arxiv: 2605.09108 · v1 · submitted 2026-05-09 · ❄️ cond-mat.mtrl-sci

Recognition: 2 theorem links

· Lean Theorem

Band alignment of grafted diamond/GaN p-n heterojunctions interfaced with ALD Al2O3 and SiNx/Al2O3

Aaron Hardy, Carolina Adamo, Chenyu Wang, Connor S Bailey, Dong Liu, Jiarui Gong, Jung-Hun Seo, Katherine Fountaine, Luke Suter, Matthias Muehle, Michael Eller, Stephanie Liu, Tsung-Han Tsai, Vincent Gambin, Xuanyu Zhou, Yang Liu, Zhenqiang Ma

Authors on Pith no claims yet

Pith reviewed 2026-05-12 01:49 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci

keywords diamondGaNband alignmentheterojunctionXPSinterfacial engineeringp-n junctionALD

0 comments

The pith

Inserting an SiNx layer increases band offsets by 0.42 eV in diamond/GaN heterojunctions with Al2O3.

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

The paper establishes that diamond/Al2O3/GaN and diamond/Al2O3/SiNx/GaN heterostructures both show type-II band alignment when measured by X-ray photoelectron spectroscopy. Adding the SiNx layer makes the band offsets 0.42 eV larger, which the authors link to a change in interfacial electrostatic potential from fewer positive fixed charges near the diamond/Al2O3 interface. This matters because diamond and GaN have complementary doping limitations, so controlling band alignment is key to making efficient p-n diodes. The results indicate that engineering the interfacial dielectric stack offers a practical way to adjust junction properties for different device needs.

Core claim

Both the diamond/Al2O3/GaN and diamond/Al2O3/SiNx/GaN structures exhibit type-II band alignment, but the band offsets in the version with the inserted SiNx layer are larger by 0.42 eV. This difference is attributed to a modification of the interfacial electrostatic potential, which may arise from a reduced density of positive fixed charges in the interfacial dielectric near the diamond/Al2O3 interface after insertion of the SiNx layer.

What carries the argument

X-ray photoelectron spectroscopy measurement of valence and conduction band offsets across the grafted interfaces, with SiNx insertion modifying the electrostatic potential in the ALD dielectric stack.

Load-bearing premise

The 0.42 eV difference in band offsets arises primarily from reduced positive fixed charge density near the diamond/Al2O3 interface due to SiNx insertion, rather than interface dipoles, defect states, or XPS artifacts.

What would settle it

Direct measurement of fixed charge density in the Al2O3 layer showing no reduction after SiNx addition, or finding identical band offsets in both structures.

Figures

Figures reproduced from arXiv: 2605.09108 by Aaron Hardy, Carolina Adamo, Chenyu Wang, Connor S Bailey, Dong Liu, Jiarui Gong, Jung-Hun Seo, Katherine Fountaine, Luke Suter, Matthias Muehle, Michael Eller, Stephanie Liu, Tsung-Han Tsai, Vincent Gambin, Xuanyu Zhou, Yang Liu, Zhenqiang Ma.

**Figure 1.** Figure 1: Process flow for the fabrication of grafted diamond/GaN heterostructures used for band-alignment studies. (a) Release of the SCD NM by hydrogen ion implantation followed by EC. In panel (a3), Pt and W denote platinum and tungsten needles used to supply the electrical current.(b) Preparation of n-GaN host substrates with (b1) Al₂O₃ and (b2) Al₂O₃/SiNₓ interfacial layers, followed by formation of the grafted… view at source ↗

**Figure 2.** Figure 2: Prepared samples used for XPS characterization: (a) bare n-GaN reference sample; heterostructure samples of (b1) diamond/Al₂O₃/GaN and (b2) diamond/Al₂O₃/SiNₓ/GaN; and interface samples of (c1) grafted diamond/Al₂O₃/GaN and (c2) diamond/Al₂O₃/SiNₓ/GaN [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗

**Figure 3.** Figure 3: XPS spectra used for band alignment analysis. (a1) Ga 3d XPS spectrum of the bare n-GaN reference sample with deconvoluted components. (a2) Valence-band XPS spectrum of the bare n-GaN reference sample with extraction of the VBM value. (b1) C 1s XPS spectrum of the SCD NM from the grafted diamond/Al₂O₃/GaN heterostructure with deconvoluted components. (b2) Valence-band XPS spectrum of the SCD NM from the gr… view at source ↗

read the original abstract

Diamond and gallium nitride are complementary semiconductors for forming p-n junctions because of their respective doping limitations. Understanding the band alignment of grafted diamond/GaN heterojunctions is therefore essential for optimizing diode performance. In this study, the band alignment of diamond/Al2O3/GaN and diamond/Al2O3/SiNx/GaN heterostructures was determined by X-ray photoelectron spectroscopy. Both structures exhibit type-II band alignment, but with different band offsets. The band offsets of the diamond/Al2O3/SiNx/GaN heterojunction are larger by 0.42 eV than those of diamond/Al2O3/GaN. This difference is attributed to a modification of the interfacial electrostatic potential, which may arise from a reduced density of positive fixed charges in the interfacial dielectric near the diamond/Al2O3 interface after insertion of the SiNx layer. These results demonstrate that interfacial-layer engineering provides an effective strategy for tailoring the band alignment of grafted diamond/GaN heterojunctions, offering guidance for the design of p-n diodes with tunable rectifying characteristics.

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

XPS measurements show a 0.42 eV increase in band offsets for the diamond/Al2O3/SiNx/GaN stack versus diamond/Al2O3/GaN, both type-II, but the fixed-charge mechanism is not directly tested.

read the letter

The central result here is straightforward XPS data on grafted diamond/GaN heterojunctions: both the Al2O3-only and Al2O3/SiNx versions show type-II alignment, but the offsets are 0.42 eV larger when the SiNx layer is present. The authors link this to a change in interfacial electrostatic potential, possibly from fewer positive fixed charges near the diamond/Al2O3 boundary after SiNx insertion. This is new quantitative information for these specific grafted stacks and ALD layers, extending earlier work on diamond and GaN interfaces. For device work in power electronics, where diamond and GaN are paired because of their complementary doping limits, the numbers give a concrete tuning option via interfacial engineering. The experimental execution looks competent; XPS is the right tool for band offsets, and the direct comparison between the two structures is clean. The paper does a good job reporting the measured values without overclaiming the device implications. The soft spot is the mechanism. The attribution to reduced fixed charge density is plausible but rests on inference rather than direct evidence. No C-V, Kelvin probe, or charge-density measurements are referenced to isolate that effect, and alternatives such as interface dipoles or small XPS referencing shifts from the SiNx are not addressed. The abstract itself uses cautious language (“may arise”), which matches the strength of the data. If the full manuscript includes solid error bars, detailed fitting procedures, and controls for surface contamination or band-bending artifacts, that would tighten the result. As it stands, the offset difference itself is the reliable part; the explanation is secondary. This work is aimed at researchers growing or modeling wide-bandgap p-n junctions, especially those already working with grafted or ALD-interfaced diamond/GaN. A reader who needs offset values for simulation or diode design will get immediate use from the numbers. The experimental core is solid enough that the paper deserves peer review; referees can check the raw spectra, sample preparation, and whether additional charge measurements would strengthen the interpretation. I would send it forward rather than desk-reject.

Referee Report

2 major / 1 minor

Summary. The manuscript reports X-ray photoelectron spectroscopy (XPS) determination of band alignment in grafted diamond/Al2O3/GaN and diamond/Al2O3/SiNx/GaN heterostructures. Both exhibit type-II alignment, but the diamond/Al2O3/SiNx/GaN structure shows 0.42 eV larger valence- and conduction-band offsets. This difference is attributed to modification of the interfacial electrostatic potential, possibly from reduced positive fixed-charge density near the diamond/Al2O3 interface after SiNx insertion. The work concludes that interfacial-layer engineering can tailor band alignment for diamond/GaN p-n diodes.

Significance. If the measured offset difference is robust, the results provide concrete experimental values for band alignment in a technologically relevant wide-bandgap heterosystem where direct p-n formation is limited by doping constraints. The demonstration that an ALD SiNx interlayer can shift offsets by 0.42 eV offers a practical route for band-engineering in grafted junctions, with potential impact on diode rectification and carrier injection. The grafting-plus-ALD approach itself is device-relevant.

major comments (2)

[Abstract and Discussion] Abstract and Discussion section: The attribution of the 0.42 eV offset increase to reduced positive fixed-charge density at the diamond/Al2O3 interface is presented as the primary explanation, yet remains inferential. No C-V, Kelvin-probe, or direct charge-density measurements are reported to isolate fixed charge from possible interface-dipole changes, defect-state shifts, or SiNx-induced chemical effects on XPS core-level referencing. This mechanism is load-bearing for the claim that interfacial engineering provides an 'effective strategy' for tunable rectifying characteristics.
[Results and Experimental] Results and Experimental sections: The abstract and main text provide no quantitative error bars, standard deviations, or details on XPS peak-fitting procedures, inelastic mean-free-path assumptions, or controls for surface band bending and contamination. Without these, the statistical significance of the reported 0.42 eV difference cannot be assessed, undermining the central experimental claim.

minor comments (1)

[Abstract] Abstract: The hedging phrase 'may arise' is appropriate, but the abstract would benefit from a clearer separation between the directly measured offsets and the proposed (unverified) mechanism.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive feedback and positive assessment of the significance of our XPS band-alignment measurements in grafted diamond/GaN heterostructures. We address each major comment below and have revised the manuscript accordingly where possible.

read point-by-point responses

Referee: [Abstract and Discussion] Abstract and Discussion section: The attribution of the 0.42 eV offset increase to reduced positive fixed-charge density at the diamond/Al2O3 interface is presented as the primary explanation, yet remains inferential. No C-V, Kelvin-probe, or direct charge-density measurements are reported to isolate fixed charge from possible interface-dipole changes, defect-state shifts, or SiNx-induced chemical effects on XPS core-level referencing. This mechanism is load-bearing for the claim that interfacial engineering provides an 'effective strategy' for tunable rectifying characteristics.

Authors: We agree that the proposed mechanism remains inferential, as the manuscript does not include direct measurements of fixed charge (C-V or Kelvin probe). The original text already employs cautious phrasing ('may arise from'), but we accept that the discussion could more explicitly acknowledge alternative contributions such as interface dipoles or SiNx-induced chemical shifts. In the revised manuscript we will expand the Discussion section to list these alternatives, moderate the abstract and conclusion to emphasize the experimentally observed 0.42 eV offset increase rather than the specific charge-density hypothesis, and retain the claim that interfacial-layer engineering is effective on the basis of the measured shift itself. We do not possess C-V or Kelvin-probe data for this study. revision: partial
Referee: [Results and Experimental] Results and Experimental sections: The abstract and main text provide no quantitative error bars, standard deviations, or details on XPS peak-fitting procedures, inelastic mean-free-path assumptions, or controls for surface band bending and contamination. Without these, the statistical significance of the reported 0.42 eV difference cannot be assessed, undermining the central experimental claim.

Authors: We acknowledge this omission. The revised manuscript will include quantitative error bars on all reported band-offset values (derived from XPS peak-fitting uncertainties and replicate measurements), standard deviations where multiple samples were measured, and expanded Experimental-section details on peak-fitting routines, inelastic-mean-free-path values and sources, and procedures used to minimize or correct for surface band bending and adventitious contamination. These additions will enable readers to evaluate the significance of the 0.42 eV difference. revision: yes

Circularity Check

0 steps flagged

No circularity: experimental XPS band-offset measurements with inferential attribution only

full rationale

The manuscript is a pure experimental study reporting X-ray photoelectron spectroscopy (XPS) core-level shifts to extract valence-band offsets for diamond/Al2O3/GaN and diamond/Al2O3/SiNx/GaN stacks. The 0.42 eV difference is stated as an observation and then attributed (with hedging language 'may arise') to possible fixed-charge modification; no equations, fitted parameters, predictions, or derivations are presented. No self-citations are invoked to justify uniqueness theorems or ansatzes. The result is therefore self-contained against external benchmarks (XPS data) and receives the default non-circularity score.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the standard assumption that XPS valence band spectra can be used to extract band offsets in oxide-interfaced heterostructures without significant artifacts from surface contamination or charging.

axioms (1)

domain assumption XPS measurements accurately reflect bulk band offsets at the buried interfaces after accounting for surface effects
Invoked implicitly in the determination of band alignment from photoelectron spectra.

pith-pipeline@v0.9.0 · 5563 in / 1312 out tokens · 46346 ms · 2026-05-12T01:49:51.185769+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/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
The band offsets of the diamond/Al2O3/SiNx/GaN heterojunction are larger by 0.42 eV than those of diamond/Al2O3/GaN. This difference is attributed to a modification of the interfacial electrostatic potential, which may arise from a reduced density of positive fixed charges...
IndisputableMonolith/Foundation/AbsoluteFloorClosure.lean reality_from_one_distinction unclear
Using the Kraut method... VBO ... 0.63 eV ... CBO ... 2.70 eV

Reference graph

Works this paper leans on

9 extracted references · 9 canonical work pages

[1]

Prospectives for AlN electronics and optoelectronics and the important role of alternative synthesis,

1 W. A. Doolittle, C. M. Matthews, H. Ahmad, K. Motoki, S. Lee, A. Ghosh, and P. D. Yoder, “Prospectives for AlN electronics and optoelectronics and the important role of alternative synthesis,” Appl. Phys. Lett. 123, 070501 (2023). 2 S. J. Pearton, F. Ren, M. Tadjer, and J. Kim, “Perspective: Ga₂O₃ for ultra -high power rectifiers and MOSFETs,” J. Appl. ...

work page 2023
[2]

Characteristics of native oxides-interfaced GaAs/Ge np diodes,

Wang, F. Xia, and Z. Ma, “Characteristics of native oxides-interfaced GaAs/Ge np diodes,” IEEE Electron Device Lett. 45, 1669 (2024). 19 J. Zhou, H. Wang, P.R. Huang, S. Xu, Y. Liu, J. Gong, J. Shen, D. Vicent, S. Haessly, A. Abrand, P.K. Mohseni, M. Kim, S.- Q. Yu, G. -E. Chang, X. Gong, and Z. Ma, “GaAs/GeSn/Ge n–i –p diodes and light emitting diodes fo...

work page 2024
[3]

Lattice -mismatched semiconductor heterostructures,

Baboli, J. Park, J. Bong, I.-K. Lee, J. Gong, S. Mikael, J. H. Ryu, P. K. Mohseni, X. Li, S. Gong, X. Wang, and Z. Ma, “Lattice -mismatched semiconductor heterostructures,” arXiv:1812.10225 (2018). 21 J. Gong, D. Kim, H. Jang, F. Alema, Q. Wang, J. Zhou, Y. Li, T.K. Ng, S. Qiu, Y. Liu, M

work page arXiv 2018
[4]

Characteristics of grafted monocrystalline Si/β-Ga2O3 p–n heterojunction,

Sheikhi, Y. Lu, R. Singh, X. Su, H.N. Abbasi, Q. Lin, S. Xie, K. Chabak, G. Jessen, C. Cheung, V. Gambin, S.S. Pasayat, A. Osinsky, B.S. Ooi, C. Gupta, and Z. Ma, “Characteristics of grafted monocrystalline Si/β-Ga2O3 p–n heterojunction,” Appl. Phys. Lett. 124, 262101 (2024). 22 J. Gong, J. Zhou, A. Dheenan, M. Sheikhi, F. Alema, T. K. Ng, S. S. Pasayat, ...

work page arXiv 2024
[5]

Band alignment and leakage mechanism analysis of p -Si/n-AlN heterojunction diodes with the Al ₂O₃ interlayer,

Liu, Z. Liu, M. Nong, K. Fountaine, V. Gambin, X. Li, and Z. Ma, “Band alignment and leakage mechanism analysis of p -Si/n-AlN heterojunction diodes with the Al ₂O₃ interlayer,” ACS Appl. Electron. Mater. 7, 9700–9709 (2025). 29 J. Zhou, H. Wang, Y. Guo, A. Abrand, Y. Li, Y. Liu, J. Gong, P.R. Huang, J. Shen, and S. Xu, “Optoelectronically Active GaAs/GeS...

work page arXiv 2025
[6]

Grafted GaAs/Si Heterojunction Tunnel Diodes with Improved Peak- to-valley Current Ratio (PVCR) of up to 36.38,

Lu, T. - H. Tsai, Y. Li, V. Gambin, T. K. Ng, B. S. Ooi, and Z. Ma, “Grafted GaAs/Si Heterojunction Tunnel Diodes with Improved Peak- to-valley Current Ratio (PVCR) of up to 36.38,” IEEE Electron Device Lett. 1–1 (2025). 31 J. Zhou, Y. Lu, C. Wang, L. Suter, A. Hardy, T. K. Ng, K. Sun, Y. Guo, Y. Liu, T. -H. Tsai, X

work page 2025
[7]

Preliminary demonstration of diamond– GaN p–n diodes via grafting,

Zhou, C. S. Bailey, M. Eller, S. Liu, Z. Mi, B. S. Ooi, M. Muehle, K. Fountaine, V. Gambin, J.-H. Seo, and Z. Ma, “Preliminary demonstration of diamond– GaN p–n diodes via grafting,” arXiv 2510.25028 (2025). 32 E. A. Kraut, R. W. Grant, J. R. Waldrop, and S. P. Kowalczyk, “Precise determination of the valence-band edge in x-ray photoemission spectra: Appl...

work page arXiv 2025
[8]

Low -temperature processing to obtain cont act resistance of <0.03 Ohm mm to boron-doped diamond,

Myszka, M. A. Hollis, T. G. Ivanov, A. G. Birdwell, D. Ruzmetov, J. Weil, N. R. Jankowski, T. A. Grotjohn, and B. Zhang, “Low -temperature processing to obtain cont act resistance of <0.03 Ohm mm to boron-doped diamond,” Phys. Status Solidi A 222, 2400636 (2025). 36 A. Hardy, M. Muehle, C. Herrera-Rodriguez, M. Becker, E. Drown, N. Baule, M. Tompkins, T. ...

work page 2025
[9]

Investigation of ultrathin surface passivation layers for GaN: a comparative analysis of Al₂O₃, SiO₂, and SiNₓ in reducing surface recombination,

Liu, D. Liu, S. Qiu, B. S. Ooi, I. Roqan, and Z. Ma, “Investigation of ultrathin surface passivation layers for GaN: a comparative analysis of Al₂O₃, SiO₂, and SiNₓ in reducing surface recombination,” Appl. Phys. Lett. 127, 011601 (2025). 50 Z. Yatabe, J. Asubar, and T. Hashizume, “Insulated gate and surface passivation structures for GaN-based power tran...

work page 2025