Lecture 27: Single Molecule Manipulation on Surfaces

IIT Kanpur July 201
20 followers
Dec 2, 2023
Welcome back to 27. So, in this lecture, we will have a look at something called single
molecule manipulation. So, you recollect that in the previous lecture, we had looked at the
atom manipulation on surface. At that time, I have told that we will create this kind of atomic
lattice on the surface using manipulation and by basically just creating atom or not adding an atom
there. So, you can basically also create something known as an atom ROM or an information bit based
on atom. But the problem, although the promise is interesting. So, yes, you have already seen
the number of bits that you can collect within the centimeter square space is very, very large
compared to a typical modern blue ray disc. The problem is actually to stabilize these atoms on
surface. I already told you in all the examples that we have talked about or discussed, mainly
the problem was actually that the imaging was done at very low temperature. The issue is as
soon as you increase the temperature, these atoms start to move around. That means, although it is
a promising application, maybe in the future it is going to be interesting or there are people
thinking of creating atom in a slightly different way. That is going to be another prospective,
which I am not planning to go into detail. But in any case, low temperature is somewhat
seemingly sufficient and mandatory at the moment. Therefore, people thought we can also use
something called molecules as an information bit or we can try to manipulate molecular electronic
structure. We could also basically use them as some kind of an information bit because they
can still be operated at room temperature because you cannot imagine having an atom
row operating at 4 Kelvin. So, this is practically impossible. Well, in the future, nobody knows what
is going to come up. But still, at the moment, it is not seemingly a feasible technology. But
this is something in prospect one can think about. Therefore, people thought about molecular
machines or molecular information bits. So, the example that I am going to talk about today is
actually tiny molecules that actually can change their shape or conformations and depending on the
conformation, we can basically think about possible applications based on these molecular
molecules on the surface. So, that is also something like an interesting aspect, a prospect
about surface science itself when you deposit these kind of molecules on surface.
There is also molecular machines also known right now. They also can do some kind of work
and you can also basically just make them operate in a way you want. So, that is also
was one of the Nobel Prize winning discovery in the last couple of years before. This was
actually one major discovery and they actually won the Nobel Prize for that. But I am just going
to put it in a slightly different perspective which is actually using molecules in kind of
as a molecular memory. So, what do you have here? You have here a thalassinin molecule,
but the difference here is that you have a different type of metal. So, that is also
something interesting about thalassinin in general that you can actually just put many
different type of metal atom in the middle and their electronic structure actually changes.
So, that is also the reason why people consider or people use this thalassinins a lot in
application. So, here tin is the case. But now when you put the tin in the middle, so the problem
is that the tin, the radius or the ionic radius of tin is much more than cobalt or iron or those
examples what we have seen in the previous case. There the molecules still remain like planar,
but here in this case the atom, the tin atom is a bit protruded out of the plane of the molecule.
So, that means when the molecule adsorb on the surface, you have two possibilities. So,
you have actually a possibility of having tin up or you have a possibility of having tin down with
respect to the surface. Well, what is so exciting about this? So, that is the question. Well,
these are two independent configurations of the molecule or conformation of the molecule with
respect to the surface. In gas phase it does not matter or in solution it does not matter because
they will be identical. But on surface you clearly see that the adsorption energy will
be different here. You also see that the geometry is different because in one case it is like this,
in the other case it is like this. So, that is the difference. And at the same time the metal
is also pointing in one case towards the surface and the other is actually upwards.
So, that means chemical interaction between the molecule to the surface as well as their
geometry is different on the surface. So, that is why we have basically two different type of
adsorption on the surface. Now, what is so interesting about this? What can we do with
this? Well, now something interesting we can do using STM. So, this is actually the molecule with
the tin pointing upward and this is actually the molecule with tin pointing down. So,
now the interesting thing is that obviously as you would expect that the tin pointing upward
where the molecule is having a bright protrusion at the centre. So, that is actually due to the
fact that the tin is pointing upward and in the other case you see that the centre is actually
appearing rather dark that is due to the fact that the tin is down. Now, what is something
very spectacular is that you can actually position your tip on top of this tin atom
and apply a voltage like 2.5 volt. So, you apply a voltage and then what happens is a tin atom
moves downward. And then you can again go on to the molecule which is actually the tin atom
pointing down you again stay with your tip there on top of that molecule and apply a bias then
what happens the tin basically comes upward that is what is shown by this double headed
arrow that means you can basically using the STM tip push the tin down and take the tin backward.
So, that is something like a switch, yeah that is why it is actually called as a switch and
why it is also switch in the context here is that you are basically switching from one
conformation to another conformation, yeah. Well, I told you these conformations are
absolutely identical in solution or in gas phase but on surface they are very,
very different because you can also measure the DI by DV spectrum that means the tunneling
spectrum on this molecule and you will also see that their electronic structure is different
because in one case the tin is directly interacting with the surface and the other
not that actually causes them to have a strong difference in the electronic structure.
Now, what you are seeing here is actually kind of a time lapse, a time dependent current
measurement by taking the molecule so you place your tip on top and then just apply
the 2.5 volt between the tip and the molecule and then keep on measuring the current as a
function of time. What you nicely see is that there is some kind of a telegraphic noise down,
up, down, up and so on. What is actually happening is that while you are measuring
the tin atom is going downward, upward, downward, upward within the molecule.
So, without changing anything within the framework of the molecule the tin atom is
going down, upward, down, upward and so on. Since the tin atom is moving down and up there
is a strong change in the current at the tunneling junction and that is what you are actually just
measuring like this. So, ideally you would see some kind of a telegraphic noise that
is looking like that but this state is actually corresponding to something like a tin up state
and this state is actually corresponding to something like a tin down state, yeah.
So, you can now control that is the interesting thing, you need to have a threshold voltage for
that if you have the threshold voltage you can basically push the tin through the molecule
downward and you can pick the tin backward and you can basically switch them up and down. Now,
that is something very interesting, so can we do something more interesting with that just
than manipulating? Yes, we can, we can do something interesting. So, now what I have
here is actually is an array of molecules, so the molecule you should identify like this,
so this is basically a molecule, yeah. So, they are not very clear in this image but each of this
you can see a clearly a pattern, so that is corresponding to a molecule and now what I am
showing you here is just an array of 4 molecules, so that means a cross section or just a section
from this part is being moved here, yeah and then what I have right now is that all the molecule
having the tin down, so all the molecule the tin is basically now pointing downward in all the
cases, yeah. So, this is the first molecule, this is a second molecule, this is a third molecule
and this is the fourth molecule and in all molecule the tin is basically pointing down,
so that means it is an array of molecule where all the tin is down, we call it a 0 because you
also have seen in the current signal that down is actually representing something like a lower
current, so that is why I am calling it as actually 0. Now, with the STM tip I can actually just go
to the centre of a molecule, so this particular molecule and I can basically just apply the
voltage and the tin goes up, the current increases therefore I am calling it as 1,
yeah. So, then I can actually start doing it in a systematic way and you see that I have basically
just using this row of atom, row of molecule I have created a 000 or 0001 or 1001 information
bit in a way I want, yeah. So, here I have actually just come to something like a completely
all upstate here, all the molecules are actually pointing upward that means the tin is basically
upward in this case and then I can put them all back and here I have come back to something like
all downstate, it is clear and beautiful, so you can clearly do this manipulation of tin upward
and backward in a very systematic way and what you have done is basically you have created an
information array with a molecule. So, what is so interesting about creating array of information
with the molecule, the molecule is small, it is again something in the order of a nanometre,
so that means it is again you can pack more information in a smaller space compared to
the current technology that you are using that would mean that if you would ever use something
called a molecular binary array, you can basically store more information inside and of course the
switchability is also very very high, the speed of switching is actually in the order of picoseconds,
you do not even measure this here, you can basically just measure them systematically
and you can also nicely create this kind of array, so this is something spectacular that
you can do. So, that is actually a bit in the future, you may also be using something called
molecular computers where each information bit is actually substituted by a molecule itself than
what you are using currently is a transistor. Now, I will show you one more interesting
example, so of course you can create now there is a huge amount of research which is going
on in this field where people try to create this kind of information bits and hoping that
one can actually use molecule as one day in information storage and things like that.
So, I just want to show you a slightly different example, so this is again a molecule which is
actually called as a porphyrin molecule, so this is similar to the phthalocyanin molecule,
a slight difference in the skeleton in the carbon skeleton but now you can also see that the
molecule is actually sketched here, so it is not so important how exactly the molecule looking
like, it is a 4-fold symmetric molecule and here in the background what you are seeing is actually
the image of a molecule, so this is nothing but a single porphyrin molecule and it is an
iron porphyrin molecule. So, the iron porphyrin molecule what you can do is you can dope with
chlorine and you can create basically like a chlorinated iron porphyrin molecule,
so this green dots are corresponding to those molecule which are having actually a chlorine
atom in the middle of the molecule. Now, I have here an array, a random array of
molecules having no chlorine, so this is actually something with no chlorine and this is the one
with chlorine, so that is the interesting thing. Now, you can also see here in the schematic,
so I have basically here some molecule with chlorine, so this is basically chlorine and
some other with no chlorine basically, so there is no chlorine. Now, what I can do is I can take
my STM tip, go close to the chlorine atom, switch the bias of the STM tip and then you can pick the
chlorine atom from one molecule and you can actually put it to another molecule, so that
basically means I can easily now take these chlorine atoms from one place to another place.
What is the advantage of that? That is exactly the same like we have talked about,
I can basically use this again as an information array where chlorine atom containing state will
be kind of 1 and the non-chlorinated state will be something like a 0, yeah. I will also make
you convince that in the next slide, so ideally you can do it many, many times,
so that is something very interesting you can do. You see here again a sequence of images
starting from 1 till 10 for example, you go like this and then here and then finally like this,
so you can basically see that I have actually manipulated chlorine atom in a very systematic
way, so you can notice this particular chlorine molecule with the chlorine atom, so what I do,
I go on top of it, I switch the voltage and then I pick the chlorine atom and then you can see now
the contrast of that particular molecule change because the chlorine atom is now is on the tip.
On the same molecule you can put the chlorine atom back and you can actually get the contrast
back again, so as minor changes in between but then again you can remove it, add it,
remove it, add it or you can even move from one atom, one molecule to another molecule and so on,
so there is enormous possibility, so it is up to you what you want to manipulate and you can
clearly go around and do this sequence of manipulation and you can clearly change the
atoms from one molecule to another. Now let us look at the electronic structure,
so when you look at the electronic structure you would understand what is really something we can
do. Now what is done here is actually again the DI by DV, so that means it is a tunneling spectrum,
so this is really the measurement STS and this is some kind of a theoretical calculation,
so this is a DFT based theoretical calculation, so just for your understanding.
So now when I measure the spectrum on top of the chlorinated molecule you see I have
clearly this nice peaks here, so I am just calling it as 1, 2 and 3, so these are the
nothing but the HOMO, the HOMO minus 1, HOMO minus 2 and so on, so that is nice I have 3 peaks
here and on the right side I have basically the LUMO and typically that is what I see within the
window of my measurement. But now when you remove a chlorine atom and then you do again the
measurement on that molecule where you remove the chlorine atom, then you suddenly see that
the electronic structure changes drastically and particularly this peak is missing now.
So the peak which was actually we call it the HOMO of the chlorinated porphyrin molecule is
simply missing and also there is one more peak here is missing and the only thing which is
remaining here is actually the 2, I am just calling it as 2 prime, so that is actually the
one for the iron porphyrin and on the LUMO side you see it is more or less same but there is tiny
differences. So what is exactly happening? So we then basically just did some kind of
theoretical calculations to understand it and using theory you can kind of reproduce the same,
so when you look at the chlorine states, so this is of course something called a
projected density of state that will give you a hint about how the density of state
of the molecule is being contributed from the elements separately.
So if you look at the chlorine P state, so you clearly see that there is something here which
I am calling it as 1 and here which I am calling it as 3, now the D state from iron is actually
contributing to a peak here which let us call it as 3, so now you see the correspondence 1,
2, 3 on the experiment and the 1, 2, 3 in the calculation, so they are nicely corresponding
and as soon as you remove the chlorine from the iron then you would expect that these two peaks
would vanish and then the only thing that would remain is basically 2 and that is exactly what
you see, so this is something very spectacular. Now you can also just see how the electronic
structure of a molecule itself is building up depending on what you are actually just
having on the molecule or not, so it is like you know dissecting the molecular electronic
structure, that is a great possibility that you cannot do with any other technique so far
existing and you can only do with scanning tunneling microscopy, you can really dissect
the molecule and you can look at the electronic structure.
Now coming back to the perspective, it is like you see the molecule that is having chlorine is
having a very strong electronic state corresponding to the chlorine states and you also see that the
homo-lumo gap is actually much smaller than that actually of the normal iron porphyrin molecule,
so this is also showing you that the electronic structure of the molecule is strongly changing
depending on the state of the molecule whether having a chlorine or not having a chlorine and
that is exactly why I told that again we can use these kind of array of molecule with chlorine
without chlorine as some kind of a molecular information array or the information bit for
example where the chlorinated state will be called as 1 because it is a high conducting state and the
non-chlorinated state will be called as 0 because it is actually a low conducting state, so that is
quite spectacular, so you can basically use this kind of molecular array in the future it may be
something that we would be using in our computer, good.
So you also see that there is something like a strong difference in the magnetic
property also of the molecule, the chlorinated one is some kind of a high spin state and the
non-chlorinated one is a low spin state and that also means that we could also use it as a as a
magnetic switch in order to store information, so this is quite practical, so that means in this
kind of application what you would expect is basically that your STM tip is going to be some
kind of a head that would write information and read information and then your surface with the
molecules or atom array is going to be actually the component where you are going to store your
information basically, so that is the aspect of it, so that is the way you have to see the
application but of course there are lot of limitations from the point of view of doing
this is actually what is limiting currently using them in any technology but who knows that may be
in the future we are going to see applications based on these kind of systems, good.
Now I want to switch a little bit to one more interesting application of tunneling
spectroscopy, this is actually the final thing that we are going to discuss with the tunneling
spectroscopy, this is not very celebrated but it can actually give you very powerful
information about the interface and also the surface itself but of course technically it
is very very challenging to do this experiment and that is why it is not very commonly done
but tunneling spectroscopy is something very commonly done, so what is this?
This is called inelastic tunneling spectroscopy, what is so inelastic in it? So now when we talk
about the other spectroscopy where we always said that the electron that is tunneling from
the tip to the surface or from surface to the substrate to the tip in whichever way,
you always showed that the electron is basically just staying at the same energy,
that means on the route the electron is not losing its energy, so that means wherever it
started it basically end up in the same level and then it finally just go into the bulk of
the material, so the energy of the electron was never lost while tunneling but what if the electron
that tunnel actually just uses its energy to do some kind of an excitation of the interface and
then of course it will lose some energy, so that kind of thing could happen then you can
actually have something called an inelastic component inside the tunneling spectroscopy.
Well the percentage of inelastic component is basically very very very less, it is even less
than 1 percentage and that is the reason why this kind of a spectroscopy is extremely tough to do
and therefore not very much celebrated in the technicality, technically it is very challenging
but otherwise you will see it is very powerful. So now imagine a situation that I have some
adsorbate molecules, so some adsorbate molecule that are sticking on the surface and now while
the electron is tunneling through if that electron energy is actually matching with some kind of
excitation that is present in the molecule like a vibrational excitation or a spin excitation
whatsoever then what can happen is basically that the tunneling electron can lose their energy and
therefore you can have two components called the elastic component and the inelastic component.
Well if that is the case then what will eventually happen is that when you do the
tunneling spectroscopy you will eventually measure something like both the components,
the elastic and the inelastic component. So generally you would have expected that
there is only elastic component but that is not true, there is always a very very small
amount of inelastic component so that means if the instrument is capable of measuring the change
then in the Di by dV spectrum you should basically able to see a kind of step like
thing which is indicating that it will be symmetric with respect to your 0 voltage,
yeah at both voltage you should be basically able to see a kind of symmetric step and that
symmetric step is basically representing the energy required that means this particular
amount is actually the energy required or the U inelastic amount energy required to
actually just do some excitation of the interface whatever is at the interface.
So now let us look at a particular example, so this is a particular example here we have
a carbon monoxide sticking on top of a metallic surface so it is actually copper 111 surface,
we have carbon monoxide sitting on the surface. Now when you place your tip on top of the carbon
monoxide and ram a tunnelling spectrum then what you see is that there are clearly 2 steps here so
this is quite symmetric with respect to 0 and there is another one which is also symmetric
with respect to 0 so that actually happens somewhere at 35 millielectron volt and 4
millielectron volt. So what are they? So this is exactly how you are going to see that Di by dV
because you will see some kind of a very symmetric step like features.
Then this turns out to be that these are actually some kind of vibrational excitation of the
molecules that are actually present at the surface so that means the tunnelling electron
while it tunnelled it actually gives a part of its energy to the molecule and then the
molecule basically start to do some vibration so it actually excites some vibration of the
molecule. So here in one case it is some kind of a vibration like this and the other case where
the oxygen goes on other side and the carbon goes on the other side so it is some kind of
vibration like this. So this is of course a high energy vibration so that is why it is at 35
millielectron volt and this is a low energy vibration 4 millielectron volt and you can
basically identify them in the tunnelling spectrum. So now well I can also do some kind
of a mapping of this so when I do the mapping what I clearly see is that if I would make an
array of carbon monoxide on the surface so it is about 513 carbon monoxide I have deposited on the
surface and now when you basically just measure the tunnelling spectrum of independent molecule
what you find is that there are some molecules showing a slight difference between each other.
So then the question that you ask what is this difference so why should there be slight difference
between different carbon monoxide molecule? Well imagine that you have actually just created a
carbon monoxide molecule with a 12 mass and another one with 13 mass then you would have
actually just expected that there is a difference in the vibrational excitation energy that you
require because it is a different kind of diatomic molecule with different mass.
You recollect that from vibrational spectroscopy the frequency is strongly dependent on the mass
of the atoms that is actually composed in the molecule. So therefore what you would
expect is actually for C13 you would require lesser energy to excite and for C12 oxygen you
would require slightly higher energy to excite a particular vibration that is basically the
vibration of the oxygen and carbon moving in this way. So if you map basically the di by dv
image at a given voltage then you would find that some of the carbon monoxide atoms are looking
brighter and some other are looking less brighter. Why is that? It is actually because the natural
abundance of the C13 is about 1.1 percentage and if you count this is about 6 bright dots
and that is basically representing an exactly 1.1 percentage of the C13 carbon.
So that means when you actually deposit 513 carbon monoxide molecule on the surface you
exactly have about 1.1 percent is actually just the C13 and that is why you see this.
So this is something interesting so I will also show you in the next lecture a few more example
so that you can understand a little bit additional perspective of this but here what we clearly see
is that you can basically just measure some of the vibrational excitation of a single molecule
that is adsorbed at the interface and you can even detect the isotope effect because you can
clearly see that difference in the tunneling spectroscopy. So with this I would like to
conclude the lecture and I see you in the next lecture with probably one more example from this
and then we will try to conclude the scanning tunneling microscopy part. Thank you very much.
