Improving sustainability via a new developer and light source

05 March 2024
Hank Hogan
A slide from Wisdom Pool Research's Seiji Morita, who discussed a biomass-based developer at SPIE Advanced Lithography + Patterning 2024
A slide from Wisdom Pool Research's Seiji Morita, who discussed a biomass-based developer at SPIE Advanced Lithography + Patterning 2024.

When it comes to sustainability, semiconductor equipment and material suppliers are working different angles. On the material side, one emphasis is sourcing while equipment suppliers work on efficiency, among other things. Both sustainability aspects appeared in presentations given on Thursday at the 2024 SPIE Advanced Lithography + Patterning conference.

For instance, Seiji Morita, chief research officer at the Wisdom Pool Research Institute, discussed a biomass-based developer. In the semiconductor lithography process, tools coat a wafer with photoresist, expose it to light, and then use a developer to finish transferring the pattern onto the wafer.

After mentioning the need to cut greenhouse gas emissions, Morita predicted changes in the materials used in this process, with a focus on the developer. “A biomass developer made from plants will be necessary,” he said.

His research team developed a process to manufacture such a developer. This manufacturing flow begins by liquifying plants and extracting oils from them. The next step involves filtering to remove metals and other contaminants, with the result used to produce the developer. The process is based on one devised earlier for biomass solvents, with steps in place to ensure batch-to-batch uniformity.

Calculations show about a 75 percent reduction in greenhouse gas emissions, Morita said. An even larger reduction is possible because the used developer can be a biofuel.

Morita’s research team tested three formulations of the developer, comparing them in their ability to print lines and spaces down to a 20-nanometer half pitch. They found one version of the developer outperformed the others, producing results comparable to a conventional developer regarding critical dimension uniformity and line edge roughness. The biomass-based developer did well with wide or narrow features, as well as those that were tall, Morita said.

As for equipment suppliers, xLight aims to improve sustainability, among other benefits, by changing the light source used in advanced lithography. A startup, the company aims to build the ultimate illumination source through system engineering of existing accelerator technology.

In a Thursday presentation, Chris Anderson, director of photonic systems at xLight, said the company’s products would be efficient and could provide multiple wavelengths. The light source would also supply light with tunable parameters, such as adjustable polarization.

“We have developed a source with a capability set and a level of sustainability that we believe can forge an even better future for our industry,” Anderson said.

The light comes from a free-electron laser, a technology where bunches of electrons pass through a magnetic structure that makes them wiggle. That undulation makes the electrons emit photons, and that radiation interacts with the electrons. The result is bright coherent light — in other words a laser.

One twist xLight uses in its approach is that it splits the electrons into separate beams, with these then generating the light. This technique avoids the problem of splitting the light, difficult because the optical element that does the splitting experiences a lot of heat. The splitting is necessary because xLight’s sources will be expensive, with estimates putting them about as costly as a high-NA EUV tool. By splitting the light, one source can power multiple EUV machines, spreading the cost around.

Being able to adjust polarization, wavelength, and power would give lithographers more knobs to tweak. The result could be higher throughput, finer feature patterning, some combination of these, or another benefit.

Anderson noted that the light source isn’t yet available. The company plans to achieve first light, a demonstration system, in 2027.

If that happens, then the new light source could help enhance the rollout of hyper NA EUV, or EUV systems with a numerical aperture of 0.7 or higher. Such tools might be necessary as feature sizes shrink.

“The ability to deliver polarized light is really going to help here,” Anderson said of the hyper NA era.

Hank Hogan is a science writer based in Reno, Nevada.

Enjoy this article?
Get similar news in your inbox
Get more stories from SPIE
Recent News
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research