Skip to content
AI-Daily-Builder

2026-06-06 — views $LRCX · Lam Research · Kallisto / Phoenix electrochemical deposition (ECD) platforms · advanced packaging / panel-level packaging (back-end)

Lam Research opens Salzburg panel-level packaging hub as Rapidus eyes 600mm glass interposers

Lam Research opened a Panel-Level Packaging Center of Excellence in Salzburg, Austria on May 20, 2026, showcasing its Kallisto ECD and high-throughput Phoenix electroplating platforms — and trade press reports Japan's Rapidus is tapping Lam's panel tools to build redistribution

Lam moves the packaging fight from round wafers to square panels

On May 20, 2026, Lam Research opened a Panel-Level Packaging (PLP) Center of Excellence in Salzburg, Austria — an R&D and customer-collaboration lab built specifically around processing square and rectangular substrates instead of the 300mm round wafers that have defined back-end packaging for decades. “The expansion of our Salzburg site reflects a long-term investment in advanced packaging,” said Aaron Fellis, Corporate Vice President and General Manager of Wet Equipment Technology Systems. Salzburg’s regional governor, Karoline Edtstadler, called the opening “a testament to our standing as a high-tech location.”

The Salzburg lab is the showcase for two electrochemical-deposition (ECD) platforms that anchor Lam’s panel ambitions:

PlatformRoleSubstrate handling (per Lam product pages)
KallistoR&D / low-volume ECD; fine-line plating below 10 microns on organic and glass-core substrates300x300mm up to 1100x1300mm (Gen5.1)
PhoenixHigh-volume manufacturing ECD for bumps, pillars, RDL, TGV, FLIoptimized for 510x515mm panels

Trade press (TrendForce, Digitimes) reports the Phoenix line lifts throughput roughly fourfold to about 120 panels per hour versus the R&D-class cadence of Kallisto — Lam’s own pages list the process menu and panel formats but do not publish the panels-per-hour figures, so treat the throughput numbers as vendor-guided estimates carried by the trade press rather than spec-sheet values.

Why panels, and why now

The economic logic of panel-level packaging is area utilization. A round wafer wastes the corners; a rectangular panel tiles far more die-and-package real estate per pass, which matters enormously as AI accelerators push package sizes well beyond a single reticle and stack ever more high-bandwidth memory. Lam framed the broader opportunity on its fiscal Q3 2026 earnings call (April 22, 2026), where management said advanced-packaging revenue is “expected to exceed 50% in calendar year 2026,” driven by copper plating and through-silicon-via etch demand tied to HBM. That is an upward revision from the “over 40%” the company had guided earlier in the cycle.

The Rapidus signal

The Salzburg news landed alongside a more concrete pull-through. On May 26, 2026, TrendForce — citing EE Times Japan and Nikkei reporting — said Japan’s Rapidus is tapping Lam’s panel-level solutions for a 2.xD packaging scheme that forms redistribution layers directly on 600mm-square glass carriers. The reported roadmap pairs 2nm logic mass production in 2027 with glass-interposer mass production in 2028, with Kallisto’s support for 600mm panels and sub-10-micron plating called out as the enabling piece.

Glass as an interposer material is the under-appreciated thread here. Glass panels offer better dimensional stability, tunable thermal expansion, and large-format scalability versus organic laminate or silicon interposers — but they also introduce warpage, co-planarity, and handling headaches that wet-process tools have to solve at panel scale. That is precisely the problem set Lam is positioning Kallisto and Phoenix against, and it is why a foundry-scale customer commitment (even a reported one) matters more than another lab opening.

Practitioner note

For anyone tracking the fab-equipment back-end, the read is that the packaging ECD race is consolidating into a panel-format arms race, and Lam is planting a flag in glass. Watch three confirmations: (1) whether the Rapidus engagement converts from “reportedly” into a named tool-of-record disclosure, (2) whether Lam’s “over 50%” advanced-packaging growth holds through its next earnings update, and (3) whether competitors respond with their own large-panel ECD/plating roadmaps rather than ceding the format. The capex story has shifted — the marginal dollar of packaging spend increasingly buys square-panel wet tools, not just front-end deposition and etch.

Under-considered angle: the geography. A flagship PLP center in Salzburg, plus a reported Japanese foundry customer on glass, quietly diversifies advanced-packaging know-how away from its heavy Taiwan/Korea concentration. If panel-level packaging on glass becomes the substrate of choice for next-generation AI modules, the locus of back-end value could migrate toward whoever industrializes the wet-chemistry and handling first — and that is a region-and-supplier map that looks meaningfully different from the front-end EUV one everyone watches.

Sources

Tip