So in lecture 13 we're going to have a look at how single molecule conductance studies have produced some interesting results in molecular electronics.

So this is a bit of a precis of what we're going to have a look at in the remaining few lectures.

We're going to begin by having a look at the sorts of molecular structure.

That carry charge well, or at least carry charge best as molecules go.

And we're going to examine the question of whether we can actually make something that we could reasonably describe as a molecular wire.

And as part of that, we're going to look at the mechanism by which molecules carry charge.

Then we're going to have a look at whether it's possible to make anything more sophisticated than just a molecular wire, or perhaps given, as we'll see, conductances of molecules are quite low. Maybe we ought to describe those as molecular resistors.

For example.

Could we make a molecular diode? Could we actually do what the other, um Ratner experiment, proposed by controlling the structure of the molecule?

And if we can do that, can we make that more sophisticated by proceeding to make a molecular switch? Something that behaves like a molecular transistor?

And finally, we'll have a look at.

An interesting question that's arisen as a result of all of these studies, which is the effect of the environment on the conductance of molecules in junctions.

'cause clearly even if we, even if we can fabricate a metal molecule metal junction where there is just one molecule between the two metal Contacts, that molecule has to be surrounded by something it might be surrounded by air. It might be in a vacuum. It might be in a solvent or in an electrolyte, and so the question arises.



How does the interaction of the molecule with its neighbors affect its conductance?

Becausw

if one can modulate the conductance of a molecule by controlling its surroundings, that opens the door to making extremely sensitive.

Molecular sensors.

So let's begin by having a look at the sorts of molecular structure that carry charge in metal molecule metal junctions best.

And as we'll see when we look at the mechanism of charge conduction.

It's worth stating at the outset that 4.

Short molecules I molecules of less than about 3 or 4 nanometers length.

The mechanisms nearly always being found to be some sort of tunneling.

And so.

What people use very often to to analyze experiments where they varying molecular length and looking at conductance or current is a basic sort of tunneling equation where the tunneling currents for a given bias voltage.

Is equal to some pre exponential factor called a.

Times E to the power minus beta L.

Where a is found to depend pretty well on the particular contact chemistry being used. In fact, it's sometimes loosely referred to as the contact resistance.

Although more recently, it's been it's becoming fairly clear that it's not only dependent on the contact chemistry, it does have some dependence on the molecule as well.

And beta is the so-called decay constant.

And that's a characteristic, largely anyway, not completely, but largely.

Of the structure of the molecular backbone between the Contacts.

And Tabita has units of reciprocal length because L.

The other factor in the equation there is the length of the molecule.

So to to see what this data usually looks like.

This is a plot from a fairly early paper in single molecule conductance experiments.

Where the office we're looking at, the conductance of a simple series of alkane guy files. And of course being physical chemists. They like to use these molecules because you can just buy a whole range of them from Aldrich.

And so they they use the STM break junction to measure the conductance of these. And then they're plotting the log to the base C of the conductance expressed in Nanas Emmons as a function of the. In this case, rather than the actual molecular length in nanometers. Here they're plotting actually the number of methylene groups in the backbone.

And you can see that as you make the molecules longer, the conductance decays exponentially because the log plot is linear.

And from the slope of that line you can extract beta.

And if you extrapolate the line.

Up to the Y axis point, where N here is zero. That would give you the pre exponential factor A.

The so-called contact resistance.

The length of the molecules, well, that can be obtained either using X Ray crystallography if the crystal structure of the molecule exists or if it doesn't, you just simply do some molecular orbital calculations. It's fairly easy to estimate the lengths of the molecules between the Contacts.

Now in this area I'm not going to talk in these lectures very much about contact chemistry. I'm only going to mention it peripherally, where if it's important. I will mention it.

But it's worth noting that quite a lot of different groups have been used in molecular electronics to make Contacts to gold electrodes.

In the early days, particularly, files were the favored ones, because files, as we've seen already.

Give a very strong.

Bond to gold electrodes.

And.

If you use a thio ether then.

The evidence is pretty firm that it coordinates to the Golders Rs minus as thiolate.

And the proton is lost.

And similarly with a range of protected files. So for example with a thio acetate.

The acetate group is quite readily lost and so those will also bind through thiolate and even something like iser fire cyanide. The cyanide group will fall off, 'cause that's actually that will also strongly coordinate to gold.

And so the the sulfur carbon bond breaks and the the dialate coordinates to the gold then.

So those make covalent Contacts.

And if you also, if you like to make her a very strong contact to the goal, then as an alternative which is arisen more recently.

It's been shown that if you take if you take trimethyltin functionalized conjugated molecules, because tin is a relatively heavy Atom.

That in carbon bond is pretty weak and if you expose these to a bag old surface.

That in carbon bonds can break. What happens to that in is not entirely clear, but that forms a covalent gold carbon bond, which also is a very strong and interaction.

Sometimes it's it's more advantage, advantageous actually, to use a slightly weaker contact, and so various neutral ligands can be used as well to form Contacts to gold.

And we'll see later on examples of thio ethers, in particular being used. There are the nice Contacts.

And parading for period I'll groups have been used quite often as well.

Other workers, particularly group in the US.

Favor what they call week Contacts.

Because they say it gives better statistical data.

In in these experiments, so they used a means. Now a means of. If you think about your coordination chemistry a bit.

A means are pretty hard. Ligonnes, pretty hard donors and gold is very soft.

Zero valent gold, gold, metal, of course, will be very soft.

And so that's a hard soft interaction, so it's not particularly strong.

And so a means, and so on. Only form quite weak bonds to gold.

So we've seen their beta value.

The beta value for alkane distyle.

With something.

Quite large alkanes don't conduct charge very well.

And if you workout the beta value in nanometers to the minus one for the data that I've just.

Zone.

It comes out as 7.6 per nanometer in that experiment.

So.

We could check that result.

From a completely independent technique that that's is dependent on totally different physical phenomena.

And in order to check whether we're really doing what we think we're doing, making metal molecule metal junctions and investigating their conductance.

So.

For quite a long time now, people have done experiments in solution using a totally different photophysical method.

So what they do people who are interested in this area, what they do is they make molecules.

Which have got a photoactive group.

That's relatively electron rich.

Unzen electron deficient.

Group.

And then they use the backbone, in this case an alkane to separate the two functions.

And then they rely on photochemistry.

So what lay in this particular instance, what they do? Porphyrins are very strongly colored. They're quite easily radiated and excited.

So you get promotion of an electron from the home over to the LUMO of the electron rich zinc to porphyrin unit.

And then you've got a what's essentially a tunneling barrier between this group.

On the electron deficiency 60 and that's represented by this dotted box here.

And.

So now once this thing porphyrins being photo excited.

Does an incentive for that electron to tunnel over the barrier, because then it can go into the lower lying LUMO on the C 60.

And if you synthesize, arrange.

This can be measured spectroscopically and we're using time.

Time resolved techniques. The rate at which that happens can be measured.

And so if you do that for a range of a range of these molecules with different length connectors between the donor and acceptor, then again you can extract a value of beta from the decay in the rate constant, and that's being done for this family molecules. And it turns out to be around 9 per nanometer measured in this way.

So considering the totally different physical phenomenon phenomenon that are involved in this and in the metal molecule metal junction work.

On the fact that the.

Contact chemistry is possibly also playing a role in the case of the metal molecule metal junctions. That's pretty good agreement.

So alkanes are lousy conductors, and we'd expect that becausw.

The conductance is by tunneling and for something which has purely Sigma bonding as a very large homolo low energy separation.

So what about conjugated molecules?

If we have π conjugated molecules, we expect a smaller **** LUMO gap.

And therefore, we'd expect a more conductive system, and would expect the conductance maybe to decay less less sharply as we increase the length.



So there will be a smaller separation between the Fermi energy of the metal, which is the key thing in determining the conductance and the frontier orbital on the molecule. Whether that turns out to be the home or the luma, whichever is closer to the Fermi energy.

And also if you're looking at a range of molecules of increasing length, obviously when you make a conjugated molecule longer.

You're going to increase the degree of conjugation, and that will actually lower the home olumo separation. This molecule gets longer.

So you'd expect that maybe to offset the increase in the length of the molecule.

So those two factors working together should mean a smaller beta and to look at a fairly recent example of a study.

Well, this has been established for look at these simpler Liga Thiophenes here with thio methyl groups as the Contacts.

And so gold electrodes, again as usual.

The point about the increasing conjugation as a function of length can be seen in these UV visible Spectra of these molecules, so this is looking at the.

The lowest energy band in the UV visible Spectra, which is the π to π star bands, which is delocalized over the whole molecule.

Or involves the the whole conjugation?

And so the the peaks here and normalized in intensity just to allow easy comparison. And you can see that as we make the molecules longer as we'd expect, the π to π star.

Transition comes down in energy.

Moves to longer wavelengths. So in fact, the by siphoning molecules colorless because it's only absorbing in the blue part of the spectrum.

But the quantify thing is quite strongly orange, because that's absorbing in the blue part of the spectrum.

So here's some single molecule conductance data for these molecules.

Now this is.

Collected using the scanning tunneling microscope break junction technique.

And in this plot, we're not actually seeing that the little steps in the in the log of conductance plot due to the breaking of the gold Contacts.

Because with they've been missed off the top here. The range only starts at 10 to the minus one G nought, so we're plotting the conductance. All the log of the conductance, rather in units of G nought here.

And so you can see that for the Python ethene molecule, if you trap a bitthief in molecule here in the gap as it forms, you gotta conductance plateau where conductances through the molecule.

And if you make the molecules longer, then this moves to lower conductance.

And if you do thousands of these experiments, and this is what's usually done.

'cause you want a statistical picture of the conductance?

You then plot the number of times a particular conductance value occurs against the conductance value.

So you've got a histogram plot.

And so you get a peak in the frequency of occurrence.

Which corresponds to where these plateaus lie most often.

And so you can read off from that the single molecule conductance, the average single molecule conductance for the molecule.

Andso

if you then plot that.

As a function of molecular length, you can see that it's a straight line, as we'd expect it's conducting. These molecules are conducting by tunneling, but actually the beta value in this experiment turned out to be quite large, unexpectedly large. For a nice conjugated system like a 5 theme.

It was five per nanometer.

But similar experiments have been done for by now for many other families of conjugated molecules.

And.

You can see that these Lego family any final in.

Systems

have a beater of around 3.7 per nanometer bit lower.

And Alego Fenile, surprisingly, perhaps appeared to have a lower beta value than illegal thiophenes, although the beta volume might actually be slightly influenced by the fact that these molecules were contacted with the stronger violate contact groups that might actually lower the beta value a bit.

But weak aiming Contacts surprisingly, perhaps have an even lower beta value, at least if you believe this data.

If you go to what's effectively like polygamous of polyacetylene.

The beta value is significantly lower.

Down these backbones that got aromatic units in.

So these molecules here.

These are actually.

Prepared from naturally occurring molecules that you can see that from the structure of these, these look like the things like beta carotene.

Molecules of that type that are found in natural systems, and that's happened. That's in fact, how these molecules originate.

So conjugated molecules have lower beta values than than alkanes, for sure.

So at this point, we'll stop for the first part of lecture 13.

And in the in the next part we'll have a look at what happens when we try and tune the structure of the molecules. Knowing the lessons that we've just learned to try to achieve as lower beta value as possible.