A 2024 proposal by Huang et al. introduced a programmable matrix of spin-transfer-torque (STT) magnetic devices integrated above a two-dimensional topological superconductor (2D TSC), whose local stray fields would pin, move, braid, and fuse superconducting vortices and the Majorana zero modes (MZMs) they bind, with in-principle feasibility established by time-dependent Ginzburg-Landau (TDGL) and time-dependent Bogoliubov-de Gennes (TDBdG) simulations. The proposal itself leaves its central feasibility questions open: whether STT stacks can operate at cryogenic temperature with stray fields of the amplitude and spatial profile needed to control individual vortices without prohibitive Joule heating, how MZM parity readout can be integrated and calibrated, whether the phenomenological TDGL vortex dynamics is quantitatively reliable for candidate 2D-TSC materials, and what braiding fidelity survives quasiparticle poisoning (stochastic quasiparticle events that flip the encoded parity), cross-talk, and fabrication variability in a real device. No physical braiding of vortex MZMs has been demonstrated in any platform.

Background

Majorana zero modes (MZMs) are zero-energy quasiparticles that are their own antiparticles; when localized in pairs they encode a nonlocal fermionic parity that is protected against local perturbations, making them candidate building blocks of a fault-tolerant topological quantum computer. A natural two-dimensional host is a topological superconductor (TSC) — a superconductor whose quasiparticle spectrum is topologically nontrivial — in which each magnetic vortex, the whirl of suppressed superconducting order parameter and circulating supercurrent that forms around a quantized magnetic flux line, binds one MZM in its core, as established for the Fu-Kane platform (a conventional superconductor depositing a superconducting pairing gap onto the surface states of a topological insulator) and for the iron-based superconductor Fe(Te,Se), where scanning-tunneling evidence for vortex-bound MZMs was reported in 2018. Exchanging (braiding) two vortices implements a non-Abelian unitary transformation on the encoded parity — one whose effect depends on the order in which exchanges are performed, which is what makes it the elementary topological gate — and observing this exchange statistics is widely regarded as the decisive test of MZM physics; as of 2019 it was characterized as an unreached goal, and no physical vortex exchange has been demonstrated in a superconductor.

The practical obstacle is vortex control: vortices are pinned by material disorder, global magnetic fields move all vortices at once, and tip-based manipulation is slow and non-scalable. Huang et al. (arXiv:2404.18578; published in Science China Physics, Mechanics & Astronomy, 2025) proposed to place a programmable matrix of spin-transfer-torque (STT) devices — magnetic tunnel junction elements, mature from magnetoresistive random-access memory (MRAM) technology, in which a spin-polarized current pulse switches the magnetization of a thin free magnetic layer and thereby turns a local magnetic stray field on or off — above the TSC layer. Switching individual STT elements would deterministically create, pin, move, and fuse vortices and their MZMs on chip. The authors support the proposal with a simulation chain: finite-element analysis of the STT stray fields; time-dependent Ginzburg-Landau (TDGL) equations — the phenomenological relaxational dynamics of the superconducting order parameter, rigorously grounded only for dirty (impurity-dominated) superconductors near the critical temperature TcT_c — for the vortex motion; and time-dependent Bogoliubov-de Gennes (TDBdG) equations — the microscopic mean-field dynamics of the quasiparticles — for the MZM evolution, showing high-fidelity vortex manipulation, braiding dynamics limited by vortex acceleration and finite MZM coupling, and controlled fusion with a measurable energy splitting.

The paper is explicit that this is an in-principle demonstration. Its Appendix D reports only a room-temperature electrical characterization of a first eight-element STT matrix, stating that low-temperature device compatibility, achievable stray-field strengths, and write-pulse Joule heating "need to be further investigated"; the concrete implementation of parity readout for fused MZMs (e.g., charge-based methods from prior 2D-TSC proposals) is declared beyond the scope of the work; and the TDGL description is justified by qualitative arguments (e.g., the magnetic penetration depth λ\lambda being much larger than the superconducting coherence length ξ\xi, the scale over which the order parameter recovers inside a vortex core, in iron-based superconductors) rather than a microscopic validation for specific candidate materials, several of which (Fe(Te,Se) in particular) are multiband and anisotropic.

Problem Statement

Determine whether the STT-matrix scheme of Huang et al. can physically control and braid vortex-hosted MZMs in a real two-dimensional topological superconductor at cryogenic temperature with quantified fidelity. Concretely, and jointly in service of that single objective: (1) demonstrate an integrated STT-matrix/2D-TSC device operating at cryogenic temperature whose individually addressable elements produce stray fields of the amplitude and spatial profile required to deterministically pin and move single vortices, with write-pulse Joule heating low enough that superconductivity, and the vortices binding the MZMs, survive every write-pulse cycle; (2) integrate and calibrate a parity-readout protocol for fused MZMs within the STT architecture, with quantitative noise and readout-fidelity budgets; (3) establish, against microscopic theory or calibrated simulation, how accurately the phenomenological TDGL vortex dynamics — used to predict the scheme's braiding performance — describes the candidate 2D-TSC materials (multiband or strongly anisotropic systems such as iron-based superconductors), including the consequences of the neglected microscopic effects; and (4) quantify the braiding fidelity achievable in a realistic device, including the effects of quasiparticle poisoning (stochastic quasiparticle events that flip the encoded parity), cross-talk between neighboring STT elements, and fabrication variability, assessing whether operation in the fault-tolerant regime — error rates below what quantum error correction can tolerate — is attainable at realistic temperatures and on realistic timescales. A complete answer either demonstrates the above in an experimental device with measured fidelities and validates the modeling, or establishes quantitatively (by measured device parameters, heating/field budgets, or microscopic bounds) that some component forbids the scheme from reaching the fidelity regime it targets. Stated part by part: a positive resolution requires all of (1)-(4) together, including the modeling validation of (3); the negative branch is resolved by a quantitative bound on any single physical component of the scheme — the cryogenic STT-matrix operation of (1), the parity readout of (2), or the achievable braiding fidelity of (4) — showing that this component forbids the targeted fidelity regime; part (3), being a claim about the reliability of the modeling rather than about such a component, qualifies how far the positive branch's device and fidelity predictions can be trusted without by itself resolving the task negatively.

References

  1. Huang, Guangyao; Fu, Jibang; Yi, Xiaofeng; Wang, Weichen; Ren, Bo; Yang, Zhaohua; Xue, Shichao; Zhang, Xinfang; Deng, Mingtang. Scheme for braiding Majorana zero modes in vortices using an STT-matrix. arXiv:2404.18578 (2024); published as 'Scheme for braiding Majorana zero modes in vortices using STT-matrix', Science China Physics, Mechanics & Astronomy (2025), DOI 10.1007/s11433-025-2797-y. https://arxiv.org/abs/2404.18578
  2. Fu, Liang; Kane, C. L. Superconducting proximity effect and Majorana fermions at the surface of a topological insulator. Phys. Rev. Lett. 100, 096407 (2008), DOI 10.1103/PhysRevLett.100.096407; arXiv:0707.1692. https://arxiv.org/abs/0707.1692
  3. Wang, Dongfei; Kong, Lingyuan; Fan, Peng; Chen, Hui; Zhu, Shiyu; Liu, Wenyao; Cao, Lu; Sun, Yujie; Du, Shixue; Schneeloch, John; Zhong, Ruidan; Gu, Genda; Fu, Liang; Ding, Hong; Gao, Hong-Jun. Evidence for Majorana bound states in an iron-based superconductor. Science (2018), DOI 10.1126/science.aao1797; arXiv:1706.06074. https://arxiv.org/abs/1706.06074
  4. Buss, Maxwell; Bedow, Jasmin; Gupta, Prashant; Morr, Dirk K. Braiding of Majorana Zero Modes in Vortex Cores. arXiv:2504.16841 (2025). https://arxiv.org/abs/2504.16841
  5. Nothhelfer, Jonas; Díaz, Sebastián A.; Kessler, Stephan; Meng, Tobias; Rizzi, Matteo; Hals, Kjetil M. D.; Everschor-Sitte, Karin. Steering Majorana braiding via skyrmion-vortex pairs: a scalable platform. Phys. Rev. B 105, 224509 (2022), DOI 10.1103/PhysRevB.105.224509; arXiv:2110.13983. https://arxiv.org/abs/2110.13983
  6. Beenakker, C. W. J. Search for non-Abelian Majorana braiding statistics in superconductors. SciPost Phys. Lect. Notes 15 (2020), DOI 10.21468/SciPostPhysLectNotes.15; arXiv:1907.06497 (2019). https://arxiv.org/abs/1907.06497
  7. Wang, Dongfei; Ortuzar, Jon; Massee, Freek; Zhong, Ruidan; Gu, Genda; Xiao, Wende; Yao, Yugui; Wiesendanger, Roland. Distinguishing Majorana zero modes from trivial defect states in an iron-based superconductor. arXiv:2606.17499 (2026). https://arxiv.org/abs/2606.17499