Moving Quantum Bits Doubles Coherence Time, Paving the Way for Robust Quantum Computers

Scientists are tackling the critical challenge of maintaining quantum coherence in mobile spin qubits, a key requirement for building scalable quantum computers. Jan A. Krzywda (Leiden University), Yuta Matsumoto (Delft University of Technology), and Maxim De Smet (Delft University of Technology) et al. report a significant advance in preserving the delicate quantum states of electron spins during movement within a silicon-based device. Their research demonstrates systematic mitigation of noise during qubit shuttling, extending coherence times by optimising magnetic field gradients and exploiting motional narrowing techniques. Crucially, they achieve coherence exceeding typical gate operation timescales, establishing mobile spin qubits as a promising pathway towards practical, scalable silicon quantum processors. Scientists at TNO, Delft, The Netherlands, are investigating mobile spin qubit architectures to enable flexible connectivity for efficient quantum error correction and relaxed device layout constraints. The viability of these architectures depends on preserving spin coherence during transport. Shuttling can transform spatial disorder into time-dependent noise, but its net impact on spin coherence is currently unknown. Researchers demonstrate systematic noise mitigation during spin shuttling in a linear Si/SiGe quantum dot device. Initially, by passively reducing magnetic field gradients, they minimise charge-noise coupling to the spin and double the spatially averaged dephasing time T∗2. Mobile qubit coherence mapping and noise characterisation via dynamic electron spin sensing Scientists utilise a mobile electron spin qubit as a quantum sensor to investigate noise mitigation strategies in silicon quantum processors. The experimental setup involves a Si/SiGe device with a linear quantum dot array, plunger and barrier gates, sensing dots, and a cobalt micromagnet, as depicted in Fig0.1(a) and (b). The qubit’s coherence is initially mapped as a static probe by preparing a superposition state, shuttling it to a specific position x along the conveyor, performing coherent oscillations with a variable waiting time τ, and then projecting onto the initial state after returning the electron. Decoherence is modelled as being dominated by random fluctuations of the spin splitting δωq(x, t), leading to a random phase accumulation during the waiting time. The return probability is fitted to an analytical expression to extract the coherence decay time T∗2(xn) and the decay exponent α, providing insight into the temporal noise correlations. Results in Fig0.2(a) demonstrate a strong spatial dependence of T∗2(xn) in a high magnetic field of 260 mT, with the lowest coherence times occurring at larger xn where magnetic field gradients are larger. This suggests that charge noise, coupled with the magnetic field gradient, is the predominant dephasing mechanism. Averaged over position, dephasing times are almost twice as long in a low-field regime of -30 mT, due to a reduced magnetic field gradient from the demagnetized micromagnet. Extracted decay exponents indicate low-frequency noise with α ≈ 2 in the high-field configuration and faster noise with α ≈ 1.5 in the low-field. To investigate spatial noise correlations, a two-point Ramsey experiment is performed, splitting the free evolution time between a fixed reference point x0 and a variable point xn. The envelope of the return probability is fitted to determine the two-point decoherence time T∗,0n 2 and the exponent α0n. Results in Fig0.3 show the measured T∗,0n 2 and correlation r0n as a function of the distance between the two points, revealing the spatial length scale of the noise correlations. Enhanced qubit coherence via gradient reduction and dynamical decoupling during silicon shuttling Researchers demonstrated noise mitigation during quantum bit shuttling in a silicon-based device. Initially, by reducing magnetic field gradients, the spatially averaged dephasing time was doubled from 4.4μs to 8.5μs. Further enhancement of coherence time was achieved through motional narrowing by periodically shuttling the quantum bit, reaching up to 15μs. Incorporation of dynamical decoupling techniques during shuttling over distances exceeding 200nm resulted in a coherence time of 15μs. Dressed-state shuttling provided robust protection against low-frequency noise with a decay time of 200μs, enabling protection during one-way transport without requiring pulsed control. Measurements of the dephasing time, T∗2(xn), as a function of position along the device revealed a strong spatial dependence in the high-field regime, with the lowest coherence times occurring at larger positions where magnetic field gradients were also larger. The extracted decay exponent α indicated low-frequency noise with a value of approximately 2 in the high-field configuration, compared to approximately 1.5 in the low-field configuration. Two-point Ramsey measurements showed a fitted dephasing time, T∗,0n 2, as a function of variable position xn, with the other point fixed at x0. Analysis of spatial noise correlations revealed a constant positive correlation of approximately 0.8 for xn less than 100nm in the high-field case. In the low-field case, the correlation was weaker, remaining around 0.1 for the first 100nm, and becoming significantly negative for larger displacements, eventually decaying to zero above 170nm. The extracted noise correlation coefficient, r0n, ranged from -1 to 1, indicating the degree of correlation between noise at different positions. These findings establish mobile quantum bits as a viable solution for scalable silicon quantum processors by preserving coherence over timescales exceeding typical gate and readout operations. Enhanced coherence in mobile silicon quantum dots via magnetic field control and dynamical decoupling Scientists have demonstrated strategies to preserve coherence in mobile silicon quantum dots, establishing their viability for scalable quantum processors. Through a combination of passive magnetic field gradient reduction and motional narrowing via periodic shuttling, coherence times were significantly enhanced in a linear silicon/silicon-germanium quantum dot device. Initial optimisation of the magnetic environment doubled the spatially averaged dephasing time, followed by a further improvement achieved through controlled qubit movement. Furthermore, incorporation of dynamical decoupling techniques during shuttling over distances exceeding one micron resulted in coherence exceeding typical gate and readout operation timescales. The research also highlights the effectiveness of dressed-state shuttling, providing robust protection against low-frequency noise without requiring complex pulsed control, and enabling protection during one-way transport. Although the authors acknowledge limitations related to the complexity of scaling these techniques to larger qubit arrays, they suggest future work could focus on optimising device fabrication and control protocols to further extend coherence times and improve qubit fidelity. These findings represent a substantial step towards realising mobile architectures for fault-tolerant quantum computation in silicon, offering a pathway to overcome spatial constraints and enhance connectivity in future quantum processors. 👉 More information 🗞 Coherence Protection for Mobile Spin Qubits in Silicon 🧠 ArXiv: https://arxiv.org/abs/2602.09179 Tags: