Good to hear from you Wally. And it looks like you are up to your old tricks: asking probing questions<GGGG>. But now for the surprise. I believe I can keep this response to under 10,000 words<GGG>.
Your initial assessment is correct relative to steppers. Unfortunately lithography tools can also include the resist application and develop tools. I want to touch on this for a few moments with an analogy. Hopefully everyone's stove has two sizes of heating elements and each household has a variety (in diameter) of fry pans. If you wanted to make the perfect pancake, there are a number of things you need to control so it comes out unfiorm all the time. A 4" pancake will proabably vcome out okay on either stove heating element with any of the fry pans. As the diameter of pancake increases you run into challenges in order to make the perfect pancake. A larger heating element that can control temperature uniformly across the entire diameter of the heating element surface. You also need larger skillets that can also uniformly cook the larger pancake. You may also need better equipment to pour the batter into the skillets to get unfirm thickness and consistency of the product as it goes on the ever larger elements and skillets. This is the major issue when it comes to applying and developing the light sensitive materials on larger diameter wafers. this is also the case when you are diffusing, oxidizing, depositing, etching or sputtering films on larger diameter wafers. Just increasing the size of the unit does not necessarily guarantee process capability and uniformity. These are engineering challenges to overcome.
As you so clearly pointed out, steppers are not a real engineering challenge. The field size is much smaller than the diameter of a wafer and multiple exposures take place per wafer. Therefore the area to control is much more discrete. The challenges occur as follows:
1. If you want to increase field size you need to design and create new lens systems
2. If you want to change the reduction ration of the stepper (5x to 4X) you pretty much have to go back to the optics drawing board.
3. If you want to do step and scan to increase throughput of system on a per wafer basis and maybe on a per sq-cm of silicon basis, you need another type of innovation like what the SVGL step and scan system does.
4. IF you desire to improve the resolution capabilties of the system for next generation feature sizes.
The above 4 items are technological engineering challenges. However, a standard system that is upgraded to 300mm wafer diameter has "minor rennovations" required. The stepper motors need to extend in all directions the extra 100mm of required travel. This may mean beefing up the wafer chuck and having a uniform flatness across the chuck. the entire wafer stage would probably need to be beefed up. None of these are real technological challenges. To me, it is more sledgehammer type work. Because you are exposing smaller discrete areas on a wafer, any issues with the chuck or stage travel tilt COULD be compensated for by site by site alignment, levelling, or focus. I do not mean to say it would be a walk in the park but it does not require the invention of a new type of wheel. Let's just say it is a much more minor challenge.
There are other issues that you encounter that would require engineering improvements but I do think Canon's PR is much ado about a minor situation. Why build a 300mm machine when there are no 300mm fabs or pilot lines. There is no real big rush. Canon now has it but mark my words, there will be few issues with its system and it will be vewry easy for its competitors to jump on the band wagon. Yes, Canon may be first but there is a whole lot of equipment and process issues that need to be overcome with inventions, complete re-designs, etal to allow 300mm processing to go into full scale production. Much more major hurdles need to be overcome outside of lithography.
JMO but very little development is required relative to the non litho tools. The biggest bang for the buck in litho is improved throughput and resolution.
Andrew |
