we place grafin on a substrate. If this substrate is magnetic and semi-conducting, it will induce some magnetism in the grafin, but it has to be shown, objuz. And if we want the material, the substrate material with a spin-orbitka, coupling the hybrid material, we'll also inherit that. And in addition, we can apply an external field, or we use a ferroelectric in order to tune the bands. And so that's the idea of putting grafin on top of a magnetoelectric substrate. So in practice how do we do it? So as I said, I'm using first principle techniques. And for these words in particular, two codes have been used, CESTA and FLIR, for different targets, either for to deal with large system or to check on the accuracy and so on. But in general, the combination of these two codes give us very detailed accurate information on electronic strata, magnetic properties and so on that we can use to study the details of the interaction. And then also combine to other codes, like the Vanier 90 to get topological properties or with a code by FIVOS to get spin dynamics. So we started easy from collinear spin calculation. We said our choice for the substrate is barium manganese oxide. We want something with manganese because it has T-electrons. It binds pretty well with the graphene so it's a practical choice. It's hexagonal so you can put it on the graphene. And most important is semiconducting. So we will not short-circ with your graphene. Because then when you go and talk with your experimentalist friends and you say I'm going to do an interface. First question is, is the substrate insulating or a metal? Because otherwise, if it's a metal and all the electrons go through the substrate and you can't do much. So here it's antiferromagnetic between the layers. So let's say the yellow atoms that you don't see here are in the middle of these let's say red things. So these are the manganese and they are antiferromagnetic in the C direction here and in plane they will form some triangular lattice. We will see more later with spin orbit coupling calculation about that. So first we do a slab just to check that actually, I mean the system stays semiconducting and it is. It's spin polarized so that you can see it also in the bands red and black are the two spin polarization. And what is interesting to see is that at the top of the slab, at the two let's say edges of the slab you have extra charge so you have basically two more electron per manganese atom and this gives some extra magnetic moment with respect to the bulk. And this is also useful because it means the system, it's ready to donate electron to make some binding towards something. In this case it will be binding to the graphene. So this is how the system looks like and there are several possible configurations but after some careful study one can see that let's say that the lowest energy one is where the manganese tend to be in the hollow sides here and the binding is very strong. Sorry, I skipped one. So once we have this structure we can calculate the electronic structure and the spin polarized as we want to and one can also think about to use this to make spin polarized transport. So around the region here close to the Fermi energy so there is the first interesting thing because one has been generally it's conducting the other, it's not. So there is a gap for one and no gap for the other as it says spin filter. Overall all the bands that are here let's say not so far from the Fermi level they are contributed by the interface atom only. So these are already just properties of the interface and it's of course the graphene but the manganese are the surface the oxygen near to the manganese and all the rest more the inner ones they do not contribute. In particular barium does not contribute at all because it's like 5s electron very low in energy so any other rare art manganese oxide will do the job because what is important here is the interface with the manganese. So last but not least the imobility region of course we love graphene because of the imobility region the Dirac points and here we notice that they are split by spin and the splitting is quite large so they are divided by 1,000,000 electron volt and eventually one can use doping to access this region. So this is what we have just seen if you doped for instance with boron, with an acceptor you can raise it up and if you put more you can raise it up a bit more so you can in experiment access this region here. Also with doping the velocity of the career so there is an extra bonus for the mobility here. So now it's quite clear what happens from the electronic point of view there is some charge that goes from the manganese atoms into the the graphene and this is reflected in the magnetic properties of the system. So here I have the spin density so spin up minus spin down for our system, so graphene with the substrate from the side and from the top and first thing one can notice is the magnetic moment on the manganese atoms they change after the graphene is placed on the substrate and that's because the electron they jump into the graphene and induce magnetization in the graphene. It's not strained, there are no defects so the only reason to have a magnetic moment which is not zero in the graphene is because of the proximity with the substrate. So magnetization is induced by the manganese. But also the story does not end here this is already very nice and was published in like one year and half ago but then so we decided honestly because I'm curious and I wanted to see to go a bit farther in this story, it's not just when you have an interaction it's never just one that interacts on the other but you can also have the other way around so the graphene affects the manganese and you can already see it here because the magnetic moment that goes instead of being four is about three at the top so something is happening also from the other point of view so what is graphene doing and that's the question that I want to answer here. For this we perform the calculation with spin orbit coupling with CST and FLIR and we found a remarkably nice agreement between the two codes and we recover of course for the bulk what is known in experiment so this is calculated so these are not the arrows that we put by hand they are really the result of the calculation so you can really see anti-ferromagnetic alignment on the manganese between plane and the triangular arrangement of the spin in plane we have delovsky-mojdi interaction as Stefan showed before and you have actually spin spirals here in the plane as you can see the easy axis is basically an easy plane in plane here so what happen now when we cut the slab so for a bare manganese oxide slab we basically recover the bulk properties so still in plane easy axis but when we put the graphene on top of it things change dramatically because the easy axis of the system becomes let's say in the z direction instead of being in plane and a reason for this is can be seen in the interaction between the graphene and the manganese atom so we saw that there were these electrons on the 2 per manganese atom we saw that there is charge transfer and spin also transfer and in a sense let's say the graphene are like spinning I mean take the spin of the manganese to change to be out of plane and this is exactly due to the let's say close interaction between the two so this is quite a strong effect now this is good but again we wanted more so we took the we calculated the bands for spin orbit coupling these are what we have seen before we focus on the Dirac cones so this is collinear sorry and this is with spin orbit coupling so we have spin orbit splitting so as Stefan said here we are breaking the inversion symmetry of the graphene and so there is some gaps that are opening here and at the k point at the k point is about 8 milliretron volt and right and left is about 2 you see ok, this is small but if you think about pristine graphene and you apply a feed and so on and you have the spin orbit splitting these are like 200 order of magnitude larger that one what you have in graphene so it's a way practical way to affect heavily the properties so this hybrid system has relevant spin orbit splitting so this was done with siesta and then we did we studied the interface the bands with flow as well and we got the same result basically we have here is just the bands at the interface so we got again a gap about 10 milliretron volt at the k point the small gaps right and left are about 2-3 milliretron volts are similar and then we used the Vanier 90 to calculate all the topological properties so here is the anomalous hole conductivity and one can see that is quantized so you have in correspondence to the energy of this gap at k, it tops at 1 and so we have we have calculated a certain number which is 1, so this I all thank the introduction of the fund because it's basically all the fundamental things I explained before and this system is a very nice system where we can actually see this effect and this is also very curvature computed so where you can see that at k point you nicely have the effect so the system has topological and non-trivial properties and that we have learned a lot of things with the spin orbit calculation, we wanted to go again a bit farther and see the spin dynamics at finite temperature and this is the last part of the work so so in order to study this slab system we did different spin configuration we extracted exchange coupling parameters between the planes and in plane like this and we put them in Monte Carlo code and we got the spin dynamics so this is how it looks like in general and for instance we did it with bulk first so this is an example of configuration that one can choose and we calculate the J and the agreement between the two is really impressive so we used the same configuration total energy calculation for the bulk and we get basically the same the same strength for the interruption main message here ok, it's antiferromagnetic coupling in between the planes which we all agree antiferro in plane so this calculation are assuming collinear spins so it's frustrated the system so you cannot get anything better than that but the point is we test also in this way against flour and we encourage with this we can go with larger system and that's what we get for the larger system so the full slab without the graph in and with the graph in so in any case the antiferromagnetic coupling between the layers is preserved but what happens here so actually because of this extra charge at the top of the manganese atoms you have that the spin starts to be not completely in plane but it has a component out of plane but still most of it is let's say you have a projection in plane which is not negligible but when you put the graph in on top the spin they all go let's say out of plane and basically what you have in plane is negligible so this is let's say isquir and also with the picture that we got from spin orbit coupling calculation so what do we have here putting the graph in on top of this surface has a very strong effect also in the magnetic direction in two different approaches overall other commented one can see here is that we have a magnetic softening of all the J parameters and another important information is that the graph in also affect not just the first layer at the interface but also the penultimate layer of the back material so you have to go like two layer down from the interface what happens here there is a change in sign in the J parameters so it goes from antiferos to ferromagnetic coupling so qualitatively the concept is if you want to model the interface you should not stop at the first layer just below the interface you have to go and see what happens a bit further because the interaction is so strong that it also influence like one or two layer below with this I will basically summarize what we have seen what we have seen in this talk so we are using we are designing an interface between grafing and the multiferos material to announce let's say the magnetic properties of grafing and actually we found the substrate influence these magnetic properties so we have charge and spin transfer into the grafing and you can tune it with doping or applied field as you like but also the grafing influence the substrate and we have a change in the easy axis of the substrate we have a magnetic softening of the substrate we have that interaction goes down couple of layer below below the interface we have spin orbit splitings logical insulating nontrivial properties so it's a very rich system very rich physics that you can try to explore by change the ingredients for instance different kind of substrate like different kind of monolayer material and so on and go and calculate the transport properties for these systems because you can see it from the band it's going to be interesting in this I thank the collaborators so there is a large let's say the floor part of the work is done in Yulich FIVOS with the Monte Carlo code and also for the Eisenberg and so on so everybody is discussing and I'm bugging everybody to give their work done and I did all the siesta part of this so and also funding European, German and computing time before I go have an announcement please come to Milan you're organizing the European theoretical spectroscopy conference it's going to be many body theories some connection with chemistry, data science and so on we have a list of invited speakers they are mostly here we are welcoming Milan in September to participate to our let's say b-annual event on theoretical spectroscopy and with this I thank you for your attention