REVIEW 3 major objections 5 minor 39 references
Centimeter-scale fully suspended metal and metal oxide thin films by one-step transfer-free liquid metal capillary forming
T0 review · 3 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash
Pith's one-line read The paper claims that a liquid metal's native surface oxide can act as a surfactant to form centimeter-scale fully suspended metal and metal-oxide thin films in a one-step, transfer-free capillary process.
desk verdict Novel capillary-forming route to cm-scale suspended metal/oxide films; credible fabrication work, but the 'fully continuous' few-nm oxide film claim needs stronger large-area evidence. 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 self-limiting native metal oxide monolayer (for example, GaOx) that forms on each surface of the liquid metal film within seconds. It acts as a viscoelastic surfactant bilayer: it laminates the high-surface-tension liquid metal into a stable capillary film (the liquid-metal thin film, l-MTF), and after dewetting drains the metal away, the two oxide layers zip together without an observable interface, forming the few-nanometer suspended metal oxide thin film (MOTF). The paper also uses the classical capillary-film framework—Frankel's law and the Reynolds-lubrication drainage equation—to explain the film's thickness profile and the onset of dewetting at the dimpl
What would settle it
To test the continuity claim, one could measure gas permeation or lateral electrical conduction across a centimeter-scale suspended MOTF: if the film contains hidden pinholes or merely loosely adhered layers, permeation would be detectable and lateral conductivity would differ from a seamless oxide; alternatively, in-situ transmission electron microscopy of the drainage event could directly verify the zippering process.
Extended reading notes
Core claim
The central claim is that the native metal oxide surface layer, which develops spontaneously on liquid metals in air, can play the role of a surfactant in forming stable capillary films. The authors demonstrate this by drawing a ring-shaped frame out of a liquid metal bath, producing a micrometer-thick liquid metal film (l-MTF) sandwiched between two oxide monolayers, and then letting the liquid metal drain away through dewetting. Instead of collapsing, the two oxide monolayers 'zip' together into a continuous few-nanometer-thick suspended metal oxide thin film (MOTF), showing no observable interface in cross-section. The paper reports circular GaOx films up to 2.5 cm in diameter and rectang
Load-bearing premise
The claim that the suspended film is one continuous, pinhole-free, few-nanometer oxide membrane across the whole centimeter-scale area rests on characterization of small transferred samples; the 'atomic zippering' of the two oxide layers is inferred from TEM and AFM, not directly proven in situ.
Editorial extensions
If this is right
- The method produces fully suspended metal and metal-oxide films without any substrate or transfer step, avoiding substrate-induced defects and interface effects.
- The resulting MOTFs are amorphous, transparent (above ~80% transmittance), and centimeter-scale, making them candidates for large-area free-standing membranes in nanomechanical and optical devices.
- The demonstrated acoustic detection sensitivity of 1.94×10^-4 m/Pa, with a pressure detection limit near 2 mPa, suggests these films could serve as ultra-sensitive microphone diaphragms.
- The generality across liquid metals and alloys points to a scalable route for making suspended oxide films with tailored compositions, including p-type and high-entropy oxides.
- The ability to solidify l-MTFs into complex 3D minimal-surface geometries in seconds offers a rapid prototyping method for thin-walled metallic structures.
Reading between the lines
- If the atomic-zippering mechanism holds, the MOTF may have electronic and mechanical properties distinct from two stacked monolayers; this could be probed by in-plane transport or local conductivity measurements across the membrane.
- The method suggests a general strategy for creating free-standing amorphous oxide membranes of arbitrary alloy composition, limited mainly by the existence of a native surface oxide and a workable melting point.
- One testable extension is to use bath temperature and withdrawal speed to control MOTF thickness and residual nanodroplet density, potentially tuning the mechanical and optical response.
- The sensitivity scaling with lateral size implies that even larger frames, if dewetting can be controlled, could push acoustic detection limits into the sub-millipascal range.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports a one-step, transfer-free, substrate-free fabrication of fully suspended liquid-metal (l-MTF), solid-metal (s-MTF), and metal-oxide (MOTF) thin films by withdrawing a Cu frame from a liquid-metal bath, using the native surface oxide as a surfactant-like stabilizer. The authors show that l-MTFs have soap-film-like thickness profiles (Plateau, dimpled, Frankel zones), can solidify into s-MTFs, or can dewet into few-nanometer-thick 'zipped' oxide films that reach circular diameters up to 2.5 cm and rectangular areas up to ~4 cm², corresponding to a claimed lateral size-to-thickness ratio of order 10^7. They test Frankel's law with a fixed constant k=1.8933, extend the method to several metals and alloys, fabricate minimal-surface 3D metal structures, and demonstrate suspended MOTFs as acoustic diaphragms, extracting a membrane tension τ* ≈ 4.60 mN/m. The central claim is the existence of centimeter-scale, continuous, fully suspended oxide films produced without substrate or transfer.
Significance. If the continuity claim survives scrutiny, this would be a notable advance in suspended thin-film fabrication, moving beyond substrate-dependent growth/transfer and enabling centimeter-scale free-standing oxide membranes with a size-to-thickness ratio that exceeds prior suspended 2D films by at least an order of magnitude. The paper has concrete strengths: direct visual and micro-CT evidence for the metallic films, use of an external Frankel-law constant rather than a fitted parameter, extension to multiple metals/alloys, and a functional acoustic-detection demonstration with a quantitative tension extraction. The main weakness is that the flagship cm-scale continuous MOTF and the 'atomic zippering' mechanism are inferred from small, transferred, ex-situ specimens; the full-area state of the suspended film is not directly evidenced. The work is therefore significant and promising, but the key claim needs additional support.
major comments (3)
- [Formation and characterization of MOTFs (Fig. 3)] The central claim of a fully suspended, continuous few-nanometer MOTF with S/h~10^7 is not directly supported. Photographs (Fig. 3A) have insufficient resolution to exclude cracks, holes, or residual liquid-metal channels; the low-magnification TEM (Fig. 3D,G) and AFM thickness profiles (Fig. 3J-M) were acquired after transfer onto grids or Si, and the cross-sections (Fig. 3F,I) after FIB slicing. These cannot establish the state of the as-formed cm-scale suspended film. A network of oxide rafts bridged by nanodroplets or a partially unzipped film would still look transparent at optical resolution. Please supply large-area continuity evidence (e.g., full-area SEM/optical stitching over the entire frame, gas-leak or pressure-hold test, or interference mapping) or explicitly temper the S/h and acoustic-diaphragm claims.
- [Formation mechanism and thickness profile (Eq. 1, Fig. 2E,F)] The comparison between calculated and experimental thickness profiles is under-specified. Eq. (1) is a time-evolution equation, but the text and Fig. 2E,F do not state the drainage time t used for the computed lines, nor the initial/boundary conditions or the numerical values of σ, η, and ρ. Without these, the apparent agreement cannot be reproduced or quantitatively assessed, and the claim that the profiles are 'well fitted' is not fully verifiable. Please report t and all parameters, or clarify that the plotted profiles are steady-state solutions.
- [Formation and characterization of MOTFs (atomic zippering)] The 'atomic zippering' mechanism is inferred from a single transferred, FIB-milled cross-section showing no interface (Fig. 3F,I) and from AFM thicknesses below twice the monolayer value (Fig. 3J-M). These observations are also compatible with an adhered bilayer that has interfacial roughness, trapped contamination, or partial interdiffusion. Because the mechanism is used to explain the thickness anomaly and implicitly the full-area coalescence of the MOTFs, it needs direct interfacial evidence (e.g., XPS/EELS line scans, statistical AFM over many samples, or in-situ observation) or it should be clearly labeled a speculative hypothesis.
minor comments (5)
- [Fig. 2G] The Frankel-law fit should report experimental uncertainties, the number of samples, and the exact l_c and Ca values used. The visual 'well fitted' statement would be improved by a quantitative residual analysis, especially for D = 10 and 15 mm where deviations for D = 5 mm are acknowledged.
- [Fig. 3P] The transmittance spectra lack error bars, baseline/subtraction details, and sample-to-sample variation. It would also be useful to specify the film thicknesses corresponding to the two spectra, since the near-constant visible/NIR transmittance is a central optical claim.
- [Eq. (1) and general text] Several mathematical symbols in Eq. (1) appear garbled or nonstandard in the submitted text (e.g., '௫', '௧', ''). Please ensure the equation is typeset with conventional notation and that all variables are defined in the text and caption.
- [References] Reference 28 is dated 2026 (Annu. Rev. Fluid Mech. 58, 111-138, 2026). If this is an in-press article, please add the appropriate online-doi or 'in press' notation; otherwise the citation could not be verified.
- [Acoustic detection (Fig. 4)] The detection limits '80 nm and 2 mPa' should be defined with respect to the measurement noise floor and bandwidth. As written, they appear to be read from the measurement resolution rather than a quantified sensitivity criterion. Also clarify how the (0,1) mode assignment was confirmed against other modes or edge effects.
Circularity Check
No circular derivation: the central fabrication claim and the supporting thickness/vibration analyses are self-contained or anchored to external standards.
full rationale
The derivation chain is not circular. The core claim—centimeter-scale suspended metal and metal-oxide films formed by one-step liquid metal capillary forming—is a direct experimental demonstration via photographs, TEM, AFM, and micro-CT, not a quantity derived from an assumed model. The thickness-profile modeling is anchored to external standards: the Reynolds lubrication equation (Eq. 1) is solved, and the Frankel-zone thickness is compared with Frankel's law hF = k lc Ca^(2/3) using the literature constant k=1.8933. The paper states that the normalized thickness 'can be well fitted by Frankel's law'; this is a consistency test of an external formula, not a fitted-parameter prediction. The MOTF 'atomic zippering' inference from TEM cross-sections and AFM thickness is an interpretation of images, not a claim forced by definition. The membrane tension τ* = 4.60 mN/m is extracted from a standard drumhead relation f0,1 = λ(2πa)^-1 (τ*/ρ*)^1/2 using experimentally measured resonance frequency and areal density, which is parameter extraction, not circular prediction. Self-citations (refs. 26 and 30) provide supporting wetting and solidification observations; they are not load-bearing definitions, uniqueness constraints, or substitutes for the paper's own measurements. The reader-identified concern about full-area continuity of the centimeter-scale MOTF is an evidence-strength limitation, not circularity: the paper does not define the MOTF's existence in terms of the same data used to prove it. The mild ambiguity about the time variable used for the Fig. 2E,F calculated profiles is not, on the text available, a demonstrated fitting-to-data reduction, so under the no-speculation rule it is not scored as circular.
Assumptions & free parameters
free parameters (1)
- drainage time t for model profiles
assumptions (3)
- domain assumption Reynolds lubrication approximation (Eq. 1) with constant effective surface tension σ and viscosity η describes the drainage of the oxide-coated liquid-metal film.
- domain assumption The native oxide thickness in the zipped MOTF equals the literature native-oxide thickness and the film density equals bulk oxide density when computing ρ* and τ*.
- domain assumption The suspended film is a clamped circular membrane obeying the thin-plate/membrane dispersion relation f0,1 = λ0,1(2πa)^-1(τ*/ρ*)^1/2 with λ0,1=4.808.
Cite this review
Pith. "Pith review of Centimeter-scale fully suspended metal and metal oxide thin films by one-step transfer-free liquid metal capillary forming." pith.science (2026). https://pith.science/paper/QNZF3MGD
@misc{pith2026260800112,
author = {Pith},
title = {Pith review of: Centimeter-scale fully suspended metal and metal oxide thin films by one-step transfer-free liquid metal capillary forming},
year = {2026},
howpublished = {\url{https://pith.science/paper/QNZF3MGD}},
note = {Machine review of arXiv:2608.00112}
}
read the original abstract
Fully suspended thin films can decouple substrate effects and provide additional tuning degrees of freedom compared with their substrate-supported counterparts, making them unique platforms for next-generation thin film devices. Here we report one-step, transfer-free and substrate-free fabrication of centimeter-scale ultrathin fully suspended metal and metal oxide film structures via liquid metal capillary forming. We show that, analogous to soap film formation, the instantaneously developed few-nanometer-thick native surface oxide can laminate various liquid metals into micrometer-thick metallic films. Surprisingly, the surfactant-like metal oxide bilayer can survive dewetting-induced liquid metal drainage, forming suspended two-dimensional films featuring an enormous lateral size-to-thickness ratio on the order of 10^7. We further demonstrate rapid prototyping of metallic minimal-surface thin-walled structures and ultra-sensitive acoustic wave detection with these suspended thin film platforms.
Figures
Figures from the paper (1 more)
Reference graph
Works this paper leans on
-
[1]
Materials Science of Thin Films: Deposition and Structure
Ohring, M. Materials Science of Thin Films: Deposition and Structure . (Academic press, 2002)
2002
-
[2]
& Ma rtins, R
Fortunato, E., Barquinha, P. & Ma rtins, R. Oxide semiconductor thin ‐film transistors: A review of recent advances. Adv. Mater. 24, 2945–2986 (2012)
2012
-
[3]
Liu, F. et al. Emerging inorganic compound thin fi lm photovoltaic materials: Progress, challenges and strategies. Mater. Today 41, 120–142 (2020)
2020
-
[4]
Xue, G. et al. Large-area epitaxial growth of transition metal dichalcogenides. Chem. Rev. 124, 9785–9865 (2024)
2024
-
[5]
Huang, Y. et al. Universal mechanical exfoliation of large-area 2D crystals. Nat. Commun. 11, 2453 (2020)
2020
-
[6]
Zavabeti, A. et al. A liquid metal reaction environment for the room-temperature synthesis of atomically thin metal oxides. Science 358, 332–335 (2017). 12
2017
-
[7]
& Liu, J
Li, Q., Du, B.-D., Gao, J.-Y. & Liu, J. Liquid metal gallium-based printing of Cu-doped p- type Ga 2O3 semiconductor and Ga 2O3 homojunction diodes. Appl. Phys. Rev. 10, 011402 (2023)
2023
-
[8]
Kong, M. et al. Ambient printing of native oxides for ultrathin transparent flexible circuit boards. Science 385, 731–737 (2024)
2024
Show all 39 references
-
[9]
Zhao, J. et al. Realization of 2D metals at the ångström thickness limit. Nature 639, 354–359 (2025)
2025
-
[10]
Yi, K. et al. Integration of high- κ native oxides of gallium for two-dimensional transistors. Nat. Electron. 7, 1126–1136 (2024)
2024
-
[11]
Zhang, C., Lu, W., Xu, Y., Zeng, K. & Ho, G. W. Mechanistic formulation of inorganic membranes at the air–liquid interface. Nature 616, 293–299 (2023)
2023
-
[12]
Meyer, J. C. et al. The structure of suspended graphene sheets. Nature 446, 60–63 (2007)
2007
-
[13]
Lee, C., Wei, X., Kysar, J. W. & Hone, J. Meas urement of the elastic properties and intrinsic strength of monolayer graphene. Science 321, 385–388 (2008)
2008
-
[14]
Li, N. et al. Direct observation of highly confined phonon polaritons in suspended monolayer hexagonal boron nitride. Nat. Mater. 20, 43–48 (2021)
2021
-
[15]
Ji, D. et al. Freestanding crystalline oxide pero vskites down to the monolayer limit. Nature 570, 87–90 (2019)
2019
-
[16]
Hou, Y. et al. Tuning instability in suspe nded monolayer 2D materials. Nat. Commun. 15, 4033 (2024)
2024
-
[17]
Sun, X. et al. Enhanced interactions of interlayer excitons in free-stan ding heterobilayers. Nature 610, 478–484 (2022)
2022
-
[18]
Chen, C. et al. Performance of monolayer graphene nanomechanical resonators with electrical readout. Nat. Nanotechnol. 4, 861–867 (2009)
2009
-
[19]
& Lee, C
Jin, T., Kang, J., Su Kim, E., Lee, S. & Lee, C. Suspended single-layer MoS2 devices. J. Appl. Phys. 114, 164509 (2013)
2013
-
[20]
Castellanos-Gomez, A. et al. Deterministic transfer of two-dimensional materials by all-dry viscoelastic stamping. 2D Mater. 1, 011002 (2014)
2014
-
[21]
& Quéré, D
De Gennes, P.-G., Brochard-Wyart, F. & Quéré, D. Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, Waves. (Springer New York, New York, NY, 2004)
2004
-
[22]
Saulnier, L. et al. A study of generation and rupture of soap films. Soft Matter 10, 2899 (2014)
2014
-
[23]
& Vermant, J
Chatzigiannakis, E., Jaensson, N. & Vermant, J. Thin liquid film s: Where hydrodynamics, capillarity, surface stresses and intermolecular forces meet. Curr. Opin. Colloid Interface Sci. 53, 101441 (2021)
2021
-
[24]
& Liu, J
Yuan, B., Sun, X., Wang, H. & Liu, J. Liquid metal bubbles. Appl. Mater. Today 24, 101151 (2021)
2021
-
[25]
& Burton, J
Frazier, S., Jiang, X. & Burton, J. C. How to make a giant bubble. Phys. Rev. Fluids 5, 013304 (2020)
2020
-
[26]
& Liu, J
Tang, J., Zhao, X., Li, J., Zhou, Y. & Liu, J. Liquid metal phagocytosis: Intermetallic wetting induced particle internalization. Adv. Sci. 4, 1700024 (2017)
2017
-
[27]
J., Dickey, M
Larsen, R. J., Dickey, M. D., Whitesides, G. M. & Weitz, D. A. Viscoelastic properties of oxide-coated liquid metals. J. Rheol. 53, 1305–1326 (2009)
2009
-
[28]
& Rio, E
Cantat, I., Pertant, S., Raufaste, C. & Rio, E. Capillary drainage in horizontal soap films: Theoretical review and experimental illustrations. Annu. Rev. Fluid Mech. 58, 111–38 (2026)
2026
-
[29]
& Sonin, A
Langevin, D. & Sonin, A. A. Thinning of soap films. Adv. Colloid Interface Sci. 51, 1–27 (1994)
1994
-
[30]
Tang, J. et al. Unique surface patterns emerging during solidification of liquid metal alloys. Nat. Nanotechnol. 16, 431–439 (2021). 13
2021
-
[31]
Bardet, L. et al. High ‐entropy liquid metal process for transparent ultrathin p‐type gallium oxide. Adv. Funct. Mater. 2425108 (2025)
2025
-
[32]
& Cantat, I
Trégouët, C. & Cantat, I. Inst ability of the one-dimensional thickness profile at the edge of a horizontal foam film and its Plateau border. Phys. Rev. Fluids 6, 114005 (2021)
2021
-
[33]
M., Quere, D
Bruinsma, R., Di Meglio, J. M., Quere, D. & Cohen-Addad, S. Formation of soap films from polymer solutions. Langmuir 8, 3161–3167 (1992)
1992
-
[34]
A., Scheid, B
Van Nierop, E. A., Scheid, B. & Stone, H. A. On the thickness of soap films: An alternative to Frankel’s law. J. Fluid Mech. 602, 119–127 (2008)
2008
-
[35]
Zhao, S., Zhang, J. & Fu, L. Liquid metals: A novel possibility of fabricating 2D metal oxides. Adv. Mater. 33, 2005544 (2021)
2021
-
[36]
Malinovsky, V. K. & Sokolov, A. P. The nature of boson peak in Raman scattering in glasses. Solid State Commun. 57, 757–761 (1986)
1986
-
[37]
E., Larson, D
Thompson, R. E., Larson, D. R. & Webb, W. W. Precise nanometer localization analysis for individual fluorescent probes. Biophys. J. 82, 2775–2783 (2002)
2002
-
[38]
G., Dolleman, R
Steeneken, P. G., Dolleman, R. J., Davidovikj, D., Alijani, F. & Van Der Zant , H. S. J. Dynamics of 2D material membranes. 2D Mater. 8, 042001 (2021)
2021
-
[39]
R., Van Der Donk, A
Scheeper, P. R., Van Der Donk, A. G. H., Olthuis, W. & Ber gveld, P. A review of silicon microphones. Sens. Actuator A-Phys. 44, 1–11 (1994). Acknowledgments: The TEM, AFM, Raman, UV-Vis, XPS characterizations of this project were carried out in the Instrumentation and Service...
1994
Reviewed August 4, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.