Name: Stanford Nanofabrication Facility: Dry Etching - Basics of Plasmas & Types of Tools (Part 2 of 4)
Uploaded: 2025-06-07T15:32:33.000Z
Duration: 23 min 18 s
Description: Thousands of YouTube videos with English-Chinese subtitles! Now you can learn to understand native speakers, expand your vocabulary, and improve your pronunciation...

Hi, I'm Jim McVitie with the Stanford Nanofabrication Facility here at Stanford University.

This is part two of a four-part lecture series on dry etching.

In this part, I'm going to cover the basics of plasmas and the types of dry etching tools.

Previously, I talked about an introduction to dry etching, and following lectures will cover dry etching mechanisms and choosing dry etching processes and tools.

In this presentation, I'm going to cover the basics of plasmas, plasmas and sheath reasons, parallel plate configurations, inductive coupled plasmas or ICP configuration, and finally, downstream configuration.

Okay, let's start with the basics of plasmas.

Plasma is commonly used for fluorescent lighting.

Over here in the diagram, I show basically a vacuum chamber.

We're going to be limiting our discussion to RF power since we're DC, you can also generate a plasma using DC, but from a plasma etching standpoint, it can cause a DC plasma to cause damage to wafers or device wafers.

So we're going to be only talking about RF.

I show here the gases coming in, I show gases going out by way of the vacuum pump, I see light coming out, and in the plasma, I have neutrals, and then radicals, the initial gases, and sometimes I have multiple ions, and then I have electrons, and of course, I already talked about radicals.

Okay, the plasmas are a steady state process that results after a gas breakdown.

The electrons gain energy from the applied field, so I'm going to be putting RF power that can generate fields that we're going to basically heat up or give energy to my electrons.

The higher energy electrons through collisions will have ionization and excitation and fragmentation of our source gas molecules.

Okay, like I said, plasmas are a steady state process resulting after a breakdown.

In the plasma region, we mainly see the glow.

We have a charged balance where the density of electrons are going to be balanced by equal to the density of positive ions.

And typically, in the processing plasmas, we're talking about partially ionized gases, so there's less than 1% of the molecules or atoms in the chamber are ionized, so it's mainly neutral species.

The glow is from the decay of short lifetime excited species.

So any place you see, when I say short lifetime, I'm talking about nanoseconds.

So any place you see glow, that's where electrons are heating, are hot, they're basically breaking down and fragmenting the gas.

And when it's dark, it means we don't have any hot electrons.

Typically, such a plasma at 20 millitorr, we're talking about neutrals in the range of 10 to the 15th per cubic centimeter, a ratio of radicals to neutrals from 50% down to 10%, and ion to neutral ratios from 10 to the minus 3 to 10 to the minus 6.

So basically, it's mainly neutrals in the chamber.

Now I'm going to talk about RF plasmas and the sheath reasons.

The diagram here, I show a parallel plate type configuration.

So we're going to be basically showing that I have one electrode that can be grounded, and the opposite electrode, well, I'm going to be applying RF power.

And after the breakdown of the gas equilibrium, what I'm going to see, I'm going to have a central glow region, and then we have dark spaces on both sides of the plasma, between the plasma and the electrode, basically between the plasma and all surfaces.

And the reason we developed the dark spaces is, well, the reason is the electrons, being very small, travel much faster than the ions.

So numbers are like electrons have velocities about 100 times faster than ions.

So it means when we have the glow region, the first thing that's going to happen is the electron is going to start streaming out of this region, and they're going to charge up all the surfaces.

So I'm showing charge here on the two electrodes.

As the surfaces charge up, we're going to develop an electric field that's going to be retarding for electrons.

So if the electric field is this way, electrons are going to want to start turning around and going the opposite direction.

So we mean, so I'm showing here schematically, I have a high current of electrons coming out of the plasma, but most of that, those electrons get returned to the plasma, and only a small amount makes it to the surface.

But the amount that makes it to the surface, it can be balanced by the same amount of ions that make it to the surface.

So at the surface, the electron current and ion current will be equal.

And so we have the electrons and ions move from the plasma towards all surfaces.

The electrons want to leave much faster because they're going, let's say, 100 times faster.

And to maintain steady state balance in the plasma, we form a boundary layer that we're going to call the sheath or depletion region.

And this makes the electron and ion current or loss rates from the plasma equal so we can maintain our steady state.

Here again, I'm showing our diagram of the plasma and the sheath region.

And so I talked about the electrons and how the electric field turned most of the electrons back so we have the balance at the electrodes.

Well, the ions now are accelerated by the electric field.

And so they accelerate towards the surface, and they gain both energy and directionality.

So the ions come in isotropically into the boundary between the plasma and the sheath.

And at the surface, they're very directional.

In addition, they gain energy, and there can be hungry EV coming from the ions crocking the sheath.

So the ion gain, the energy gain, depends on the voltage across the sheath and the collisions.

So if I go low pressure, I'm going to have a few collisions or no collisions, and the ions are going to get the full energy across the sheath.

And this energy, so if I integrate the electric field here, I have what we call the sheath voltage across the sheath, and that's basically the voltage between the plasma and the electrode.

And with no collisions, the ions will basically gain the sheath voltage in terms of energy.

So lower pressure, less collisions, more directionality, more ion energy.

The sheath voltage depends on the system configuration, the applied power and pressure, and we're going to talk a little bit more of that in the next slide.

Oh, and I'll talk about it actually in a couple slides.

So first I want to talk about, give you an overview of the four configurations I'll be going into more detail.

So I have, first I have what we call a CCP plasma, and this can mean charged coupled plasma.

And I'll talk about a capacitive coupled plasma, and I'll talk about the meaning of that in the following slides.

And I have what we call a plasma mode, where you have equal electrodes.

RIE mode, where you have one electrode much smaller than the other electrode.

And then I'll talk about ICP case, inductive coupled plasmas.

And finally I'll talk about downstream plasmas.

So let's start talking about the CCP case.

So we have the CCP, or capacitive coupled plasmas, and again I'm showing the parallel plate case, where our plasma is sitting between the electrodes with a sheath on both sides.

So the plasma is driven by RF currents going through the electrodes.

So we have the RF supply here, the RF current going through the top electrode, it goes across the sheath region, through the plasma, and then out the other side.

And this is RF current, if we're going to be driving, keeping the plasma going.

So we never use DC supply for action because of the damage issue.

RF, we normally use 13.56 megahertz, it's an inductive plasma frequency, set up by the

FCCC, and although some of the time, there are special cases where we use lower frequency at the wafer.

We also use lower frequency for the ICP, but at the wafer, typically 13 megahertz.

RF current flow determines the sheath voltage, okay, let's talk about what this means.

So the electrons carry the RF current in the plasma region.

So in the plasma, I have a lot of electrons, they're very mobile, they carry the current.

But at the two sheath region, the depletion region, there's not enough electrons to carry the RF current.

And so how does the RF current get across the region?

And by what we call displacement current, or capacitive current.

And that's a way RF current gets across a capacitor, where we build up a charge on one side of a capacitor, and the two plates of the capacitor, the charge on one side induces a charge on the other side, and we have basically an oscillating charge across the capacitor.

So if we start to look at our sheath region, we have electrodes here, so it's easy to have an oscillating charge building up and going away on the metal electrode.

But at the boundary layer, I don't have any electrodes.

So the way I get my oscillating charge, the QDT, is by the sheath oscillating.

I deplete into the plasma, and then I undeplete.

And so the sheath will be oscillating up and down, and that's going to give me a charge per unit of time that's going to allow the displacement current, the capacitive current to go across the sheath region.

Now if we go in and integrate that charge, that gives us a voltage across it.

So what I'm going to get at here is that RF current, this oscillating charge here, is what gives us our voltage across the sheath.

And that, basically if you work out the details of it, you're going to get that the sheath voltage goes into RF current density divided by the electron density.

So if I have a higher RF current density, I'm going to have a higher voltage.

And this is where the area comes into play, because now if I change the area of my electrode,

I can increase my current density and increase my voltage.

So let's talk about it a little bit more.

So we're seeing that the RF of the sheath voltage depends on the RF current density.

Reducing the electrode area increases the sheath voltage, and therefore I'm going to be able to increase my ion energy across the sheath.

Okay, there's going to be two ways we do this.

So I forgot to mention that the capacitive coupling across the sheath, the displacement current or capacitive current is where we get the name.

Capacitive coupled plasma means that it refers to the capacitive current across the sheath.

The plasma mode is where we have parallel plate electrodes but equal area.

So I have the two electrodes at equal area, and in this case we're going to have, we get relatively low voltage across the sheath, and so we get low ion energies.

Subtitles ListPlay Video

Stanford Nanofabrication Facility: Dry Etching - Basics of Plasmas & Types of Tools (Part 2 of 4)

basically

typically

bias

amount