450mm Wafer Manufacturing: The Semiconductor Industry at a Crossroads

The debate over 450mm wafer manufacturing has been raging for years within the semiconductor industry. Should it happen? Will it happen? When? Who will be the first to take the plunge? Development costs and fab costs are so high that only a select few companies are even capable of making it a reality. What does that mean for the rest of the industry?

For the equipment companies, the risks are higher than ever. There are fewer potential customers for 450mm equipment, and with lower production volumes, fewer tools will be needed. How can the equipment vendors hope to recoup their development costs?

It is true that there are fewer companies maintaining fabs, but the industry continues to demand chips with more functionality, and fabs are needed to produce those chips in larger and larger volumes. Based on chip unit volumes Semico believes there will be a point when building one 450mm wafer fab will be more cost effective than two 300mm fabs.

But there are no guarantees. The 450mm transition will not come easily. How many more 300mm fabs will be built? What will the industry do in the meantime, before 450mm takes off? This report steps through the history of 300mm wafer adoption, the forecasted unit and wafer demand which will drive the need for more advanced capacity, and takes a rational look at what this industry needs to do to continue on the path of innovation.

The dynamics of this wafer transition cycle are not that unusual. The variable that makes this cycle slightly different is the notion that if the adoption of 450mm wafers is pushed out beyond 7nm, it is highly likely that the industry will move to a completely different manufacturing option. Carbon nanotubes or something to replace CMOS becomes a real potential.