Lecture 31: X-ray Photo-electron Spectroscopy (XPS)

IIT Kanpur July 201
20 followers
Dec 2, 2023
Hello everyone, welcome back to lecture number 31. In this we will start to look at the X-ray
photoelectron spectroscopy. So in the last lecture we had looked at ultraviolet photoelectron
spectroscopy, we have seen the capabilities of it and now we will move on to X-ray photoelectron
spectroscopy and we will also look at the applications of X-ray photoelectron spectroscopy
to understand in greater detail what this spectroscopy can do particularly in the surface
science perspective.
Now you remember that something we have already discussed in the previous class, so what is
here going to happen is that we are going to use an X-ray photon that means a photon
that is higher in energy which is going to be basically shining on to the or impacting
the material and then you are going to basically kick out a core electron, yes so that is the
interesting aspect.
So you will also see that typically the energy that we are talking about in the X-ray photoelectron
spectroscopy is in the order of a few hundreds of electron volt, in the previous case in
the ultraviolet photoelectron spectroscope what we dealt was from a few electron volt
to a few tens of electron volt but now today or in this class we will be looking at a spectrum
which is actually much, much deeper in binding energy, so that is the difference.
But it has some importance as I told you that this spectroscopy is actually something which
we can do something like a chemical analysis of the surface, films, interface, material
whatsoever you are investigating, yes because the important thing is that the core levels
are quite sensitive to the chemical environment of a given element, so if you are looking
at the core level binding energy of a given element that is quite strongly dependent on
the chemical environment of this molecule, so that means we can understand formally the
oxidation state of the element itself.
So let us do the experiment, so like we can take an example and then try to understand
it in greater detail, so I have here a polyurethane film, so this is a thick film, so here it
is nothing to do with our surface or interface, so I am just taken a very nice example, so
that is the reason why I do that, so what you do first you take the film and then you
basically scan the energy over a wide range, so of course here you see that we are starting
now from 0 as always, then we are going up to about 100 electron volt, you remember in
the last class when we talked about the UPS, so typically we were just looking at this
range here and that is also something they are actually showing, well I can see also
the valence band peak, but that is not something we are really interested in this case, what
we are really interested is about the core levels of different elements.
So polyurethane, so this is basically a single unit of the polyurethane or the chemical composition
of the polyurethane and what do I have here in this molecule or in this polymer chain
is nothing but carbon, nitrogen and oxygen, well you see this scan which is spanning over
several 100s of electron volt is typically known as a survey scan, so that is something
you first do to detect what are the elements present inside, no surprise you see carbon,
nitrogen and oxygen, no surprise, but what is quite striking here is that the carbon
peak is the most intense and the nitrogen is being intermediate and oxygen is actually
somewhere, sorry the nitrogen is the lowest and oxygen being the intermediate intensity.
That clearly tells me that this is some kind of a proportional chemical analysis, so carbon
is the highest that is what I expect, nitrogen the lowest, oxygen the intermediate, so clearly
I see that I can detect the elements that are present which are carbon, nitrogen and
oxygen and then I also see some kind of a composition of the carbon, nitrogen and oxygen
and that is exactly what you see by the peak intensity, good, already the survey scan is
quite promising but now we can look into each of them separately, so zoom into these small
window of energy and if we do that then we can basically understand each of the element
in a greater detail, so let us do that.
So polyurethane what I am doing is basically I am zooming into the energy of the carbon
and this is particularly the carbon 1S level, so always you would find that in the XPS analysis
there will be a particular assignment like this, this is indicating it is a carbon 1S
level, core level that you are actually looking at and each of the core level will have a
characteristic energy but within the characteristic energy range you will also see multiple peaks
depending on the type of carbon that is available in this particular case, depending on the
case.
So here you clearly see that we will look in greater detail in the next slide but here
you quickly look into the system, you already see that I have basically carbon, alkyl carbon
I have, then I have aromatic carbon I have and you clearly see that I can also identify
many different peaks and each of these peaks are actually corresponding to different type
of carbons that are present within the polymer chain.
We will look at the oxygen, I also expect to see like 2 different type of oxygen and
that is also something I see which is actually at a slightly higher energy, this is the oxygen
1S level and you can also look at the N1S level, there you see only one particular peak
and that is actually corresponding to this aminic nitrogen peak, that is the only one
nitrogen that is actually present within the molecule and that is the one you are actually
detecting but now you see when you do this kind of an elaborated or some kind of a fine
scanning into the region which you are interested in, you can basically detect not just the
element, you can actually just detect the different type of carbon and even their proportion
because always the intensity of the peak is kind of representation, it is not an absolute
representation please keep that in mind but it is a kind of representation of the amount
of the element that is present.
So that means I have more of that alkyl carbon and then I have something like an intermediate
amount of aromatic carbon and aromatic nitrogen and that you also clearly see and then I also
have some kind of carbon which is also connected to oxygen, so that is the carbon that is actually
present here, so that is the interesting thing.
So now oxygen when you look there is only 2 and nitrogen 1, so this is good, so you
can now identify the elements and also the different type of elements and also each element
but their different chemical environment you can analyse, so that is the interesting thing,
so we are basically just looking at not only the element and also their different possible
oxidation states within the system.
Now let us try to understand this in greater detail, so that you have a clear idea how
to interpret the spectra itself, now this is actually a table that will give you an
overview kind of chart telling you the typical energy of carbon and also like what would
happen if you connect the carbon to a different element and then how it will basically be
just looking at, so this is of course a small guide but of course there are bigger table
available in the XPS database that is also available online and there you can look for
more details.
So now what I have here is typically a hydrocarbon, carbon is present at around 285 electron volt,
so this is the carbon 1S, please keep it in mind carbon 1S, so that is about 285 electron
volt but the moment the carbon is actually connected to nitrogen, you see the binding
energy has actually increased or when the carbon is connected to an oxygen or when carbon
is connected to more and more oxygen you see that the binding energy has increased tremendously
compared to hydrocarbon that is what I said, compared to carbon connected to carbon you
see about a 5 electron volt shift in the energy, in the binding energy is observed
when carbon is connected to more and more oxygen or when carbon is connected to fluorine
for example or chlorine you see also the binding energy is increasing.
What is going on actually or why is this happening?
You clearly see the carbon is connected to another element and that element is actually
more electronegative in nature, so carbon is connected to oxygen that is more electronegative,
so what happens is that the oxygen basically draws the electron from carbon and because
of this the carbon is actually getting more oxidized, so that means when carbon is actually
connected to more oxygen then eventually what happens to carbon is that the carbon is actually
more oxidized, when you have a highly oxidized species of course what happens is the core
electron is getting lower screening and therefore it is harder to remove that particular core
electron compared to a normal carbon, so the moment the carbon is connected to more oxygen
so what happens is a carbon is getting more oxidized, so when it getting more oxidized
that would eventually means it is harder to basically just kick that electron, it is also
to do something like this that when the outer electrons are actually being pulled that means
valence electrons are being pulled by the neighboring elements then the electrons that
are present in the core will basically start to bind more closer to the nucleus, so their
binding energy basically increases.
So in general whenever an element is actually connected to a more electronegative element
then what happen their binding energy of that particular electronic state in this case the
carbon 1s state is actually shifting to a higher binding energy, so in general the binding
energy value is an indicator of whether the element that you are looking is more oxidized
higher binding energy or reduced that means lower binding energy.
Now we can look back into this spectrum again of the carbon element and there you clearly
see that the hydrocarbon carbon is actually the one which is actually the least oxidized
or that should be the most reduced in this case which should give you a binding energy
the lowest, so that is something you can see here it is about 285 electron volt and
then when the carbon is connected to oxygen or nitrogen you clearly see that the binding
energy shifted as per this table, that is also as per your expectation and this particular
carbon which is actually the one which is connected to oxygen and also one nitrogen
therefore you would expect it to be shifted to a much, much higher binding energy because
that carbon is actually in a highly oxidized state, yeah.
Now you see the reason why these particular carbons are actually just looking at different
energies and also now it makes sense for you to how to analyze or how to identify that
particular carbon in this case or depending on the system that you are looking at, well
therefore you can clearly, clearly identify the element and also the elements chemical
environment and that is why I call it that you can basically understand the oxidation
state of the formal oxidation state of a given element.
Now we can also just see the same for oxygen 1S state but interesting thing here is in
this case you can see that the change in the binding energy as a function of the functional
groups or elements that are coming in the neighborhood is not as strong as you would
have expected it for carbon, yeah because oxygen itself is a strongly electronegative
element, therefore you are not expecting to see a huge change but nonetheless you can
see that if you have a carbonyl or an alcohol type oxygen you can basically identify these
oxygens clearly, yeah.
So that is the interesting aspect, so you can of course it also has to do with the energy
resolution of the device itself but in the modern XPS technique we can basically just
measures at a resolution of about 0.1 electron volt, this is not a problem, so therefore
you can see that this is actually quite identifiable, so we can clearly measure the details at an
elemental level and also we can identify the oxidation state of each of the element.
Now let me just show you also one important table, so this is quite useful for you, this
will actually be an identifier or kind of a chart which you can use in looking for the
position of energy and the type of element and its core level that you want to detect,
so if you want to detect for example a carbon 1S state you know that typically the energy
that you are expecting is in the order of 284 electron volt, yeah so all the energies
are by the way in electron volt.
Now if it is nitrogen you see it is about 399 electron volt, for oxygen it is 532 but
oxygen also can have basically the 1S core level and also can have the 2S core level
but that comes at a much lower energy, so typically you do not look at that what is
more interesting is basically this guy here which is the 1S level of oxygen.
But now you see also in the bracket here there is some kind of an indicator given which is
actually known as a cross section or the relative intensity, yeah so carbon is just taken here
as 1 so that means typically the oxygen resonance or the oxygen XPS resonance of 1S level would
be more intense, so 3 times intense than the carbon for example, so that is known as the
cross section that is also what is given, so for example fluorine you can look or when
you take higher elements gold I know that I can look at this range here which is basically
the 4F or 4D levels I can look at and so on.
So you can basically select which energy you want to look for and what element you want
to basically identify, so typically you would have a slight idea when you start doing your
experiment what are the type of element that you are going to detect and once you have
that idea then you can use this kind of a table in understanding the position of the
energy, yeah so this is kind of a help.
But you also notice that for a few elements like oxygen, fluorine or many of them you
actually have not just one type of 2P it also has something like a strange notation, so
what is basically that representing?
So for bigger elements you are also expecting basically that means core levels that are
not just S states you are also expecting something called as a spin orbit coupling, that means
you are basically exciting an electron and once the electron is excited then there is
actually a spin orbital coupling and due to that spin orbital coupling you will have basically
new quantum number that represents, so that is actually the spin orbit coupling quantum
number is basically known as J which is actually L plus S where L represent basically the spin
orbital angular momentum and S represent basically the spin angular momentum and for P state
you will have basically like L is equal to 1 and I also have basically plus or minus
half for the spin of electron that basically means I can have 2J values, the J values are
half and 3 by 2, so that is the reason why you see basically the 2P states are represented
either half or 2P 3 by 2.
So these are basically the 2 spin orbital states and this would be basically always
be observed in the elemental analysis or when you do the XPS spectrum of higher elements,
so for example if I would do for aluminum or if you do it for phosphorus whatsoever
you can see basically you can find it in this table you would start to find this.
And now there is also something interesting that you have to notice that that means the
2P state is basically now split into, so this is basically the 2P and due to the spin orbit
coupling so it is basically split into 2P 1 by 2 and 2P 3 by 2 and now you can also
calculate basically the degeneracy of this states and the degeneracy can be calculated
using 2J plus 1, so you can plug in the J values here and then you would find that the
degeneracy is actually equal to 2 for this and the degeneracy for this is actually 4.
This would means that always the P 3 by 2 level will be much more intense compared to
the P 1 by 2 level, so in this particular case it will be a ratio of 1 is to 1 between
the half P 1 by 2 and P 3 by 2, so this is basically representing due to the degeneracy
of the different level.
So please also keep that it in mind, so I will show you some spectrum where you can
clearly see this kind of spin orbital coupling and also the different peaks that are originating
from this spin orbital coupling and we can clearly use this kind of understanding in
measuring the elements inside.
So we need to keep this in mind while we look at the elements particularly the elements
like this one, so for carbon and nitrogen it does not matter because we are looking
at the 1S core level, yeah good.
So now let us also look at a few important aspect here, so we have already talked about
in the beginning that this kind of technique the photoelectron spectroscopy is quite surface
sensitive because most of the time what you are basically measuring is all in this region,
so that means it is typically a few nanometer around the surface, so that is what you are
basically measuring.
So that means it is a surface sensitive technique but now imagine that you have a kind of sample
which is like this that within this 10 nanometer, so this is 10 nanometer I have 5 different
type of material that are actually just sticking on top of each other, so that means I have
actually just made within this 10 nanometer 5 different material each having 2, 2 nanometer
thick, well here we are having a trouble.
Of course if it is just one type of material sticking on a surface no problem we can nicely
do the experiment and we can get the spectrum of that particular material on the surface
but this is a very special case where I try to generate a thin films of different materials
which are stacked on top of the other, yeah.
So now what happens if I would measure the spectrum I would get some kind of a spectrum
which is basically kind of the sum of all these different type of spectrum, so what
I would have expected from this blue material is this, from the green material is this and
from the magenta material is this and from the cyan is this and the red is this.
But now what I am basically seeing in the actual measurement is basically a sum of all
these things together right that is bad, so if I have special requirement where I want
to basically measure these kind of heterostructures then I need to make actually the XPS much
more surface sensitive that means I want to basically measure just materials property
from a very, very thin layer that means a monolayer or from bilayer, is it possible?
Yes it is we can do a technique which is actually known as an angle resolved XPS so that is
something which is generally known as angle resolved XPS what you do is that you basically
just rotate the sample with respect to the detector, yeah.
So now whenever I talk about general XPS detection is always about the surface being like that
and the detector is just on the top, yeah so this is something known as a take off angle
of 90 degree so that means the electrons are taking off from the surface perpendicular
to the surface so it is actually known as a take off angle of 90 degree.
But now I know that typically that everything within this region is something that I am
going to measure it is of course 8 nanometer it is surface sensitive but you see like it
is about 80 nanometer but the moment I tilt the sample so that means if I would just do
an experiment at a take off angle of 45 degree then what happens is the length that the electron
travels is still the same as previously but since the electrons that are started from
here in this part is somewhat not escaping the surface particularly when you take the
electrons that are coming at a take off angle of 45 degree or when I have a take off angle
of 10 degree you basically see everything that is at the interface is not detected but
something which is at the closer to the surface is what is detected.
So that means when you do an angle resolved X-ray photoelectron spectroscopy you are extremely
sensitive to the surface, yeah so that is quite striking and interesting because now
I can basically measure the sample that I have showed you in the previous case extremely
something at the surface, so I have also given a scaling here as you make actually
the angle smaller and smaller the so called take off angle smaller and smaller your detection
is basically becoming extremely closer to the surface, good.
So now I have an example here so I have taken it from an internet source so this is not
a publication so far but it is a beautiful example where what they have done is XPS on
a silicon dioxide deposited on silicon surface so what typically the thickness of this silicon
dioxide is about 10 nanometer on top of the silicon so this is the natural silicon dioxide
on top of a silicon and now what I am doing is an experiment where I am looking at the
2P core level of the silicon and I am doing the experiment as a function of the take off
angle.
So the yellow spectrum that means this spectrum here is the one which is actually recorded
at a normal take off angle so that means the detector is exactly on the top and then you
start to basically detect the rotate the sample and then start to detect so the take off angle
is actually getting smaller and smaller, so what you see is something very striking when
you have a take off angle of 15 degree you see that this is the spectrum where the silicon
resonance, so the first resonance the one at a higher binding energy is actually from
the silicon dioxide because you would expect it that silicon is connected to oxygen so
the silicon 2P should be at a higher binding energy silicon is about 99 electron volt that
is because it is pure silicon and when you have a take off angle of 15 degree what you
clearly see is that I can only detect the silicon dioxide and the silicon peak is almost
vanished that means this was somehow the construction of our material so this is basically the silicon
and then I have basically on top of it the silicon dioxide yeah this is the silicon dioxide
and now you see as you tilt the sample I am only getting information from the silicon
dioxide and the silicon is actually just kind of not possible to detect.
So this is quite beautiful so if you do the angle resolved XPS you are basically more
sensitive to the surface, so you are extremely sensitive to the surface and therefore you
can basically detect what is going on at the surface directly and that is why XPS is quite
a celebrated technique in surface and so you will also see like I will show you a couple
of more examples and also I would familiarize you a few more new interesting materials that
is clearly forming on surfaces there we will use XPS as a tool in detecting whether we
can in understanding the chemical nature of those materials that we form yeah.
Thank you very much for your attention and I see you in the next class with more examples
on XPS, thank you very much.
