In the first part of lecture four, we're going to begin by looking at transistors.

How they work?

How their assessed, how their performance is assessed.

And the particular architecture of transistors that can be used most conveniently for organic semiconductors.

So a transistor is set to be an active component. All the previous components, diodes, capacitors, resistors or what are known as passive components. The transistors active becausw. What it does is controlled by the output from another device.

And it has a third electrode.

And there are lots of different architectures of transistor that can be made with inorganic semiconductors, like Silicon.

And they all they all employ.

Some kind of doping of small areas in the Silicon to make them heavily P type or heavily N type or create PN junctions.

But there are some architectures which don't do that.

And one of these architectures is the metal oxide semiconductor field effect transistor.

And it's this architecture that's most often used when we're dealing with organic semiconductors, for reasons which we'll go into later.

So what's a metal oxide semiconductor field effect transistor or mosfet to give it its usual acronym?

Well, the important thing about these transistors, which is significant for organic electronics.

Is that you only need one type of semiconductor?

To fabricate one of these devices, either N type or P type.

You don't need a PN junction.

Or at least.

A PN Junction is not necessary for the operation of the device.

So this is this diagram shows how how a mosfet is made and in this particular masseffect.

We're showing what's called an N Channel Mosfet. Now it's equally possible to make a P Channel Mosfet. It would just be exactly the same device, but with everything inverted.

And so the way this is done.

Is.

If you're fabricating an end channel mosfet, you start with a substrate of lightly P doped Silicon. The light P doping is to make sure there are no accidental charge carriers in there that might interfere with the operation of the device when it's when it's in its off state.

You can be sure.

That if it's an N Channel Mosfet and the substrate is likely P doped.

You're not going to get any charge flowing through the device if the gate is not switched on.

So you start with that.

You then.

Either thermally or in some other way, you oxidize the top layer of the Silicon substrate to Silicon dioxide.

And Silicon dioxide.

Is, of course, an insulating material. It's what's called the dielectric in the.

In the transistor, and, incidentally, This is why capacitance.

Is an important parameter in the operation of transistors because they always involve some sort of dielectric.

And.

Once you've done that, you then drill two holes through the.

Through the gate oxide.

And then you use thermal implantation.

To heavily N dope.

The Silicon.

Immediately at the interface with that whole so.

The yellow regions in the diagram are heavily anti porn, therefore they're very conducting.

So they they water acting as the source and drain Contacts.

Because silicones are semiconductor, we have to do this to in any fact make it as if it was metallic.

And then we evaporate metal.

To make the sauce and drain Contacts.

And also to form the third electrode, the gate electrode. That's a layer of metal sitting on top of the Silicon dioxide layer.

So if the device is off. In other words, if we're not applying a voltage to the gate electrode, then it doesn't matter.

How large a bias would put between the source and the drain?

We won't be able to drive any current through the device because this is an N Channel Mosfet.

And the substrate is likely P doped.

Now if we apply a positive voltage to the gate electrode.

And that can be a voltage that's triggered by another device in the circuit.

Then what happens?

Is we induce a negative charge in the Silicon dioxide dielectric layer at the interface with that metal?

OK, and because the Silicon dioxide is a dielectric material, there's a dipole across the thin layer, and so at the other interface at the interface with the semiconductor.

We induce a positive charge. In other words, the thing behaves like a capacitor.

And because we induce a positive charge in the dielectric.

In turn, because Silicon itself.

Is pretty pretty much a dielectric material. We induce a negative charge.

In the layer of Silicon, immediately next to the Silicon dioxide.

Search what negative charges there are in the Silicon will migrate to there and accumulate in that region.

So in effect, what we've done by putting the positive voltage on the gate electrode.

Is to make the channel region. In other words the very thin layer of Silicon immediately next to the dielectric.

We've made that behave as if it's heavily an type.

And so now if we apply a voltage between the source and the drain.

Will be will have a current flowing between the source and the drain, so the device is on.

So this is.

This is how a masseffect device works, and the fact that we induce.

We induce a channel region that's of the opposite type to the original substrate means that this is called inversion.

It operates by inversion. That's what electronic engineers say.

So This is why.

Masseffect type architectures are useful for testing organic materials as semiconductors.

Be cause the problem with organic semiconductors is that unlike Silicon, which is an ambipolar material, in other words, it doesn't mind whether it transmits holes or electrons.

Organic semiconductors are usually unipolar, meaning they will only carry one kind of charge carrier.

Unhappy type.

Poly fighting, for example, is a very electron rich material, so it's amenable to oxidation. In other words, the generation of positive charges within it and therefore P type, but it's not really amenable to reduction very easily.

So effectively polici things are unipolar and P type.

And that means.

We couldn't make transistors with organic semiconductors that use PN junctions.

So we need the mosfet type architecture which can work.

For organic materials, because we only need one kind of charge carrier.

Incidentally, this has been something of a roadblock for the development of organic electronic circuits using transistors. Because electronic engineers are most familiar with.

PN junction type devices. Because of course they are trained on Silicon. They're brought up with Silicon.

It starts really being something of a problem for the the area of organic electronics over the years.

There is a lot of research going on at the moment in into attempts to find.

Ambipolar organic semiconductors, and there's been a lot of success actually in the last five to 10 years.

And we'll say some of that research later on when we talk about Poly Thiophenes in transistors.

In fact, that researchers now got to such a pitch that people have even now made conjugated organic polymers that are organic semiconductors, but are highly electron deficient and only carry electrons. So they uni polar in end type.

And at some point this may open the door later to the development of PN junction type devices using organic materials, but as far as I'm aware, nobody's really managed to do this yet.

So most facts can be used for organic materials as semiconductors.

So how are they employed to test organic materials as semiconductors?

While there are some slight differences with the classic metal oxide.

Semiconductor field effect transistor. Because organic materials are much more chemically delicate than Silicon.

So.

Effectively we have to design.

Devices.

Where?

All of the high temperature and aggressive chemical reagent type steps that are used to fabricate the device are done first.

And the organic semiconductor is only.

Added or put down as.

Somewhere near the last step in the process. So this is the way that people typically make.

A mosfet type device.

This employs an organic materials transistor.

So again, we're starting with a Silicon substrate.

And it doesn't matter whether it's P or N doped, it would probably actually be lightly P doped.

And then the whole of the top layer of the Silicon is first of all ion implanted with N type dopant to make it heavily N type.

And therefore very conducting.

Next

that's.

Top anti player is subjected to oxidation so that the top layer of the heavily N type region becomes Silicon dioxide. So that's now insulating and that's going to act as the gate dielectric in the device.

Next

we have operate thermally source and drain electrode.

Untypically for organic electronic materials.

It's normal to use gold actually as the source and drain electrode. It just happens that for most kind of electron rich organic semiconductors, the Fermi Energy of gold is a good match.

For the **** of the organic semiconductor.

And that means that the contact between the metal and the semiconductor is an ohmic contact, so we don't have any problems with the device.

Not behaving as we expect because of unexpected schottky barriers cropping up.

Now the.

The device I've shown here uses a Silicon substrate and Silicon dioxide as the dielectric. Other people like to use aluminium as the gate electrode and alumina on the top of the aluminium as as the as the insulator, but it doesn't really matter. The principle is the same.

So the important thing about this architecture is you can do all of those high temperature or chemically aggressive steps first.

On the red organic semiconductor layer can be put down last.

OK, if it's a soluble conjugated polymer that might be by solvent casting. In other words, putting down a solution and then evaporating the solvent, or if it's a small molecule, it could be thermally evaporated as long as it's volatile enough.

So that's the great advantage of this, so you can see it's I've talked before about organic electronics. The possibility of using all organic materials to make electronic devices. This is not such a device, it's a hybrid where we're using Silicon technology to fabricate the device as well. Even though the Silicon is not acting as a semiconductor.

But this isn't this is a good way of just test making devices to test organic materials.

A semiconductors.

So having said that, how do we assess?

And organic material is a semiconductor.

Well, the way transistors are assessed.

The key parameter that you want to measure for a thin film transistor like this.

Is the so called field effect mobility?

That is how readily charge carriers can diffuse through the material with the application of a voltage.

And field effect mobility is denoted by the Greek letter mu.

And if you look at papers about organic electronics and transistors.

When people make a good organic transistor, they usually boast about having a high field effect Mobility Himu.

But actually an electronic engineer.

Is not only interested in the field effect mobility, becausw the performance of the transistor.

Is governed by the ratio between the field effect mobility.

And the square of the source drain distance L. So that's the channel length. That's called the channel length. But it's essentially the distance between the source and drain, and that's why it's important to try to miniaturize transistors, and it's why, with the enormous developments in Silicon technology, the size of transistors has consistently shrunk, because the performance of the transistor goes up if the source drain distance comes down.

And actually, that's a more important parameter than the field effect Mobility Becausw.

The term is a square term.

So Mu is the field effect mobility, which is essentially how fast charge carriers can move through the material between the source and the drain, and L is the channel length, which is the source drain distance.

So let's have a look at some transistor characteristics now.

Incidentally, if you want to know more about this or you would like an alternative source to my notes.

There's quite a good tutorial style review of the whole area of organic field effect transistors.

So little old now considering the developments in this field, but it's still a very good introduction to the subject.

So in this particular example, which I've plucked from the literature because it just happened to have nice pictures in the paper.

We're going to be looking at an organic field effect transistor that was made using small molecule pentacene.

Which will consider a little bit later on in the course.

So the first thing that you do with any transistor to assess the performance of the.

Device.

Is.

What's called an output plot. So what you're measuring here what you're doing is you're.

Keeping the cake. The gate voltage constant, you can see that then these are negative gate voltages because pentacene.

Is again a uni polar P type organic material, so we need a negative gate voltage to switch the transistor on.

And that's being fixed.

And then we're varying the source drain voltage and measuring the source drain current through the channel.

And.

Family, typically what happens in this case is that as you start to increase the source drain voltage for a given gate voltage.

The current will increase the more source drain bias you replied, the faster you can drive charge through the device.

But then at some point.

That increased levels off until it becomes essentially constant, so it doesn't matter after that if you increase the source drain voltage, you won't be able to drive anymore current through the device.

So the initial form of the plot, where current is increasing linearly as a function of gate voltage source drain voltage, sorry.

That is called the linear region.

And when it's leveled off and it's not varying anymore with source string voltage.

The transistor device is set to be saturated, not set to be the saturation regime.

And.

The reason for this behavior is fairly easily understood just based on the architecture of a simple simple thin film transistor that the.

Channel region where the charge is accumulated by applying a voltage to the gate. that Channel region is very narrow.

UN, especially inorganic materials, where the dielectric constant of the material is very low.

It's that Channel region can be very narrow, and in fact it's been said that for some organic materials it might be no more than a single molecule thick.

With a low gate voltage.

So if anyway, regardless of that, it is very narrow indeed.

And so there's just a physical limitation on how many charge carriers you can get between the source and the drain by applying the source drain voltage.

And that will.

It will be a natural limit for that.

So that's one reason. Just simply the architecture.

But the other reason reason is that if you.

You need a negative gate voltage to switch the transistor on, but if you've got a bias between the source and a drain, then clearly one of those electrodes.

Is going to be positive.

On the other one is going to be negative.

So the contact that's negative.

Well, we're trying to get holes through this device, and so if we make the.

If we make that electrode negative, it's going to cancel out the holes that were induced by the gate in the channel and so.

You're going to end up with a situation that looks something like this. Weather Channel becomes wedge shaped, it's thicker.

By the positive electrode, but much narrower by.

By the negative electrode, so that's where that's that's why the current starts to level off effectively.

And organic electronic engineers call that the pinch off effect. They say the device is pinched off.

And that's that's the other reason why the source train current.

Levels off even if you increase the source drain voltage.

So that's the first kind of plots. That's the output of the transistor.

Once you've done that.

You then do. You then measure what's called the transfer characteristics.

And what you do here is you use.

A substantial source drain voltage and you keep that fixed now.

And this time you vary the gate voltage and measure the source drain current.

And you can plot the source drain current at the log of the source drain current against the gate voltage. That's the pink line here.

And you also calculate the square root of the source drain current and plot that as a function of the gate voltage.

Now that plot should be linear.

But you can see that in this pentacene example here.

Initially.

You increase the.

Gate voltage not very much current flows through the device and it barely takes off at all and it's only around this point here around minus 10 volts or so that the Clock actually becomes linear. And that's because, like most organic semiconductors.

The pentacene will probably have some impurities in it. They probably cancel out.

The charge carriers that are induced by the low gate voltage.

And it takes a certain gate voltage before the transistor actually can switch on because there are enough holes there to enable the charge to be carried.

And.

So what we do when we do one of these transfer plots is we extrapolate the linear region back to 0.

Only current axis here and we read off the gate voltage and in this case it's about minus seven and a half volts, and that's set to be the threshold voltage. That's the voltage you need to apply before the transistor starts to operate, and that's an important parameter to measure.

Ideally you want that to be as low as possible.

So A is called the output Connector Istics B is called the transfer characteristics.

What about the equations that you use to calculate the field effect mobility from these measurements? Well, that's shown here.

In this particular paper.

The source, the channel length, the source drain separation in other words was ten microns in these devices.

And the width of the channel was .4 millimeters.

Now, that's not unusual. That's not often is the case, particularly with test type transistors. But the channel length, the source train separation is actually much smaller than the width.

The two equations that we need are shown here.

Now these are fairly empirical equations. I don't think they come from any deep theory.

They're just equations that electronic engineers have come up with over the years to characterize transistor characteristics.

The first equation, equation 1.

That's for the saturation regime of one of the output characteristics, so that's where the source drain voltage is not varying as a function of source drain at the source drain. Current is not varying as a function of source drain voltage.

And equation 2 is for the linear regime where it is varying with the source drain voltage.

So in this, in these expressions, the field effect mobility is mu.

C is the gate dielectrics capacitance per unit area, and in this particular example, this is one of the reasons why I use this paper. 'cause this is not always quoted in the information in the paper, but here it was.

It's 10 nanofarads per square centimeter.

W is the channel width and L is the channel length.

The threshold voltage is Vt, and notice that only appears, of course.

In the xpression for the linear part of the output plot.

And sorry, no, it it appears in both. It's the, it's the.

Gate voltage VD that only appears in in this part in this equation 'cause of course this is where the the source drain current is varying with the source drain voltage in the saturation regime equation one it's not, so it doesn't appear.

So let's have a look how we use this equation, or one of these equations. Let's look at the output characteristics and just pick this point here. This data point here, which is in the linear part of the curve. So we have to use this equation, equation 2.

So for either of these equations we need threshold voltage, so we have to get that from the transfer characteristics plot.

Um so.

Vt is minus 7 1/2 volts.

VG is the gate voltage.

So in this particular curve here, the gate voltage was minus 40 volts.

And.

Substituting the numbers in, reading off the current here, that was a.

120 microamps

is equal to field effect mobility times the capacitance. 10 Nanofarads, 10 * 10 to the minus 9 fire adds per square centimeter.

Channel widths let's put the width and the length in the same units meters, so the council.

So .4 millimeters.

10 microns on the bottom there.

And then there's the expression for all of the voltages here.

So working out these numbers and moving that over to the other side of the equation, we come up with this 300 equals the field effect mobility times all of these voltages put into the expression there. If we evaluate that.

It comes to 275.

So again, taking that over to the other side, the field effect mobilities 300 / 275. Actually in this case.

So roughly, it's one.

And the units are field effect mobility.

With all of the units that we have in this equation here come out as centimeter squared per Volt 2nd.

Right, those are the units for field effect mobility in the thin film transistor.

Now that's actually a pretty good field effect. Mobility for an organic semiconductor.

To give you some ballpark figures of what we're looking for here.

To make circuits.

With transistors.

We need at least.

.1 Centimeter squared per Volt second as a bare minimum for any semiconductor.

And it's actually really quite difficult to hit that number for organic materials, as we'll see later.

So 10 to the minus, one centimeter squared provolt seconds is a ballpark figure.

That we need to achieve if we really want to make circuits with transistors, and if we really want to do organic electronics with organic semiconductors.

The pixels in a flat panel display device, such as the one on your laptop.

That uses porous Silicon typically as the semiconductor that's chemically.

A chemically deposited form of Silicon. So it's not very pure.

And its crystal structure is not very ordered.

And so the field effect mobility for that material is typically around.

1 Centimeter squared, provolt second or maybe even less.

So the worst kind of Silicon that's useful in technology.

Reaches about this kind of value. OK, single crystal Silicon that's used in Silicon chips and the like has a very much higher field effect mobility.

OK, so that concludes what I'm going to say about characterizing electronically characterizing.

Thinfilm transistors.

And in the next presentation will start to consider Poly alkyl thiophenes are semiconductors and the sort of considerations that we need in order to use an organic material reliably as a semiconductor material in a transistor or other electronic device.