Introduction

Hello. I'm Hidetoshi Matsumura from the Quantum Laboratory at Fujitsu Research. At our lab, we're dedicated to exploring the exciting world of quantum computers. These aren't just faster versions of what we have now; they operate on entirely different principles, promising revolutionary computational power with the potential to transform countless industries.

In this article, I'll be diving deeper into our press release from March 24, 2025, titled "Fujitsu and QuTech realize high-precision quantum gates" [1]. The goal is to explain the value and significance of the technology we unveiled. With so many incredible advancements being announced in the quantum computing space lately, I hope you'll come away thinking, "Wow, Fujitsu's diamond spin quantum computer is pretty impressive too!"

What is a Diamond Spin Quantum Computer?

Let's start with the basics. I'll explain the unique characteristics of the diamond spin quantum computer, which we're developing in collaboration with our research partners Delft University of Technology and its quantum technology research institute QuTech[2].

This method leverages spins that reside in structures called "color centers," which are formed when carbon atoms in a diamond crystal are replaced with vacancies and atoms like nitrogen or tin. The spins are electron spins, the nuclear spins of substituted atoms, and the nuclear spins of several ¹³C carbon isotopes around the color center. The press release is about the technology for nitrogen-vacancy center (NV center). You can see a schematic diagram of this in Figure 1. This is the fundamental building block, the unit module, for our quantum computer.

Figure 1. A unit module of diamond spin based quantum computer (with NV Center)

This type of quantum computer has two key advantages. First, our electron spin qubits can communicate with each other using light. Unlike our approach, many other quantum computing technologies rely on strict physical arrangements of qubits to achieve their interaction, which can hinder scalability. However, with optical connections, we have more flexibility, making it easier to scale up the system by arranging these unit modules. Second, our system can operate at relatively high temperatures – above 1 Kelvin. Compared to superconducting quantum computers, which require extremely low temperatures around 20 milliKelvin (mK), our method needs about 1/1000 of the cooling energy, leading to significantly simpler cooling infrastructure. Fujitsu believes these technical advantages position this method as a good candidate for realizing large-scale quantum computing systems in the future, and we are conducting its research and development.

Building a quantum computer from these diamond crystal unit modules involves more than just its manufacturing. It demands a range of technological innovations (the terms in parentheses below correspond to the technical field names in Figure 2). This includes: developing error correction algorithms tailored for this method (Quantum algorithm, Quantum error correction); creating compilers that translate user-defined quantum algorithms into quantum gate operation sequences executable on our system (Compiler, Microarchitecture); determining the physical operations – electromagnetic waves, magnetic fields, lasers, and so on – required for each quantum gate on the unit module (Quantum gate design – the topic of the press release); developing electronic circuits (Interface electronics) and optical components (Diamond spin quantum chip) that realizes the electromagnetic waves, magnetic fields, and lasers needed for quantum gate operations; and integrating all of these components into a system (3D integration). Fujitsu is partnering with Delft University of Technology, which possesses world-leading technology in diamond spin control, to conduct joint research across all of these areas.

Figure 2. Overview of research collaboration between Fujitsu and Delft University of Technology

For more on our diamond spin quantum computer approach and our collaborative research with Delft University of Technology, please check out our other Fujitsu Tech Blog articles: [3]-[6] (Sorry, we only have Japanese ones).

Technical Deep Dive into the Press Release

Now, let's break down the key elements of our recent press release. I'll walk you through each point, providing more context and explanation.

Key Technological Innovations
1. Reduced environmental noise using high-purity diamonds
2. Decoupling gates designed to mitigate environmental noise
3. Application of gate set tomography

The Achievement
A world's first: an error probability below 0.1% for both 1-qubit and 2-qubit gates on diamond spin qubits.

Point 1: Reduced environmental noise using high-purity diamonds

As I mentioned earlier, the nuclear spins of ¹³C atoms surrounding the color center also can be used as qubits. This is because the ¹³C nuclear spin interacts electromagnetically with the electron spin of the color center. However, this "electromagnetic interaction" can become a source of environmental noise for quantum gates implemented using electromagnetic radiation, if it's not carefully controlled. Standard diamond samples contain approximately 1% ¹³C, the natural abundance, and the resulting noise is significant. Therefore, a high-purity diamond with the ¹³C concentration reduced to about 1/100 of its natural abundance are used to perform experiments with a two-qubit system based on an electron spin and a nitrogen nuclear spin.

Point 2: Decoupling gates designed to mitigate environmental noise

Even with high-purity diamond, it is hard to completely eliminate the effects of ¹³C. Furthermore, the nitrogen nuclear spins also interact with the electron spins, so their influence need to be minimized as well. In addition, factors like the misalignment of the static magnetic field that is necessary to use the color centers as qubits, and inaccuracies in the control of the electromagnetic waves themselves, can also introduce noise. To combat these various noise sources, a technique called "decoupling" is employed. This involves periodically irradiating the electron spin with microwave pulses that flip its state, effectively inverting the influence of environmental noise, which cancels out the noise over the long term. The implemented quantum gate operations are based on this decoupling approach. Through the use of gate set tomography (which will be explained next), a detailed analysis of the underlying noise mechanisms was conducted, and microwave parameters – frequency, intensity, interval of spin-flip pulses, and so on – were optimized to minimize errors as much as possible.

Point 3: Application of gate set tomography

To optimize quantum gate operations as described above, it is necessary to understand the types of errors that occur in the implemented quantum gates. A technique called "gate set tomography" was introduced for this purpose. While it requires a large amount of data collection (many experimental runs), it provides more detailed and complete error information than other methods. To the best of our knowledge, this is the first time gate set tomography has been applied to diamond spin qubits anywhere in the world.

The Value and Significance of the Achievement

Achieving "the world's first error probability below 0.1% with diamond spin qubits" is a huge step our consistent efforts to improve quantum gate operations towards building practical quantum computers. That's why we had a press release for this topic. Let's dive into why this is such a big deal.

Qubits are inherently susceptible to "errors" caused by external noise and imperfections in gate operations. Without any measures, if we simply scale up the size of current quantum computers (number of qubits, number of quantum gates) to the point where they could outperform classical computers in real-world applications, the accumulation of these errors would make the final results meaningless. Therefore, for quantum computers to truly realize their potential, we need to implement "quantum error correction," which allows us to detect and correct errors in parallel with the computation.

Quantum error correction, in its simplest form, involves representing a single logical qubit using multiple physical qubits. This redundancy allows us to detect and correct errors that occur in the physical qubits while simultaneously performing the desired computation. Even considering the overhead associated with this redundancy – the increased number of qubits and the additional quantum gate operations required for error detection and correction – it's known that we can make the errors in quantum computation arbitrarily small if the error probability of the physical qubits is kept below a certain value (the threshold). The exact value of this threshold depends on various factors, such as the method used to encode the logical qubit with physical qubits and the types of errors that occur. However, it's generally accepted that achieving an error probability of around 0.1% is a good target.

Actually, other research groups have reported similar error probability on diamond spin qubits. However, the result didn't include the full set of qubit gates needed for performing error correction. In contrast, our work demonstrates the implementation of a single-qubit gate that can realize arbitrary quantum states and a two-qubit gate that implements the CNOT operation. This is a significant step forward because it "demonstrates, for the first time, the possibility of a quantum computer based on diamond spin qubits performing quantum computation while simultaneously executing quantum error correction." Furthermore, the achieved error probability below 0.1% is among the lowest reported for any quantum computing platform, making it a compelling demonstration of the potential of the diamond spin approach.

Conclusion

As outlined in this article, we've demonstrated the potential and promise of diamond spin-based quantum computing for practical applications. Of course, there are still many technical challenges to overcome before we can realize fully functional quantum computers. For example, our current work focuses on quantum gate operations 'within' a single module of a quantum computer. We need to achieve the same high level of precision for quantum gate operations 'between' modules. In addition, as we integrate the various elemental technologies into a complete system, we'll need to find optimal tradeoffs and address new, unforeseen problems.

We're committed to tackling these challenges head-on, with the ultimate goal of revolutionizing the world with diamond spin quantum computers. We hope you'll continue to follow our progress and look forward to the future of Fujitsu's diamond quantum spin computer!