Pith. sign in

REVIEW 4 major objections 5 minor 48 references

XHEMTs on Ultrawide Bandgap Single-Crystal AlN Substrates

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A transistor built on single-crystal AlN delivers 5.92 W/mm at 10 GHz by using a silicon donor sheet to fill the polarization hole gas that otherwise causes current collapse.

desk verdict A credible first RF demonstration of thin-channel AlN/GaN/AlN HEMTs on bulk AlN, with an empirical improvement that stands even though the Pauli-blocking mechanism is underdetermined. read the letter →

arxiv 2506.16670 v1 pith:57EJ4ZKZ submitted 2025-06-20 cond-mat.mtrl-sci physics.app-ph

Eungkyun Kim , Yu-Hsin Chen , Naomi Pieczulewski , Jimy Encomendero , David Anthony Muller , Debdeep Jena , Huili Grace Xing This is my paper
classification cond-mat.mtrl-sciphysics.app-ph
keywords AlNXHEMTsingle-crystalsubstratesGaHEMTstwo-dimensionalholegascurrentcollapseRFpoweramplifierssilicondeltadopingpseudomorphicchannel
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper introduces the AlN single-crystal high electron mobility transistor (XHEMT): a thin, pseudomorphic GaN channel sandwiched between AlN layers and grown on a single-crystal AlN substrate. The central claim is that inserting a sheet of silicon donors near the bottom GaN/AlN interface compensates the polarization-induced two-dimensional hole gas (2DHG), which eliminates the severe current collapse seen in otherwise identical undoped devices. First-generation XHEMTs reach 5.92 W/mm output power and 65% peak power-added efficiency at 10 GHz under a 17 V drain bias, performance on par with state-of-the-art GaN HEMTs but on a platform with far fewer dislocations and much better heat extraction. The result matters because it offers a path to thin-channel, high-frequency nitride transistors that do not need thick, defect-laden buffer layers.

What carries the argument

The load-bearing object is the XHEMT epitaxial stack: a 20 nm coherently strained GaN channel between an AlN top barrier and an AlN buffer on a single-crystal AlN substrate, plus a silicon $\delta$-doping sheet placed one nanometer above the bottom GaN/AlN interface. The silicon donors supply electrons that compensate the net negative polarization bound charge at that interface, converting what would be a 2DHG into a filled valence band. The named mechanism is Pauli blocking: with the valence-band states at the bottom interface already occupied, hot electrons can no longer be captured there during RF swings, so the 2DEG is not partially depleted. The same doping raises the two-dimensional electron gas density through modulation doping, and the resulting lower vertical field in the quantum well reduces interface-roughness scattering.

What would settle it

Perform deep-level transient spectroscopy or admittance spectroscopy on the bottom GaN/AlN interface in both undoped QW HEMTs and Si-$\delta$-doped XHEMTs: if current collapse disappears while a hole-related trap signature at that interface remains unchanged, or if a different electron trap governs the dispersion, the Pauli-blocking explanation is falsified. A second check is channel-thickness scaling: a thin doped channel should stay dispersion-free, whereas the same doping in a thick channel should show the known low dispersion regardless of the hole gas.

Watch

Extended reading notes

Core claim

The central discovery is that charge trapping in AlN/GaN/AlN quantum-well transistors can be switched off by doping, not by thickening the buffer. In the undoped structure, polarization charges at the bottom GaN/AlN interface create a 2DHG roughly 18 to 20 nm below the electron channel; under large-signal RF drive, hot electrons are captured into empty valence-band states at that interface and take milliseconds or longer to escape, collapsing the drain current (over 30% collapse under a -5 V/20 V stress bias, and output power limited to 0.90 W/mm with 20% PAE at 8 GHz). Adding a silicon $\delta$-doping sheet of $5\times10^{13}$ cm$^{-2}$ one nanometer above the bottom interface fills those valence-band states, Pauli-blocking the electron capture, and simultaneously raises the 2DEG density by modulation doping. The resulting AlN XHEMT shows negligible current collapse, a sheet resistance reduction of about 37% (to roughly 307 ohm per square), and continuous-wave output power of 5.92 W/mm with 65% peak PAE at 10 GHz under 17 V bias. The authors present this as the first large-signal operation of a double-heterostructure HEMT with a channel at or below 20 nm, enabled by single-crystal AlN substrates with dislocation densities below $10^4$ cm$^{-2}$.

Load-bearing premise

The load-bearing assumption is that the current collapse in the undoped devices comes from hot electrons being captured into empty valence-band states at the bottom GaN/AlN interface, and that the silicon sheet removes the collapse by filling those states; if trapping actually has a different cause, the empirical improvement could still hold but the design principle would need to be revised.

Editorial extensions

If this is right

  • First-generation XHEMTs already deliver RF output power and peak PAE on par with state-of-the-art GaN HEMTs while operating at a modest 17 V drain bias, improving on identical undoped devices by roughly 6x in output power and 3x in PAE.
  • Dispersion-free operation becomes possible in pseudomorphic GaN channels 20 nm or thinner, opening the double-heterostructure design space that thicker relaxed channels could not enter without reintroducing dislocations.
  • Because the AlN substrate contributes dislocation densities near $10^4$ cm$^{-2}$ and high thermal conductivity, the platform promises thinner epitaxial layers and much lower thermal boundary resistance than GaN on SiC or silicon.
  • The remaining limit on output power is off-state breakdown at the gate edge; improving electric-field management should raise the usable drain bias and push output power further.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same donor-compensation trick could plausibly be transferred to other polarization-charge interfaces, such as AlGaN back barriers, N-polar heterostructures, or other III-nitride channels, wherever a fixed sheet of compensating charge sits below the electron channel.
  • If the Pauli-blocking mechanism is correct, thinning the GaN channel toward its pseudomorphic limit of a few nanometers should not reintroduce RF dispersion, because the bottom-interface states remain filled; this is a testable channel-scaling prediction.
  • Since the evidence for the 2DHG trapping path is indirect, a direct trap-spectroscopy measurement on undoped versus doped samples would separate the Pauli-blocking effect from any incidental improvement in material quality caused by the doping step.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper reports the epitaxial growth, fabrication, and characterization of AlN/GaN/AlN double-heterostructure high-electron-mobility transistors ('XHEMTs') on single-crystal AlN substrates, comparing an undoped quantum-well HEMT with a Si delta-doped XHEMT. The authors show that the undoped structure exhibits a second capacitance plateau attributed to a polarization-induced two-dimensional hole gas at the bottom GaN/AlN interface, and that this structure suffers severe current collapse in pulsed I-V and load-pull measurements. Introducing a Si delta-doping sheet nominally eliminates the 2DHG, raises the 2DEG density and mobility, reduces current collapse, and yields a maximum output power density of 5.92 W/mm with 65% peak power-added efficiency at 10 GHz under a 17 V drain bias. The central mechanistic claim is that Si donors fill empty valence-band states at the bottom interface, Pauli-blocking electron capture that would otherwise cause RF dispersion. The paper also benchmarks the device against published GaN HEMTs and discusses the advantages of the AlN substrate platform.

Significance. If the reported device performance and its mechanistic interpretation hold up, this work represents an important first demonstration of high-frequency large-signal operation of pseudomorphic AlN/GaN/AlN transistors on bulk single-crystal AlN substrates. The platform promises very low dislocation density, high thermal conductivity, and elimination of thermal boundary resistance and impurity-doped buffers, which are genuine advantages over conventional GaN-on-SiC or GaN-on-Si transistors. The direct comparison between undoped and Si delta-doped structures with otherwise identical epitaxy and processing provides a controlled experimental test of the role of the 2DHG, and the measured 6-fold improvement in output power is a strong empirical result. The paper also includes useful materials characterization (STEM, Hall, C-V) and a transparent, detailed fabrication section.

major comments (4)
  1. [§2.3, Fig. 4(b),(e)] The Pauli-blocking mechanism is the load-bearing claim of the paper but is explicitly presented as a hypothesis ('We hypothesize...'). The experimental evidence—disappearance of the C-V second plateau and reduced pulsed I-V dispersion—is consistent with the hypothesis but does not uniquely support it. Si delta-doping also reduces the average vertical field in the GaN channel from 3.80 to 2.84 MV/cm (§2.1) and increases the 2DEG density by roughly 75%, both of which could independently suppress trapping or reduce its relative impact on drain current. To make the mechanistic claim convincing, the authors should provide more direct evidence for the trap states, for example temperature-dependent pulsed I-V, gate-lag/transient measurements with different quiescent biases, or low-frequency noise spectroscopy, or alternatively soften the claim to state that Si doping suppresses dispersion by a combination of 2DHG compensation, field reduction, and channel-density increase.
  2. [§2.3, Fig. 4(c),(f)] The large-signal comparison between the undoped QW HEMT and the AlN XHEMT is confounded by different measurement frequencies (8 GHz vs. 10 GHz), different quiescent drain currents (115 vs. 260 mA/mm), and different load- and source-reflection coefficients. Since current collapse and gain compression are strongly frequency- and bias-dependent, the reported 6-fold and 3-fold improvements in Pout and PAE may partly reflect these differing measurement conditions rather than the trapping mechanism alone. The authors should measure the undoped device at the same frequency and, ideally, at comparable bias conditions to support the direct comparison.
  3. [§2.3, §4 (large-signal characterization)] The central RF result appears to be based on a single device for each structure, with no indication of device-to-device variability or confidence intervals. Given that the benchmarking claims position the AlN XHEMT as a record-holder at VDSq ≤ 20 V, single-device data are insufficient for a quantitative claim. At minimum, the authors should report measurements from at least three devices per structure and state the statistical spread, and they should state explicitly if load-pull was performed on only one device.
  4. [§2.1, Fig. 2(d)] The C-V second plateau in the undoped QW HEMT is attributed specifically to a 2DHG, but a capacitance plateau only demonstrates a mobile sheet charge at a certain depth; it does not identify the charge as holes in the valence band. Earlier work on separately grown samples supports the existence of 2DHGs in AlN/GaN heterostructures, but in the present device stack the identification is indirect. The authors should either provide direct evidence (e.g., Hall or magnetotransport on the undoped QW sample) or explicitly state that the plateau is consistent with, but not proof of, a 2DHG.
minor comments (5)
  1. [Fig. 3 caption and §2.2] The TLM analysis in Figure 3(g) reports Rc = 0.34 Ω·mm, but the text states an average Rc of 0.39 ± 0.04 Ω·mm; please reconcile these numbers and report whether the figure shows a representative or averaged measurement.
  2. [Fig. 3(d)] There is a typo in the figure caption: 'Atomic-resoluation' should read 'Atomic-resolution'.
  3. [Fig. 5 and Table S1] The benchmark plots mix CW and pulsed data points without a clear distinction in all panels; while hollow and solid symbols are defined, the captions should remind the reader that pulsed and CW measurements are not directly comparable, especially for 'Pout at peak PAE' comparisons.
  4. [Abstract and §1] The abstract states that XHEMT performance is 'on par with the state-of-the-art GaN HEMTs', while the benchmarking in §2.4 more carefully places the result as the highest Pout at VDSq ≤ 20 V with a caveat about frequency. Please make the abstract wording consistent with the more qualified benchmark statement.
  5. [§2.3] The phrase 'these are no empty states in the valence band' in the Figure 4 caption should be reworded to 'there are no empty states in the valence band'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the device performance is directly measured, and the mechanistic claim is explicitly labeled a hypothesis rather than derived from fitted inputs.

full rationale

The paper's central results are experimental and self-contained: the AlN XHEMTs are grown, fabricated, and characterized, and the headline values (5.92 W/mm output power, 65% peak PAE at 10 GHz) are direct load-pull measurements, not outputs of a model fitted to the data. The band-structure simulations use standard III-nitride polarization parameters and are validated against independent Hall and C-V measurements; no simulation parameter is tuned to reproduce the RF performance. The existence of the 2DHG is supported both by a measured second C-V plateau and by earlier experimental reports from the same group, but the plateau is independently observed here and the prior reports are used only as corroborating identification, not as a load-bearing derivation of the RF result. The Pauli-blocking mechanism is explicitly introduced as a hypothesis ('We hypothesize...') and is not presented as a mathematical consequence of the measurements, so its underdetermination relative to other effects of Si doping is a question of experimental evidence rather than circular reasoning. No fitted parameter is renamed as a prediction, and no self-citation chain forces the conclusion. The benchmarking against published GaN HEMTs is external comparison, not a circular input. Therefore no circular step meeting the evidentiary standard is present.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The central claim depends on the chosen delta-doping density and on standard III-nitride material parameters used in the simulations. The device metrics themselves are measured, not derived. The main nonstandard assumption is the trapping mechanism, which is inferred from C-V and simulations.

free parameters (1)
  • Si delta-doping density = 5 x 10^13 cm^-2
    A hand-chosen experimental input intended to compensate the net negative polarization sheet charge at the bottom GaN/AlN interface. The central design parameter of the XHEMT; not fitted to the measured RF performance.
assumptions (3)
  • domain assumption Standard III-nitride polarization charges and GaN/AlN band offsets used in the band simulations
    The simulated band diagrams and charge profiles in Figures 2 and 4 rely on literature values for spontaneous and piezoelectric polarization and band offsets; they set the predicted 2DHG density.
  • domain assumption Undoped AlN buffer is semi-insulating with ultrahigh resistivity and low RF loss
    The XHEMT design assumes the AlN buffer does not introduce conductive paths or traps; supported by the authors' previous work (ref [20]), not verified in this paper.
  • domain assumption Single-crystal AlN substrate properties (dislocation density <10^4 cm^-2, thermal conductivity ~340 W/m·K) are as quoted from supplier/literature
    These material advantages motivate the design but are not measured in this work.

how reviews work

0 comments
Cite this review

Pith. "Pith review of XHEMTs on Ultrawide Bandgap Single-Crystal AlN Substrates." pith.science (2026). https://pith.science/paper/57EJ4ZKZ

@misc{pith2026250616670,
  author       = {Pith},
  title        = {Pith review of: XHEMTs on Ultrawide Bandgap Single-Crystal AlN Substrates},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/57EJ4ZKZ}},
  note         = {Machine review of arXiv:2506.16670}
}
abstract

AlN has the largest bandgap in the wurtzite III-nitride semiconductor family, making it an ideal barrier for a thin GaN channel to achieve strong carrier confinement in field-effect transistors, analogous to silicon-on-insulator technology. Unlike SiO$_2$/Si/SiO$_2$, AlN/GaN/AlN can be grown fully epitaxially, enabling high carrier mobilities suitable for high-frequency applications. However, developing these heterostructures and related devices has been hindered by challenges in strain management, polarization effects, defect control and charge trapping. Here, the AlN single-crystal high electron mobility transistor (XHEMT) is introduced, a new nitride transistor technology designed to address these issues. The XHEMT structure features a pseudomorphic GaN channel sandwiched between AlN layers, grown on single-crystal AlN substrates. First-generation XHEMTs demonstrate RF performance on par with the state-of-the-art GaN HEMTs, achieving 5.92 W/mm output power and 65% peak power-added efficiency at 10 GHz under 17 V drain bias. These devices overcome several limitations present in conventional GaN HEMTs, which are grown on lattice-mismatched foreign substrates that introduce undesirable dislocations and exacerbated thermal resistance. With the recent availability of 100-mm AlN substrates and AlN's high thermal conductivity (340 W/m$\cdot$K), XHEMTs show strong potential for next-generation RF electronics.

Figures

Figures reproduced from arXiv: 2506.16670 by the authors.

Figure 1
Figure 1. A tale of three substrates for nitride HEMTs. (a) Spider plot comparing the key material properties of sub [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Undoped QW HEMT on AlN (top row) vs AlN XHEMT (bottom row). Layer structure of the (a) undoped QW [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. AlN XHEMT. (a) Three-dimensional representation and (b) SEM image of a fully fabricated AlN XHEMT. (c) [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Undoped QW HEMT on AlN (top row) vs AlN XHEMT (bottom row), both fabricated using the same process [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Benchmarking AlN XHEMT with metal-polar, single-channel nitride HEMTs on large signal RF power am [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

48 extracted references · 48 canonical work pages

  1. [1]

    R. T. Bondokov, K. Hogan, G. Q. Norbury, S. Matsumoto, J. Grandusky, physica status solidi (b) 2500032

    work page

  2. [2]

    R. T. Bondokov, S. P. Branagan, N. Ishigami, J. Grandusky, T. Nagatomi, K. Tatsuta, T. Miebach, J. J. Chen, ECS Meeting Abstracts 2021, MA2021-02, 34 985

    work page 2021

  3. [3]

    Zhang, M

    Z. Zhang, M. Kushimoto, T. Sakai, N. Sugiyama, L. J. Schowalter, C. Sasaoka, H. Amano, Applied Physics Express 2019, 12, 12 124003

    work page 2019

  4. [4]

    Rounds, B

    R. Rounds, B. Sarkar, A. Klump, C. Hartmann, T. Nagashima, R. Kirste, A. Franke, M. Bicker- mann, Y. Kumagai, Z. Sitar, R. Collazo, Applied Physics Express 2018, 11, 7 071001

    work page 2018

  5. [5]

    Y.-H. Chen, J. Encomendero, C. Savant, V. Protasenko, H. G. Xing, D. Jena, Applied Physics Let- ters 2024, 124, 15 152111

    work page 2024

  6. [6]

    Y.-H. Chen, J. Encomendero, C. Savant, V. Protasenko, H. G. Xing, D. Jena, Applied Physics Let- ters 2024, 125, 14 142110

    work page 2024

  7. [7]

    M. Qi, G. Li, S. Ganguly, P. Zhao, X. Yan, J. Verma, B. Song, M. Zhu, K. Nomoto, H. G. Xing, D. Jena, Applied Physics Letters 2017, 110, 6 063501

    work page 2017

  8. [8]

    Zhang, J

    Z. Zhang, J. Encomendero, R. Chaudhuri, Y. Cho, V. Protasenko, K. Nomoto, K. Lee, M. Toita, H. G. Xing, D. Jena, Applied Physics Letters 2021, 119, 16 162104

    work page 2021

Show all 48 references
  1. [9]

    K. J. Chen, O. H¨ aberlen, A. Lidow, C. l. Tsai, T. Ueda, Y. Uemoto, Y. Wu, IEEE Transactions on Electron Devices 2017, 64, 3 779

    2017

  2. [10]

    Y. Tang, K. Shinohara, D. Regan, A. Corrion, D. Brown, J. Wong, A. Schmitz, H. Fung, S. Kim, M. Micovic, IEEE Electron Device Letters 2015, 36, 6 549

    2015

  3. [11]

    K. M. Bothe, S. Ganguly, J. Guo, Y. Liu, A. Niyonzima, O. Tornblad, J. Fisher, D. A. Gajewski, S. T. Sheppard, B. Noori, IEEE Electron Device Letters 2022, 43, 3 354

    2022

  4. [12]

    Ishikawa, G.-Y

    H. Ishikawa, G.-Y. Zhao, N. Nakada, T. Egawa, T. Jimbo, M. Umeno, Japanese Journal of Applied Physics 1999, 38, 5A L492

    1999

  5. [13]

    Lawrence Selvaraj, T

    S. Lawrence Selvaraj, T. Suzue, T. Egawa, IEEE Electron Device Letters 2009, 30, 6 587

    2009

  6. [14]

    M. J. Uren, J. Moreke, M. Kuball, IEEE Transactions on Electron Devices 2012, 59, 12 3327

    2012

  7. [15]

    Meneghini, I

    M. Meneghini, I. Rossetto, D. Bisi, A. Stocco, A. Chini, A. Pantellini, C. Lanzieri, A. Nanni, G. Meneghesso, E. Zanoni, IEEE Transactions on Electron Devices 2014, 61, 12 4070

    2014

  8. [16]

    Sarua, H

    A. Sarua, H. Ji, K. P. Hilton, D. J. Wallis, M. J. Uren, T. Martin, M. Kuball, IEEE Transactions on Electron Devices 2007, 54, 12 3152

    2007

  9. [17]

    Manoi, J

    A. Manoi, J. W. Pomeroy, N. Killat, M. Kuball, IEEE Electron Device Letters 2010, 31, 12 1395

    2010

  10. [18]

    S. G. Mueller, R. T. Bondokov, K. E. Morgan, G. A. Slack, S. B. Schujman, J. Grandusky, J. A. Smart, L. J. Schowalter, physica status solidi (a) 2009, 206, 6 1153

    2009

  11. [19]

    Alvarez-Escalante, R

    G. Alvarez-Escalante, R. Page, R. Hu, H. G. Xing, D. Jena, Z. Tian, APL Materials 2022, 10, 1 011115

    2022

  12. [20]

    E. Kim, Z. Zhang, J. Encomendero, J. Singhal, K. Nomoto, A. Hickman, C. Wang, P. Fay, M. Toita, D. Jena, H. G. Xing, Applied Physics Letters 2023, 122, 9 092104. 17

    2023

  13. [21]

    C. F. C. Chang, J. E. Dill, Z. Zhang, J.-C. Chen, N. Pieczulewski, S. J. Bader, O. A. Valenzuela, S. A. Crooker, F. F. Balakirev, R. D. McDonald, J. Encomendero, D. A. Muller, F. Giustino, D. Jena, H. G. Xing, Quantum oscillations of holes in GaN, 2025

    2025

  14. [22]

    Chaudhuri, S

    R. Chaudhuri, S. J. Bader, Z. Chen, D. A. Muller, H. G. Xing, D. Jena, Science 2019, 365, 6460 1454

    2019

  15. [23]

    Binari, K

    S. Binari, K. Ikossi, J. Roussos, W. Kruppa, D. Park, H. Dietrich, D. Koleske, A. Wickenden, R. Henry, IEEE Transactions on Electron Devices 2001, 48, 3 465

    2001

  16. [24]

    Vetury, N

    R. Vetury, N. Zhang, S. Keller, U. Mishra, IEEE Transactions on Electron Devices 2001, 48, 3 560

    2001

  17. [25]

    W. Li, Z. Hu, K. Nomoto, D. Jena, H. G. Xing, Unleashing the promise of gallium oxide, 2019, Compound Semiconductor

    2019

  18. [26]

    Kim, Y.-H

    E. Kim, Y.-H. Chen, J. Encomendero, D. Jena, H. G. Xing, In 2024 Device Research Conference (DRC). 2024 1–2

    2024

  19. [27]

    Malmros, P

    A. Malmros, P. Gamarra, M. Thorsell, H. Hjelmgren, C. Lacam, S. L. Delage, H. Zirath, N. Rors- man, IEEE Transactions on Electron Devices 2019, 66, 1 364

    2019

  20. [28]

    Harrouche, S

    K. Harrouche, S. Venkatachalam, L. Ben-Hammou, F. Grandpierron, E. Okada, F. Medjdoub, Mi- cromachines 2023, 14, 2

    2023

  21. [29]

    S. Kaun, M. Wong, J. Lu, U. Mishra, J. Speck, Electronics Letters 2013, 49, 14 893

    2013

  22. [30]

    Hickman, R

    A. Hickman, R. Chaudhuri, L. Li, K. Nomoto, S. J. Bader, J. C. M. Hwang, H. G. Xing, D. Jena, IEEE Journal of the Electron Devices Society 2021, 9 121

    2021

  23. [31]

    Kotani, K

    J. Kotani, K. Makiyama, T. Ohki, S. Ozaki, N. Okamoto, Y. Minoura, M. Sato, N. Nakamura, Y. Miyamoto, Electronics Letters 2023, 59, 4 e12715

    2023

  24. [32]

    Micovic, P

    M. Micovic, P. Hashimoto, M. Hu, I. Milosavljevic, J. Duvall, P. Willadsen, W.-S. Wong, A. Con- way, A. Kurdoghlian, P. Deelman, J.-S. Moon, A. Schmitz, M. Delaney, In IEDM Technical Digest. IEEE International Electron Devices Meeting, 2004. 2004 807–810

    2004

  25. [33]

    Schafer, M

    S. Schafer, M. Litchfield, A. Zai, Z. Popov ´ ıc, C. Campbell, In 2013 IEEE MTT-S International Mi- crowave Symposium Digest (MTT) . 2013 1–3

    2013

  26. [34]

    Sardin, T

    D. Sardin, T. Reveyrand, Z. Popovi´ c, In 2014 44th European Microwave Conference . 2014 1337– 1340

    2014

  27. [35]

    C. Lee, P. Saunier, H.-Q. Tserng, In IEEE Compound Semiconductor Integrated Circuit Symposium,

  28. [36]

    P.-C. Chao, K. Chu, C. Creamer, J. Diaz, T. Yurovchak, M. Shur, R. Kallaher, C. McGray, G. D. Via, J. D. Blevins, IEEE Transactions on Electron Devices 2015, 62, 11 3658

    2015

  29. [37]

    Y.-F. Wu, A. Saxler, M. Moore, R. Smith, S. Sheppard, P. Chavarkar, T. Wisleder, U. Mishra, P. Parikh, IEEE Electron Device Letters 2004, 25, 3 117

    2004

  30. [38]

    X. Luo, S. Halder, W. R. Curtice, J. C. M. Hwang, K. D. Chabak, D. E. Walker, A. M. Dabiran, In 2011 IEEE International Symposium on Radio-Frequency Integration Technology . 2011 209–212

    2011

  31. [39]

    M. Peng, Y. Zheng, W. Luo, X. Liu, Solid-State Electronics 2011, 64, 1 63

    2011

  32. [40]

    H.-Q. Tao, W. Hong, B. Zhang, X.-M. Yu, IEEE Microwave and Wireless Components Letters 2017, 27, 1 73

    2017

  33. [41]

    Resca, A

    D. Resca, A. Raffo, S. Di Falco, F. Scappaviva, V. Vadal` a, G. Vannini, IEEE Microwave and Wire- less Components Letters 2014, 24, 4 266. 18

    2014

  34. [42]

    Piotrowicz, Z

    S. Piotrowicz, Z. Ouarch, E. Chartier, R. Aubry, G. Callet, D. Floriot, J. Jacquet, O. Jardel, E. Morvan, T. Reveyrand, N. Sarazin, S. Delage, In 2010 IEEE MTT-S International Microwave Symposium. 2010 505–508

    2010

  35. [43]

    Piotrowicz, E

    S. Piotrowicz, E. Morvan, R. Aubry, S. Bansropun, T. Bouvet, E. Chartier, T. Dean, O. Drisse, C. Dua, D. Floriot, M. A. diForte Poisson, Y. Gourdel, A. J. Hydes, J. C. Jacquet, O. Jardel, D. Lancereau, J. O. Mc Lean, G. Lecoustre, A. Martin, Z. Ouarch, T. Reveyrand, M. Richard...

    2008

  36. [44]

    K. Chu, P. Chao, M. Pizzella, R. Actis, D. Meharry, K. Nichols, R. Vaudo, X. Xu, J. Flynn, J. Dion, G. Brandes, IEEE Electron Device Letters 2004, 25, 9 596

    2004

  37. [45]

    J. S. Moon, H. Moyer, P. Macdonald, D. Wong, M. Antcliffe, M. Hu, P. Willadsen, P. Hashimoto, C. McGuire, M. Micovic, M. Wetzel, D. Chow, In 2012 IEEE Topical Conference on Power Ampli- fiers for Wireless and Radio Applications . 2012 9–12

    2012

  38. [46]

    Ozaki, J

    S. Ozaki, J. Yaita, A. Yamada, Y. Kumazaki, Y. Minoura, T. Ohki, N. Okamoto, N. Nakamura, J. Kotani, Applied Physics Express 2021, 14, 4 041004

    2021

  39. [47]

    Kotani, J

    J. Kotani, J. Yaita, K. Homma, S. Ozaki, A. Yamada, M. Sato, T. Ohki, N. Nakamura, IEEE Jour- nal of the Electron Devices Society 2023, 11 101. 19

    2023

  40. [2005]

    2005 157–160

    CSIC ’05. 2005 157–160

    2005

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.