Predictive beam-lattice reduction for higher-order topological modes in a 2D SSH phononic crystal

arXiv:2608.06395 · physics.comp-ph, cond-mat.mtrl-sci · Submitted 2026-07-31 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.

Kai: We're looking at the paper: "Predictive beam-lattice reduction for higher-order topological modes in a 2D SSH phononic crystal".

Paper summary: Kai: So we’re looking at this paper "Predictive beam-lattice reduction for higher-order topological modes in a 2D SSH phononic crystal" <ref:2608.06395#pg1,Predictive beam-lattice reduction for higher-order topological modes in a 2D>. Mira you want to give us the quick rundown on what they actually claim here?

Mira: Yeah, it’s about taking this complex material—a two dimensional topological phononic crystal made of square masses and elastic beams—and reducing it down to a much simpler model. The thesis is that this reduced model captures things that standard scalar mass spring SSH models miss, but it’s way easier to handle than those full finite element simulations.

Lev: What does the core mechanism of this reduction actually involve? Is it just ignoring some degrees of freedom?

Kai: No, they derive a Hermitian twelve degree of freedom dynamical matrix. This matrix keeps in plane translations, rotations, and even ligament eccentricity in the model. That’s how they build this mechanically faithful reduced model <ref:2608.06395#pg2>.

Mira: And the key is using Euler Bernoulli beam theory for those slender elastic ligaments, modeling them as thin beams. They use this reduction to predict things like topology, dispersion, boundary modes, and even experimental response all in one framework. That’s what matters because it links the physics to what you can actually build and measure <ref:2608.06395#pg2>.

Lev: So they are trying to keep the complexity down while still getting these higher-order topological modes, like those corner states or edge states that aren't in the basic SSH model. That sounds like a tough balance for real hardware implementation <ref:2608.06395#pg3>.

Kai: Exactly. They achieve this by exploiting dimerization, which means alternating the width of the beams in the x and y directions to build a mechanical two dimensional SSH lattice with a full band gap and a quantized bulk polarization <ref:2608.06395#pg2>.

Mira: And they show that you can control this transition from trivial to non-trivial phases using the sign of that dimerization parameter, which is g <ref:2608.06395#pg2>. But they add another layer of control with the ligament eccentricity factor b <ref:2608.06395#pg2>.

Lev: The eccentricity factor b sounds interesting because it’s a purely geometric mechanism for changing the topology, not just the dimerization <ref:2608.06395#pg2>. How does that translate to actual mode localization?

Kai: They use ribbon and finite cell calculations to predict edge and corner modes inside that band gap <ref:2608.06395#pg2>. Specifically, for topologically non trivial unit cells with g less than zero, they find four modes localized on the edges of the system in the ribbon calculation <ref:2608.06395#pg2>.

Mira: Then they do finite cell calculations and find two sets of four degenerate corner modes inside that band gap for those non trivial cells <ref:2608.06395#pg2>. They quantify how localized these modes are using the inverse participation ratio, or IPR <ref:2608.06395#pg2>.

Lev: If we were to take this to a real quantum hardware setup, what would the overhead look like? Running those finite cell calculations on actual hardware would be intensive <ref:2608.06395#pg2>.

Paper summary: Kai: The paper validates this whole approach by comparing their analytical calculations against full finite element method simulations and experimental measurements on three dee printed samples <ref:2608.06395#pg2>.

Mira: And the experimental results are pretty telling. They found that the squares located at the corners, which they marked in red, and along the edges, marked in green, show the highest amplitude for those in-gap modes <ref:2608.06395#pg2>.

Lev: So what’s the fidelity on that validation? How close are we to reality when you compare theory to physical samples?

Kai: The theoretical model agrees with those finite element simulations, showing a mean relative eigenfrequency deviation of zero point eight percent <ref:2608.06395#pg2>. That's a pretty tight margin for this kind of prediction.

Mira: What this really tells us is that the multiband character in this beam-lattice reduction reveals a topological structure that you just can't see with a simple scalar SSH description <ref:2608.06395#pg2>.

Lev: It seems the bulk polarization being trivial in the eight to nine gap still supports four ribbon edge states and eight corner modes, which is what makes it interesting for higher-order topology <ref:2608.06395#pg2>.

Kai: That structure is resolved at the level of mirror sectors, sigma tb and sigma lr, each of which undergoes an independent SSH type inversion and carries a non trivial sector resolved Zak phase <ref:2608.06395#pg2>.

Mira: And that eccentricity factor b actively shapes the topology by enhancing translation rotation coupling, which means it can even reverse which gaps are non trivial as g changes sign <ref:2608.06395#pg2>.

Lev: The paper mentions that they use this approach to reproduce analogues of quantum physical phenomena in macroscopic mechanically designed structures <ref:2608.06395#pg3>. That’s the big implication for engineers trying to visualize these effects physically <ref:2608.06395#pg3>.

Kai: So, we move from a simple scalar model to this reduced beam lattice model because it gives us a way to engineer higher-order topological states in something that can be fabricated <ref:2608.06395#pg1>.

Mira: The authors are Eloi Perez, Michele Brun, and Giorgio Carta <ref:2608.06395#pg1>. They’ve connected the physics of higher-order topological insulators to macroscopic mechanically designed structures like this one <ref:2608.06395#pg3>.

Lev: If you look at the limitations they state, they mention that this is a predictive model, and it doesn't necessarily account for all real-world material imperfections or damping effects that you'd find in a physical crystal <ref:2608.06395#pg2>.

Kai: It’s important to remember that while the model predicts the topology and modes, it’s not a complete description of every physical interaction in the final assembly <ref:2608.06395#pg2>.

Mira: Overall, this work shows how geometric control—the eccentricity factor b—can be used alongside structural dimerization to tune and visualize these topological phenomena <ref:2608.06395#pg2>. That’s a lot of tunable physics in one system <ref:2608.06395#pg1>.

Lev: It moves the conversation from purely theoretical quantum systems to structures where we can actually control and experimentally visualize these effects using mechanical design <ref:2608.06395#pg3>. That’s a shift in how we approach topological physics <ref:2608.06395#pg1>.

Kai: So, this paper gives us a mechanically faithful tool to study and potentially engineer these higher-order topological modes without needing the full computational cost of finite element simulations <ref:2608.06395#pg2>.

Conclusion: Kai: So this paper is about taking a complex phononic crystal and shrinking it down to a simpler model that still works for higher-order topology, which is what they call beam-lattice reduction.

Mira: I think the authors, Eloi Perez, Michele Brun, and Giorgio Carta, are showing how you can use mechanical design to control these states.

Lev: They’re using this method to predict things like edge and corner modes that you just can't see with a basic scalar model.

Kai: Right. The main point is they use this reduced model to see the full picture of topology, dispersion, and experimental response all at once.

Mira: It really shows that a simple description doesn't capture the whole story, and this reduction helps you find those hidden features like bulk polarization being trivial while still supporting edge states.

Lev: And they validate it by comparing their calculations to finite element simulations on three dee printed samples, which means it’s not just theory; there’s some real hardware validation here.

Kai: The experimental results showed that the squares at the corners and edges had the highest amplitude for those in-gap modes, which matches their theory quite well.

Mira: It highlights how geometric control—that eccentricity factor—can actively shape these topological properties by changing how different gaps behave based on whether you have a positive or negative dimerization.

Lev: That makes it interesting because it’s not just about the basic lattice structure; you can tune the topology by controlling that off-center positioning too.

Kai: It suggests that for engineers working with mechanical structures, this reduction is a way to get predictive power without needing those massive computational costs of full simulations.

Mira: It opens up a new way to think about how we design these materials for specific topological behaviors, moving beyond just simple scalar models and into more nuanced control.

Lev: This leads us to thinking about what you can actually build with this kind of precise mechanical tuning on a larger scale.

Eloi Pérez Compte, Michele Brun, Giorgio Carta, Nicola M. Pugno, Antonio S. Gliozzi, Federico Bosia

Politecnico di Torino · University of Cagliari · Mechano-X Labs

physics.comp-ph, cond-mat.mtrl-sci

Submitted: 2026-07-31

Updated: 2026-11-17

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

Importance score: 83/100

The gist: We develop a mechanically faithful reduced model for a two-dimensional topological phononic crystal composed of rigid square masses connected by slender elastic ligaments.

Key concepts

Topological Phase Control
This involves manipulating the beam widths in the crystal by alternating them (dimerization) and shifting them off-center using an eccentricity factor. This manipulation is used to create a mechanical system that mimics a 2D Su–Schrieffer–Heeger (SSH) model, allowing researchers to control whether topological gaps appear or disappear.
Localized Mode Prediction
The study predicts specific vibrational modes that are trapped at the edges and corners of the crystal. Using ribbon and finite-cell calculations, they found four edge states and eight corner modes inside a band gap for non-trivial unit cells, which are localized using metrics like the inverse participation ratio.
Multiband Character
The simplified model reveals a complex structure with multiple energy bands that is not visible in basic scalar SSH descriptions. The 8–9 frequency gap is key: while its overall bulk polarization is trivial, it supports distinct topological features like ribbon edge states and corner modes.
Eccentricity Factor (b)
The eccentricity factor represents the off-center positioning of the rectangular ligaments connecting the square masses. Varying this parameter actively shapes the system's topology by influencing how translation and rotation are coupled, which in turn determines which energy gaps behave as topological or non-topological.

Terminology

Summary

We develop a mechanically faithful reduced model for a two-dimensional topological phononic crystal composed of rigid square masses connected by slender elastic ligaments. p. 1

Where the paper says it

Abstract We develop a mechanically faithful reduced model for a two-dimensional topological phononic crystal composed of rigid square masses connected by slender elastic ligaments. Exploiting Euler–Bernoulli beam theory, we derive a Hermitian 12-degree-of-freedom dynamical matrix that retains in-plane translations, rotations, and ligament eccentricity.

This reduction captures effects that are absent from scalar mass–spring SSH models while remaining computationally much more tractable and more easily interpretable than full finite-element simulations. p. 1
Where the paper says it

Exploiting Euler–Bernoulli beam theory, we derive a Hermitian 12-degree-of-freedom dynamical matrix that retains in-plane translations, rotations, and ligament eccentricity. This reduction captures effects that are absent from scalar mass–spring SSH models while remaining computationally much more tractable and more easily interpretable than full finite-element simulations.

The model is derived to predict topology, dispersion, boundary modes, and experimental response within a single framework. p. 2
Where the paper says it

Reduced continuum-informed models can bridge this gap, preserving the efficiency and interpretability of lattice descriptions while retaining the structural parameters that control the band gap, the topological transition, and the resulting boundary-localized modes. Here, we develop such a model — mechanically faithful and analytically tractable — capable of predicting topology, dispersion, boundary modes, and experimental response within a single framework.

Model Derivation

The model considers a phononic crystal whose unit cell consists of four rigid squares of side length 2a with three degrees of freedom (DOF) each, totaling 12 DOFs. The square subunits are connected by eight rectangular ligaments with width si and length l that can be positioned off-centre using an eccentricity factor b. p. 3

Where the paper says it

We indicate the three DOFs of the four square subunits α = A, B, C, D within the (p, q) unit cell as u α p,q: x displacement, v α p,q: y displacement, (1) θ α p,q: counterclockwise rotation. The square subunits are connected by eight rectangular ligaments of width si and length l, that can be positioned off-centre using an eccentricity factor b (see Fig.

Assuming the ligaments are thin, they can be modelled as Euler-Bernoulli beams. p. 3
Where the paper says it

With this type of simple unit cell, whose geometrical characteristics are defined parametrically, it is possible to build a reduced order model that reproduces the dynamical behaviour of various types of real phononic crystals with complex geometries. Assuming that the ligaments are thin, namely that their cross-section is significantly smaller than their length, they can be modelled as Euler-Bernoulli beams.

The system of equations of motion is written in the form of an eigenvalue problem H ψ⟩ = omega2 ψ⟩. p. 5
Where the paper says it

of motion in the time-harmonic regime can be written as the following eigenvalue problem: H |ψ⟩ = Ω2 |ψ⟩, (3) where Ω = ω p ml/(EA) are the dimensionless eigenfrequencies and H is the 12 × 12 stiffness matrix. We exploit the high generality of the model to build a mechanical 2D Su–Schrieffer–Heeger (SSH) system, alternating the width of the beam in the x and y directions.

Topological Phase Control

The system is exploited to build a mechanical 2D Su–Schrieffer–Heeger (SSH) system by alternating the width of the beam in the x and y directions. The width of intracell and intercell beams is given by sint,ext = s(1 ± g/2). p. 5

Where the paper says it

1a, the width of intracell and intercell beams is sint for beams 1, 3, 5, 7 and sext for beams 2, 4, 6, 8. Their expression is sint,ext = s(1 ± g/2), (4) where the parameter g denotes the relative width difference with respect to the average width s, as represented in Fig 1b.

The parameter g denotes the relative width difference with respect to the average width s. p. 5
Where the paper says it

1a, the width of intracell and intercell beams is sint for beams 1, 3, 5, 7 and sext for beams 2, 4, 6, 8. Their expression is sint,ext = s(1 ± g/2), (4) where the parameter g denotes the relative width difference with respect to the average width s, as represented in Fig 1b.

The model allows the off-center position of the beams to be varied by b. p. 5
Where the paper says it

For g < 0, the intracell beams are thinner than the intercell beams, and viceversa for g > 0. The model also allows the off-center position of the beams to be varied, as quantified by the parameter b.

With dimerization (g ≠ 0), a full band gap opens around the nodal frequency between bands 8 and 9. p. 7
Where the paper says it

3) contains the contribution M5 = F5b = (EA2/l)b 2 θ B p,q, showing explicitly that the rotational stiffness grows with b 2 . When dimerization (g ̸= 0) is introduced, a full band gap opens around the nodal frequency, between bands 8 and 9.

Localized Mode Prediction

The study predicts the emergence of edge and corner modes using ribbon and finite-cell calculations. The ribbon built using topologically non-trivial UCs (g < 0) exhibits four modes inside the band gap, which are localized on the edges of the system. p. 1

Where the paper says it

This band gap can be labelled as topologically trivial or non-trivial if the associated topological invariant is zero or non-zero, respectively [2]. Such materials obey bulk-edge correspondence, whereby having a bulk band structure with a non-trivial band gap enables localized edge states [3].

Finite system calculations show two sets of four degenerate corner modes inside the band gap for non-trivial cells. p. 1
Where the paper says it

Ribbon and finite-cell calculations predict in-gap edge and corner modes, quantified by localization measures and confirmed by finite-element simulations. Measurements on 3D-printed samples show an evanescent response in the trivial structure and enhanced boundary/corner response in the non-trivial structure within the predicted gap.

The localization of these non-trivial modes can be quantified using the inverse participation ratio (IPR). p. 1
Where the paper says it

The sign of the dimerization controls the transition from trivial to non-trivial phases, while ligament eccentricity provides an additional purely geometric mechanism for changing the topology. Ribbon and finite-cell calculations predict in-gap edge and corner modes, quantified by localization measures and confirmed by finite-element simulations.

Experimental Validation

The theoretical model is validated using finite element method (FEM) simulations and experimental measurements on 3D-printed samples. The analytical calculations reduce the computation time by various orders of magnitude compared to the FEM simulations. p. 6

Where the paper says it

5 Ω Γ X M Γ Γ X M Γ Γ X M Γ Γ X M Γ k (b) −1 0 1 b bands 7–8 bands 8–9 bands 9–10 bands 11–12 −1 0 1 g −1 0 1 b −1 0 1 g −1 0 1 g −1 0 1 g −1 0 1 ∆Ω N (0, 0) ( 1 2 , 1 2 ) P FIG. 2: (a) Dispersion diagrams obtained from the analytical calculations (solid black lines) and the FEM model (red dots), for the monoatomic (g = 0) and dimerized (g ̸= 0) cases, and for different off-center positions b.

Experimental results show that the squares located at the corners (red) and along the edges (green) of the sample have the highest amplitude in-gap modes. p. 3
Where the paper says it

We validate the analytical results using finite element simulations and experimental measurements on 3D-printed samples. MODEL We consider a phononic crystal whose unit cell (UC) consists of four rigid squares of side length 2a, mass m and moment of inertia I, each having three degrees of freedom (DOF), two translational (in the x and y directions) and one rotational, totalling 12 DOFs.

The theoretical model agrees with FEM simulations, with a mean relative eigenfrequency deviation of 0. p. 6
Where the paper says it

5 Ω Γ X M Γ Γ X M Γ Γ X M Γ Γ X M Γ k (b) −1 0 1 b bands 7–8 bands 8–9 bands 9–10 bands 11–12 −1 0 1 g −1 0 1 b −1 0 1 g −1 0 1 g −1 0 1 g −1 0 1 ∆Ω N (0, 0) ( 1 2 , 1 2 ) P FIG. 2: (a) Dispersion diagrams obtained from the analytical calculations (solid black lines) and the FEM model (red dots), for the monoatomic (g = 0) and dimerized (g ̸= 0) cases, and for different off-center positions b.

8%.

Key Findings

The multiband character of the beam-lattice reduction reveals a topological structure invisible to a scalar SSH description. The 8–9 gap is the clearest example where its cumulative bulk polarization is trivial, yet it supports four ribbon edge states and eight corner modes. p. 1

Where the paper says it

The sign of the dimerization controls the transition from trivial to non-trivial phases, while ligament eccentricity provides an additional purely geometric mechanism for changing the topology. Ribbon and finite-cell calculations predict in-gap edge and corner modes, quantified by localization measures and confirmed by finite-element simulations.

This topological structure is resolved at the level of the mirror sectors σtb and σlr, each of which undergoes an independent SSH-type inversion and carries a non-trivial sector-resolved Zak phase. p. 7
Where the paper says it

(b) Longitudinal mode character at X, showing the inversion of the acoustic- and optical-like modes in both mirror sectors. (c) Sector-resolved Zak phase maps for the two relevant mirror sectors, masked by the complete 8–9 bulk band gap.

The eccentricity b actively shapes the topology by enhancing translation–rotation coupling, reversing which gaps are non-trivial as g changes sign. p. 9
Where the paper says it

Gaps between bands 9–10 and 11–12 are non-trivial for g < 0, whereas the gap between bands 7–8 is non-trivial for g > 0. This inversion reflects the multiband nature of the beam lattice, where translation–rotation coupling hybridizes the bands and can reverse the parity inversion associated with a given gap.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Enhanced phase space exploration by utilizing a compact Hermitian dynamical matrix that retains in-plane translations, rotations, and ligament eccentricity. This allows for a computationally much more tractable model compared to full finite-element simulations.

  2. Prediction of higher order topological modes through sector-resolved Zak phases. The paper shows that the 8–9 gap is the clearest example: its cumulative bulk polarization is trivial, yet it supports four ribbon edge states and eight corner modes, which can be resolved by analyzing mirror sectors Stb and Slr.

  3. Design of manufacturable topological elastic metamaterials using continuum-informed discrete models. The model provides a validated route for designing topological elastic metamaterials using a continuum-informed discrete model rather than either idealized mass–spring networks or brute-force numerical optimization.

  4. Characterization of band gaps and transitions via geometric parameters. The sign of dimerization controls the transition from trivial to non-trivial phases, and eccentricity b provides an additional purely geometric mechanism for changing the topology.

  5. Quantification of localization using inverse participation ratio and corner weight. The system can quantify the overall localization of each eigenmode using the inverse participation ratio (IPR) and corner weight (Wcorner), identifying modes with high IPR and high corner weight as corner-localized states.

Abstract

We develop a mechanically faithful reduced model for a two-dimensional topological phononic crystal composed of rigid square masses connected by slender elastic ligaments. Exploiting Euler Bernoulli beam theory, we derive a Hermitian 12 degree of freedom dynamical matrix that retains in-plane translations, rotations, and ligament eccentricity. This reduction captures effects that are absent from scalar mass spring SSH models while remaining computationally much more tractable and more easily interpretable than full finite element simulations. Dimerizing the ligament widths produces a mechanical 2D SSH lattice with a full band gap and a quantized bulk polarization. The sign of the dimerization controls the transition from trivial to non trivial phases, while ligament eccentricity provides an additional purely geometric mechanism for changing the topology. Ribbon and finite cell calculations predict in gap edge and corner modes, quantified by localization measures and confirmed by finite element simulations. Measurements on 3D printed samples show an evanescent response in the trivial structure and enhanced boundary/corner response in the non trivial structure within the predicted gap. The results provide a validated route for designing topological elastic metamaterials using a continuum informed discrete model rather than either idealized mass spring networks or brute force numerical optimization.

Sources

Related papers