 Thank you for this opportunity to be presenting you guys some of the things we have been doing here in Brazil, right? So this is for those of you. I think some of you may already know our synchrotron, our new synchrotron. So this is a picture of the new synchrotron, right? And the Carnauba Inline is like this one, right? The last that we have here, the one, the longest one, right? So I'm going to present you some of the things that we have. All of those things are not a single person work. So this is, of course, a presentation on behalf of all of them, right? So just before I start, I'm going to present you the idea of the campus. So we are in a campus called CN Paint. So it's a research material and energy institute, right? And there we have different labs, national labs. We have a nanotechnology lab, a biology lab, and one related to renewable renewables and bioenergy, right? And then, of course, we have also the synchrotron, Brazilian synchrotron light source, right? And I like this picture just to demonstrate kind of the huge step we have been doing, we have done in the last years coming. This is the old source that we have, right? So I will show just a quick picture of it. But you can actually, at least for me, because it was such a huge step, we can actually see that the entire ring of the previous one fits on this new machine, this new building that we just finishing, right? So after so many years constructing and building and perfecting this machine, we were able to get this resource and make this huge step from a second generation machine to this fourth generation that we have here. So before we start, I just want to show this slide. This is a picture of the old synchrotron. But the reason why I'm showing that's not because I made my PEG and I'm kind of nostalgic about it, was actually to show this graph here, which is just related to a work that we have been done in the imaging beam line that we used to have there, right? So it's a tomography beam line, regular full field illumination, then you take the measurement and this would be the four-year shared correlation that we have just to show the kind of resolution we have, the kind of image we used to provide to users in the old machine, right? And also give you the scope of how huge it was, how huge it will be for this Brazilian community, this step forward that we're doing in terms of imaging, right? Because in the end, I'm a microscopy, I'm a guy that likes to make nice photos of things. So that's the kind of thing that I like to show. Concerning the new source that we have here, it's a four-band, five-band acrobat, sorry, quadruples. More information, of course, I'm not the one with the accelerator, so of course please contact the responsible ones, but it's quite similar in that sense, machine today one we have in Sweden, right? And Nanomax and you can see, I think this seminar is for exactly this kind of discussion, discussions on what we expect for this new machines and how we can better enjoy it and how we can better take part of this development. Now focusing on the beam line, right? That's a little bit more comfortable to me. So this is the kind of a beam line. The idea is that it's a coherent x-ray, nano-focused beam line, right? Then we have like a source, an on the later source, some focusing. Interesting, in this case, our focusing to create this secondary source is only in the horizontal, in the vertical direction, we actually see the source inside the ring, right? And then of course we have some leads, a four-band monochromator, right? For stability and also it's the way it's supposed, it's demonstrated here, it's mounted in the horizontal direction to increase. And then we have a representation here that actually is the representation of the end station. But we actually have two end stations, right? So this KB mirror focusing device and all the detectors that are presented here actually is split it in two. Though those are sequential beam lines, so you need to choose which one you're gonna use. But nonetheless, they represent different opportunities for measurements. Both, anyway, in both beam lines, the idea is that you would be able to measure fluorescence, absorptions, scattering diffraction, and even luminescence from your sample. And it's as an imaging beam line, the idea is to do 2D mapping, 3D tomography, going for typography, and even some images in bright condition, right? Now, some details about the flux and the energy, you can see that, for example, the energy, the beam size, you can see here that we are kind of diffraction limited to some energy, but you can already see the difference between the two beam lines. So the total line station that it's already under commissioning and I will present in some of the, how can I say, brand new results that come out. It's not, it's more versatile and it has a bigger focus, right? And then for the nano probe, really, not nano, yeah, that's why we call it like a super micron probe, right? And then for the nano probe, the Sapo-T station, then we have like 30 nanometers focusing around, depending on the energy, but that's the second stage that we'll be developing after we finish the taro main station, right? In terms of flux, so we have 10 to 12, 10 to 11, 10 to 12 flux, those would be fully coherent flux because we had this lead that select on the coherent part of the beam, right? So now I'm gonna focus on the taro main station, the end station itself, that is 136 meters from the source, right? So it's a very crowded space, right? You can already see here that you can't really tell where things are, as I need to point a little bit. So the beam will be coming behind these luminescence detector that we have here, right? And then we have the sample. And then you can see how difficult it is to get so much contrast because of the number of detectors that we have to pack everything together. That's part of the challenging, but that's okay. So we're managing that. And then we can see that we have like two fluorescence X-ray detectors. We have a diffraction detector. With some angular acceptance transmission one that will be of course used for tachyography. And just here in the middle, oh yeah, of course. And then we have just optical microscope for helping navigating in such a difficult, such a compact environment. But then we have here the sample environment that's zoomed here. Just this is just an example to point out how these things can be small. I think most of you are used to but always impressed me the size of the scale of things. Because from this part of the design it's easy to see such a draw like this, but usually you forget how small thing is. And now when you put your hands on it, you can actually see how difficult it will be to make it fit, to make it the proper measurement. So, but basically this is the basically design. I think it's traditional one. We have like a normal pin then where you can put your sample. But of course we are also working on different sample holders that will fit different demands for different samples. And I forgot to mention that the tunnel mine station was designed to be a linear microscope. Meaning it can make it easier for most of the experiments. You don't have to worry about cryogenic conditions. You don't have to worry about vacuum. And the energy that it works kind of allow us for to having that one. Currently the current situation that we have here. This is a picture. It should be a couple of weeks old already but shows most of the things already assembled. It's not so packed, right? Because of course some of the detectors are not in here. So we only have one vortex fluorescence detector installed. The other one is in the way. The microscopes are set, one of them. The luminescence is not in assemble yet. But that's the way of things goes, right? So we assemble things. You can see how huge the structure is and how tiny the sample is compared to everything around, right? So this is the main idea. I think this will be one of interesting slide in a sense that I try to bring the numbers for this station. So the transmission detector, for example, it's 1.1 meter. Sorry, this is millimeter. Sorry, not that much. So it's not a success. Be mindful that this one should be millimeter, right? So it's one meter from the sample, right? And then it would have an angular range from zero to 40 degrees. If you look closer to the sample, then we can see the detectors. So the fluorescence detectors, they have different arrangement. This was in purpose that we have two once fluorescent detectors so that we can actually make fluorescence tomography, right? The rotation stage is not capable to do 360 rotation. So we went the other way around and make those two detectors working together so that we can actually make 360 rotation, right? For the fluorescence tomography. For the transmission tomography is more than 180. So also it's gonna be fine, right? In that sense, we also have this other detector here that's a little bit shadow cover here. This is an also an error detector. So that would allow us to measure not only the transmitted beam but also diffraction simultaneously. Of course, you can also choose to do this diffraction measurement using the main detector. This detector that we have here is any house development. I show data what we have been working on but it has like more than 1,500 pixels, right? And then we have of course this small one that's only 512 pixels, pixel side, right? And then the microscopes and the luminescence one that we are developing. So I think this is one shows a little bit of the possibilities that we envisions with no techniques, with the possibility of doing simultaneous measurements. So the scope of this beam minus you do almost all as much as we can simultaneous measurements, right? So the detectors are designed not to interfere with one another, that's why we're so packed together. In terms of detector, it was an in-house development. So this is actually, this slide is not mine. This is from the Jampoli, which is the head of the detector group. He's developing together with a Brazilian company. This area detector for us, right? And then it comes on different sizes, right? But I think the most interesting part is the pixel size which is 555 microns, kind of traditional ones, right? But they are in modular ones, meaning we can make small ones like this one and bigger ones that I think even Carla when she presented like the Caterete beam line, she probably mentioned, she did mention this detector here, which is a huge one. And that's the good thing to have this Brazilian company joining us on this development. Now we have our own detectors, which is a critical point and serves almost all beam lines that we have at the Sears right now. Summer environment. I mentioned that the taro man is an in-air microscope, right? So it's now that, of course, we can measure regular samples, we wanted to do more than that. So we have different sample environments. Most of them are already in prototyping, meaning I'm showing these ones, I'm showing those are actually technical drawings with a lot of details. Some of them are already revolving simulations for thermal and mechanical stabilites and actually being manufactured already in prototyping and everything. This one I'm showing here that was supposed to be rotating. It's a special case that I'm here. I just brought here because it's one of the projects that I'm mostly involved. So this is a prototype that the idea is to create an environment where you could make the tomography of a root system, right? The plant and then you can focus on the capillary where you would guide your roots and make a proper tomography, not only for imaging because remember as I mentioned, fluorescence is a thing. So guys would love to see the elements turning, going back and forth around the roots and everything, right? This is the cryogenic setup. That's another one that I'm mostly involved. You can see it because it's an in-air microscope means you have a lot of... It's not easy to be... You don't really need to have cryogenicis but of course you're gonna want that, right? So we need to come up with a solution so that we use a more... I mean, it's a traditional crystream setup with a lot of things around it. So to make it work on this geometry, right? It took a lot of simulations to see the best way of taking this cold air away from the staging and protecting and even developing some extra protection to the sample so that the flux, although it keeps everything safe from freezing, from icing, it may of course disturb the sample a little bit so we add also to protect not only the stage but if necessary, we could protect the sample inside a small dome that we are developing here. Atmosphere and electrochemistry is of course part of the deal and we have a lot of developments on that too, right? And of course the whole idea of having this bunch of sample environments on this microscope is because we are trying to cover the main scientific cases that we identify, some of them being even more important for Brazil, like for example, this soil science and plant science that we have environmental and agricultural science that we have here, right? So now finally for what exactly do we have right now, right? We have a little bit more, of course we have more than what I'm showing here. These ones are results from the last year, right? These things have already happened from there, but I decided to brought these ones because it was like a nice story to tell, okay? So the Kanauba bean line had its first light on October, right? The end of October, right? And then we start to do the commission. When I say we, I said a bunch of people, not only the bean line staff, but we have a whole engineering and support groups for doing that, then we start commissioning and the guys managed to make it work until in December 10th, we managed to have our first image. I know it may look a little bit silly, but at least for us, I was there. It was a nice moment because we finally saw something that reflect all the effort that we had been doing on designing and everything. The resolution wasn't so great. The resolution was actually worse than the one we had before, not that much worse, but the one we had in the previous one, but it was still very nice to see things coming to be. To be honest, actually, we had a scanning fluorescence on the old source and the resolution was 25 microns. So I guess we're ready to make a pretty good job in a sense that we are pushing things already further than what we had before. But of course, people would not be happy if you only stay on the micrometer. So after some efforts, the beam was reduced to around 500 nanometer, and then we managed to make some nice fluorescence imaging of this Siemens start here that it's kind of a pattern that we have here. And this was done just before, in the last week, in the last week that we had available on 2020, and it was an effort. And finally, we managed to gather in December 16, we managed to gather something like this that I'm showing here. It's not a beautiful scattering, has a lot of things around, has a lot of things happening, but gave us the possibility to do ticography. So we managed to make some ticography, and this was actually at the last day of the year. It was the last beam that we had. We even, we wasn't able to make, to be sure that this would happen. Of course, seeing structure like this is scattering around, maybe it made us some promise, it was kind of a promising, but actually the imaging took us a while to be reconstructed. And the reconstruction was done by the scientific computing group. So that's, thank you, more specifically to a guy called Giovanni, right? It's a very nice guy. I always talk to him and everything, we discuss this kind of things. So a lot of tries and error, trying to find the proper parameters. It's a very intense, but he managed to find the proper parameters to reconstruct the Siemens star to a much higher resolution. Of course, this 30 nanometer makes no sense. It's just like a pixel size. It's not the proper resolution. If you look closer to the image, the resolution is not as a tropic. Of course, I have been looking for this picture for a while, but that's what's very nice. And then of course we have a lot of incoherent modes, which just show that there's a lot of work to be done to clean everything in the beam line, but it works. I think this was the main message that I would like to bring here. So it is working, comparing with the fluorescence that we have here, it works and we had a very nice news for the end of the year. Although independent of how it was difficult, it turned out very nice by the end. So that was Taruman. This is just a slide presentation for the second one, the Saputi, which would be for scanning and techography, all done and also integrated with tomography. It's an in vacuum, ultra high vacuum kerosene microscope. The same detectors fluorescence one transmission that I'm sure here. But of course inside we are finding ways to put it, the Xeon optical luminescence inside. And of course some of others detector for intensity, if it's necessary to make the measurements in vacuum regarding for example, the transmission, right? It's not on our shell. I'm showing the shell, but I decided to show these two images here that showing that we already have stuff inside. Actually, we already have stuff that had been built and designed and are part of the commissioning stage right in there. And this one would be that old pin that I showed you guys before here, just to give some sense of where things will fix. Right? So, and finally, just before... I have a quick question. Does it have a cryo transfer where you bring the sample in already frozen? Yes. Yes. So this yellow part here, it's cryo transfer chamber, right? If you look at closer here, I don't know if you guys are familiar. This would be the Leica VCT 500 transfer system. Yeah. Right? For the vacuum bag. Vacuum case, sorry. Thanks. Yeah. That's the main point. This is not only for radiation damage, but also to keep the sample on cryogenic condition for the entire life cycle, right? Of the sample. Any questions? That's fine. Okay. That's good. Thank you. Yeah. Okay. So, and now the questions. What do we do while we don't have a beam line, right? So this is just the two plate around that I made. I mean, with numbers, right? This one is motivated by the guys from the soil that I mentioned to you. So the soil people, they would love to measure phosphorus, fluorescence from phosphorus, but the phosphorus is like the hard thing to do, right? Because the energy, the fluorescence energy is so low that you most likely won't be able to actually see anything, right? Except for the surface of it. But the guys keep asking, wow, I would really, really love to have some kind of true information. That would be very nice to have some kind of tomography. And then while I have some, some discussions, he kind of, I kind of convincing him that it would be okay to have some kind of tomography, not in the entire surface, but at least something that would be around it, like a shell through the information. Like that's why we call it annular tomography, some kind of tomography on this annular region. And that said, okay, it's better than nothing, better than just doing 2D mapping that's usually done. And of course this would apply for other cases, like if you really don't have to measure this region, you can save that. And if you really don't really to touch that region because of those, you don't have to concern. Of course you will still have fluorescence around this region, but at least the direct beam will not be touching anything in this region, right? And then we made some simulations. It turns out to be very okay to do those simulations. Some optimizations had to be done, right? You have a proper cover to actually measure all the pixels that you are interested. And I just realized that I put the wrong, sorry about that. You can see that this one is not the optimized beam coverage here. It has some a little bit of miss regions, but nonetheless, you can actually see that the reconstruction look fine, right? So it's okay. Probably it's going to work. We are waiting for the beam line to actually test this idea and see if we manage to do that. No big deal. Just playing around with numbers. And then we have this another development, this other development here doing conjunction with the optics group and the beam line design group. The idea is to see how far we can go on actually aligning the beam line, right? So the idea is to use a tachographic measurement and from that use this unique polynomial composition to actually understand where it is and then retrofit it using machine learning to actually align the beam line. What we have done so far, so far, because we're of course waiting for the beam line, we have to create the base, right? For you to do this training. Let's put it that way. And for that we use like beam line simulation using SRW package. It's one of the version of it is this nice coding, like connecting things, but this is not exactly just for illustration here. But the idea is that you misalign your optics, then you go for a proper, the composition on this square aperture of course, because you're using KB murals. And then you can actually see what this misaligned meant in terms of different aberrations that you would have. So far then we have the assembly error, meaning that this is all the errors with respect to the both KB murals that are inside the beam line. And how is the assembly error concerning all the possible strategies that they have been developed in terms of misaligning on the position of the KB murals. Sorry. And then of course we run some quick simulations on that. We track all most of these positions. You can see there's a lot of possibilities, a lot of things going on, but this would be the first machine learning validation that we have in that, meaning that of course this has also simulated data because you use for validation, but they are not used for training, meaning that in this case, for all those cases, it had like a maximum error of 3.5%. So we would be missing any of these numbers here, as so far by a 3.5%. But we start developing, still going for not only of course measurements, to see if we can actually take this from the tachography measurement, but also try to increase the errors and how we manage to fit this problem. So with that I think I'm finished. I want to say thank you. Thank you. This is not as light. This is actually as light for the people back home here in Brazil. So this is part of the, this is the Carnival team, right? And just part of it, you can see how big it is. And this is part of the people, optical group, engineering and so forth. So all of those guys actually made it this happen, right? And manage and manage most of the things work. So and now for the other audience, I say thank you too. And I'm open for questions. I hope I haven't been too fast. Thank you so much. This was very nice, concise, and so there is lots of space for discussion. We have already a question from Richard Sandberg. Well, when will you get beam, at the beam line? Yeah, so, oh, I'm sorry. How can I stop this posture? No, no. Okay, stop here. Can you guys see me? Yes. Okay. So that depends. We already have a beam at the beam line, of course, but that was made in a little bit rush. So for this first part of the year, we kind of reworked most of the things that had been done, but the beam is already, should be in the, the, the, the Tatlman station already, right? I think the proper question, question is when we're going to start doing measurements. And that would be most likely in the next months, because we now are going to run proper experiments for validation and actually trying to see things that may have some scientific impact in that sense. And of course, make sure the beam line needs, it's working on all of its different possibilities. So I would say by this semester, we may be starting experiments for users. Well, that's, that's another question that's, I think it's too uncertain to actually put anything like that. Maybe I don't know if any of my colleagues from home would like to, to point that to say anything, but I personally don't like to really put a date when we'll be open because of what actually happened the last years and is still happening. Thank you. Actually, I would like to congratulate with you and the whole team for this. It's quite a, quite an achievement in small time from October to December to get all this. And yeah, all the best for the future. We have a question from Lukas. Lukas, you can unmute yourself. Thank you. Yeah. So first, a nice, nice presentation. Thank you. My question is regarding the tarman station. It was not clear for me which rotation axis you have for example. For instance, I mean, I wonder if you, if you can rotation, if you can rotate your simple round kite like around the opto axis or. No, that was my fault. Actually, I missed on mention that. So we have like the possible, the beam, I'm itself, it's made to do only the tomography rotation. Right. So it would be that's the beam idea what I showed you, but we have space for gonometers and we actually are developing the, the, the, the necessary environment, environment to have some rotations in the guy and the other ones. Right. Because we understand that of course before proposing to do the fraction, we need to have all these other freedoms for actually online. So we have, it won't be like a full guy, but it's close. What, what does it mean is that you would need to pre align your sample outside of the beam way. Right. But then we would have some free, some degree for you to optimize that in respect to the rest. Okay. At least a few degrees, but. Oh yeah. Sorry, sorry. Can I make a follow up question? Yeah. So you, you said that the, the time one station is designed for tomography, but the supporting station. I also didn't, didn't follow if you have this. No, both of the beam line and design for tomography. Right. The idea is that it's a microscopy beam line and what we understand that x-ray microscopy eats tomography. Right. Being that transmission tomography or fluorescence tomography. Right. So all of them have the, the thing is some, because of the cables and the stability we're looking for, we are not able to do 360 rotation. Right. That should be okay for transmission tomography, but it's kind of a problem for fluorescence one. That's why we have two detectors and they should complement one another mean giving you a fake 360 view. Right. 360 rotation of your sample. But the supporting station is, you said it's for scanning the fraction. So you probably have like a translation or sort of sample. Oh yeah. Of course you have translation in the, in the plane. We, we of course, because it's the scanning beam line, meaning that you're going to have to scan the sample on front of the beam. Yeah. It's fine. Yeah. But the supporting station is a scanning in typography one. Right. That you can position your detector for some diffraction angle. Aha. But do you, you have only one degree of freedom for your detector. Yeah. In that sense, yes. Okay. That would be the, the, the, the, the Teta direction for the detector. Okay. Thank you. We have a question from Salvador Ferrer. What about the control of the sample drift in the nano probe? Do you have interferometry? Yes. So we have a bunch of detectors. So both for it's closed loop, of course, and everything for both stations. And for the supper tea, it's even title minds also has some interferometric sensors. Redundance is also one thing that we try to, to, to make it. Right. Because. Okay. For the supper tea one, we have only small range and everything. So it's, it's, it's kind of limited in that sense. So more for special samples. That's why, why. I, we have focused more entitlement because it's much more versatile and should. Target's much. Easier, much easier. Right. Easily different scientific problems. Right. So we have different stages, the core stage and fine stage. Both of them have. We have either capacitive sensors or interferometric sensors. Right. And for the supper tea station, we also have interferometric sensors. And the position. One thing that we had a lot of put it, they people put a lot of effort on was actually to make a proper. Termal. To look at what, what, what, what they would be the implications of any Tamil drift. And what would you get with this Tamil models and everything. Right. So the supper tea, of course, is going for the highest resolution because we have much more control in it. Right. It's the cryogenic condition is not only because of the, the sample itself, but also to have a proper reservoir to cool down and control the temperature. Or most of the things inside. Right. So it's not only, of course, we understand that the motors generate heat, but then we use this code has a war to control the temperature of almost all the involved components. In that minimizing drifts and everything. So that was one of the concerns that we have. Right. That's why you're aiming for so aiming for very, very high resolution at the. The supper tea station. Thank you. Next, I think was Manuel. Can you unmute yourself? Yes. Thank you. Carlos. It was very nice talk. Thank you very much. And also I think I want to extend in as congratulations because it's very impressive to have from first light to first typography in three months. But I have two, two point one thing. This is first light, but the guys manage a very nice work on arrow budgeting. So I that, that's a table that I showed you. They have that same table for almost all components on the beam line. Right. So they actually know what they are doing when they are putting things in the, the bean path. I was very impressed too. We have a few things to learn for the upgrade of SLS as well to, to manage such efficiency. But I wanted to ask you for the supper tea beam line. When do you think you would have already typographic tomography running there or estimate? I do not know. I do not know. I think I do not know. I think the first in these days is very difficult to know with much certainty. And I also was curious, maybe I missed it. What are the target resolution and sample sizes that you have for the high resolution in 3D imaging? So not this year. All right. So we are actually put the, the, the supper tea project on hold, right? Right now because we're focusing on the time, which is a good thing. And we plan to, to, to, to restart, let's put it that way, to, to start, to, to get back to the project by the next semester. Right? But in any case, it won't be operational. I mean, maybe we can try to push like we did last, last year. Right? But as I said, it was just like a way of showing, right? To end the year was such a hard year. So we, we need to have something to, to show in something to, to be right to presenting, right? And that was the goal. That's why we kind of revisit most of the things. So it's not going to be this, this next year. The best thing I can mention to you, it's actually what would be the numerical aperture of the detection, right? Cause you said resolution. That's it. I think unless we. The specification for the positioning would be relevant. So the detector. Yeah. No, no, that's everything is will be around nanometer. Right? So the resolution is around nanometer. And then the position is also going for that target. Right? So I would say units of nanometers. That's, that's the goal in terms of positioning and in terms of detector. If you, of course, we have signal to that up to that, in terms of scattering, that's another issue. But we are trying to go for active one nanometer. That's that's like for the numbers. One of these would be great for us. Thanks. Thanks for that. Constantine. Yeah, thank you. Yeah. Hello. I guess you will be using variable polarization. Can you tell how? Because we are right now, we don't have that, but the beam line was designed with a delta on the later. Yeah, this, this is why I'm asking this. Yeah, exactly. So yeah, we are developing our delta on the later in how is any house development. So we, so remember when I mentioned the previous beam, the previous source, it wasn't like a second generation ish because we had an on the later in that. So we developed the on the later for that. And now we are developing the delta on the later for this new beam line. So you can actually see that the fraction is in plane. It's not in the vertical axis. That's one of the reasons to. So we push for stability in that sense. Then we choose to change the polarization and then be able to make the diffraction in plane. But it is, it will still be linear. No, yeah. I mean, you could choose with that. So we have the possibility to have linear polarization in the vertical direction. Okay. Thank you. Next in line is Virginia. Please unmute yourself. Hi. Hello. Hello. Thank you very much for, for the nice talk. You mentioned in your presentation that you are going to implement bright tech graphy. And I would like to, to, to know a little bit more how you are planning to do it and, and, and, and where and so on. Well, first of all, asking for our help. Right. For trying because it's a very impressive presentation. So I was even discussed with Dina. So if I could mention that, that the beam line, of course, when was first designed and conceptualize all these techniques were boring, right? It may sometimes do not even exist. And we try to push some, some, for some design that's versatile in that sense. Right. Concerning bike, black, likeography, I definitely would need some help. Right. For developing. I asked my supervisor to, to, to be, to, to, to go in that direction. He said, of course, it's a very nice technique. You should try. I don't know if I made the right choice, but then the idea is, of course, to, to try to follow your steps. Right. In that sense. And from that, I would definitely need your help. And I hope, of course, you would, if you have any comments, you have any, if you need any further details, to, to, to on that, we can discuss. And I hope that the hardware that we have, it's, it's good enough for, for your technique. Okay. It wasn't really not the, the, the direction of my questions. I was really just curious to, to hear your thoughts and so on, because as Steven, as, as mentioned in his previous presentation, his current seminar, right? You know, it's quite tricky and it really requires some, some manpower actually for finding the right, the right button where to push more than on the experimental power actually on the inversion path. So this is where the, the manpower should be on the synchrotron side, I guess. But so yes, of course. Yeah. Luckily for us, we have a very strong scientific computing. We have a very strong scientific computing. Yeah, luckily for us, we have a very strong scientific computing division, right? That also would be interesting on going for the algorithm, because I remember you said this is kind of the, the one of the critical parts. Yeah. Sorry. But I understand that comes together with a proper instrumentation. Okay. Yeah. I hope we can actually develop much more discussions on that. Okay. I'm glad I get your attention. Okay.