Controls reshape the TAM by squeezing whatever is narrowest, so it is worth naming every narrow piece of EUV — and the scanner is only the most visible one. A scanner exposes nothing useful without a photoresist that responds correctly to extreme-ultraviolet light, and it places nothing accurately without a stage that moves the wafer to nanometer precision. Both are specialized, and both are patent-dense.
Start with the chemistry. A photoresist is the light-sensitive film that records the pattern; at EUV wavelengths and the tiny doses involved, the resist has to react cleanly to very few, very energetic photons without blurring. TSMC's grant US12645145B2, "Additive for lithography" (issued June 2, 2026; TAIWAN SEMICONDUCTOR MANUFACTURING COMPANY, LTD.; CPC G03F7/38 and G03F7/0042 resist classes), is exactly this kind of materials work — an additive that tunes how the resist behaves. The fact that the leading foundry, not only the chemical suppliers, is filing on resist additives tells you the formulation is close enough to the yield to be strategic.
Now the motion. EUV overlay budgets are measured in fractions of a nanometer, which means the wafer stage has to know where it is to that accuracy, constantly, while moving fast. ASML's grant US12645069B2, "High accuracy temperature-compensated piezoresistive position sensing system" (issued June 2, 2026; ASML Netherlands B.V.; CPC G03F7/70116 and G03F7/70291 lithography positioning classes), is about sensing position with that precision — and crucially, compensating for temperature, because thermal drift at this scale is enormous relative to the budget.
The pattern across both grants is the same lesson the optics piece teaches: EUV is a system of narrow dependencies, not one machine. The scanner gets the headlines and the export-control attention, but a fab that somehow obtained a scanner without the matching resist chemistry and stage metrology would still print nothing economically. Each subsystem is its own small supply chain.
This is why "can country X build its own EUV?" is the wrong question. The right question is whether it can build the scanner and the optics and the resist and the nanometer stage — each of which is dominated by one or a few suppliers. The system-level concentration is what makes the chokepoint durable and what gives controls their bite. Upstream is where the leverage is, and there are several upstreams.
So credit the resist and the stage. The 2026 record from TSMC and ASML shows active, recent IP on both — the chemistry that records the pattern and the metrology that places it. EUV is a relay, and every leg is run by a near-monopoly. Name them all and the chokepoint stops looking like a single box and starts looking like what it is: an entire narrow ecosystem.