Upcoming Challenges And Changes In Semiconductor Materials

Semiconductor Engineering sat down with Dan Brewer and Srikanth Kommu, co-CEOs at Brewer Science, to talk about current and future changes in materials used in semiconductor manufacturing and adjacent markets. What follows are excerpts of that conversation.

SE: What was behind the decision to have co-CEOs instead of just one?

Brewer: We see a lot of value to having multiple perspectives behind the leadership of the company and doing that in a collaborative fashion. It really is a true collaborative co-CEO effort, where we each bring in our areas of expertise and perspectives, working together to lead the company.

Kommu: It’s also a succession plan for Terry Brewer. He established the company over 40 years ago. He’s still actively involved, and he’s always going to be the go-to person for us.

SE: How are materials changing to extend deep UV lithography, and how are the materials different for EUV lithography?

Kommu: ‘More than Moore’ is continuing, driving the demand for higher quality and better performance. This paradigm of high performance and high quality continues, whether you want to extend the life of deep UV or the ArF litho space, or even EUV. Getting to the next level of performance and quality requires sustained investments in people, capabilities, and innovation, either by tweaking existing chemistries or through the development of new material platforms.

SE: Brewer Science was putting a lot of effort into flexible substrates in the past. Where is that technology now?

Brewer: We still have a program associated with sensor devices in printed electronics, and we see that as a very significant growth opportunity for the company. It takes us into a new market space with a whole new set of products than we’ve had in the past, and we see it as a very large growth opportunity.

SE: For which markets?

Brewer: We are really excited by our water quality sensors that are being developed to monitor impurity levels in water — especially industrial water. Companies want to monitor their effluent water quality to ensure compliance with EPA or city or state-local regulations. We’re also getting some interest from companies with water filtration systems to detect how well that water filtration system is working and when the filter needs to be replaced.

SE: Some of this is aimed at legacy nodes. What else is happening there?

Kommu: Our customers are demanding high quality materials with a stable and secure supply chain. They’re adding capacity, they’re onshoring, and they’re expanding factories, which is great. And that’s where POR (process of record) positioning is so important. And we have POR wins in the litho space, so when they add capacity, you grow with that. And then, for the advanced nodes, you obviously need to improve performance along with quality. There is also a lot happening with sacrificial and permanent materials. Especially with the boom in advanced packaging, there is an incredible need for innovative materials that are cost-effective and achieve certain performance benefits. In die thinning, for example, stress management becomes a critical performance attribute. If you have a polymeric material, you can manage these stresses better. Similarly, if you can take out an oxide and replace it with a polymeric material, you can get better insulation, better thermal properties, and better flexibility. I’m simplifying that, of course, but there are a lot of opportunities for newer materials that can be permanent. Now, these materials need to have the same or improved reliability as existing deposition types of materials.

SE: A related challenge is getting rid of the material when you don’t want it there, right?

Kommu: Yes. There are lot of possibilities, and that’s what is exciting in terms of creating new materials, both on the sacrificial side and on the permanent side.

SE: For hybrid bonding, your materials for bond/debond are pretty standard in the industry. Are those going to higher and/or lower temperatures?

Kommu: In general, they’re going to a lower temperature, but if you go into the specifics, it’s a combination of both. With hybrid bonding, stress management is very critical, and that’s where temperature management comes in, and the precise engineering and design of material properties such as modulus comes in. The expectation for sacrificial bonding materials is that the adhesive can be easily applied in one step, resulting in a defect-free, flat film that covers any bumps without planarizing. The adhesive needs to be easily removed when you don’t want it because that affects the cycle time of the process. So how do I ensure my quality is good and my cost of ownership is low in terms of throughput and cycle times and removal of materials? That’s where innovation comes in, because there are only certain materials and processes that can meet all these boundary conditions.

SE: When you’re developing a new material, how long does that take? And if you’re dealing with the next generation of devices, the next node, will they stay with the same chemistry?

Kommu: With ‘More than Moore,’ it’s typically within a certain platform network. You don’t want to make fundamental changes, because the EUV tools are expensive, so people want minimal chemistry changes. You don’t have to qualify new chemistries, but you do want to optimize them. It’s different on the packaging side. That’s still evolving, and there are opportunities for several new platforms. Also, back-end qualifications happen at different time scales compared to the front end. The flip side is that material development can take a long time, especially with new chemistries. For a new entrant, that could be 10 or 20 years. We have a database of material platforms, which allows us to come up with new platform in time spans of one to three years. Of course, the customer still needs to qualify and check the reliability, especially with permanent materials. That takes time. But customers are willing to work with us to expedite those or to pull in those timelines.

SE: Have you been developing a lot of materials that have not made it into production yet, but you think will be in production in the future?

Kommu: Yes. For example, in advanced packaging there are many possibilities, and also many needs, and even the customers don’t know exactly which materials will work. We see a lot of materials in our portfolio that will be productized and commercialized down the road.

SE: What do you see as your biggest growth areas?

Kommu: It’s materials overall. We have expertise in lithography and in packaging, of course, but we’re also seeing growth in cleaning chemistries and high purity bulk chemicals. Similarly, we see opportunities in environmental monitoring that involve our sensors. There also could be opportunities in other industries, such as pharmaceuticals, where they require high quality and where AI is providing new ideas.

Brewer: In any area where there is a lot of change to a process, or a new process is required, there is a lot of materials growth potential. That could be advanced packaging or sensors. New materials often are needed to support or enable those processes and good outcomes.

SE: People always select their tool before they select their material, which is limiting for the industry as a whole. Do you see that changing?

Kommu: That is not changing because the industry still goes by the spend. Equipment requires a lot of capital, and I don’t know if that will ever change because the customer base is shrinking. There are only a few big players, and how you manage your costs is very important. As these tools need specialized material chemistries to deliver the performance and cost of ownership expectations of the device makers, there is increased recognition of the importance of the materials supply chain across the industry.

Brewer: For the equipment vendors, there’s a recognition of the growing importance of materials for enabling their customers’ processes to be successful with that equipment. So equipment vendors are more open to recognizing the role materials are playing, in some cases even more than the customers recognize.

SE: As we get down to the most the advanced nodes, and also for safety- and mission-critical types of applications, there has been more of a focus on the purity of the materials. Does the supply chain conform to that, or are there still suppliers outside the semiconductor industry who don’t have the same concerns for quality?

Kommu: The quality is getting better, but the quality requirements are getting more and more stringent, especially on the front end. On the packaging side, traditionally they were not as strict, but there is a convergence happening. The quality requirements are getting tighter, and the people who understand the front end have an advantage. Across the supply chain, a few companies manage their supplier very, very well. Is everyone doing that? I don’t know. But we are, and that is making a difference.

Brewer: You have to partner with other suppliers to be good at supply chain management. This is a strength that we bring to the table, and we definitely have seen positive outcomes for the customer as a result of our ability to better manage our supply chain.

SE: How do geopolitics affect this?

Brewer: Geopolitics is absolutely playing a part. We are somewhat unique as a supplier here in the U.S. We are a U.S. company making U.S. products, and there are not many of those around — and certainly not in our field. But from an incentives point of view, I wish we had more.

SE: So is there any material you think is going to be in short supply due to geopolitics, and are there alternatives?

Kommu: I don’t want to name any specific material, but that’s a constant concern. Even our customers are paranoid. On top of that, there is an emphasis on PFAS-free (polyfluoroalkyl substances-free) materials. That’s where innovation comes in. We are actively involved in ensuring the supply of critical materials but also making sure they are PFAS-free.

Brewer: It goes back to process innovation and the diversity of materials required, especially in packaging and other processes. That drives material development, material diversity, and different material needs. It’s very situational.

SE: Were any of your materials hard to get during the pandemic?

Brewer: With the pandemic, supply chain management became critical to business success.

We did not delay any of our shipments due to a lack of materials. There absolutely were materials — sometimes bottles, sometimes filters, all kinds of stuff — that were extraordinarily difficult to get during the pandemic, but we were able to navigate that without any impact to our customers whatsoever.

SE: Was that just a matter of multi-sourcing everything?

Kommu: Multi-sourcing was one way. Another was better planning and inventory management. But if you have too much inventory it adds cost, so that’s a balancing act.

SE: Looking forward, what do you see as your next big opportunities?

Kommu: Within semiconductors, the industry has become very prescriptive, particularly on the front end. A customer will tell you, ‘These are your specifications.’ You have to meet them, and there is not enough room to change the game. We follow those, and we just have to make sure we’re better than the competition in quality, because there are only a few big customers. Packaging is an area where there are new opportunities even the customers don’t know about. They know at a high level what could be needed to solve a problem, but they don’t know exactly what that is. On the sensor side there is an incredible opportunity in water quality monitoring, and we have been active in that area, investing quite a bit for the last several years.

SE: Is that adjacencies, or completely different markets?

Brewer: Some of both. As a company, we’re driven by the technology for the customer. So where do customers highly value that technology innovation? Where is it really important to what the customer is trying to do for their business? The two that are obvious to us right now are the packaging and sensor areas. As the need for performance grows in both of those, that drives new materials needs. But there may be others that are completely different market spaces.

Related Reading
2D Semiconductors Make Progress, But So Does Silicon
Can 2D materials be fabricated consistently at a cost that competes with silicon?
New Materials Are in High Demand
Development methodologies combine old and new techniques, but getting any new material into high-volume manufacturing is a complex process.