Insider Brief
- Researchers from, Quantinuum, Harvard and Caltech have created what they are referring to as the first-ever experimentally demonstrated “true topological qubit,” using a Z₃ toric code to encode quantum information in a way that is more resistant to errors.
- The research builds on work extending back to 2015, and shows how exotic quasiparticles called non-Abelian anyons can be manipulated to protect quantum information, potentially reducing the resource burden of error correction in quantum computing.
- Conducted on Quantinuum’s H2 ion-trap quantum processor, the experiments confirm theoretical predictions from a 2015 paper and mark a significant step toward scalable, fault-tolerant quantum computing.
Topological quantum computing is widely regarded as one of the most promising approaches to overcoming the fragility and error-prone nature of quantum systems. By encoding information into the spatial and braiding properties of exotic quasiparticles, this method offers intrinsic protection against certain types of errors, potentially reducing the overhead for error correction.
Achieving a true topological qubit has been a goal in the field, but it’s also one that — because it requires a delicate balance of theoretical precision and experimental control — has remained tantalizingly out of reach for the scientific community.
Now, an international research team led by Quantinuum report they have taken a step in that they have created a true topological qubit, an initial step that could pave the way for more scalable and robust quantum computers. The topological qubit they have created meets the criteria established and accepted in this 2015 paper.
The achievement, which the team detailed in a study on the pre-print server ArXiv and in a blog post, demonstrates how non-Abelian anyons — an exotic class of quasiparticles that are both rare and difficult to observe — can be harnessed to encode quantum information. The topological qubit is capable of being implemented in the existing Quantinuum processor, and the next steps now include exhibiting universality.
Preparing The Topological State
The researchers write that they succeeded in preparing a topological state known as the “Z₃” toric code in a qutrit-based Hilbert space. To break this down a bit, Hilbert space is simply a mathematical framework where quantum states are represented. Essentially, the researchers encoded and protected quantum information by creating a grid of quantum systems with three states — qutrits — where the information is stored in patterns of relationships between the systems rather than in the systems themselves, making it more resistant to errors.
Unlike traditional quantum bits, or qubits, that are prone to errors from environmental noise, topological qubits rely on properties of their spatial arrangement to resist interference, offering a potential solution to one of quantum computing’s toughest challenges. By entangling defects in this state, researchers demonstrated fundamental elements of quantum error correction based on non-Abelian principles, a key step toward making quantum computers reliable and scalable.
The concept of topological qubits was first defined in a 2015 paper, referred to above, and that remains a cornerstone of the field. It proposed encoding quantum information into systems of non-Abelian quasiparticles, whose fusion outcomes depend not only on their types but also on the sequence in which they are combined.
According to the researchers, their work bridges the gap between theoretical physics and practical quantum technology, underscoring the promise of topological qubits to reduce the resource overhead of error correction significantly.
Ilyas Khan, Founder and Chief Product Officer, Quantinuum, said this study demonstrates for the first time that such a state of matter can be realized and manipulated in a controlled environment.
“I initiated this area of research way back in 2015 with a small team in Cambridge, as we dreamt, in those days, of a quantum processor that might be powerful enough to fulfill the criteria that Michael Freedman and others had highlighted in previous work,” said Khan. “There are different pathways to building topological qubits and in this respect we are sitting alongside, but separate, from the long standing research programme at Microsoft.”
In practical terms, error correction in conventional quantum computers can consume as much as 90% of a system’s computational resources for tasks such as executing complex algorithms. Non-Abelian anyons, however, provide a universal gate set that simplifies these operations.
“In a new paper in collaboration with Harvard and CalTech, our team is one step closer to realizing fault-tolerant non-Abelian quantum computing. This paper is a follow-up to our recent work published in Nature, where we demonstrated control of non-Abelian anyons. This marks a key step toward non-Abelian computing, and we are the only company who has achieved this,” the team wrote in the blog post
Ion-Trap Quantum Processor
The experiments were conducted on Quantinuum’s H2 ion-trap quantum processor, an industry-leading platform that features 56 fully connected qubits and gate fidelities exceeding 99.8%. Using the processor’s high connectivity and precision, the researchers encoded qubits into qutrits to construct a lattice representing the Z₃ toric code. This setup enabled them to study parafermion and charge-conjugation defects, which are are exotic quantum structures in topological systems that manipulate particle-like excitations to encode and process quantum information. These structures are also theoretical precursors to non-Abelian anyons. These defects were then entangled into Bell pairs, a hallmark of quantum error correction and a demonstration of the viability of topological systems.
One of the study’s key innovations was the ability to fuse defects and observe their interactions with anyons, quasiparticles that are neither fermions nor bosons and can only exist in systems with topological order. By verifying the predicted behaviors of these interactions, such as their ability to exchange charge and flux anyons, the researchers provided direct evidence for the computational utility of non-Abelian systems. The study also confirmed that these systems can encode quantum information in a way that is inherently protected against certain types of errors, a critical requirement for practical quantum computing.
The Research Road Ahead
The research also highlights several challenges and next steps for scientists. The demonstrated qutrit fidelity was quite high — at 96.5% — but would need to be pushed higher to reach thresholds required for fault-tolerant quantum computing. Scaling these systems to larger dimensions may also require advances in hardware and algorithms, as manipulating non-Abelian defects is far more complex than controlling simpler point-like anyons.
While the Z₃ toric code represents a critical step, the researchers note in the paper that next steps would include achieving a universal non-Abelian gate set — a fundamental building block for large-scale quantum systems — which has been an ongoing challenge.
With those challenges and limitations in mind, the researchers aim to extend their work to more advanced topological codes, such as the S₃ toric code, which could provide universal computational capabilities. They also plan to explore adaptive error correction protocols, which could further enhance the robustness of these systems. Integrating these developments with existing quantum error correction techniques, such as repeat-until-success protocols, could create a more comprehensive framework for building reliable quantum computers.
Beyond The Technical
The implications of this study go beyond the technical domain, offering insights into how topological quantum systems could reshape the future of computing. By reducing the computational overhead of error correction, topological qubits could make quantum computing more accessible and practical for solving complex problems in fields such as cryptography, materials science, and artificial intelligence. The study’s findings add momentum to ongoing efforts to overcome the scalability challenges that have long limited quantum technologies.
This milestone continues Quantinuum’s track record of advancing quantum error correction, following earlier achievements such as fault-tolerant teleportation of logical qubits and the first implementation of the Quantum Fourier Transform with error correction. The company’s QCCD architecture, combined with contributions from collaborators at Harvard and Caltech, demonstrates the value of interdisciplinary research in tackling the most difficult problems in quantum science.
For a deeper, more technical view of the work, which this article cannot provide, please see the paper in ArXiv. Studies in pre-print servers have not been officially peer-reviewed, but give scientists the ability to receive early feedback on their work before seeking peer-review. It’s particularly useful in fields, such as high-energy physics and quantum computing, when rapid dissemination is required.
Ilyas Khan stated: “In this particular field we have led from the front by ensuring timely publication of all the aspects of our research that matter. It is a widely held view that Topological Qubits, when available, are without question the best qubits possible. Creating non-Abelian anyons as a route to demonstrating the first ever true topological qubit could only be done because we now have the world’s most powerful quantum computer. Our capability will increase even more in 2025 and 2026 and I am excited to see how this work develops alongside our other research in QEC – all powered by a quantum processor that is years ahead of the competition”
The research team incuded Mohsin Iqbal and Henrik Dreyer, both from Quantinuum in Munich, Germany; Anasuya Lyons, Chiu Fan Bowen Lo, and Ashvin Vishwanath, all from Harvard University; and Nathanan Tantivasadakarn from the Walter Burke Institute for Theoretical Physics and California Institute of Technology. From Quantinuum’s facility in Broomfield, Colorado, contributors included Joan Dreiling, Cameron Foltz, Thomas M. Gatterman, Dan Gresh, Nathan Hewitt, Craig A. Holliman, Jacob Johansen, Brian Neyenhuis, Yohei Matsuoka, Michael Mills, Steven A. Moses, and Peter Siegfried. Ruben Verresen, affiliated with both the Pritzker School of Molecular Engineering at the University of Chicago and Harvard University, also contributed to this work..