Pith. sign in

REVIEW 3 major objections 5 minor 27 references

Ge$_{1-x}$Si$_{x}$ single crystals for Ge hole spin qubit integration

T0 review · 3 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A bulk Ge0.85Si0.15 crystal is a homogeneous random alloy with Ge-like valence bands.

desk verdict Solid multi-technique characterization of a bulk GeSi crystal, but the random-alloy conclusion overreaches what the XPD comparison can actually show. read the letter →

arxiv 2504.15943 v1 pith:YMDCPIW2 submitted 2025-04-22 cond-mat.mtrl-sci

Andreas Fuhrberg , Pia M. Düring , Olena Fedchenko , Olena Tkach , Yaryna Lytvynenko , Kevin Gradwohl , Sergii Chernov , Andrei Gloskovskii
show 4 more authors
Christoph Schlueter Gerd Schönhense Hans-Joachim Elmers Martina Müller
This is my paper · ORCID
classification cond-mat.mtrl-sci
keywords GeSialloyholespinqubitsHAXPEShardX-raymomentummicroscopyphotoelectrondiffractionrandomvalencebandstructureCzochralskigrowth
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Ge-based hole-spin qubits need low-dislocation GeSi substrates, but the usual strain-relaxed buffer layers on silicon wafers are defect-rich. This paper reports a Czochralski-grown Ge0.85Si0.15 bulk single crystal and claims, on the basis of three synchrotron X-ray techniques, that it is chemically homogeneous, electronically uniform, and free of phase separation, with Si atoms sitting on Ge lattice sites in a random alloy. The measured valence band resembles pure Ge, including low effective hole masses, and the spin-orbit split-off gap is reduced to 0.35 eV. If correct, this provides a bulk wafer platform that could combine minimal lattice strain at x = 0.15 with low dislocation density for strained-Ge quantum wells. The promised benefit of longer spin lifetimes is inferred from this quality, not directly measured.

What carries the argument

The central mechanism is element-specific X-ray photoelectron diffraction (XPD): photoelectrons emitted from Si 2p and Ge 3p core levels are scattered by the surrounding lattice, producing angular patterns that depend on the emitter site and species and on the local geometry. Because the Si 2p pattern of the alloy closely resembles the Ge 3p pattern of the pure Ge reference, and because Bloch-wave simulations reproduce the measured patterns, the comparison places Si at Ge lattice sites and, in the authors' reading, excludes short-range Si order. The two companion measurements supply the chemical and electronic context: HAXPES core-level intensities at two photon energies give the Ge:Si composition, and hard X-ray momentum microscopy maps the valence band dispersion along high-symmetry directions, from which band curvatures and effective masses are extracted.

What would settle it

A quantitative comparison of the measured Si 2p XPD pattern with Bloch-wave simulations of short-range-ordered supercells (for example, Si–Si pairs in the first or second shell) would settle the random-alloy claim; if an ordered model fits as well as or better than the random model, the conclusion fails. Alternatively, Si K-edge EXAFS would directly count Si neighbours in the first coordination shell: a random alloy should show about four Ge neighbours, while clustering would produce Si neighbours.

Watch

Extended reading notes

Core claim

Using hard X-ray photoelectron spectroscopy (HAXPES), hard X-ray momentum microscopy, and hard X-ray photoelectron diffraction (hXPD), the authors find that a bulk Ge0.85Si0.15 single crystal has a uniform valence band structure with no signs of phase separation. The heavy/light-hole and split-off bands resemble those of pure Ge, with effective masses $m^*_{\mathrm{LH/HH}} = 0.07(2)\,m_0$ and $m^*_{\mathrm{SO}} = 0.05(3)\,m_0$, and the spin-orbit split-off gap is 0.35(10) eV compared with 0.5(1) eV for Ge. The Si 2p XPD pattern matches the Ge 3p pattern of the Ge reference and agrees with Bloch-wave calculations, which the authors take as evidence that Si occupies Ge lattice sites in a random alloy with no short-range order. Their conclusion is that the crystal is electronically uniform, the Si distribution is homogeneous, and phase segregation can be ruled out.

Load-bearing premise

The load-bearing premise is that the Bloch-wave XPD simulations used for comparison can detect short-range ordering of Si atoms; the agreement reported in the paper is visual and qualitative, so if those simulations start from a random-alloy model, the measurement alone cannot exclude short-range order.

Editorial extensions

If this is right

  • Strained Ge quantum wells grown on such a bulk substrate would start from a chemically uniform template, removing composition fluctuations as a source of disorder.
  • The low effective masses close to those of pure Ge imply high hole mobility in the quantum well, which is favourable for qubit control.
  • The demonstrated homogeneity of a bulk random alloy at x = 0.15 means the minimal-strain composition can be stabilised outside the thin-film buffer geometry.
  • The combination of HAXPES, momentum microscopy, and XPD provides a general certification route for the electronic and structural uniformity of qubit substrate materials.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not measure spin lifetimes; the claim that this crystalline quality 'enables long spin lifetimes' is a forward-looking motivation. A decisive test would be to grow a strained Ge quantum well on this substrate and measure $T_1$ or $T_2^*$ against a conventional strain-relaxed buffer.
  • The random-alloy conclusion rests on the assumption that the Bloch-wave simulations are sensitive to short-range order. A quantitative check would be Si K-edge EXAFS or diffuse X-ray scattering, which directly probe Si–Si versus Si–Ge neighbour statistics.
  • The measured reduction of the split-off gap with 15% Si suggests the same momentum-microscopy protocol could map how effective masses and SO splitting vary with Si fraction, giving direct input for k·p models of Ge/GeSi heterostructures.
  • Because the bulk crystal is grown by Czochralski feeding rather than epitaxial buffer deposition, the method, if reproducible, points toward wafer-scale substrates for quantum devices rather than small MBE-grown pieces.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The manuscript reports a multi-technique synchrotron study of a Czochralski-grown Ge0.85Si0.15 (001) single crystal intended as a substrate for strained-Ge hole-spin quantum wells. Hard X-ray photoelectron spectroscopy is used for chemical composition, hard X-ray momentum microscopy for valence-band dispersion and effective masses, and hard X-ray photoelectron diffraction with Bloch-wave simulations for local atomic structure. The authors claim that the crystal is electronically uniform, that Si is homogeneously distributed, that Si occupies Ge lattice sites in a random alloy, that no phase separation or short-range order is present, and that the low effective Ge-like masses make the substrate suitable for qubit integration.

Significance. If the central claims hold, the paper provides a useful materials characterization template for an alternative to strain-relaxed GeSi buffer layers, with potential relevance to Ge hole-spin qubit scalability. The combination of HAXPES, momentum microscopy, and XPD on the same sample is a strength, and the direct comparison with semiconductor-grade Ge and Si references is valuable. The claim that Si forms a random alloy rather than a short-range-ordered or clustered distribution is, however, the load-bearing conclusion and is not established by the evidence presented. The reported data are also interesting in their own right, especially the valence-band maps, but the manuscript currently overstates what the XPD comparison can prove.

major comments (3)
  1. [II, Structural insights ... by X-ray photoelectron diffraction; Conclusion] The conclusion that 'good agreement of experimental and calculated diffraction pattern also excludes the presence of a short range order of Si atoms' is not supported by the evidence. The manuscript never states how the Bloch-wave calculations represent the Si distribution: the simulations could assume a random site-occupation model, a virtual-crystal approximation, or a pure-Ge lattice. If the reference calculation already assumes a random alloy or a Ge-like environment, the comparison cannot discriminate random from short-range-ordered or clustered Si distributions. Moreover, the agreement is assessed only visually, with no quantified residual, R-factor, or sensitivity test. To support the random-alloy conclusion, the authors should specify the alloy model used in the simulations, include simulated patterns for at least one ordered or clustered Si configuration, and quantify the agreement between measured and calculated patterns.
  2. [I.A and SI A] The dislocation density is reported inconsistently: the main text states 1 × 10−5 cm−2, while the Supplementary Information gives 'around 1 × 105 cm−2 in the center' and 'below 1 × 106 cm−2'. These values differ by many orders of magnitude and directly affect the central substrate-quality narrative that the crystal is superior to strain-relaxed buffer layers. The authors should correct the discrepancy, report the measurement method, and clarify whether different positions or different crystals are being described.
  3. [II.B.1] The HAXPES composition analysis yields Ge0.85Si0.15 at 6 keV and Ge0.8Si0.2 at 2.8 keV, with a stated accuracy of 'about 2%'. The two extracted Si fractions differ by 5 percentage points, which exceeds the quoted accuracy if the 2% is interpreted as an absolute uncertainty. Since the manuscript uses this depth-dependent composition comparison to support the homogeneity claim, the full uncertainty budget (cross-section uncertainties, background model, oxide correction) should be given and the discrepancy reconciled.
minor comments (5)
  1. [Conclusion] The phrase 'by alloying the crystal alloyed with 15% Si' is grammatically awkward and should be revised, for example to 'by alloying the crystal with 15% Si'.
  2. [Conclusion] There is a typo: 'real-word building blocks' should read 'real-world building blocks'.
  3. [II.B.3 / Fig. 3 caption] The caption reports varying kinetic energies (3203 eV, 3180 eV, 3167 eV) while the text around Fig. 3 describes 'Ekin ≈ 3200 eV'; please clarify which kinetic energy corresponds to which spectrum and simulation.
  4. [Fig. 3 caption] The caption labels the simulated panels as 'Simulated XPD core level spectra at Ekin = 3166 eV' but does not state which core level or alloy model each panel represents; adding this information would help readers.
  5. [Abstract] The abstract states the results 'enable long spin lifetimes' in Ge-based hole-spin qubits, but no spin-lifetime measurement is reported; the claim should be softened to a materials-level implication rather than a demonstrated device property.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper reports direct measurements and compares them to independent, external Bloch-wave calculations; no result is constructed from parameters fitted to the conclusion.

full rationale

The paper's derivation chain is measurement-driven: chemical composition comes from HAXPES cross-section-weighted peak areas, the valence band structure and effective masses come from momentum-microscopy band curvatures, and the XPD structural conclusion comes from comparing measured diffraction patterns to Bloch-wave dynamical calculations cited from the external literature (Refs. 14, 15). No free parameter is fitted to the data in a way that would force the conclusions about homogeneity or the random-alloy nature. The self-citations (e.g., Refs. 8, 20, 21) are methodological and are not load-bearing for the central scientific claims. The only concern is that the conclusion that 'good agreement ... also excludes the presence of a short range order' requires the Bloch-wave simulation to be sensitive to short-range order, and the manuscript does not state how the Si distribution was modeled in the calculation. This is an inference-quality gap, not a circularity: the agreement is an external model test, and the conclusion is not mathematically identical to the simulation input by construction. A separate numerical inconsistency exists for the dislocation density (1e-5 cm^-2 in the main text vs. 1e5-1e6 cm^-2 in the SI), but that affects the substrate-quality narrative, not the circularity of the derivation.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The paper introduces no free parameters or invented entities. It rests on standard theoretical models (direct transitions, Bloch-wave XPD, parabolic band fitting) and on the stated assumption that the surface oxide is inert for the momentum-resolved measurements. The main unverified premise is the XPD simulation's sensitivity to short-range order, which is the load-bearing assumption behind the 'random alloy' conclusion.

assumptions (4)
  • domain assumption Direct-transition model for k_z assignment in momentum microscopy
    Used in Section I.A to choose photon energies that map specific k_z planes (Fig. 2). If the model or the assumed inner potential is wrong, band energies at X and L points would be misassigned.
  • standard math Bloch-wave dynamical theory for XPD simulation
    Used for the simulated XPD patterns in Fig. 3(e-h); the correctness of the comparison depends on this standard theory.
  • domain assumption Parabolic approximation for effective mass extraction
    Used to convert band curvature near Gamma into m* values; the approximation is stated in Section I.B.2 and yields values with large error bars.
  • domain assumption Amorphous oxide overlayer adds only background and no coherent VB signal
    Stated in SI.B; if the thin surface oxide contributed coherently, the measured VB maps and XPD patterns could be distorted.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Ge$_{1-x}$Si$_{x}$ single crystals for Ge hole spin qubit integration." pith.science (2026). https://pith.science/paper/YMDCPIW2

@misc{pith2026250415943,
  author       = {Pith},
  title        = {Pith review of: Ge$_1-x$Si$_x$ single crystals for Ge hole spin qubit integration},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YMDCPIW2}},
  note         = {Machine review of arXiv:2504.15943}
}
abstract

Spin qubits are fundamental building blocks of modern quantum computing devices. The path of Ge-based hole-spin qubits has several advantages over Si-based electron-spin systems, such as the absence of valley band degeneracy, the possibility of efficient field control due to large spin-orbit coupling, and smaller effective masses. Among the possible Ge qubit devices, Ge/GeSi planar heterostructures have proven to be favourable for upscaling and fabrication. The Si concentration of the straining GeSi buffer serves as an important tuning parameter for the electronic structure of Ge/GeSi qubits. A particularly low Si concentration of x = 0.15 of the Ge$_{0.85}$Si$_{0.15}$ crystal should enable minimal lattice strain for spin qubit heterostructures, which is difficult to stabilize as a random alloy. We present a synchrotron-based study to investigate the chemical composition, valence band electronic structure and local atomic structure of a Ge$_{0.85}$Si$_{0.15}$ single crystal using the advanced combination of hard X-ray photoelectron spectroscopy (HAXPES), hard X-ray momentum microscopy (HarMoMic) and X-ray photoelectron diffraction (XPD). We found that the Ge$_{0.85}$Si$_{0.15}$ crystal has an individual, uniform valence band structure, with no signs of phase separation. The shapes of the valence bands resemble those of pure Ge, as do the low effective masses. XPD experiments and Bloch wave calculations, show the Si atoms located at Ge lattice sites within the crystal, forming a random alloy. This high chemical, electronic and structural quality of Ge$_{0.85}$Si$_{0.15}$ single-crystal substrates is of crucial importance for their implementation to enable long spin lifetimes in Ge-based hole-spin qubits. The results emphasise the power of combined X-ray spectromicroscopy techniques, which provide key insights into the qubit building blocks that form the basis of quantum technologies.

Figures

Figures reproduced from arXiv: 2504.15943 by the authors.

Figure 1
Figure 1. Chemical properties of the GeSi single crystal determined by HAXPES. Photon energies of 2.8 keV and 6 keV enable increased surface and bulk sensitivity, respectively. The Ge 3s (left) and Si 2s (right) core levels are shown, both of which exhibit slightly increased surface oxidation. Ge:Si composition analysis yields Ge0.85Si0.15 for 6 keV. 6 [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Momentum-resolved photoelectron microscopy of Ge, Ge0.85Si0.15 and Si single crystalline samples. (a) Γ-plane of the Ge0.85Si0.15 sample. (b) Sliced E(k) maps from the Γ- to the L-plane for Ge0.85Si0.15. (c)–(e) Valence band electronic structure of the Ge, Ge0.85Si0.15 and Si single crystal along L − Γ − X direction. Next, we turn to the results of the band structure measurement of the Ge crystal, shown in [PITH_FU… view at source ↗
Figure 3
Figure 3. High-resolution X-ray photoelecton diffraction pattern of Ge 3p3/2 and Si 2p3/2 core levels of the (a) Ge, (b),(c) Ge0.85Si0.15 and (d) Si single crystals. The photon energies were varied to achieve kinetic energies around Ekin ≈ 3200 eV. (e)-(h) Simulated XPD core level spectra at Ekin = 3166 eV using Bloch wave calculations [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Oxidation state of the plain Ge and Si substrates, determined by HAXPES. Photon [PITH_FULL_IMAGE:figures/full_fig_p016_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 17 canonical work pages

  1. [1]

    author Hendrickx, N. W. et al. title A four-qubit germanium quantum processor . journal Nature volume 591 , pages 580–585 ( year 2021 ). ://dx.doi.org/10.1038/s41586-021-03332-6

    doi:10.1038/s41586-021-03332-6 2021

  2. [2]

    author Philips, S. G. J. et al. title Universal control of a six-qubit quantum processor in silicon . journal Nature volume 609 , pages 919–924 ( year 2022 ). ://dx.doi.org/10.1038/s41586-022-05117-x

    doi:10.1038/s41586-022-05117-x 2022

  3. [3]

    author George, H. C. et al. title 12-spin-qubit arrays fabricated on a 300 mm semiconductor manufacturing line . journal Nano Letters volume 25 , pages 793–799 ( year 2024 ). ://dx.doi.org/10.1021/acs.nanolett.4c05205

    doi:10.1021/acs.nanolett.4c05205 2024

  4. [4]

    author Scappucci, G. et al. title The germanium quantum information route . journal Nature Reviews Materials volume 6 , pages 926--943 ( year 2020 ). ://www.nature.com/articles/s41578-020-00262-z

    work page 2020

  5. [5]

    author Stehouwer, L. E. A. et al. title Germanium wafers for strained quantum wells with low disorder . journal Applied Physics Letters volume 123 ( year 2023 ). ://dx.doi.org/10.1063/5.0158262

    doi:10.1063/5.0158262 2023

  6. [6]

    author Subramanian, A. N. et al. title On the Czochralski growth of Si _ x Ge _ 1-x crystals as substrates for strained ge quantum well heterostructures . journal Journal of Applied Physics volume 137 ( year 2025 ). ://dx.doi.org/10.1063/5.0238533

    work page doi:10.1063/5.0238533 2025

  7. [7]

    author Schlueter, C. et al. title The new dedicated HAXPES beamline P22 at PETRAIII . journal AIP Conference Proceedings volume 2054 , pages 040010 ( year 2019 )

    work page 2019

  8. [8]

    author Müller, M. et al. title Hard x-ray photoelectron spectroscopy of tunable oxide interfaces . journal Journal of Vacuum Science and Technology A volume 40 , pages 013215 ( year 2022 ). ://dx.doi.org/10.1116/6.0001491

    doi:10.1116/6.0001491 2022

Show all 27 references
  1. [9]

    , author Smekal, W

    author Werner, W. , author Smekal, W. & author Powell, C. J. title Simulation of Electron Spectra for Surface Analysis (SESSA) Version 2.2.2 User's Guide ( year 2024 ). ://doi.org/10.6028/NIST.NSRDS.100-2024

    2024 doi

  2. [10]

    author Chernov, S. et al. title Anomalous d-like surface resonances on Mo(110) analyzed by time-of-flight momentum microscopy . journal Ultramicroscopy volume 159 , pages 453--463 ( year 2015 )

    2015

  3. [11]

    author Medjanik, K. et al. title Progress in HAXPES performance combining full-field k -imaging with time-of-flight recording . journal Journal of Synchrotron Radiation volume 26 , pages 1996--2012 ( year 2019 ). ://doi.org/10.1107/S1600577519012773

    1996 doi

  4. [12]

    author Rosenberger, P. et al. title Spin polarization of the two-dimensional electron gas at the EuO/SrTiO _3 interface ( year 2024 ). note Preprint at https://arxiv.org/abs/2410.23804

    2024

  5. [13]

    author Medjanik, K. et al. title Direct 3D mapping of the Fermi surface and Fermi velocity . journal Nature Materials volume 16 , pages 615--621 ( year 2017 ). ://www.nature.com/articles/nmat4875

    2017

  6. [14]

    , author Fadley, C

    author Winkelmann, A. , author Fadley, C. S. & author Garcia de Abajo, F. J. title High-energy photoelectron diffraction: model calculations and future possibilities . journal New Journal of Physics volume 10 , pages 113002 ( year 2008 )

    2008

  7. [15]

    , author Schröter, B

    author Winkelmann, A. , author Schröter, B. & author Richter, W. title Simulation of high energy photoelectron diffraction using many-beam dynamical Kikuchi-band theory . journal Physical Review B volume 69 , pages 245417 ( year 2004 )

    2004

  8. [16]

    author Schindelin, J. et al. title Fiji: an open-source platform for biological-image analysis . journal Nature Methods volume 9 , pages 676–682 ( year 2012 ). ://dx.doi.org/10.1038/nmeth.2019

    2012 doi

  9. [17]

    & author Yarzhemsky, V

    author Trzhaskovskaya, M. & author Yarzhemsky, V. title Dirac–fock photoionization parameters for HAXPES applications . journal Atomic Data and Nuclear Data Tables volume 119 , pages 99--174 ( year 2018 ). ://www.sciencedirect.com/science/article/pii/S0092640X16300596

    2018

  10. [18]

    title High-mobility si and ge structures

    author Schäffler, F. title High-mobility si and ge structures . journal Semiconductor Science and Technology volume 12 , pages 1515 ( year 1997 ). ://dx.doi.org/10.1088/0268-1242/12/12/001

    1997 doi

  11. [19]

    author Chelikowsky, J. R. & author Cohen, M. L. title Electronic structure of silicon . journal Phys. Rev. B volume 10 , pages 5095--5107 ( year 1974 ). ://link.aps.org/doi/10.1103/PhysRevB.10.5095

    1974 doi

  12. [20]

    author Fedchenko, O. et al. title High-resolution hard-x-ray photoelectron diffraction in a momentum microscope—the model case of graphite . journal New Journal of Physics volume 21 , pages 113031 ( year 2019 )

    2019

  13. [21]

    author Medjanik, K. et al. title Site-specific atomic order and band structure tailoring in the diluted magnetic semiconductor (In,Ga,Mn)As . journal Physical Review B volume 103 , pages 075107 ( year 2021 )

    2021

  14. [22]

    author Hoesch, M. et al. title Active sites of Te-hyperdoped silicon by hard x-ray photoelectron spectroscopy . journal Applied Physics Letters volume 122 ( year 2023 )

    2023

  15. [23]

    title Comprehensive Handbook of Chemical Bond Energies ( publisher CRC Press , year 2007 )

    author Luo, Y.-R. title Comprehensive Handbook of Chemical Bond Energies ( publisher CRC Press , year 2007 ). ://dx.doi.org/10.1201/9781420007282

    2007 doi

  16. [24]

    author Bera, M. K. et al. title Rapid thermal oxidation of Ge-rich Si _ 1-x Ge _ x heterolayers . journal Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films volume 24 , pages 84–90 ( year 2005 ). ://dx.doi.org/10.1116/1.2137329

    2005 doi

  17. [25]

    author Binder, J. F. , author Broqvist, P. , author Komsa, H.-P. & author Pasquarello, A. title Germanium core-level shifts at Ge/GeO _ 2 interfaces through hybrid functionals . journal Phys. Rev. B volume 85 , pages 245305 ( year 2012 ). ://link.aps.org/doi/10.1103/PhysRevB.85.245305

    2012 doi

  18. [26]

    , " * write output.state after.block =

    ENTRY address author booktitle chapter edition editor eid howpublished institution isbn issn journal key month note number organization pages publisher school series title type url volume year label INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCT...

  19. [27]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.