High-precision quantum gates with diamond spin qubits

Researchers at QuTech, in collaboration with Fujitsu and Element Six, have demonstrated a complete set of quantum gates with error probabilities below 0.1%. While many challenges remain, being able to perform basic gate operations with errors occurring below this threshold, satisfies an important condition for future large-scale quantum computation. The research is published in Physical Review Applied on 21 March 2025.

Quantum computers are anticipated to be able to solve important problems that are beyond the capabilities of classical computers. Quantum computations are performed through a large sequence of basic operations, called quantum gates. For a quantum computer to function, it is essential that all quantum gates are highly precise. The probability of an error during the gates must be below a threshold, typically of the order 0.1 to 1%. Only then, errors are rare enough for error correction methods to work successfully and ensure reliable computation with noisy components. 

Diamond spins

Spins in diamond are a type of qubit that shows promise for quantum computation. These qubits consist of electron and nuclear spins associated with atomic defects, for example a nitrogen atom replacing a carbon atom in the diamond. They operate at relatively high temperatures up to 10 Kelvin and are well protected from noise. Also, their natural connection to photons—the elementary particles of light—enables distributed computation over quantum networks. However, realising a complete set of quantum gates with low enough error rates has remained a challenge until now. 

Precise universal gates

Researchers at QuTech, the interfaculty quantum technology research institute of Delft University of Technology, have now demonstrated a highly precise universal set of quantum gates using a diamond quantum chip. The researchers used a system of two qubits, one formed by the electron spin of the defect center, the other by its nuclear spin. Each type of gate in this two-qubit system operates at an error below 0.1%, and the best gates even reach errors as low as 0.001%. 

"To realise such highly precise gates we had to systematically remove sources of errors. The first step was to use ultrapure diamonds that have a lower concentration of carbon-13 isotopes as these cause noise”, says Hans Bartling, lead author. The second key step was to design gates that carefully decouple the spin qubits from each other and from interactions with the remaining noise in the environment. 

Characterising, optimising and testing the quantum gates

A final challenge was to find tools to reliably characterise the gates and optimise their parameters. For this, the team turned to a method called ‘gate set tomography’, which provides the full quantum description of the gates. “It was essential that our characterisation provided complete and precise information about the gate errors, as this enabled us to systematically find imperfections and optimise all the gate parameters”, says co-author Jiwon Yun. 

Ultimately, the researchers put the quantum gates and their characterisation to the test by performing an artificial algorithm with a large sequence of gates. After 800 gate operations the result could be accurately predicted from the team’s knowledge of the individual gates, indicating that the gate operations were now both precise and well understood.

The road ahead

While high-precision universal gates are a key prerequisite towards quantum computation, there is still a long way to go to large scale computation. “Our demonstration was on a two-qubit system and using a particular type of defect”, says Tim Taminiau who supervised the research. “A key challenge is to maintain and further improve the gate quality when moving to chip-scale integrated optics and electronics and scaling to many more qubits.”

Realising such larger processors is the focus of the research effort at QuTech and of its collaboration with Fujitsu. The team takes a full stack approach, in which not only improved quantum bits are studied, but also the required control electronics, scalable fabrication methods and new types of quantum computer architectures. “Making the next big step will require bringing together scientists, engineers and industry”, says Taminiau.