The unglamorous side of EUV is housekeeping. Each laser pulse on a tin droplet produces the 13.5 nm light you want and a spray of tin debris you do not. That debris drifts toward the optics, and the EUV collector mirror is among the most expensive, hardest-to-replace components in the entire fab.
TSMC's grant US11297710B2 (issued April 5, 2022; TAIWAN SEMICONDUCTOR MANUFACTURING CO., LTD.; CPC H05G 2/008 EUV-source debris/plasma handling, G03F 7/70033 EUV illumination, G03F 7/70916 contamination control) claims a heated tin-vane bucket with a heated cover - a structure to catch stray tin before it reaches the optics. The H05G 2/008 code is precisely about source debris management.
Why heated? Tin solidifies as it cools, and solid tin on the wrong surface is a maintenance nightmare. Keeping the catch structure hot keeps the tin molten so it flows and collects in the bucket instead of plating onto sensitive surfaces - a small thermal trick with large uptime consequences.
Uptime is the hidden EUV economics. A scanner that has to stop for optics cleaning or collector replacement is a scanner not printing wafers, and at EUV capital costs, downtime is brutally expensive. Debris management like this directly protects throughput - the same throughput levers seen across TSMC's EUV portfolio.
It is again notable that the foundry, not just the equipment vendor, is filing here. TSMC running so many EUV tools has every incentive to patent its own debris-handling improvements, because the marginal value of keeping each scanner clean and running is enormous at its scale.
For the equipment-and-materials reader, the lesson is that the EUV chokepoint is not only optics and source brightness - it is also keeping vaporized metal off the mirrors. TSMC's 2022 tin-bucket grant is a concrete claim on that mundane but mission-critical problem.