Luuk Berkelaar and Marc Geers, ACCESS

How are particles detached from silicon wafers? How do they propagate in low vacuum systems? And how can they be removed using electromagnetic fields? These are the three topics addressed by ACCESS (Active Contamination Control for Equipment and SubstrateS), a project in which TU Eindhoven and VDL-ETG combine their expertise to tackle particle contamination.

For low vacuum equipment such as lithography machines and electron microscopes, contamination by miniscule particles can cause enormous disruptions. In the semiconductor industry, the handling of silicon wafers is a key source of such contaminants, as TU/e Professor Marc Geers explains. “In the Mechanics of Materials research group, we have a long tradition of working on damage and fracture mechanics. This project is a niche in that area and touches on nano-tribology, which is an established discipline on the wear of materials upon contact. This case is special because we’re talking about minute contact points that remove nano-sized particles from silicon.”

As one of ACCESS’s academic leaders, Marc’s research team focuses on particle formulation. This requires experiments to determine parameters as well as computational tools to turn these into predictions. “What is particularly challenging is that you need to span the whole range from the perfect continuum – the silicon wafer – to a fully-isolated discrete particle. Additionally, experiments are not trivial at this scale: the set-ups, forces and contact area are too large. For that reason, we’re looking for a dedicated set-up with VDL.”

“For us, it’s important to stop particles from traveling where they can damage the process,” says Luuk Berkelaar. As a group leader within VDL’s technology and development, he works for customers in both the semiconductor and analytical worlds, each with high demands on cleanliness. He therefore hopes to translate research from the university into concrete solutions for industry. “The project’s third topic is basically about creating a shield using plasma. We use a lot of robots with joints that generate wear particles, which we want to keep in the same place rather than going outside the arm. Different seals are in operation but are very expensive. We think we can do it smarter by applying plasma, creating a charge on the particles and using the opposite charge to catch them.”

Having started the project in 2019, this collaboration is set to last four years. For the university, it serves as an opportunity to connect with industrial partners that go beyond typical engineering questions, digging into fundamentals for which standard tools are insufficient. VDL is one such partner: in return for investments such as the ‘nano-scratcher’ (Bruker Hysitron PI 88 SEM PicoIndenter, the most advanced in-situ testing instrument for quantitative nanomechanical characterization), they gain access to state-of-the-art fundamental knowledge.

For Luuk, another prime motivation is simply the enjoyment factor. “In my group, we have a lot of ex-PhD students of Physics. Their work is more interesting if they can collaborate with the university and go deeper than our department normally does. It’s very worthwhile to make their jobs more interesting and give the opportunity to develop themselves.” 

Marc is also keen to emphasize the role of High Tech Systems Center (HTSC), as collaboration within the university is another core component for the project’s success. “This goes beyond one academic group; there are several groups working in the team. HTSC has played a pivotal role in orchestrating this by setting up a collaboration beyond a single topic or partner and finding a joint collaboration between industry and several academic groups, which have been selected on the basis of their expertise.”

None of this would be possible without the hard work of students. For each topic, one PDEng is responsible for conceptual design while PhD students Sven Sperling, Judith van Huijstee and Ralf Reinartz take care of the more in-depth research. Additionally, a number of master’s students are working on theses related to the project’s sub-problems. “It’s always more interesting for students if they are involved in a collaborative effort,” Marc notes.

“The real benefit is that they see the broad picture. There are also areas that touch on one another. For example, we expect to say something about the initial velocity of particles once they are removed through wear. This serves as an initial condition for the next PhD student who has to deal with the transport of particles in particular flow regimes.”

“I recognize that very well,” adds Luuk. “In ACCESS, we have internal progress meetings every three weeks with the PhDs and PDEngs plus three or four master’s students who work at our side. They discuss problems encountered and we all try to help them. For example, we had a student who was already doing his MSc thesis on charging particles and trapping them elsewhere. One of the PhD students was busy with a theoretical set-up for this, so they could have an open discussion to help each other draw the right conclusions. As Marc says, this is a nice benefit for everyone. The only challenges are corona-related as students cannot travel between the university and our company – but that’s true for everyone at the moment.”

With a supplementary proposal having been granted for an additional three PDEngs, both parties are satisfied with the project’s balance and direction. “The financial support comes from VDL but they lean on our expertise to operate the experimental tools. You can’t just put that on a desk in your office! It’s far more complex,” concludes Marc. “The prime outcome is that the young people carrying out PhD and PDEng projects flow into the company later – a real transfer of knowledge. In this way, we can offer something to both academia and industry.”

As for the future, Luuk is optimistic about what ACCESS will achieve. “At the end, I hope we have a model with which we can predict particle generation based on the type of contact (material combination, surface condition, contact shape) between the wafer and pins/gripper. This will enable us to optimize our designs regarding particle generation. If we can better understand the way in which plasma charges the particles, we can design a capturing system to keep them inside the contaminated area. That’s where we want to go – or sometimes have to go. It’s the start of what I hope is a long-term collaboration with the university at this level.”