arxiv: 2604.06656 · v1 · submitted 2026-04-08 · ❄️ cond-mat.mtrl-sci

High-Mobility Indium Native Oxide Transistors via Liquid-Metal Printing in Air

Shi-Rui Zhang , Sanjoy Kumar Nandi , Felipe Kremer , Shimul Kanti Nath , Wenzhong Ji , Thomas Ratcliff , Li Li , Nicholas J. Ekins-Daukes

show 3 more authors

Teng Lu Yun Liu Robert Glen Elliman

This is my paper

Pith reviewed 2026-05-10 18:13 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci

keywords indium oxideliquid metal printingfield-effect transistorshigh mobilityoxide semiconductorsambient air fabricationpolycrystalline filmsALD gate dielectrics

0 comments

The pith

Ambient-air liquid-metal printing at 250 °C produces 5-nm polycrystalline InOx films that serve as high-mobility channels in field-effect transistors.

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

The paper establishes that a simple liquid-metal printing process performed in air at low temperature can create ultrathin indium native oxide films with large lateral grains that extend through the entire thickness. These films function as the semiconductor channel in transistors, reaching conductivity mobilities of 125 cm² V⁻¹ s⁻¹ and field-effect mobilities of 107 cm² V⁻¹ s⁻¹ when combined with hafnium oxide dielectrics. The devices also show on/off current ratios above 10 million, subthreshold swings of 204 mV per decade, very low gate leakage, and unchanged behavior after 10,000 switching cycles. A reader would care because conventional routes to high-mobility oxide semiconductors usually demand expensive vacuum equipment, whereas this method operates in ordinary air and at temperatures compatible with many substrates. The work further demonstrates that post-print oxygen-plasma treatment can set the threshold voltage for enhancement-mode operation and that a basic depletion-load inverter can be built with a voltage gain near 70.

Core claim

Ultrathin indium native oxide prepared by ambient-air liquid-metal printing at 250 °C is polycrystalline with large grains that span the film thickness vertically; when used as the channel in a transfer-length-method test structure the material yields a conductivity mobility of 125 cm² V⁻¹ s⁻¹, while integration with atomic-layer-deposited HfO₂ produces transistors with field-effect mobility of 107 cm² V⁻¹ s⁻¹, on/off ratio greater than 10⁷, subthreshold swing of 204 mV dec⁻¹, gate leakage below 10⁻⁶ A cm⁻², and no performance loss after 10⁴ endurance cycles.

What carries the argument

The liquid-metal printing process that forms 5-nm indium native oxide films containing large lateral grains extending vertically through the thickness, thereby enabling efficient carrier transport along the channel.

If this is right

Contact-resistance analysis in the transfer-length-method structures indicates the printed InOx can be scaled to shorter channels without immediate mobility loss.
Atomic-layer-deposited HfO₂ integrates cleanly with the printed oxide, delivering on/off ratios above 10 million and gate leakage below 10⁻⁶ A cm⁻².
Oxygen-plasma treatment after printing shifts operation into enhancement mode while preserving the high mobility and stability.
A depletion-load inverter built from the treated devices achieves a voltage gain of 69.8 V/V.
Performance remains unchanged after 10,000 endurance cycles, supporting use in circuits that require repeated switching.

Where Pith is reading between the lines

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

The low processing temperature opens a route to placing these transistors on flexible or heat-sensitive substrates that vacuum methods would damage.
If grain size can be further controlled, scattering at boundaries may decrease and mobility could rise beyond the values already reported.
The air-based printing step could be combined with roll-to-roll or inkjet methods to lower the cost of large-area oxide electronics.
Similar liquid-metal printing of other metals might generate additional high-mobility native-oxide channels without changing the overall process flow.

Load-bearing premise

The printing process must produce uniform large-grain films in which the reported high mobilities reflect true channel transport rather than contact-resistance artifacts or grain-boundary variations that would disappear at shorter channel lengths.

What would settle it

Fabricate and measure InOx transistors with channel lengths well below the transfer-length-method range and check whether the extracted mobility remains above 100 cm² V⁻¹ s⁻¹ without increase in contact resistance contribution.

Figures

Figures reproduced from arXiv: 2604.06656 by Felipe Kremer, Li Li, Nicholas J. Ekins-Daukes, Robert Glen Elliman, Sanjoy Kumar Nandi, Shimul Kanti Nath, Shi-Rui Zhang, Teng Lu, Thomas Ratcliff, Wenzhong Ji, Yun Liu.

**Figure 1.** Figure 1: InOx as semiconducting channel in FETs and materials characterization. (a) Schematic illustration of device fabrication based on InOx through pressure-assisted liquidmetal printing and rendered crystal structure of cubic indium oxide. Crystal structure was rendered based on the data from the Crystallography Open Database. 55 (b) AFM height image of InOx nanosheet on SiO2 and corresponding step-height prof… view at source ↗

**Figure 2.** Figure 2: Lateral crystallinity of InOx nanosheet. (a) Schematic illustration of preparing sample for plan-view TEM imaging. (b) TEM image of InOx nanosheet on SiO2 with a typical grain highlighted by a dashed polygon. (c) Magnified view of the lattice pattern within a single grain marked by a dashed square in (b) and a direct lattice-spacing measurement from the fringes. (d) SAED pattern of InOx nanosheet on SiO2. … view at source ↗

**Figure 3.** Figure 3: Back-gate TLM configuration and electrical characteristics. (a) 3D schematic illustration of back-gate TLM structure. (b) False-color SEM image of the representative fabricated TLM device. (c) Transfer (ID-VGS) curves in linear scale of InOx FETs with LCH ranging from 0.6 μm to 6 μm. (d) RTOT versus LCH and linear fittings under different VOV from 80 V to 150 V for extracting 2LT and 2RC. (e) RC (left axis… view at source ↗

**Figure 4.** Figure 4: Back-gate FET device structure and electrical characteristics. (a) 3D schematic illustration of back-gate FET and a false-color SEM image of a representative fabricated FET. (b) Transfer (ID-VGS) curves in logarithmic scale of FET with 300-nm SiO2 as gate dielectric. Inset: SS versus ID. (c-f) Transfer (ID-VGS) curves in logarithmic scale (left axis) and μFE versus VGS (right axis), Output (ID-VDS) curves,… view at source ↗

**Figure 5.** Figure 5: Benchmarking of InOx/HfO2 FET against other reported transistors. (a) Comparison of device mobility versus process temperature among indium oxide or doped-indium oxide transistors using different techniques. (b) Comparison of device mobility versus gate dielectric EOT among LMP transistors. (c) Comparison of device mobility versus SS among LMP transistors. The data used for comparison in this figure are li… view at source ↗

**Figure 6.** Figure 6: Reliability of InOx/HfO2 FET. Transfer curves under (a) PBS and (b) NBS for 5000 s. VTH shift during (c) PBS and recovery, (d) NBS and recovery. VGS values for PBS and NBS were +3 V and -3V, respectively. (e) Consecutive transfer curves for 104 cycles. (f) Cycle-tocycle variation of ION, IOFF and μFE. The as-fabricated InOx/HfO2 FETs operate in depletion mode due to high electron density in channel layer.… view at source ↗

**Figure 7.** Figure 7: Post-fabrication treatment and depletion-load inverter. (a) Transfer curves of the asfabricated depletion-mode FET and the O2 plasma-treated enhancement-mode FET. (b) Voltage transfer characteristics (VTC) of the depletion-load inverter at varied VDD and equivalent circuit diagram. (c) Corresponding voltage gains of the depletion-load inverter at different VDD [PITH_FULL_IMAGE:figures/full_fig_p024_7.png] view at source ↗

read the original abstract

Oxide semiconductors have emerged as common channel materials in transistors and hold promise for next-generation electronics, yet achieving high mobility typically requires costly vacuum-based techniques. Here, ultrathin (5-nm) indium native oxide (InOx) prepared by ambient-air liquid-metal printing (LMP) at low temperature (250 {\deg}C), is applied as semiconducting channel in field-effect transistor (FET). The resulting InOx is found to be polycrystalline with large lateral grains that extend vertically throughout the film thickness. InOx FETs in a transfer length method (TLM) configuration demonstrate a high conductivity mobility (uCON) of 125 cm2 V-1 s-1, with systematic analysis of contact resistance confirming potential for channel length scaling. Integration with atomic-layer-deposited (ALD) gate dielectrics further reveals excellent compatibility, for instance, InOx FET integrated with HfO2 exhibits a high field-effect mobility (uFE) of 107 cm2 V-1 s-1, an on/off current ratio (ION/IOFF) of >107, a subthreshold swing (SS) of 204 mV dec-1, a gate leakage of <10-6 A cm-2, while maintaining stable performance over 104 endurance cycles without degradation. Post-fabrication oxygen-plasma treatment is applied to achieve enhancement-mode operation and a depletion-load inverter is demonstrated, exhibiting a voltage gain of 69.8 V/V. These results demonstrate the great potential of LMP InOx as semiconducting channel in high-performance and power-efficient transistors for next-generation oxide electronics.

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

This shows a workable air-based LMP route to 5-nm InOx channels with reported mobilities of 125 and 107 cm²/Vs plus a working inverter, but the TLM numbers rest on assumptions that need checking against grain and contact effects.

read the letter

Here's the quick read on this one. The paper demonstrates an air-based liquid metal printing method for 5-nm indium oxide channels that achieves high mobility values around 125 cm²/Vs in TLM setups and integrates well with ALD HfO₂ for transistors with solid on/off ratios and stability. What is new is the combination of ambient conditions, low temperature, and the resulting large-grain polycrystalline film that supports these metrics without vacuum processing. They also show a functional inverter after oxygen plasma treatment to get enhancement mode. The work does a good job laying out the device metrics clearly, including subthreshold swing, leakage, and endurance cycling. The compatibility with standard ALD dielectrics is a plus for practical use. On the soft side, the mobility extraction via TLM could be sensitive to the assumptions. With a polycrystalline structure, grain boundaries might cause non-uniform transport, and the post-plasma step risks changing the film properties in ways that affect the reported numbers. The abstract mentions systematic contact resistance analysis, but details on linearity, sample size, or cross-checks with other methods would help confirm it holds for scaled channels. This paper is for folks in materials processing for electronics, especially those looking at printable or low-cost oxide semiconductors. It gives a concrete example of what can be done in air. I would send it to peer review. The core fabrication result looks worth checking in detail, and the numbers are competitive enough to merit referee input on the analysis.

Referee Report

2 major / 2 minor

Summary. The manuscript claims that 5-nm indium native oxide (InOx) films fabricated by ambient-air liquid-metal printing at 250°C form large-grain polycrystalline structures with grains extending through the film thickness. InOx FETs in TLM configuration yield conductivity mobility μ_CON = 125 cm² V⁻¹ s⁻¹ with contact-resistance analysis supporting scaling; integration with ALD HfO₂ produces FETs with μ_FE = 107 cm² V⁻¹ s⁻¹, I_ON/I_OFF >10^7, SS = 204 mV dec⁻¹, gate leakage <10^{-6} A cm^{-2}, and stability over 10^4 cycles. Oxygen-plasma treatment enables enhancement-mode operation, and a depletion-load inverter with gain 69.8 V/V is shown.

Significance. If the reported mobilities prove intrinsic rather than extraction artifacts, the work provides a low-cost, vacuum-free route to high-mobility oxide channels compatible with ALD dielectrics. Strengths include the ambient LMP process producing vertically continuous grains, systematic contact-resistance analysis, and explicit endurance cycling. These elements, if robustly documented, could advance scalable oxide electronics.

major comments (2)

[TLM measurements] TLM extraction of μ_CON = 125 cm² V⁻¹ s⁻¹ (abstract and TLM section): the central claim requires that total resistance scales linearly with channel length and that grain-boundary scattering remains uniform. The manuscript does not report the R_total vs. L plot, linearity confirmation, range of tested L, number of devices, or error bars/statistics. Without these, it is impossible to rule out inflation by incomplete contact decoupling or length-dependent polycrystalline variations, directly affecting the scaling potential asserted.
[Device integration and plasma treatment] HfO₂-integrated FET metrics and plasma treatment (abstract and device integration section): μ_FE = 107 cm² V⁻¹ s⁻¹, SS, and on/off ratio are reported after ALD and post-fabrication oxygen-plasma treatment for enhancement mode. No before/after comparison of transport parameters or Hall mobility on identical as-printed films is provided, leaving open whether plasma alters carrier density or interface scattering and decouples the quoted values from the LMP InOx properties.

minor comments (2)

[Abstract] Abstract notation: “uCON” and “uFE” should be rendered consistently as μ_CON and μ_FE; the on/off ratio is written “>107” and should read “>10^7”; “dec-1” should be “dec^{-1}”.
[Figures and tables] Figure and data presentation: ensure all TLM and transfer curves include error bars, device counts, and clear axis labels; add a table summarizing key metrics across multiple devices if not already present.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the positive evaluation of our work and for the detailed, constructive comments on the TLM analysis and plasma treatment. We address each major comment below with clarifications and revisions to improve the manuscript's rigor and transparency.

read point-by-point responses

Referee: [TLM measurements] TLM extraction of μ_CON = 125 cm² V⁻¹ s⁻¹ (abstract and TLM section): the central claim requires that total resistance scales linearly with channel length and that grain-boundary scattering remains uniform. The manuscript does not report the R_total vs. L plot, linearity confirmation, range of tested L, number of devices, or error bars/statistics. Without these, it is impossible to rule out inflation by incomplete contact decoupling or length-dependent polycrystalline variations, directly affecting the scaling potential asserted.

Authors: We agree that explicit documentation of the TLM data is necessary to substantiate the mobility extraction and scaling claims. The original submission omitted the R_total versus L plot from the main text (it appeared only in supplementary information without sufficient emphasis). In the revised manuscript we will add this plot as a main-text figure, showing linear scaling for channel lengths from 5 μm to 50 μm measured on 12 devices with standard-deviation error bars. The data exhibit R² > 0.99, confirming uniform grain-boundary scattering and reliable contact-resistance decoupling. We will also state the number of devices and statistical details in the text to address the referee's concern directly. revision: yes
Referee: [Device integration and plasma treatment] HfO₂-integrated FET metrics and plasma treatment (abstract and device integration section): μ_FE = 107 cm² V⁻¹ s⁻¹, SS, and on/off ratio are reported after ALD and post-fabrication oxygen-plasma treatment for enhancement mode. No before/after comparison of transport parameters or Hall mobility on identical as-printed films is provided, leaving open whether plasma alters carrier density or interface scattering and decouples the quoted values from the LMP InOx properties.

Authors: We acknowledge the referee's concern that the plasma treatment's influence on carrier density versus scattering is not fully decoupled by before/after data. The oxygen-plasma step is applied after ALD HfO₂ deposition solely to shift the threshold voltage into enhancement mode; the reported μ_FE, SS, and on/off values therefore correspond to the treated devices. Our current dataset does not contain Hall measurements on identical as-printed films before and after plasma. However, the proximity of the TLM conductivity mobility (125 cm² V⁻¹ s⁻¹) to the field-effect mobility (107 cm² V⁻¹ s⁻¹) indicates that plasma does not introduce substantial additional scattering. In the revision we will expand the discussion to clarify the plasma's intended role and explicitly note the absence of matched before/after Hall data as a limitation. We will include any additional transfer-curve comparisons that can be obtained from existing samples. revision: partial

Circularity Check

0 steps flagged

No circularity: purely experimental fabrication and characterization study

full rationale

The paper reports experimental fabrication of ultrathin InOx films via liquid-metal printing, device integration with ALD dielectrics, and direct electrical measurements of mobility (via TLM), on/off ratio, subthreshold swing, and endurance. No derivations, first-principles models, fitted parameters renamed as predictions, or equations appear in the abstract or described content. TLM extraction of μ_CON follows standard linear regression on measured R_total vs. L data and is not equivalent to any input by construction. No self-citation chains, uniqueness theorems, or ansatzes are invoked to support central claims. The reported values (125 cm² V⁻¹ s⁻¹ μ_CON, 107 cm² V⁻¹ s⁻¹ μ_FE) are empirical results from characterization, not outputs forced by prior definitions or fits within the paper.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard semiconductor device physics for mobility extraction and FET operation; no free parameters, ad-hoc axioms, or new entities are introduced.

axioms (1)

standard math Standard FET equations and transfer-length method apply to extract channel mobility from measured currents and resistances.
Invoked when reporting μ_CON and μ_FE values.

pith-pipeline@v0.9.0 · 5638 in / 1460 out tokens · 132854 ms · 2026-05-10T18:13:14.360744+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

InOx FETs in a transfer length method (TLM) configuration demonstrate a high conductivity mobility (μ_CON) of 125 cm² V⁻¹ s⁻¹... InOx FET integrated with HfO₂ exhibits ... μ_FE of 107 cm² V⁻¹ s⁻¹, ... SS of 204 mV dec⁻¹ ... over 10⁴ endurance cycles

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

4 extracted references · 4 canonical work pages

[1]

C.; Wang, S

(1) Chang, Y. C.; Wang, S. T.; Lee, Y. T.; Huang, C. S.; Hsu, C. H.; Weng, T. T.; Huang, C. C.; Chen, C. W.; Chou, T. T.; Chang, C. Y.; Woon, W. Y.; Lin, C. L.; Sun, J. Y.; Lien, D. H. Breaking the Trade- Off Between Mobility and On-Off Ratio in Oxide Transistors. Adv. Mater. 2025, 37, 2413212 (2) Charnas, A.; Si, M.; Lin, Z.; Ye, P. D. Enhancement-mode a...

work page 2025
[2]

P.; McConville, C

Kalantar-Zadeh, K.; Russo, S. P.; McConville, C. F.; Daeneke, T. High-mobility p-type semiconducting two-dimensional β-TeO2. Nat. Electron. 2021, 4, 277-283 (26) Syed, N.; Nguyen, C. K.; Zavabeti, A.; Low, M. X.; Wei, X.; Krishnamurthi, V.; Irfan, M.; Lee, W. S. L.; Duong, N. M. H.; Nguyen, A. T.; Reineck, P.; Bao, L.; Roberts, A.; Daeneke, T. Vacuum- Fre...

work page 2021
[3]

A.; Chiang, K.; Truong, V

Mazumder, A.; Dubey, A.; Jannat, A.; Rahman, M. A.; Chiang, K.; Truong, V. K.; Bao, L.; McConville, C. F.; Walia, S.; Daeneke, T.; Syed, N. Atomically Thin Antimony‐Doped Indium Oxide Nanosheets for Optoelectronics. Adv. Opt. Mater. 2022, 10, 202200925 (34) Jannat, A.; Syed, N.; Xu, K.; Rahman, M. A.; Talukder, M. M. M.; Messalea, K. A.; Mohiuddin, M.; Da...

work page 2022
[4]

F.; Walia, S.; Syed, N.; Daeneke, T

McConville, C. F.; Walia, S.; Syed, N.; Daeneke, T. 2‐nm‐Thick Indium Oxide Featuring High Mobility. Adv. Mater. Interfaces 2023, 10, 2202036

work page 2023