Pith. sign in

REVIEW 2 major objections 6 minor 1 cited by

Flip-chip bonding integrates a superconducting resonator with a SiMOS quantum dot for dispersive readout.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-02 01:43 UTC pith:RPLE7OU4

load-bearing objection A solid first demonstration of flip-chip indium-bump integration for SiMOS dispersive readout, but single-device evidence and a simulated mode identification keep it an existence proof rather than a general validation. the 2 major comments →

arxiv 2607.14559 v1 pith:RPLE7OU4 submitted 2026-07-16 cond-mat.mes-hall

Dispersive Readout of a SiMOS Quantum Dot Using a Flip-Chip Integrated Microwave Resonator

Vo Kim Hieu Van , Santiago Serrano , C\'edric Boh\'emier , Ajit Dash , Fay E. Hudson , Tuomo Tanttu , Chih Hwan Yang , MengKe Feng
show 6 more authors
Ensar Vahapoglu Florian K. Unseld Wee Han Lim Andrea Morello Andrew S. Dzurak Kok Wai Chan
This is my paper
classification cond-mat.mes-hall PACS 85.35.Gv84.40.-x85.25.Cp73.63.Kv
keywords SiMOS quantum dotdispersive readoutflip-chip bondingindium bumpsuperconducting resonatorcharge sensingspin qubitheterogeneous integration
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper demonstrates that a silicon metal-oxide-semiconductor (SiMOS) double quantum dot chip can be flip-chip bonded to a separately fabricated superconducting aluminum resonator chip using indium bump interconnects, allowing the quantum dot's gate to serve as the capacitive readout node. The authors show that after bonding, the resonator still operates at 4.482 GHz with quality factors Qi ≈ 1798 and Qc ≈ 20900, and that charge transitions in the quantum dot system are detected through the resonator's dispersive response. They measure a signal-to-noise ratio that scales as the square root of integration time, reaching SNR = 1 in about 0.3 ms, and the readout remains functional up to roughly 640 mK and 10 mT in-plane magnetic field. The central claim is that separate fabrication and packaging of the semiconductor device and microwave circuit is a viable route for scalable spin-qubit readout, eliminating the need to co-fabricate resonators with the quantum dots.

Core claim

The central discovery is that a microwave resonator on a separate chip can be connected to a SiMOS quantum dot gate through indium-bump flip-chip bonding, and that the resulting assembly performs dispersive readout while preserving both dc gate control and microwave charge sensitivity. The authors observe resonator-based charge-stability maps in both reservoir-coupled and isolated regimes, demonstrate tunability of interdot coupling through the exchange gate, and quantify readout performance with an SNR that follows a sqrt(t) law, reaching unity at about 0.3 ms. They also show that the readout sensitivity degrades at higher temperature and magnetic field in a manner consistent with the super

What carries the argument

The key mechanism is the indium-bump flip-chip interconnect that capacitively couples the P2 plunger gate of the SiMOS double quantum dot to the superconducting aluminum resonator. This interconnect places the quantum dot gate in the resonator's electric-field distribution, so charge rearrangements in the dot alter the resonator's effective capacitance and hence its microwave reflection response. The complex scattering data are fit to extract internal and coupling quality factors, which confirm that microwave coupling is preserved across the bond.

Load-bearing premise

The assumption that the indium-bump interconnect between the plunger gate and the aluminum resonator is a low-loss, reproducibly aligned microwave path, inferred from a single assembled device (Qi = 1798, Qc = 20900) and electromagnetic simulation, rather than measured across many bonds.

What would settle it

Fabricate several flip-chip assemblies from the same process and measure their resonator transmission; if the indium bumps introduce variable loss or misalignment, the internal quality factor, resonance frequency, or readout SNR would vary significantly from device to device, showing that the result is not reproducible.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Quantum dot devices and superconducting resonators can be fabricated and optimized independently, then assembled by packaging, which may simplify manufacturing.
  • Existing SiMOS devices can gain microwave readout by bonding on a resonator chip, without adding lithographic steps to the device itself.
  • Resonator-based readout on a gate electrode is compatible with dense spin-qubit arrays, reducing the wiring fan-out per qubit.
  • The measured SNR scaling predicts single-shot charge readout at integration times around 0.3 ms, suitable for spin-qubit readout.
  • The demonstrated operating range (up to ~640 mK and ~10 mT) covers the typical parameter space for silicon spin qubits in dilution refrigerators.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the indium-bump interconnect proves low-loss and reproducible across many devices, this packaging route could align silicon spin-qubit systems with standard chip-scale assembly, a step toward industrial scalability.
  • A testable next step is to vary the resonator design (for example, using high-impedance resonators) while keeping the same flip-chip process, to see whether the bump bond or the gate lever arm sets the readout sensitivity limit.
  • The reported electrochemical-potential noise spectrum (approximately 1/f with 134 µeV/sqrt(Hz) at 1 Hz) could serve as a benchmark; comparing it with a non-integrated device would isolate any extra charge noise introduced by the flip-chip packaging.
  • The same assembly could be used for spin-photon coupling experiments, since the resonator is already directly connected to the gate; the resonator's quality factor would then determine the cooperativity.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 6 minor

Summary. The paper reports an experimental demonstration of flip-chip integration of a SiMOS double quantum dot with a superconducting aluminum microwave resonator using indium bump interconnects. The assembled device is characterized at 60 mK; a resonance at 4.482 GHz with internal and coupling quality factors Qi = 1798 and Qc = 20900 is observed, close to CST predictions. Charge-stability maps in reservoir-coupled and isolated regimes show resonator-detected charge transitions. The SNR of an interdot transition is reported as a function of microwave power, local-oscillator power, and integration time, with a sqrt(t) scaling and an extrapolated SNR = 1 at ~0.3 ms. The device is further characterized versus temperature and magnetic field, and an electrochemical-potential noise spectrum ~1/f with amplitude 134 ueV/sqrt(Hz) is extracted. The authors conclude that flip-chip bonding is a viable integration route for SiMOS quantum dots with microwave readout.

Significance. The main contribution is the integration platform: a separately fabricated superconducting resonator is bonded to a SiMOS gate stack with indium bumps and is shown to provide both dc gate control and dispersive charge sensitivity at millikelvin temperatures. This addresses a practical packaging bottleneck for scalable spin-qubit arrays. The measurements are internally consistent: the resonator parameters are obtained from standard complex-S21 fits; the charge-stability maps demonstrate gate-controlled charge transitions; and the SNR and noise data follow expected scaling. The CST simulation provides a plausible mode assignment, and the paper is honest about the limitations of simulating Qi. If the claims hold, the work is a useful step toward 3D-integrated silicon spin qubits. However, the quantitative headline (SNR = 1 at 0.3 ms) is an extrapolation, and the generality of the flip-chip approach rests on a single assembly, so the strength of the conclusions should be calibrated accordingly.

major comments (2)
  1. [Abstract; Section on SNR (Fig. 2(f))] The abstract states that the readout reaches SNR = 1 at approximately 0.3 ms, but the text says this is an estimated integration time from a fit to SNR ∝ √t. If 0.3 ms is not a directly measured point, the abstract overstates the result. Please state the shortest measured integration time, the fit parameters and their uncertainties, and either show the data point or explicitly label the 0.3 ms value as an extrapolation.
  2. [Resonator characterization (Fig. 1(c,d)); Conclusions] The identification of the 4.482 GHz mode as the designed resonator-interconnect mode is based on one assembled device and a CST simulation that, as the authors note, cannot capture Qi, the loss channel most sensitive to the indium-bump interconnect. Charge-stability maps prove dc connection and charge sensitivity, but do not independently rule out a parasitic stack mode. Since the paper's conclusion generalizes to a 'viable integration approach' for large-scale SiMOS devices, the evidence base (n = 1, no pre-bond baseline, no second device) should be strengthened, or the conclusions should be explicitly limited to the single assembly demonstrated. A dummy-chip or pre-bond S21 control, or a second device, would substantially increase confidence.
minor comments (6)
  1. [Fig. 3(f) and noise characterization] The notation '1/f^{1.04/2}' is confusing. Clarify whether S_µ is an amplitude or power spectral density and state explicitly which quantity the exponent β = 1.04 refers to.
  2. [Fig. 2(f) and methods] The SNR-vs-time fit is described only as 'a fit to the form of SNR ∝ √t'. Provide the number of data points, fitting range, and confidence intervals for the extracted 0.3 ms value.
  3. [Fig. 3(c,e)] The dc transport current data are not quantified. Specify the current scale, bias conditions, and measurement procedure so that 'remains visible within fluctuations' is meaningful.
  4. [Device fabrication and bonding] Reference [11] is a conference paper; for reproducibility, include key bonding parameters (temperature, pressure, time) or refer to a more detailed description if available.
  5. [Fig. 2(e)] The claim of 'SNR > 10 for LO powers 12, 15, 17 dBm and drive powers between -10 and 5 dBm' should be tied to the plotted data; specify which curves correspond to which LO power and how the threshold was determined.
  6. [General] Minor language issues: 'preserves both of the dc gate control' should be 'preserves both dc gate control', and 'has a superconducting critical temperature' should be 'have a superconducting critical temperature'.

Circularity Check

0 steps flagged

Experimental demonstration: direct measurements and transparent fits; no circular derivation chain.

full rationale

The paper's derivation chain is an experimental demonstration, not a first-principles derivation, so the classical circularity modes do not apply. The resonator mode (fr = 4.482 GHz, Qi = 1798, Qc = 20900) is directly measured and fitted; the CST simulation is a forward comparison (fr,sim=4.371 GHz, Qc,sim=21704) with the acknowledged limitation that Qi cannot be simulated, which limits the identification's robustness but does not make it circular. Charge-transition detection is directly evidenced by gate-voltage-dependent resonator response maps in reservoir and isolated modes. The SNR∝√t statement is presented explicitly as a fit ('A fit to the form of SNR∝√t gives an estimated integration time...'), not as a prediction from an independent input, so the 0.3 ms SNR=1 time is a transparent transformation of the same data rather than a fitted input disguised as a prediction. The only self-citation that could be questioned is the flip-chip bonding process of Ref. [11], cited for the assembly method; this is not load-bearing because the present paper's own measured resonator response, charge-stability maps, and temperature/field data independently demonstrate the bonding outcome. Remaining concerns—single assembled device, no pre-bond S21 baseline, mode identity resting partly on CST—are external-validity and correctness risks, not circularity.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 0 invented entities

No new physical entities or ad hoc corrections are introduced. The central claims rest on the integrity of the flip-chip interconnect, the assignment of the measured resonance to the designed mode, and standard resonator/SNR fitting—these are the assumptions a referee would want tested on more devices and with raw data.

free parameters (4)
  • Resonator internal quality factor Q_i = 1798
    Extracted from a complex fit to the measured S21 using the method of Ref. [12]; used to claim the resonator survives flip-chip bonding.
  • Resonator coupling quality factor Q_c = 20900
    Same complex fit; compared with CST value 21704.
  • SNR-vs-time scaling amplitude A (SNR = A√t) = not quoted; fixed by fit in Fig. 2(f)
    Used to estimate t(SNR=1) ≈ 0.3 ms; the abstract presents this extrapolated point as 'reaching SNR=1'.
  • Electrochemical-potential noise amplitude A (S_µ = A/f^(β/2)) = 134.01 µeV/√Hz at 1 Hz
    Fit to measured noise spectrum (β≈1.04); gives charge-noise benchmark.
axioms (4)
  • domain assumption The observed 4.482 GHz resonance is the designed compound mode with an anti-node at the quantum dots, as supported by CST simulation.
    If the resonance were a spurious package mode, the assigned Qi/Qc and coupling to P2 would be invalid. The simulation is the only check of the field distribution (Fig. 1d).
  • domain assumption The indium bump interconnects are electrically low-loss and superconducting (Tc≈3.4 K) at the operating temperature and field, so they do not dominate the resonator loss.
    Relied on throughout; the authors note indium Tc≈3.4 K [5,21], but do not directly measure bump resistance or loss.
  • domain assumption Gaussian-peak SNR extraction and baseline selection give an unbiased estimate of readout sensitivity.
    The SNR values in Fig. 2 depend on the baseline region and Gaussian fit; no validation against simulated or repeated-shot data is provided.
  • domain assumption The reflectometry chain (−88 dB input attenuation, 46 dB gain, IQ demodulation) is linear and its noise is dominated by white noise in the integration-time scaling.
    Used to interpret SNR∝√t; deviations at short t are expected if 1/f or amplifier noise contributes, but are not analyzed.

pith-pipeline@v1.3.0-alltime-deepseek · 7102 in / 11750 out tokens · 113822 ms · 2026-08-02T01:43:22.620552+00:00 · methodology

0 comments
read the original abstract

Heterogeneous integration provides a promising route to combine semiconductor quantum dot devices and superconducting microwave circuits, while allowing each component to be fabricated using an optimized process flow. Here, we demonstrate a flip-chip integrated platform for dispersive readout of silicon metal-oxide semiconductor (SiMOS) quantum dot devices. A SiMOS double quantum dot chip is bonded to a superconducting aluminum resonator chip using indium bump interconnects to enable microwave coupling to the quantum dot gate. We show that the developed flip-chip process is compatible with cryogenic operation of both the SiMOS device and the superconducting resonator, and demonstrate resonator-based detection of charge transitions in the quantum dot system. The readout signal-to-noise ratio follows a dependence of $\sqrt{t}$ with the integration time, reaching SNR = 1 at an integration time of approximately 0.3 ms. These results establish flip-chip bonding as a viable integration approach for SiMOS quantum dot devices operating at both dc and microwave frequencies, with potential applications for resonator-based techniques such as spin-photon coupling.

Figures

Figures reproduced from arXiv: 2607.14559 by Ajit Dash, Andrea Morello, Andrew S. Dzurak, C\'edric Boh\'emier, Chih Hwan Yang, Ensar Vahapoglu, Fay E. Hudson, Florian K. Unseld, Kok Wai Chan, MengKe Feng, Santiago Serrano, Tuomo Tanttu, Vo Kim Hieu Van, Wee Han Lim.

Figure 1
Figure 1. Figure 1: FIG. 1. Flip-chip integrated SiMOS quantum dot-resonator [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Dispersive readout of the flip-chip integrated SiMOS double quantum dot. (a-c) Resonator response in reservoir mode [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Cryogenic performance and noise characterization of the flip-chip integrated dispersive readout device. (a) Mi [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Spin-cQED with bulk germanium spin qubits

    cond-mat.mes-hall 2026-07 accept novelty 6.0

    Bulk unstrained Ge hole double quantum dots are predicted to reach gs/2π ≳ 100 MHz spin-photon coupling with broader B-field tolerance than strained Ge heterostructures.

Reference graph

Works this paper leans on

22 extracted references · 1 linked inside Pith · cited by 1 Pith paper

  1. [1]

    Veldhorst, C

    M. Veldhorst, C. Yang, J. Hwang, W. Huang, J. Dehol- lain, J. Muhonen, S. Simmons, A. Laucht, F. Hudson, K. M. Itoh,et al., A two-qubit logic gate in silicon, Na- ture526, 410 (2015)

    2015

  2. [2]

    Huang, C

    W. Huang, C. Yang, K. Chan, T. Tanttu, B. Hensen, R. Leon, M. Fogarty, J. Hwang, F. Hudson, K. M. Itoh, et al., Fidelity benchmarks for two-qubit gates in silicon, Nature569, 532 (2019)

    2019

  3. [3]

    Steinacker, N

    P. Steinacker, N. Dumoulin Stuyck, W. H. Lim, T. Tanttu, M. Feng, S. Serrano, A. Nickl, M. Can- dido, J. D. Cifuentes, E. Vahapoglu,et al., Industry- compatible silicon spin-qubit unit cells exceeding 99% fidelity, Nature646, 81 (2025)

    2025

  4. [4]

    Bonen, U

    S. Bonen, U. Alakusu, Y. Duan, M. Gong, M. Dadash, L. Lucci, D. Daughton, G. Adam, S. Iord˘ anescu, M. Pˇ a¸ steanu,et al., Cryogenic characterization of 22-nm fdsoi cmos technology for quantum computing ics, IEEE Electron Device Letters40, 127 (2018)

    2018

  5. [5]

    Foxen, J

    B. Foxen, J. Mutus, E. Lucero, R. Graff, A. Megrant, Y. Chen, C. Quintana, B. Burkett, J. Kelly, E. Jeffrey, et al., Qubit compatible superconducting interconnects, Quantum Science and Technology3, 014005 (2018)

    2018

  6. [6]

    Holman, D

    N. Holman, D. Rosenberg, D. Yost, J. Yoder, R. Das, W. D. Oliver, R. McDermott, and M. Eriksson, 3d in- tegration and measurement of a semiconductor double quantum dot with a high-impedance tin resonator, npj Quantum Information7, 137 (2021)

    2021

  7. [7]

    Petersson, C

    K. Petersson, C. Smith, D. Anderson, P. Atkinson, G. Jones, and D. Ritchie, Charge and spin state readout of a double quantum dot coupled to a resonator, Nano letters10, 2789 (2010)

    2010

  8. [8]

    Rossi, R

    A. Rossi, R. Zhao, A. Dzurak, and M. Gonzalez-Zalba, Dispersive readout of a silicon quantum dot with an accumulation-mode gate sensor, Applied Physics Letters 110(2017)

    2017

  9. [9]

    Granel, F

    S. Granel, F. Gustavo, J.-L. Thomassin, H. Niebojew- ski, B. Bertrand, F. Berger, A. Gueugnot, C. Mhamdi, E. Dumur, R. Maurand,et al., 3d integration of a hybrid quantum dot circuit-qed device for fast gate dispersive charge readout and coherent spin-photon coupling, arXiv preprint arXiv:2604.25871 (2026)

    Pith/arXiv arXiv 2026

  10. [10]

    K. Chan, W. Huang, C. Yang, J. Hwang, B. Hensen, T. Tanttu, F. Hudson, K. M. Itoh, A. Laucht, A. Morello, et al., Assessment of a silicon quantum dot spin qubit environment via noise spectroscopy, Physical Review Ap- plied10, 044017 (2018)

    2018

  11. [11]

    V. K. H. Van, C. Boh´ emier, S. S. Ramirez, R. Wrey- ford, A. Dash, A. Dickie, S. A. Yianni, F. E. Hudson, W. H. Lim, A. S. Dzurak, and K. W. Chan, Development of a flip-chip bonding process for scalable silicon-based quantum processors, in2026 10th IEEE Electron De- vices Technology & Manufacturing Conference (EDTM) (IEEE, 2026) pp. 1–3

    2026

  12. [12]

    Probst, F

    S. Probst, F. B. Song, P. A. Bushev, A. V. Ustinov, and M. Weides, Efficient and robust analysis of complex scat- tering data under noise in microwave resonators, Review of Scientific Instruments86, 024706 (2015)

    2015

  13. [13]

    J. Gao, M. Daal, A. Vayonakis, S. Kumar, J. Zmuidzinas, B. Sadoulet, B. A. Mazin, P. K. Day, and H. G. Leduc, Experimental evidence for a surface distribution of two- 6 level systems in superconducting lithographed microwave resonators, Applied Physics Letters92, 152505 (2008)

    2008

  14. [14]

    N. S. Lai, W. Lim, C. Yang, F. Zwanenburg, W. Coish, F. Qassemi, A. Morello, and A. Dzurak, Pauli spin block- ade in a highly tunable silicon double quantum dot, Sci- entific reports1, 110 (2011)

    2011

  15. [15]

    C. H. Yang, R. Leon, J. Hwang, A. Saraiva, T. Tanttu, W. Huang, J. Camirand Lemyre, K. W. Chan, K. Tan, F. E. Hudson,et al., Operation of a silicon quantum pro- cessor unit cell above one kelvin, Nature580, 350 (2020)

    2020

  16. [16]

    Han Lim, T

    W. Han Lim, T. Tanttu, T. Youn, J. Y. Huang, S. Ser- rano, A. Dickie, S. Yianni, F. E. Hudson, C. C. Escott, C. H. Yang,et al., A 2×2 quantum dot array in sili- con with fully tunable pairwise interdot coupling, Nano Letters25, 10263 (2025)

    2025

  17. [17]

    Zheng, N

    G. Zheng, N. Samkharadze, M. L. Noordam, N. Kalhor, D. Brousse, A. Sammak, G. Scappucci, and L. M. Van- dersypen, Rapid gate-based spin read-out in silicon us- ing an on-chip resonator, Nature nanotechnology14, 742 (2019)

    2019

  18. [18]

    A. West, B. Hensen, A. Jouan, T. Tanttu, C.-H. Yang, A. Rossi, M. F. Gonzalez-Zalba, F. Hudson, A. Morello, D. J. Reilly,et al., Gate-based single-shot readout of spins in silicon, Nature nanotechnology14, 437 (2019)

    2019

  19. [19]

    Khukhareva, The superconducting properties of thin aluminum films, Soviet Physics JETP16, 828 (1963)

    I. Khukhareva, The superconducting properties of thin aluminum films, Soviet Physics JETP16, 828 (1963)

    1963

  20. [20]

    Caplan and G

    S. Caplan and G. Chanin, Critical-field study of super- conducting aluminum, Phys. Rev.138, A1428 (1965)

    1965

  21. [21]

    C. Y. Ho, R. W. Powell, and P. E. Liley, Thermal conduc- tivity of the elements, Journal of physical and chemical reference data1, 279 (1972)

    1972

  22. [22]

    T. J. Wilson and H. Jiang, Fast and sensitive readout of a semiconductor quantum dot using an in-situ microwave resonator with enhanced gate lever arm, arXiv preprint arXiv:2510.00765 (2025)

    arXiv 2025