Introduction
The quest for a practical quantum computer hinges on building large arrays of high-quality qubits that can be interconnected to form error-corrected logical qubits. Two broad strategies have emerged: solid-state qubits embedded in semiconductor devices (like quantum dots or superconducting circuits) and atomic-scale qubits (trapped ions, neutral atoms, or photons). Each approach has distinct trade-offs—solid-state offers manufacturability and scalability, while atomic qubits provide consistent behavior and the ability to move qubits for flexible connectivity. Now, a new study published this week demonstrates a way to combine the best of both worlds by moving qubits within a quantum dot array without losing quantum information.
The Fixed vs. Mobile Qubit Divide
In solid-state quantum computing platforms—such as those based on superconducting circuits or semiconductor quantum dots—qubits are physically anchored to specific locations on a chip. They are wired into a fixed topology during fabrication, which limits which qubits can directly interact. While this approach benefits from mature semiconductor manufacturing, the lack of reconfigurable connectivity poses challenges for implementing efficient quantum error correction codes, which often require entangling arbitrary pairs of qubits.
In contrast, atomic and ionic qubits are naturally mobile. Trapped ions can be shuttled through an ion trap, and neutral atoms can be moved with optical tweezers. This mobility allows any two qubits to be brought together to perform entangling operations, enabling the flexible, all-to-all connectivity that simplifies error correction and many quantum algorithms. However, the overhead of trapping and controlling individual atoms is substantial, and scaling to millions of qubits remains difficult.
The Quantum Dot Solution
Quantum dots are nanoscale semiconductor structures that can confine single electrons. The spin of a trapped electron can serve as a qubit—often called a spin qubit. These qubits are fabricated using standard lithographic techniques, promising a path to large-scale integration. Until recently, however, spin qubits were essentially stationary: once placed in a quantum dot, they could only interact with immediate neighbors via tunnel coupling. The new research overcomes this limitation.
How Quantum Dots Work
A quantum dot acts as an artificial atom, holding a well-defined number of electrons. By applying voltages to tiny electrodes, researchers can control the energy levels and the position of the dot. A single electron spin in the lowest orbital state forms a robust qubit. The spin-up and spin-down states represent |0⟩ and |1⟩. These qubits have demonstrated long coherence times and high-fidelity gate operations, but their immobility was the major bottleneck for building scalable quantum processors.
The Achievement: Coherent Transport
The new work, published in Nature Physics, shows that an electron spin can be physically moved from one quantum dot to another while preserving its quantum state. The researchers formed a linear array of quantum dots and used carefully timed voltage pulses to shift the electron between adjacent dots—like a bucket brigade. They verified that the spin remained coherent during the transfer, meaning the quantum information was not destroyed. This coherent transport over several dots represents a crucial step toward reconfigurable qubit arrays.
The experiment achieved >99% transfer fidelity for a single hop, and the team successfully shuttled the qubit across three dots. By adjusting the pulse sequence, they could also swap two spin qubits in the chain, demonstrating basic reconfigurability. This is essentially the same mobility that atomic qubits enjoy, but within a fully manufacturable semiconductor system.
Implications for Error Correction
One of the main advantages of mobile qubits—whether ions or now spin qubits—is the ability to create flexible connectivity patterns. In error correction codes like the surface code or Bacon-Shor code, certain operations require a qubit to interact with many distant neighbors. With stationary qubits, this forces a two-dimensional grid with many dedicated connections, which can be wasteful. With mobile qubits, a single ‘shuttle’ can bring any qubit to any other, dramatically reducing the number of physical qubits needed for a given logical qubit.
For example, if you have a linear chain of mobile spin qubits, you can simulate two-dimensional connectivity by repeatedly moving qubits. This simplifies chip design and could lead to more efficient use of quantum resources. The new result shows that such a paradigm is now experimentally accessible in a solid-state platform.
Future Outlook
While the demonstrated mobility is limited to a few dots, the team is already working on longer chains and integrating the technique with high-fidelity gates. The long-term vision is a hybrid quantum processor: arrays of stationary spin qubits for computation, with a few ‘mobile’ qubits that can carry entanglement across the chip. This would combine the scalability of manufactured qubits with the connectivity of atomic systems.
Challenges remain: maintaining coherence over many hops, reducing errors during transport, and implementing quantum error correction protocols on the fly. Nonetheless, this breakthrough opens a new direction in quantum computing hardware, blurring the line between solid-state and atomic approaches. As manufacturing techniques improve, we may soon see chips with thousands of mobile spin qubits, each able to interact with any other—a true quantum fabric for error-corrected logical qubits.
Read the full article