 In this lecture, we will be looking at a specific spectroscopy tool which has been widely used in materials chemistry and as the title suggests, there are two issues that are involved in this spectroscopy. The name suggests that it is a x-ray photoelectron spectroscopy therefore, there is a x-ray that is produced when a photoelectron is ejected. So a combination of two processes that are happening brings about a very rich chemistry that one can learn. Material scientists, physicists and chemists have used this technique extensively and this is now a very established tool for characterization materials in the last three decades improvised and very sophisticated instruments have come. In the last two decades, the machine could actually occupy a very big room but today the sophistication of this instrumentation has moved such a way that even a benched up x-ray photoelectron spectrophotometer can be used. So we will look into this issue of what this principle behind the spectroscopic tool is. And if there is something very highly complementary to x-ray diffraction, it is the spectroscopic tools and as you would see in this lecture, there are several spectroscopic issues that are manifested when x-rays are emitted and how this can be mapped and how we can translate this into quantifying some results we will see in the next few slides. And to start with let us see how we create x-rays and how the spectrum of a x-ray tube is generated. Now when an incident electron beam is incident on a core electron that is in the location, then there are several things that can happen. Electron can be ejected out of the to the vacuum level which means it is totally out of the binding force or binding energy of the metal or it can actually have a electron loss energy situation that is involved or electrons can be produced x-rays can be produced from one of the L-shells to a K-shell when electron jumps from one to the other. So we can actually look at the particular x-ray that is produced during this transfer of this electron from L-shell to K and that can be a characteristically mapped for a particular element and this is the characteristic radiation that one observes which has two components K alpha and K beta but usually we try to use K alpha because the intensity of K alpha is much higher than K beta. So we will now look at the process that involves generation of x-rays which is typical of a material and this has to do with the core shell electrons instead of the valence band electrons therefore this is particularly called as x-ray photoelectron spectroscopy. Now when we look at this analysis tool it is mainly classified or generally referred to as surface analysis and the study of the outermost layers of materials can be studied therefore in thin film may be through the bulk but in the bulk materials you can only penetrate to very few nanometers therefore it is mostly a surface phenomena rather than a bulk phenomena that we can look for but generally this can bring about rich chemistry because all the impurities that are occluded to a given bulk material can be analyzed because the depth that it can go is only for few nanometers. Now depending on the manifestation of the interaction of the beam with the sample there are several things that can happen one is based on electron spectroscopy and another one is based on ion spectroscopy. So a series of characterization tools can emerge out of the approach that we take if you are going to use a electron beam to interact then the manifestations are XPS that is x-ray photoelectron spectroscopy or another secondary process that happens during the XPS which is Auger electron spectroscopy or there could be another secondary effect which is a electron energy loss spectroscopy all this when electron beam is interacting with the core shell electrons. We can also have ion interaction with the matter and as a result we can come across several spectroscopic tools which can carry useful information one is a SIMS that is secondary ion mass spectroscopy can give you very refined information into all the corresponding elements or impurities that are present in the surface or the Rutherford back scattering which is another very sophisticated tool where helium ion is sent into the lattice and if the lattice is actually ordered then you have a good channeling that is happening but if it is a disordered material the helium ion will get scattered back and therefore you can look at the length of this ordering of the lattice or how orderly the lattices are developed during the process can be mapped and specially those who are involved in thin film technology use RBS as a very sophisticated tool which can give you a clear insight into the epitaxial growth of the film whether the film is growing along a particular axis or not can be mapped using RBS and then you also have ions scattering spectroscopy. So a range of spectroscopic tools are categories in the form of surface analysis and today we will look little bit more deeper into the XPS technique. So what is XPS X-ray photo electron spectroscopy is actually based on the photoelectric effect that is why it is called X-ray photo electron spectroscopy and this was developed in mid 60 by Kai Seibann and his research group at the university of Uppsala in Sweden and since then this has been a pioneering probe tool. So what is XPS and how historically we can identify this 1887 it was actually the discovery of photoelectric effect and this was not explained based on classical physics because what we saw from the photoelectric effect was not commensurate to the theory that was existing about light. So in 1905 it was Einstein who used the issue of quantization proposed by Max Planck and he explained what this photoelectric effect means and the first formulation of basic XPS theory by Rutherford in 1914 and then photoelectron spectra of different elements were recorded in 1920 and then on it has been a major effort as a surface probe tool. Now what really happens in the photoelectric process as you would see the incident X-ray actually falls on the core shell electron that is K shell and this is immediately ejected as a photoelectron. So this is the photoelectron that we are trying to map but in the process there are several things that can happen once the hole is created here then the neighboring shells that is in the L1, L2 or L3 will quickly translate one of the electrons to the lower shell that is K shell and thereby secondary processes do emerge. So XPS spectral lines are identified by the shell from which the electron was ejected it could be from K shell or L or M and so on. The ejected photoelectron has a kinetic energy which is h nu minus binding energy minus phi and this is actually from the instrument. We will look at this equation because we are familiar with this expression from photoelectric effect which is a work function plus half mv square this is your basic expression from photoelectric effect but what you would see here is an extra parameter that is coming which we will try to understand as to why we do that. Following this process the atom will release energy by emission of an ogre electron. So where this ogre electron is coming to picture the L electron falls to fill the vacancy which is step 1. So once the photoelectron is ejected out from which was from here it the photoelectron was ejected and in that place one of the L 2 or L 3 electrons will actually come and fill this place and to conserve energy one of the other electron will be emitted. As a result you are actually getting a ogre electron which is actually not the photoelectron which was ejected in the first step therefore two different processes do occur. One is ogre electron which is coming and the other one is the photoelectron extra photoelectron. So the kinetic energy of the ogre electron is actually different from the kinetic energy of the photoelectron. In photoelectron we are bringing in other phi and this binding energy is also a work function of the sample. This is typical of the sample and this is typical of the machine or the electron gun which will come to it. Therefore there are two work functions that are related when we talk about XPS. We are talking about the kinetic energy of the ogre electron which is linked to the energy of the electrons in different levels. So both are different the information that you get is different. So the x-ray induced ogre electrons can be actually mapped in this way. So when we have this x-ray photoelectron then the electron actually is ejected from the core shell and this the core shell electron which is gone. So that means the binding energy is the energy that we go from here to the valence band and from valence band to the vacuum level. So that is your energy that is associated with the kinetic energy of the ejected electron. And when we think of the ogre electron it is the energy that is translated from one of these shells to the vacuum level. So the kinetic energy of the ogre electron therefore is going to be completely different or will be less than the kinetic energy of the photoelectron. However the kinetic energy is actually independent of the x-ray photo photon energy as far as the ogre electron is concerned. However in the binding energy scale ogre peak positions depend on the x-ray source. We will also look at it shortly from now. When we look at a typical x-ray photoelectron spectra then there are some notations spectroscopic notations which are included and they are actually summed up either as k L1, L2, L3 or it is actually abbreviated as 1 is half, 2 is half, 2 p half, 2 p 3 by 2. So we need to understand in perspective what this exactly means. The suffix that we are looking at is mainly coming from the J and J is nothing but spin orbit coupling. So where L is your orbital momentum or orbital angular momentum number, quantum number therefore the L and the spin of the electron will couple together and give half. For example if your spin is half then 0 plus half will give you half. If it is 0 in this case if this is going to be 1 then 1 minus half will give you half or 1 plus half will give 3 by 2. So the suffix is actually a product of both L plus S which is denoted as J. So depending on the shell we can abbreviate this as L1. If we say L1 then we talk about 2 is half, L2 is 2 p half, L3 is 2 p 3 by 2. So by this way we can actually map all the shells or the electrons related to that particular shell and these are having very characteristic binding energy which we will see shortly from now. So this is the spin orbit coupling that we picked up from the earlier slide. So Ls coupling is nothing but your J coupling which is L plus or minus half. So if L plus half is there then we talk about half or 3 by 2 or L minus half will be half. So this is the way it is denoted your 2 is nothing but your principal quantum number that determines which shell the electron is involved in this spectra and this is to do with the coupling. So if you are actually looking at a 2 p half or 2 p 3 by 2 spectra then there will be certainly a ratio of the area for this peak. So if we talk about 2 p 3 by 2 then the area under the curve will be approximately 2 times more than the 2 p half. And similarly between 3 d 3 by 2 and 3 d 5 by 2 you will have a intensity ratio of 3 is to 2 and same is the case for 4 of electrons 4 of 5 by 2 and 4 of 7 by 2 will have a ratio of 3 is to 4. So looking at this we can easily map even without a notation looking at the area under the peak we will be able to immediately say what is the sort of peak that we are looking at. So typically the peak will look like this and this is the case for gold 4 d 5 by 2 and gold 4 d 3 by 2 as I told you in the previous case that if it is a d electron then the ratio would be somewhere between 2 is to 3 and you can see that very well here and the area under this curve will be the measure. So the separation between the 2 peaks are named as spin orbit coupling and the values of spin orbital splitting of a core level of an element in different compounds are nearly the same. So this splitting will nearly be the same for almost all the elements and the peak ratios of the core level of an element in different compounds are also nearly the same because it does not depend on the material but it depends on the spin orbit coupling. So these are very unique to the quantum states. So what we can identify therefore is in spin orbit splitting the peak area will assist in the element identifications. Now how will X-ray photoelectron spectra chamber look like? The analyzer which will actually analyze the X-ray photoelectron will actually be a semi hemispherical chamber and this is how it goes. If you have a X-ray source which is hitting the target that is your sample. Now electron is ejected out beyond the vacuum level and this is actually translated through a semi hemispherical chamber which is the inner layer is actually coated with gold. So outer chamber might look metallic but then it is actually plated with gold. So gold is your analyzer and this is actually channelized so as to get the stuff directly to the detector through a photomultiplier tube. So this is in essence a simple setup of a X-ray photoelectron spectra, spectrophotometer. So because we are actually using an analyzer which is gold in this case now this has to be since the gold is acting as a reference detector where all the photoelectrons are collected the work function of the gold is also to be equated or grounded to the sample. Therefore the Fermi level of both the sample and the analyzer has to be same and as a result we need to include the work function of gold electrode also and this is a typical picture of a assembled analyzer that we are talking about. So you would have the sample fitted here and then it will go through the detector and the photoelectron will be mapped. So this is a cartoon that tells how simple or how intricate the assembly is made. Now when we talk about XPS energy scale there are two things that we need to understand. One is the XPS energy with respect to kinetic energy and the XPS energy scale with respect to binding energy two things are there. So when we talk about kinetic energy the expression that we follow is kinetic energy is equal to the incident energy of the electron beam and the binding energy of your sample minus the work function of your spectrophotometer which is nothing but gold. So we usually take the gold work function and we minus that because that is also involved in the analyzing process. Photoelectron line energies therefore are dependent on photon energy whereas the Auger electron line energies are not dependent on photon energy. This is a very important principle. So only the XPS lines are dependent on the photon energy which we are talking about this and the binding energy is not dependent on this h nu. So why we use spectrophotometer work function? Because electrons are detected when reaching the detector's Fermi level kinetic energy is therefore measured relative to the detector's zero point. That is why we use the work function and in addition sample may be on a different potential that is it can be charged. When we talk about the energy scale in terms of binding energy then you will easily find out that we are talking now about binding energy which is equal to h nu minus kinetic energy minus phi spectrophotometer. So in this case the photoelectron line energy is not dependent on photon energy whereas Auger electron is dependent on photon energy when we think about binding energy. So the binding energy can be calibrated by standards. Typically we always use carbon 1S standard because when we are running the vacuum system there will be some amount of carbon in the vacuum which will get deposited on the surface of the sample. Therefore carbon impurity will always be there. However you treat your sample and therefore carbon 1S which is coming from the vacuum chamber will actually serve as an internal standard. How much ever we try to flush out this sample yet we will see a very clear carbon 1S spectra and that is typically 285 electron volt and that is used as a reference point. Then we also have gold 4 or 7 that is 7 by 2 which appears at 84 as a reference point. If suppose these binding energies are differing by one electron volt or two electron volt then we need to use this as a correction factor when we are analyzing our sample. So that is the way we do. The two things that we saw is the binding energy reference or the kinetic energy reference. Now this is the process that we talked about taking the electron from the core level to the Fermi level which is the binding energy plus the work function of the sample from where you take the electron from the Fermi level to the vacuum level and then this electron is actually put into the vacuum level of the analyzer therefore we are talking about the work function of the spectrophotometer. So when we look at the kinetic energy of the photo electron then we are talking about H nu H nu minus B minus the sample and similarly we have to take into consideration the work function of the sample minus the spectrophotometer. Therefore if we merge all these four parameters H nu minus B minus phi sample minus the difference of sample and spectrophotometer then we boil down to this expression. And similarly for binding energy we have this equation H nu minus kinetic energy minus phi. That is the reason why we have two different ways of evaluating the ejected electrons. And this is a map which tells us the relative binding energy of different elements and the 1s electrons usually have a range between 0 to 1100 electron old and this is for the atoms with atomic number 0 to 10. And as you progressively increase then you can see all the binding energies can be mapped the 2s and 2b and these are all very specific to the core shell of different elements. So each number will have a very precise count. So there would not be any change and therefore we can easily try to analyze the local state of each of this metal or elements. But when you go to higher energies as you can see here there are some gaps for 4f and for 4d electrons. Mainly coming because of the heavier electrons and the screening and the penetration effect which will actually distort the way the of the principle describes. Therefore there are some gaps here for which we can see how we can account for those regions. Mainly those are heavier metals therefore you would see those missing data present here. So bulkier the atoms then you have this of the principle dictating the binding energy. When we have a survey spectrum taken for indium phosphide for example then these are the different peaks that one would see and each peak and every hump that is shown here has some message to convey. For example if you look at indium 3d 5 by 2 and indium 3d 3 by 2 these two peaks are coming because of the photo electrons without energy loss whereas there will be a broadening or a small satellite feature that is coming which is actually a background effect coming because of the photo electrons energy loss. And as you would see here you can pinpoint each of this peaks which is characteristic for indium 3s 3p 3d and phosphorous 2s 2p and indium 4d. So each of this peak can be monitored precisely and this is a survey spectrum of a nickel foil which clearly gives you the peak for 2p and also for 3s and 3p electrons and in between you also see the ogre lines which are coming and typically you cannot avoid the ogre lines in a survey spectrum but you can easily make out which is coming from a ogre spectra mainly because they will be asymmetric and it will be broad. So those peaks which are asymmetric and broad are coming due to the ogre effect whereas the sharp ones are due to the photo electrons. So in a survey spectrum you can easily pinpoint on those features and if you try to expand or enlarge this portion you can see this in the inset you can clearly get the nature of the peak whether it is symmetric or asymmetric and what is the broadening and is there any other extra features that are coming all this can be seen when we try to expand this scale. So every satellite feature that you see to the primary feature will give some idea about the local state and the oxidation state of those elements. One more thing that we need to understand why this is a surface technique is because of the sampling depth that is involved the sampling depth is actually defined as the depth from which 95 percent of all photo electrons are scattered by the time they reach the surface. So when electrons are ejected the time taken or the distance it travels by the by the time 95 percent of the photo electrons can be scattered is the sampling depth and that is of the order of three lambda most lambda's wavelengths are in the range of 1.12 3.5 nanometer for a alpha aluminum k alpha radiation usually we use aluminum k alpha or magnesium k alpha as the radiation for XPS. So the sampling depth is approximately 3 lambda for XPS and this is condition therefore your sampling depth can be ascertained to be anywhere between 3 to 10 nanometers and that is all we can prove. One of the very useful information that you get from XPS is the oxidation state we will see several examples in the next few slides where we will we can understand what are the intricacies as we study those numbers. This is the case for lithium metal and this is the case for lithium oxide as you see here the binding energy is actually lower for lithium metal because of the increased screening of the 2 S electrons and therefore the binding energy will always be on the lower side and that is what you see here in the two peak spectrum if you have lithium oxide and lithium metal together then you see two peaks coming one corresponding to lithium metal and the higher binding energy peak is actually corresponding to lithium oxide and that is because the lithium 2 S electron has been given to oxygen in the bonding and therefore the binding energy is much more stronger for lithium oxide. So you would get a two peak spectra which clearly tells what is the center and what is the coordination site that is involved in this compounds. So examples of other chemical shifts we can see here this is a compound with the fluorine as substituted groups and then you have carbon hydrogen bonds carbon oxygen bonds. So clearly we have three different range of substitutions in this polymer and if you take a XPS spectra you can clearly find out that carbon 1 spectra will have three peaks and all these three peaks correspond to different bonding CF will give at a very higher binding energy than CO than CH. So clearly the peak separation suggests that there are three different sets of carbon that is present and similarly you take a spectra of silicon in comparison with silicon fluoride of different fluorine substitutions SAF or SAF 2 SAF 3 SAF 4 then you can clearly see the silicon spectra gives different features and this is what we call as satellite feature these two are called as satellite features and if we properly deconvolute then we can quantify what is actually present in this sample. So if you map this in each case for example silicon the oxidation state is 0 and then silicon fluoride oxidation state is 1 silicon F2 is actually 2 plus silicon F3 here 3 plus. So if you make a comparison between the core shell shift here versus the oxidation state it is actually linear and as you would see here for each oxidation state the increase in the core shell shift is by the order of 1 electron volt. So roughly we can take this to heart that for every oxidation state the jump would be approximately 1 electron volt. So if carbon or any other element for example oxygen is 2 minus if you are looking for oxygen 1 minus then you would only see a difference of around 1 electron volt or less than 1 electron volt there it cannot be 2 then there is something else which is happening. Therefore this calibration clearly gives you an idea that it has to be somewhere around 1 electron volt for a change in oxidation state by plus 1 or minus 1. This is another case if you have CF4 or CO2 or CH4 as we saw from the previous graph we can clearly see the carbon 1s peaks are shifted core level chemical shifts these are related to the overall change on the atom and the reduced charge which means increased binding energy there are number of substitutes which can be evident from the number of peaks. The electronegativity of the substitution also is magnified between oxygen, fluorine and hydrogen. So the shift is very clear and the formal oxidation state can also be traced. In fact for many of the organic compounds or material with organic substitutions just like IR which gives fingerprint region so is the case for these functional groups. In the case of carbon nitrogen single bond you can see 286 in the case of keto group or ester group it is 289 and when we go for fluorocarbons it is 290 and when we go for complete substitution of CF3 you can go to 293. As you see here the range of shift in the binding energy for carbon 1s can vary up to 285 to 293 which is nearly 12 electron volt and that totally depends on the substitution that is going along with the carbon atoms. So this can become a very useful tool because as I told you carbon 1s 285 is a very good calibrant which is the hydrocarbon. So compared to that any other functional groups that are there you can easily pick out selectively which particular carbon atom is there. For example if you take a compound which is hygroscopic it will also absorb carbon dioxide so it will form carbonates. So easily you can map based on the position of your carbon 1s whether it is a carbonate impurity is present. So the core level chemical shift for carbon 1s can be easily found out and therefore the chemical environment can easily be ascertained in our materials. In the next example we will take example of metal oxide in comparison with pure metal and try to see what is the approximate shift and what is the nature of shift that you can expect. As you can see here this is iron 2p spectra and in this case you can see metallic iron actually shows a much sharper peak but once we talk about oxides or iron oxide then we see a more broader peak and the difference between the iron oxide peak Fe 2p peak from that of the pure metal is approximately of 3 electron old. And this 3 electron old suggests that the oxidation state of iron actually has shifted from 0 to plus 3 and similarly let us take the case of a partially oxidized magnetite and if you map the Fe 2p line spectra of this magnetite you would find out apart from the Fe 3 peak which is visible here very clearly you also see a small inflection and there is a asymmetry in the XPS peak. The moment we see asymmetry in the peak then we got to be cautious the base line shift here can actually come from many issues which we will also see but one of the things that we need to find out is the sort of a satellite feature that is bent here and if we try to deconvolute we would see a proportion of this peak and a smaller proportion of this peak coming into picture which suggests that it is partially converted to Fe 2 plus. So the nature of the XPS peak will clearly tell us not only about the proportion but also about the mixed valence states. Take for example the case of oxygen 1s spectrum and in the case of metal oxide you would see a very clear oxygen 1s whereas because of surface hydration you will also see a satellite peak which is coming. So you can easily map whether the ion oxide is pure or it is having some water of hydration associated with it. Similarly in the case of cubic boron nitrate crystal if you try to take this crystal and look at the boron 1s or nitrogen 1s you would be confused with the sort of spectra that you are seeing here mainly such an unsymmetric broad spectrum is coming because of the surface impurity and in the case of boron nitrate you would actually have an oxide impurity which brings about this broadening and this multiplied feature and similarly the case of nitrogen you can see that there is a ambiguity there. What you do here is as a protocol in XPS before we map the 1s or 2s or 2p spectra as a regular protocol it is suggested to sputter the sample with 200 electron volt organ ions and this sputtering actually is bound to remove all the surface occlusions or adsorbed molecules. Therefore after sputtering you can clearly see that the boron 1s is giving a very sharp Gaussian peak. Similarly nitrogen 1s gives a very resolved peak therefore it is very important when we study XPS that we just do not map your spectra rather you do the surface treatment in the first place but one should also understand this organ sputtering can modify the oxidation states of the elements which are already present therefore we need to be very careful to have a rough idea about such a impact of the sputtering ions on the sample therefore if we look for look for a mixed valency then we need to be doing this in a much more careful way but generally the organ sputtering will try to take out all the surface impurities which are settled on the sample. Let us take the case of aluminum oxide. Aluminum oxide actually we can not only map the trivalent state of the zero valency in this case the metal and the aluminum oxide clearly shows a two peak spectra so we know that both aluminum oxide and the aluminum is there but apart from that we also get to see that the oxide thickness can be mapped which is of the order of 3.7 nanometer how do we do that we have already seen what is the length depth that we can scan which is ranging up to 10 nanometer based on that when the oxide is actually 9 nanometers it is possible to distinguish the contribution from both the oxide and metal photo electrons and the expression that is given here is d which is the depth or the thickness of the oxide layer which is given by 2.8 ln 1.4 and this is the ratio of your oxide versus the metal photo electron peaks therefore we can actually quantify what is the amount of surface oxidation that has occurred on a aluminum foil. So XPS can actually give you a quantized picture of that situation we will also look at some more examples of mixed valence states in mixed metal it is actually metal oxides. Take for example the perovskites the transition metal 2p bands showing both plus 3 and plus 4 states the first peak what we see here in this case is actually coming from the plus 3 state and then we also see a satellite feature. So for all compounds whether it is ferrites or manganites or cobaltites we can always look at the mixed valence states of the B cation and therefore this can become a very useful way to analyze it and as you see here when we have mixed valency the 2p 3 by 2 is asymmetric with the satellite feature therefore one has to deconvolute this spectra in order to quantify how much of 3 plus and how much of 4 plus ions are there and similarly in this case all the perovskite compounds like lanthanum nickelate, lanthanum ferrite, lanthanum vanadate all this show asymmetry related to the 3 plus and 4 plus states. Here is another cartoon which gives us a range of oxides starting from LA-MNO3, cobaltites, ferrites everything can be mapped for example these 2 compounds can serve as a standard when we are specially analyzing mixed valence compounds or substituted ones as you know here a pure LA-MNO3 actually should have Mn in 3 plus but when you substitute calcium in lanthanum site Mn3 and Mn4 ratio will be present. So how do we map that we can take the individual spectra of Mn2O3 which is predominantly 3 plus and MnO2 which is predominantly 4 and therefore these 2 can act as a standard to find out how much of 3 and 4 is present and looking at the peak positions we can clearly map whether the 3-4 combination exist. Similarly in the case of ferrites we can use Fe2O3 as a standard to see how much of reduction in the oxidation state or oxidation has happened from Fe3 to Fe4 and here again there are other examples given only thing in this case it is about K absorption H therefore you can see the values are much less because we are talking about K shell so this K absorption H can also serve as a useful number when we try to map between substituted ones and the parent perovskate oxides. When we look at the XPS there are other advantages that we have for example if we look at lanthanum nickelate and substituted with more than 2 metallions all that we can observe from the lanthanum 4D 5 by 2 and 4D 3 by 2 you can see here hardly there is any change in other words the lanthanum valency is not affected in the process when substitution is going on therefore rareth positions peak positions can also serve as a internal standard because they do not exhibit mixed valency except for presiodimium which shows 2 plus 6 plus and 4 plus oxidation state lanthanum in general can show a very precise number and so is the case for oxygen 1S you would hardly see any change happening unless or otherwise a peroxo species is present. So when we have this perovskate compounds taking a look at lanthanum and oxygen position itself can be a useful calibrant to see whether the effect is predominantly coming from the mixed valence states. So the features of such XPS spectres are given here and this can also be used for a methodical study. For example if we are going to make a systematic substitution of zirconium into Siria then this can use as this technique can be a very successful one as you would see here with substitution of zirconium you can see this 890 EV peak is actually decreasing in intensity why because zirconium is in 4 plus and cerium can actually exist in both 4 as well as 3. So when you substitute zirconium then there is a fallout of one of the oxidation states of cerium as a result you can clearly see a systematic trend and so is the case for zirconium you can see the position or the peak intensity varying as we substitute in Siria. So we can clearly map or make a very detailed study especially when we are looking at solid solutions of two dissimilar metals. And similarly XPS has been used almost to crack the chemistry behind the superconductors and I will be dealing with the superconducting oxides in one of the lectures in module 5. But here quickly I want to bring out the usefulness of X-ray photoelectron spectra in elucidating the electronic states of the copper valency for example in the case of the 34k superconductor which is lanthanum strontium cuprate compared with the 90k superconductor the conductivity as we see both are metallic but if you look at the formal valence of both the 34k and 90k superconductor this is actually nearly the same compared to a semiconductor and the formal valence also goes beyond to suggest that copper is not exactly in copper 2 plus but it is in a higher valence state. So if you make a comparison with a standard compound which is cupric chloride which is a well studied system and if you try to reposition this to match with the copper 2p core level spectra you can clearly see that there is the copper 2p 3 by 2 and 2p half spectra's they are both divided in other words they show a split in each peak. So the split in each peak clearly shows that copper is not remaining totally in copper 2 plus but a part of copper 2 plus is oxidized to copper 3 plus and that is happening in order to bring the charge neutrality into this lattice and therefore this valence state of copper 2 plus and copper 3 plus has really helped the community to ascertain what is the conduction mechanism which actually drives this superconducting oxides to show absolute zero resistance. So XPS is a very useful tool therefore to provide the mechanism in a much more elegant way and this is another example which I will be talking about in module 1 under combustion synthesis where noble metal ionic catalyst has been proved to be existing because if you think of any catalysts with noble metals like platinum on silica platinum and Syria or platinum on alumina it is always recognized that these noble metals are in zero valence state and that is why they show very high efficiency in converting the effluent gases in automobile exhaust. So till today most of the companies are using this platinum coated solid supports for catalytic conversion but what has been found recently that if you take platinum and actually treated with Syria in a combustion mode platinum is not actually sitting on the surface but it is actually getting precisely doped in the cerium site and as you would see here very clearly these are the valence bands spectra for platinum, palladium and other noble metals whereas when palladium is actually doped into Syria you can see the spectra is completely different from the noble metals. So what it suggests that platinum is actually getting doped and it is in two place state and because of that one can see that 2 percent and 1 percent platinum doped in titanium shows a tremendous increase in the conversion compared to the undoped Syria state. So a mechanism has been very clearly elucidated based on XP spectra where the substitution of this noble metal precisely into Syria and the valence state has been ascertained using XP spectra and similarly if you take alloy and then try to see what is the influence of doping gold along with aluminum we can clearly find out that if you have gold then the oxidation of aluminum to aluminum hydroxide or aluminum oxide can be inhibited largely because of this alloy formation with gold. So if you take the XP spectra of aluminum and you do a comparison with aluminum gold alloy you can clearly see that the conversion rate is much faster in this case compared to gold alloyed aluminum. So when we try to look at the degradation process of alloys XP's again can prove very useful. So in next few slides I just leave with some parameters which we need to bear in mind what to look for and what not to look for or what to avoid in XP's data. Number one as we know that we are looking at the splitting and as a result we are mapping the unusual oxidation states and we will also be observing the auger lines in the XP spectra but apart from that we will also get X-ray satellites and surface charging effects, intrinsic satellite effects all this can come into picture. So one has to be careful while trying to analyze the nature of XP spectra. For example surface charging can become a very important issue when a specimen's neutralizer is actually on you can see this is the peak due to AL2O3 and this is due to aluminum metal but once you remove the specimen neutralizer because of the surface charging this peak actually gets shifted here but the aluminum peak will remain the same but this peak has actually moved here. So we need to be very careful about the charging effects which can totally mislead our analysis and also the XP's data analysis has to be viewed carefully now we can try to use the peak area analysis in this way or we can just draw a baseline this way and try to analyze but what is actually expected is a optimization of your baseline which has to be in this form. Now there are several softwares which can help us in adjusting the baseline correction so that the area under this peak can be mapped to quantify the amount of this oxidation state that is present. Therefore the peak analysis or the deconvoluted peaks need to be analyzed with proper precision. So this can prove to be a very useful technique especially when you play with the heavier metals. So more than organic samples the inorganic samples have taken a much better attention as far as XP's is concerned and lot more information can be drawn with XP's and this can be quantified. So this is one of the spectroscopy where it is not only quantitative but it is also a quantitative picture. In the next few lectures we will also look at other spectroscopic tools which can be used for surface analysis.