 So good afternoon to everybody. Today we will continue with some special topics of neutron imaging using energy selective option at the state sources. And after that, Robin will present the same technique at post sources like spallation sources at ESS, SNS and so on. So I'll start with the consideration here that the wave particle duality is the concept in quantum mechanics that every particle may be partially described in terms not only of particles but also of waves. In this way, actually the particles, the material particles like neutron also can be represented as a wave propagation. So it has wave properties and we can assign wavelength and frequency to neutrons with certain energy. So therefore it is possible to plot such diagrams where we have the energy distribution of neutrons coming from the neutron source or the so-known spectrums. Spectra in terms of not only of energy but also of neutron wavelengths as shown here. So we see here the representation in neutron wavelengths of thermal and cold spectrum which were measured at the source fRM1 in Munich when I was a PhD student there. So we see the broadening of the cold neutron spectrum in the direction of the longer wavelengths. So this means lower energies. In addition to this, we can see that the transmission properties of probably crystalline materials like iron here, iron material is also influenced from the wave nature of the neutron in order that the strong changes in the transmission can be related to diffraction contrast. This means that the wavelength of the neutrons is in the range of the latest spacing in the polycrystalline material or in the crystalline material. And therefore for certain wavelengths we can expect also diffraction. And the diffraction signal is actually not recorded by the neutron imaging detector. So that's why we have some loss of intensity as shown here of transmitted intensity through the sample. But I'll go in details to this later. So how this looks like when we image, when we make an image of polycrystalline material at certain wavelength, for example, here at around four angstrom, we see that there is a very strong change in the transmission property of the material or this is actually the attenuation coefficient shown here. So we have very rapid and big change of the attenuation coefficient for iron around four angstrom. And here is an example of radiography at four angstrom of a weld between two steel plates where actually the structure of the weld can be seen quite well. So the contrast what we are observing here can be related to diffraction contrast due to the fact that in areas where we see some dark contrast or let's say where we lose intensity, in these areas we have some diffraction where the neutrons which are going for, which are scattered from their primary trajectory are lost for the imaging process. And therefore we see, we observe some decreased intensity for these areas. For example, you can see here and here and so on. So of course we have some, we cannot have perfect monochromatization of the beam. So we have some wavelength resolution where actually the delta lambda is the spectral broadening of the beam. The spectral broadening depends on the way of neutron monochromatization or how we select certain wavelength from the polychromatic neutron beam or from the white beam coming from the neutron source. There are different ways of monochromatization. I think Robin spent yesterday some time on this, but here I'll mention them again. So these methods for monochromatization are typical for steady state sources. For example, velocity selector or double crystal monochromator. And here I'm showing also the time of white method which is typical for post sources, but it can be also implemented at steady state sources using choppers. So we have continuous beam coming from the steady state source. And then with choppers we can chop the beam in post sequences or we have some post beam structure which actually propagates. And with the time we have just extension of the pulse where actually the faster neutrons or the shorter wavelengths are arriving first at the detector and the low energy or longer wavelengths are arriving later. So the very important feature here is that we are needing or it is essential to have a detector which can take pictures in very short times in order to make the discrimination between different neutron wavelengths. So our focus on every of these kinds of methods for monochromatization starting with the velocity selector. So this is how the velocity selector looks like. It is a turbine which rotates at about 2000 revolutions per minute and it can select actually certain wavelength because it allows neutrons of defined velocity to pass through these channels of the turbine. So these channels are defined by this lamella or by the blades of the selector. And then at certain velocity of rotation we can accept one velocity of neutrons passing just through this like a screw shaped structure. So in this way we can just set conditions where only certain velocity from the neutron beam is accepted and at the end we have just transmission for neutrons with this certain velocity. This means that we can select certain wavelength from the beam. So this is a very important feature Blades or this lamella are manufactured by are just manufactured by very light material like for example, some plastic material which is coated with a strongly neutron absorbing material like boron 10. So every neutron which hits these blades will be absorbed and in this way we get the chance only of get the chance for neutrons having defined speed to transmit this selector. So I can see here on the image how the experiment arrangement looks like we have a neutron guide where the neutrons are transported to the certain position. So this is the velocity selector. It is very compact turbine and here it is the detector with a sample in front of it. By changing of the angle of the velocity selector in respect to the beam, we can tune actually the wavelength resolution and select a broader or narrower spectrum coming from the beam. So this method is quite good when we need actually high intensity and low wavelength resolution. So as you can see here on the bottom, so the resolution of this method is about between 15 and 30% which is quite let's say a bad resolution but we transmit quite large number of neutrons because we are accepting quite broad energy band for our experiment. So this method is very beneficial if we are needing actually a lot of intensity and we can just relax the condition for monochromacity. So some methods don't need really very strong or very, very defined wavelength for the experiment. So for such kind of experiments, we can use this method. The next device which can be used for monochromatization is the double crystal monochromator. And here we can improve the resolution down to 3 to 10%. So here we have much better wavelength resolution due to the fact that we are using two monochromators or two monochromatic monocrystals but in order to gain a bit more intensity and relax the resolution, we can use monochromators with a certain mosaicity this means that we have just not a single crystal but combination of small crystallites, small single crystals which have some preferential orientation but they are misaligned a bit if we consider them as a plane. So this means that the misalignment between the crystals can be in the range of 0.8 to 3.5 degrees in respect to the surface of the monochromator. So in this way, we can make the band or the beam spectrum, the band spectrum which is transmitted or reflected from the crystals much broader. And in this case, reducing the wavelength resolution we can gain more intensity. So if we use a single crystal monochromator then the resolution will increase and will become below 1% but with the mosaicity of the crystals we decrease the resolution, let's say and gain more intensity which is very important for neutron imaging experiments where the intensities are not very high. So how this device works actually from the polychromatic beam we reflect with the first crystal a certain wavelength which is dependent on the orientation of the crystal or rotation of the crystal. So using different rotation angles here we can select different wavelengths from one to six angstrom using Pyrrolithic graphite monochromators. And then with the second crystal we can double reflect the beam and keep the same beam orientation as the initial beam which is coming from the reactor. In this way, just by rotating the crystals and translating the second analyzer crystal it is possible to set wavelengths from one to six angstrom, for example, continuously. So this means that we can perform even scans by just setting different conditions for the orientation of the crystals. So this method is very, how to say very frequently used at the steady state sources, neutron sources because we can tune the wavelength in a quite broad range. We don't need the, let's say, a very sophisticated controlling system for velocity selector where you need a vacuum and also water cooling for the velocity selector here. We have a quite simple mechanical device which is very robust. And even here we can tune the wavelength resolution by changing the crystals and using different mosaicity. So for the time of flight method which I already mentioned that can be used also even at steady state sources, we need the pulsed beam which can be just produced from the neutron source. If it's a pulsed neutron source or using choppers at the steady state neutron source and this pulsed beam is propagating at a certain distance where actually it is transmitting the samples and behind that we have a time resolved or detector with very high time resolution which can take pictures in very short time. And after that from this stack of collected images we can reconstruct the wavelength where this image was taken at. So here you can see we are in the range of microseconds. So this experiment, performed experiments from Anton Trenzin was performed with multi-channel plate detector. And the resolution which was achieved here is microsecond. So at the end we are able to see the change of the contrast or the transmission through the samples with very high wavelength. Resolution of below 1%. So you can see here that with this method we can achieve resolution of 0.1 to 1%. So Robin will go into detail after my talk and just focused, will focus in this method and give you a lot of examples. So how these different types of monochromatization look like when we compare them. So here you see the transmission through iron plate where actually the neutron monochromatization was performed with velocity selector, double crystal monochromator with mosaicity of 3.5 degrees and double crystal monochromator with improved mosaicity or less mosaicity of 0.8 degrees. So you see that with the velocity selector having wavelength resolution of about 15 to 30% which is quite a bit. We have quite smeared spectrum or transmission spectrum or here are the attenuation coefficients actually for iron. So the wavelength dependence here is not really sharp. We have roughly, we can guess the shape or where we have the very strong rack cutoff in the transmission. We can catch it with the velocity selector but of course the smearing of the transmission or this rack edge curve is quite strong. Using a double crystal monochromator we can catch even the small here cutoffs or edges and but if we compare this to the better mosaic or small mosaicity of the crystals we see that with 3.5 degrees we have much more smearing of the spectrum and with mosaicity of 0.8 degrees of the crystals we have quite sharp edge here. And therefore this method is preferential for rack edge mapping or imaging where we want to take pictures and reconstruct the rack edge shown here precisely. So some examples will be given later and here is a summary about the different methods of monochromatization. So you can see here the achievable wavelength resolutions which we have with each of these methods and also the exposure times at imaging facilities where we can take one image in a range of seconds when we are using a velocity selector with a double crystal monochromator increasing the wavelength resolution. This means that the intensity of the monochromatic beam goes down and then we have minutes. And in time of white because of the time stamping which is necessary just to be able to take position sensitive images in very short exposure times then the exposure is in the range of microseconds. So I mentioned already that in polycrystalline materials the neutron beam attenuation coefficient may always sum of its wavelength dependence to the fact that some neutrons are scattered out of the incident beam by a diffraction. So in this case at certain wavelengths in analog to analog to break peaks in a diffractometer break edges are observed and also say that imaging method is hence often termed a break edge imaging and it is carried out using a tunable monochromatic neutron beam at the steady state neutron sources. So here I would like to just give you an example I would like to just remind you again how the diffraction with neutrons works like. I guess that Robin has shown this already in his lecture but nevertheless, so in the neutron diffraction geometry we have incident beam, we have scattered neutron beam and here we have just the preservation of the moment expressed here where we have a preservation of energy and momentum of the neutron. And then of course, the break diffraction is obeying the break law where the the latest basings in the crystal are related to the wavelength through the angle of the incidence angle of the neutrons. So if we change the latest spacing by applying for example, tension on the crystalline material then the diffraction peak will move at a certain extent and it can be measured in a diffraction mode where actually the residual stress analysis method can help us to find such kind of shifts and quantitatively to estimate such kind of residual stresses in the crystal lattice. So usually in the diffraction experiments we are not interested in spatial information or in position sensitive information coming from the material. So we are using a scanning method so we can scan point by point and get the scattering spectrum or the diffraction spectrum for each point and then reconstruct actually the residual stress position sensitivity. For this purpose, the detector which is used here is a counting tube which doesn't have any usually any spatial resolution. In some instruments recently these kind of detectors were replaced by position sensitive gas detectors where the resolution is in the range of few millimeters but if you are trying to extract resolution in the range of micrometers as in the case of neutron imaging then you should replace this counting tube by position sensitive detector which we are using for imaging and this is shown here. Actually from the research reactor we are using we can select a certain wavelength from the beam coming from the reactor by a monochromator then we are directing the monochromatic beam to the crystalline material and in the diffraction experiment we are using a counting tube to count the scattered intensity at different angles. So this is the diffractogram which at the end we can use for estimation of the crystalline structure. So we can replace this counting tube by imaging detector and just observe the scattering signal scattered signal coming from this kind of crystalline material position sensitively. So this was done here in diffraction configuration at Paul Scherer Institute when I was a PhD student. So this was one trial how to use imaging detector in neutron diffraction experiments. You can see this is the monochromator in this drum. So the beam is directed to the sample and between the sample and the position sensitive detector here we have collimators in X and Y direction. So we can just project the scattered beam over the position sensitive detector. And at behind this imaging detector we have just the counting tube here in this container. So this means that we can measure simultaneously the image of the diffracted radiation and also to take the spectrum, the diffracted spectrum by the counting tube. So as a sample, we used a monochromator just one, let's say crystal which is reflecting the neutron beam. And it was known that there is some mosaicity in the crystal. So this means that there are regions which are scattering the radiation under different angles because there are some misalignment of different regions in the crystal. So here this is the result. So you see that actually the diffraction curve has two peaks at two different angles which are coming from two different regions in the monochromator, which are misaligned by let's say one, two degrees. What is the picture or what is the let's say map of the scattered radiation is shown on the bottom. This is actually the signal from recorded from the imaging detector and you can see immediately the areas of scattering position sensitively. So you can identify already the areas which are scattering the radiation at about 43 degrees. So it's in this area and the bigger area at about 43 degrees is this one. So this means that this monochromator consists of two regions or let's say three regions. Two of them are misaligned to the central region by 1.5 degrees about. So looking only at the diffraction spectrum, it is not possible to identify what is the distribution of these areas in the monochromat or in the this, let's say single crystal material, but by using of imaging, it is possible to see exactly which areas from the sample are diffracting or reflecting. And after that, if we are interested to, for example, to cut an area which is a single crystalline, then we can really see from the map which area we should cut and then we have a single crystalline material. So this was quite interesting, but the main thing what we learned from here is that imaging detectors can be used in combination with diffraction metals. So this was actually application of imaging detector instead of diffraction counting tube, but we can actually change a bit the geometry and use the imaging detector for a transmission. So this means here we have the scattered neutrons and if we take a picture from the sample in transmission mode, then this means that the regions of diffraction will appear as a dark areas. So in this configuration, this actually the diffracted neutrons or the diffracting areas are seen as high intensity areas because we are detecting just the scattered neutrons, but in case of transmission mode or transmission geometry, these scattered neutrons are no more transmitted so we see them as dark area. So this is the way what we want actually to explore. And for imaging purposes, we are using the double crystal monochromator in our case and then we can select one certain wavelength from the white beam coming from the reactor as I mentioned before. So if we put some polycrystalline material in the beam, then with the double crystal monochromator we can catch actually very precisely the shape of the attenuation dependence of this polycrystalline material with the wavelength and you can see here that we can really catch the break cut off very well and even have a good representation of the break edges at different wavelengths. So how we can use this effect for imaging? So if we take one sample like this step wedge and put it in the beam, then we can observe actually very simple, change in the transmission before and above the break edge. So if we perform an experiment with wavelength of 4.5 angstrom and wavelength of 3.5 angstrom, you see that actually here the attenuation coefficient is much higher than the attenuation coefficient above the break edge. So this means that here we have much better transmission because the attenuation is lower for iron. So this is just some kind of quality measurement So this is just some kind of qualitative exploration of this feature of polycrystalline materials to obey different attenuation coefficients and different energies. So we can extend this experiment and be more consistent just by taking images at different wavelengths. And you can see that for example for steel which is dominant or preferentially consist of iron or there is a big portion of iron up to four angstrom we have quite strong attenuation and above for angstrom at 4.3 the attenuation becomes much less. So this means that here we have better transmission. So for each element we have different behavior. So for aluminum you see that here we have the highest attenuation and then the attenuation decreases. So this means that for each element we can find this break edge or break cutoff at different positions which is actually the feature that can be explained to this feature with the crystalline structure of this metals. So just we continue with the qualitative exploration of this break edge effects or break edge shaped transmission properties of the materials. And we can play with this and just make like a magic and we can make that some materials lose contrast. So if we take copper and iron so here in this image is very difficult to distinguish between copper and iron. They are quite similar and similar behavior but if we look at the wavelength dependence of their attenuation coefficients we can find for two different wavelengths the same attenuation coefficient for copper as you see here or for iron. So this means that if we take images at these two wavelengths and divide these images one by the other then this means that due to the fact that the certain element has the same attenuation for these two wavelengths then it will disappear and this is the case here actually as you see for copper it's in the ratio of the images of 3.4 and 4.2 angstrom copper is disappeared. So we can make the same for iron and enhance the contrast for the other elements for example here for brass and copper we can enhance the contrast and we can make that iron is disappearing and you can see that polyethylene is also disappearing because it has also very similar attenuation coefficients for the very short wavelength range of less than one angstrom. So this means that playing with the wavelengths qualitatively we can change the contrast in our images and reduce or enhance the contrast for different elements it depends of course on what we want to enhance. So I'll go back to the chat and just I saw that there are a few questions. So here the question about the weld I'll go back to this example later when I'm showing the reason for the black edges. So the ratio of two images is just division so we take two images at two different wavelengths and after that we divide these images one by the other and in this case the transmission will be the same for the element the same for the two wavelengths if the attenuation coefficient is the same. So this is just a division of the pictures. So here we can use this change of the contrast in order to enhance some structures in different samples so this is an ancient Roman brush which was taken which was imaged by thermonutrients and we can guess that there is some structure inside in the brush and it was known that there is some crustacean in the brush made by silver so if we compare the attenuation coefficient for the silver and copper we can see that increasing the wavelength actually the attenuation properties of copper and silver just are very different and that's why at 6 angstrom we are able to see this structure with much better detail so this is the case when we just play with the wavelength so I'm going back to the chat are the attenuation coefficients sorry are the attenuation coefficients a different wavelength for different elements available somewhere yes there is a database and also there is this online calculator I don't know if Robin has shown this to you already but at the end of the lecture I can demonstrate to you how you can find out the attenuation coefficients for different wavelengths there is an online library at NIST which can be used I can... I shot it partly yesterday but I will also put the NXS plotter actually on the Indico page so people can download it with some examples as well yeah good Robin, thank you so let's continue further now I'm going to explain why we see the breakage Robin has shown this already but I need to show it again I mean it is for better understanding so if we have a polycrystalline material and with a double crystal monochromator we select a certain wavelength and send this wavelength to the polycrystalline material then due to the break scattering in the sample some of the crystallites which are presented here and have the orientation of a certain orientation to the incident beam will scatter this wavelength so this means that it will happen for a certain HCL family of latest planes and by increasing the wavelength then we are coming to the case where we have back scattering or this is the maximum angle of orientation let's say to the incident beam which we could have so this means for this HCL family of latest spacing the next wavelength or the wavelength which is bigger than this one will not cause any more scattering in this or will be not more scattered from this latest spacing in the crystal in this way actually we have very strong change in the attenuation properties of the material at certain wavelengths and these are exactly the positions where we are expecting to see actually break scattering in diffraction experiments and the positions of this of these edges are at if we consider that sinus from sign from 90 degrees is one so it is the double the latest spacing in the crystal so this means that in a transmission mode we have some sense about the crystalline structure of our material where the positions of the break edges are pointing out to the to the wavelengths where we have actually twice the latest spacing in the crystal so this case here is representing iron and you see that the maximum the maximum break cutoff is at about 4.1 angstrom so this is the maximum two times the maximum latest spacing in iron crystal so I'm coming back to the weld joint here between the iron plates and this is the transmission spectrum which we measured this is just the defined tabulated spectrum for iron and if we take pictures of this weld joint at different wavelengths we can see different pictures so this is a 3.8 4.0 and 4.2 angstrom and you can see definitely that at 4.2 angstrom we don't have any diffraction contrast more so this means that here the wavelength is too broad or too large in order to fulfill the rack scattering glow therefore here we have just pure absorption so no more scattering from the sample so what is the reason for these structures which we observe here so if we consider for example texture in the sample so this means that for different orientation of the crystallites we'll have diffraction at different wavelengths and it can be described here actually if we have a certain orientation of the crystallite at the angle theta so then for different orientation of these crystallites we'll have scattering so if we consider for example if we consider some texture then we can say okay the dark areas are corresponding for example to 24 degrees in relation to the incident beam so you can just make a mapping of texture in the sample if you don't have texture and see such kind of changes in your sample this means that there is some phase transition for example what we can expect here is that by the high temperature in the welding process we have some phase transitions and change from for example martensitic to austenitic state in the sample so there are different reasons for the contrast which we can observe in such kind of energy selective imaging and this should be studied in much more detail even with diffraction pure diffraction experiments so what we can learn actually from such kind of break edge imaging so what we can learn is phase transitions so you can see here from the dark or from the black curve just to the curve with the circles and the triangles we have change actually of the phase from I think from austenitic to martensitic state and of course the position of the break edge is shifting with the time so this is just by heating of the sample and taking just the transmission spectrum the other thing which we can where we have sense is actually texture because then we have change of the height of the break edge with the time strain can be also detected if we have shift of the break edges because as we said in the beginning the position of the break edges can be correlated with the position of the diffraction peaks in diffractograms so as it was shown in the beginning the position of the diffraction peaks or the shift of the diffraction peaks can be related to some residual stresses or some stress in the crystal lattice and here it will be just related in this case to shift of the position of the break edges so in order to be competitive to diffraction we need here actually very good energy resolution in order to pick this very small shifts of the diffraction peaks and the resolution which we are aiming in this case is just for example 0.4 angstrom which is quite high resolution and it can be achieved for the moment only by using of time of light method so I'm going to several examples now so how we can just use this information for position sensitive mapping of phase transitions in metals in this case it is a very famous example where Robin performed this nice experiment here at our facility so here a trip steel was mechanically how to say it was stressed by using tension and torsion and in this trip steels we have phase transition from austenitic to martensitic state which is which is dependent on the applied stress let's say in the sample or applied mechanical pressure so what we can observe here that in this phase transition we have change of the crystalline structure or of the symmetry of the crystalline structure from from example from austenitic to martensitic from phase centered cubic structure to body centered cubic structure so this means that the positions of the bracket edges will appear at different wavelengths so if we set the wavelength in a way that we have very different attenuation coefficients for the two phases the tomography experiment will show us the distribution of the two phases because the tomography is showing the 3D matrix of the attenuation coefficients so this is the case here and you see that in the torsion sample where we have torsion we have some radio dependent phase transition where at the most distant let's say position from the center we have phase transition from austenitic to martensitic state due to the fact that in the torsion we have the highest displacement between two points in the sample in a distance which is far away from the center so you see that there is some gradient of the phase transition with the radios in the sample where we have tension we have the highest stress actually at the position where we have some let's say deformation of the sample and here you can see also the phase transition map in 3D so this is a very simple method we need just monochromatic beam we should find a proper position for our wavelength and then we have automatically actually the different contrast for the different phases in 3D so it's very powerful method because the contrast which we have here is not related anymore just to the different materials but it is related to the different phases crystallographic phases of one material so here the main feature or the main effect is the diffraction contrast or the scattering so here we can compare this method with diffraction and we see actually very good agreement between the signal from the tomography experiment and just standard diffraction experiment so the question here is if we can compete the standard diffraction experiment no it's not the case because for these samples we were sure that there is no texture in the sample so coming back to this question which was asked before if we have a texture this means that the signal here or the transmission will depend also on the orientation of the sample which will actually produce artifacts in the tomography experiments where we are expecting that attenuation is only due to the different attenuation coefficients in the sample and we can see that there is a difference in the texture effect so here I mean I'll go a bit further so the texture effects can be seen also here where actually we rotate the sample in the monochromatic beam and you can see in the twisted samples then that we have changing of the contrast with the rotation so here the sample was multiple twisted and this was twisted less at the less degree so you can see how the contrast in the sample changes in the rotation so this means that actually the imaging can be used as a complementary method to the diffraction in order to identify which region of the sample can be used for diffraction for example because the texture is also a problem in standard diffraction experiments so here I would like to show more sophisticated method where actually we don't use one single wavelength but take tomographies at different wavelengths then we can actually see the break edges for voxels of our sample for let's say 3D map of the break edges in the sample and in this way it is possible actually if we select one voxel it is possible to take the first derivative of the break edge and to find the position of the break edge so due to the fact that the break edge position is different for austenitic and martensitic states then we could be much more quantitative and reconstruct actually in 3D the positions of the break edges which gives us the chance really to estimate the in percentage the amount of one or other phase under development for the moment so you see that the paper is quite recent so here another just example so Steve do I have 3 minutes more I guess yeah I guess so okay so this is quite interesting method because it's archaeology and archaeology is always very fascinating so we got some mesopotamian seals to investigate and we knew that they are produced by hematite mineral and from diffraction experiments it was found out that there is also magnetite amount in the samples so it was the question which we had how to identify samples containing magnetite in addition to the main material which is hematite and is it possible to map the phase distribution of these phases actually in the sample in 3D so these are the transmission spectra through all the samples and you see that there is some strong break edge at 4.4 angstrom and only for one sample this is not really the case so here we have some different structure so even in transmission mode we are sensitive to the crystalline structure of the samples and it was found out by diffraction that this is calcium carbonite so there was one fake sample which by monitoring of all the samples was found immediately so we see also in addition to this that this main break edge for hematite has some splitting here and one additional peak appears at around 5 angstrom here or not peak but break edge which can be related to magnetite actually with dashed lines I put the expected positions for break edges for hematite which actually correlates very well to the measured data and also this line or expected break edge for magnetite is correlating very well with this new break edge here so what we have done is just we set the wavelength at 5 angstrom and then we have completely different attenuation coefficients for magnetite and hematite so it was possible really to get the distribution of the two phases in 3D and be more quantitative in order to estimate the weight percentage volume percentage actually of the hematite and here I want to point out that there is another method of Newton beam of magnetization, the time of white method which will be presented more in detail by Robin so just stay awake and after short break I think Robin will step in and present this in more detail so thank you for your attention