TSMC Arizona Achieves Yield Parity with Taiwanese Facilities, Production Remains on Schedule

[AleksandarK]

TSMC has reportedly managed to produce yields at its Arizona facility that are on par with yields back home in Taiwan, making its expansion efforts successful. According to Bloomberg, TSMC did a trial production, a multi-month effort, to produce N4 node wafers with low defect rates. With wafers now in TSMC's labs for testing, it is reported that Arizona facility yields have achieved parity with their Taiwanese facilities back home. This indicates that TSMC's efforts to expand in the US are so far considered a success, as advanced chipmaking is a very complex process that is only done by a few makers and in very few locations. With TSMC expanding in the US now and proving that its technology can work on US soil, the company has a green light to start volume production in the first half of 2025.

However, this is only the beginning of TSMC's Arizona expansion. The Taiwanese giant plans to have a second fab operational by 2028 and produce 2 nm and 3 nm chips in the state. Additionally, there will be a third facility for 2 nm and more advanced nodes in Phoenix, bringing the total value of TSMC's US expansion efforts to $65 billion, with $6.6 billion from the CHIPS Act grants and $5 billion in loans from the US government. If upcoming fabs follow the lead of the first facility, US-based production needs will possibly be satisfied.



View at TechPowerUp Main Site | Source

 

[Daven]

With TSMC expanding so quickly and IFS halting most of its expansion plans, TSMC is going to steam roll right over Intel.

 

[GenericUsername2001]

With TSMC expanding so quickly and IFS halting most of its expansion plans, TSMC is going to steam roll right over Intel.

Very true; Intel's one 'advantage' in the fab building race is the perception that the US government will be willing to subsidize them heavily so that the US has advanced fabs; but with TSMC showing that it can take federal money & get an actual useful fab working on US soil in a reasonable time frame, well, that makes giving Intel money all the more questionable.

 

[Wirko]

Guess there's enough sand in a half-mile radius to keep making chips forever...

 

[dir_d]

Now we just need to know where packaging is going to be. Double capacity is cool but the bottle next of packaging is still there. I bet that is still a concern of the US Government as well.

 

[Minus Infinity]

Guess there's enough sand in a half-mile radius to keep making chips forever...

Nope, the type of silica that's required for chips is not found in 99.99% of sand. Only one place currently produces the right sand and it's running out. We are in trouble if no new source is found.

 

[Wirko]

Nope, the type of silica that's required for chips is not found in 99.99% of sand. Only one place currently produces the right sand and it's running out. We are in trouble if no new source is found.

Yes I'm aware it's not that simple, and also that rock on the Earth's surface is mostly sedimentary rock, which doesn't contain silicon. Still, as is the case with many other natural resources, we're not "running out" of silicon. We're just running out of silicon that's cheap to mine and process.

 

[Minus Infinity]

Yes I'm aware it's not that simple, and also that rock on the Earth's surface is mostly sedimentary rock, which doesn't contain silicon. Still, as is the case with many other natural resources, we're not "running out" of silicon. We're just running out of silicon that's cheap to mine and process.

Yeah and the costs for processing normal Silica into ultra-pure silicon is huge. It's a complex process. I was getting at the fact there is a quarry in North Carolina that already has very pure quartz and that's the one the chip indsutry turns to, but it's not going to keep going a lot longer. So sure whe can find silica/quartz anywhere but costs of chips will increase a lot if we have to do rcomplex refining. Solar cells need
99.999999% pure silicon and semiconductors 99.999999999% pure silicon.

 

[Wirko]

Yeah and the costs for processing normal Silica into ultra-pure silicon is huge. It's a complex process. I was getting at the fact there is a quarry in North Carolina that already has very pure quartz and that's the one the chip indsutry turns to, but it's not going to keep going a lot longer. So sure whe can find silica/quartz anywhere but costs of chips will increase a lot if we have to do rcomplex refining. Solar cells need
99.999999% pure silicon and semiconductors 99.999999999% pure silicon.

I'm trying to find some info but all data seems quite unreliable, especially on the price.

Price: up to $5000 for a 300mm wafer

Purity: you say 11N (eleven nines) but I've seen 9N mentioned in several places, the only mention of 11N being here but that's for polycrystalline Si, which is supposed to be used for semiconductors - huh?

Sources of silica: int's true what you say but there are usable deposits in several other places in the US, Australia, Canada, Russia and elsewhere. China will turn to Russia of course if and when they need more. A lot of info here.

Scale: Spruce Pine Mine produces 10 million tons of silica (SiO2) per year. That yields 4.7 million tons of silicon. A wafer weighs 0.128 kg. The mining capacity is thus 37 billion wafers yearly. The semiconductor industry's capacity is predicted to reach 115 million wafers yearly in 2026, or 15 thousand metric tons, which amounts to 0.3% of silicon production. I here assumed that most of the silicon wasted in the process (cutting, grinding) can be recovered. Sure, the greatest part goes into solar, but solar can use lower purity deposits too - actually it has to because I can buy solar panels full of wafers for ~100 € per square meter, and it would be stupid to use almost all of the precious pure ore for that.

All in all, unless I've missed something gross, the semiconductor industry can't possibly run out of wafer supply at current manufacturing costs (which doesn't mean a stable market price).