 Most of the second half of the 20th century accelerator based high energy physics was the paradigmatic kind of physics and we got ever bigger accelerators built. Synchrotron radiation was just a waste product that just happened to be created whenever you accelerated electrons and access to those synchrotron radiation beams was granted more or less gradually by the high energy physicists to other less fundamental physicists like solid state physicists or even molecular biologists. That has changed quite a bit in the last few decades. It's becoming more and more common to build large accelerators, large particle accelerators, not to study the particles by smashing them into one another but to use these electromagnetic radiation that was originally considered a waste for other purposes like material sciences or biology. Hubert Seidenhandel is the chairman and managing director of the European Excel in Hamburg, which operates one of the most advanced of these new kind of accelerators and he wants to tell us about the new paradigm in science that these devices enable. Have much fun with Hubert Seidenhandel. Thank you very much and thank you for the introduction. I hope that the technical difficulties are over now, so we'll see. So I will tell about X-ray field so laser as I call my talk a new paradigm in science. I hope that the translator can follow because I'm a fast speaker. So on this first slide you see the opening of the field for laser with the green laser from the Elbphilharmonie. You see the freedom so laser in the middle and you see the directors of European Exfol to the right. You see me with the that's the guy with the gray hair in the back. So I'll give you a personal view on where we are with all this here. So European Exfol, it's your European, it's your European facility. You see the shallow countries, it's X-rays. We want to see atoms and molecules so you need X-rays for the short wavelengths. We use free electrons and we have a laser and I think it's quite timely to give the talk now because it's actually nearly 125 years ago we had the first anniversary for the first X-ray image which was taken on the 22nd of December 1895. So this is what I'm going to tell about what is what are free electron lasers, what is European X-ray, what's your organization, what's the layout, how do we actually do science and then at the very very end tell a little bit about why it's important to do this kind of science. So my background here, I have nearly 40 years of experience in X-ray science. I'm coming from Denmark. I'm actually sitting now in Roskilde in Denmark right now. I'm a professor at the Niels Bohr Institute in Copenhagen. I've been in Hamburg since 2017 and as you see from I don't know how many physicists are among you then the physicists probably know PSD comics and this is Professor Smith's and you see I'm nearly looking the same and as Professor Smith I've not done experiments or data analysis for years so I'm just the managing director. So as you heard in the introduction X-rays for science is today generated by central radiation facilities and lately also by X-ray free insulators. So you see this plot here which shows the quality factor or the power of X-rays called average brilliance as a function of years and you see from about 1960 the quality of X-rays has increased dramatically. So this increase here which is basically a logarithmic growth here, it's the growth faster than Moore's law and all you know Moore's law and this year it even grows faster which give us problems because our data generation is basically proportional with the brilliance so we produce data faster than the increase in Moore's law and I have been in the field since 1981 where I was a student, a summer student that they see in Hamburg and I can tell you it has been a fantastic experience following this increase of more than 10 orders of magnitude during your career. So what are X-ray free insulators? So X-ray free insulators as you heard in the introduction this is the accelerator, it's based on accelerators so you have an accelerator of electrons which on the left you have the electron emitter where you emit electrons by the laser you have a linac to accelerate the electrons, you have a bunch compression to make them very short, you have an electron bunch which is very short, you have another linac, you put the electrons through an undulator where the electrons do a sinusoidal path where they produce X-rays and then you throw away the electrons in an electron beam dog and you need to have X-rays for the kind of science we want to do because you want to see atoms and molecules and X-rays are very energetic, light particles, photons and therefore you need a long and big accelerator so the undulators that produce light looking like this year these are magnetic devices where the electrons go when you go through the the undulator they take a sinusoidal path we have about three of the we have three of these undulators and each of them is about 150 meters long so so this it's a huge piece of infrastructure with very very many single magnets so the whole trick about laser assist you need Korean light you need to have the the the particles that emit light to be organized in a certain array on the left you see a non-Korean bunch which are electrons normally generating light in a synchrotron we want to align them to have a Korean bunch so they all send out light that is been amplified coherently and the way you do that is you do it by lasing in one pass so we use a principle called self-amplified spontaneous emission which was introduced by Afgani Saldin on the left you see him on the or the left where he received a prize he invented this idea in Novosibirsk in the late 70s in the last century and the principle was then further developed the idea is that you have a the long undulator you see here the undulator with the green and yellow the green and red magnets the electrons go through the through the the the undulator the electron follow the yellow path they wiggle up and down they emit light every time they accelerate that they emit light the lights get stronger and stronger they didn't act with the electron beam and then aligns up the electron so you see that in the lower charge here the electron cloud coming in is homogeneous and it goes along the undulator where it starts to get sliced up into into certain slices and at the very end you have a micro bunch the electron cloud where all the electrons emit light to hear and you get a extremely powerful x-ray beam coming out and then you throw away the electron so we only use the electrons once so here's the here's the european x-file in the nutshell at in hamburg at schenefeld you see the daisy barnfield campus where we start and then in schenefeld just across the the border to slagio holstein you then see the the european x-file so the electron injector starts in in barnfeld where the electrons are injected into the accelerator it's a superconducting accelerator so the accelerator is it's cooled down to two kelvin and that have special properties which i'll come back to here you see a small movie i hope you can see the movie can anybody tell me whether your movie can be seen no okay i hope you can see okay yeah you see the movie okay good so you see the electrons are being accelerated through the two kelvin cold naopin cavities then we have the the undulator systems were here in ostover born and then we hacked the experimental stations out here for research and you see here again this is the undulator system i showed you before the electrons are passing through the undulator and then you can can follow them this is your undulator and soon you see the the electrons are going there you can see due to the magnetic structure they wiggle far from back far from back and you see the the electron the light is being emitted that you see as a the the white cloud and then at the end they then send out a big a big burst of photons and here's the experimental station the experimental hall where we do experiments so you have three undulators we call them sars one sars two and sars three and then we have six experimental stations where we use this very powerful exo beam coming out to do experiments you see here the the two beam dumps where we throw away the electrons called xx du one and xx du two that's the the beam dumps and then we have the six experiments where we do a different kind of of experiments i'll i'll come back to that here's a typical instrument station the x-rays are coming from the right going towards the left you at the first you see some some various optics where you can manipulate the x-ray beam you can you can focus it you can measure how intense this is then in the middle with this blue bar you have an experiment where the x-rays x-rays are scattered off a sample then you have a a detector and i will come back to the the detector here this is this bit piece with the yellow front and then you have a monitoring system at the very end so this is a a a a a a typical person will be as as as big of as this detector this robot arm is actually to um manipulate another another detector the The sample is sitting just in front of the robot arm. And here's the detectors. I'll come back to the detectors later on. They produce a tremendous amount of data, which we then need to analyze. This is a large pixel detector, a 1 million pixels, and it's delivered by the UK. And here's the people we have all this for. This is a user community. So we are a user facility where people can book in and do an experiment via proposal. A typical group here, you see six, seven, eight different groups, they come in and do an experiment for about five days or 60 hours, five times 12 hours. You either measure during the night or during the day. And in the end, we hope that we can do 200 of these user experiments per year. So a user group is basically 10 to 100 people consisting of different universities or different institutes that come and do the experiment together with us. So here's a little bit about the facts of the European Exfol. As I said, it's a user facility. So the science we do is determined by our users. We have 12 member countries. Germany pays a little bit more than half. 58% Russia, which is a little bit unusual, pays 27%. So about a quarter. And all the other countries, the remaining 10 countries, contribute between 1% and 3% of the construction and operation costs. Daisy runs the accelerator for us. We have about 500 people on the Schoenefeld site for doing the experiments and the maintenance and administration, et cetera. And Daisy has about 250 people running the accelerators. It costed about 1.5 billion euros to build it in 2018 prizes. And the running cost is about $135 million a year. And you see our end goal will be to do about 200 user experiments per year. We are not there yet, but we're all a bit operating for about three years. So where are we in the world? We are not the only one who have a hard X-ray frills or laser. The first one was in Stanford at LCLS. In 2009, they opened. Then the Japanese came two years after in 2011. Then the Swiss in 2016 and the Koreans also in 2016. And the number you should take note of here is how many X-ray pulses per second will come. The X-ray facility is a pulse source. So you get a number of pulses per second. And these facilities deliver between 60 and 120 pulses per second. Here, European X-ray is very particular because we have the superconducting accelerator where we have a megahertz punch structure. So we have a burst mode where we have 10 bursts of very strong X-rays coming per second. And in these bursts, we have up to 2,700 X-ray pulses. So in total, we can deliver 27,000 X-ray pulses per second. But they come in this burst mode where every 10th of a second we have a new burst of X-rays coming in. And that makes us the most powerful X-ray source in the world. So we want to measure atomic motion. We want to see how atoms and molecules are moving. We want to measure on very short time scale. But what is the time scale of the atoms? Now, for humans, you would say the time scale, our time scale is actually milliseconds because that's sort of the time scale we move on. If you drive a car and the car in front of you stops, you need to react within milliseconds. Now, the interesting time scale, while the interesting time scale for humans is milliseconds, it's femtoseconds for atoms. And you can see that by the speed of sound in the material. It's 3,000 meters per second. The distance between atoms is 3 angstroms. Then we can say that sort of the interesting time scale is about 100 femtoseconds. So that is the time scale you want to do experiments. If you want to see atoms move, as you see in this acoustic vibration, you see up there. So that's the interesting time scale. And that is exactly the time scale of the pulses coming out of our fritillation lasers. If you look at the singleton-racing light on the left, which is a picture facility at Desi, there you get about 10 to the 9 photons light particles per pulse. And the pulse length is about 100 picoseconds, which is much longer than 100 femtoseconds. So the fel will give you 10 to the 13 photons in about 2 to 150 seconds. So we have an enormous amount of photons on a very short time scale. And that's the difference between X-ray lasers and singleton-racing. So we measure the dynamics on a very short time scale. This also makes that we have an extremely and exceptionally strong X-ray beam. Normally, you would say you can stop X-rays, by, for instance, lead that can shield the X-ray beam, so it's not harmful. But with this X-ray beam we have, which is a very strong laser beam, it can be focused to a few microns if you do that, then you can drill a hole through a piece of steel in less than half a minute. You can see here a piece of steel on the left, it's 5 centimeters thick, and you can see the entrance hole to the lower right and the exit hole to the upper right. And within half a minute, the X-ray beam has basically drilled a hole through this piece of steel because it's so intense and it comes with such a short pulse that it drains the plasma and the material evaporates. So if you focus the beam, nothing can actually stop it. And therefore, we have to have very special safety requirements around our experiments. Now this here, I don't know whether it actually works. This could show you the X-ray beam where we have a tube on the right. And now I let it run and then I just hope you can see it. And then you would see in a few seconds the X-ray beam come. So this is the X-ray beam that goes into air. Ionized is the air. And it basically drills hole in the air. Penetration here is much longer than who would have anticipated. And the noise which you might hear is the 10 hertz noise. So every time you see a burst coming in, the ionized is the air and shows this kind of light here. I hope you saw it. Did you see it? Can somebody tell me where you actually saw this small movie? Yes, there was some movement. The background played a bit into the audio, but OK. OK, good, good. OK, so how do we then do the experiments? Now I told you about the whole setup. Now I go a little bit back and look to see how did I do experiments when I was younger? How did you do data collection the traditional way? So I took here the example of the double headaches. On the left, you have Rosalind Franklin, who was the woman who actually took the first diffraction data of DNA, which you see in the middle there. Diffraction data. You saw a sample, which was a kind of a few crystals. And from such diffraction data, you could actually solve the structure. And you see the double helix structure here with James Watson and Frank Siegrig to the right. By the way, our guest house at the Jennifer campus, you have Rosalind Franklin on the front of the building. But basically, in those days, you had one crystal, you took one diffraction data, and you took one crystal off. So this is how I looked in the 80s, early 90s. When we did the experiments at DC, we had one sample, which we spent. We prepared that for weeks, and prepared, and shine, and so in this case, it was a precious gold sample. Then we studied this only sample for one or two weeks, and then we went home with a small data set afterwards. This is not how we do the experiments today. We do it in a radically different manner. Now, but how do we do experiments? Because I told you this exceptionally strong beam that will drill holes in a piece of steel. So how can you study anything with such a powerful beam because it will get destroyed? And yes, the samples are destroyed. And this was shown here about 20 years ago in one of the first studies. Can you at all study samples with such a strong beam? And here is a sample of a protein, the exploding protein, where you need to beat the radiation damage, because I showed you that the sample will evaporate. So can you take data from the sample before it evaporates? And here you see a protein you put into a beam, and then you see after the beam hits it at 0 times equal to 0, then you see how the sample looks at 2 femtoseconds, 5 femtoseconds, 10 femtoseconds, 20 femtoseconds. And it shows you have to be extremely fast to actually do the experiment to after reconstruct how the sample actually looks. But you need a very short pulse in order to do that. And that's exactly what we have a pulse, which is a few tens of femtoseconds long. So can you take the picture of the sample? So the first experiment here was done also at Daisy, where they used a smaller x-ray facility than the one we have, where you had a piece of silicon, where you had an engraved two men and a son. You can see a scanning electron microscopy picture of it here. Then you put it into the sample, put it into the beam, recorded the diffraction image, and you see that over here on the upper left. And from that diffraction image, you can actually reconstruct how the sample looked, and that you see to the right, where you see the reconstruction of the two men and the son. Now, if you take a second shot on this sample, then you see the scattering image on the lower left, where there's actually nothing, because the sample has completely evaporated. So it shows the first beam is so powerful that it destroys your sample, but you can actually collect data fast enough that you can reconstruct the sample and calculate how it actually looked. So this was an extremely important proof of principle that you are able to take your scattering data so fast that they can actually see how the sample looked. So the way the experiments are done today is done in a completely different manner. So here's a sketch where you have this kind of yellow thing coming in, and these are the extra pulses, which are coming in this burst mode with 10 of these extra pulses per second. Then you have a jet of sample, which is shown in green. These are small crystals of some kind of protein you want to study. And then you look at the diffraction image to the upper right. And then you just shoot. Instead of only one sample, you have millions of samples and millions of scattering images. You don't know how the atoms are looking, how they're organized, but that you can sort of afterwards if you have a sufficient amount of data. So here's what you call in several 50 second crystallography. It's the same experiment. You have a lot of different scattering images here. You have the samples coming in for this jet GDVN that the jet coming in. You can see if it works. I don't know how the movie works. You can see a real jet on the lower right on my screen. It's a movie where you can see a jet coming in from the right. When the X-ray beam hits the jet, it evaporates and there you see it gets assisted up. And from these millions of scattering images, you can then calculate a structure of this protein they wanted to study here on the right, where you see the different atoms and molecules. But you need to collect a lot of data. So here's one example of science that has been done where they did a proof of principle. Instead of looking at proteins, you look at millions of gold nanocrystals. You can see them on the upper right. You then spray them very fast into the X-ray beam. And then you collect just everything you can. In this case, 10 million diffraction images were collected. And then you sort out the data afterwards and see how did your samples actually look. And then you can classify in this case that you had four different kinds of samples that you can do afterwards. You don't need to do that in forehand. You just take all the samples you have, throw them into the beam, and then via the software afterwards, you can then calculate what kind of structures they actually are. And that is what I mean, that it's a completely different way of doing science and instead of one sample, you have millions of samples. And then you look afterwards what kind of samples did you actually have and it doesn't need to be homogeneous. Here's a maybe a little bit clumsy overhead information of water bubbles and experiments that were done about a year ago at the European X-well where they looked at water bubbles, creation of water bubbles. How is a water bubble actually formed in the very first nanoseconds or femtoseconds after it's created? And they had a laser to create water bubbles. They looked at it with X-rays and then they just look at millions of water bubbles and then see by looking at the images of the water bubble how they actually grow, how fast they do what's the density and they can see that the water bubble is completely empty in the middle and just at the rim of the water bubble, the density of water is 40 to 50% higher than normal water. Then you can study the statistics by looking at these millions of water bubbles. Another type of experiments we're doing is we're looking at fundamental interaction and time scales, something which is also important for you is a magnetism where you store data. You need to understand how is the interaction between electrons, the spins of the electrons and the lattice and to see how is this interaction in order to design new magnetic devices that can switch much faster than the ones we have today. So magnetism is actually one of the key areas that we are looking into. And here we had an experiment where we looked at magnetic modulators by a group from Itmo in St. Petersburg where you can see the magnetic structure is changing as a function of time on the picosect time scale. So I gave you some examples of what we're doing. The sort of what we really want to do is to look at the molecular movies where we, for instance, trigger a process in a photo-action protein. You see that on the right where you have a protein. It's excited by an orange optical laser. Then you have a blue X-ray probe coming in. You scatter off and from that you can then calculate the structure of the protein. On the left, you see the way we're doing experiments where we put in many, many proteins into the beam, measure a certain time delay, calculate the structure. And if you then measure several time delays, so you see what happens after 10 femtoseconds, 20 femtoseconds, 100 femtoseconds after you hit the protein by the optical laser, you can then put together a number of still pictures and make molecular movies. And that is what we really want to do at our facility. And the first of these kind of studies were done here. This is a paper from, oh, I think it was this year by Mario Schmidt from Arizona where they looked at the photo-active yellow protein where you can see from the left. It's a three picoseconds, 10 picoseconds, 30, 80 and 100 picoseconds after you have excited this protein with a laser, a normal optical laser pulse. And that's important to understand how photosensors actually works in nature in these kind of proteins. So that will be one of the main focus areas for our facility. So here you can see the data that has been taken at DOP next level. Good, so now I told you a little bit about the science we are doing. I hope you've got a glimpse of the biology from nanoscience to magnetism to biocrystallography and we do many other things. But maybe I'll also show a little bit about the complexity of experiments we are actually doing with a special focus on the data. So here you have all the different or part of the different things that you need to have control of. You have the accelerator which actually works exceptionally well. We have X-ray optics to focus X-ray beam. We need to measure photon diagnostic to see why is the beam how intense it is. We need optical lasers to excite the samples as it does show you. We need a sample delivery, this jet of sample that is coming in with a very high speed. And that's also a lot of technology. And then we need our detectors where we measure the scattered photons and then we need to analyze the immense amount of data we're actually taking. And we have a data department which is in charge of detectors, the electronics, the controls, the data management and the data analysis. And we are about, typically about 100 staff working on this at UP NextFirm. So here's the, Judge and Pete, here's the way the experiments is done. You have the high rep rate of X-ray bursts coming into the, hitting the sample jet. You have the detectors where you collect all the data and we collect free files of 520 frames per second in this detector here. So here's the adaptive gain detector. It can switch the game. It's a detector which has about a million calibration parameters. So it's a very complicated beast. Here's the pulse structure. And we have these pulses coming in in this burst mode. And within the burst, you don't have time to read off the detector. So you need to store, this detector can store 352 frames on the detector. And then between the bursts where you nearly have a tenth of a second, then you can read out all this data and store them away. So here's the piece of this, AGIB detector, here's the second one coming. So it's really, this is the first experiment done with this detector, you'll see we're very pleased. And it produces a lot of data. Now, I'll go a little bit back and look here. When I started data analysis in real time, so when I started here, you see on this plot on the left, the data came out on a piece of paper, this tailor type, you see the guy on the right. And then we looked, we had the numbers plotted here on this piece of paper, took a pocket calculator and then calculated how many counts did we actually have in this peak as you check the background. The data didn't come much faster than you could do that with a pocket calculator. The data storage, by the way, was a little bit old fashioned because we actually stored the data on paper. So you saw this guy on the right, where he's doing some data analysis, looking for a particular scan. And then we collected this data in boxes and then we drove home with 10 boxes of paper, which was the data we actually took. And then we have analyzed the data on the way with the pocket calculator and nicely written them down into a log book. Here you can see, we also got a little bit more advanced and made a small script where you can actually automatically calculate the intensity of the peak. We are far from that now. We have, there's been a tremendous development. We have, when I started, we had detectors that measure whether it was a photon or not. So today we have two dimensional detectors where you can also see where on the detector did the photon actually hit. And they are time-resolved. We have detectors which measure now with the megahertz repetition rate. So here's the data storage when about 10, 10, or actually 10, 20 years ago, we had the more data storage about 2011. And I still remember that when I went home from an experiment samples in Switzerland with an experiment, we have bought 10 of these data devices and we carried them home in anti-plastic boxes. So we say this is completely outrageous today. So here's how it works at European X-File. We have on the left, you can see how much data we have collected 2019, about 10 paid about a little bit less in 2020 due to the pandemic. So we could not really collect data. You can see on the plot on the right, you have the data size in petabyte, then you can see the date and then you can see how the amount of data we have stored, the raw data we have stored is actually increasing. And every note here is where the flat horizontal plateau that's where we stopped the summer due to the COVID-19. We had no experiments running. And all the vertical lines, that's where the agent detector starts to work. And then we really produce a lot of raw data. So this is one of the biggest challenges we have is to manage and manipulate and analyze a huge amount of data. So where are we actually with this here? You can see the detectors we have the various, we have several types of Egypt detectors. We have a D detector I also showed you and could see sort of the data rates that we are collecting the amount of raw data. And for comparison, you can see, I also listed here the amount of data that comes out of the CERN detector. So we basically produce data at the moment is at the same amount of CERN. Now, one of our biggest challenges is data reduction that's built in in the detectors at CERN that they only measure the good events where we collect all the data that the detectors actually measure without checking whether it's good or bad. And there's a special challenge here that physicists are emotionally attached to the data. Now, as soon as they have the data, you can take away their husband on their wife. You can take away the kids or whatever, but you cannot take away the data. Then they take you to court for the human rights. So if you have to analyze the data and produce the data before you give it to the users and that's one of the challenges that we're actually working on up because at the moment we have the data wage which is quite large. We also have to the different requirements we need to store the data. We need to process the data. We need to transport the data. We have local resources at UP Nextville. Then the data is transported to Daisy nearby where they also stored. And then we have recently made a also fast data connection to NCBJ near Warsaw which can have a 100 gigabytes per second data. So they could also participate in the data analysis and data management. So this is a very interesting part of the challenge. We have a UP Nextville, the high data rates we are working on. Now, before I end here, I told you about some of the science we have been doing. I told you something about the data challenge we are actually having. And please, if you're interested, look on our website if you want to look for jobs. We're always looking for good scientists and data and engineers. But then I'll just end up saying we have six user experiments or six experiments. They're all in user operation. You can see them here. The first started in September 17. The next started in the end of 18 and the last they got online in 19 and now they're all producing data and data is being published. We have many high impact publications. So we had actually a very good start. So that was about UP Nextville. But before I end, I will also tell a little bit about why is it that we at all do science? What is it, but actually comes out. And here I will send up a few slides on the importance of large scale facilities. And then I will put the question here. What is the most important outcome of CERN? What's the most important thing they have done? The first of all, they have discovered the Higgs boson. And second of all, they have created the World Wide Web. So what CERN has been doing is doing at all large scale facilities are doing. We do science and we do technology. And I think you have to understand both things are extremely important for outcome of large scale facilities. I have here taken a little bit, looking at disrupting technologies over the last 100 years. And then we can speculate what's the role of large scale infrastructures. Now disrupt technology, I've taken a personal view here. So this is my grandfather Bartolt, born in 1800 and 1880 is something like that. He participated in the First World War, born in the northern part of Germany. That's how I'm now in the southern part of Denmark. The most important for him, the most disruptive technology for him was the car. Now the car changed his life. This is my father, Gunther. He was born in 29, 1929. And the thing that changed his life was the television. Now this is me, some years younger. And what changed my life was really was that the disruptive technology was a laptop. But I know this, I have been, when I did my master thesis, it was still on a normal typewriter where they have a correcting water, et cetera. I had all drawings had to be hand drawn and photocopied in. This is my son, Nikolai, what changed his life is the iPhone, the handy. This has been his disruptive technologies. This is my grandson, Aska. Something will change his life. Something will be disruptive, what he's doing that his parents and grandparents did not do. I don't know what it is, but there will be something. Now, if you look and see, this all these here contains basic sciences. The car, there's a lot of metallurgy and thermodynamics. The television, there's a go, there are Nobel prizes into the television, the discovery of the electron, the transistor and stuff, the integrated circuits in my computer, the CCD camera in the handy. I'm pretty sure that whatever will change the life of Aska, there'll be some Nobel prizes behind it. And our role in all this here is that we study new materials because all of these technologies I'll just show you, the basis of them is some new kind of materials, silicon or CCD devices, something that will create new technologies. Touch screens, for instance, is something that has completely changed our life. And we are in for do. We do here, our role is to do advanced characterization techniques. And I show you here, it's not only European Maxwell, there are singleton-relation sources, electron microscopes, there's neutron sources. We all are there to study materials and also biomolecules. That is what we are there for. And to end up with the very last slide is the COVID-19. I think you can hardly do a talk these days without talking about the pandemic. So I allow myself here to have a slide on the epidemic and large-scale facilities. You here have seen these COVID-19 viruses. You have seen them a million times already, but you might have not thought about where do these images actually come from because somebody has actually studied these viruses, the different parts of them, the different proteins, and then constructed these images. They are basis of science. And when the COVID-19 started, a lot of different groups started to study the various proteins, the viruses, the electron microscopes, and with singleton-relation. And here's some of the very first articles that came out in Science and Nature in March, April. The Chinese immediately went to their singleton, studied some structure-based, the sign of candidates targeting this COVID-19 virus. There was a very nice study done in Germany by the Gruppe Wildenfels, the way he also looked at some crystal structures. All that was done at the Basie II Singleton in Berlin. And of course, the Americans also used their advanced photon source, and our synchronization source in Argonne to do some of these studies here. And at least every region needs to have one of these facilities to do this kind of science. So what I say here, we have a vaccine based on science, a vaccine which I'll be proud of taking. I know it's based on a lot of science that has been done. And as I said, every region needs to have some of these large-scale facilities. Could you imagine that Mrs. Merkel would have to call to China and say, oh, by the way, we have to study this. Can we borrow your singleton? Or could you imagine Merkel would call Trump and say, oh, could I lend you a singleton for a couple of days to study this protein? No, it's unthinkable, unthinkable. Every region needs to have these large-scale facilities, singletrons, x-ray fields, neutron sources, and electron microclips. An essential part of the infrastructure to be able to attack such crisis that we had with the pandemic. And I think with that, I would say end my talk. It's a little bit too late to say Merry Christmas, but I can at least say Happy New Year. And now I don't know what's going to happen. OK, thank you for your talk. Now we'll have a short Q&A session. There have been some questions from the audience and Twitter and IOC. I think most are asking probably a bit personal question, what made you choose this venue for your talk? Well, most probably don't expect the chairman of the International Research Institution at a Hacker Conference. So we're glad you're here, but why did you choose it? I said I want to tell what we're actually doing. I want to tell what kind of science we're doing. And then I hope I give you a flavor of the challenges that we have with our detectors, with our electronics, and with our data analysis. Yes, and another question. What practical applications would the financing partners be mostly interested in? I think you have partly answered that one before in the last two slides. I think the financing partners, what they are interested in is they want to be part of the most advanced technology you can actually do. And these are accelerators, free interlaces, detectors, because only by participating in this kind of advanced infrastructure you really get the technology. You need to do it yourself. And then you also, they open up, these facilities are then a tool for the user groups, for the university groups, that they could do the most advanced science that is possible these days. And thereby also train in the best way their students. OK, and then one more. In one of your slides, you had this tiny bubble called x-ray optics. How do you do optics with a beam that can evaporate on its steel? Yes, that's a very good question. Actually, the beam, when the beam is born, then it's about, in the undulator, it's about some few microns big. But then it's slightly divergent. So then we let it run for about a kilometer. And after a kilometer, the beam is bigger. And then it could be maybe a millimeter. And then you can put in a lens. And you can now have a lens or a mirror, or a lens to focus the beam again. And when you focus, and then there, the lens can actually take the beam. The lens is made up of a beryllium, which absorbs the x-rays. Or you have a mirror, which is reflecting all the x-rays. So very little is penetrating the mirror. So there you can actually handle the beam. But as soon as you focus the beam, you cannot stop it anymore. Yes, OK. That concludes the questions. Thank you again for the talk.