China Just Built a 1 Kilowatt Accelerator That Fixes ASML’s 98% Light Loss

Inside ASML's EUV machine, 98% of the light it generates never reaches the silicon. It gets created, then thrown away, bounced into oblivion by the very mirrors that are supposed to deliver it. That's not a flaw. It's the unavoidable tax of the only method anyone has ever gotten to work at commercial scale. A team at Tsinghua University looked at that number and asked the one question ASML never had a reason to ask: what if the light source never needed those mirrors at all? This video walks through steady-state microbunching, the real, peer-reviewed physics behind China's accelerator-based alternative to ASML's plasma method — published in Nature back in 2021, not invented for a thumbnail. You'll see the actual design target (it's not what the headlines say), why a shared light source could rewrite the economics of an entire chip fab, and the honest gap that's still standing in the way: China's own mirror reflectivity, sitting at roughly 65% against Zeiss's 70%, a difference that compounds across ten or twelve reflections exactly the way ASML's own loss does. No exaggeration. No dismissal. Just the physics, the real numbers, and the project's own lead scientist telling you plainly there's still a long way to go. Iron Circuit exists for exactly this kind of story. Subscribe if you want the honest version before it becomes a headline. 📌 Covered: Steady-state microbunching · Tsinghua EUV accelerator · ASML mirror efficiency · Carl Zeiss Mo/Si mirrors · China EUV research · Xiong'an test facility · EUV lithography physics #ChipWar #ASML #China #Semiconductors #IronCircuit