Eigenmode analysis of a half-mode uniplanar metamaterial-inspired substrate integrated waveguide

Kyriakos Katsaros-Gkouskos; Maria-Thaleia Passia; Traianos V. Yioultsis

REVIEW 2 major objections 5 minor 8 references

Eigenmode analysis of a half-mode uniplanar metamaterial-inspired substrate integrated waveguide

T0 review · 2 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read A half-mode, via-free substrate integrated waveguide built on complementary split-ring resonators can be cut to nearly half its transverse size while keeping the same propagation and loss performance as the full-width version.

desk verdict A clean, useful eigenmode study of a half-mode CSRR SIW that plausibly halves the footprint at similar loss, but the radiation-loss modeling is underspecified and the headline comparison rests on simulation only. read the letter →

arxiv 2607.17403 v1 pith:3KV4HG2W submitted 2026-07-19 physics.app-ph physics.comp-ph

Maria-Thaleia Passia , Kyriakos Katsaros-Gkouskos , Traianos V. Yioultsis This is my paper · ORCID

classification physics.app-phphysics.comp-ph

keywords half-modesubstrateintegratedwaveguidecomplementarysplit-ringresonatorseigenmodeanalysiscomplexpropagationconstantdispersiondiagramattenuationmmWavetransmissionlinemetamaterial-inspired

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

Half-mode uniplanar CSRR SIW is a transmission line that replaces metalized via rows with a single row of complementary split-ring resonators and then cuts the full waveguide in half along the magnetic-wall plane of its dominant mode. The paper's central claim is that this halved line keeps the performance of the full-width line: near 26 GHz it has similar or slightly lower attenuation and a similar propagation constant. This matters because the design is cheaper to fabricate than a via-based SIW and more compact than the existing full CSRR version, easing integration in mmWave circuits. To establish the claim, the paper solves an omega-k eigenproblem that returns both propagation and attenuation constants with dielectric, conductor, and radiation losses included, and it tunes the ring width to a low-loss operating point.

What carries the argument

The two load-bearing pieces are (1) the complementary split-ring resonator (CSRR), a pair of concentric slots etched in the ground plane that acts as an electric wall and replaces a row of metalized vias, and (2) the half-mode principle: the full waveguide's TE10 mode has its electric-field maximum at the center, so the center plane behaves as a magnetic wall; slicing there leaves a half-width waveguide whose dominant mode is TE0.5,0. The eigenmode formulation treats the propagation constant as the eigenvalue of a quadratic eigenvalue problem, allowing both beta and alpha to be extracted instead of a fixed-frequency beta-only sweep.

What would settle it

Fabricate the half-mode and full CSRR SIW unit cells or transmission lines with the stated dimensions and measure their insertion loss over 23–29 GHz; if the half-mode line's measured attenuation constant exceeds the full line's by more than the simulation predicts, the central claim fails.

Watch

Extended reading notes

Core claim

The central claim is that a substrate integrated waveguide can be cut in half along the symmetry plane of its dominant TE10 mode, with the open side acting as a magnetic wall, and the resulting half-mode line — whose electric walls are formed by a single row of grounded complementary split-ring resonators instead of metalized vias — shows similar or slightly lower attenuation and a similar propagation constant to the full-width uniplanar CSRR SIW around 26 GHz. The paper supports this by solving an omega-k eigenproblem for the complex wavenumber k = beta − j alpha with dielectric, conductor, and radiation losses included, and by a parametric study of the CSRR ring width c that selects c = 0.

Load-bearing premise

The comparison assumes the eigenmode simulation correctly captures radiation loss from the open side of the half-mode waveguide; if it under-counts that loss, the half-mode design's similarly low attenuation may not hold in practice.

Editorial extensions

If this is right

Half-mode CSRR SIWs can be fabricated with ordinary PCB lithography and no metalized vias, reducing both cost and footprint.
The design offers roughly half the transverse size of the full uniplanar CSRR SIW at similar attenuation near 26 GHz.
The parametric result that increasing CSRR ring width shifts the low-loss band upward provides a direct tuning rule for synthesizing the line.
An omega-k eigenmode solve with all losses included can be used in place of fabrication-intensive cut-and-measure iterations to screen future half-mode metamaterial waveguides.

Reading between the lines

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

If radiation loss from the open center aperture is indeed the dominant extra loss mechanism, then enclosing or optimizing that aperture (e.g., with a superstrate) could push the half-mode design below the full-mode losses at frequencies well above 26 GHz.
The magnetic-wall splitting argument should degrade for narrow substrates where the width-to-height ratio is small; a numerical sweep of w/h would show the validity envelope.
The same split-along-the-magnetic-wall construction can be applied to other uniplanar metamaterial SIWs, such as multi-row or nonuniform metasurface walls, potentially yielding further miniaturization.
A two-port measurement of a fabricated prototype, extracting attenuation from S-parameters over 23–29 GHz, would confirm or refute the simulated loss parity and is a natural next step.

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.

Desk Editor's Note

read the letter

Colleague,

The useful core here is concrete: the paper takes the established full uniplanar single-CSRR SIW from the authors' prior work [3] and cuts it along the magnetic wall to make a half-mode version that is roughly half the width. The eigenmode method is the right tool for this — an omega-k formulation with complex propagation constant, implemented in COMSOL, with copper-loss impedance boundary conditions and complex permittivity for dielectric loss. The parametric sweep over the CSRR ring width c and the resulting dispersion and attenuation curves are clean and internally consistent. The comparison against the full version at 26 GHz is the right kind of benchmark: it is anchored to a device that was experimentally characterized in [3], so the central claim is not circular.

What is genuinely new is the half-mode topology itself, not the half-mode concept (HMSIW and HM-CSIW exist) and not the CSRR concept. The value is the specific combination plus the loss-inclusive characterization. If the simulated performance holds, this is a practically useful footprint reduction for mmWave PCB transmission lines. I would call that a legitimate incremental contribution within its subfield, not a breakthrough.

Now the soft spots, in proportion. The main one is the radiation-loss question, and the stress-test note is on target. Section II-B asserts that dielectric, conductor, and radiation losses are all included, but the formulation in II-A only spells out the IBC for conductor loss and complex permittivity for dielectric loss. There is no PML, scattering boundary, or other open-domain radiation model described, and the only geometric comment is that the grounded substrate is slightly extended to enclose fringing fields — which addresses ground-plane leakage, not radiation from the open aperture into air. In an eigenmode solver without an absorbing boundary, radiation is not a loss channel, so the attenuation constant α in Fig. 2(a) could be underestimated for the half-mode structure. Since the full SIW is essentially closed, this biases the comparison in favor of the half-mode. That is a genuine concern, but it is also an addressable one: state the boundary treatment explicitly or add a lossy air domain. I would not call the whole result false — the magnitude of the radiation loss at 26 GHz for this geometry may be small — but the paper currently does not demonstrate it.

Secondary gaps: no mesh-convergence study, no experimental validation (fine for a design-analysis paper, but worth saying), and the data are not released. The number of free parameters in the sweep is modest, and the chosen geometry is clearly presented.

Who benefits: anyone designing via-less SIWs at mmWave, and anyone using eigenmode solvers to characterize lossy periodic transmission lines. The paper deserves a serious referee: the method is sound, the topology is new, and the missing radiation modeling is fixable rather than fatal. Send it out, but ask the authors to either show the radiation-loss treatment or soften the "similar or slightly lower losses" claim to "material-loss limited."

Recommendation: accept for review with request for clarification on the radiation model.

Referee Report

2 major / 5 minor

Summary. The paper presents an eigenmode analysis of a half-mode uniplanar single-CSRR substrate integrated waveguide (SIW). The authors use a finite-element ω−k eigenmode formulation that solves for the complex propagation constant k = β − jα, rather than the more common β−ω approach, and include conductor losses via impedance boundary conditions (Eqs. 2–3) and dielectric losses via complex permittivity. A parametric sweep over CSRR radius, ring width, gap, and half-mode width selects dimensions that place a low-loss window near 26 GHz. Dispersion diagrams for the dominant mode are compared with those of the previously reported full uniplanar single-CSRR SIW. The central claim is that the half-mode design retains the performance of the full design, with similar or slightly lower losses and a similar propagation constant, while occupying roughly half the transverse footprint.

Significance. If the central claim is correct, the proposed half-mode CSRR SIW is a useful compact, via-less alternative for mmWave transmission lines, and the complex-k eigenmode methodology is appropriate for extracting attenuation constants. A notable strength is that the comparison is grounded in a previously experimentally characterized full design [3], so the comparison is not circular. The parametric design study and the inclusion of material losses are also valuable. However, the quantitative loss comparison depends on an unverified radiation-loss model and on numerical convergence, neither of which is demonstrated. The significance is therefore conditional: the design concept and methodology are sound, but the headline loss claim needs additional support.

major comments (2)

[§II-A, Eq. (1)–(3); §II-B] The statement that 'all loss mechanisms, dielectric, conductor, and radiation losses, are considered' is not supported by the described formulation. The eigenproblem uses Floquet periodic boundaries, impedance boundary conditions (2)–(3), and complex permittivity, but no open/absorbing boundary (PML, scattering boundary, or air-domain truncation) is specified for the open aperture at the center plane. Extending the grounded substrate 'to enclose any fringing fields' confines fields through the ground but does not model radiation into the air half-space. Without such a boundary, the complex eigenvalue cannot include a radiation loss channel, so α in Fig. 2(a) may be underestimated, biasing the headline comparison in favor of the half-mode design. Please specify the computational domain/boundary condition and quantify radiation loss (e.g., a comparison with and without an air/PML region),
[Fig. 2 / §II-B] No mesh-convergence study or discretization parameters are reported. Attenuation constants computed with an impedance boundary condition are sensitive to mesh resolution near metal edges and to the skin depth (about 0.4 µm for copper at 26 GHz). Without a convergence check, the quantitative claim of 'similar or slightly lower losses' in Fig. 2 is not robust. Add a convergence study of β and α at the operating point (e.g., a table varying mesh density or near-field refinement), and state the mesh statistics in the final description.

minor comments (5)

[Abstract / §I] Grammar: 'the via are substituted' should be 'the vias are substituted'; similar wording appears in the introduction. Please revise.
[Eq. (1)] The matrices A, B, and C in the generalized eigenvalue problem are not defined or derived. For self-containedness, define them or give a clear reference to the assembly procedure in [7], [8].
[§II-B] The mode designation 'TE0.5,0' is nonstandard. A brief explanation of why a fractional transverse index is used for the half-mode structure would help readers unfamiliar with half-mode SIW terminology.
[Fig. 1(b)] The label 'w=2' in the figure appears inconsistent with the stated half-mode width w=6.5 mm; presumably 'w/2' is intended. Please check the label.
[Conclusion] The conclusion restates the central performance claim without noting the modeling assumptions (radiation-loss modeling, lack of experimental validation of the half-mode structure). A one-sentence caveat would improve accuracy.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: half-mode loss comparison is a self-contained eigenmode simulation against an externally characterized baseline.

full rationale

The paper's derivation chain is an eigenmode analysis: the unknown complex wavenumber k is computed from a stated quadratic eigenvalue problem (Eq. 1) with explicit impedance boundary conditions (Eqs. 2-3) and complex permittivity. The half-mode geometry is obtained by a symmetry cut of the full CSRR SIW, a concept supported by external references [4]-[6], not by a claimed uniqueness theorem. The comparison baseline (uniplanar single-CSRR SIW) comes from the authors' prior experimental work [3], but the current paper re-simulates that baseline with the same solver, so the comparison is not a fitted parameter renamed as a prediction. The geometric parameters are selected by a parametric sweep to place the low-loss window around 26 GHz; this is design optimization, not fitting a target output into an input. The headline claim 'similar or slightly lower losses' is an output of the simulation, not equivalent to the model inputs by construction. The only notable gap is the assertion in Section II-B that radiation losses are considered, while Section II-A specifies no absorbing boundary for the open center aperture; if radiation is under-modeled, the attenuation constant for the half-mode may be underestimated. This is a modeling/completeness caveat, not a circular step, because the comparison is still a genuine computation under the stated (if incomplete) model. Self-citations [3] and [8] provide background and method, but neither is load-bearing in a way that forces the central result. Therefore no significant circularity is found.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The paper's central claim depends on hand-chosen geometric parameters (r,c,g,w, plus an unquantified substrate extension) and on six modeling assumptions. No new physical entity is introduced. The design parameters are not fitted to external data—they are selected to put the low-loss window at 26 GHz—so they are free parameters rather than evidence of data fitting.

free parameters (5)

CSRR external radius r = 0.93 mm
Chosen by parametric analysis to place low losses near 26 GHz (§II-B).
CSRR ring width c = 0.32 mm (swept 0.3-0.4 mm)
Chosen by parametric analysis; the comparison uses half-mode c=0.32 vs full-mode c=0.3 (§II-B, Fig. 2).
CSRR gap g = 0.3 mm
Chosen by parametric analysis (§II-B).
Half-mode waveguide width w = 6.5 mm
Chosen by parametric analysis (§II-B).
Substrate extension beyond center line = not quantified
The text says the grounded substrate is 'slightly extended beyond the center to enclose any fringing fields' (§II-B), but the extent is not given. It affects fringing and radiation loss and is effectively a hand-chosen free parameter.

assumptions (6)

domain assumption The FEM eigenvalue formulation of [7],[8] (Eqs. (1)-(3)) correctly computes the complex propagation constant k=β-jα for the periodic unit cell.
The paper adopts this numerical method without independent validation in this manuscript; it is established in the cited literature, but the implementation is not verified against measurement or a reference solution.
domain assumption Conductor loss is modeled by the impedance boundary condition with Zs=(1+j)Rs, Rs=sqrt(ωμ/2σc), σc=5.8e7 S/m (Eqs. 2-3).
IBC is valid when skin depth is small relative to conductor dimensions; standard at 26 GHz but not stated as a validity check.
domain assumption Dielectric loss is modeled by complex relative permittivity εr=2.18(1-j0.0009).
Material data are taken as given; no uncertainty or temperature dependence is considered.
domain assumption The center plane of the full CSRR SIW behaves as an equivalent magnetic wall, so the half structure supports the dominant mode independently (§II-B, citing [5]).
This is the design premise of all half-mode SIWs; it is imported from [5] and not re-verified for the CSRR-loaded geometry.
ad hoc to paper Radiation loss from the open center aperture is included in the eigenmode solution, although no open-boundary condition is specified.
The text only says the grounded substrate is extended beyond the center to enclose fringing fields; the numerical treatment of the open side is not described, making this a paper-specific unstated modeling premise.
domain assumption The unit-cell Floquet model (periodic field transformation E=ee^{-jk k·r}) represents the propagating mode in the infinite periodic waveguide.
Standard for eigenmode analysis; assumes no coupling between periods beyond the assumed phase shift.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Eigenmode analysis of a half-mode uniplanar metamaterial-inspired substrate integrated waveguide." pith.science (2026). https://pith.science/paper/3KV4HG2W

@misc{pith2026260717403,
  author       = {Pith},
  title        = {Pith review of: Eigenmode analysis of a half-mode uniplanar metamaterial-inspired substrate integrated waveguide},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3KV4HG2W}},
  note         = {Machine review of arXiv:2607.17403}
}

read the original abstract

In this work, we systematically analyze the propagation characteristics of a new half-mode uniplanar substrate integrated waveguide (SIW) based on complementary split-ring resonators (CSRR), using a finite element method (FEM) eigenmode solver. The proposed half-mode CSRR SIW has a simpler fabrication than the SIW, since the via are substituted by CSRRs, and is more compact than the existing full uniplanar CSRR SIW, since its transverse size is reduced almost by half. To gain insight into the propagation characteristics of the proposed half-mode CSRR SIW and guide its synthesis process, we solve an eigenvalue problem that determines the complex propagation constant of the supported modes. By calculating the dispersion diagrams of the dominant mode, with all loss mechanisms included, we demonstrate that the half-mode uniplanar CSRR SIW retains the performance of the existing full CSRR SIW.

Figures

Figures reproduced from arXiv: 2607.17403 by the authors.

**Figure 2.** The attenuation constant of the half-mode single-CSRR SIW, varying [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

8 extracted references

[3]

Passia and T.V

M.-T. Passia and T.V . Yioultsis, ”Analysis and experimental char- acterization of a uniplanar metamaterial-inspired substrate integrated waveguide for easy-to-implement mmWave components”,IEEE Open J. Antennas Propag., vol. 6, no. 4, 2025

2025
[1]

Xu and K

F. Xu and K. Wu, ”Guided-wave and leakage characteristics of substrate integrated waveguide”,IEEE Trans. Microw. Theory Tech., vol. 53, no. 1, 2005

2005
[2]

Eccleston, ”Mode analysis of the corrugated substrate integrated waveguide”,IEEE Trans

K.W. Eccleston, ”Mode analysis of the corrugated substrate integrated waveguide”,IEEE Trans. Microw. Theory Tech., vol. 60, no. 10, 2012

2012
[4]

Q. Lai, C. Fumeaux, W. Hong and R. Vahldieck, ”Characterization of the propagation properties of the half-mode substrate integrated waveguide”, IEEE Trans. Microw. Theory Tech., vol. 57, no. 8, 2009

2009
[5]

Cheng, W

Y . Cheng, W. Hong and K. Wu, ”Half Mode Substrate Integrated Waveguide (HMSIW) Directional Filter”,IEEE Microw. Wirel. Compon. Lett., vol. 17, no. 7, 2007

2007
[6]

Patel, A

A.K. Patel, A. Bansal, C. Panagamuwa and W. Whittow, ”Half mode corrugated substrate integrated waveguide (HM-CSIW) band-stop filter using hexagonal ring resonators”,in Proc. Eur . Conf. Antennas Propag. (EuCAP), Glasgow, United Kingdom, 2024

2024
[7]

Fietz and Y

C. Fietz and Y . Urzhumov and G. Shvets, ”Complex k band diagrams of 3D metamaterial/photonic crystals”,Opt. Express, vol. 19, no. 20, 2011

2011
[8]

Nitas, V

M. Nitas, V . Salonikios, C.S. Antonopoulos and T.V . Yioultsis, ”Numer- ical calculation of dispersion diagrams and field distributions of waves in 3-D periodic split-ring resonator media”,IEEE Trans. Magn., vol. 55, no. 12, 2019

2019

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.

[1] [3]

Passia and T.V

M.-T. Passia and T.V . Yioultsis, ”Analysis and experimental char- acterization of a uniplanar metamaterial-inspired substrate integrated waveguide for easy-to-implement mmWave components”,IEEE Open J. Antennas Propag., vol. 6, no. 4, 2025

2025

[2] [1]

Xu and K

F. Xu and K. Wu, ”Guided-wave and leakage characteristics of substrate integrated waveguide”,IEEE Trans. Microw. Theory Tech., vol. 53, no. 1, 2005

2005

[3] [2]

Eccleston, ”Mode analysis of the corrugated substrate integrated waveguide”,IEEE Trans

K.W. Eccleston, ”Mode analysis of the corrugated substrate integrated waveguide”,IEEE Trans. Microw. Theory Tech., vol. 60, no. 10, 2012

2012

[4] [4]

Q. Lai, C. Fumeaux, W. Hong and R. Vahldieck, ”Characterization of the propagation properties of the half-mode substrate integrated waveguide”, IEEE Trans. Microw. Theory Tech., vol. 57, no. 8, 2009

2009

[5] [5]

Cheng, W

Y . Cheng, W. Hong and K. Wu, ”Half Mode Substrate Integrated Waveguide (HMSIW) Directional Filter”,IEEE Microw. Wirel. Compon. Lett., vol. 17, no. 7, 2007

2007

[6] [6]

Patel, A

A.K. Patel, A. Bansal, C. Panagamuwa and W. Whittow, ”Half mode corrugated substrate integrated waveguide (HM-CSIW) band-stop filter using hexagonal ring resonators”,in Proc. Eur . Conf. Antennas Propag. (EuCAP), Glasgow, United Kingdom, 2024

2024

[7] [7]

Fietz and Y

C. Fietz and Y . Urzhumov and G. Shvets, ”Complex k band diagrams of 3D metamaterial/photonic crystals”,Opt. Express, vol. 19, no. 20, 2011

2011

[8] [8]

Nitas, V

M. Nitas, V . Salonikios, C.S. Antonopoulos and T.V . Yioultsis, ”Numer- ical calculation of dispersion diagrams and field distributions of waves in 3-D periodic split-ring resonator media”,IEEE Trans. Magn., vol. 55, no. 12, 2019

2019