ZEISS Optics: High-NA EUV Ready for Mass Production, Orders Booked Through 2027
nashnova research
ZEISS semiconductor optics chief Christoph Hensche says High-NA EUV is ready for volume production from an optics standpoint, with orders locked through 2026–2027; an AI-driven demand surge is forcing ZEISS to expand capacity to avoid becoming a bottleneck.
Is High-NA EUV actually ready?
ZEISS SMO president Hensche says High-NA's mirror technology, lifetime, and stability all build on Low-NA systems already running in volume production.
This means → High-NA is not a clean-sheet design — it is an upgrade on a proven platform. The technical bar is lower than the market assumes.
In plain terms = the market worries High-NA is "too new, too fragile." ZEISS's argument: the core optics have already logged real fab hours.
Orders through 2027 — where is the bottleneck?
Demand for 2026 and 2027 is locked in. Capacity at ZEISS and some other suppliers is fully booked.
An AI-driven chip demand surge created an "unusually sudden rise," stretching optical-component lead times.
This means → ZEISS is proactively expanding capacity; the industry expects growth to continue through 2028–2030. Whether ZEISS can scale on schedule is the key variable for those orders to ship.
TSMC, Intel, Samsung — who moves first?
TSMC leads in market share and customer trust. It can keep using Low-NA plus double patterning — exposing the same layer twice to print finer features — so its urgency to adopt High-NA is the lowest.
Intel is the most aggressive: it is deploying High-NA on products that could also run on Low-NA, deliberately building real production experience to close the gap with rivals.
Samsung sits between the two. This reflects a clear split among the big three on *when* the upgrade becomes unavoidable — the leader can wait; the chasers must move early.
What is Hyper-NA? How long will High-NA last?
Hensche identified Hyper-NA at a numerical aperture of 0.75 — the metric for how much light an optical system captures; higher means finer resolution — as the next technology step.
He expects High-NA to remain the mainstream advanced-lithography technology for several more years; Hyper-NA still needs to mature before entering product development.
In plain terms = the pace of semiconductor shrinks has slowed, stretching High-NA's "lead-technology window" — no replacement is imminent.
Why is ZEISS moving into advanced packaging?
ZEISS is extending from front-end lithography into process control for advanced packaging — the back-end step of stacking or tiling multiple chips together.
Its lineup includes the NLX 3D X-ray inspection system (roughly 50 measurement points per hour) and the DUNE 100 wafer-warpage correction system, the latter deployed on production lines in Taiwan since September 2026.
This means → ZEISS is no longer just selling lithography optics; it is expanding into packaging inspection — and the growth of the advanced-packaging market gives it a second revenue curve.
市场有风险，内容仅供研究参考，不构成投资建议。