ZDNet Korea reported on August 11 that at NGL 2026, held in Suwon, South Korea, Park Chang-min, a master at Samsung's semiconductor research division, explained that the company intends to deploy 0.55 NA High-NA EUV in mass production starting with the 1nm-class A10 node, expected around 2030.
What this report conveys is not whether Samsung will use EUV, but which node Samsung has explained it will incorporate the High-NA exposure method into mass production processes. Samsung has stated it wants to apply High-NA to 2nm and 1.4nm as well, but according to the report, technical gaps still need to be addressed at this stage. There is a difference between the stage of advancing research and process evaluation and the stage of establishing a process at a mass-production node.
In 2024, Samsung set a target of mass-producing SF1.4 in 2027. In its Q1 2026 earnings call, the company also stated that 1.4nm development was proceeding as planned. Neither of these official statements mentions the timing for High-NA's introduction into mass production. This latest report should be treated not as an official announcement changing the existing 1.4nm development target, but as newly reported information about the point at which High-NA would enter mass production.
What the Report on Starting from the 1nm Generation Means
NGL 2026 was held from August 10 to 11. According to ZDNet Korea, Park indicated Samsung's intention to use High-NA EUV for 2nm and 1.4nm while explaining that it is a technology that will become necessary from A10 and beyond. He said Samsung is pursuing joint development with partner companies.
The "introduction" of High-NA is not a single event. Research-stage exposure, process evaluation including materials and masks, and qualification of layers for mass-production use are each separate tasks. What this statement indicates is not that Samsung has abandoned research or evaluation, but that it is placing the mass-production introduction point at A10.
For this reason, the "1nm-class" designation cannot be equated with actual transistor dimensions. It has not been disclosed which interconnect layers or patterns will be exposed using High-NA, how many units of equipment will be installed, or which fab will use them. Mass-production introduction at A10 represents a point on the roadmap, not qualification for mass production today.
Multi-Patterning with 0.33 NA Bridges the Gap
Samsung's immediate mainstay is multi-patterning using 0.33 NA EUV. ZDNet Korea reported that this method will remain central from 1.4nm through the 1nm generation, after which High-NA is intended to become the primary means. This represents a path of extending the existing EUV infrastructure while transitioning to exposure tools with higher numerical aperture.
In multi-patterning, critical layers requiring finer features are exposed across multiple passes. By constructing what cannot be achieved through single exposure using a combination of processes, near-term equipment transitions can be minimized. On the other hand, this increases the number of exposure passes and process interdependencies. In mass production, each process step must be stacked and stabilized, including overlay accuracy.
ASML has also explained that the value of transitioning to High-NA lies in its potential to replace multi-patterning with single exposure. Reducing the number of process steps could help control cycle time and defects. ASML states this can also reduce cost, energy consumption, and use of chemicals and water. However, this represents general advantages of the exposure platform, not figures demonstrating Samsung's actual yield or cost outcomes.
For Samsung, this also represents a period for perfecting the next exposure method while using its fleet of 0.33 NA tools. While accelerating the switch to High-NA could reduce process steps, if mass-production conditions are not in place, it could destabilize the entire process. Choosing multi-patterning is not a return to older-generation technology—it is a choice to buy time for advancing leading-edge nodes using current EUV capabilities.
That said, a shift to High-NA does not necessarily mean existing 0.33 NA EUV tools will be replaced all at once. Samsung has not disclosed which layers of A10 will use High-NA, nor how usage will be differentiated by layer. The actual process configuration—including the possibility of reducing process steps for layers that can shift to single exposure while continuing to use conventional methods elsewhere—will not become clear until mass-production qualification.
Where 0.55 NA Delivers Value in Mass Production
ASML's High-NA tool, the "TWINSCAN EXE:5200B," raises numerical aperture from 0.33 in the current NXE series to 0.55. ASML states it offers 8nm resolution, enabling features 1.7 times smaller in a single exposure compared to NXE, and can increase transistor density by up to 2.9 times.
Looking at the numbers alone, High-NA appears to be a shortcut to next-generation nodes. However, the optical performance of an exposure tool does not directly translate into fab-level results. Behind Samsung's mention of technical gaps lies the work of aligning materials—including masks, pellicles, and resists—along with metrology, overlay, and process integration under mass-production conditions. This is not something that can be judged by equipment installation counts alone.
ASML positions High-NA EUV as a technology aimed at future logic and memory production. Even so, what matters for Samsung's mass-production process is not High-NA's raw performance itself, but how far it can qualify layers for transition to single exposure. Even if process steps can be reduced for a given layer, unless surrounding conditions are aligned, this will not translate into mass production of actual products. The content designating A10 as the mass-production introduction point is presented as Samsung's explanation as reported by ZDNet Korea.
ASML has tied its EXE platform to mass-production readiness in 2025–2026, outlining a vision of expanding from 2nm-class logic to memory with comparable density. In other words, the timing at which equipment makers prepare mass-production-ready models does not align with when Samsung qualifies its own specific layers for mass production. The figure of "around 2030" reported by ZDNet Korea does not refer to the emergence of High-NA itself, but rather to Samsung's A10 mass-production introduction point as indicated in that same report.
This is precisely why Samsung's continued use of Low-NA multi-patterning makes sense. While pursuing High-NA's advantages through future process simplification, the near-term nodes are being established by deepening operation of existing EUV. The generational shift in exposure technology and the mass-production technology needed to actually ship products do not advance at the same pace.
Official Targets Versus the Reported Mass-Production Introduction Point
The mass-production target for SF1.4 that Samsung presented at SFF 2024 was 2027. In its subsequent Q1 2026 earnings call, the company stated that 1.4nm development remained on schedule and also touched on expanding its major 2nm customer base. Within the scope of official announcements, it had not been disclosed at which node High-NA would be introduced into mass production.
The statement at NGL 2026 in August filled this gap by providing a sequence for mass production. ZDNet Korea also referenced the current roadmap placing SF1.4 around 2029 and SF1.4+ around 2030. These are not new official Samsung press releases, but reports based on on-site coverage.
From the 2024 official target and the schedule reported this time alone, it is not possible to immediately judge the success or failure of development. Samsung's official information indicates continuity of 1.4nm development, while the timing of High-NA's mass-production introduction remains undisclosed. The next point of focus is whether Samsung will officially disclose both the timing of A10 mass production and the scope of High-NA qualification.
If Samsung is to genuinely deploy High-NA at A10, it will need to align not just the exposure tool itself but also the surrounding processes. What will determine the outcome is not the smallness of the node name, but whether qualified layers can be repeatedly processed in mass production. In evaluating the reported A10 introduction around 2030, this scope and timing of mass-production qualification will serve as the key criteria.