arXiv:2610.08411
A 3D-Printed GHz Microwave-Resonator Paul Trap for Electron Confinement and Millisecond Spin-Qubit Coherence
Niklas V. Lausti, Vineet Kumar, Ivan Hudák, Jiří Hajnyš, Peter Kúš, Radek Plašil, Michal Hejduk
Abstract
We report a 3D-printed microwave-resonator Paul trap that operates at the GHz frequencies required for electron confinement while preserving an open geometry suitable for laser access and imaging ion Coulomb crystals. The resonator exhibits moderate quality factors of approximately 1000, enabling large confining electric-field amplitudes at low input power, and is in good agreement with finite-element-method (FEM) predictions despite manufacturing imperfections, which we experimentally characterize. The validated FEM model is then used to calculate the electromagnetic fields in the trapping region. These field distributions, together with surface-roughness measurements, are incorporated into a model of electron spin-qubit decoherence to estimate the coherence time. Our analysis predicts that electron spin-qubit coherence times on the order of 10 ms should be achievable under the assumption that motional heating is dominated by Johnson noise, making such systems relevant for quantum information processing. The method of obtaining the coherence time is applicable to any electron Paul trap.
AI Summary
The paper addresses whether a GHz Paul-trap resonator can combine fields suitable for electron confinement with optical access for hybrid electron–ion experiments, and how its intrinsic oscillating magnetic fields affect electron-spin coherence. An open resonator is fabricated from AlSi10Mg by laser powder bed fusion and polished to an RMS surface roughness of $2.5\pm0.7\,\mu\mathrm{m}$. Microwave spectroscopy without the ion endcaps gives a resonance near $2.3\,\mathrm{GHz}$ and quality factor $Q=990\pm10$, compared with $1076\pm9$ in finite-element simulations; resonance frequencies agree within approximately $2\,\mathrm{MHz}$. Simulations of the assembled device at $10\,\mathrm{W}$ input predict a $2.21\,\mathrm{GHz}$ trapping resonance, electron pseudopotential depth around $3500\,\mathrm{K}$, and secular frequencies of approximately $77$, $77$, and $154\,\mathrm{MHz}$. The simulated magnetic-field gradients enter an effective spin–motion Hamiltonian and heating master equation, for which an analytical coherence expression is derived. Assuming Johnson-noise-dominated heating, the estimated heating rate is $101\pm24\,\mathrm{s}^{-1}$. For an initial axial Fock state with $n_0=2750$ and a $10\,\mathrm{mT}$ quantization field, the main analysis predicts a coherence time of $21\pm7\,\mathrm{ms}$. The conclusion instead quotes approximately $60\,\mathrm{ms}$ and $30\,\mathrm{s}^{-1}$, without reconciling these values. Electron confinement, simultaneous electron–ion trapping, and spin coherence are not experimentally demonstrated. The coherence estimate omits additional noise mechanisms, assumes a prepared motional Fock state, and treats the parent ion’s Coulomb interaction as negligible at sufficiently high excitation. A rigorous coupled electron–ion treatment remains future work; additional experimental decoherence channels are expected to shorten the prediction.
Central claim may not hold: The stated magnetic gradients and coherence model predict negligible coherence loss at 21 ms, rather than the reported 1/e decay, leaving the central quantitative coherence estimate unsupported. The conclusion reports approximately 60 ms coherence and 30 s⁻¹ heating without reconciling them with the detailed results of 21 ± 7 ms and 101 ± 24 s⁻¹. The printed quartic coefficient βz = 79 J/mm⁴ yields Kz/ℏ ≈ 1.6 × 10²² s⁻¹ under the paper’s conversion formula, contradicting the reported 0.9 s⁻¹.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 14 pages, 6 figures