Pith. sign in

REVIEW 3 major objections 4 minor 21 references

Evaluation of Switching Technologies for Reflective and Transmissive RISs at Sub-THz Frequencies

T0 review · 3 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash

Pith's one-line read This paper claims that Schottky diodes, memristors, liquid metal, phase-change materials, and RF-SOI switches can replace PIN diodes in D-band reconfigurable intelligent surfaces.

desk verdict One measured liquid-metal beamsplitter plus four simulation-only switch studies: the measurement is solid, the simulations are plausible but the 'demonstrated potential' claim overreaches, and the RF-SOI bandwidth numbers are internally inconsistent. read the letter →

arxiv 2504.20175 v1 pith:A3EOFFRF submitted 2025-04-28 cs.ET eess.SP

classification cs.ETeess.SP
keywords reconfigurableintelligentsurfaceD-bandswitchingtechnologiesSchottkydiodememristorliquidmetalphasechangematerialRF-SOI
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

The paper argues that the usual RIS switches—PIN diodes and RF-MEMS—become lossy and slow at D-band, and that five alternatives can take their place: Schottky diodes, memristors, liquid metal, phase-change materials (GeTe), and 45 nm RF-SOI CMOS switches. For each technology it designs a unit cell and checks reflection or transmission across 110–170 GHz with full-wave simulations; for the liquid-metal design it also fabricates a beamsplitter and measures it at 150 GHz. If the simulations are right, these results give hardware designers a menu of practical bit-level phase controls for sub-THz reconfigurable surfaces, which matters because 6G networks are expected to rely on such surfaces for beam steering.

What carries the argument

The carrying objects are the unit cells themselves, each paired with a different switching mechanism. Reflective cells use a planar tightly coupled bowtie dipole with the switch bridging the two halves, so 'ON' and 'OFF' states produce about 180 degrees of phase difference; the liquid-metal cell instead changes the period of a strip grating by pumping EGaIn in and out of microchannels. Transmissive cells use a GeTe-based O-slot patch that toggles between amorphous and crystalline states, and an RF-SOI Fabry-Perot resonator whose two I-shaped rotators are shunted by CMOS switches. In every case the mechanism is 1-bit phase quantization: the surface steers beams by flipping each element between two phase states, and the paper's simulations measure the insertion loss and phase error this quantization introduces.

What would settle it

Fabricate the Schottky, memristor, PCM, and RF-SOI unit cells and measure their two-state S-parameters at 140 GHz. If the ON/OFF phase difference deviates substantially from 180 degrees or insertion/return loss exceeds the simulated values—for example, PCM insertion loss above 1.5 dB at 140 GHz or RF-SOI bandwidth narrower than 121–158 GHz—the paper's central viability claim would be refuted.

Watch

Extended reading notes

Core claim

The paper's central claim, stated in the abstract, is that advanced switching components are viable alternatives to traditional technologies in D-band RISs. Concretely: a Schottky-diode reflective cell and a memristor reflective cell each steer a $20\times20$ aperture to 30 degrees; a liquid-metal grating acts as a reconfigurable beamsplitter, splitting one normally incident beam into three or five beams as channel spacing grows from 2 mm to 4 or 6 mm, confirmed by measurement at 150 GHz; a GeTe (germanium telluride) phase-change transmissive cell gives two states separated by 180 degrees with 0.69 dB insertion loss at 140 GHz and under 1.5 dB over 27% bandwidth; and an RF-SOI (silicon-on-insulator) transmissive cell offers a 37 GHz transmission band with less than 2 dB return loss. All non-liquid-metal results are simulation-based.

Load-bearing premise

The weakest load-bearing premise is that the full-wave simulations, with their assumed substrate losses, metal properties, and switch equivalent circuits, accurately predict how real D-band devices behave; only the liquid-metal beamsplitter has measurement behind it.

Editorial extensions

If this is right

  • D-band RISs can be designed without PIN diodes or MEMS: Schottky and memristor cells show simulated beam steering to 30 degrees with only the phase error expected from 1-bit quantization.
  • Phase-change transmissive RISs can reach low insertion loss at 140 GHz—0.69 dB at center, below 1.5 dB over a 27% bandwidth—while retaining non-volatile state retention.
  • RF-SOI CMOS switches support a wide 37 GHz transmission band (121–158 GHz) with less than 2 dB return loss, making them suitable for broadband D-band operation.
  • Liquid-metal gratings can serve as reconfigurable beamsplitters at 150 GHz, with the number of output beams controlled by the channel period; this is the only design in the paper confirmed by measurement.

Reading between the lines

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

  • Because all non-liquid-metal results rest on full-wave simulation with assumed switch equivalent circuits, a natural next step is to fabricate the Schottky, memristor, PCM, and RF-SOI cells and compare measured reflection/transmission phase and loss; this would tighten or revise the switch models.
  • The technologies differ sharply in control speed: Schottky and RF-SOI switch in nanoseconds or faster, while liquid metal and PCM are slower but non-volatile, suggesting hybrid RIS designs could pair a fast electronic array with a non-volatile reconfiguration layer.
  • The same unit-cell topologies might extend beyond the D-band, but switch parasitics and substrate losses grow with frequency, so the viability ranking could change near 300 GHz; the Fabry-Perot transmitarray reference already points to 300 GHz operation with PCB technology.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The manuscript evaluates five switching technologies for D-band (110-170 GHz) reconfigurable intelligent surfaces (RISs): Schottky diodes, memristor switches, liquid metal, phase-change materials (PCM), and 45 nm RF-SOI. It presents unit-cell and small-array full-wave simulations for reflective (Schottky, memristor, liquid metal) and transmissive (PCM, RF-SOI) designs, with 1-bit phase control in the electronic and PCM cases, and reports one experimental validation for a liquid-metal-based reconfigurable beamsplitter at 150 GHz. The abstract and conclusion claim that these advanced components are viable alternatives to traditional PIN-diode or varactor approaches for sub-THz RISs.

Significance. If the reported simulations faithfully represent realistic device behavior, the paper provides a useful comparative view of emerging switching technologies for D-band RIS design, a topic of current interest for 6G. The liquid-metal beamsplitter is a concrete, measured demonstration at 150 GHz, and the RF-SOI design gives explicit switch equivalent-circuit values. However, the central claim of viability rests primarily on simulation-only results: four of the five technologies have no experimental backing, and for three of those the switch models are not disclosed. The paper's value as an 'evaluation' is therefore limited unless the simulation inputs are provided and the conclusions are scaled to match the evidence.

major comments (3)
  1. [Section II-B2, last paragraph before Fig. 12] The text states that Fig. 12(a) indicates "a wide transmission bandwidth of 30 GHz with less than 2 dB return loss over 121 GHz-158 GHz with absolute bandwidth of 37 GHz." This is internally inconsistent: the range 121-158 GHz spans 37 GHz, and in a transmissive unit-cell the relevant loss metric is insertion loss (|S21|), not return loss. A return loss below 2 dB would indicate poor matching. The sentence needs to be rewritten to specify which S-parameter is meant (presumably |S21| < 2 dB over 121-158 GHz) and to reconcile the 30 GHz and 37 GHz bandwidth numbers.
  2. [Sections II-A1, II-A2, II-B1] The Schottky diode, memristor, and PCM switch models are not specified. For the Schottky design the text mentions only that a diode is integrated; for the memristor no equivalent-circuit parameters are given; for the PCM design the GeTe switches are described qualitatively (amorphous/crystalline) but the electrical conductivity or sheet resistance in each state is not reported. These parameters are direct inputs to the full-wave simulations and determine the simulated insertion loss and phase response. Without them, the reader cannot assess whether the simulated performance reflects realistic D-band devices or idealized switch behavior, which is load-bearing for the claim that these technologies are viable alternatives. Please report the switch models and parameter values for all three cases.
  3. [Section III, Conclusion] The conclusion states "We provided both simulation and experimental validation of various RIS architectures," but the only experimental validation in the paper is the liquid-metal beamsplitter at 150 GHz. The Schottky, memristor, PCM, and RF-SOI designs are simulation-only. This overstatement should be corrected so that the conclusion accurately reflects the validation status of each technology, and the abstract should be aligned as well if it implies broader experimental support.
minor comments (4)
  1. [Section II-B1, Fig. 9 description] The text says "losses below 1.5 dB over a 27% bandwidth" and also reports 0.69 dB at 140 GHz; please state explicitly whether the 27% bandwidth refers to the condition that insertion loss stays below 1.5 dB or to another criterion, and define the center frequency used for the percentage.
  2. [Section II-A3, Fig. 5] The phrase "each liquid channel has an in/outlet" should read "inlet/outlet," and the sentence "which is much higher than water and ten times below copper" is awkward; it should be rewritten for clarity, e.g., "its conductivity is much higher than that of water and about one order of magnitude below that of copper."
  3. [Section II-B2, Fig. 12 caption] The caption in the text reads "(a) Unit-cell with an air-gap of 50 µm (b) S-parameter over 110-180 GHz"; it would be clearer to state which S-parameters are shown in (b) (e.g., |S21| and phase) for the two switch states.
  4. [Section II-A2, memristor paragraph] The claim that memristors provide "non-volatile switching ... eliminating the need for continuous voltage application" could be misread as eliminating any voltage during operation; non-volatile devices still require a programming pulse to change state. The wording can be made more precise.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the unit-cell and array results are full-wave simulations with stated design inputs, plus one independent liquid-metal measurement; the central claim does not reduce to its own assumptions.

full rationale

The paper is an engineering evaluation of five switching technologies for D-band RISs. Its central claim, that Schottky diodes, memristors, liquid metal, phase-change materials, and RF-SOI are viable alternatives, rests on unit-cell designs simulated in HFSS or CST, with the liquid-metal beamsplitter additionally measured at 150 GHz. I find no step in which a predicted quantity is defined in terms of the target result, no fitted parameter renamed as a prediction, and no load-bearing self-citation chain. The RF-SOI switch values (Ron=6.13 Omega, Con=18.5 fF, Roff=4300 Omega, Coff=19.0 fF) are stated as approximate equivalent-circuit inputs, not extracted from the simulated transmission response. Likewise, the PCM unit-cell's 180-degree phase difference is a symmetric design outcome of the ON/OFF GeTe states, not a fitted reproduction of the S-parameters. The Schottky and memristor unit-cells are simulated with assumed diode/memristor behavior; the paper does not disclose the full switch models, which is a reproducibility and correctness-risk concern, but not circularity. The citation [19] for the in-house MATLAB synthesis model is a self-citation in the broad sense of the TERRAMETA community, but it is not load-bearing: the same unit-cell is independently evaluated with HFSS, and the paper reports a cross-tool comparison (16.6 dBi in MATLAB vs 15.8 dBi in HFSS), so the claim does not reduce to the cited prior work. There is no uniqueness theorem imported from the authors' earlier papers and no ansatz smuggled in via citation. The liquid-metal measurement provides external, non-fitted validation of the simulation methodology. Under the stated review rules, a self-contained comparison against simulations and one experiment is the typical honest finding, so the circularity score is 0.

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

The central claims rest on unverified full-wave simulations for four of five designs, with switch models and material parameters taken as inputs. Only the liquid metal beamsplitter has measured validation, so the ledger is small but the unmeasured simulations carry the burden.

free parameters (1)
  • RF-SOI switch equivalent circuit values = Ron=6.13 Ω, Con=18.5 fF, Roff=4300 Ω, Coff=19.0 fF
    These values are used as inputs in the CST simulation in Section II-B2; no derivation or citation is given, and they directly determine the reported transmission phase and loss.
assumptions (3)
  • domain assumption Local periodicity principle
    Invoked in Section II to justify analyzing each unit-cell as part of an infinite periodic array; standard approximation in metasurface design.
  • domain assumption Full-wave solver accuracy
    HFSS and CST results are treated as correct for unmeasured designs in Sections II-A and II-B; no experimental verification for four of five technologies.
  • domain assumption Switch models represent real D-band behavior
    The RF-SOI RC model and Schottky/memristor switching behavior are assumed valid at D-band without measured verification; see Section II-B2.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Evaluation of Switching Technologies for Reflective and Transmissive RISs at Sub-THz Frequencies." pith.science (2026). https://pith.science/paper/A3EOFFRF

@misc{pith2026250420175,
  author       = {Pith},
  title        = {Pith review of: Evaluation of Switching Technologies for Reflective and Transmissive RISs at Sub-THz Frequencies},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A3EOFFRF}},
  note         = {Machine review of arXiv:2504.20175}
}
read the original abstract

For the upcoming 6G wireless networks, reconfigurable intelligent surfaces are an essential technology, enabling dynamic beamforming and signal manipulation in both reflective and transmissive modes. It is expected to utilize frequency bands in the millimeter-wave and THz, which presents unique opportunities but also significant challenges. The selection of switching technologies that can support high-frequency operation with minimal loss and high efficiency is particularly complex. In this work, we demonstrate the potential of advanced components such as Schottky diodes, memristor switches, liquid metal-based switches, phase change materials, and RF-SOI technology in RIS designs as an alternative to overcome limitations inherent in traditional technologies in D-band (110-170 GHz).

Figures

Figures reproduced from arXiv: 2504.20175 by the authors.

Figure 1
Figure 1. (a) Configuration of the designed element and simulated (b) magnitude [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. 2D RCS of the metasurface scanning at an angle of 30 [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. (a) Configuration of the designed element and simulated (b) magnitude [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: 2D RCS of the metasurface scanning at an angle of 30 [PITH_FULL_IMAGE:figures/full_fig_p003_4.png]
Figure 6
Figure 6. Figure 6: Simulation results of the proposed LM structure at various frequencies [PITH_FULL_IMAGE:figures/full_fig_p003_6.png]
Figure 8
Figure 8. Figure 8: Concept of the PCM-based T-RIS unit-cell and its proposed fabrication [PITH_FULL_IMAGE:figures/full_fig_p004_8.png]
Figure 7
Figure 7. Figure 7: (a) Measurement setup and (b) results at 150 GHz for different [PITH_FULL_IMAGE:figures/full_fig_p004_7.png]
Figure 9
Figure 9. Figure 9: Simulated scattering parameters of the proposed reconfigurable unit-cell [PITH_FULL_IMAGE:figures/full_fig_p005_9.png]
Figure 11
Figure 11. Figure 11: Front and side view of the design T-RIS unit-cell ( [PITH_FULL_IMAGE:figures/full_fig_p005_11.png]
Figure 12
Figure 12. Figure 12: (a) Unit-cell with an air-gap of 50 µm (b) S-parameter over 110- 180 GHz. III. CONCLUSION In this paper, we presented several innovative switching technologies for RIS, encompassing both reflective and trans￾missive designs, that operate within the D-band. We pro￾vide…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 17 canonical work pages

  1. [19]

    Design of a binary programmable transmitarray based on phase change material for beam steering applications in D- band,

    S. Gharbieh et al. , “Design of a binary programmable transmitarray based on phase change material for beam steering applications in D- band,” Sci. Rep., vol. 14, no. 2966, 2024

  2. [1]

    Reconfigurable intelligent surfaces for energy effi- ciency in wireless communication,

    C. Huang et al. , “Reconfigurable intelligent surfaces for energy effi- ciency in wireless communication,” IEEE Trans. Wireless Commun. , vol. 18, no. 8, pp. 4157–4170, 2019

  3. [2]

    RIS-enabled smart wireless environ- ments: deployment scenarios, network architecture, bandwidth and area of influence,

    G. C. Alexandropoulos et al. , “RIS-enabled smart wireless environ- ments: deployment scenarios, network architecture, bandwidth and area of influence,” EURASIP J. Wireless Commun. Netw., vol. 103, pp. 1–38, 2023

  4. [3]

    Terahertz modeling (THz); identification of use cases for THz com- munication systems,

    “Terahertz modeling (THz); identification of use cases for THz com- munication systems,” ETSI GR THz 001, ETSI, 2024

  5. [4]

    Characterization of indoor RIS-assisted channels at 304 GHz: Experimental measurements, challenges, and future directions,

    G. C. Alexandropoulos et al., “Characterization of indoor RIS-assisted channels at 304 GHz: Experimental measurements, challenges, and future directions,” arXiv preprint:2412.07359, 2024

  6. [5]

    Reconfigurable intelligent surfaces for THz: Signal processing and hardware design challenges,

    ——, “Reconfigurable intelligent surfaces for THz: Signal processing and hardware design challenges,” in Proc. EUCAP, Glasgow, Scotland, 2024

  7. [6]

    Reconfigurable intelligent surfaces for 6G: Emerging hardware architectures, applications, and open challenges,

    E. Basar et al., “Reconfigurable intelligent surfaces for 6G: Emerging hardware architectures, applications, and open challenges,” IEEE Veh. Technol. Mag., vol. 19, no. 3, pp. 27–47, 2024

  8. [7]

    Reconfigurable intelligent surfaces for THz: Hardware impairments and switching technologies,

    S. Matos et al., “Reconfigurable intelligent surfaces for THz: Hardware impairments and switching technologies,” in Proc. ICEAA , Lisbon, Portugal, 2024

Show all 21 references
  1. [8]

    Holographic MIMO communications: Theoretical foun- dations, enabling technologies, and future directions,

    T. Gong et al., “Holographic MIMO communications: Theoretical foun- dations, enabling technologies, and future directions,” IEEE Commun. Surveys & Tuts., vol. 26, no. 1, pp. 196–257, 2024

  2. [9]

    Hybrid reconfigurable intelligent meta- surfaces: Enabling simultaneous tunable reflections and sensing for 6G wireless communications,

    G. C. Alexandropoulos et al., “Hybrid reconfigurable intelligent meta- surfaces: Enabling simultaneous tunable reflections and sensing for 6G wireless communications,” IEEE Veh. Technol. Mag., vol. 19, no. 1, pp. 75–84, 2024

  3. [10]

    A high-speed programmable and scalable terahertz holographic metasurface based on tiled cmos chips,

    S. Venkatesh, X. Lu, H. Saeidi, and K. Sengupta, “A high-speed programmable and scalable terahertz holographic metasurface based on tiled cmos chips,” Nature electronics, vol. 3, no. 12, pp. 785–793, 2020

  4. [11]

    Non-volatile memristor-based 1-bit reconfigurable intelligent surface towards a greener 6g,

    M. Elsaid and L. M. Pessoa, “Non-volatile memristor-based 1-bit reconfigurable intelligent surface towards a greener 6g,” in 2024 18th European Conference on Antennas and Propagation (EuCAP) . IEEE, 2024, pp. 1–5

  5. [12]

    Terahertz reconfigurable in- telligent surfaces (riss) for 6g communication links,

    F. Yang, P. Pitchappa, and N. Wang, “Terahertz reconfigurable in- telligent surfaces (riss) for 6g communication links,” Micromachines, vol. 13, no. 2, p. 285, 2022

  6. [13]

    Phased antenna arrays based on non-volatile resistive switches,

    M. Dragoman, M. Aldrigo, and G. Adam, “Phased antenna arrays based on non-volatile resistive switches,” IET Microwaves, Antennas & Propagation, vol. 11, no. 8, pp. 1169–1173, 2017

  7. [14]

    Design of millimeter wave microstrip reflectarrays,

    D. M. Pozar, S. D. Targonski, and H. Syrigos, “Design of millimeter wave microstrip reflectarrays,” IEEE transactions on antennas and propagation, vol. 45, no. 2, pp. 287–296, 1997

  8. [15]

    Reconfigurable intelligent surfaces: Principles and opportu- nities,

    Y . Liu, X. Liu, X. Mu, T. Hou, J. Xu, M. Di Renzo, and N. Al- Dhahir, “Reconfigurable intelligent surfaces: Principles and opportu- nities,” IEEE communications surveys & tutorials , vol. 23, no. 3, pp. 1546–1577, 2021

  9. [16]

    Air-bridged schottky diodes for dynamically tunable millimeter- wave metamaterial phase shifters,

    E. Vassos, J. Churm, J. Powell, C. Viegas, B. Alderman, and A. Fere- sidis, “Air-bridged schottky diodes for dynamically tunable millimeter- wave metamaterial phase shifters,” Scientific Reports, vol. 11, no. 1, p. 5988, 2021

  10. [17]

    Sub-terahertz transmissive reconfigurable in- telligent surface for integrated beam steering and self-ook-modulation,

    D. Shen, F. Lan, L. Wang, T. Song, M. Yang, T. Hu, Y . Li, X. Nie, J. Yang, S. Liang et al., “Sub-terahertz transmissive reconfigurable in- telligent surface for integrated beam steering and self-ook-modulation,” Light: Science & Applications , vol. 14, no. 1, p. 13, 2025

  11. [18]

    Reversible switching in phase-change materials,

    W. Wełnic and M. Wuttig, “Reversible switching in phase-change materials,” materials today, vol. 11, no. 6, pp. 20–27, 2008

  12. [20]

    140–220 GHz SPST and SPDT switches in 45 nm CMOS SOI,

    M. Uzunkol and G. M. Rebeiz, “140–220 GHz SPST and SPDT switches in 45 nm CMOS SOI,” IEEE Microw. Wireless Compon. Lett., vol. 22, no. 8, p. 412–414, 2012

  13. [21]

    Experimental demonstration of a 43-dBi gain transmitarray in PCB technology for backhauling in the 300-GHz band,

    O. Koutsos, F. F. Manzillo, M. Caillet, R. Sauleau, and A. Clemente, “Experimental demonstration of a 43-dBi gain transmitarray in PCB technology for backhauling in the 300-GHz band,” IEEE Trans. Tera- hertz Sci. Technol., vol. 13, no. 5, pp. 485–492, 2023

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.