Manufacturing Bits: Aug. 10

EUV mask cleaning process; mask distortions; EUV blisters.


EUV mask cleaning process
TSMC has developed a new dry-clean technology for photomasks used in extreme ultraviolet (EUV) lithography, a move that appears to solve some major problems in the fab.

TSMC and Samsung are in production with EUV lithography at advanced nodes, but there are still several challenges with the photomasks and other parts of the technology. Using 13.5nm wavelengths, EUV is a next-generation technology that patterns tiny features on a chip in a fab.

Before chips are produced in a fab, chipmakers require a component called a photomask. For this, a device maker designs a chip, which is then translated into a file format. Using various equipment in a photomask facility, the file format is transformed into a mask, which is basically a master template for an IC design.

In the fab, the mask and wafer are placed in a lithography scanner. The scanner then projects light through the mask onto the wafer, creating patterns on the wafer.

During the EUV process as well as other steps, chipmakers would like to use an EUV pellicle. A pellicle is a thin membrane that covers the mask. The pellicle is like a dust cover, preventing particles from landing on the mask.

The problem? ASML is developing EUV pellicles, but they are not quite ready. Imec and others are developing EUV pellicles in R&D.

This presents a problem. Let’s say a particle lands on the mask or the mask has a defect. During the EUV process, the irregularities might get printed on the wafer, which in turn can impact yield or kill a die.

Chipmakers have two options here. First, one could wait for EUV pellicles, but it’s still unclear when they will be ready for prime time. Second, one could move into production without EUV pellicles. This is possible, but it presents some challenges. In the fab, a chipmaker must clean the EUV mask often to get rid of the particles on the mask. Then, a vendor must inspect the mask more often to ensure there are no defects on the structure. This translates into time and money.

For some time, there was speculation that TSMC moved into EUV production without a pellicle. And in a recent blog, TSMC confirmed it and disclosed it has developed a new EUV mask cleaning technology.

“Depending on process requirements, EUV photomask is divided into two types – with pellicle and without pellicle. TSMC has chosen EUV mask without pellicle to enhance optical transmittance, thus reducing energy loss during exposure process,” according to TSMC researchers James Chu, Ivence Hu and Jenna Chang in the blog.

So to mitigate the defects on the mask, TSMC has developed a new “dry-clean technique” for EUV masks. “Instead of using traditional wet clean process with ultrapure water and chemicals, fall-on particles are rapidly removed by such a dry clean technique. Meanwhile, the fall-on source is precisely located by sub-nanometer analysis technique and therefore contaminations can be excluded thoroughly. With persistent tests and optimization, the fall-on particle reduction rate achieved more than 99% in 2020,” according to TSMC’s researchers in the blog.

“By means of fall-on analysis and contamination source elimination, the fall-on count of each 10,000 wafers decreased from hundreds of particles to single-digit particles, achieving 99% of reduction rate. Since its introduction, the amount of water saving and chemical usage saving has reached about 735 metric tons and 36 metric tons, respectively,” they said.

In the blog, though, TSMC was vague in terms of how the EUV mask cleaning process works. That’s not surprising. Chipmakers often don’t want to reveal their secrets for competitive reasons. However, TSMC filed a patent in the arena, which might provide some clues.

EUV mask distortions
In the August version of the BACUS Newsletter, Hanyang University has presented a paper that explores pattern distortions and temperature variations in EUV masks. (To find the paper, go to this link. Then, download the August PDF newsletter on the page.)

An EUV mask itself consists of 40 to 50 alternating layers of silicon and molybdenum on top of a substrate, resulting in a multi-layer stack that is 250nm to 350nm thick. A ruthenium-based capping layer is deposited on the stack, followed by an absorber based on tantalum.

In operation, an EUV scanner generates photons, which bounce off several mirrors and then hits the mask at high temperatures. A mishap in the process can cause defects and unwanted variations.

“Due to the high absorptivity of (the) EUV source, thermal deformation can occur in the mask as well as in mirrors, pellicles, and wafers. Thermal and structural deformation of EUV mask during the exposure process can be an important issue since these masks are subject to strict image placement and flatness variations. EUV masks are composed of complex layered structures, which absorb energy from each layer during exposure, increasing the temperature of the mask. This can lead to thermomechanical deformation, which can lead to poor pattern quality,” according to Chung-Hyun Ban, Eun-Sang Park, Ui-Jeong Ha, Chae-Yun Lim, and Hye-Keun Oh from Hanyang University in the BACUS Newsletter. BACUS is an international technical group of SPIE dedicated to the advancement of photomask technology.

In the paper, researchers looked how the mask is deformed in the process. For this, researchers developed a simulation model. Then, a finite element method (FEM) is applied to this model. The results and conclusion of the study are intriguing.

EUV blisters
The Dutch Institute for Fundamental Energy Research (DIFFER), Eindhoven University of Technology and University of Twente have explained the so-called blistering effect in EUV.

As stated, in operation, an EUV scanner generates photons, which bounce off several mirrors and then hits the mask. The photons are originally generated from tin plasma. The mirrors are multi-layer structures, in which the top layer is based on a ruthenium.

The problem? At times, a layer of tin contamination allows hydrogen into the underlying ruthenium layer, but blocks it from leaving again, according to researchers.

This in turn forms blisters, which are high-pressure pockets of hydrogen underneath the ruthenium capping layer. “Our earlier work had established that hydrogen and tin stick readily to the ruthenium surface, and that tin proximity aids hydrogen penetration into the ruthenium,” said Chidozie Onwudinanti, a researcher in a paper. “However, hydrogen solubility in ruthenium is low. So we faced the question: how do so many hydrogen atoms get into, and through the ruthenium layer to form blisters?”

Leave a Reply

(Note: This name will be displayed publicly)