 Hello there, today we are going to learn how to record an EPR spectrum and what are the ways that we can get the best quality spectrum. To start with of course we need to prepare our sample and sample could be of various kind it could be solid, it could be liquid, it could even be gaseous sample. So for different types of samples we need to have sample tubes, so if it is solid or liquid one could use tube of this kind, it is a narrow tube typically in a dimension of maybe 2 to 3 millimeter and keep the solid here or fill the liquid maybe up to 2, 3 centimeter and insert it inside the cavity. Now often in liquid sample most of the solvents dissolve some oxygen and you all know that oxygen is a paramagnetic species, so this presence of oxygen increases the spin lattice relaxation rate of the paramagnetic species that is we are studying. So effectively it causes line-roddening, so also it can even kill the free radical that is present there if oxygen reacts with that. So oxygen is something undesirable, we do not want oxygen in the sample, so how to remove oxygen from the liquid sample, so what we can do is to use a sample tube that is shown here this kind of tube, now here we use a vacuum line to remove the dissolved air, so that will remove both oxygen and nitrogen. So this type of tube which has a small vessel here where the liquid sample is kept and the side arm is similar to this standard appear tube which can be actually inserted in the cavity, now this standard joint here is used to connect this to a vacuum line and after degassing this is the place where this could be sealed off and permanently kept under vacuum. Now how do I remove the dissolved air from this for that these are the steps, so that is called degassing of liquid EPR samples, so first we connect this our EPR sample tube after placing the sample here and connected to the vacuum line and you keep it closed first and this is the stock of which can either connect to the vacuum line or disconnect it, so first we freeze this sample by placing it in a liquid nitrogen dewar, so freeze the sample first and then once the sample is frozen we connect this sample tube to vacuum line by opening the stock of and remove the air which was trapped there. This is evacuated 7-century well again close the stopcock and then remove this liquid nitrogen from the sample and allow it to thaw to room temperature, as it thaws this frozen liquid melts and because the pressure is low here the dissolved air comes out of this and remains here, so again I come back to this step freeze it keeping the stopcock closed and once it is frozen I pump the released air from the sample here and wait for sometime till the vacuum is established again close the vacuum line thaw it and when it thaws whatever residual gas was there or air was there that will again be released, so that way if you keep on doing it 3-4 times and if the vacuum line gives reasonably good vacuum let us say 10 to the power minus 4 or minus 5 thaw then this solvent will almost be free from dissolved air and that is the sample we like to work on and then if after having reached the desired level of freeze pump and thaw and the degassing of the sample we can seal it here, so this is the way the sealed sample tube this place is sealed, so for doing the experiment this liquid is transferred here and then that could be kept inside the cavity. Now liquid then of course we can use sample tube of this kind provided this liquid is not very polar we have seen earlier that polar solvent absorbs micro air and undergoes electric dipole transition that is nothing but the rotational transition and that is bad we do not want that to happen, so if the sample is reasonably polar then this sort of sample tube this item here or here they may be sufficiently larger in dimension so that the microwave electric field might be influencing the sample here and we can get this sort of transition and that will make the whole thing very insensitive, we do not want this to happen all you want is the magnetic dipole transition to take place, so one of the common solvents for biological experiment is the water, water you all know is highly polar, so experiment that needs water as a solvent cannot be done in this type of sample tubes that I have shown just now, so for that one can get around in two ways, one is of course use a capillary, so the dimension of this tube here instead of 2 to 3 millimeter diameter it will reduce to let us say 1 millimeter diameter and then it may be reasonably acceptable for recording the EPR spectrum, but then very difficult to handle this capillary tubes, so what one normally does is to use the standard EPR tube like this and then insert the capillary inside there, so this is the capillary which contains the aqueous sample then one could possibly record the EPR spectrum of this sample kept inside the capillary, but this is not quite ideal because the amount of sample that is present here is very small and like any other spectroscopic technique the intensity of the signal will depend on how much sample I put in, so the smaller the amount less will be the signal, so unless the sample concentration is very high it is very difficult to get decent signal using a capillary tube, so there is another way of doing it that is use different type of sample tube and that is shown here this is the sample tube is called a flat cell, this flat cell is a rectangular chamber of approximately this dimension 5 centimeter long and 1 centimeter wide and this thickness of this is about 0.5 millimeter, so it is like a rectangular box these two tubes are to hold the sample tube inside the cavity, now the design is such that this will be a flat surface and the rectangular dimension and the thin region is only 0.5 millimeter, so if you take a let us say cross section of this kind, so I can ensure that the electric field actually sees only this much of the sample, but the magnetic field cannot see this much of the sample, so that way I can minimize the interaction of the electric field with the sample, so that it has to be placed in a cavity such way that that condition is satisfied, so the wider dimension of this should be facing the maximum of the magnetic field, the thinner dimension same time will be at the minimum of the electric field, so that is not very difficult to understand where to place it, this is the cavity, rectangular cavity of TE 102 mode and we know that at this place exactly here this plane the magnetic field is maximum and electric field is minimum, so all I need to do is at this plane I must position this flat cell, so that I achieve the requirement that the sample will see very electric field and maximum of the magnetic field. For holding the sample in this cavity there are of course holders available which are mounted here and here and for any of this tubes there are corresponding holders, for holding the EPR sample this flat cell also special holders are available, because it is important to realize that it is not only that the flat cell has to be exactly in the center of the cavity it also has to be oriented properly, so what I mean by that is that suppose this is the cavity and the center of this is a center of this place of maximum magnetic field is somewhere here, so I need that this plane must be aligned here, but it also has to be aligned such a way that it need perpendicular to this surface, if there is little bit of mismatch of this kind then the sample from the edges will see the electric field and that is not good, so not only it will be a center also has to be exactly made parallel to this and in this lateral direction front and back also it has to be exactly the center of cavity, how does one position that it is not very difficult to do that if we see the cavity mode by reflect modulating the micro frequency, now this is the let us say Pleistone mode and the cavity mode is superimposed on that, so this is the cavity mode, when we insert this one the frequency of the cavity of course going to change, so to start with we take the flat shell and approximately insert in this position, so that the front surface of the flat shell is almost parallel to this that is the way we start with and we get the cavity mode in the middle of this, now see that because of the rectangular nature it has certain symmetry if you see it once again here, it is symmetric with respect to this plane on both side similar this is also symmetric with respect to this plane at the center, so having approximately started with this cavity and the flat shell here in this position, if I now gently orient it in this direction then this frequency is going to shift in one direction, it will reach an extreme position again go back in the harder direction, why is that so? Because exactly when the flat shell is parallel to the surface that is what the frequency is going to be extreme, if it goes this way or that way the shift in frequency will be same, so that we can say this is going this way if we keep turning in one direction and then up to some it will come out and come back to this opposite direction, so that way I can find out the optimum placement of this cavity in the flat shell to get the extremum of this, assumption in that you must start with the initial placement that this is approximately parallel to the desired plane that is the first part, second part is to how to ensure that this is indeed center by adjusting this back and forth here, the holders at the top of and bottom of the cavity allows one to have that movement also, so once again because the cavity has this sort of a symmetry plane with this here and so if I the flat shell from the center if it comes this way or that way that change of frequency will be in the similar direction, so again exactly at the center of the flat shell and it is parallel to this plane this placement twice appearance of this will be at the extremum that is the way one decides the correct position of the flat shell, for cylindrical cavity there is only a cylindrical symmetry so it is not very easy to use similar argument but the aim is that in the cylindrical cavity this is the deep let us say, so when the electric field interaction is minimum this deep will be maximum, so I must orient the flat shell inside the cylindrical cavity such that I maximize this deep, more electric field is seen by the sample, poorer will be the cube and this will become poorer, so I can either rotate in the in this axis and see that this becomes deeper and deeper and try to reach the extreme position and then also move it back and forth to maximize this that is the way to place the sample. If the sample happens to a gas then it is very simple all we need to do is to have a just a tube and which I insert kept inside the cavity and let the gas be inside but then this parametric molecules when they collide with each other they change their angle momentum direction, so that way lines become broad, so to minimize the collision at the same time to get a descent signal to know as ratio one has to work at a moderate pressure, so this has to be connected to a vacuum line or a pump, so vacuum pump, here the sample goes in, so that way one can do the experiment, having inserted the sample and next is to turn on the micro frequency and match the micro frequency to the cavities one characteristic frequency for that we modulate the micro power and see where the cavity deep is, but in general we may not see any deep, so we tune the micro frequency using the appropriate knob here and then we should be able to see the deep here it started appearing, we want this frequency to be at the maximum of the micro power that is emitted by the klystone, so this is the klystone mode and this is the cavity deep which is coming there, so we tune the frequency, so more or less this is the correct place, now if it is, if the micro source is a solid state source a Garn oscillator then that mode will not look like this, but it will look like this sort of thing this is the frequency axis and this is the power, so this is flat that is the only difference otherwise it is the same thing, now so we get the micro frequency to the maximum of the klystone mode and then we have to optimize the coupling that is this should be very nearly the critical coupled condition for that, so we have to adjust the iris tuning screw which is here, so we adjust the iris tuning screw so that this deep becomes maximum, but we should not make it over coupled, so that is just to turn the screw, this screw is going inside and the deep is increasing and increasing, increasing, keep tuning then deep is increasing, now this is almost critically coupled, suppose you go beyond that you see it started going up again, so this has become over coupled we do not want that, go back here, so this is almost critically coupled we stop here, so having obtained the nearly critically coupled condition we can now switch of the modulation and turn on the automatic frequency control, but here one important point to be kept in mind is that the bias power that is used to bias the detector is present or not while we are doing this tuning up, so if you remember let us recall our design of the spectrometer, this is the simplest arrangement, this is the source of microwave comes here goes to a circulator, this is the cavity it is kept here, this is the detector let us call it D and we have a bias power coming from here and through this again mixing there and we have got to appropriate attenuator and at phase scepter, attenuator phase scepter, so if the bias power is present, now we have drawn here then detector sees that power all the time, so in this tuning this will not reach to the bottom of this one, because that will give a constant micro power there, so that has to be kept in mind, some spectrometer allows you to switch of this bias power during the tuning period, so there could be some symbolically speaking there could be some switch here to either disconnect this or connect this, so if that provision is there this is better to switch of the bias power and do the tuning, in that case there will not be any confusion when the critically coupled condition is peraged, but if a this provision is not there then one should try to bring it to as low as possible and then watch that this does not go up again, so this dip will probably go down here, so this may be the bias power, so you will not be able to bring it below that one, so then again it will go up, this also assumes that the phase of this is same as this one otherwise this two power will try to cancel each other if the phase is opposite, so one has to also adjust the phase of the bias power such that I get maximum signal here and I can see the maximum here also, that way also one can start with the bias power to have the appropriate and correct phase with respect to the micro power that is going there, now having done that we now switch on the AFC and then ready to start recording the spectrum, first we need to use certain starting conditions micro power, how much micro power to use to see the spectrum, this is something one does not know a priori, but if you have some idea the sample relaxation time is fairly long and it saturates very easily, then one uses low observing micro power, but in the sample does not saturate very easily one uses high micro power, so one starts some intermediate value of the micro power, next the magnetic field modulation amplitude typically 100 kilo watts, so here again one has to decide how much modulation number is to use, so to start with some intermediate value to start with, then the magnetic field range to where the signal is likely to appear magnetic, now here and its center, so these two parameters mean that the region of interest where I expect the signal to appear, so let us say this much is the magnetic field I want to scan and search for signal and this is the my center field and this is the I call it scan range, these two parameters need to be adjusted, so how does one decide that, again if one has no idea what sort of g value the sample has, then one really does not know what center field to use and how much scan range to use, so one usually starts with a reasonably large scan range and assume the g is equal to 2 or so, then keep the center field approximately there and the c f signal is appearing, having set up this conditions, if you look for signal and scan the magnetic field and find that there is no signal appearing there, then what happens, what could have gone wrong or can we really change some of the settings and look for signal, one possibility is that the range of this magnetic field is not right, may be the g value is such that it is somewhere else, so if one suspects the g value is equal to 2 or any other known values, then from the micro frequency new you can find out the center field by this formula, so if you have some idea what this is, then keep the center field approximately the value that you calculate from the micro frequency and put it here and then have a large scan range and then try again see if the signal appears there or it is possible that a that I may be scanning too little or then you can increase the scan range or it is possible that even then the spectrum is somewhere else, that is I can whether it is assumed g value is not what I what is right, so I can go somewhere else then change the center field somewhere else again scan it, it is possible then that the modulation amplitude was not appropriate enough, may be it was giving too little small signal, we will see later that it may even give broadening of line if the modulation amplitude is very high, so again you change this parameter and see if the signal comes or not, may be the micro power was either too high which is causing saturation and the signal was not appearing or maybe it is too little, so the signal also was not coming, so keep on adjusting this and look for signal, so in spite of all this suppose the signal really is not coming then what can one conclude, one obvious conclusion will be the sample is not paramagnetic either all the preparation of samples and whatever you have done to bring the sample to the cavity may have just died that is too bad you have to start over again on the other hand if there is strong reason to believe that it is indeed paramagnetic and still the sample is not giving appear signal the way you have done it and what I of course implicitly assume that we are doing it at room temperature it is possible therefore that the relaxation time is so fast or the that it does not give signal at a room temperature then one has to go to lower temperature may be liquid nitrogen temperature which is 7 to 7 Kelvin or even liquid helium which corresponds to 4 Kelvin, so maybe we will get some signal that time now having done all this let us say we have seen some signal now what are the adjustments we can do to optimize the quality of the spectrum first thing is to do is to adjust this suppose our signal has appeared in this fashion this is the range of magnetic flux can and it is appearing some sort of signal appears here this sort of thing. So, first I do is to which obvious that the scan range was too much and also that it is not the center of the spectrum, so I first change the center field this was the center field earlier I now bring the center field here, so center field is brought here then again I scan the magnetic field and get the spectrum this time it will look like this it will look like this type of thing. So, now this much of magnetic field is not doing anything it is simply compressing the spectrum here, so to do proper measurement let us have probably coupling constant or line position in other words to get a more resolved spectrum we do not need to scan this region, so reduce the scan range instead of this now new scan range could be let us have this much new scan range again you do record the spectrum. So, this time this may look like this is a new scan range which has been expanded this fashion, so it may look like this type of thing this is much better now we can do some measurement, but is that all or we can improve it further as I said earlier now we can try to increase the micro power and see if we can get better quality spectrum, micro power will cause more efficient transition and signal intensity will go up provided recall our discussion earlier of the saturation of the spin system and the relaxation mechanism which are present there. So, these two will also work such a way that you cannot increase the micro power too much without bringing in saturation, so the way it will behave is that if your signal intensity intensity and the micro power if the square root of that this will increase linearly first increase the micro power the signal height will go up and up and up in this in this fashion and then it will start swing some sort of saturation and goes down here. So, in this place where the relaxation mechanism is not able to maintain the population difference, so we cannot use this much power and lose the signal height intensity of the signal. So, we have to decide that somewhere here we can stop and not increase the micro power, so we get bigger signal in the process. Next is this modulation amplitude, how to optimize that? This is the principle of magnetic field modulation and phase sender detection. So, here we have come across this slide earlier, but let us recall once again this is the amplitude of the magnetic field modulation and this is the corresponding response of the sample to the modulated magnetic field. So, it is obvious that not only this amplitude depends on the position of the magnetic field and that is the reason of course for the derivative presentation of the EPR signal it also amplitude of this also depends on how big this is. If I increase this amplitude this will also increase this if this is increased this will also increase. So, the EPR signal is going to increase with the increase of the modulation amplitude, but to get a true line shape that is line shape is exactly the derivative of this absorption profile this should not be too much. So, the signal will have this sort of dependence on the magnetic field modulation amplitude. So, this way then it will go down here by the way by intensity I mean here the height of the signal not the area of that this height of this signal here. So, this is linearly increasing and then it will start going down when the modulation amplitude becomes comparable to the line width. So, when that happens the intrinsic line width of this here this is the intrinsic line width and correspondingly there will be derivative line width that will start showing distortion and the EPR signal will become broad. Suppose this is the EPR signal at a moderate amplitude of the modulation this will height will go up and up as you increase the modulation amplitude, but then when this modulation amplitude becomes comparable to this width this will show broadening of this kind. So, that is not desirable. So, this has to make some compromise the way let us say another axis here this is a width let us call this delta V peak to peak is defined to be this delta V peak to peak that is the width of the derivative line. So, that will have this sort of dependence on the modulation amplitude it will remain almost constant then it will starts going up. So, the width remains constant so long as the amplitude of modulation is much smaller than its own width here. So, that is the true line width of the sample, but as signal goes up this also up to remains constant. So, this is a desirable region of operation this are the acceptable this is ok, but somewhere here now intensity is going down and width is going up is not acceptable because this gives distortion. So, that way one decides how much magnetic modulation is to be used here sometimes one can sacrifice the line shape and get the IPR signal because when you are trying to struggling to find if at all there is IPR line or not you really do not care if the shape is correct or not. So, then once work somewhere here this region where there is some sort of distortion of the width, but nevertheless signal is somewhat bigger than what is supposed to be if it was not if one was using smaller amplitude of modulation. So, some distortion, but slightly higher intensity is preferable or even acceptable if one interested only in detecting the presence of the radical and getting some sort of IPR signal there. And then after adjusting this next parameter to be adjusted how much time I should use to scan this magnetic field range this is called the scan time. The scan time how many minutes or how many seconds depends on how quickly the magnetic field is going through this lines. If the lines are very sharp then one should spend enough time for spectrometer to respond sufficiently quickly to the changing signal here. Now how does one decide that quantitatively after the phase and direction and the derivative signal has been recorded one usually puts a what is called the low frequency filter time constant. Low frequency filter with a time constant let us say t that means any signal or noise whatsoever oscillations whose frequency is faster than one by t will not be given out, but will be filtered out. For example, if I t is equal to one second time constant that means any oscillations or instability which is more than one hertz one hertz will be effectively reduced. So, depending on what time constant I use here I must allow several times this time constant for the magnetic field to go through this line. So, the narrow at the line let us say that is one line is this narrow other is broad. So, it has to go through sufficiently slowly through this one compared to this time constant so that this is faithfully reproduced. So, typically time to go through a line should be let us say about greater than 10 times the time constant. This is time constant which is used to filter out the signal so that the spectrometer has sufficient time to slowly go through this. So, that depends on the how broad and how narrow is it is very narrow then it will take it should once you allow longer time if it is a broad then one does not need to have that much time. So, here is a small guide number suppose the line width is 0.5 gauss and you scan 10 gauss and time constant one second. So, I should give 10 second to scan 0.5 gauss 10 second to scan 0.5 gauss 10 second to scan 0.5 times time constant to go through this one. So, that means to scan 10 gauss which is the range from this to this I have to give a time of 200 second to scan 10 gauss that is the way one decides the scan time and a little bit more of final adjustment is sometimes necessary not always is the phase of the modulation frequency which is usually set up by the manufacturer. And some often one does not change that very often, but if one can see things have got disturbed or not one can just tweak the reference phase of this one to maximize the signal. And also the bias power phase which we have said that we could optimize by looking at the cavity mode which may also need to be adjusted little bit to maximize the EPR signal. So, these are the various parameters which need to be optimized to get the best quality signal to noise ratio. With this we come to end of this session.