 My name is Milko Aksic. I came from the, let's see, from the Ruzier Boskovich Institute, which is the National Institute for Research in Old Sciences, Chemistry, Physics, Biology and Medicine. So we have a good environment to do many applications. And my field of work is mostly application of IMB accelerators. You have last week probably several lectures describing IMB analysis and also you visited Ljubljana, which is facility that has very similar capabilities as we have. And you also listened about PICSA, RBS, these classical IMB analysis techniques. So my task was to try to find some special topics in IMB analysis. And today I will speak about the secondary and mass spectrometry technique, which is the only one, almost the only one, IMB analysis technique that can analyze molecules. So this is the outline of the talk. I will speak a little bit about IMB micropro, but you visited microbeam in Ljubljana, so this is quite similar. And I think even they have very similar equipment as we have concerning mass spectrometry. I will speak a little bit about interaction of heavy ions of the MEV energy range with material. And about this secondary and mass spectrometry, some examples and also some conclusions. And at the end I will give you some information about the access to accelerator facilities. Okay, so as most of my work is connected with micro probe, so that we are using focused ion beams, I will just try to give few basics of way how the ion beam can be focused to very small dimensions. Normal focusing of charged particles like electron microscope are using solenoid lenses, but unfortunately to focus any of the ions because of their mass you cannot use solenoid lenses because of that the only way or the best way to do focusing is using quadruples. And if you consider quadruple, one quadruple is actually focusing only in one dimension, so you have to have two quadruples, at least two quadruples in series in order to get finally focused ion beam. I mean, in all the accelerators you have a lot of quadruple systems that focus the beam for beam transport system. In the case of micro beam, it is important to have very clean quadruple field and many other requirements, but basically if you have some systems you can put it some kind of lens and the performance of the system will be mostly defined by the magnification. So if you have opening of the slit, let's say like a source, you have some square or aperture of let's say 100 by 100 microns. If you have the magnification of 100, then after this lens at very short range here, you may have 100 times less dimensions of the beam. So basically if you can put current of let's say 1 nanohm or 100 picohms into this field, all of these ions will come to the very, very small dimensions. Of course, ion beam focusing is very difficult because of many other things that one has to take into account. Ion source is important, of course the net current is also important. The aberrations of the elements involved are important, vibrations, misalignment, distances, this distance and this distance, vacuum and so many different things have to be taken into account, but I will not have time to do much about that. I'll just give you maybe an example of the traces of the ions that go through the system of let's say 2 quadruples or 3 quadruples. In the most simple way, you just focus, this is the doublet, so this is just like this configuration. First quadruple focus the beam, the second one is defocusing, this is X and this is Y, so X and Y. And depending on the ratio between the working distances, you can calculate somehow the magnification. In this particular example of very old picture from 30 years ago, this was in Heidelberg system which was very short, but today if you have doublet maybe you can use 6 meters or so and then you have the magnification of the order of 10 to maybe 50. But so far the Oxford couple triplet seems to be the best configuration and most of the microbeams in the world are using this kind of arrangement. There are certain varieties of this system that to display this quadruple a little bit further and then you can get higher the magnifications. But for a moment we don't need to go into details of the way how this is done. You just consider that at the end basically you have something like one micron focused beam. It's a triplet made by differences which are not equal. Yeah, this is probably one of the unique examples because two quadruples are from Oxford company both and the final one is from Melbourne. The reason for that was first that we didn't have money at the beginning so we bought just two but one of the important parameters to get high the magnification is this distance and Oxford lenses cannot be used to make a very short focus but these lens from Melbourne have a cut pole so you can go with the detector from the side. But essentially the important part is the same so the distance is 10 centimeters of each the distance between the poles is the same so essentially they look outside different but they are physically point of view the same only two power sources. One power source is for this and one for that. So because of this final lens we have only working distance of 11 centimeters because we could come with the quadruple much closer to the center of the chamber which is spherical here you'll see later. So this is the chamber we have. So this last lens you can even maybe see the conical shape of the chamber so that the pole can come very close to the center of the chamber. But you have seen the microbeam how this actually works. If you have some detector you collect the product of this detector you connect some certain energy or some certain product with the position which is determined by the scanner coils and by data acquisition you select all the peaks whatever the detector is to some images of the property that you actually looking with the detector. In this case this is particularly the use extra emission but any other detector RBS gamma ray whatever you are using you can create in the same way these images. So one thing which is when you are doing ion beam analysis you have to try to take all the possible processes into account and try to get the best about the sample on the basis of the product or on the basic of this interaction. So if you look at the stream traces of the ions that go through the material you see that the majority of events is ionization and only very few at the end of the scattering with nuclei not in terms of nuclear reaction but just the scattering of atoms together so they are just I don't know the best word but they are displaced actually they are displaced from the position these green parts are displaced of atoms from their original position. So if you look at the end you have this final green is distribution of defects which is always at the end of the range and then you have ionization and the curve depends really on the energy in this case we have oxygen of 4 MeV so majority of the ionization is similar over the depth if you look at the let's say protons you'll have this break peak at the end which is similar close to the break this damage profile peak. If you go to the different processes you can combine them let's say all those that are based on the elastic scattering you have rather for best scattering and elastic here they call detection so this is the elastic scattering of the target ion with the nuclei if you make ionization then you can create inertial ionization and create x-ray spectrometry or if you create some nuclear reaction this is a big end on a nuclear reaction analysis but there is also one technique that relies on the ionization which is scanning transmission ion microscopy if the sample is in some relative thin you can by placing detector here you can measure energy loss in this part I'll tomorrow say something about that but all these processes are not sensitive to molecular content of the ions of course there are some effects in a pixel spectroscopy the x-ray lines are a little bit distorted there are some fine structure of the K-beta, L-beta lines and where you can maybe say something about molecular structure but in terms of analysis of molecules none of these techniques can make this so I'll go now back to the smaller energies to this energy loss curve and to this particular part where the nuclear stopping means scattering bit nuclei are the highest so these are few Kv region I think that this is some heavy ion like xenon or something like that and if you look at this part what is nuclear stopping means that the primary ion makes a lot of scattering with the surface layer of the sample and some of these ions are ejected into the system, into the vacuum but if you make not just one ion but you have beams of heavy ions of Kv region you are making such a huge interaction with the surface that it spatters away part of the sample so it is like a drill it is spattering material from the surface unfortunately in mass, secondary ion mass spectrometry which is based on this principle you create a lot of different ions so these are not necessary molecules these are individual atoms these could be fragments of the molecules or in very, very few cases they can be molecular ions and if you look at the system this secondary ion mass spectrometry seems to be a basically tabletop instrument with the ion gun from the side you have a sample and then you have spectrometer you are first pulsing the beam over the aperture you are sending the beam very well focused to the sample and then you have extractor which is few kV potential that will suck in the ions into the spectrometer so these ions that I extracted are going to this ion mirror and are reflected to the detector and once you have a pulse from the beam goes through this aperture the time starts and in spectrum you are having different times when the ion will arrive so this is time and as you are going further in time you are having heavier and heavier fragments coming to the detector the problem is in top species that you have a huge amount of peaks in some cases this is a very nice technique the majority of cases this is really excellent technique and can be used in different applications but if you are looking at some organic molecule which is large big molecules you will have in this case this is polyethylene some organic compound only this peak belongs to the whole molecule all other peaks belong to the fragments like this part is green this orange one is this peak and all these are different fragments of the molecule so in order to analyze this spectrum you really need a lot of experience a lot of databases and to explain the system so yeah oops now what is that sorry yeah student can you explain about the mechanism of extractor and ion mirror in this part okay so you have short pulse of the beam beam goes through this is scanning through the aperture so in certain point defined by one nanosecond let's say you have now ions are going through when this ion goes through you hit the target and you extract you eject a lot of atoms from the surface and then you apply 5 kilowatts let's say negative to attract positive ions and they become they start to fly in this tube when they come here to the this mirror electrostatic mirror this is probably so if we have positive ions it will be positive they will be reflected to the detector I will later explain why this is needed let's now consider that we have just detector here so the start probably this is hydrogen the first ion that can come will be hydrogen it's just proton it will be fastest to come and then later on comes heavier and heavier fragments which are slower so we are measuring this is time of light and then as they are coming more peaks are forming and then you have of course repetition many times many times hundreds, thousands of times in one second to do so so actually here is you are from Lebanon but the first time I belong to Sully you know him about 1999 proposed to buy mass spectrometer for accelerator in Lebanon I said come on why you need mass spectrometer for accelerator but he was right unfortunately because of situation Lebanon that was never purchased but that was the first idea that I heard about to do mass spectrometer with accelerator and later on I'll explain in next slide maybe I can skip to now okay I will come to that later on so what about MED ions can these MED ions also do the same so the difference between KV ions that are sputtering this material and MED ions is that nuclear stopping is completely low so the probability to scatter whole atoms is very low but we have extremely high electronic stopping power which means that a lot of ionizations are made and this is big difference you know this is like every ion will create let's say iodine of energy of one MED per ammeter so let's say 120 MEDs iodine will create 10,000 charge pairs in one nanometer so this is a huge concentration of charge and this effect can maybe give us some way how to ionize the whole molecule so what is going on this is a slide about radiation damage in material by arms but what I wanted to show is this one when you have heavy arms going through the material it creates a lot of ionization so this is also one kind of artistic image of what is going on when this heavy ion goes through and one of the first application of this property is that if you are going through the let's say organic material you create almost holes so you create a lot of damage in very very narrow tube along the way that ion is going through so this process could be the basis of analysis of masses so what is going on the first mass spectrometer based on heavy ions is PDMS maybe 30 years ago so 74 so even more californium source produce a lot of fission fragments with the energies of about let's say 70 to 100 MEVs so these are huge energies and when these energies come to the sample ions from the surface, molecular ions from the surface are disorbed and they are again accelerated to the detector so this is something similar as today's technique MEV sims 2009 MEV sims, Japan Kyoto University this was the first accelerator based mass spectrometer so in this year I reminded discussion with Bilal who had the same idea 10 years before so that was unfortunately never completed idea but I admire him very much because he had excellent idea to make something that later on showed to be quite successful anyway, instead of the radioactive source from Japan used heavy ions for the accelerator, microbeam, the sample and a very simple mass spectrometer the same kind of spectrometer we made also in a few years later and I'll explain you how this is done in our case so you have the beam coming from the accelerator, here are lenses that focus the beam to the sample, here is the scanner but here is a beam chopper, this beam chopper consists of two deflectors, one in X and one in Y so what computer control we made, the kind of circle that pulses are going through the slits you have here the slit, you have the reflector, there are another pair of slits here so the beam is moving in this way and only every 100 microseconds you have a pulse of 2 nanoseconds wide so in 2 nanoseconds actually you don't have a lot of ions, if you look at this kind of animation or slide that if you take, that you have here 100 picombs 100 picombs is 620 ions in 1 microsecond, so if you want to have 2 nanoseconds you'll have just 1 ion per nanosecond so when this beam is scanning over the aperture basically you'll have 1 or 2 or 0 ions in pulse but as this time is very well determined by 2 nanoseconds when such a beam, when this moment occurs, we have a start signal for the data acquisition so this start is coming from the beam chopper and the stop will be the ion that will come to micro channel play detector after extracting and flying all the way to the detector one can do one up and down and this is possible but as we were having heavy ions the vacuum in our system was not sufficient so basically you have all the time some beam halo so maybe 1% of ions will go through the system in any time so you have to increase as much as possible deflection to get rid of beam halo, even today we have problems with beam halo when we decrease so this is the kind of main reason to go as far away as possible so then you have this mass spectrometer you have so the in our chamber you have this extractor and then time of light and here is micro channel plate, so this is this extractor, some lenses is 400 mm so this is the flight to the micro channel plate, in this case we don't have this reflection another thing which is very important that when you're looking the capabilities of MEV sims in terms of analysis of the masses quite well positioned for ions between let's say 101,000 maybe a few more thousand KV sims is optimal for very low masses and then there are this Maldi very very commercial, very very used in biomedicine field spectrometer that covers all the very very heavy molecules but the problem of Maldi is that it needs some kind of sample preparation as it works with laser light evaporates part of the sample and then you analyze the masses but this is destructive technique and this is also destructive, only this one is almost not destructive because the ions are very very rare and another point is that the efficiency is very high so molecular yield for this leucine molecule is about 1% which means that you need 100 ions to get one ion out and this is extremely very sensitive because then if you have let's say femtomp current femtomp is 6000 ions per second you'll have let's say 60 events in one second and if you remember the pixel currents this is femtoms so femtoms don't do anything with the sample so sample is essentially non-destructive technique so this is our first try by scanning over some kind of leucine target which was deposited above the grid so there is no sample here and sample is in this region so we made a scan, we get the spectra this is a different, this is a mass of the molecule, see the highest peak is just a clear molecule and these are multiple so two molecules, three molecules, four and so on and unfortunately the resolution was not that good, it was two by five microns because you have to start with very high current but nevertheless we did a lot of different applications one of them is imaging, for instance this is imaging this was one cross section of the pigments in painting so you could see different masses that are connected to different areas of the sample so one can find the binders or pigments from one single sample another application was that you can identify if you have two inks, for instance you have one line with one ink, another line with another ink, you can determine which ink is at the surface on the basis of analysis of this, see this blue one is over the red one here, this is one article about this application and then we did a lot of different examples, this is like a finger print finger print I think and you can get a lot of different peaks, the problem with the M&V Sims is how to identify all this zoo of so many different peaks which is very difficult to interpret but there are some possibilities so this is one project that we were doing two or three years ago is about analysis of pigments in modern paints and these modern paints are often degrading by time so they were looking for what is going on with different kind of pigments by different times and what were some of these pigments were, this is just identification of different pigments, I will not go into details because this is very difficult to interpret the chemistry so we need the chemist bio, we are physicists, we don't really like such a complicated spectra but ok, we found some chemists that are working with and then we made different tests you can see for instance that this yellow paint was aged by UV light for let's say two months, four months and so the picture the green color became yellow blue became too yellow so they are changing the color and you can see that some of the peaks by this aging are becoming bigger and some of them are becoming smaller so this indicates persons that are knowing these things what is going on the problem that we have is this better resolution so we wanted to find some way that we can improve resolution in terms of spatial dimensions so then we came to one idea that we can use that we can analyze by Toph Sim's technique samples that are very thin so if we don't pulse the beam but we pass the beam through the sample to the pin diode to the detector, this detector can give you a star signal and then the stop signal will give you a micro channel plate in this way we obtain much better spatial resolution it was below one micron so this 0.3 microns was probably the best I think nobody repeated that good resolution with MEV sims so you can really come to the point where you can analyze very small sample and as we are going to these small sizes we try to analyze individual cells and of course again we are physicists, we don't have biologists around and then we just took from Ornian school example of the cells and we try to analyze this sample and indeed we could see this individual cell, there we obtain some spectra of some different elements and these are sodium so you can see also that you can make elemental analysis, very sensitive elemental analysis this kind of sodium image cannot be done by pixels with such small current so this is extremely useful for sodium, potassium and calcium and so on but also proteins can be imaged, the only problem with the molecules is that you are not actually analyzing cell inside but you are analyzing cell membrane because the ions don't go through they go through but they will not emit ions so certain sample preparation will be needed to cut, maybe to freeze down and then to make real analysis of the cell nucleus let's say If possible, if you are making, all experiment was made at the same time, I mean with the same manuals you may ask them on the time of play because there is an efficiency of one and the time of play with a line of one meter should have a very low efficiency because most of the ions you are not going to receive it in the... Oh no, no, no, this one percent efficiency is the efficiency in the spectrometer, not emitted so if you have one, let's say that we have this load scene sample from the before okay, I have to go too far away yeah, anyway so one ion, so we need 100 ions to eject one molecule of load scene that will go into spectrometer so if you want to get one x-ray from sodium you need one million of ions, one billion of ions to go through the sample in order to make ionization, in this case you need just 100 there are some molecules that you can easily, two or three ions that will emit, so this is really sensitive and okay, maybe later I can explain you more details anyway, this is just possibility that we can analyze single cells, but there are still some disadvantages anyway, we try to do also some analysis of cancer cells with the biologists in our institute and this will really analyze some oh, this is cancer cell and this is single cell sodium, potassium, lipid, okay, just feasibility study we don't know much about the biology of the system, but basically we prove the proof of concept that we can do analysis of single molecules how did you prepare the sample this is in the vacuum, so we first dried, so this sample is a position at very thin silicon nitride foil, very thin foil they evaporate, just make dry put it in a vacuum and then analyze, so this is essentially not biologists will not do that in that way, biologists will probably freeze make a cut, do a lot of chemistry I don't know, and then maybe we can do that, so this is kind of how physicists look like, simplify the system but okay, I think proof of principle was made and there were also some advantages, so another thing which we were trying, in our system you see this is our accelerator, the microbeam is here so at the microbeam we cannot bend very heavy ions from accelerator to this beam line, so because of that we try to get the beams of very heavy ions to the central beam line and on this central beam line we don't have micro boroprop, but we have capillary capillary focus, so you see there is a needle glass capillary with very small exit foil, so we can put the beam into this space and by doing this, so this is that one we could scan the sample over the capillary and we can make images from the sample, why we do that because very heavy ions produce even much higher sensitivity, probably one to one and this is this large ion can be used for that, and in addition of this system, we put this reflector on geometry, spectrometer now I will try to explain why we need the reflection so this is the chamber, the beam comes from that side and then you have this spectrometer, so here is the sample and imagine that you have two molecules of the same weight, but because of the interaction they may have different energy they could get energy from the, while the surface, so the faster the faster ion will go further into this reflector on electrostatic mirror while the slower particles will go shallower but at the end by proper adjustments of the geometry and voltages both of ions will come to the same time at the end and what this happens so this green peak was the peak from the previous spectrometer where we didn't have this reflection and now with this new spectrometer in this one peak we could identify four different masses, so this is much better energy resolution actually timing resolution, so this spectrometer is very very useful because it changes by factor of 10 resolution to much better here is also some another example where we try to compare again the problem of the physics which printing here is at the surface, as you know that if you are working with pixie, pixie will integrate almost everything and you will not be able to see from pixie images what is on the top, so we use MEV sims but the problem with MEV sims was another one that we again had thousands of peaks, so we have a PhD student Marco Barats and Catherine Moore, they developed principle component analysis and displayed the different components of this mathematical procedure to different images while scanning of different components and each component consists of tens of peaks, not just one peak so in that way this is done much better and this is just the description of the same thing with pixie pixie integrates all the contribution from all the layers and you cannot distinguish them, although there are possibilities in x-rays like silicon or sodium which are absorbed because of their low energies and this is the same thing done of two different lines from the printer so you can see in this case we have inkjet is this one, this is laser so this is laser jet and this is inkjet line, so obviously the laser one is at the top okay and these are principle component maps, this is not a single element map but a physical component of component that take both pixel spectra and spectra from the map sims okay and the same problem appears in another project we had we tried to collaborate with some biologists to establish can MEV sims indicate when they are better created so we did a lot of analysis of different blood glucose levels in mice and you have a lot of different spectra, of course you cannot distinguish anything from this spectra without the computer so there was again some multiviral empirical base and time series analysis, I never heard about that but basically a student works on that and I will show you the results so these are, this is a graphical presentation of different peaks, these are different peaks in spectrum so each of the lines is different peak, this is control and this is deceased mice the thing is that these evens in these peaks are predicting that the high glucose level, this is the time, you see, so when the mice shows that there is the bitis, this is this time but these peaks suggested that even before illness developed, you see some new or some different pattern in peaks so this was I think one of the first works where actually you could predict that somebody will develop the bitis so this is running on research but obviously mathematical analysis of all these large number of data is quite important okay so these are some conclusions about MEV sims, I will not go, okay I can try to oops, what is this, yeah so if you have accelerator, if you have a microbeam I mean MEV sims is not very difficult part to add so one can say that it is cheap relatively because if you look at the Maldi MEV sims, KV sims is a half a million or million euro instruments but if you already have accelerator I think that this is a very, very good application we presented different kind of applications, one of some of them can be used and we could also show that we could have very high spatial resolution so I think I spoke enough about MEV sims at the end I will just show you a little bit about our laboratory so we have two accelerators, one is small and another one is very old 6 megawatt tandem hopefully we will replace it with the new accelerator soon these are the beam lines that we have and from the point of view of the whole laboratory I could say that our beams can provide plenty of unique information and unique samples so in addition to classical IBA techniques we have this new MEV sims and also which I will speak more tomorrow about single line techniques which you can characterize charge transport property, crystal structure, density and so on and the final two slides are dedicated to our target room, it has a lot of different chambers some of them are pretty unique, one of them is dual beam irradiation stations, as we have two accelerators we can irradiate a sample with two beams at the same time, so two of them for fusion materials, we have time of light Erda which is a very, very useful machine and the capillary micro problem and the other micro problem here one of the very important things for accelerators is to have enough users, in our case we have now we are part of the Radiate Consortium so mostly European researchers but also there are possibilities from non-European can propose the project through this Radiate Consortium and get the free beam time at our accelerator and in addition there is a Central European Research Infrastructure Consortium which is institution based in Trieste in Sigrotron and through this Eric we can also provide beam time and finally probably you heard about that, that Atomic Energy Agency is having CRP about access to different accelerators around the world and we are also part of this CRP and hopefully I think we will have the first run through this CRP in November and hopefully this will go further because I think there is not enough possibilities for people from developing countries where there is not enough facilities even if there are accelerators, techniques are different from different laboratories so you can probably find different labs with different capabilities and that's our group in Zagreb behind us is Accelerator and we are funded by several of the projects of European Union at the Atomic Energy Agency.