The Quantum Spring Array (QSA) proposal arranges ion strings on a two-dimensional lattice of surface traps and achieves connectivity by Coulomb coupling between neighboring strings whose separation is tuned along both the string axis and the radial direction. A 2024/2025 demonstration (Valentini et al., Phys. Rev. X 15, 041023) realized the key elements — axial and radial well-to-well coupling, an entangling gate across radially separated traps, RF-controlled radial spacing, numerically demonstrated parallel transversal gates on 2×4 ions, and mappings to error-correcting codes — but explicitly left open the question of whether these elements can be scaled to a fault-tolerant processor. This problem gathers the demonstration's own open requirements into one architecture-level question: identify the noise mechanism responsible for the anomalously high motional heating observed whenever radial transport is driven by two independent RF resonators; determine the control hardware a large two-dimensional array needs (number of independent RF channels, length scale for local DC stray-field compensation, wiring/control electronics); extend the transversal light-shift gate scheme beyond 2×4 ions under realistic imperfections; and compute architecture-level fault-tolerance thresholds and resource overheads for the QSA code mappings under realistic noise including transport errors. As of 2026 the heating mechanism is still under investigation, and no QSA-specific threshold or gate-scaling analysis exists.

Background

In a segmented surface-electrode ion trap, ions hover above a microfabricated chip; radio-frequency (RF) voltages on rails create a trapping pseudopotential (the time-averaged force on the ion from the inhomogeneous RF field), while DC electrodes provide confinement and shuttling along the remaining direction. Motional heating — the growth of the ions' phonon number from fluctuating electric fields, reviewed by Brownnutt et al. (Rev. Mod. Phys. 87, 1419 (2015)) — degrades the motional coherence on which entangling gates rely. The Quantum Spring Array (QSA) architecture (Holz et al., Adv. Quantum Technol. 3, 2000031 (2020); Valentini et al., Phys. Rev. X 15, 041023 (2025)) proposes a rectangular two-dimensional lattice in which each site holds a short linear ion string; neighboring strings couple through the Coulomb interaction at a rate tunable via their separation, giving connectivity without merging ions into one well. Valentini et al. demonstrated the building blocks in two dedicated traps: axial double-well coupling up to 39 kHz (6 ions per well) with phonon exchange, and radial coupling up to 15 kHz (up to 3 ions per well) with a Mølmer–Sørensen (MS) entangling gate (a bichromatic laser pulse entangling qubits via shared motion) between ions in two distinct traps, reaching 0.99 numerical fidelity on a parallel (CZ)⊗4(CZ)^{\otimes 4} transversal gate over 2×4 ions. A transversal gate acts qubit-wise across a code block and is prized in quantum error correction (QEC) because it does not spread errors within a block. Their scheme uses a light-shift gate in which the laser beams are angled so that the difference of the laser wavevectors along the chain, ΔkL,z\Delta k_{L,z}, satisfies ΔkL,z=(2p+1)π/(2∣z1−z2∣)\Delta k_{L,z}=(2p+1)\pi/(2|z_1-z_2|) for integer pp (with z1,z2z_1, z_2 the equilibrium positions of ions in the two wells), which cancels unwanted nearest-neighbor couplings, plus a single half-chain spin-echo refocusing pulse (a π\pi rotation applied to half the ions that reverses accumulated unwanted phases) to suppress residual next-nearest-neighbor couplings. The paper maps the QSA onto QEC primitives: Steane-type error correction via transversal CNOTs, a surface code (a topological code whose parity checks couple neighboring qubits on a two-dimensional grid) concatenated with a [[4,2,2]][[4,2,2]] code (a 4-ion code encoding 2 logical qubits), and two-dimensionally embeddable quantum low-density-parity-check (qLDPC) codes promising constant encoding rate. The demonstration also revealed the architecture's main obstacle: tuning the radial separation by changing the ratio ζ\zeta of two independent RF voltages (15–110 µm range) works, but in that dual-RF-resonator trap the heating jumps to tens of phonons per millisecond — roughly three to four orders of magnitude above the 18(3) and 2.6(4) ph/s measured in the statically coupled radial trap — which the authors attribute, without proof, to uncorrelated phase and/or amplitude fluctuations of the RF electrode voltages, and which blocks coherent operations during RF transport. A direct precursor double-well trap driven by shaped RF was reported by Tanaka et al. (Quantum Sci. Technol. 6, 024010 (2021)).

Problem Statement

The single objective is a quantitative feasibility-and-requirements assessment for scaling the demonstrated Quantum Spring Array (QSA) connectivity elements to a fault-tolerant ion-trap processor, comprising four linked questions whose answers feed one another (heating bounds gate fidelity; gate fidelity and control hardware bound thresholds and overheads): (1) Noise physics of radial transport — identify and characterize the mechanism behind the anomalously high motional heating (tens of phonons per millisecond) in the dual-RF-drive configuration used for radial transport in Valentini et al., for example through heating-rate measurements as functions of RF amplitude ratio, relative RF phase, resonator bandwidth/quality factor, and noise spectral content and correlation structure, so as to confirm or refute the conjectured uncorrelated RF phase/amplitude voltage fluctuations and to establish whether RF-tuned spacing is compatible with coherent operations, or must be replaced (e.g. by the DC-only tuning route of Pfeifer et al.); (2) Control scaling — determine how many independently driven RF channels a large 2D QSA array practically requires for local versus global string-spacing control, over what length scale locally addressable DC compensation of stray fields must act (stray fields displace ions from the RF-field minimum and induce micromotion — residual driven motion at the RF drive frequency — which such compensation removes; the paper expects this compensation length to be comparable to the ion-surface distance), and what wiring/control-electronics overhead (e.g. multiplexing) a many-site array implies; (3) Gate scaling — extend the transversal light-shift gate with geometric cancellation and refocusing sequences beyond the numerically demonstrated 2×4 ions to larger chains, quantifying how gate fidelity degrades through imperfect geometric cancellation for non-uniform ion spacings, residual higher-order couplings, realistic laser amplitude/phase errors, and motional heating, and optimizing sequences (tailored modulation, more complex dynamical decoupling) accordingly; (4) Architecture-level fault tolerance — compute fault-tolerance thresholds and resource overheads for the proposed QSA code mappings (transversal-CNOT Steane-type correction, [[4,2,2]][[4,2,2]]-concatenated surface code, two-dimensionally embeddable qLDPC codes) under realistic noise models that include the measured gate fidelities, transport errors, heating, and control-electronics overhead, thereby evaluating whether the parallel transversal-gate resource savings survive a complete noise model. A complete answer states, with experimental or simulation evidence for each component, whether and under what hardware parameters the QSA architecture crosses the fault-tolerance threshold at acceptable overhead. Each of (1)–(4) is an independently sufficient deliverable — a rigorous result on any single component (radial-transport heating mechanism, control-channel and DC-compensation requirements, beyond-2×4-ion gate scaling, or architecture-level thresholds and overheads) resolves the task on its own — and a complete answer addresses all four.

References

  1. Marco Valentini, Martin W. van Mourik, Friederike Butt, Jakob Wahl, Matthias Dietl, Michael Pfeifer, Fabian Anmasser, Yves Colombe, Clemens Rössler, Philip Holz, Rainer Blatt, Alejandro Bermudez, Markus Müller, Thomas Monz, Philipp Schindler, "Demonstration of two-dimensional connectivity for a scalable error-corrected ion-trap quantum processor architecture", Physical Review X 15, 041023 (2025), DOI 10.1103/b9s1-6r44, arXiv:2406.02406, https://arxiv.org/abs/2406.02406
  2. Michael Pfeifer, Marco Valentini, Matthias Dietl, Fabian Anmasser, Simon Schey, Jakob Wahl, Philip C. Holz, Clemens Rössler, Yves Colombe, Philipp Schindler, "Demonstration of transport in an ion trap design for two-dimensional lattices", arXiv:2605.16543 (2026), https://arxiv.org/abs/2605.16543
  3. F. Anmasser, M. Abu Zahra, K. Schüppert, M. Pototschnig, M. Wahl, M. Dietl, M. Pfeifer, Y. Colombe, J. Repp, M. Brandl, P. Schindler, C. Rössler, "Demonstration of a Multiplexing Trapped Ion Quantum Processing Unit", arXiv:2605.16010 (2026), https://arxiv.org/abs/2605.16010
  4. P. C. Holz, S. Auchter, G. Stocker, M. Valentini, K. Lakhmanskiy, C. Rössler, P. Stampfer, S. Sgouridis, E. Aschauer, Y. Colombe, R. Blatt, "2D linear trap array for quantum information processing", Advanced Quantum Technologies 3, 2000031 (2020), DOI 10.1002/qute.202000031, https://doi.org/10.1002/qute.202000031
  5. U. Tanaka, M. Nakamura, K. Hayasaka, A. Bautista-Salvador, C. Ospelkaus, T. E. Mehlstäubler, "Creation of double-well potentials in a surface-electrode trap towards a nanofriction model emulator", Quantum Science and Technology 6, 024010 (2021), DOI 10.1088/2058-9565/abe51a, https://doi.org/10.1088/2058-9565/abe51a
  6. Bassem Badawi, Philip C. Holz, Michael Raffetseder, Nicolas Jungwirth, Juris Ulmanis, Hans-Joachim Quenzer, Dirk Kähler, Thomas Monz, Philipp Schindler, "Chiplet technology for large-scale trapped-ion quantum processors", arXiv:2512.02645 (2025), https://arxiv.org/abs/2512.02645
  7. M. Malinowski, D. T. C. Allcock, C. J. Ballance, "How to wire a 1000-qubit trapped-ion quantum computer", PRX Quantum 4, 040313 (2023), DOI 10.1103/PRXQuantum.4.040313, https://doi.org/10.1103/PRXQuantum.4.040313
  8. Hudson Leone, Thinh Le, S. Srikara, Simon Devitt, "Resource overheads and attainable rates for trapped-ion lattice surgery", Physical Review Research 7, 023088 (2025), DOI 10.1103/PhysRevResearch.7.023088, arXiv:2406.18764, https://arxiv.org/abs/2406.18764
  9. Scott Jones, Prakash Murali, "Architecting Scalable Trapped Ion Quantum Computers using Surface Codes", arXiv:2510.23519 (2025), DOI 10.1145/3779212.3790128, https://arxiv.org/abs/2510.23519
  10. Ian D. Thorvaldson, Jeffrey Marshall, Jack R. Craig, Samuel K. Gorman, Charles D. Hill, Michelle Y. Simmons, "The Magic Scroll: Leveraging biased noise to improve magic state cultivation in register-based architectures", arXiv:2608.09018 (2026), https://arxiv.org/abs/2608.09018
  11. M. Brownnutt, M. Kumph, P. Rabl, R. Blatt, "Ion-trap measurements of electric-field noise near surfaces", Reviews of Modern Physics 87, 1419 (2015), DOI 10.1103/RevModPhys.87.1419, https://doi.org/10.1103/RevModPhys.87.1419