IN Brief:
- Xanadu and ASML will develop lithography processes aimed at reducing optical loss in photonic quantum hardware.
- The work focuses on line-edge roughness and its relationship with patterning quality and photon propagation.
- Xanadu places fabrication loss alongside packaging and error correction in its fault-tolerant hardware roadmap.
Xanadu and ASML have begun collaborating on lithography processes for photonic quantum hardware, examining whether tighter patterning control can reduce optical losses in the structures used to transport quantum information as light.
The programme brings a photonic quantum-computing developer together with the semiconductor industry’s largest lithography-equipment supplier. The objective is not a conventional logic-node shrink, but improved control of photonic structures where small variations in fabricated waveguides can affect photon propagation across a circuit.
Line-edge roughness is a particular focus. Lithography and subsequent process steps determine how accurately patterned features are transferred, and microscopic variations along a waveguide edge can scatter light as photons travel through the device. In a quantum processor, that loss contributes directly to the error burden the architecture has to manage.
Xanadu identifies optical loss as one of the central constraints on fault-tolerant photonic computing. Its current technology roadmap targets a reduction in a key loss indicator from 24.1x in 2026 to 1.0x by 2030, alongside a fault-tolerance milestone in 2028–2029 and an objective of more than 1,000 logical qubits by 2031.
Those remain company targets rather than demonstrated outcomes, but they give the ASML collaboration a measurable engineering context. Improvements in lithography would have to translate into repeatably lower optical loss before they could contribute materially to the wider fault-tolerance programme.
Dr Christian Weedbrook, Founder and Chief Executive Officer of Xanadu, said: “Fault tolerance and error correction are arguably the most important technical objectives for the quantum computing industry.” The work with ASML will examine how lithography conditions and patterning quality influence the photonic structures on which Xanadu’s architecture depends.
Manufacturing variation becomes increasingly important as photonic quantum hardware grows beyond laboratory-scale devices. A small circuit can demonstrate a useful optical structure without proving that the same geometry can be reproduced across larger dies, complete wafers, multiple production lots, and increasingly complex assemblies.
Once a processor relies on large numbers of optical paths and devices operating within narrow tolerances, fabrication becomes part of system architecture. Loss in a single waveguide may be manageable; accumulated loss across sources, switches, couplers, detectors, and interconnects can become a system-level constraint.
ASML’s role is therefore centred on process control rather than simply applying the smallest available feature size. Photonic integrated circuits impose different requirements from mainstream logic and memory, particularly around sidewall quality, geometry, optical coupling, and propagation loss. Quantum photonics raises the penalty further because the loss of individual photons can affect computation directly.
The companies have not disclosed which ASML lithography platform will be used, the process geometry, manufacturing location, or a qualification timetable. They have also made no claim that the collaboration has already produced a lower-loss quantum chip, leaving the programme at the process-development stage.
Xanadu is pursuing the lithography work alongside materials, packaging, system architecture, and quantum error correction. Its manufacturing roadmap also includes a 158,000-square-foot Toronto facility intended for testing, heterogeneous integration, photonic integrated-circuit packaging, and rack-level module assembly.
Improved lithography cannot remove losses introduced elsewhere in the system, nor solve constraints around photon sources, detectors, packaging, control electronics, and error correction. Its contribution can, however, be measured directly. Evidence that tighter patterning reduces optical loss consistently across repeated devices would turn lithography from a general manufacturing requirement into a quantified element of Xanadu’s fault-tolerance roadmap.
