arxiv: 2604.13364 · v1 · submitted 2026-04-15 · ❄️ cond-mat.mes-hall · physics.app-ph

Recognition: unknown

Cryogenic Loss Limits in Microwave Epitaxial AlN Acoustic Resonators

Hemant Gulupalli , Navnil Choudhury , Jiacheng Xie , Yufeng Wu , Huili Grace Xing , Hong X. Tang , Debdeep Jena , Kanad Basu

show 1 more author

Wenwen Zhao

Authors on Pith no claims yet

Pith reviewed 2026-05-10 13:17 UTC · model grok-4.3

classification ❄️ cond-mat.mes-hall physics.app-ph

keywords aluminum nitrideacoustic resonatorsquality factorcryogenic temperaturesFBARHBARloss mechanismsmicrowave

0 comments

The pith

A physics-based model accurately predicts how temperature limits the quality factor of epitaxial AlN acoustic resonators down to 6.5 K.

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

The authors fabricate a 16 GHz epitaxial aluminum nitride thin-film bulk acoustic resonator and measure its loaded quality factor across temperatures from 6.5 K to room temperature. They construct a model that adds temperature-dependent intrinsic material losses to an analytical calculation of radiation losses leaving through the resonator anchors. The calculated quality-factor curve matches the measured drop from roughly 1589 at the coldest point to 363 at 294 K. The same model also reproduces previously published data on a 23 GHz high-overtone resonator. This supplies a concrete way to forecast and reduce loss in compact acoustic devices intended for quantum hardware and millimeter-wave systems.

Core claim

Measurements on a fabricated 16 GHz epitaxial AlN FBAR show the loaded quality factor decreasing monotonically from approximately 1589 at 6.5 K to 363 at 294 K. A physics-based model that includes both intrinsic temperature-dependent dissipation and an analytical anchor-radiation loss expression for bulk acoustic wave resonators reproduces this trend without post-hoc fitting and is further validated on a 23 GHz HBAR.

What carries the argument

The physics-based loss model that combines temperature-dependent intrinsic dissipation with an analytical anchor-radiation loss term for bulk acoustic wave resonators.

If this is right

The Qf product reaches 24.79 THz at 6.5 K for the 16 GHz FBAR.
Separate quantification of intrinsic and radiation losses guides geometry and material choices to raise cryogenic Q.
The framework applies across FBAR and HBAR geometries, supporting its use for other low-loss resonators.
Cryogenic microwave filter elements for superconducting quantum hardware can be designed by targeting the dominant loss mechanism at each temperature.

Where Pith is reading between the lines

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

If the analytical radiation term remains accurate for altered anchor designs, device layouts can be optimized analytically before fabrication.
Extension of the same loss accounting to millikelvin temperatures would directly inform integration with superconducting qubits.
Comparison of the model against resonators on different substrates could reveal whether substrate losses are already included or still missing.
The temperature scaling extracted here supplies a benchmark for testing new piezoelectric films beyond AlN.

Load-bearing premise

The model captures all relevant loss channels and the analytical anchor-radiation loss expression accurately predicts extrinsic losses from the given MIM stack and geometry without adjustments to the measured data.

What would settle it

A measured quality factor at 6.5 K that substantially exceeds the model's predicted upper bound for the same geometry and stack would falsify the claim that the listed mechanisms set the limit.

Figures

Figures reproduced from arXiv: 2604.13364 by Debdeep Jena, Hemant Gulupalli, Hong X. Tang, Huili Grace Xing, Jiacheng Xie, Kanad Basu, Navnil Choudhury, Wenwen Zhao, Yufeng Wu.

**Figure 1.** Figure 1: compares the quality factor versus frequency for various acoustic device architectures based on AlN, SiC, quartz, and LiNbO3, etc. The Q · f limits at 6.5 K and 300 K are shown for AlN resonators operating in longitudinal modes utilizing the piezoelectric coefficient d33. Contourmode resonators (CMRs), predominantly realized using AlN and Al1−xScxN thin films, exhibit room-temperature quality factors typ… view at source ↗

**Figure 2.** Figure 2: FIG. 2. (a) Cross-sectional schematic of a suspended AlN FBAR on [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Figure 3: FIG. 3. (a) Representative measured reflective response around the FBAR resonance without de-embedding: magnitude [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

**Figure 4.** Figure 4: FIG. 4. Temperature dependence of the relaxation times as [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗

read the original abstract

Aluminum nitride (AlN)-based thin-film bulk acoustic wave resonators (FBARs) are promising compact platforms for 6G communications and quantum memory hardware, enabled by their integrable acoustic modes with high quality factors. However, temperature-dependent acoustic dissipation ultimately limits device performance. In this work, we fabricated a 16 GHz epitaxial AlN FBAR as a test platform, performed small-signal RF measurements from 6.5 K to 300 K, and developed a physics-based model to estimate the fundamental quality-factor limits of FBARs to cryogenic temperatures. The proposed model incorporates both intrinsic and extrinsic loss mechanisms, including an analytical anchor-radiation loss model for bulk acoustic wave resonators, rather than relying solely on finite-element simulations. Measured loaded quality factor (Q) decreases monotonically with temperature, from Qmax of approximately 1589 (Qf=24.79 THz) at 6.5 K to 363 at 294K (Qf=5.66 THz). This trend is consistent with the theoretical limit based on the resonator geometry and the chosen Metal-Insulator-Metal (MIM) stack. To demonstrate the generality of the physics-based framework, we further validate it by benchmarking against a 23 GHz high-overtone bulk acoustic resonator (HBAR) using previously reported data. The validated model provides a practical, transferable framework to interpret Q(T) limits in low-loss resonators by quantifying the temperature-dependent mechanisms that constrain Q, enabling the design of cryogenic microwave filter elements for superconducting quantum hardware.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

New low-temp Q data on epitaxial AlN FBARs plus an analytical anchor-radiation model that tracks the measurements and an external HBAR set, but the abstract leaves open whether the model runs with zero adjustable constants.

read the letter

The paper's core contribution is a set of RF measurements on a 16 GHz epitaxial AlN FBAR from 6.5 K to 300 K, showing loaded Q falling from roughly 1590 to 363, together with a physics-based loss model that adds an analytical expression for anchor-radiation losses on top of intrinsic terms. They check the same framework against an independent 23 GHz HBAR dataset from prior work. That combination of new cryogenic data and a non-FEM-only model is the useful part; it gives device designers in quantum acoustics a concrete starting point for estimating Q(T) limits in MIM stacks without having to run full simulations for every geometry tweak. The validation step on the HBAR data is a reasonable check for generality within the subfield. The main soft spot is transparency on the model itself. The abstract says the measured trend is consistent with the theoretical limit from geometry and the MIM stack, yet it does not state whether the anchor-radiation formula was evaluated with strictly zero free parameters on the primary 16 GHz device or whether any effective boundary conditions were adjusted to match the observed Qmax. Without error bars on the Q values or an explicit derivation showing the expression contains only material and geometric inputs, the risk remains that some scaling was applied after the fact. If the full text demonstrates that the analytical term was fixed in advance and still reproduces both datasets, that concern disappears; otherwise the transferability claim rests on thinner ground than the abstract suggests. This work is aimed at researchers building cryogenic microwave filters or acoustic quantum memories who need practical Q estimates rather than pure theory. It is solid enough on the measurement side and modest enough on the modeling advance that it deserves a serious referee who can ask for the missing parameter details and error analysis. I would send it to review.

Referee Report

1 major / 3 minor

Summary. The manuscript reports fabrication and cryogenic (6.5–300 K) RF measurements of a 16 GHz epitaxial AlN thin-film bulk acoustic resonator (FBAR) in a MIM stack, with loaded Q decreasing from ~1589 at 6.5 K to 363 at 294 K. A physics-based model is presented that combines intrinsic loss channels with an analytical expression for anchor-radiation loss derived from the resonator geometry and stack; the measured Q(T) is stated to follow the resulting theoretical limit. The framework is further benchmarked against an independent 23 GHz HBAR dataset from prior literature.

Significance. If the central claim holds without post-hoc adjustment, the work supplies a practical, transferable model for quantifying temperature-dependent Q limits in low-loss acoustic resonators. This is relevant for cryogenic microwave components in superconducting quantum hardware. The analytical (rather than purely numerical) treatment of anchor-radiation loss and the external HBAR validation are concrete strengths that support generality beyond the primary device.

major comments (1)

[Model/theory section] Model/theory section: The abstract asserts that the measured Q(T) trend is 'consistent with the theoretical limit based on the resonator geometry and the chosen MIM stack' and that the model is 'physics-based' with an 'analytical anchor-radiation loss model'. The manuscript must explicitly present the anchor-radiation formula (including any boundary conditions or effective parameters) and demonstrate that it contains zero adjustable constants fitted to the 16 GHz FBAR data (e.g., to the reported Qmax ≈ 1589 at 6.5 K). If any scaling or effective value is determined from the primary dataset, the consistency claim becomes circular and the transferability asserted in the abstract is unproven.

minor comments (3)

[Abstract/Results] Abstract and results: Replace 'approximately 1589' with the precise measured value and state whether it is loaded or unloaded Q. Include error bars or uncertainty estimates on all Q(T) data points.
[Model section] Model section: List the numerical values and sources of all material parameters (e.g., intrinsic loss coefficients, acoustic velocities) used in the calculations; clarify whether any were obtained by fitting to the FBAR measurements.
[Validation section] Validation: The HBAR benchmark should report quantitative agreement metrics (e.g., RMS deviation or R^{2}) rather than qualitative consistency alone.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful review and constructive feedback on our manuscript. We address the major comment below and have revised the manuscript to strengthen the presentation of the model.

read point-by-point responses

Referee: [Model/theory section] Model/theory section: The abstract asserts that the measured Q(T) trend is 'consistent with the theoretical limit based on the resonator geometry and the chosen MIM stack' and that the model is 'physics-based' with an 'analytical anchor-radiation loss model'. The manuscript must explicitly present the anchor-radiation formula (including any boundary conditions or effective parameters) and demonstrate that it contains zero adjustable constants fitted to the 16 GHz FBAR data (e.g., to the reported Qmax ≈ 1589 at 6.5 K). If any scaling or effective value is determined from the primary dataset, the consistency claim becomes circular and the transferability asserted in the abstract is unproven.

Authors: We agree that an explicit derivation of the anchor-radiation loss term is necessary to fully substantiate the physics-based claim and eliminate any perception of circularity. In the revised manuscript we have added a dedicated subsection to the theory section that presents the complete analytical formula for anchor-radiation loss. The expression is obtained from the acoustic power radiated into the supporting anchors, using the resonator lateral dimensions, electrode and AlN thicknesses, and the acoustic impedances and velocities of the MIM stack layers. Boundary conditions are set by continuity of stress and displacement at the anchor interfaces; no effective scaling factors or adjustable constants are introduced. All numerical inputs are taken from independent literature values for material properties and from the lithographic mask dimensions. The low-temperature Q limit is therefore a forward prediction of the model rather than a fitted parameter. The same geometry-only framework, with only the HBAR-specific layer thicknesses and lateral dimensions substituted, reproduces the independent 23 GHz HBAR dataset without further adjustment, confirming transferability. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation remains independent of target data

full rationale

The paper constructs a physics-based model from intrinsic material loss mechanisms plus an analytical anchor-radiation expression derived directly from the stated MIM stack geometry and boundary conditions. Measured Q(T) is then compared to the resulting theoretical limit calculated from those geometry and material inputs; the abstract explicitly frames this as consistency rather than a fit. Validation is performed on an independent external HBAR dataset. No equations or statements in the provided text reduce the anchor-radiation term or overall limit to a parameter tuned against the 16 GHz FBAR measurements, nor do any load-bearing steps rely on self-citation chains. The framework therefore stands as a genuine forward prediction from geometry and physics, not a renaming or re-fitting of the observed values.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Abstract provides insufficient detail to enumerate all free parameters or axioms; the model rests on standard acoustic loss mechanisms plus a domain assumption that anchor radiation admits an analytical closed-form expression derived from geometry and the MIM stack.

free parameters (1)

intrinsic and extrinsic loss coefficients
Physics-based model must incorporate temperature-dependent material loss parameters and geometry-specific factors whose exact values are not stated in the abstract.

axioms (1)

domain assumption Anchor radiation loss for bulk acoustic wave resonators can be expressed analytically from geometry and MIM stack without requiring finite-element simulation
Explicitly adopted in place of FEM to estimate extrinsic losses.

pith-pipeline@v0.9.0 · 5611 in / 1520 out tokens · 76742 ms · 2026-05-10T13:17:58.188608+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

7 extracted references · 3 canonical work pages · 1 internal anchor

[1]

Cryogenic Loss Limits in Microwave Epitaxial AlN Acoustic Resonators

Experimentally,Q max of- ten denotes the maximum extractedQamong nearby oper- ating points and should be viewed as a device-level upper bound. Importantly,Q max can be raised by reducing extrinsic losses even if intrinsic material losses are unchanged13. Con- versely,Q max can remain low in excellent material if energy leaks through anchors or electrodes....

work page internal anchor Pith review Pith/arXiv arXiv 2026
[2]

erio ically pole aluminum scan ium nitri e bul acoustic wave resonators an filters for communications in the era,

Temperature dependence of the relaxation times as- sociated with phonon–phonon scattering and thermoelastic dissi- pation. The phonon relaxation time is modeled asτ phonon = 3κ(T)/ Cv(T)c 2 D , while the thermoelastic relaxation time is τTED = (b/π) 2 Cp(T)/κ(T). Here,bis the film thickness,κ(T)is the thermal conductivity71,C v(T)is the volumetric heat ca...

work page doi:10.1038/s41378-024-00857-4 2025
[3]

A loss mechanism study of a very highQsilicon mi- cromechanical oscillator,

pp. 74–77. 10X. Liu, J. F. Vignola, H. J. Simpson, B. R. Lemon, B. H. Houston, and D. M. Photiadis, “A loss mechanism study of a very highQsilicon mi- cromechanical oscillator,” Journal of Applied Physics97, 023524 (2005). 11M. Kazemi, S. Nabavi, M. Gratuze, and F. Nabki, “Anchor Loss Reduc- tion in Micro-Electro Mechanical Systems Flexural Beam Resonator...

2005
[4]

High-Q Capacitive- Piezoelectric AlN Lamb Wave Resonators,

21T.-T. Yen, A. P. Pisano, and C. T.-C. Nguyen, “High-Q Capacitive- Piezoelectric AlN Lamb Wave Resonators,” inProceedings of the 26th IEEE International Conference on Micro Electro Mechanical Systems (MEMS)(2013) pp. 114–117. 22R. Vetury, M. D. Hodge, and J. B. Shealy, “High Power, Wideband Single Crystal XBAW Technology for Sub-6 GHz Micro RF Filter App...

2013
[5]

Characteristics of an AlN-based Bulk Acoustic Wave Resonator in the Super High Fre- quency Range,

23K. Umeda, H. Kawamura, M. Takeuchi, and Y . Yoshino, “Characteristics of an AlN-based Bulk Acoustic Wave Resonator in the Super High Fre- quency Range,” Vacuum83, 672–674 (2008), selected papers revised from the Proceedings of the Ninth International Symposium on Sputtering and Plasma Processes (ISSP 2007), Kanazawa, Japan, 6–8 June

2008
[6]

Analysis and Removal of Spurious Re- sponse in SH0 Lithium Niobate MEMS Resonators,

24Y .-H. Song, R. Lu, and S. Gong, “Analysis and Removal of Spurious Re- sponse in SH0 Lithium Niobate MEMS Resonators,” IEEE Transactions on Electron Devices63, 2066–2073 (2016). 25J. Rodriguez, S. A. Chandorkar, C. A. Watson, and et al., “Direct Detection of Akhiezer Damping in a Silicon MEMS Resonator,” Scientific Reports9, 2244 (2019). 26M. Rinaldi, C...

2066
[7]

Quality factors in micron- and submicron-thick can- tilevers,

68K. Y . Yasumura, T. D. Stowe, E. M. Chow, T. Pfafman, T. W. Kenny, B. C. Stipe, and D. Rugar, “Quality factors in micron- and submicron-thick can- tilevers,” Journal of Microelectromechanical Systems9, 117–125 (2000). 69D. Sedmidubský, J. Leitner, P. Svoboda, Z. Sofer, and J. Machá ˇcek, “Heat capacity and phonon spectra of AIII nitrides,” Journal of Th...

work page doi:10.1038/srep02132 2000