Fujitsu Corporation has unveiled a working prototype of a quantum computer built on diamond spin qubits. This is not just a laboratory experiment, but a real attempt to solve the main problem of quantum systems—scaling. Instead of increasing the number of qubits within a single device, engineers propose connecting individual modules via photonic channels, paving the way for distributed quantum networks.
The Diamond Foundation: Why SnV Centers
At the core of the development are tin-vacancy centers (SnV) in the diamond crystal lattice, which serve as qubits. These defects are integrated with photonic integrated circuits, enabling quantum states to be read out optically. The choice of SnV centers is no accident: unlike the more common NV centers, they possess high structural symmetry and exhibit lower susceptibility to external noise, which is critical for stable operation.
The prototype operates at a temperature of about −271.6 °C. This is slightly warmer than the environment for superconducting counterparts (approximately −273.13 °C), which simplifies operation and reduces requirements for cryogenic equipment. The system is managed through Fujitsu's proprietary hybrid platform, which abstracts the user from the hardware architecture, making interaction intuitive.
Technological Details and the Path to Modularity
The key innovation is the connection of diamond structures with photonic circuits. To achieve this, engineers had to solve a nontrivial problem: reducing the thickness of the diamond layer from hundreds of micrometers to hundreds of nanometers and developing methods for bonding diamond with other materials. This is precisely what enables the creation of compact, optically linked modules.
Fujitsu has already announced plans to build a multi-module version of the quantum computer in 2027. In parallel, work is beginning on hybridizing diamond and superconducting architectures, which could lead to the creation of universal computing complexes. The project is the result of a collaboration with Delft University of Technology and the QuTech research center, which began back in 2020, and is supported by the Dutch Holland High Tech program.
Fujitsu's roadmap is ambitious: by 2030, it aims to achieve 250 logical qubits, and by 2035—1,000. However, it is worth emphasizing that these are goals, not current achievements. For now, the prototype demonstrates only basic functionality, but the very fact of using photonic links for scaling is a strategically sound step that could radically change the economics of quantum computing in the coming decade.
My view: the transition to modular systems with optical connectivity is the only realistic path to building fault-tolerant quantum machines. If Fujitsu succeeds in integrating diamond and superconducting technologies, it could become the foundation for a new industry standard.