REVIEW 4 major objections 5 minor 1 cited by
Discovery of Niobium Hydride Precipitates in Superconducting Qubits
T0 review · 4 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash
Pith's one-line read Niobium hydride precipitates form in superconducting qubit films and act as a previously unaccounted source of decoherence.
desk verdict Plausible case for niobium hydrides in qubit films, but the structural evidence rests on a single XRD peak on a surrogate film—worth refereeing, not a slam dunk. 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 load-bearing object is the niobium hydride precipitate, indexed as orthorhombic NbH0.89 with a (111) spacing d = 2.45 Å in space group Pnnn. It is a normal or weak-gap inclusion embedded in the superconducting niobium film and does its damage through proximity coupling: microwave fields experience a locally suppressed superconducting order parameter, and the inclusions act as quasiparticle sinks that increase dissipation and can trigger the high-field Q slope. The paper also uses a 1.3 GHz bulk niobium cavity with controlled hydrogen loading as a clean single-interface proxy to quantify how much RF loss a given hydrogen concentration produces.
What would settle it
Cryogenic transmission electron microscopy or scanning nano-X-ray diffraction on an actual qubit chip cooled below 150 K should resolve multiple NbH0.89 lattice reflections and precipitates whose number density scales with locally measured hydrogen content and inversely with measured T1. If the only hydride fingerprint remains the single 2.45 Å reflection, or if no precipitates are found inside the device film, the structural identification is not supported.
Extended reading notes
Core claim
On the basis of cryogenic AFM, synchrotron grazing-incidence X-ray diffraction, and ToF-SIMS, the paper asserts that niobium films on silicon in superconducting qubits can precipitate nanocrystalline niobium hydride, indexed as orthorhombic NbH0.89, during cooldown beginning below about 150 K. The precipitates are non-superconducting inclusions; proximity coupling between them and the surrounding superconducting niobium degrades microwave performance. The same physics is demonstrated semiquantitatively in a 3D bulk niobium cavity whose hydrogen content was increased by mechanical grinding and electrochemical loading: roughly a 2.5-fold rise in hydrogen concentration produced an order-of-magn
Load-bearing premise
The single weak X-ray reflection at d = 2.45 Å, seen only between about 100 and 150 K and assigned to the (111) plane of orthorhombic NbH0.89, is genuinely niobium hydride, and it is representative of the actual qubit film rather than only a co-fabricated witness film.
Editorial extensions
If this is right
- Cooldown-to-cooldown variation in qubit coherence can result from hydride precipitates that change size, morphology, and location on each thermal cycle.
- Hydride regions have a suppressed superconducting gap and act as quasiparticle sinks, so even low-power operation suffers dissipation through the proximity effect.
- Because hydrides form throughout the film thickness, changing film thickness alone will not eliminate this decoherence channel.
- Hydrogen introduced during deposition or during wet and mechanical processing steps can be retained under the surface oxide and later precipitate, making fabrication controls a direct lever on qubit performance.
- Introducing higher oxygen concentration into niobium films may trap hydrogen and suppress hydride formation, a practical mitigation route proposed by the paper.
Reading between the lines
- If the identification holds, hydrogen should be treated as a first-class materials variable in qubit fabrication: controlled hydrogen-concentration measurements on actual devices would help separate cooldown-dependent hydride loss from stationary two-level-system loss.
- A natural test would be to fabricate matched qubit chips with deliberately different hydrogen loads and track T1 and T2 across repeated cooldowns; the bulk-resonator data predict a monotone, cooldown-history-dependent degradation.
- The mechanism suggests that 'aging' effects and sporadic coherence jumps may have a reversible microscopic cause—re-cooling into a different hydride configuration—rather than permanent damage, which cycling experiments could distinguish.
- Because the authors note their bulk-cavity experiment cannot assess two-level-system losses from hydrides, extending the same hydrogen-loading protocol to film-based resonators would determine how much of qubit TLS loss overlaps with the hydride channel.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports evidence for niobium hydride precipitates in HiPIMS-deposited Nb films on Si substrates in superconducting qubit chips fabricated at Rigetti Computing. The evidence is drawn from room-temperature and cryogenic AFM showing surface features that appear on cooling, ToF-SIMS depth profiles showing hydrogen enrichment beneath the native oxide, synchrotron GIXRD on a co-fabricated film showing a weak reflection at d = 2.45 Å assigned to NbH0.89 (111), and RF measurements on a 3D bulk Nb cavity after deliberate hydrogen loading showing degradation of the quality factor. The central claim is that these hydrides are a previously unidentified source of decoherence in superconducting qubits, contributing to quasiparticle and possibly TLS losses and explaining cooldown-to-cooldown variability.
Significance. If the central claim holds, the paper identifies a practically important decoherence mechanism for planar superconducting qubits, connecting established bulk SRF cavity hydride physics to the 2D qubit geometry. It proposes a concrete mitigation pathway (oxygen doping) and provides a plausible explanation for qubit performance variability across cooldowns. The study has notable strengths: it uses actual Rigetti qubit chips for AFM and SIMS, acquires temperature-dependent cryo-AFM and synchrotron XRD data, and leverages a controlled bulk-cavity hydrogen-loading experiment to quantify RF loss. The authors also explicitly acknowledge several limitations, including that their bulk study does not address TLS losses and that further film-based resonator studies are needed. These candid statements are to their credit, but they also delimit the strength of the conclusions as currently stated.
major comments (4)
- [§II, Fig. 3] The entire crystallographic identification rests on a single weak reflection at d = 2.45 Å, assigned to the (111) plane of orthorhombic NbH0.89. The text explicitly states that no other hydride reflections were observed. The figure shows no error bars, peak intensity, signal-to-noise ratio, or detection limits, and the Rietveld refinement is mentioned without parameters or residuals. Alternative assignments (e.g., NbO, Nb2O5, NbC, strain-modified Nb reflections) must be quantitatively excluded. In addition, the XRD was performed on a co-fabricated film, not on the actual Rigetti qubit chip; the representativeness of this film for the qubit Nb pads needs justification. As written, this one peak carries the structural identification, which is a load-bearing point for the paper's central claim.
- [§II, Fig. 1 caption] The AFM features on the actual qubit chip are labeled 'unknown structures' in the caption and are not chemically identified. Their interpretation as niobium hydrides relies solely on morphological similarity to features previously observed in bulk Nb SRF cavities. Since the XRD was not performed on this chip, the connection between the AFM features and the hydride phase is circumstantial. Direct compositional or structural evidence on the same features (e.g., TEM/EELS, EDX, or nano-SIMS) is needed to substantiate the claim that these are hydrides. If such evidence is not available, the text should explicitly state that the AFM features are consistent with, but not proof of, hydride precipitates.
- [§II, ToF-SIMS (Fig. 2)] The ToF-SIMS depth profile shows hydrogen enrichment beneath the Nb2O5 layer, but hydrogen in interstitial solid solution cannot be distinguished from hydrogen in a hydride phase by this measurement. The statement that this hydrogen 'potentially drives Nb hydride formation' is a reasonable hypothesis but is not tested. No spatial correlation is shown between the H-enriched regions and the AFM features, and no estimate is given for whether the measured H concentration exceeds the solubility limit at the temperatures where the features appear. This limits the directness of the chemical evidence.
- [§II, Fig. 4 and §III Discussion] The bulk 3D Nb cavity data demonstrate that externally introduced hydrogen degrades Q0, but the quantitative extrapolation to qubit films is not derived. The abstract states that the authors 'quantify RF dissipation on a superconducting qubit' from these bulk-cavity measurements, yet the geometry, field regime, and dimensionality differ (bulk cavity vs. 100–200 nm film; <1 MV/m vs. single-photon powers). No model connects the measured H concentration and Q0 reduction to a hydride volume fraction or to qubit T1/T2. The paper itself concedes that the bulk study does not address TLS losses and that film-based resonator tests are needed. The wording of the abstract and conclusion should be tempered to reflect that the RF quantification is on a bulk model system, not on a qubit.
minor comments (5)
- [Abstract and §III] The phrase 'quantify RF dissipation on a superconducting qubit' overstates what was measured; consider saying 'on a bulk Nb resonator as a model system for qubit-relevant RF dissipation.'
- [§II, GIXRD setup] The text says 'X-ray penetration depth (λ = 0.68 Å)'; λ is the wavelength, not the penetration depth. Please correct and report the estimated penetration depth separately.
- [§II, Fig. 3] The shift of the hydride peak by ~0.02 Å between cooldowns is reported without an uncertainty estimate. Provide the fit uncertainty and the temperature at which each measurement was taken.
- [§II, Fig. 1] The AFM features are called 'unknown structures' in the caption but 'hydride-related' in the text. This inconsistency should be resolved once the features are identified or explicitly labeled as presumed hydrides.
- [General] There is a typo: 'Regetti qubit device' should be 'Rigetti qubit device.'
Circularity Check
No significant circularity: the hydride identification rests on new XRD, AFM, and SIMS measurements, and the RF-dissipation argument is an extrapolation from an independent bulk-cavity experiment, not a fitted prediction.
full rationale
The paper's central claim is that Nb hydride precipitates form in Rigetti qubit Nb films. The evidence chain is: (1) cryo-AFM observes surface features on the actual chip; (2) GIXRD on a co-fabricated film shows a peak at d=2.45 Å assigned by standard crystallography to NbH0.89 (111); (3) ToF-SIMS shows H accumulation near the Nb surface; (4) a bulk Nb cavity with controlled H loading shows Q0 degradation. None of these steps defines the conclusion in terms of an input or fits a parameter and then 'predicts' a closely related quantity. The XRD assignment uses an external reference pattern, not the authors' prior work. The interpretation of AFM features as hydrides invokes the authors' earlier bulk-Nb direct observation [31], but that is an independent experimental result, not an unverified self-citation or a uniqueness theorem. The bulk-cavity RF measurement is a separate experiment on a different geometry; extrapolating it to qubit films is an interpretation (flagged by the authors as requiring further Nb-film resonator work), not a circular reduction. No self-definitional, fitted-input, ansatz-smuggling, or renaming pattern is present. The paper's weaknesses—single XRD reflection, co-fabricated film, uncharacterized AFM structures—are evidence-sufficiency concerns, not circularity. Therefore the circularity score is 0.
Assumptions & free parameters
assumptions (5)
- domain assumption Nb hydrides are non-superconducting and suppress superconductivity in surrounding Nb via the proximity effect.
- domain assumption The weak XRD reflection at d = 2.45 Å observed between 100 and 150 K corresponds to NbH0.89 (111) in the Pnnn space group.
- domain assumption Surface topographical features seen by cryo-AFM are due to Nb hydride precipitates.
- ad hoc to paper The HiPIMS Nb film used for GIXRD, co-fabricated with the qubit device, is representative of the actual Rigetti qubit chips.
- ad hoc to paper Microwave dissipation measured in a 3D bulk Nb cavity quantifies RF loss from hydrides in 2D qubit Nb films.
Cite this review
Pith. "Pith review of Discovery of Niobium Hydride Precipitates in Superconducting Qubits." pith.science (2026). https://pith.science/paper/QZWRAGET
@misc{pith2026250810889,
author = {Pith},
title = {Pith review of: Discovery of Niobium Hydride Precipitates in Superconducting Qubits},
year = {2026},
howpublished = {\url{https://pith.science/paper/QZWRAGET}},
note = {Machine review of arXiv:2508.10889}
}
read the original abstract
We report the evidence of the formation of niobium hydride phase within niobium films on silicon substrates in superconducting qubits fabricated at Rigetti Computing. For this study, we combined complementary techniques, including room-temperature and cryogenic atomic force microscopy (AFM), synchrotron Xray diffraction, and time of flight secondary ion mass spectroscopy (ToF-SIMS), to directly reveal the existence of niobium hydride precipitates in the Rigetti chip area. Upon cryogenic cooling, we observed variation in the size and morphology of the hydrides, ranging from small (5 nm) irregular shapes to large (~10-100 nm) domain within the Nb grains, fully converted to niobium hydrides. Since niobium hydrides are non-superconducting and can easily change in size and location upon different cooldowns to cryogenic temperature, our finding highlights a new and previously unknown source of decoherence in superconducting qubits. This contributes to both quasiparticle and two level system (TLS) losses, offering a potential explanation for changes in qubit performance upon cooldowns. Finally, by leveraging the RF performance of a 3D bulk Nb resonator, we can quantify RF dissipation on a superconducting qubit, caused by hydrogen concentration variation, and are able to propose a practical engineering pathway to mitigate the formation of the Nb hydrides for superconducting qubit applications.
Figures
Forward citations
Cited by 1 Pith paper
-
Coulomb blockade in microscopic material defects as a source of decoherence and noise in solid-state quantum circuits
Metallic grains in thin-film superconducting circuits cause microwave-driven Coulomb-blockade dissipation, a newly identified decoherence mechanism as common as two-level-system defects.
Reference graph
Works this paper leans on
-
[1]
M. A. Tanatar et al., Anisotropic Superconductivity of Niobium Based on Its Response to Non-Magnetic Disorder, arXiv:2207.14395
-
[2]
K. R. Joshi et al., Quasiparticle Spectroscopy, Transport, and Magnetic Properties of Nb Films Used in Superconducting Transmon Qubits, arXiv:2207.11616
-
[3]
Premkumar et al., Microscopic relaxation channels in materials for superconducting qubits, Commun
A. Premkumar et al., Microscopic relaxation channels in materials for superconducting qubits, Commun. Mater. 2, 72 (2021)
work page 2021
- [4]
-
[5]
Zmuidzinas, Superconducting Microresonators: Physics and Applications, Annu
J. Zmuidzinas, Superconducting Microresonators: Physics and Applications, Annu. Rev. Condens. Matter Phys. 3, 169 (2012)
work page 2012
-
[6]
J. M. Martinis et al., Decoherence in Josephson Qubits from Dielectric Loss, Phys. Rev. Lett. 95, 210503 (2005)
work page 2005
-
[7]
McDermott, Materials Origins of Decoherence in Superconducting Qubits, IEEE Trans
R. McDermott, Materials Origins of Decoherence in Superconducting Qubits, IEEE Trans. Appl. Supercond. 19, 2 (2009)
work page 2009
-
[8]
D. P. Pappas, M. R. Vissers, D. S. Wisbey, J. S. Kline, and J. Gao, Two Level System Loss in Superconducting Microwave Resonators, IEEE Trans. Appl. Supercond. 21, 871 (2011)
work page 2011
Show all 47 references
-
[9]
A. A. Murthy, J. Lee, C. Kopas, M. J. Reagor, A. P. McFadden, D. P. Pappas, M. Checchin, A. Grassellino, and A. Romanenko, TOF-SIMS analysis of decoherence sources in superconducting qubits, Appl. Phys. Lett. 120, 044002 (2022)
2022
-
[10]
Bafia, A
D. Bafia, A. Murthy, A. Grassellino, and A. Romanenko, Oxygen vacancies in niobium pentoxide as a source of two-level system losses in superconducting niobium, Phys. Rev. Appl. 22, 024035 (2024)
2024
-
[11]
M. V. P. Altoé et al., Localization and Reduction of Superconducting Quantum Coherent Circuit Losses, https://arxiv.org/abs/2012.07604v1
2012 arXiv
-
[12]
Romanenko and D
A. Romanenko and D. I. Schuster, Understanding Quality Factor Degradation in Superconducting Niobium Cavities at Low Microwave Field Amplitudes, Phys. Rev. Lett. 119, 264801 (2017)
2017
-
[13]
Megrant et al., Planar superconducting resonators with internal quality factors above one million, Appl
A. Megrant et al., Planar superconducting resonators with internal quality factors above one million, Appl. Phys. Lett. 100, 113510 (2012)
2012
-
[14]
C. T. Earnest, J. H. Béjanin, T. G. McConkey, E. A. Peters, A. Korinek, H. Yuan, and M. Mariantoni, Substrate surface engineering for high-quality silicon/aluminum superconducting resonators, Supercond. Sci. Technol. 31, 125013 (2018)
2018
-
[15]
Wenner et al., Surface loss simulations of superconducting coplanar waveguide resonators, Appl
J. Wenner et al., Surface loss simulations of superconducting coplanar waveguide resonators, Appl. Phys. Lett. 99, 113513 (2011)
2011
-
[16]
Romanenko, R
A. Romanenko, R. Pilipenko, S. Zorzetti, D. Frolov, M. Awida, S. Belomestnykh, S. Posen, and A. Grassellino, Three-Dimensional Superconducting Resonators at T < 20 mK with Photon Lifetimes up to τ = 2 s, Phys. Rev. Appl. 13, 034032 (2020)
2020
-
[17]
Serniak, M
K. Serniak, M. Hays, G. de Lange, S. Diamond, S. Shankar, L. D. Burkhart, L. Frunzio, M. Houzet, and M. H. Devoret, Hot Nonequilibrium Quasiparticles in Transmon Qubits, Phys. Rev. Lett. 121, 157701 (2018)
2018
-
[18]
A. P. Vepsäläinen et al., Impact of ionizing radiation on superconducting qubit coherence, Nature 584, 551 (2020)
2020
-
[19]
C. D. Wilen et al., Correlated charge noise and relaxation errors in superconducting qubits, Nature 594, 369 (2021)
2021
-
[20]
Barkov, A
F. Barkov, A. Romanenko, Y. Trenikhina, and A. Grassellino, Precipitation of hydrides in high purity niobium after different treatments, J. Appl. Phys. 114, 164904 (2013)
2013
-
[21]
Barkov, A
F. Barkov, A. Romanenko, and A. Grassellino, Direct observation of hydrides formation in cavity-grade niobium, Phys. Rev. Spec. Top. - Accel. Beams 15, 122001 (2012)
2012
-
[22]
Romanenko, F
A. Romanenko, F. Barkov, L. D. Cooley, and A. Grassellino, Proximity breakdown of hydrides in superconducting niobium cavities, Supercond. Sci. Technol. 26, 035003 (2013)
2013
-
[23]
Trenikhina, A
Y. Trenikhina, A. Romanenko, J. Kwon, J.-M. Zuo, and J. F. Zasadzinski, Nanostructural features degrading the performance of superconducting radio frequency niobium cavities revealed by transmission electron microscopy and electron energy loss spectroscopy, J. Appl. Phys. 117,...
2015
-
[24]
Schober and H
T. Schober and H. Wenzl, The Systems NbH(D), TaH(D), VH(D) : Structures, Phase Diagrams, Morphologies, Methods of Preparation, in Hydrogen in Metals II, edited by G. Alefeld and J. Völkl, Vol. 29 (Springer Berlin Heidelberg, Berlin, Heidelberg, 1978), pp. 11–71
1978
-
[25]
C. G. Torres‐Castanedo et al., Formation and Microwave Losses of Hydrides in Superconducting Niobium Thin Films Resulting †zsung@fnal.gov from Fluoride Chemical Processing, Adv. Funct. Mater. 34, 2401365 (2024)
2024
-
[26]
Hauck, Ordering of hydrogen in niobium hydride phases, Acta Crystallogr
J. Hauck, Ordering of hydrogen in niobium hydride phases, Acta Crystallogr. Sect. A 33, 208 (1977)
1977
-
[27]
Makenas and H
B. Makenas and H. K. Birnbaum, On the techniques for preparing refractory metal — Hydrogen alloys for T.E.M. studies, Scr. Metall. 11, 699 (1977)
1977
-
[28]
Isagawa, Hydrogen absorption and its effect on low‐temperature electric properties of niobium, J
S. Isagawa, Hydrogen absorption and its effect on low‐temperature electric properties of niobium, J. Appl. Phys. 51, 4460 (1980)
1980
-
[29]
Q Disease
J. Knobloch, The “Q Disease” in Superconducting Niobium RF Cavities, in AIP Conference Proceedings, Vol. 671 (AIP, Newport News, Virginia (USA), 2003), pp. 133– 150
2003
-
[30]
Nersisyan et al., Manufacturing Low Dissipation Superconducting Quantum Processors, in 2019 IEEE International Electron Devices Meeting (IEDM) (2019), p
A. Nersisyan et al., Manufacturing Low Dissipation Superconducting Quantum Processors, in 2019 IEEE International Electron Devices Meeting (IEDM) (2019), p. 31.1.1- 31.1.4
2019
-
[31]
Z. Sung, A. Cano, A. Murthy, D. Bafia, E. Karapetrova, M. Martinello, J. Lee, A. Grassellino, and A. Romanenko, Direct observation of nanometer size hydride precipitations in superconducting niobium, Sci. Rep. 14, 26916 (2024)
2024
-
[32]
Riffaud, M.-C
J. Riffaud, M.-C. Lépy, Y. Ménesguen, and A. Novikova, Measurement of K fluorescence yields of niobium and rhodium using monochromatic radiation, X-Ray Spectrom. 46, 341 (2017)
2017
-
[33]
L. G. Parratt, Surface Studies of Solids by Total Reflection of X-Rays, Phys. Rev. 95, 359 (1954)
1954
-
[34]
Melnychuk, A
O. Melnychuk, A. Grassellino, and A. Romanenko, Error analysis for intrinsic quality factor measurement in superconducting radio frequency resonators, Rev. Sci. Instrum. 85, 124705 (2014)
2014
-
[35]
H. H. Shen, X. T. Zu, B. Chen, C. Q. Huang, and K. Sun, Direct observation of hydrogenation and dehydrogenation of a zirconium alloy, J. Alloys Compd. 659, 23 (2016)
2016
-
[36]
Higuchi, K
T. Higuchi, K. Saito, Y. Yamazaki, T. Ikeda and S. Ohgushi, HYDROGENQ- DISEASEANDELECTROPOLISHING, The 10th Workshop on RF Superconductivity, 2001, Tsukuba, Japan pr02, (n.d.)
2001
-
[37]
Faber and H
K. Faber and H. Schultz, Hydrogen contamination in tantalum and niobium following UHV-degassing, Scr. Metall. 6, 1065 (1972)
1972
-
[38]
Higuchi, G
T. Higuchi, G. U. A. Studies, and N. P. C. Ltd, Centrifugal Barrel Polishing of L-band Niobium cavities, 2 (2001)
2001
-
[39]
Balachandran, A
S. Balachandran, A. Polyanskii, S. Chetri, P. Dhakal, Y.-F. Su, Z.-H. Sung, and P. J. Lee, Direct evidence of microstructure dependence of magnetic flux trapping in niobium, Sci. Rep. 11, 5364 (2021)
2021
-
[40]
A. H. Cottrell and B. A. Bilby, Dislocation Theory of Yielding and Strain Ageing of Iron, Proc. Phys. Soc. Sect. A 62, 49 (1949)
1949
-
[41]
J. A. Rodrigues and R. Kirchheim, More evidence for the formation of a dense cottrell cloud of hydrogen (hydride) at dislocations in niobium and palladium, Scr. Metall. 17, 159 (1983)
1983
-
[42]
Visentin, M
B. Visentin, M. F. Barthe, V. Moineau, and P. Desgardin, Involvement of hydrogen-vacancy complexes in the baking effect of niobium cavities, Phys. Rev. Spec. Top. - Accel. Beams 13, 052002 (2010)
2010
-
[43]
Pfeiffer and H
G. Pfeiffer and H. Wipf, The trapping of hydrogen in niobium by nitrogen interstitials, J. Phys. F Met. Phys. 6, 167 (1976)
1976
-
[44]
Cox, Hydrogen trapping by oxygen and dislocations in zirconium alloys, J
B. Cox, Hydrogen trapping by oxygen and dislocations in zirconium alloys, J. Alloys Compd. 256, L4 (1997)
1997
-
[45]
Bafia, A
D. Bafia, A. Grassellino, and A. Romanenko, The Role of Oxygen Concentration in Enabling High Gradients in Niobium SRF Cavities, in The Role of Oxygen Concentration in Enabling High Gradients in Niobium SRF Cavities (US DOE, 2021)
2021
-
[46]
Padamsee, The science and technology of superconducting cavities for accelerators, Supercond
H. Padamsee, The science and technology of superconducting cavities for accelerators, Supercond. Sci. Technol. 14, R28 (2001)
2001
-
[47]
Hasan Padamsee, Jens Knobloch, Tomas Hays, RF Superconductivity for Accelerators, 2nd Edition | Wiley (2008)
2008
Reviewed August 5, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.