Xanadu and Mitsubishi Chemical are advancing quantum simulation for semiconductor fabrication, aiming to address radiation-induced blur in extreme ultraviolet lithography with support from Canadian and Japanese research funding
Photonic quantum computing developer Xanadu and Japanese chemical manufacturer Mitsubishi Chemical have launched the second phase of their collaborative research program, focused on simulating and mitigating radiation-induced blur in extreme ultraviolet (EUV) lithography. The project is supported by binational funding from the National Research Council of Canada Industrial Research Assistance Program (NRC IRAP) and Japan's Strategic Innovation Promotion Program (SIP), with technical leadership from Japan's National Institute of Advanced Industrial Science and Technology (AIST) and the Global Research and Development Center for Business by Quantum-AI Technology (G-QuAT).
Quantum Simulation for Semiconductor Materials
The central technical challenge addressed by the partnership is the spatial resolution limit in EUV lithography, which uses 92 electronvolt (eV) photons to etch sub-nanometer features onto silicon wafers for advanced semiconductor devices. At these energies, photoresist materials experience quantum mechanical effects-such as photoabsorption, secondary electron cascades, and Auger decay-that lead to blurring of circuit patterns. Classical simulation methods struggle to capture these effects with sufficient accuracy, motivating the use of quantum algorithms to model the underlying light-matter interactions.
In the first phase of the collaboration, the teams demonstrated that quantum algorithms could simulate the optical properties and photoabsorption behavior of candidate photoresist materials. The second phase aims to integrate these quantum-derived parameters into Mitsubishi Chemical's multi-scale industrial workflow, enabling more accurate prediction of electron blur and supporting the search for new blur-resistant photoresists.
Technical Architecture and Resource Constraints
The workflow architecture combines Xanadu's fault-tolerant quantum computing (FTQC) algorithm pipeline with Mitsubishi Chemical's industrial modeling tools. Parameters calculated by quantum simulation-such as 92 eV photoabsorption cross-sections-are intended to feed directly into macroscopic models of electron transport and blur. The project targets early fault-tolerant quantum hardware, with resource estimates suggesting that relevant simulations could be performed with fewer than 500 physical or logical qubits, depending on the algorithm and error-correction overhead.
Key features of the workflow include the integration of Mitsubishi Chemical's proprietary blur models, photonic quantum processing unit (QPU) execution on Xanadu hardware, and cross-border technical support from both Canadian and Japanese research agencies. The binational funding structure is designed to accelerate industrial adoption and facilitate technology transfer between North America and East Asia.
Funding, Collaboration, and Industry Context
NRC IRAP has expanded its direct financial support for Xanadu's quantum application roadmap, building on more than $800,000 in previous advisory and funding commitments. On the Japanese side, the SIP program-under the direction of Dr. Masahiro Horibe at AIST and G-QuAT-prioritizes the deployment of quantum simulation in domestic materials manufacturing supply chains. The collaboration is led by Xanadu Founder & CEO Dr. Christian Weedbrook and Mitsubishi Chemical Senior Chief Scientist Dr. Qi Gao, with the goal of demonstrating commercial utility for quantum simulation in semiconductor fabrication workflows.
While the project is positioned as a step toward practical quantum advantage in industrial chemistry, it remains at the stage of integrating quantum simulation outputs into established classical modeling pipelines. The approach reflects a broader trend in the field, where quantum algorithms are increasingly being tested for their ability to improve materials modeling and device design. For context, recent reporting on AI-driven optimization of quantum algorithms for hardware compatibility highlights the growing intersection of quantum computing, artificial intelligence, and industrial application.
Limitations and Open Questions
Despite the technical ambition, several limitations remain. The accuracy of quantum simulations for complex photoresist materials is still constrained by available qubit counts, gate fidelities, and error-correction capabilities on current hardware. The integration of quantum-derived parameters into multi-scale industrial models introduces additional sources of uncertainty, including the translation of microscopic quantum effects into macroscopic device performance. The project has not yet demonstrated a full end-to-end workflow that delivers a measurable improvement in semiconductor fabrication outcomes, and independent verification of the approach is not yet available in peer-reviewed literature.
As with many quantum-industry initiatives, the distinction between laboratory demonstration, simulation, and commercial deployment remains critical. The current phase focuses on computational modeling and workflow integration rather than direct fabrication or device testing. The extent to which quantum simulation can deliver predictive accuracy beyond classical methods for industrially relevant materials will depend on continued advances in hardware, algorithm design, and cross-disciplinary collaboration.
Quantum simulation refers to the use of quantum computers to model the behavior of physical systems-such as molecules or materials-whose complexity makes them difficult or impossible to simulate accurately on classical computers. In the context of semiconductor fabrication, quantum simulation aims to capture the quantum mechanical processes that govern light-matter interaction, electron transport, and chemical reactions at the nanoscale. The practical value of quantum simulation depends on the ability to perform calculations with sufficient accuracy, scale, and reliability to inform real-world engineering decisions. Current efforts focus on integrating quantum-derived insights into established industrial workflows, with the goal of improving predictive modeling and enabling the discovery of new materials with desirable properties.
