[Photoresist Series Part 15] How will High-NA EUV change photoresists?
After EUV, comes EUV again.
However, it is not the same EUV.
The NA of the EUV exposure equipment currently used in mass production is 0.33.
In the next-generation High-NA EUV, the NA will increase to 0.55.
The wavelength is 13.5nm for both.
Even so, the resolution will change significantly.
And if the exposure equipment changes, the photoresist cannot remain the same either.
What is High-NA?
Conceptually, the resolution of lithography is,
CD ≈ k1 × λ / NA
can be expressed as.
By shortening the wavelength λ or increasing the NA, smaller patterns can be formed.
Low-NA EUV is 0.33.
High-NA EUV is 0.55.
The NA is approximately 67% higher.
According to imec, High-NA EUV has the optical capability to target 16nm pitch, or 8nm CD-class resolution.
No longer just an 'experimental' technology
High-NA has long been discussed as a future technology.
However, the situation has changed as of 2026.
ASML explains that by the end of 2025, it had shipped a total of 8 High-NA systems to multiple customers, with 6 in operation, and that the second-generation EXE:5200B has also met full specifications at customer sites.
The company is maturing High-NA for introduction into customer mass production processes by 2027-2028.
In other words, there is not much time left for material manufacturers.
Resist printed with 16nm pitch
In 2024, imec used ASML's TWINSCAN EXE:5000 to form 16nm pitch line/space patterns with High-NA EUV using single exposure.
The material used at this time was Metal Oxide Resist (MOR).
Furthermore,
24nm pitch contact holes: CAR
24nm pitch pillars: MOR
have also been demonstrated.
What is important here is that
MOR has not replaced everything, and CAR has not disappeared
is the case.
The optimal material may differ depending on the pattern shape and process.
Confirming electrical yield by 2025
Just because a resist image looks clean does not mean it can be used for mass production.
What is truly important is whether it can be processed into metal wiring and function electrically afterward.
In 2025, imec conducted electrical evaluations on 20nm pitch metal lines formed by single-exposure High-NA EUV and reported an initial electrical yield exceeding 90% for structures using MOR.
This means that High-NA has moved from
"getting a resist image"
to
"evaluating the potential for use in device manufacturing."
This means that we have moved to the next stage.
Why does High-NA make resists more difficult?
If the contrast of the exposure image becomes higher, it might seem that the resist would become easier to handle.
In fact, the high image contrast of High-NA is advantageous for improving LCDU and roughness, as well as reducing the required exposure dose.
However, at the same time, as the pattern dimensions become smaller, other problems become more significant.
Thinning
When fine lines are created with a thick resist, the aspect ratio becomes high, making them prone to collapse.
Therefore, a thin resist film becomes important for High-NA.
However, when the film is made thinner, it cannot withstand the etching process.
In other words,
I want to make it thinner, but I don't want it to be etched away
This creates a contradiction.
Thinking in terms of a "stack" rather than just the resist
The key to solving this problem is the underlayer and hard mask.
With High-NA,
resist
↓
underlayer
↓
hard mask
↓
The importance of designing performance across the entire material stack, known as the
target film, will increase even further.
In imec's High-NA demonstrations, not only the resist but also the underlayer, mask, illumination conditions, OPC, and etching are being optimized simultaneously.
In other words, next-generation photoresist manufacturers must change from
companies that only sell resist solutions to companies that provide patterning material systems
.
Stochastic defects will not disappear even with High-NA
One of the biggest problems with EUV is stochastic defects.
Because photon absorption, secondary electrons, and acid generation occur discretely,
line break
bridge
missing contact
merged contact
and other random defects occur.
With High-NA, the aerial image contrast increases, and improvements can be expected under certain conditions.
However, because the patterns themselves become even smaller, the influence of stochastic fluctuations does not disappear.
Even in the High-NA era,
optimization of dose, roughness, and defectivity
remains a central challenge.
Why MOR is attracting attention
There are several reasons why MOR is attracting attention with High-NA.
EUV absorbency
High resolution
Low LER
Pattern transferability in thin films
High etching resistance due to inorganic components
are the reasons.
In 2026, imec also reported that the dose response can be improved by controlling the oxygen concentration during the PEB process of MOR.
This is an interesting result.
This is because not only the chemical formula of the material,
but also the atmosphere after exposure influences resist performance
.
With High-NA, the boundary between materials and processes will become even more blurred.
Will CAR survive?
There is a strong possibility that it will survive.
CAR has a long track record of mass production.
There is a vast amount of know-how regarding raw materials, coating, PEB, development, and defect management.
Also, in imec's High-NA evaluations, CAR is being used for contact hole formation and other applications.
Therefore, for the time being,
an all-out war between CAR and MOR
rather than,
selective use and competition layer by layer
It is more reasonable to view it as becoming.
The third direction: dry resist
Furthermore, there are approaches other than liquid resists.
Lam Research is developing a dry resist technology formed through a vapor-phase process.
In 2025, it was announced that JSR/Inpria and Lam would collaborate on next-generation patterning, including High-NA EUV, MOR, and dry resist.
This indicates that future resist competition will span across
CAR
MOR
dry resist
underlayer
etching
.
Is High-NA a tailwind for Japanese companies?
There will likely be significant opportunities for Japanese material companies.
It is not just resists that will be required for High-NA.
underlayer
rinse
developer
high-purity solvent
hard mask
cleaning chemical
filtration
The required standards for all of these will increase.
These are fields where the high-purity, precision synthesis, and joint customer development that Japanese companies excel at will be put to use.
However, if the technical system changes, such as with MOR or dry resist, existing market share is not guaranteed.
Summary
High-NA EUV is not just a simple performance upgrade for EUV exposure equipment.
As patterns become smaller due to 0.55 NA,
resist thinning
etch resistance
stochastic defect
underlayer
MOR
CAR
dry resist
must all be redesigned simultaneously.
The winner in the High-NA era will not be the 'highest resolution resist,' but the material system that can provide the highest yield across the entire mass production process.
The next installment will be the final one.
Based on the 16 installments so far, I will predict how the photoresist market will change through 2035.
Reference Materials
Hashtags
#Semiconductor #Photoresist #HighNAEUV #EUV #MOR #CAR #Lithography #imec #ASML