 So I will start to thank all the people involved in the BIM line, especially the BIM line team. So Roland with our assistant engineer, Oria, who is a scientist in charge of the current scattering in station, Alessandro Nicolaou, who is in charge of the Riggs in station and for sure the real people that are working with Eric and Cyril who are postdoc, who are postdoc and PhD student of the BIM line. And something really important, maybe especially we have a delay that we have what we call the research associate and at the BIM line we have quite a lot of them that are somehow involved in the operation on BIM line development, in particular Sorinthus Bayan and Emmanuel Jean from the LCPM in Paris, Jean-Marc Angérôme from Manistinale and Mauricio from in Spain, Paris and the new Monbelle, which is Luc Ortega that is at the LPS laboratory in Sackley. So there was my talk, I will really briefly introduce Resonant Elastici, so scattering because I think everybody knows what we are speaking but I will just introduce the mighty cross-section. Then I will also briefly introduce Soleil and the BIM line and then I will move to the two parts of my talk with the first one is not linked to Crayons but it's simply Resonant scattering and how we can programmatic carry it in Synfilm and you can do it at ultrafast time scale. And then I will move to holography and tichography at the BIM line, what we are doing now and what we want to do in the future and this will be a link with the first part and then from perspective for future. So for the current scatter, I mean the elastic scattering cross-section, this is a one non-sinter paper, a famous paper in the early, I mean in 88 from Anna, non-co-workers, that if you only consider the depolar transition and you expand the, I mean you write the internal addressing, so you can write the Resonant scattering cross-section in three terms. The first term is a so-called Resonance, I mean Thompson plus the Resonant-Charles scattering term which is the one that most people use in the Ardix range and there was two other terms that are proportional to M which is the magnetic moment when you are at the alleged, for example, of the 3d meters and these two terms are, the first term is proportional to M and it has a dependence in polarization when you have an incoming and outgoing polarization and the second term is proportional to M squared somehow so it's can give information on anti-formatic compounds and these are also different polarization dependents. And while it's interesting to do this in the third X-ray range, despite that we have a quite limited eval sphere, but it's because you can probe the alleged of 3d meters on the edge of flowers which are the magnetic atoms and when you sit at Resonance you have an element and site and balance specificity. It's easy in the third X-ray range to produce polarized X-rays and using circular X-rays you can probe the magnetic orbital spin contribution and you can resolve the structure. The third X-ray range is perfectly suitable because you have a huge sensitivity to let's say monolayer or very thin materials down to 1 nanometers and the wavelength it's range of 1 nanometers and then if you add the current radiation that the most important part for today is that you can recover the image of your sample and you can do it time-resolve depending on the source you are using. That will be all the first things. Now I will just briefly introduce Soleil. Soleil is a French national facility which is covered by a wide energy range from the terrarium up to let's say a very hard X-ray range because we have one terrarium beam line and we have a beam line that goes down for a few tens of kilovolts and you can see that if you look at the beam lines as a function of energy we have let's say let's say roughly 50% of the beam line in the soft in the so-called third X-ray range and 50% of them in the hard X-ray range with some particular beam line like the Lucia and Sirius that are focused on the tender X-ray range. Since 2015 we reached the goal that was to store 500 milliamps in the storage ring and to have I don't remember now we have 29 beam line in operation and that's one of the highest current for last generation machine and then since we reached the original goal of the of the machine now we in that time we start for a possible discussion for an upgrade and particularly to do like Max for a diffraction-limited storage ring that something is still ongoing because we are quite slow in France in the process and regarding the beam line so it's a soft X-ray beam line we cover right energy run from 50 electron volts a very soft up to 1.7 kilo electron volts with we try to maximize the flux and to have on paper above 510 to the 12 photon per second with the 10,000 reserving power and we use two undulators as a source because the idea was to maximize the flux on the purity of the polarization in a small spot so we only use the first harmonic of the undulator for doing this. In terms of real flux we have another one paper that the beam line flux measurement adds a sample on the recent station which is a more demanding experiment for flux and basically up to 1 kilo electron volts we can see that we are in excess of 10 to the 12 photon per second and then at higher energy the flux go down quite seriously. In terms of currents, in terms of currents that the measurement we done using a theory of all really in using focus ion beam in an opaque membrane and we make it in a way that they are non-redonant and then if you take this kind of sample and you you make just a Fourier transform of the of this object you have this hologram and then if you make a Fourier transform back you should recover something like this. Now if you put this sample on the real beam line with the full acceptance of the beam line this is the scattering pattern we measure and you see it somehow look like the theoretical one and if we do a Fourier transform back you see that we recover a lot of point in the vertical direction because the source of sole is quite narrow in the vertical direction but you can see clearly that in the horizontal direction we don't have a lot of this diffraction spot indicating that the current length in the horizontal direction is really worse than in the vertical direction and if we repeat the same by reducing the acceptance of the beam line at the expense of the two order of magnitude in photon flux you see that we improve quite a lot in terms of scattering pattern and then do the Fourier transform back you see that we we have a current length that is we estimate around 25 around 35 micrometers in both directions but this is a we have a price to pay and then we come back at the end on this on this on this part. So now the main first part is to that's a study that we started a few years ago and there is something in that's a probability in sinfim and this collaboration between the beam line at sole and the group of Vincent Crois and Nicolas Rennes at Senora Stales and I put in right the real people that were that were joining show was a share postdoc William was a PhD Eric was also a share postdoc and Cyril uh this is a share student and Yanis and Matthew that just started his PhD a few months ago and then we have also to manage the funds to do this work so I may maybe most of you are not familiar with the magnetic magnetism so I will just briefly try to work out to to recall what are the term and if you look at the magnetic system you have thorough magnetic interaction that you have to account the first one and the most important one is the so-called exchange energy and this energy term can be right in this kind of form and you see from simple mathematics that in fact this term for the matrix the spin to be parallel and to be I mean so to have a formatic alignment you have also a deep polar interaction which is a next question on this point and then this deep polar interaction is somehow the field that expands from the atom and you have to close the field as usual in the magnet and this induce that this neighboring magnetic moment want to be uh anti-parallel and in fact in a real system uh in the magnetic state you have a competition between these two terms and this leads to the creation of magnetic domains and the size of the magnetic domains is directly related to the term of I mean the competition of these two these two terms in the system and then if you have you have you look a bit I mean smaller energy term you have the interface anisotropy term which is something that has been discovered a long time ago that when you send on the system like for example you take a cobalt layer and when you send on the system you you move from from an in-plane for a thick cobalt layer to out of plane anisotropy and then you spin move out of plane anisotropy and the last term it's just existing in the in the Burke this jerry-schimoy interaction and it's worked from Albert a few years ago so that in fact when you you couple two magnetic two materials like a magnetic thin layer to a large spin orbit coping material like a platinum, tungsten or edgium you you create at the interface between these two layers an equivalent of a jerry-schimoy interaction term it's so now it's called nowadays a interfacial that jerry-schimoy interaction but that's the the the the way you express this term is the same and you look from the mathematics that this term wants that the spin start to bend and then you have a bending of the spin neighboring spin around a so-called jerry-schimoy vector and that's because that's the source of the chirality in in in such a materials so if you increase this dmi term in fact you i mean if you have no dmi term in the system you have a what's this closed domain walls are as blood type and those if you move from from out of plane a magnetic moment to i mean upwards and downwards you you have somehow to to move from this part to this one in the in the two domain and you have the so-called domain wall and in most of the bulk system on most of six uh films you have a block type domain walls that your magnetization rotates uh in the direction that you have a domain wall which is somehow the spin rotates and make a rotation in plane in this direction if you increase this dmi term you move through a nail wall which are much less natural in nature that you rotate with a winding which is you rotate in this direction directly you see that you have a what sense of rotation which is autogonal and then if you increase it even bigger that's your dmi term start to be a dominant term or one of the most dominant and you you end up with no no no no no more domain but with the perfect spin spiral in the system that's something you can create uh in artificially in in symptom so just to briefly show you uh and why this is important i mean because this uh when you have this kind of system when you you have you have this kind of nail tab domain wall you can stabilize by just applying external field uh uh transform the domain wall which is a one-day object to a 2d object with the so-called skirmjohns and that's such a potential for let's say application and uh but there is because you can move it and you can store information very very in a very small uh i mean surface and you have also you can uh enjoying this and next in orbital on chiral domain and that's a really uh something that in it's been trying a lot of laboratory i'm working in nowadays but then you need to to understand the study this you need to access to the full magnetic distribution and in particular is the distribution of the magnetization at the domain walls that's something we study uh in in our system so as i say we we typically grow a multi-layer on this time so we stack a magnetic layer very thin it's typically all 0.8 nanometers and we put a spinobic coupling uh matter i mean is matter with high spinobic coupling like platinum on the iridium and we deliberately choose to to to not have the same at the 20th i mean the 20th of the cobalt layer and then when you go that but you can have magnetic domains as on a demarc state like a maze domain or you can if you demarc the system with the in-plane magnetic field you you can create perfectly a great a grating of magnetic domains that it's uh depending and that is something uh now we are looking and what we do in our case we come from a from uh we take soft x-ray x-rays and we we take the put the wavelength at the cobalt l3 edge and we just send the light on the sample and we look at the specular i mean the scattering intensity using a 2d detectors and basically uh if you take a i mean like a maze domain uh system uh it's like a border it's completely disordered but you have a current lens you have a typical lens and that give you a ring of scattering intensity like in polar diffraction and if you reverse the left and right in coming polarization you see that you have a strength change in the distribution along the the ring uh intensity distribution along the ring and if you do the different between these two you can see that you have a huge difference in what we call a circular diacrisis in scattering intensity and in that case as you can see it's a french flag so you can also do the same with the stripy domain so you don't have any more ring you have a bright spot but like here you have the first order the third order and everything but the the idea remain the same but you see a nice diacrisis and I would try to show you uh that this diacrisis is directly related to the to the domain wall chirality so if you take a an azimuthal angle of this of this of this diacrisis and you you plot this as a function of the azimuthal angle you see that you have a like a almost a sine function and if you you take the paper from anon and you calculate the scattering intensity uh and in the bond approximation and to make it simple and you try to modelize uh what is the nail domain wall so I show you you you up from up to down domain and you have to move in this direction and you can simply make a modelization of this domain wall by uh saccharid and winding because as you know in the fraction we it's very easy to calculate this cosine and sine function and if you do the simulation just like you obtain this if you take a clock or as nail wall you have this kind of simulation if you take a clock who has block domain walls the one which the block is on that direction you see that we have a phase shift of 90 degrees so it's basically just by a simple comparison that in your experiment we can directly claim that we have a nail type domain wall in the system and it's a clockwise uh his winding sense is clockwise if this is because we have a iridium at the bottom and platinum at the top if we reverse the sense I mean that you can do with the simulation for the opposite if we reverse the stacking in your in your system and we start with the platinum before putting the cobalt and iridium at the top you see that experimentally we reverse the sign of the direct reason which means that in that sense by just changing the in fact changing the ordering of your gross you change the sign of the dmi interaction and so you reverse uh the sense of rotation of your domain walls and that is directly seen in a in a scattering experiment so that's the first result uh but now I just to show you that if you do it for a time 5 multi-layer that's it's perfectly reproducible you reverse the dmi interaction you reverse the back you do it for time 10th repetition it's the same and then you increase the repetition is not working anymore and that was a big surprise for us and unfortunately we start with this system so we spend quite a lot of time to understand what happened and this is very surprising in fact what we realize that what happened is that if you make me chromatic simulation on the system for times a 5 repetition period you see that you have a cycling rotation which is constant in the 5 cobalt layer if you do it for time 10 it's almost the same but you see that at the top something happened and then if you do it for the 20 repetition you see that you have a sense of rotation at the bottom you arrive at some points that you have the spin in the domain walls are pointing orthogonal to the one here and then you rotate in the other sense that in fact what we call it's a do is simply due uh because you have uh when you yeah and and as you we use soft x-rays we many probe the top it means that we always have this uh independently of the stacking layer and it's because in fact when you start to have a sample which is thick enough the deeper our field start to be uh start to play a role and you need to cloth the field and this induce uh what's something what we call uh an hybrid domain walls and uh in fact we have an hybrid charity with one charity at the top uh the opposite uh at the bottom and uh in the middle we have a block uh tight domain wall and that's something what's completely unknown and unexpected at the end and that's something that uh we discover using soft x-ray scattering and then and then I don't spend too much time but that's something you can control by applying in playing field along the block part you can move and expand or vanish the block part so somehow you can externally control the the clarity or the type of clarity of your system just by applying an extent that I have no time today to to show you this and that I said you can do it time resolved and that's something we do so recent more recently we we take our sample and then we go to free turn laser and or case the Fermi a free turn laser in three years and you repeat exactly the same experiment you you use uh x-rays uh on the sample and you measure the scattering intensity and you pump with an infrared laser in this case uh the x-rays uh that we are looking at the funny edge uh it's a cobalt m2 free edge it's not a very high resonance uh I mean it's quite a smooth edge but you can recover exactly the same information that you have a magnetic ring and then if you do the difference between left and right position you have a a dichroism if you do that as a function of time uh you point you demig the sense so if we look at the intensity of the magnetic ring uh when you start to pound the system at t equal zero you demig the system and then you're the magnetic intensity scattering intensity uh vanish a lot and recover if you're doing at the longer time scale so you zoom and you're doing up to 900 because when you see that in fact the blue curve which is a dichroism recover faster than the sun which is the magnetic moment let's say the magnetic scattering intensity is basically this shows that the domain will magnetization recover much faster than the main domain magnetization in the main domain and that's something we we measure perfectly but that's something has been also measured and published before us by the group of uh christian roots in zingen and matrix chlorine in mines but we just confirm the same experiment and in your case we we focus on the ultrafast time scale and now if we plot the asymmetry ratio uh well so we divide the difference between the sum uh its magnetic scattering intensity and we plot it as a function of time we have this white and black dots and we in the paper we have a third experiment so we repeat several times the experiment to to be sure and you see that uh the asymmetry ratio is one uh let's say which is normalized before the pumping and start to deviate from one at the in the first two picosecond and recover a wrong one at uh let's say at five picoseconds another idea was to explain this deviation and uh so we we we make several models in the simulation so let's say that you demag this the you demagnetize the system in the same amount in the domain and the domain manual that all the spin I mean you you reduce the magnetization in both at the same time uh simultaneously this is a simulation where you expect a straight line going to one and that's not what we observe if now you consider that something we have done few years ago uh uh flash and it's a paper from basian for uh that you you you have in the same time you demag the system we we we see a domain or expansion a slightly domain or expansion and if you do this simulation in fact uh just simply you consider that as we measured before you have a slight expansion of the main wall in the first few picosecond you see that the asymmetry ratio uh go above one and this is typically because it's easy to understand because the asymmetry ratio just give you the ratio between the moment the number of moments that are in the domain wall versus the moment that are in the domain it's it's a direct measurement of the ratio between the spin in the domain wall and in the in the domain and that is something that's not upon and then the uh uh we we we try to modelize what happened if uh we induce a spin talk because when you demag what you you send a huge power uh infrared laser and you create a lot of hot electrons you will take a lot of electrons from the Fermi level and these electrons are spin polarized and they they they they travel in the in the system and and but they are spin polarized with a direction from the domain and when they cross the the domain wall so there you have a spin in the domain which is up and it's ejected and moved through the domain wall where the spin is in plane and then you have what we call a torque the domain walls the spin in the normal adducer torque on the on the on the on the spin of the electrons that are extra that's something that is known uh from from quite a long time from a bit of work from nishelver and oh and that's something we try to account and this is the the red the let's say the green curve which is just below the red one and perfectly you can uh reproduce the experimental data uh including this spin torque uh due to the hot electrons and then if you include the manual expansion that the red uh you can also go a bit slightly above one if you believe that these two points are real that's basically what i say is that when when you you you demand the system you you have a hot electrons in the system that are spin polarized and they come from the domain mainly because the domain are much bigger than the domain wall and they induce a torque on the spin in the domain walls and then you see that just with a tilt angle or precision angle of the spin up to eight degrees we can reproduce the symmetry ratio that we experimentally measure so it's clearly that we show that during the first two because the ground after optical optical pumping we have a change of the domain wall charity and we in fact we change from a pure nail to a transit block nail block and it's simply because the spin from these domain that are up travel in this direction and just a torque in this direction but the spin that are the electrons i mean the spin of the electron that are eject here from the down domain are down and they use an opposite torque and it's mean that your uh your domain walls in fact as a as a precession that are opposite in the two direction so we have a block nail block domain walls that's something uh we discover and that's something now we we would like to to to study using current scattering experiments so just just to recall uh actually and now i will move to uh the what we are doing in the beam line and what we want to do especially in the view of today's this this charity in the future but the first experiment we do is a ectree holography so it's basically it's a pionic paper from from stefan icebeat published quite a long time ago you take an object which is your sample let's say basically you you let a hole in your sample you drill a hole in your sample and you have a reference bin and interfere with the detector position and you have another graph just simple mathematics make that you can uh make a few a transform of your hologram intensity and what you obtain at the end is basically three i mean four uh things the first one is a bright spot at the middle very small which is a convolution of the the reference by itself which you you make the reference very small or much smaller than your object in the middle also of the of the Fourier transform you have a the object by itself and you see that the size of the object is twice the original size and on the side you have the object converted by the reference on the reference committed by the object i mean it's complex conjugate and in fact this is an image of your object seen with the resolution given by the size of your reference or in practice if you do this uh we can do this in uh so in the first and station so we can do this in transmission you know this is a comet and station which is currently installed at the beam line and i put the paper uh from oria that described the experiment and station and we have also a scattering and station uh that we try to to do this kind of experiment but in scattering geometry and i will show the few results we have nowadays but in terms of in and station so we have a and station that we can do Fourier transform holography and tucography and i will show you some example we can have either we have this master sample approach uh whatever we can teach the sample uh if you want to i mean if you put the sample perpendicular to the beam you put the auto plane uh magnetic moment if you tilt the sample you can have a sensibility to in-plane magnetic moment also we can apply the magnetic field up to 0.9 tesla in any direction uh in the resistive plane we can cool down the sample uh from 30 k to higher than temperature and we have a special resolution which is typically 20 nanometers and uh this is a view of the magnet so here it's uh typically it's not working okay it's funny it was working in the test so we we use permanent magnets that we can uh change the distance between the magnets uh to change the magnetic field and you can rotate the magnet each magnet is rotating along its direction to change the the direction of the magnetization that's something you can see that we as a for example a system a classical system with magnetic domains when you start from saturation and you reduce the field you see the creation of domains and can do the same for one of our sample uh which are uh which have a chiral nail type domain and you can see that we can start we move from domain to add some fit some small dots that are the the skemions I speak before and uh so this typical example from from users that come from the v9 so the group of peter atlant for example it's a bulk materials the syndrome using like a transmission electron microscope uh techniques the syndrome the bulk system and you can see that other functional field which can see a nice skemion lattice and this you also some samples are removed depending the field from a nickel state to a the skemion lattice phase and the conical phase and this is really made using current scattering holography this is a reconstruction from the hologram uh the one I show you uh here it's you see the small dot and we show you a bit more in detail but uh it can be as small as 20 nanometers and you are not limited to to magnetism I mean you can also as in spectroscopy you can also probe the metal interstate of transition for example because uh in the in our dichroism you can have an access and it is an example of a phase transition view to and don't make the group of my golden uh using also holography but using linear polarization we can do it time result at the beam line so this is a for example a vortex uh so you have a nano structure of one microns or three microns by three microns that is a lando type uh the vortex structure you have in plane magnetization which are in in this direction here in the opposite direction and here in down and up so you have a close flux of the magnetization and when you pump it with a electric field you send the pulse current and you see that you can move the vortex which is something so you have a rotation a duration of the vortex core you can follow on measure using holography and you can also probe on the in plane magnetization and in fact the in plane magnetization uh sorry the auto plane because the moment are in plane the auto plane if you really align the sample perpendicular to I mean to the beam you see that in fact you see a cross here because in the domain walls between the the four part you have an in plane component and you see I mean it's hard to see but in the supplementary matters of the paper you see that we we you have a change in the color along the domain wall during the precession and that's something you can reproduce in the simulation in fact you have a spin wave when you start to make the the core processing you create a spin wave in the domain wall that propagates uh from the corner to have the center and that's something we measure in this time result of holography experiment and now I will show you the new development because we are we are doing now that's the main point and the the rollback in the south x-rays is the detectors I mean the the detectors that usually people use is a ccd detectors and they have a very long reading time it's typically in our case we we typically measure on about millisecond acquisition time and we have a five-second reading time with the detectors that's something is very painful uh so it's it's limited quite a lot and smack that to make a good image from an holographic experiment or tico rafi I mean you have to to let's accumulate during one hour but 99 percent of the beam is lost just by reading the ccd so we develop new CMOS detectors so basically we don't develop we buy a cheap CMOS chip from a chinese directly from a chinese company and we adapt it to make it uh u hv compatible and and we modify the orientation to uh this is a new detector chamber that in fighting this new detector chamber we can have uh three detectors and we can select the one we want so here we have the classical princeton ccd camera here is a new CMOS and we have a search space for time result uh delay line in cp detectors uh for for that we we will install for time result experiment so this uh detectors has been developed in collaboration with max four during the max four solid collaboration the first prototype now now it's uh commercially available through the axis photonic company based in montreal so they develop and they make it a bit more professional uh based on on our design uh and we fully characterize the detectors and we also show that this detector can be used at the free electron laser and especially at fermi we can pick uh because it's a fast detector that can run up to 50 hertz and fermi is running at 50 hertz we can use the CMOS to pick pulse by pulse the the fell intensity and this is a typical measurement we have done uh to show that we can reproduce the same uh quality of data that we have in ccd with a gain of time acquisition time again a factor of 30 in term of acquisition time and so fermi when we do this test experiment we just buy one now it's currently installed in the institution we also take a try to make a comparison the quality of the image because the CMOS and the ccd doesn't have exactly the same dynamic the same noise and there's slightly two different detector and this is a comparison on the same sample so it's one of our samples you can see this bubble like a scan of like objects and this is a image this is a hologram we have and the the die-cook hologram uh you can see here we change the polarization and this is an image reconstructed with the with the CMOS and this has been done in collaboration with all better on philx but now from agd and this is the image we get from the CMOS and this is the same image uh we got uh using the classical ccd detector so basically maybe the contrast is a bit bigger uh better in the ccd but the main difference is that for this image we it's a total acquisition time of 50 minutes not here we have eight minutes so we really gain a lot in in time and i will show you that it could be useful for for photographing but we do the same for our sample this is a scamyon lattice measure of that uh you know multi-layer this is a scattering pattern you see that we have a six fold symmetry characteristic from a perfect scheme almost perfect scheme of lattice and we can image it as a function of field i show you and then uh the beauty that in fact by carefully changing the field we can uh reduce the size of the scamyons onto 20 nanometers which is basically the experiment the resolution of our experiment at the time which means that we can stabilize magnetic object at very with a very small size which is very important for potential for future application that's something we are not doing time so now we i will just show you the alternative to to not do this kind of experiment in transmission because it's very limiting the soft x-ray range because the soft x-rays uh you need very thin system sample so we try to do it in scattering geometry so we use this yama and station at the beam line and we make friendly samples so we make uh so we have the sample here and we put in front uh in top a mask for holography and this is a membrane where you can see that using the trick in the field we can uh cut a part of the membrane and by playing with the field in the direction we can move up the membrane like leaf of the membrane and just playing a bit and then we make the reference and object all for holography and then we put this uh mask holography right on the sample and then we send the beam in between the two and we put the ccd detectors at 90 degrees so that's i mean if i say that sounds easy in fact is i think we use 10 or i don't know a number of nanometers because then you have to put the sample at the right position then you put the mask at the right angle and then move it down to the sample as close as possible to because your current length is not infinite so you your mask has to be as close as possible to the sample and that's something we try and then we get a hologram in scattering intensity and then we try to make the reconstruction so we use a test sample which was a nano micrometer 90 dots the system and this is the image we get at the end so we are far from being able to do a magnetic holography in scattering geometric but we are working on and uh we also try to do uh i don't show you i have no time today but uh we also try you we can also do it in ticography we just uh have a small beam we put a pinhole in our case and we move the sample in front of the beam and this is a typical hologram uh this is a square tuned by two nanometers square a micrometer square and you see here the vertical dots are the magnetic domains so typically nowadays we have a 15 nanometer special resolution just limited because we have a quite a big beam but it's was using the ccd it was taking uh let's say days before getting this image now with the CMOS we can uh expect to to do it uh in continuous scan and that's something now the CMOS has been installed for example in the states and station at soleil and they do quite a regularly ticographic experiment as this takes them we will do the same uh in the next in the near future the big difference in our case that we focus on magnetic materials and we want to keep uh the capacity to apply a strong magnetic field electric field or uh even do do timers of experiments so that's something uh we really build a dedicated instation for magnetic materials uh now i hope that in the future of the soleil of great uh we can uh move to we can gain a lot that's my dream and then we can move from from doing such experiments ticography or holography in transmission it's easy it's easy to do i mean quite easy to do now with the actual source but to do it in scattering geometry that will open the way to make a new experiment and particularly for example in my case i'm interesting to image the magnetic clarity not just do a magnetic scattering or secular bikerism scattering i just want to have an image of the clarity and that's something you can directly do it in scattering but you need a much higher current flux and that's something we start to work on that so our machine group is working for a new two meters long in vacuum appetite on the laters that we can gain a lot of flux at the source and uh so basically we can uh with the on the laters we can up to gain a factor 10 in flux in terms of current flux at i take a lot easy we will gain a factor on dot just from the source point of view if we it will be even more easy to focus because now as i show you in holography we have a let's say a sample which is typically a few micrometers the field of view or samples but your beam is 50 micrometers so if we're we can easily focus the beam much easier and with a new source which means that you also gain useful photons and we can also expect a factor on dot and then from the new peak new on the laters gain and if we change our monochromator because for for holography or for magnetic scattering a transition method edge you we don't need this 10 000 rhetoric power and practice and uh so the idea would be that we gain basically if we do it correctly uh five other of magnitude in current useful for photon flux it's in the current photon flux that's something for me uh i hope that in the near future this will open the the the field of uh resonant elastic or resonant magnetic scattering ticography in a scattering geometry to to be able to probe all the system we want with the resolution which will be n-piper wavelength limit of the typically few few nano few 10 nanometers uh we are i mean we can also have a bulk surface sensitive just just by changing the incident angle on the sample and that's something it's really i mean you can have an access to all the relevant parameter for connotes matter like the structure if you have nano structure of materials the magnetic properties that i show you electrical like for vo2 and even the orbital ordering in the system and that's something which is very important in scattering ticography in scattering geometry in in opposite for example to stick same uh that that we don't have any constraint around the sample so basically you have your sample you have your beam that is focused and your detectors it's that's a few tens of centimeters away uh from the sample which means that you can build a sample environment dedicated to what you you want and in our case is uh i-83 field and the pump probe experiment like in pride out there and that's something i hope to develop uh that we can develop if we have the money to build an upgrade of the of soleil that's really something uh i hope and i just finish uh my turn by a just a short advertisement that we are organizing the resin elastic x-ray scattering experiment end of june uh in paris downtown and the the resist you cannot i mean the registration is still open uh until next week so you are all welcome and just uh i finish my presentation thank you so much uh nikola for this and now the session is open for questions so please raise your hand or you can type your question in the chat if you like and i can read it to nikola so i have a few questions about um the technical aspect uh you work with the with the cryostat is it a helium flow yeah it's a standard genesis cryostat so i was i was really curious to see how to know how this affects stability and how stability is important for your experiment but the the the thing that for holography the beauty is that if you have the the mask on the same sample that you i mean basically what we do in this kind of experiment you have a membrane we put the opaque layer on the side in three bits and then you have your sample deposited on the other side so both the hologram the holography mask and the sample are stick together and your beam is bigger than than the than the i mean the sample size i mean the useful sample size so even if we could down even if the be the sample move a bit in the beam that doesn't change anything that that's a big advantage of holography i mean for for temperature dependence or for time-resolved experiment uh it's holography is very interesting i mean for just make an image it's not pretty useful holography i mean the jugs go to sticks and that's that's fine but for time-resolved or temperature or i mean or if you pump your sound system with higher current holography it's much much much easier so i was actually you were showing the comparison of the resolution that you get with the ticography you're limited to 50 nanometers and with the with the footage transform a lot of you get down to a pixel size of 10 nanometers so i was wondering actually if the your limitation is not the stability rather than i mean what do you mean by beam size intensity the point that in in our case in our case i mean both i mean in our case when we do ticography we we don't have focus element we just put a pinhole of alpha i mean 500 nanometers basically which means that we have a a beam on the sample arriving this is 500 nanometers plus this divergence because it's scattered on the pinhole so we have a right i mean a big beam that sense so the ticography doesn't make me oracle at the end that's the first things and then when we do this ticography experiment as we have a pinhole the small pinhole we lose a lot of photons in addition to make it current you lose a lot of photons and then we have acquisition times that are a bit long and then we have stability issue so i don't know who is the dominant one as there have been sizes stability but both i mean most likely both of them so we are thinking to buy a focusing element not a zone plate because we want to do spectroscopy but this new um capillary optic so that's something we will put to rock over the flux and be able to to make image much faster so which beam size are you are you looking for i think a few hundred nanometers no more no i mean because then the idea will be to have i don't know 100 200 nanometers a very clean beam and then the ticography will help you to go down special resolution because i don't want to again i don't want that my optical element is closed to the sample i want to put it as far away from the sample yeah i i actually know that um capillary for soft x-rays um they're worse than specified so i raise a flag here yeah there is a question in the chat from Richard uh he says i may have missed it but when all the upgrade for the beam line could be completed i think increased flux becomes available to users uh good point the for the new undulators we we are building i mean they build i mean it's not me i mean the machine people they build uh the prototype nowadays and we expect to install it in the ring before the upgrade for sure that's the thing so we'll gain a factor 10 in flux uh in general current or not and the upgrade of soleil is officially planned to be finished in 28 i think 26 i don't remember but it's not yet officially finance so i still have no discussion so we have the scientific case down and now we are working on the technical design phase but we don't have any guarantee from from from the government that we will have the money so okay so all this they're all they're all within a large upgrade yeah yeah yes yeah the new beam lines and then because then we need to make a new end station everything would be in depth in the upgrade okay any questions from the audience please raise your hand or you can just unmute and ask nicole yeah hi nicole um i have a question about the uh metal interior transition in uh vo2 what's the nature of the contrast that you get is it is it linear decorism yeah okay i thought it was a difference of scattering altitude due to the difference of variance electrons maybe or no no linear decor is this question so it's more easy okay thanks nicole i have a i have a question about um using this uh imaging with x-rays and uh let's say standard more standard techniques like an mfm you know it's a bit of a silly question but could you highlight what is the what is in your experience the real gain of using x-rays that's a good point uh if it's just my feeling is nowadays even mfm or i mean there is a new microscope technique which is a nv magnetometry when you have a resolution which is even much better than mfm and the sensibility which is completely incredible i mean i've done an image on on this mid ferrite with basically an anti-firm magnet compound and you see the magnetization pretty well uh so if it's just getting an image of magnetic materials i think we should stop x-rays i mean it's no point uh to make just an image the the only advantage of x-rays that you can do it uh time myself with high field low temperature on all that stuff uh i mean the the main still the remaining advantage of x-rays is the capacity to to to let's say have a perturbation of your system whatever the perturbation is which is quite difficult in mfm i mean applying strong field in mfm is not easy do it time uh temperature dependent or time is always impossible so this the summer environment or your capacity to to perturb the system is the is the really important one and the time resolution is really so so what are what are the technical uh uh necessary uh items let's say technically what does what does a soft x-ray beamline need to be capable of doing microscopy uh you know competitive useful for material science uh for magnetic materials what would what are the ads on what do you mean in the sense well i mean i mean uh you know there are many uh if i had to build a soft x-ray microscopy uh beamline that uh dedicated to magnetic magnetic materials i would certainly need a possibility of applying magnetic field a cryostat um uh in my case we are building uh we are installing a laser for during time as our experiment uh and you need the most highest flux you can a coreon flux you can so which means that uh you don't need you need really to make a that's a problem of all beam line that we have a brand for inelastic scattering which is need uh high energy resolution which is completely useless for for current scattering so if you want to design you need to have a disparate monochromator that just disperse uh north if you can make it a long on you later i mean that's something i'm start to look at uh if you can make a long on the later with a small gap you can even things to use a direct beam the pink beam does the on the later pick can be narrow as few electrodes and that for 3d methods it's enough to separate the l3 and l2 and then you would gain a lot a lot of plus that's something we are looking it's why it's why we are building a prototype for in vacuum long on the laters to to to narrow the on the later gap uh the on the later pick narrow it enough that it's cover only one edge and that's something that you need to go on fix as much as possible let's hope also so they will approve yeah great any other questions we have a few more minutes for questions from the audience question or comments one other question maybe uh let's thank you this richard sandberg um does the you mentioned some advantages of soft x-ray but you didn't mention elemental selectivity isn't that another advantage being able to do it resonantly yes but i mean i mean yes for sure element selectives is an advantage but if you look for example i mean if you do like like or magnetic multi layers you can do cobalt or cobalt iron ball or whatever but you know all the spin or achieving the system out i mean if you do a cobalt iron ball system what does it mean to look at cobalt or iron i mean the boss moment are aligned in this by exchange so you don't care i mean in some particular compound like uh in fairy magnetic materials that it's very interesting yes thank you guillon yeah i guarantee is actually i get maybe you can tell the public how you get depth information information in the sir direction wonderful technique yeah you have to duration as you change the scattering angle as you know and there is something i start to to work i mean to sink on that in the soft x-ray range you eventually so curves that uh in fact you should have some 3d information in the scattering pattern itself that's something that is not i mean maybe vasan can comment but something is not a count in the in the software for for psychography or construction because you consider i mean they are coming from from addicts race and your sphere is flat in your detectors but it's not too anymore in the soft x-ray range which means that you you have even without changing the scattering and your some 3d information already why is that accepting because i where does that come from if it doesn't change the angle because you have a sphere is curved so it cannot be considered as flat on your detectors so then you have a 3d information if you are able to account that in the in the algorithm that's another story but that's yeah this has been demonstrated for the structure by uh yeah sure sure yeah yeah but in the soft the main problem is not only that you i mean it's your contours eval sphere but uh kinematic approximation will not work also it's more a bit painful to implement it but we are working on that i mean we know we are almost finishing a software development in in collaboration with uh samu fluids from from san diego university and now we are able to calculate fully the scattering pattern with dynamism dynamical and approximation including roughness and therefore on the way so we are on the way to that that's a long way it's a long way okay thanks always in time gilm sorry you are always in been time not always okay so i i would like to stop a second just to say thank you to nicolas and everybody was joined today to listen and to contribute and now we are free to stay here and to chat this is also an opportunity to to continue uh conversation informally thank you so much everyone so gilm if you want to go back to the conversation with nicole you are welcome yeah i don't know what i'm talking about anymore just want to involve anyone everyone thanks a lot thank you thank you nicole yes it was very very very nice you come on unmute yourself you can show up or do you want me to read this do you observe the beam damage or any beam induced effect with soft x-rays ask theory okay uh on the materials we are studying not not really beam damage but we crack carbon at the surface of the sample i mean if you take the outside i mean we are working i don't show any any result but we are working a lot on this material for example and then when you take the sample out of the beam it's black so you don't i don't think we change the materials but we we we crack carbon uh on the surface that's clear uh i think we are not in the limit yet of damaging the sample i mean this kind of magnetic materials uh i mean we do do did some experiment on polymer on uh we also had some bulges that tried to come and it was really funny i mean we drill all is we drill all the sample quite immediately and that's something i see in the free electron laser so in the free electron laser experiment that i show you an example we spend most of the time to be really careful to not destroy the system so basically when we do this time-resolved experiment uh on the corality if you pump either by the free electron laser beam or the infrared too much you change the roughness at the interface until you change the dmi interaction at the interface and that's something we really spend a lot of time so in the free electron laser you definitely you can burn the sample in a synchrotron i think we are a bit not yet yeah i would say wait until the upgrade no that's that's something we have to and the focusing you know i'm a nanometer actually i was curious i was curious to to know how can you distinguish from scurremians to magnetic bubbles or other topological states with the 2d uh basically the first thing is the size okay and in the the the magnetic bubbles that are known since uh long time ago i mean they are big but the scurremians in your system what they we call these scurremians is first because it's a nail type domain wall first and the second time is because we can see that the size of these objects are 20 nanometers or a bit below and that's something is not classical bubble also the uh hexagonals like the short range ordering is is a sign of scurremians and bulk material is it's known that you have a six-fold symmetry in the scurremium lattice in the in the artificial system that the one we are studying i mean i show you one example when it was very one of the very good sample we i mean my colleague from thales can can grow i mean most of the time we don't stabilize the skin on lattice i mean on the beam size i mean locally you have this six-fold symmetry but on the long range usually it's not not perfect so i'll show you the the best best result we have yes the six-fold symmetry is also an indication of the okay well i i have the feeling that the conversation is kind of calming down yeah and thank you again for your time and for everyone who has been here good luck for your experiment again thanks thanks enough for organizing this hi yeah thank you thank you for joining see you then