REVIEW 6 minor 39 references
Optimising germanium hole spin qubits with a room-temperature magnet
T0 review · 0 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash
Pith's one-line read A permanent magnet outside the cryostat can align a germanium hole spin qubit's magnetic field in-plane.
desk verdict External room-temp magnet fine-tunes a Ge hole spin qubit to the in-plane sweet spot; the demonstration is solid, with screening and hysteresis acknowledged and not load-bearing. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is an NdFeB N45 block magnet, 110.6 × 89 × 19.5 mm, mounted on a remotely controlled XYZ gantry attached below the cryostat, whose position tunes the magnetic field seen by the qubit. It supplies a coherent correction field to the uniaxial solenoid's field; because the hole g-tensor is strongly anisotropic, the qubit's Larmor frequency acts as a sensitive readout of field orientation. The minimum of the Larmor frequency as a function of magnet x-position marks the in-plane condition, where the external field cancels the out-of-plane component of the internal field. All performance gains are tied to finding and sitting at that minimum: coherence times and Rabi drive efficiency peak simultaneously there, consistent with the heavy-hole hyperfine sweet spot.
What would settle it
Place a calibrated 3D magnetometer at the sample location inside the cryostat, or compare the qubit's Larmor-frequency map versus magnet position with the solenoid on and off at the same nominal total field; if the solenoid screens the external field appreciably, the field orientation inferred from the Larmor-frequency minimum will disagree with the measured vector field beyond the stated alignment uncertainty.
Extended reading notes
Core claim
The central claim is that a room-temperature permanent magnet located outside the cryostat can replace the field-shaping role of an in-cryostat vector magnet for germanium hole spin qubits. The total field at the qubit is the uniaxial superconducting solenoid's few-tens-of-mT field plus a correction field from a movable NdFeB block magnet; translating the external magnet along x changes the out-of-plane component. Where the measured Larmor frequency reaches a minimum, the external field cancels the solenoid's out-of-plane component, leaving the total field in the sample plane, which is the hyperfine-noise sweet spot for heavy holes. At that spot the qubit shows $T_2^* = 13.41 \pm 0.53$ μs, $T_2^H = 88.77 \pm 9.99$ μs, Clifford fidelity $F_C = 99.936 \pm 0.005$%, and native gate fidelity $99.980 \pm 0.002$%, comparable to vector-magnet operation. The authors also measure the qubit's Zeeman splitting with the internal magnet off, concluding that a small permanent magnet near the sample could eventually replace the superconducting solenoid entirely.
Load-bearing premise
The magnetic field at the qubit is assumed to be the unscreened sum of the measured external magnet field and the solenoid field; if the superconducting solenoid or other cryostat components significantly screen or distort the external field, the inferred in-plane alignment and the quoted sweet-spot quantities would need re-examination.
Editorial extensions
If this is right
- A uniaxial solenoid plus a movable room-temperature magnet is sufficient for high-fidelity single-qubit control; no superconducting vector magnet is required.
- Coherence times and Rabi drive efficiency peak at the same magnet position, so locating the Larmor-frequency minimum gives a one-step calibration for optimal operation.
- Removing the internal superconducting magnet frees base-temperature sample space, which the paper identifies as a route to integrating more control wiring and cryogenic control circuitry.
- The external magnet can be paired with a small 1 T permanent magnet placed in the cryostat, with the room-temperature stage providing fine orientation tuning.
- For isotopically purified germanium, the optimal field may point a few degrees out of plane, and the external magnet can supply that tilted alignment as well.
Reading between the lines
- The qubit's Larmor frequency is itself a precise in-situ magnetometer, so the same setup could be run closed-loop, using the qubit to steer the external magnet to the sweet spot without requiring the absolute field to be known.
- The measured stage hysteresis, about 50 µm per start-stop event, is far smaller than the millimetre-scale width of the Larmor-frequency minimum, so the sweet spot is robust; adding encoder feedback would make the same position reproducible across repeated runs and across qubits.
- If the uniaxial solenoid does screen the external field, the zero-internal-field measurements provide a calibration path: comparing frequency maps taken with the solenoid on and off would quantify the screening and sharpen the claim of full in-plane alignment.
- For isotopically purified germanium, where hyperfine noise is no longer dominant, the same external stage could intentionally set a small out-of-plane angle to trade off charge-noise sensitivity, qubit control, and uniformity across the array.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper demonstrates that a movable permanent magnet placed outside a dilution refrigerator can fine-tune the magnetic field at a germanium hole spin qubit operated with a uniaxial superconducting solenoid. In a hybrid mode, the authors map the qubit Larmor frequency versus external magnet position, identify a sweet spot where the field is inferred to be in-plane, and report T2* = 13.41 ± 0.53 µs, T2H = 88.77 ± 9.99 µs, a randomized benchmarking Clifford fidelity of 99.936 ± 0.005%, and GST fidelities above 99.8%. They also show that the qubit resonance frequency can be probed with the internal solenoid switched off, using only the external magnet. The central claim is that room-temperature magnets can achieve qubit performance comparable to vector magnets while freeing cryostat space.
Significance. The result is significant for the field of semiconductor spin qubits because it offers a practical alternative to large superconducting vector magnets, whose footprint inside the cryostat is a known scalability bottleneck. The demonstration is supported by multiple independent quality metrics (Ramsey, Hahn echo, Rabi efficiency, randomized benchmarking, and gate set tomography) with consistent error bars, and the data and analysis code are openly available. The identification of the sweet spot is empirical rather than reliant on absolute field calibration, so the acknowledged uncertainty in the external field magnitude due to possible solenoid screening does not undermine the central demonstration. The zero-field-resonance measurement, though not a full qubit-operations demonstration, supports the scalability claim as a proof of principle.
minor comments (6)
- [Section I (Introduction), last paragraph] The sentence 'we also demonstrate qubit control when the superconducting magnet is turned off' overstates what is presented in Section V, which reports microwave spectroscopy of the Larmor frequency only; no Rabi oscillations, Ramsey fringes, or gate operations are shown for this mode. Please reword to 'we probe the qubit resonance frequency'.
- [Section II and Figure 1d] The assumption of no screening by the uniaxial solenoid is acknowledged in the figure caption but is not mentioned when the ≈6.2 mT field value is used in the text; given that Section V later states the solenoid 'may screen the external magnetic field,' the main text should flag the uncertainty of the quoted field strength.
- [Section IV, coherence times] The comparison 'comparable with coherence times measured at the magnetic field sweet spot in a vector magnet [6]' would be more informative if the T2* and T2H values from Ref. [6] were quoted explicitly, since the reader has no quantitative benchmark.
- [Section IV, RB/GST paragraph] The text reports that GST gives 'average fidelities of 99.95±0.02% for the X90 gate and 99.88±0.02% for the Y90 gate,' so the Y90 gate is below the 99.9% figure quoted in the abstract for the randomized benchmarking Clifford fidelity; please state explicitly that the 99.9% claim refers to RB only and not to all gates.
- [Supplementary Note 2, Figure 7 caption] Use the micro sign in '4.23 ± 0.11 us' (also in the figure text) for consistency with the rest of the manuscript.
- [Section V, last paragraph] The prediction 'magnetic fields of several tens of millitesla at the sample can be created' after removing the solenoid is plausible but untested; consider labeling it as an extrapolation based on the free-space Hall-sensor map.
Circularity Check
No circularity found: the headline numbers are direct measurements, the sweet spot is located empirically by an x-position scan, and coherence/fidelity at that spot are independent observables; the only inference (Larmor minimum = in-plane field) is a stated physical modeling assumption backed by external g-tensor work.
full rationale
This paper makes no first-principles derivation claim; the headline quantities T2* = 13.41 ± 0.53 us, T2H = 88.77 ± 9.99 us, and FC = 99.936 ± 0.005% are directly measured observables reported with fit uncertainties, not predictions produced by fitting a model to a subset of the same data. The field sweet spot is found empirically by scanning the external magnet in x while tracking the Larmor frequency minimum; T2*, T2H, Rabi efficiency, RB, and GST are then measured at that position as independent observables, so the simultaneous coherence maximum is not forced by construction. The one inference that might look circular—identifying the Larmor minimum with an in-plane field—rests on the well-established g-tensor anisotropy of germanium holes, cited to external Refs [5,6,7,14] as well as to the same group's device characterization [12]; it is a modeling assumption that labels the empirically located optimum, not a fitted parameter renamed as a prediction. The unverified 'assuming no screening effects from the uniaxial magnet' assumption (Sec II) and the later note that the solenoid 'may screen the external magnetic field' (Sec V) concern absolute field calibration; they do not affect the relative scan that locates the optimum, and the conclusion that room-temperature magnets enable high-fidelity operation therefore does not reduce to these assumptions. Self-citations to Refs [9,12,15] supply device context and comparison points, but the central claim is also benchmarked against external work (e.g., Ref [6], a vector-magnet sweet-spot result) and by independent GST, so the cited prior results are not the sole load-bearing support. The acknowledged hysteresis and magnet-position imprecision (Supplementary Note 6) are reproducibility limitations, not circularity.
Assumptions & free parameters
assumptions (4)
- domain assumption The minimum of the qubit Larmor frequency as a function of external magnet position identifies the in-plane magnetic field orientation.
- domain assumption The external magnet's field at the qubit is well approximated by Hall-sensor measurements without accounting for screening by the superconducting magnet.
- domain assumption Hyperfine interaction with nuclear spins is the dominant dephasing mechanism for natural germanium, and an in-plane field minimizes it.
- domain assumption The internal solenoid field has a 2-3 degree out-of-plane misalignment, as estimated in previous work on this device.
Cite this review
Pith. "Pith review of Optimising germanium hole spin qubits with a room-temperature magnet." pith.science (2026). https://pith.science/paper/KOVWOL6N
@misc{pith2026250703390,
author = {Pith},
title = {Pith review of: Optimising germanium hole spin qubits with a room-temperature magnet},
year = {2026},
howpublished = {\url{https://pith.science/paper/KOVWOL6N}},
note = {Machine review of arXiv:2507.03390}
}
read the original abstract
Germanium spin qubits exhibit strong spin-orbit interaction, which allow for high-fidelity qubit control, but also provide a strong dependence on the magnetic field. Superconducting vector magnets are often used to minimize dephasing due to hyperfine interactions and to maximize spin control, but these compromise the sample space and thus challenge scalability. Here, we explore whether a permanent magnet outside the cryostat can be used as an alternative. Operating in a hybrid mode with an internal and external magnet, we find that we can fine-tune the magnetic field to an in-plane orientation. We obtain a qubit dephasing time T2*=13 microseconds, Hahn-echo times T2H=88 microseconds, and an average single-qubit Clifford gate fidelity above 99.9%, from which we conclude that room temperature magnets allow for high qubit performance. Furthermore, we probe the qubit resonance frequency using only the external magnet, with the internal superconducting magnet switched off. Our approach may be used to scale semiconductor qubits and use the increased sample space for the integration of cryogenic control circuitry and wiring to advance to large-scale quantum processors.
Figures
Figures from the paper (7 more)
Reference graph
Works this paper leans on
-
[2]
H. Watzinger, J. Kukuˇ cka, L. Vukuˇ si´ c, F. Gao, T. Wang, 6 F. Sch¨ affler, J. J. Zhang, and G. Katsaros, A germanium hole spin qubit, Nature Communications 9, 1 (2018)
work page 2018
-
[3]
N. W. Hendrickx, W. I. Lawrie, M. Russ, F. van Riggelen, S. L. de Snoo, R. N. Schouten, A. Sammak, G. Scappucci, and M. Veldhorst, A four-qubit germanium quantum pro- cessor, Nature 591, 580 (2021)
work page 2021
-
[4]
C. A. Wang, V. John, H. Tidjani, C. X. Yu, A. S. Ivlev, C. D´ eprez, F. van Riggelen-Doelman, B. D. Woods, N. W. Hendrickx, W. I. Lawrie, L. E. Stehouwer, S. D. Oosterhout, A. Sammak, M. Friesen, G. Scappucci, S. L. de Snoo, M. Rimbach-Russ, F. Borsoi, and M. Veldhorst, Operating semiconductor quantum processors with hop- ping spins, Science 385, 447 (2024)
work page 2024
-
[6]
N. W. Hendrickx, L. Massai, M. Mergenthaler, F. J. Schupp, S. Paredes, S. W. Bedell, G. Salis, and A. Fuhrer, Sweet-spot operation of a germanium hole spin qubit with highly anisotropic noise sensitivity, Nature Mate- rials 23, 920 (2024)
work page 2024
- [7]
-
[8]
J. Saez-Mollejo, D. Jirovec, Y. Schell, J. Kukucka, S. Calcaterra, D. Chrastina, G. Isella, M. Rimbach-Russ, S. Bosco, and G. Katsaros, Exchange anisotropies in microwave-driven singlet-triplet qubits, Nature Commu- nications 16, 1 (2025)
work page 2025
-
[16]
J. P. Dehollain, J. T. Muhonen, R. Blume-Kohout, K. M. Rudinger, J. K. Gamble, E. Nielsen, A. Laucht, S. Sim- mons, R. Kalra, A. S. Dzurak, and A. Morello, Optimiza- tion of a solid-state electron spin qubit using gate set tomography, New Journal of Physics 18, 103018 (2016)
work page 2016
-
[21]
E. A. Chekhovich, M. N. Makhonin, A. I. Tartakovskii, A. Yacoby, H. Bluhm, K. C. Nowack, and L. M. Vander- sypen, Nuclear spin effects in semiconductor quantum dots, Nature materials 12, 494 (2013)
work page 2013
Show all 39 references
-
[22]
X. Xue, T. F. Watson, J. Helsen, D. R. Ward, D. E. Savage, M. G. Lagally, S. N. Coppersmith, M. A. Eriks- son, S. Wehner, and L. M. Vandersypen, Benchmarking Gate Fidelities in a Si/SiGe Two-Qubit Device, Physical Review X 9, 021011 (2019)
2019
-
[23]
Nielsen, J
E. Nielsen, J. K. Gamble, K. Rudinger, T. Scholten, K. Young, and R. Blume-Kohout, Gate Set Tomography, Quantum 5, 557 (2021). 7
2021
-
[24]
This qubit demonstrates the same trend in Larmor frequency and coherence times as a function of the magnet x-position
ADDITIONAL QUBIT DA T A Here we plot the coherence times as a function of the magnet position along x direction for a qubit Q3 that located on the east side of the device, see Figure-1c in the main text. This qubit demonstrates the same trend in Larmor frequency and coherence ...
-
[25]
We estimate that here the magnetic field points 2-3 degrees out-of-plane and the qubit coherence times are expected to be hyperfine limited [9, 12]
QUBIT COHERENCE TIMES WITHOUT THE EXTERNAL MAGNET In Figure 7, we measure the coherence times of qubit Q8 by operating with the internal magnet set to 25 mT while the external magnet is 700 mm away from the sample such that we assume it has no effect on the total magnetic fiel...
-
[26]
−50 −40 −30 −20 −10 x (mm) 1.0 1.1 1.2 1.3 1.4 1.5 Driving efficiency MHz/(MHz·V) 2.5 5.0 7.5 10.0 12.5 T ∗ 2 ( s) FIG
DRIVING EFFICIENCY Here we plot the Rabi drive efficiency and the dephasing time T ∗ 2 of Q8 as a function of the external magnet position in the x direction. −50 −40 −30 −20 −10 x (mm) 1.0 1.1 1.2 1.3 1.4 1.5 Driving efficiency MHz/(MHz·V) 2.5 5.0 7.5 10.0 12.5 T ∗ 2 ( s) FIG...
-
[27]
In particular, we randomly sample Clifford gates, composed from the native X90 and Y90 gates according to Table I
RANDOMISED BENCHMARKING In this work we perform single qubit RB in an identical fashion to that described in [15]. In particular, we randomly sample Clifford gates, composed from the native X90 and Y90 gates according to Table I. We then append a recovery gate, taken to be the...
-
[28]
We then use pyGSTi to fit the resulting data
GA TE SET TOMOGRAPHY To perform GST, we use the XY model implemented by pyGSTi [23], with a max circuit length of 128 gates. We then use pyGSTi to fit the resulting data. The table of individual error metrics generated in the report is shown in Table II. The Pauli transfer mat...
-
[29]
The movement of the magnet is non-linear, which introduces hysteresis when repeating runs involving multiple start- and-stop motions
HYSTERESIS EFFECT IN THE MAGNET MOVEMENT Here, we quantify the hysteresis effect observed in the magnet motion on three different qubits Q1, Q4 and Q8. The movement of the magnet is non-linear, which introduces hysteresis when repeating runs involving multiple start- and-stop ...
-
[30]
Burkard, T
G. Burkard, T. D. Ladd, A. Pan, J. M. Nichol, and J. R. Petta, Semiconductor spin qubits, Reviews of Modern Physics 95, 025003 (2023)
2023
-
[31]
Watzinger, J
H. Watzinger, J. Kukuˇ cka, L. Vukuˇ si´ c, F. Gao, T. Wang, F. Sch¨ affler, J. J. Zhang, and G. Katsaros, A germanium hole spin qubit, Nature Communications 9, 1 (2018)
2018
-
[32]
N. W. Hendrickx, W. I. Lawrie, M. Russ, F. van Riggelen, S. L. de Snoo, R. N. Schouten, A. Sammak, G. Scappucci, and M. Veldhorst, A four-qubit germanium quantum processor, Nature 591, 580 (2021)
2021
-
[33]
C. A. Wang, V. John, H. Tidjani, C. X. Yu, A. S. Ivlev, C. D´ eprez, F. van Riggelen-Doelman, B. D. Woods, N. W. Hendrickx, W. I. Lawrie, L. E. Stehouwer, S. D. Oosterhout, A. Sammak, M. Friesen, G. Scappucci, S. L. de Snoo, M. Rimbach-Russ, F. Borsoi, and M. Veldhorst, Operat...
2024
-
[34]
Scappucci, C
G. Scappucci, C. Kloeffel, F. A. Zwanenburg, D. Loss, M. Myronov, J. J. Zhang, S. De Franceschi, G. Katsaros, and M. Veldhorst, The germanium quantum information route, Nature Reviews Materials 6, 926 (2020)
2020
-
[35]
N. W. Hendrickx, L. Massai, M. Mergenthaler, F. J. Schupp, S. Paredes, S. W. Bedell, G. Salis, and A. Fuhrer, Sweet-spot operation of a germanium hole spin qubit with highly anisotropic noise sensitivity, Nature Materials 23, 920 (2024)
2024
-
[36]
Mauro, E
L. Mauro, E. A. Rodr ´ ıguez-Mena, M. Bassi, V. Schmitt, and Y. M. Niquet, Geometry of the dephasing sweet spots of spin-orbit qubits, Physical Review B 109, 155406 (2024)
2024
-
[37]
Saez-Mollejo, D
J. Saez-Mollejo, D. Jirovec, Y. Schell, J. Kukucka, S. Calcaterra, D. Chrastina, G. Isella, M. Rimbach-Russ, S. Bosco, and G. Katsaros, Exchange anisotropies in microwave-driven singlet-triplet qubits, Nature Communications 16, 1 (2025)
2025
-
[38]
V. John, C. X. Yu, B. Van Straaten, E. A. Rodr ´ ıguez-Mena, M. Rodr ´ ıguez, S. Oosterhout, L. E. A. Stehouwer, G. Scappucci, S. Bosco, M. Rimbach-Russ, Y.-M. Niquet, F. Borsoi, and M. Veldhorst, A two-dimensional 10-qubit array in germanium with robust and localised qubit co...
2024
-
[39]
Bassi, E
M. Bassi, E. A. Rodr ´ ıguez-Mena, B. Brun, S. Zihlmann, T. Nguyen, V. Champain, J. C. Abadillo-Uriel, B. Bertrand, H. Niebojewski, R. Maurand, Y.-M. Niquet, X. Jehl, S. De Franceschi, and V. Schmitt, Optimal operation of hole spin qubits, arXiv 2412.13069 (2024)
2024
-
[40]
Da Prato, Y
G. Da Prato, Y. Yu, R. Bode, and S. Gr¨ oblacher, Step-by-step design guide of a cryogenic three-axis vector magnet, arXiv 2503.05459 (2025)
2025 arXiv
-
[41]
L. E. A Stehouwer, C. X. Yu, B. van Straaten, A. Tosato, V. John, D. Degli Esposti, A. Elsayed, D. Costa, S. D. Oosterhout, N. W. Hendrickx, M. Veldhorst, F. Borsoi, and G. Scappucci, Exploiting epitaxial strained germanium for scaling low noise spin qubits at the micron-scale...
2025 arXiv
-
[42]
Jirovec, P
D. Jirovec, P. M. Mutter, A. Hofmann, A. Crippa, M. Rychetsky, D. L. Craig, J. Kukucka, F. Martins, A. Ballabio, N. Ares, D. Chrastina, G. Isella, G. Burkard, and G. Katsaros, Dynamics of Hole Singlet-Triplet Qubits with Large g -Factor Differences, Physical Review Letters 128...
2022
-
[43]
Fischer, W
J. Fischer, W. A. Coish, D. V. Bulaev, and D. Loss, Spin decoherence of a heavy hole coupled to nuclear spins in a quantum dot, Physical Review B - Condensed Matter and Materials Physics 78, 155329 (2008)
2008
-
[44]
W. I. Lawrie, M. Rimbach-Russ, F. v. Riggelen, N. W. Hendrickx, S. L. Snoo, A. Sammak, G. Scappucci, J. Helsen, and M. Veldhorst, Simultaneous single-qubit driving of semiconductor spin qubits at the fault-tolerant threshold, Nature Communications 14, 1 (2023)
2023
-
[45]
J. P. Dehollain, J. T. Muhonen, R. Blume-Kohout, K. M. Rudinger, J. K. Gamble, E. Nielsen, A. Laucht, S. Simmons, R. Kalra, A. S. Dzurak, and A. Morello, Optimization of a solid-state electron spin qubit using gate set tomography, New Journal of Physics 18, 103018 (2016)
2016
-
[46]
Mohseni, A
M. Mohseni, A. Scherer, K. G. Johnson, O. Wertheim, M. Otten, N. A. Aadit, Y. Alexeev, K. M. Bresniker, K. Y. Camsari, B. Chapman, S. Chatterjee, G. A. Dagnew, A. Esposito, F. Fahim, M. Fiorentino, A. Gajjar, A. Khalid, X. Kong, B. Kulchytskyy, E. Kyoseva, R. Li, P. A. Lott, I...
2024 arXiv
-
[47]
Adambukulam, V
C. Adambukulam, V. K. Sewani, H. G. Stemp, S. Asaad, M. T. M¸ adzik, A. Morello, and A. Laucht, An ultra-stable 1.5 T permanent magnet assembly for qubit experiments at cryogenic temperatures, Review of Scientific Instruments 92, 10.1063/5.0055318 (2021)
2021 doi
-
[48]
Rooney, Z
J. Rooney, Z. Luo, L. E. A. Stehouwer, G. Scappucci, M. Veldhorst, and H.-W. Jiang, Gate modulation of the hole singlet-triplet qubit frequency in germanium, npj Quantum Information 11, 1 (2025)
2025
-
[49]
Moutanabbir, S
O. Moutanabbir, S. Assali, A. Attiaoui, G. Daligou, P. Daoust, P. D. Vecchio, S. Koelling, L. Luo, and N. Rotaru, Nuclear Spin-Depleted, Isotopically Enriched 70Ge/28Si70Ge Quantum Wells, Advanced Materials 36, 10.1002/adma.202305703 (2024)
2024 doi
-
[50]
E. A. Chekhovich, M. N. Makhonin, A. I. Tartakovskii, A. Yacoby, H. Bluhm, K. C. Nowack, and L. M. Vandersypen, Nuclear spin effects in semiconductor quantum dots, Nature materials 12, 494 (2013)
2013
-
[51]
X. Xue, T. F. Watson, J. Helsen, D. R. Ward, D. E. Savage, M. G. Lagally, S. N. Coppersmith, M. A. Eriksson, S. Wehner, and L. M. Vandersypen, Benchmarking Gate Fidelities in a Si/SiGe Two-Qubit Device, Physical Review X 9, 021011 (2019)
2019
-
[52]
Nielsen, J
E. Nielsen, J. K. Gamble, K. Rudinger, T. Scholten, K. Young, and R. Blume-Kohout, Gate Set Tomography, Quantum 5, 557 (2021)
2021
Reviewed August 6, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.