SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?


To: pgerassi who wrote (252996)6/7/2008 5:43:12 PM
From: misenRead Replies (1) | Respond to of 275872
 
I do know of one production semiconductor process that uses 200+ scans right off the top of my head, that of making VCSELs. They need a 30-60 pairs of alternating materials to do the mirrors at the top. Each layer takes two scans. Granted it is a special structure, but the results are well worth it for very high speed high data rate comm (>1TB/s). And yes they are SESP at a relatively wide process. They usually have the drive logic adjacent to the VCSEL. Long term they are looking at it to do interchip comm.

First, if you are correct about 2 mask layers per material layer, 2x60 is not anywhere near > 200+. Even with any additional masking layers needed to create interconnects or drive circuitry, you are still far below 200.

Second, I highly doubt that a VCSEL process patterns each material layer after it is deposited. It is far more likely that all the material layers are deposited together and then they are patterned as a stack (similar to the way that the oxide/poly 1/Oxynitride/Poly 2 stack was fabricated on some older EPROM processes). Do you have a link that substantiates your claim or are you just assuming that this is the case?

I notice you didn't comment about either of my points relating to your error on masks required for metal/dielectric patterning as well as the small percentage of mask layers that would use double patterning.

Misen