Researchers help unlock a more versatile future for quantum computing

Adobe freestock image

Published in Nature Photonics, the international study introduces Clavina – a modular, programmable photonic quantum computing architecture capable of combining both linear and nonlinear quantum operations within a single system. The advance provides a practical blueprint for more powerful and adaptable quantum computers that could one day tackle problems beyond the reach of today's most advanced machines. 

Quantum computers use the principles of quantum physics to solve certain calculations far more efficiently than conventional computers. Among the different technologies being explored, photonic quantum computers—which use particles of light (photons) to process information—offer major advantages in speed, stability and energy efficiency. However, building a truly universal photonic quantum computer has remained a long-standing challenge because existing systems have struggled to incorporate the nonlinear operations needed for the most powerful quantum algorithms. 

The new Clavina architecture addresses this challenge by allowing specialised quantum modules to be added or removed as required, much like components in a conventional computer. This flexible design enables researchers to perform a much wider range of quantum computing tasks using a single platform, rather than relying on separate, purpose-built experimental systems. 

Using the new architecture, the team demonstrated several advanced quantum applications, including large-scale quantum simulations, the generation of quantum states needed for future error correction, and complex calculations that had previously been impractical for photonic hardware. The experiment was carried out in Prof Ian Walmsley and Dr Raj B. Patel’s labarotary at Imperial. 

Queen Mary's contribution, led by Dr Jinzhao Sun from the School of Physical and Chemical Sciences, focused on the theoretical aspect, in collaboration with Prof Vlatko Vedral from Oxford and Prof Myungshik Kim and Prof Roberto Bondesan from Imperial. They developed the theoretical framework underpinning one of the study's key demonstrations: the quantum simulation of the Bose-Hubbard model, an important mathematical model used to understand the behaviour of interacting quantum particles as well as quasi-deterministic breeding of Gottesman-Kitaev-Preskill (GKP) states, opening up a new avenue for universal quantum computing. Such simulations could eventually help scientists investigate new quantum materials, design novel technologies and deepen our understanding of complex physical systems. 

Dr Shang Yu, first author at Imperial, said: 

"Photonic quantum computing has enormous potential, but one of its greatest limitations has been the lack of a practical way to combine scalable optical circuits with the nonlinear operations required for universal quantum computing. This work demonstrates a flexible architecture that brings those capabilities together, creating a platform that can be adapted for many different quantum applications." 

The researchers say the modular approach means future technologies can be incorporated into the architecture as they become available, allowing the platform to evolve without requiring an entirely new system to be built. 

While practical fault-tolerant quantum computers remain a long-term goal, the research represents an important step towards scalable quantum technologies that could eventually transform fields including materials science, chemistry, optimisation and secure communications. 

The research was carried out through an international collaboration involving Queen Mary University of London, Imperial College London, the University of Oxford, the University of Hong Kong, and other partners.