Last of three parts: Building and equipping new fabs; green fabs; standards; how the transition to 450mm will likely go; modular platforms.
Semiconductor Manufacturing & Design sat down to discuss future 450mm fab and facilities challenges with Gerald Goff, director of the project management office for fab design and construction at GlobalFoundries; Joe Cestari, president of Total Facility Solutions; Ivo Raaijmakers, chief technology officer of ASM International; and Michael Brain, senior director of the Fab Solutions Business Unit at Brooks Automation. What follows are excerpts of that conversation.
SMD: How long does it take to design, build and equip a new and leading-edge fab?
Goff: The driving force is to turn an asset into a revenue generating asset as quickly as you can. From the time you put a shovel in the ground, it could be years. But from a design perspective, and to design a facility of this nature, it’s a six-month process at a minimum. Typically, it can run as much as nine months, depending on whether or not you are going to take the design all the way to completion, or whether you are going to do some type of design and development at the same time. So, you could say six to nine months for design. And then, the big debate is once you stick a shovel in the ground, how long before you can put the first tool in the fab, which we call ready for equipment. So, from ground breaking to ready for equipment is 18 to 24 months for a modern fab. If you go into some of these Asian facilities, they claim they can do it in 12 months. I question some of those metrics in straight up comparisons. In the U.S., it would be extremely aggressive to do something of this magnitude in a 12 month period.
Cestari: A rapid fab from design to finish to first silicon is still a two-year effort. The number one challenge is being able to work together to produce an accurate integrated model, meaning that our design documentation looks like what the fab looks like. Once you commit to spending $6 billion or more on a fab, you must have that facility operating efficiently and as quickly as possible. You can’t tolerate two months of idle time or an additional cost. It just breaks the entire economic model.
SMD: How do we deal with sustainability and conservation in building the next round of “green fabs”?
Goff: Everybody has different goals, based on your geographic location. We go through an analysis on the return on investment for putting in conservation programs. You also have to make a business decision. If you’re in the northeast part of the United States, for example, water is abundant. And it actually costs us more to recycle the water than it basically does to use new incoming water. If you are on an island like Taiwan, where water is expensive, they will strive towards a higher recycling methodology. The same is true for those buildings in the mid-west or desert regions in the United States. So, you have to look at things not only from an environmental perspective but also an economic perspective.
SMD: Is there a standard for becoming a green fab?
Goff: That said, we had an objective of meeting a Gold LEED standard for Fab 8 in Malta. We obtained certification for the fab, which is pretty remarkable. Those things don’t usually apply to manufacturing facilities.
SMD: The transition from 200mm to 300mm was difficult. There were many starts and stops. How do you predict the 450mm transition will go?
Goff: The simple fact is there are fewer players for 450mm, which makes that a little easier. The problem is that the few players that remain are kind of arch enemies. You continue to reduce the gene pool down to the strongest players. The strongest players will have an epic battle to the end, until it ultimately gets down to the last man standing. In 450mm, however, I really don’t see it being anywhere near as difficult as it was with 300mm. 300mm was an extreme example of having way too many cooks in the kitchen. It took a long time to fight through some of those battles. In 450mm, I think the battle will be intense and people will be leveraging what they feel will be an advantage for them during the initial discussions. But ultimately, just the fact there are fewer players, the industry should be able to push 450mm pretty quickly.
Cestari: I agree. Anytime you have fewer people playing the game, it should be easier. On our side, we have to look at a broader contingent. The rule of three doesn’t apply in the supply chain. That’s where there is a disconnect. I think it would benefit all of us if we did try and drive the rule of three on the supply chain. But on the other hand, there are a plethora of component suppliers both on the mechanical and electrical side. That makes it a challenge for us.
Raaijmakers: The consequences for a tool manufacturer are huge. If the transition does not take place for another four to six years, it means that for several technologies you will have to sustain both 300mm and 450mm developments. So that’s basically doubling, or almost doubling, the development expenses. I don’t think the equipment industry can sustain that. So I don’t think there is any choice. There has to be a coordinated transition to 450mm and we have to move as painlessly as we can.
SMD: Besides collaboration, how do we make the 450mm transition easier?
Brain: One point that didn’t come up is that I believe that any large change, whether that it be a process, material handling or automation, is going to have to be proven in 300mm first. There will have be a short generation of ‘300mm Prime,’ if we can reuse the terminology from before. We might have to do some things a little differently in 300mm to get them ready to do them in 450mm.
SMD: Tokyo Electron is proposing the idea of having an “open and modular platform” for 450mm. This would enable all fab tool vendors to develop various plug-and-play process modules for the open platform, thereby reducing costs and development times. Will that work and make the 450mm transition easier?
Raaijmakers: We have talked about it. We considered such a platform as a potential competitive differentiator. But if you go that way, the competitive differentiators go away. I think this is not good for the industry at the end.