[Photoresist Series Part 5] The World of ArF Immersion Resists and Double Patterning

In the history of semiconductor lithography, there have been many instances where it was said that a limit had been reached, only for new innovations to break through each time.

ArF immersion lithography and double patterning are prime examples of this.

ArF exposure at a wavelength of 193nm supported advanced mass production for a long time before the advent of EUV.

However, the 193nm wavelength could not be changed.

So, how were patterns much smaller than that drawn?

The answer was to mobilize everything, including:

• optical system innovations

• optimization of resist materials

• process splitting

• overlay technology

all at once.

In this installment, we will delve into the world of ArF immersion resists and double patterning from the perspectives of 'why it was necessary' and 'what makes it difficult'.

What is ArF exposure?

ArF exposure is 193nm exposure that uses an excimer laser with argon and fluorine as the light source.

It has a shorter wavelength than the 248nm of KrF, allowing for higher resolution.

However, simply moving from 248nm to 193nm did not solve everything.

193nm is still optical lithography, and there are limits to the dimensions that can be drawn.

Furthermore, in advanced logic, pattern density for wiring and gates is high, and requirements for CD control and yield are extremely strict.

Therefore, in the ArF era, it was necessary to optimize materials, equipment, and processes as a whole, rather than just updating the light source.

Why did "immersion" become necessary?

Lithography resolution depends not only on the wavelength but also on the NA (numerical aperture).

To target even finer dimensions with ArF, it was necessary to increase the effective NA without changing the 193nm wavelength.

This led to the introduction of immersion lithography.

In immersion lithography, a liquid such as ultrapure water is filled between the projection lens and the wafer.

By interposing a liquid with a higher refractive index than air, the effective NA is increased, enabling the formation of finer patterns.

While this idea itself is very elegant, its implementation is not simple.

With immersion, it is no longer just an optical problem; new challenges emerge all at once, such as:

• the interaction between the liquid and the resist surface

• components leaching from the resist

• watermark defects

• bubbles and flow instability

• drying properties after immersion

and so on.

In other words, immersion lithography cannot be achieved through equipment technology alone.A resist that can be used stably in a liquid environment was necessary.

Requirements for ArF immersion resists

ArF immersion resists require more performance than standard ArF resists.

1. Must have sufficient transparency at 193nm

If too much 193nm light is absorbed, the film cannot be exposed uniformly to the bottom.

2. Must achieve both high resolution and high sensitivity

Advanced mass production requires both miniaturization and throughput.

3. Must interact appropriately with immersion water

If water adversely affects the surface, or if surface modification is excessive, it becomes a source of defects.

4. Must suppress leaching

If low-molecular-weight components or additives in the resist leach into the liquid, it can cause equipment contamination or re-adhesion.

5. Suppressing defects

It is necessary to suppress watermarks, particles, residues, and surface unevenness that are specific to immersion lithography.

For this reason, precise design is important for ArF immersion resists, including not only the polymer but also surface control agents, PAGs, quenchers, solvents, and additives.

Why was a topcoat used?

In the early stages of immersion lithography, a topcoat was sometimes used to protect the resist surface.

A topcoat is a thin protective layer applied over the resist film.

Its main roles are as follows:

• Reducing direct contact between the immersion water and the resist

• Suppressing the leaching of resist components

• Stabilizing surface wettability

• Suppressing watermark defects

However, adding a topcoat increases the number of process steps.

Therefore, development subsequently shifted toward topcoat-less immersion resists.

This is a design philosophy that imparts immersion suitability to the resist itself.

For material manufacturers, this meant they were required to incorporate not just photosensitive materials, but "surface chemistry adapted to the immersion environment itself."

Even so, 193nm has its limits

Even if NA is increased through immersion, there is a limit to the dimensions that can be drawn at 193nm.

This is where double patterning comes in.

Double patterning is a concept where high-density patterns that cannot be drawn in a single pass are formed over multiple steps.

Simply put,

is the basic idea.

This has made it possible to further reduce the apparent pattern pitch even with 193nm exposure.

What types of double patterning are there?

Although we say "double patterning," there are several methods.

LELE (Litho-Etch-Litho-Etch)

This is the most straightforward method.

A portion of the pattern is formed through the first exposure, development, and etching, and then the intermediate patterns are filled in by performing exposure, development, and etching once more.

While the number of process steps increases, the concept is intuitive.

Spacer-based (SADP, SAQP)

This is a method where a spacer is formed on the sidewall of a previously formed core pattern, and that spacer is used to obtain a finer periodic pattern.

While it is robust for increasing density, the process becomes even more complex.

Combined use of Cut/Block masks

There is also a technique where lines are first formed densely, and then the final shape is created by cutting away unnecessary parts.

For complex wiring in advanced logic, the concept of creating the final pattern through a combination of multiple masks, rather than a simple single exposure, has become generalized.

What is difficult about double patterning?

Double patterning is not magic.

Rather, it carries numerous difficulties as the price for extending resolution.

1. Overlay Error

For patterns formed by two or more exposures, their relative positional relationship is extremely important.

Even the slightest misalignment leads to line width variation, shorts, opens, and pattern defects.

2. Increase in Process Steps

As the number of process steps for a single wafer increases, so do processing time, cost, and the opportunity for defects.

3. Material Compatibility

It is necessary to ensure that the first resist, hard mask, second resist, and surface treatments do not negatively affect each other.

4. CD Uniformity

If there is a dimensional difference between the 1st pattern and the 2nd pattern, it will affect the final periodicity and electrical characteristics.

5. Design Constraints

In layouts designed for double patterning, the circuit design side must also incorporate splitting rules and coloring constraints.

In other words, double patterning is a technology that involves not only materials, equipment, and processes, but also design itself.

Where do resist materials have an effect?

While the "process" often receives the most attention in double patterning, the quality of the resist material is actually extremely important.

It has a significant impact, particularly in the following areas:

• Resolution

• LER/LWR

• Development contrast

• Etch resistance

• Immersion suitability

• Defect performance

• Surface interaction

For example, even if the first pattern is slightly rough, that effect propagates to the subsequent hard mask and second exposure.

As a result, minute roughness can sometimes be amplified in the final pattern.

Therefore, ArF immersion resists are required not only to have high performance on their own, but also to function stably within the entire double patterning process.

Why did we work so hard before EUV?

You might think, "If it was that complicated, shouldn't we have moved to EUV sooner?"

However, it took a long time to introduce EUV into mass production.

During that time, we could not stop the miniaturization of advanced logic and memory.

That is why ArF immersion and double patterning became extremely important as "bridge technologies."

Thanks to this technology, we were able to advance to finer generations even with the existing 193nm light source.

In other words, ArF immersion and double patterning were not just auxiliary technologies that bridged the gap until EUV.

It was.

The value does not disappear even in the EUV era.

Even now that EUV mass production is underway, the knowledge of ArF immersion and multi-patterning has not gone to waste.

The reasons are as follows.

• Not all layers will be replaced by EUV.

• Mature ArF processes remain important.

• The philosophy of overlay and defect control lives on in EUV.

• Knowledge of resist surface design and high-resolution enhancement is inherited by next-generation materials.

Furthermore, there is a possibility that multi-patterning concepts may be required even in EUV in some cases.

In that sense, the integrated knowledge of materials, processes, and design cultivated during the ArF immersion era continues to hold value today.

Summary

ArF immersion resists and double patterning are technologies that pushed the limits of miniaturization within the constrained wavelength of 193nm by:

• increasing NA

• adapting to immersion environments

• optimizing surface chemistry

• splitting processes

• maximizing overlay accuracy to the limit

.

What becomes clear from this is the fact that semiconductor miniaturization is not merely a competition of equipment performance.

.

ArF immersion and double patterning can be considered symbolic examples of this.

Next time, we will delve into "stochastic defects," the biggest challenge for EUV resists.

#Semiconductor #Photoresist #ArFImmersion #DoublePatterning #Lithography #SemiconductorMaterials #EUV #AdvancedProcess