🔬 applied physics

Topology optimization of pentamode metamaterials for underwater acoustics

This study presents an automated topology optimization framework that utilizes low-frequency homogenization, the adjoint method, and the Virtual Temperature Method to design manufacturable pentamode metamaterials with fluid-like behavior for advanced underwater acoustic applications such as Luneburg lenses and invisibility cloaks.

Original authors: Sebastiano Cominelli, Matteo Pozzi, Francesco Braghin

Published 2026-03-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Sebastiano Cominelli, Matteo Pozzi, Francesco Braghin

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to build a special pair of glasses for sound. These glasses need to do two magical things:

Focus sound like a magnifying glass focuses light, so you can hear a whisper from miles away underwater.
Hide objects from sound, making a submarine or a rock completely invisible to sonar, as if it weren't there.

The problem is that nature doesn't give us materials that can do this. Water is great at carrying sound, but it's just a liquid; it can't hold a shape. Solids (like steel) hold their shape, but they bounce sound around in weird ways.

This paper introduces a new way to design "Smart Liquids" (called Pentamode Metamaterials). These are solid structures that act exactly like a custom-designed liquid, but they are strong enough to be built.

Here is how the authors did it, explained simply:

1. The Problem: The "Lego" Trap

Usually, when engineers want to build these smart materials, they start with a specific shape (like a specific Lego brick) and try to tweak it.

The Analogy: Imagine trying to design a perfect car by only being allowed to change the size of the wheels and the length of the doors. You might get a fast car, but you'll never invent a flying car because you are stuck with the "car" shape.
The Limitation: If the shape you started with isn't right, you can't get the sound properties you need. It's like trying to fit a square peg in a round hole.

2. The Solution: "Digital Sculpting" (Topology Optimization)

Instead of starting with a shape, the authors started with a blank block of digital clay. They used a computer algorithm to "sculpt" the material from the inside out.

The Analogy: Think of a 3D printer that doesn't just print a shape, but figures out the perfect internal structure to make the object float, bend, or focus sound. The computer asks: "If I remove this tiny bit of material here, and add a tiny bit there, will the sound behave the way I want?" It does this millions of times until it finds the perfect design.

3. The Challenge: The "Spaghetti" Problem

When the computer designs these structures to act like liquids, it often creates very thin, delicate connections.

The Analogy: Imagine the computer designs a bridge, but the supports are so thin they look like strands of cooked spaghetti. If you tried to build this in real life, the "spaghetti" would snap or melt during manufacturing. The design would be a beautiful picture but a useless object.
The Fix (Virtual Temperature): The authors added a clever trick called the Virtual Temperature Method.
- Imagine you are trying to keep a room warm. You turn on a heater on one side and a fan on the other. The heat has to flow through the room.
- If there is a gap in the wall, the heat escapes, and the room stays cold.
- The computer uses this "heat flow" idea to force the design to stay connected. If the structure has a gap, the "virtual heat" leaks out, and the computer knows to fix it. This ensures the final design is strong enough to be built, with no broken "spaghetti" strands.

4. The Results: Two Magical Devices

The authors tested their new "Digital Sculpting" method on two real-world challenges:

A. The Underwater Lens (The Sound Magnifier)

Goal: Take a flat wave of sound coming from far away and squeeze it all into one tiny point.
Result: They designed a lens made of thousands of tiny, unique micro-structures. When they simulated it, the sound waves bent perfectly and focused on a single spot, just like a lens focuses light.

B. The Acoustic Cloak (The Sonar Invisibility Cape)

Goal: Make a large object (like a submarine) look like a tiny pebble to sonar.
Result: They wrapped the object in a special "cape" made of their smart material. When sound hit the cape, it flowed around the object like water flowing around a rock in a stream, and reformed perfectly on the other side. To the sonar, the large object simply disappeared.

Why This Matters

This paper is a big deal because it removes the need for engineers to guess the right shape.

Before: "Let's try to tweak this specific honeycomb shape to see if it works." (Slow, limited, often fails).
Now: "Here is the sound behavior we need. The computer will invent the shape for us." (Fast, flexible, and creates designs no human would think of).

In a nutshell: The authors built a digital "magic wand" that automatically designs complex, buildable materials to control sound underwater, solving the tricky problem of keeping these delicate structures strong enough to actually exist.

1. Problem Statement

Acoustic metamaterials, particularly Pentamode Materials (PMs), offer the ability to manipulate sound waves in fluids (like water) by mimicking fluid-like behavior while retaining the structural integrity of solids. PMs are characterized by a bulk modulus ( $\kappa$ ) much higher than their shear modulus ( $G \approx 0$ ), allowing them to support pressure waves but resist shear waves.

However, designing PMs for practical underwater applications (e.g., acoustic lenses or cloaks) presents significant challenges:

Manufacturability: Achieving the extreme contrast between bulk and shear moduli requires microstructures with very thin connecting elements (hinges). Standard topology optimization often produces disconnected components or "one-node hinges" that are impossible to manufacture.
Design Limitations: Traditional parametric optimization relies on predefined base geometries. This limits the design space, risks geometric conflicts, and struggles to cover the full range of required anisotropic properties.
Connectivity: Ensuring structural connectivity in low-stiffness, high-density microstructures is difficult without compromising the target acoustic properties.

The authors aim to develop a fully automated, scalable topology optimization framework that designs PMs with precise control over effective acoustic properties (bulk modulus, density, and anisotropy) while guaranteeing structural connectivity and manufacturability.

2. Methodology

The proposed framework integrates low-frequency homogenization, the adjoint method, and a specific connectivity constraint strategy.

A. Theoretical Framework

Homogenization: The authors use low-frequency homogenization to compute the effective properties of periodic unit cells. The cell problem is solved using periodic boundary conditions to determine the homogenized elasticity tensor ( $C^h$ ) and effective density.
Pentamode Condition: For a 2D "bimode" material, the target elasticity tensor is constrained to have near-zero shear stiffness ( $C^h_{33} \approx 0$ ) while maintaining specific bulk modulus and density profiles.

B. Topology Optimization Setup

Density-Based Approach: The design variable is a density field $\mu$ (ranging from 0 to 1). The SIMP (Solid Isotropic Material with Penalization) scheme is used to penalize intermediate densities, pushing the solution toward a clear black-and-white (solid-void) design.
Regularization: A Helmholtz filter is applied to prevent mesh dependency and checkerboard patterns. A projection scheme (Heaviside projection) is used to sharpen the transition between solid and void.
Connectivity Constraint (Virtual Temperature Method - VTM): To address the issue of disconnected structures, the authors employ the Virtual Temperature Method.
- A thermal conduction problem is solved where the density field acts as thermal conductivity.
- Heat is generated on one side of the cell and sinks on the opposite side.
- The Thermal Compliance ( $C_T$ ) is minimized as an objective function. This forces the optimizer to create continuous thermal paths, which translates to continuous structural connections in the final design.

C. Optimization Problem

The core optimization problem minimizes the thermal compliance ( $C_T$ ) subject to constraints on the effective properties:

Target Bulk Modulus ( $\kappa_L$ ): $C^h_{11} \approx \kappa_L$
Target Density ( $\rho_L$ ): Effective volume $V$ matches the target.
Shear Suppression: $C^h_{33} \leq 0.01 \kappa_L$ (ensuring fluid-like behavior).
Symmetry: Symmetry constraints are imposed on the unit cell to enforce orthotropy or isotropy as required by the specific device.

The sensitivity analysis is performed using the adjoint method to efficiently compute gradients for the Method of Moving Asymptotes (MMA) optimizer.

3. Key Contributions

Automated Design Framework: The paper presents a fully automated workflow that eliminates the need for predefined base geometries, allowing for the discovery of complex, anisotropic microstructures that parametric methods might miss.
Integration of VTM for PMs: The authors successfully adapt the Virtual Temperature Method to pentamode materials. This ensures that the optimized microstructures possess well-defined, manufacturable connections (hinges) without sacrificing the target acoustic performance.
Generalization to Anisotropy: While previous work focused on isotropic PMs, this framework is generalized to design anisotropic graded-index materials, essential for complex devices like acoustic cloaks.
Validation of Full Devices: The study validates not just individual unit cells but also the performance of complete macroscopic devices (a Lüneburg lens and an acoustic cloak) through full-wave acoustic-elastic simulations.

4. Results

The framework was applied to two distinct underwater acoustic devices:

A. Lüneburg Acoustic Lens

Goal: Focus incoming plane waves to a point on the opposite side of the lens.
Design: A gradient-index (GRIN) lens requiring isotropic PMs with varying bulk modulus and density.
Outcome:
- 11 unique unit cells were optimized to cover the radial property profile.
- The optimized cells achieved the target properties with a relative error of < 3%.
- Dispersion Analysis: Confirmed the existence of a shear bandgap, ensuring shear waves are attenuated while pressure waves propagate.
- Simulation: Full-wave COMSOL simulations showed the microstructured lens successfully focused sound waves, matching the performance of an ideal continuous lens.

B. Acoustic Invisibility Cloak

Goal: Conceal a cylindrical obstacle from underwater sound waves by guiding waves around it.
Design: Requires anisotropic PMs with spatially varying properties derived from Transformation Acoustics (TA).
Outcome:
- 4 unique anisotropic unit cells were optimized and arranged in a polar grid.
- The optimized cells matched target anisotropic properties with < 3.5% error.
- Scattering Reduction: Simulations demonstrated a significant reduction in the Total Scattering Cross Section (TSCS). The microstructured cloak performed comparably to an ideal theoretical cloak, effectively hiding the obstacle from incident waves.
- Limitations: Performance degradation was observed at higher frequencies (where wavelength approaches the microstructure size), confirming the limits of the homogenization assumption.

5. Significance and Conclusion

This work represents a significant advancement in the field of acoustic metamaterials:

Scalability: The method is computationally efficient (optimizing 11 cells in ~11 minutes) and scalable to 3D geometries.
Reliability: By solving the connectivity issue via VTM, the designs are physically realizable, bridging the gap between theoretical optimization and manufacturing.
Versatility: The framework is not limited to lenses; it successfully handles complex anisotropic requirements for cloaking, proving its utility for a wide range of wave-manipulation devices.

The authors conclude that this topology optimization approach offers a powerful, flexible alternative to conventional parametric design, enabling the automated creation of complex, high-performance pentamode metamaterials for underwater acoustics. The code and simulation files are made publicly available to facilitate replication and further research.