 So I think we can get started. It's a pleasure to introduce Dina Schaefer, and she will be talking today about X-ray photon correlations spectroscopy, how dynamics could be probed using this technique. She will talk a little bit, hopefully, about the achievements and the future outlook. So a little bit about Dina, she received her specialist degree from MEPHI, which is the National Research Nuclear University in Moscow. It's a pretty good university in that. And then she got a PhD from Hamburg, University of Hamburg, where she worked on coherent X-ray scattering and application to physics. She also did a lot of work on X-ray photon correlation spectroscopy and some cross-correlation analysis. Then she did a postdoctoral research with Stefan Huskevich at ARGON, MSD. And then she also worked a little bit more on XPCS and also developed some BRCA-herent defective imaging. She is currently a staff scientist at the Advanced Photon Source, ARGON National Laboratory, where she is sitting at the micro diffraction beamline that's 34 IDE. So let's welcome Dina and listen to what she wants to tell us. All right, thanks a lot. Thank you very much for introduction. Thanks a lot for inviting me for this co-work seminar. So I would like today to show you that application of the XPCS and talk about the future of XPCS. So here is outline of my talk. I will start with a short introduction. Then I will say about the XPCS method and maybe showing a bit history. And then I will show you a few examples of application XPCS in my research studies and currently going experiments. And then I will talk about the future of XPCS. So structural heterogeneities in different type of materials, like, for example, nanostructures in complex fluids or defects like dislocations in batteries materials or also grain and grain boundaries. They govern the different phenomena in physics. And probably in that structural heterogeneities, structural and dynamical behaviors. And in my career, I have been probed the soft matter structural heterogeneities as well as in crystalline matter. And in order to probe the structure and dynamics, I mostly utilize the coherent X-ray methods. And as have been mentioned by Dina, the most previous seminars here was dedicated to show the imaging techniques to probe mostly crystalline matter, like break coherent diffractive imaging and diffraction with limited focus beam, like with nanofocus beam, the co-graphy methods. So today I would like to discuss about the dynamical aspects. And I would like to do it by showing you an example of soft matter, which is, can imagine, like, for example, nanoparticles dispersed in a liquid, which would be typical colloidal particles or more complicated nanostructures complexes on the order of nanometers. So typically, the probe, the structure in such disorder systems, we will use a X-ray beam, which would eliminate the sample. And the disorder in the system will produce interference pattern, which we can record it in the far field using 2D detectors. For example, in such a, if we use a coherent X-ray beam, the scattering pattern would consist of such grainy structure. And such a pattern would usually call those a speckle pattern. And the structure of the speckle pattern actually relates to the position of the particles in the real space. It is hard to extract the exact structure. However, if we imagine that our real structure is changing in time, like particles is moving in the water, then that diffraction pattern would also mimic that dynamics. And if we would record the fraction pattern in time and apply certain mass to analyze the diffraction patterns, we can access the dynamics which will be connected to the dynamics in the real space. And just for comparison, why we need coherence, if we would eliminate such a disorder system with incoherent beam, typically our speckles will smeared out. And we will be able to access average structure information like size, spacing, and use the entropy of the particles. So what today I would like to talk about, we can access the dynamics of the system. And X-ray photoconterlation spectroscopy allows us to do this. So by recording that speckle patterns in time, we can construct time correlation function. And the behavior of this time correlation function, which is usually called G2 function, can describe us what is the typical relaxation time in the system, as well as dependence on the wave number would tell us about whether the behavior of that dynamics is a diffusive or collective, or is it Brownian dynamics. So the XPCS method stands along with other brilliant techniques. And it's extend the dynamic light scattering method, which is an allig of the XPCS as well. It's extended to the much smaller scales. And the typically time that currently XPCS can probe, which I will show today, it can go to the microsecond regime with the current coherent flux at the APS source. And a little bit history of XPCS. So since the first speckle pattern was measured at the synchrotron in 1991, the first XPCS experiment have been performed in 1995 and 1996 by this group of people who developed these techniques and extend the X-ray region. So since at times there are, XPCS have been applied to study different types of phenomena in different materials. And here is just a few examples in which materials that techniques was applied to probe the dynamical behavior. So every year we also are trying to upgrade the capabilities of the XPCS. And this happens thanks to the upgraded X-ray sources over the world and also upgrade of the detector capabilities will allow us to bring the XPCS to the faster time scales and probing the dynamics at smaller length scales. So I would like to show you today how the fast detector capabilities is accessing the faster dynamics down to microsecond regime. So this study has just investigated the dynamics in one of the important system, which is a liquid-liquid extraction system, which is used for extracting the used ions from the industry processes. And this extraction of the science typically happened by building this complex nanostructures in liquids. And in this movie, it shows how the ions is sketched by the nanostructures in the liquid. So this is, of course, just a schematic view and how exactly this nanostructures is looking. It's a big question still for the systems. But let's just study the dynamics of that. And the liquid-liquid extraction process I'm showing here schematically on this slide. It is based on the immiscibility of the two liquids. And here you can see the red dots as representing the extracted targets. And so after mixing those two immiscible liquids, there is a formation of the nanostructures in the top part, which partially contains the extracted ions. And this, you can imagine, like water and oil, which would not be mixing together. So however, sometimes depends on the conditions, like concentration and temperature, that second phase may split into two phases. And it will produce the highlight phase and more concentrated ions phase, which is undesired in the application processes. And how we can control that process, we need to study the structure and dynamics and behavior of that second organic phase after the extraction of the ions. And how it does behave while it comes to the splitting point. And this will help us to design a new extraction molecules to make a process with better efficiency. Here is a very simplified schematic phase diagram of that organic phase after the ions have been extracted. And in order to bring that system closer to the critical point where the space splitting happens, we are cooling the system in experiment. And on reaching the critical point in the system, the critical fluctuation is occurring. And the dynamics of that system is supposed to be changing while we're cooling it. So this system is a multi-component system. And the dynamical behavior may also affect by different structures of extractant molecular or different composition. However, we try to be at the critical composition by choosing the extractant fraction, which seems to be very able to find the critical composition of the system. And so this critical point dynamics and structure and dynamical description at the critical point is also important, not only practically, but also important for the fundamentally, for the description of the theory phenomenon. And so we performed the XPCS experiment at the API, at the APS. And here is a parameter of the experiment. So we used the 11KV. And we record the diffraction pattern in the fur field. The beam size was that we were splitting our beam in order to choose a coherent portion of the beam, which is typically beam size. And XPCS experiment is now in the order of several micrometers. So what is important is here, we used the recently developed and already upgraded and commercialized FASTA x-ray camera, which can do exposures of 20 microseconds in a continuous mode. And it can, in general, achieve faster in a burst mode vision. So as you can see here, our typical single diffraction pattern is actually consistent by one photon events. And if this scattering signal is average over the thousands of patterns, then you can see a typical small angle x-ray scattering signal, which is isotropic behavior. So in analysis, first we probe the structural properties by looking in the time average signal and by analyzing everything over the different wave vectors cube, we obtain the intensity profiles. And on this slide, I'm showing that intensity profiles increase, the signal density profiles increase as we go to the critical point. So we can imagine that we have a nanostructure complexes as we go to the critical point. There are more building the kind of aggregation or they are increasing in the size. So that's why we have an increase in our scattering signal. And so we were able to describe that intensity signal at different temperatures within the critical phenomena theory. The critical phenomena theory describes the behavior in the different systems by a set of defined critical exponents, like for example, the phase transition in liquids and for example, the magnetic, ferromagnetic transitions can be described within the one system, but with a set of exponents and this exponents would have a different constants, for example. So in our case, we were able to describe the behavior, static behavior of the system at different temperatures within the 3D ising model. And we found a good agreement with the standard exponents. And also what is interesting that our multi-component system is actually behaving like a binary system. So that there is some parameters that can simplify our system. Like for example, instead of considering all of the five components, there is only two components in effect, that behavior. And by doing such analysis, and we were able to rescale all of the data to the one single master curve, which is typically within the critical phenomena theory. So then for discovering the dynamics in the experiment, in order to obtain the enough statistical averaging, in order to calculate the G2 function, we had to record a lot of sketch patterns in time. So each of the, so the detector camera, the camera was working in the continuous mode and the recording was happening in the way that it was recording the one batch, which would consist of 200,000 patterns in time. Then there is a delay in order to record all of the images and like a single delay. And then it starts working again and record the next batch. So at a different temperature, we have different scattering signals. That's why in order to obtain the G2 function, we need to average over several repetitions. And depends on the scattering signal, this batches number will vary between 200 to 500. And we analyzed the dynamics at the different wave numbers. So you can see with the increasing the wave number, we obtain the much faster decay in the G2 function, which is expected since if you go to the HQ, you probe smaller length scales and dynamics on the smaller length scales would be appeared to be faster. So in order to extract the relaxation time of the system, we fitted this G2 function data with exponential decay. And there are three parameters in this function. It's parameter B, which is called as a baseline, which is usually close to 1. Parameter B, it's called contrast. And this usually depends on the properties of the coherent beam. And in our case, so we had a 10% contrast. And tau is a relaxation time which describes the time of the system at different length scales. And here I present the extracted correlation time as a function of wave number. And it's presented for different temperatures. And you can see that the behavior of that time at different wave numbers and some temperature is different. So as we decrease the temperature and going closer to the critical point, we see the change of that slope in the Q dependence of tau. And this slope is represents the Z effective, which is another critical exponent in the dynamical critical phenomena theory. And we found, again, a good consistent with the critical phenomena theory. We've seen a certain model, which Z effective is changing from 2 to 3 as the system is approaching the critical point. So again, as for static data, we were able to rescale our relaxation time presented here as a reduced diffusivity coefficient to a one curve. And what is remarkable by doing the fast XPCS, probing the microsecond dynamics, which you can see here that we were probing microsecond regime times, we were able to probe the critical behavior, which is on the scale plot would be here. It's presented as a reduced wave number and reduced time. So in the closer to the critical temperature, it's above the X equal 1. So by doing the faster probing timescales, we were able to stay far away from the critical point in terms of the temperature. So we were able to be several degrees away. We already did the critical behavior by probing the faster timescales at their larger Q. So with this, we were able for the first time to describe within the critical phenomena theory that's interesting behavior of the liquid-liquid extraction system. And we showed that with the efficient analysis, and the availability of the high-speed detectors, we are able to probe the microsecond dynamics in fluids. So the particular outlook for this work is we are considering to probe dynamics across different compositions. And we have recently also found that by just changing a different extractant molecular type and going from different ions extraction across the lanthanide series, it is still showing that the system would behave as a two component system, which is very interesting. And we also consider to probe across a whole series of lanthanides. And we're also doing MD simulation, which will help us to connect with the experimental data. And the application of machine learning is also considering in order to breach the timescales and length scales that can be accessible in experiment and in the MD simulation. So I have shown you an example of probing the dynamics in the disorder in the soft matter. And can we do XPCS with a crystalline matter? So and the answer is yes. So we can perform the XPCS reflection type experiment. And I'd like to show you on just a few slides results of recent work, recent experimental results, which we probed in a wide angle XPCS in the conditions in situ experiment on the relaxer's ferroelectrics. So relaxer's ferroelectrics, it's crystalline material with the interesting properties and dependence of the electric constant on a driving frequency external field. And this will span between the Hertz and megahertz regimes. And it also has a remarkable temperature dependence. And this material is usually used a lot in electronics as like, for example, fast memory devices. And in the recent studies, I've been shown that this microscopic properties correlates with a distinct diffuse scattering. And diffuse scattering, it is in a simple way explanation. This is everything which is not coming to the Brex scattering. So near the Brex scattering exists diffuse scattering, which would come from the not regular crystal lattice defects and so on. So however, it is still unknown what is how we can explain the distinct behavior in the diffuse scattering, which cause the different microscopic properties in the relaxer's ferroelectrics. The problem is that static diffuse scattering does not distinguish between several different models was developed to try to describe the behavior of the relaxers. And thus, that's why we think about to probe the dynamical behavior. I have to say that thermoequilibrium XPCS studies have been also performed on different types of ferroelectrics. And it has been shown that the speckle pattern is mostly static. So it does have very slow dynamics. And it again doesn't help us to distinguish between theories. So that's why we think if we will apply external stimulus and probes, the dynamics with XPCS under different conditions of that external stimulus, like, for example, changing the electric field shape and amplitude, it will help to build additional information for, will make additional information for building the theory about the models of the relaxers. And here is just one slide of recent data that we collected. So this is showing the reflection type experiment in a wide angle XPCS. So we probe the diffuse scattering by eliminating with a coherent beam and putting our detector near the 0, 0, 2 scattering. So the material that we used is a classical PMN, relaxers ferroelectric. And here is mentioned the beam line parameters. So the electric field variation in our parameters. And I would like to show you another result presenting for dynamical data is a two-time correlation function. So in the previous result in general, we can also call it the two-time correlation function. Typically, you would do it for the good scattering signal in the XPCS experiment. And here, the T1 and T2 on this plot is showing this X and Y axis. And you can see the bottom of this plot I'm showing that it was applied a sinusoidal signal of electric field. And you can see that the correlation results is following that periodicity of that sinusoidal periodicity. And it is very interesting first observing that we actually used the electric field, which was very small compared to the typically field, which would help to pull the atoms in that materials. They will be on the order of kilovolts per centimeter. And it is very interesting that applying such a small field, we have observed the response on our sign and return memory effect. So we are still working on investigation of that properties. So I just want to show you that it is also possible to do nothing, the disorder soft matter, but also in the crystalline matter. In the following, I would like to discuss the future of the XPCS. So of course, upgraded X-ray sources and faster detectabilities will help to bring and extend XPCS to the faster time scales and probing smaller length scales. And this is because if we XPCS signal as this on the mass that it's a square as a brightness of the source because we have intensity multiplied by intensity in the formula, right? So then it will help us, I think, with the upgrade of APS, at least I know that we still able to probe the nanosecond time scale as far as we will have detectors that we're able to probe that, better to be detectors for the fast averaging and so on. But also we will have an increase in signal. What we need to consider for the future is this we need to consider with a higher brightness. We might have X-ray effect on the sample. Like sometimes in XPCS, you need to consider damaging from X-ray or X-ray can change the dynamics of your system. And we need to also create analysis pipelines, which will require to handle all of this huge amount of data, which will be recorded with a fast detector. And since the flux will be increasing, we will not need to have a fast exposure time. So with upgrade and increase flux, we can also consider that we can crop our beam to the smaller sizes. So we can also focus our beams. And this will help to probe the differentiated dynamics across especially in our system. So decreasing the beam size, we will increase the lateral resolution in our experiments. And however, we need to consider, again, the same problems of preventing the sample damage. So I very much like that there is a going assimilation studies done by Dina here. So that she showed how by probing different regions in such a model system, which has a different flows of particles across the system. And this example can be considered as the liquid jets of the system, which would have a faster in the center of the liquid jet and maybe slower in the sites. So she showed how that in this work, it's shown how the different timescales is found across the system. And also in the previous work, for example, I've been done to probe the magnetic dynamics by Oleg Sporkor. So the typical size is about 10 of microns. If we would focus our beam to probe at different positions of the sample, we may consider to find a different type of dynamics relaxation by separating the dynamics of different domains faster and slower domains. What also needs to be considered in this case, this is a beam position stability in order to probe all these dynamics from the sample spot. Another huge area of applying PCS, of course, is also the pre-electron lasers. And I will not talk into the details, but there have been a lot of studies have been already showing that it's possible to apply XPCS as a conventional way. And as the recent work at the European expel with the nanoparticles, it have been shown the huge amount of correction work for the detectors as well as they have been showing that with increasing number of pulses, the number of pulses would increase the temperature in the system. However, they have been shown that there is no damage to the system by just showing the speed up of the dynamics, which correlates with the diffuse dynamics by analyzing with Einstein-Diffuse coefficient. So there is several different techniques to analyze the dynamics at the pre-electron laser, like split and delay pulse techniques, where the diffraction pattern of a split FEL shot will be recorded in one diffraction pattern. And by wearing the delay of the splitting, we can achieve the different contrast in the diffraction pattern. And by changing the delay, we can see how interest is changing, depends on the dynamics in the system. So the one point that needs to be also considered is another methodology for, for example, by then the XPCS that it feels, since also every shot in the X-fold pulse is not identical, like it's mostly wearing a lot going from weak to strong. So here is the example of weak simulated patterns by Yuri Kau of weak and strong pulses. And we are trying, in this paper, we are showing how the proper analysis and weighted G2 function algorithm showing how to use the noise in the obtained G2 function. So with this, I would like to thank people with who I'm working on different XPCS projects. And I also would like to thank my first supervisor and your first group for introducing me to this XPCS technique. And thank you for your attention. All right. Thank you very much, Dino. Thanks for the extremely amazing talk and insight into XPCS. We're going to welcome everyone right now to ask questions and, if not, some discussion. So the floor is open. Edwin, this is Ian Robinson here. All right, Ian, please go ahead. Can I ask you a question? Yeah, definitely. Sure. Hi, Dino. And thank you for a very clear talk and very nice slides that you sent there. I'm going to bug you for some references in a few minutes. I was curious about the, I think this is the best example to date of the studying of a true critical phenomenon with XPCS. And I was curious about the exponents that you're getting because the normal diffusion exponent is Stokes law. That gives a slope of 2 to the decay time versus Q. And what you're seeing is, well, you're seeing 2 only at the very, very furthest temperature range on the very far right hand side. But most of the time, it comes out as a slope of 3. And I'm wondering what the slope of 3 corresponds to microscopically. What's happening to the diffusion? Is it strongly caged or is it collisional? Is there something long range happening in the process? So this is interesting. So I would say that we definitely can say that it's different from the Brownian dynamics. You mentioned that it's kind of close to 2. So we already even, yeah. So I would think that it's maybe a mixture of the complex dynamics in the system. And so far, as we did this experiment, we have never observed a kind of different exponent going. Like it was kind of 3.02. This was a maximum that we observed. So this is, I think it's a complicated dynamics and mixture. And this is what we're trying to figure out by doing the dynamical simulation. There is a new type of coarse grain type of simulation. And we try to bring our length scales of the simulation box closer to the experimental data. And as far as we see in the simulation, we see that mostly it is growing and the collapse of the micelles. So I would expect that it's a collective dynamics of the monastructure complexes. And again, as I mentioned, some of the people in the liquid extraction system consider that it's a micelles. But actually, it is not fully known yet what is exact structure. It can be also like polymeric sheets, like for example here. They can be not as a micelles, not like circle objects and so on. They can just make a complicated structure there. Yeah, I. It does have a very clear exponent of three. The two is more washy, but just go to two at the, I guess, the highest temperatures. But there's a clear region with three, and that's new. That's a new kind of dynamics that isn't brownie and it's something else. And it may be dynamic light scattering can help find what gives an exponent of three. That's really fascinating. Yeah, in the dynamic light scattering studies, there have been also shown that the behavior in general would be also changing to the Z-effective closer to the three. So this is happening in, this is also pretty good in the critical phenomena theory. But it doesn't exactly explain how and what exactly happening with the system. It is just some, I think it's a mixture really of the faster, slower, and collective movement of the bigger aggregates in the system. OK, but it is a prediction of theory. So that's, then we should look at that theory to understand what it means for the material. A clear behavior that you're seeing. Yeah, thank you. Thank you for the comment. All right, thanks a lot, Ian. Some more questions? Yeah, can I? Hello? Hello. Can I ask you, I mean, I will provocate your question. So for this micelle or microemulsion, what is actually the advantage compared to DLS? Because even microemulsions and binary liquid mixtures have been very well studied by light scattering in the 80s. So what is actually the advantage of using this? Xvc, yes, because the critical point, everything is scaled. So even if you don't go to the smaller size scale, I mean, it's scaled. Yeah, so I mean, the advantages, so I would like to show you here, for example, the DLS study and comparing to our Z-effective exponents that we can probe by. So the typical range of the Q that we access in the X-ray experiment, it's 0.0065 fan-stream inverse, while the maximum Q at the DLS, because of the wavelengths, typically observed is 0.0026. So this is our usually low Q. So by doing at the high Q and probing the dynamics at how the system dynamics is changing on a very high Q, that means that we probe at very low scales, which is at this Qs would be very fast, even if we are not very close to the critical point. So we are probing that critical range, starting off that critical range, already further away from the temperature, so from the critical point. So in the scales, it's several degrees compared to the milli-degrees, which usually needs to come closer to the critical point in the DLS in order to start to probe that critical behavior. That is not true, because critical fluctuation in light scattering you can see tens of degrees away from the critical point. So that is not true. You can observe very far away, because light, the contrast is much, the refractive index fluctuation is much more sensitive than the electron density fluctuation in the case of X-ray. So critical phenomena has been, if you see the old papers, people go to very far away from the critical point, you can still see the growth of the correlation length. And of course, in critical phenomena, when you go close, it is a diverging correlation length. So the relevant size is much larger. So you don't have to go to this smaller scale. So my question is, is there really an advantage compared to DLS in using ECS to study critical phenomena? So in general, if you would probe at the very far of the critical point, and you need to have a bigger scales there, which would mean that DLS might have troubles with multiple scattering in the system. So the X-ray has. The DLS is a travel of multiple scattering near the critical point. But of course, since then, the methods have developed. So there is a way to suppress the multiple scattering. And so you can also go closer. So far, less problem of the damage. So when you study micelles and microemissions, you have problem of X-ray damage. And with the light scattering, OK, you can still have a beam heating effect, which can be calibrated. So I don't see a real advantage of using XPCS to study binary mixture like a critical phenomena. So first comment is that it's very exciting that we found that it's a kind of a binary mixture while the system is like a five component system and so on. So another comment that, of course, every techniques may have some obstacles. And there are many brilliant papers of DLS which shows how to try to develop the additional subtraction of the background subtraction and so on in order to correct the critical exponents and so on to achieve that. But again, by looking in the similar micelle system in the DLS literature, to my knowledge that I have tried to find, I haven't found so many which can show that there is a critical behavior, observation far away from the critical point in such kind of multi-component system. There was a three-component system, maybe two studies that I now kind of remember. But again, they have applied a lot of also background correction in order to correct all of those multiple scattering effects. So I mean, I can send you some literature, because there's a huge literature. And of course, this is also well-established, what is called quasi-binary approximation in multi-component systems. So I mean, microemulsions and all these are multi-component systems. I mean, their critical phenomena is the same. I mean, they all follow icing like a critical exponent, like binary mixture. So this is also known from 80s. So what is called quasi-binary approximation. And I mean, in the worst case, you can have what is called Fischer scaling of the exponent in multi-component mixtures. Not, I mean, it's not so different from icing like critical exponent. So I mean, this is already there in the literature in the 80s. Yeah, yeah. And I mean, as I mentioned, the most DLS literature had deal still in the milli-degree away from the critical point here, just by comparison. And again, there are a lot of background correction applied. Didn't use any background correction in our system. We estimate also the heat and damage and possible x-ray damage to the system, which was neglectable in our case. So that's kind of I think a brilliant study of bringing and probing the faster dynamics of the low Q and you start to see the critical effect already far away from the point. So that's the purpose of why we want to probe the dynamics of the system. And again, with increasing the flux by probing the higher larger Q wavelengths, we are hoping, in general, to probe also ion dynamics that may affect whether the ions inside the mysos may be having separate dynamics inside the myosos volume or those bigger complexes can have their own dynamics. That's the goal to go with upgrades and probing faster dynamics. Thank you. I will be interested to get more literature from you that you mentioned. Thank you very much. We'll appreciate it. Peter has a question. Peter, please go ahead. Yes. Hi, Dina. Hi, Peter. I actually have a similar question to what Narayan mentioned. If you look at your transparency, I think it was 25. So where you had the results from the dynamics, yes. Because there you say that for XPCS, you actually see a larger critical region. Now, I think. I see the critical behavior earlier. Yes, but I think this is not correct. Because, I mean, for example, a long time ago in the 80s, we looked at critical phenomena with protein solutions. And there again, we used DLS. And we saw critical behavior starting at something like 10 to 15 degrees away. From the critical point, everything nicely done. You see ising exponents, 3D ising exponents. I think the point is that you actually are probably not in a good position to look at critical dynamics with XPCS ones to come closer to the critical point. Because what you measure with the quasi-elastic technique is the decay of concentration fluctuations or density fluctuations with wavelengths that correspond to the inverse of your scattering vector. And so if you come closer to the critical point, your critical fluctuations grow and grow. And they sort of grow out of your Q regime that you actually have access to with XPCS. And that will possibly also give rise to some funny Q dependence. Because you see both critical as well as non-critical contributions to your dynamics. And so I would agree with Narayan that if you really want to look at critical dynamics, either in single or multi-component systems, whether it's micromulsions or proteins or single fluids or binary fluids or whatever, at the end of the day, I wouldn't go to XPCS to look at the dynamics. If you are just interested in critical phenomena, because there I think you're much better off doing this with DLS, because this allows you to actually look at the right length scales for critical fluctuations. So we have a DLS in our experiment, but only as a pre-characterized system to kind of shortly determine where is a critical temper. So in order to probe the critical fluctuation there, so I have never at least got it by the pre-characterization. I always got the decay of G2 function, which starts to be affecting by the multiple scatters for particular this liquid limit extraction system. Yeah, but there are ways around this. I mean, either you decrease your scattering volume, you can use capillary. Once you go towards critical phenomena of so much intensity that even reflexes from walls don't matter. So you can go to small capillaries or you can use these cross-correlation techniques that allow you to suppress multiple scattering. But the point is the following. I mean, again, if I refer back to the experiments we did with proteins, in this case, we combined DLS and neutron spin echo as the neutron-prase elastic technique. If you go and look at it on the length scales of neutron spin echo, you see no contribution from critical phenomena whatsoever, because you only see diffusive dynamics from the particles on nearest-neighbour length scales. If you want to pick up critical phenomena, you have to go to DLS and then you see that dynamics is dominated by critical slowing down. And in your XPCS, you're somewhere in between, so I would still expect you to see some non-critical dynamics, diffusive dynamics, but you get a mixture, because initially, you're probably good because you see... Yes, but the problem is your correlation links, the dynamic correlation links move sort of out of your scattering vector length scale. And so you're no longer probing the right decay of fluctuations because you're out of the critical length scale. The critical fluctuation would happen in the whole system, as far as it's there, because you have always mixture. And if you go, the critical behaviour here is above the one, and if you look here as close, so the colour representing blue is, as you come closer to the critical point, so you see that critical component much stronger here in the X more than one, when you would probe that higher Q with a faster length scale, that you see that critical component already there. So this is... Is this cover something? Yes, yes. But again, I think if you really want to probe... If you want to probe, for example, dynamics close to the critical point, critical dynamics or critical slowing down close to the critical point or close to a spinodal, you want to pick up your concentration fluctuations or density fluctuations at the length scale of your dynamic correlation lengths. And not somewhere in between, because in between you'll get contributions from different processes. And so maybe this is something that goes beyond what this crowd here wants to listen to, but maybe this is something for more... Yeah, I just want to say also... In-depth discussion later. Yeah, so we also, of course, we compared it with the DLS with the closest Q that we were able to reach at DLS, so that we will be the lowest Q at the XPCS and the maximum Q that we can probe in DLS. So we saw the good agreement by looking at the plotting that temperature dependence for a particular Q and the data is consistent from the DLS and XPCS. They're just showing for the XPCS that with XPCS we can probe a larger Q for this. And we start to see that the critical component that is a mixture of critical and non-critical and there is this critical component already far away from the critical temperature when this fluctuation starts to build. That's what I'm trying to say. That's why you already start to see it even at the far away from the critical point by probing that smaller length scales when this critical fluctuation starts to build. That's like crossover behavior plus DLS, which is very consistent with what we've probed as well with XPCS in general for one particular Q that we were able to cross. Okay. I think this would require some more time. And I think it goes beyond what probably is the purpose of this webinar. I'll be happy, Peter, if you would have time that we can make a discussion with you and my colleagues as well for that particular case. That would be very interesting. Okay. I'll send you the email later. Okay. Thanks a lot, Peter. I think Sasha, Alexander has some questions. Yes, thank you very much, Khadina. Very simple. Don't worry. Just about the second part, you've shown that you apply the sinusoidal voltage and then you see the sinusoidal response. So can't you see the same if you just use, for example, less coherent beam and analyze brightness or position of the break peak or its widths, won't you see the same effect? Why do we need such complicated analysis applied here? So, yeah, as kind of we were actually surprised in general, if you just some intensity here, it depends on the amplitude field. If you just would integrate the intensity over that diffuse peak, you would start to see already that oscillant. So, imagining that you would have, this is kind of showing that example that you can apply here. We didn't know exactly what to expect that the small field as well. So, but generally closer to the break peak, you see the static behavior. So you don't see your response on the sinusoidal field and the two-time correlation function would be just a bright showing the static behavior. And for this particular system, the scattering signal is quite good for us. So that in principle, you kind of want to, you can see that response over that whole diffuse peak pattern. However, since you see the speckles of this diffraction pattern, which would reflect the structure, like the main structure in your system, we were thinking about also that there might be different dynamical behavior if it would analyze a different part of that diffuse peak. So that's why we were hoping that XPCS will show some interesting observation in the system. Okay, so basically you expect to get more information from those speckles than you already have, right? Yeah, yeah. So like I was trying. Sorry. Sorry, go. Yeah, I'm listening, sorry. So I was trying to analyze, for example, how the, especially the center of mass in different speckles would be moving and what would cause in general, whether it's changing intensity of the speckles. So if it's in itself, the intensity within speckles would change to produce this. So it is still ongoing analysis. I can definitely say that the oscillation disappears as we go closer to the break peak. However, we also consider that from where this actually effect comes, because we now trying to consider whether X-ray also may apply some additional effect to the system. So that's why I'm just showing that it's interesting as the experimentally type of experiment and for lower scattering signals, it's maybe also interesting to apply species for different speckles and different domains to extract different types of dynamics. Thank you very much. Thank you. Yeah, I think we have a couple of questions. There is Alex. But just if I get to Alex, just a quick question relating to what Sasha was asking, what type of physical phenomenon do you think it is driven by this tiny electric field? So as I said, just mentioned that we may consider that the X-ray may also affect and make also huge polarization effect also in the material. So we are working in this direction, too. So that's why maybe the small electric field applied is already creates this phenomenon by already made some charges in the system by causing by the X-ray, for example. But what could be another series that, so it's of course bigger field is needed to pull the atoms and pull the domains. The another series may be possible that we see some not flipping the electric, not flipping of the polarization. We see in the each domains, maybe smaller, like nanometer sized domains. But there might be some distortion from the circular, as you would consider, like a spin wave. When, for example, the spin moment would be kind of rotation of the spin moment, small rotations. And that small rotation would change the polarization and slightly change the structure that would cause the difference in the contrast, which is, actually, remarkably going almost like sometimes 100% changes of the contrast in the analysis that we absorb. Thanks a lot. Alex, please go ahead. Thank you. Hi, I'm Alex here from Max4. I have a super practical question. Hi. Thanks for a very nice talk. I don't know so much about XPCS, but I just wondered what detector do you use for this sort of tens of microns, tau? Oh, OK. So for this one, it's now commercialized by Rigaql. So it's 75 by 75 micrometer squared, so pixel size. And there are several publications about this detector. I can send you. Yes, great. I mean, we're looking into this also at Max4. And the Rigaql is one of the Rigaql. Yeah, it's a Rigaql. It's now bigger chips detector. So this was just before commercializing it. So it was just still in the test mode, a small camera. But then they prove it by putting more chips. So now it's, I think, 506, or even 1,000 nice pixels by 505. So something like this. And it comes in a shiny box now. I can send you the publication if you're interested later. Thank you very much. Yeah. So Edwin, I feel that you're mute. All right, definitely. Sorry about that. Go ahead. Yes, so actually, mine has comments to the last slides that you are presenting where you are talking about the MID or European Electronizer. So yes, to say that actually, the split in the day line was to say split commission on the month of May. So actually, in the next run that is starting in July, we are going to actually do XPCS with PAMX-ray to say. So this is what you were presenting. And a comment with respect to, I think, you simulate, or you say in the next slide, that you are simulating how is the pulses. When you saw that, you are taking into account the normal, to say, trains of European X-ray, or how is that stochastic noise? Yeah, this was so we were taking the exponential distribution on gamma distribution or simulating a different intensity in the pulses. So I think we were mostly taking the SACLA also. So that's not exactly particular for European X-ray. So in general, you would have a distribution in the pulses of FEL. So I didn't make a lot of experiments. So we were interested in terms of how to analyze. We did a few experiments, and it was very interesting. So actually, just to mention, when we are doing it at European X-ray with our train pulses, so depending on how many, the last May also, we arrived up to 1,000 pulses for the first time at MIT. So actually, along the trains, they are quite stable. So they are quite reproducible. They don't depend so much of the stochastic start of the machine for each single shot, but somehow inside the train, they have quite reproducible intensities. But in different trains, so that is every 10, every one, open one seconds, every train that is coming in 10 hertz, these guys can really change a lot between them. But usually, in the same train, it's really stable. Good to know. I think it's interesting for you. Yeah, I have participated in the Robinson experiments at MIT. And yeah, so I kind of don't remember the time. Was it very, in terms of analysis, I remember we had a lot of troubles there. So it's very difficult. It's complex. We have improved a lot, I think, in the last year. Yeah, OK, thank you. You're welcome. Nice talk, by the way. Thank you. All right, thanks a lot, Engel. And just to mention, we have two postdocs for XPCS at MIT, if someone is in. OK, thanks. I don't see anyone raising their hands again. So I think in the absence of any other question, let's thank the speaker. And thanks a lot, Dina, for such an amazing talk. And hopefully, we'll get to hear more from you. Yeah, I will be interested, as far as I get more, if you would be interesting, about law and diffraction. Like, I'm actually considering I'm kind of getting more and more to the law and micro and the diffraction. And I consider to put the coherent scattering methods for the sector study for E. So I'm looking for the users as well, if somebody has bigger structures and so on, who are interested to work with me. So contact me, send me emails, so chatting with me. So that's great. Thank you very much for inviting me for this nice seminar. Yeah, I really appreciate it. Yeah, thanks for the question. And I'm glad to thank everyone who has contributed also to the discussion. Actually, normally, we close officially the session. And if anyone is interested in staying a little longer, I don't know if Peter and Dina would like to take this occasion maybe to exchange emails or anything else than has been already said, it's OK. I mean, we can keep it open for a short number. Otherwise, we can close it here. Or if there is anyone else who can ask some maybe curiosity on the set up. Yeah, Dina, I actually wanted to talk to you a little bit about your fair electric relaxer if you've got some time. Yeah, I have another presentation where I just shown that the APS user meeting a bit more data and analysis. But as I said, it's really we've got an interesting behavior and data, but we're still trying to understand why language. Because it's something that I've been actually interested in for quite some time now. So you have a couple of phenomenon in this relax of materials. You have this so-called polar nanoregions that you would expect from relative displacements of, let's say, niobium and malase atoms and stuff like that. Then you also have the ferroelectric, the tweet domain. So these tweets, these are twin variants that you would expect to see, normally from needle-like domains. So the two of them actually are competing theories and even some experiments that show the possibility of these contributing to this relax of properties. And these are things that you would expect to probably see with such type of measurements that you're looking at. And the interesting thing with the twin domains is that the needle-like domains is that you need a very, very small amount of force or pressure to actually move them, to displace them, to move them around. So we had a paper that we published maybe last year where we just did BCDI to actually look at these ferroelastic tweets, ferroelastic needle-like domains. And one of our collaborators did some theory calculation just to show that the radiation pressure coming from x-ray, actually, it's sufficient to actually displace, to move around some of these needle-like structures. So you might be actually in the ballpark they're saying that there's some sort of dynamics that might come in from the x-ray. What kind of relax did you study? So we actually look at tweets at the barium titanate that had a little bit of enrichment with z-quanium. So if you dope some of these ferroelectrics with some acceptors or some donor elements, you can actually form some of these local sites that you can actually use them. Which BIMLIN did you study? We actually did this at Ross's BIMLIN. We did BCDI. I was actually, we're trying to do XPCS. We're still writing the proposal, but we'd like to do XPCS on this type of system to see if we can say anything about this so-called polonano regions and also the tweets themselves. And you would expect them to, I mean, they are different. They fluctuate at different lens scales. So you would see them at different cues. At least that's what we're expecting. So at higher cues, you would expect to see maybe this polonano regions or the tweets. And then at lower cues, maybe you see the domains. But then again, you have to, just like what you did, you have to drive them with different strengths of the field. So a large or strong electric field would really move the domains or switch them. What electric fields were applied there? Or did you apply it at all? Or did you were showing, you said that you have seen a lot of radiation pressures. Yes. So we didn't apply the field in this case. We just saw some effect from radiation pressure. The ones which we applied the field, we applied fields of about, I think, the voltage was about 20 volts. And then we had, I can't remember, the thickness of the sample I have to. May I ask you where did you make the sample? Or did you got a commercial sample? No, we have some samples from Sint, from Los Alamos. So you were growing the sample? No, I was just wondering. I think Chen is a collaborator in Los Alamos. He grew the samples for us. I see. Because we were kind of always, in general, sample preparation for the experiment. Like we got some sample which was grown, some samples which were cutted from the commercial and polished, and also making a contact on the side. So everything was so, we improved every experiment. But mostly what we were improving the technical details, like in terms of the preparation and the aligning and so on. So a part of the analysis took a lot of time to make a proper sample preparation. I have a short question to you. Yeah, Dima, go ahead. Yeah, hi. This switching you say because of the beam pressure, is it related to the ferroelastic behavior of VTO or not? It is related to the ferroelastic. So ferroelastic. So it's basically the twins switching there. This is what you think? Yeah, that's what we think. And we have one of our collaborators. He's working on the theory. So he actually showed that with a small force of, I can't remember, just a few piconewtons is sufficient enough to actually displace some of these. Of course, you have different lengths. There's a little bit of a mesoscopic lens scale of this ferroelastic domain. But it's specific for barium titanite. Yeah, it's specific for barium titanite. I don't know if for PMN, it would be very interesting to do the land-out calculation. Yeah, interesting if it differs a lot. Because recently, we published a paper where we used the twinning switching by AFM. And it took a lot of force to do this. Yeah, it's OK. Yeah, thank you. Yeah, thanks. Yeah, so one thing that we are actually considering right now is that there is a, so he came up with a paper that shows that if you can display some of these twigs, it's possible to displace the polar ions and then create magnetic fields. So this is another approach of harvesting magnetism from systems that, from oxides and complex oxides that do not predominantly have any magnetic atom or ion. So just by displacing these elastic domains. So I would send you an email, Dina, just to... Yeah, it would be very interesting. Because we start to kind of, with this very electrics, like for me, this materials was new. I was joining, I must say, this project as a postdoc and I continue working and it was very exciting to learn about the sphere of electrics at the beginning than figuring out that there are so many theories. It is like a so long years problem as well. Everything would be helpful to figure out what it's cause. And if you're interested in the XPCS or like the stuff, so I will be very interesting also to join and see how it's... Definitely. I will send you an email. We actually, we did some XPCS measurement at sector eight on copper oxide, actually copper oxide is expected to be, it's reported to have magnetoelectric properties and between two, from 213 Kelvin up to I think 230, this window is supposed to be ferroelectric. And then this ferroelectric window is has some sort of commensurate behavior with an anti-ferromagnetic phase, an AFM, AF1 and AFM2 phase. And really what we did in this experiment, we just did a normal heating and cooling. We cool it down to about, I think maybe 80 Kelvin and then heat it up again to room temperature. And then we're doing XPCS. But the interesting thing is that we have some data that we start seeing some sort of imagined something that, so we see some structure coming in. So initially you get the traditional speckle patterns and stuff like that. And once you hit that ferroelectric window and you see some stripe-like features coming in a little bit at higher cues. So it's... Kind of like a memory effect from the previous temperature. I think that's... Yeah. So it's something that I would like to discuss it with you and maybe get a little bit of understanding of... I haven't done XPCS, I've always wanted to, but I haven't. I just started getting the XPCS the past couple of years, so... Yeah, there's additional aspects when you try to do with a different temperature. Like for example, also this ferroelectric, they have also temperature dependence. So we choose the PMN and we usually now performing a room temperature experiment. So we are considering also to try wearing the PT component which will shift that the electric constant dependence in the temperature range a lot. And so for some of the materials, you need to like cryo-temperature for some, you need to heat it up. So we're looking forward also to investigate the temperature behavior, but so far we just realized this behavior and trying to understand this before going forward. But we consider also different temperature behavior might also affect this kind of different memory and so on. You know, that's another interesting direction to throw. I think it will also help for building the theory model. Yeah, Yui, Kyle, he's currently also trying to model this kind of speckle pattern. So because itself that speckle pattern is also interesting. We see some small speckles which is like one pixel size, but we see also bigger features in it. So which would, we think that it's kind of like a pudding, like rising pudding, have a bigger, maybe bigger domains with smaller domains in it. Like that's another possibility, but we're still trying to understand the structure as well as I think the simulation experiment would help. Did you see the same behavior for a different reflection? Let's say for... Yeah, we saw the set up that we used allowed us unfortunately to go only, like we tried to zero zero one as well and we saw the similar response at zero zero one and one zero three peak we also try. Unfortunately, due to this kind of setup, we were able to access only a few diffraction peaks there. So because the original setup, I can show you if you can see, we'll be interested to perform a data ID. Did you get any beam time for that XPCS experiment already? Or... Yeah, we already did. We had two measurements already for the data ID on copper outsider. So I don't know exactly which set up you were using. Let me find that. Yeah, that's a... Yes, something I was asking Eric about it. So if there's ever gonna be a possibility of having a magnetic field, you know, Oh yeah, magnetic field. I don't know exactly if they have a magnetic type of set up. So we used the one that allowed to keep the sample. So they have a cryo set up that I think can go like 70 degrees. So that will be possible with a cryo set up, I think. And it's a bicarrelium dominant. So you can access more angles. So in our case, we use this one, which has a water cooling and it's kind of disc-roundish. So the sample is sitting. So this is a becoming beam. And there is a very small arc kind of window with a certain width window. So it doesn't allow this a lot of in-kai direction. We were able to move the sita, this is a two sita. So, but the kai direction was limited. Yeah, so that was kind of tricky. We were hoping to use a cryo set up next time to see. And yeah, so we were kind of mapping the diffraction peak. So this is a butterfly diffused scattering. So we go over the two different direction of that butterfly. So we were catching the two inks. And then as I mentioned here, like if I go closer to the break peak, that's I don't see the response on the field. We also saw a different kind of response on the asymmetric square wave. So we applied not only sinusoidal, we applied the asymmetric square wave to see. And yeah, here is a little bit more. So we saw different behavior in the relaxers depends on increasing amplitude. So it gets kind of linear to nonlinear region. And yeah, so as far as I was going, I was trying to analyze where is this kind of decarrelation comes into this periodicity. So I was trying to look at the special cross correlation analysis of different patterns and see how the center of mass in general changing. Mostly changing within that. And it shows that the center of mass in general is a small, it picks the speckles, like would shift in that direction. So I'm trying now to analyze this in terms of the Q space and figure out what is in the Q space that's shifting. See whether it's a red one. But yeah, so you, I don't know exactly which sample chamber you expect, but if you would go for this one, kind of doesn't have a lot of out and go some, I don't have a picture of another sample chamber. But you can ask Eric. If you have any question about the kind of, I developed the data here, I usually using some of my codes, but they have a very good pipeline analysis as well. So, yeah, you can directly find, see that's nice. Yeah, I'll definitely touch with you with the analysis. Yeah, sure. Feel free. Interesting. Yeah. This is quite interesting.