Surface Roughness and Filler Restructuring in Magneto-Active Elastomers: Magnetically Hard versus Magnetically Soft Particles

arXiv:2608.14433 · cond-mat.soft · Submitted 2026-08-14 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: We're looking at the paper: "Surface Roughness and Filler Restructuring in Magneto-Active Elastomers".

Mira: Here's the gist of the paper. Magneto-active elastomers (MAEs) are promising for soft robotics because their shape and mechanical properties can be controlled by an applied magnetic field <ref:2608.14433#pg2>.

Kai: The MS model reproduces the magnetization curves and surface roughness of CIP-based MAEs for magnetic fields close to saturation, while maintaining the observed qualitative features at lower field strengths <ref:2608.14433#pg3>.

Mira: Particle Restructuring Mechanisms Both magnetically hard (MH) and magnetically soft (MS) elastomers converge to the same high-field state—a labyrinth of bundled, field-aligned chains—but do so through distinct pathways <ref:2608.14433#pg6>.

Kai: The MH MAEs break their zero-field chains, which lie parallel to the MAE layer plane (in-plane), and rotate them into alignment with an external magnetic field, whereas MS MAEs gradually build up field-aligned chains from neighboring particles <ref:2608.14433#pg6>.

Paper summary: Kai: So to wrap up this paper, "Surface Roughness and Filler Restructuring in Magneto-Active Elastomers: Magnetically Hard versus Magnetically Soft Particles," it confirms that the type of magnetic filler you use has a direct impact on the final surface roughness of your material Mira.

Mira: It shows that magnetically hard particles cause a rougher surface because they break those initial chains in-plane, while soft particles have different clustering dynamics Lev.

Lev: The authors provide these molecular dynamics insights into how the elastic energy, Zeeman energy, and dipolar energy work together to drive those distinct restructuring paths Kai.

Kai: It’s a useful comparison because it shows that even when two systems converge to the same high-field state, they reach it through fundamentally different physical processes Mira right?

Mira: Exactly. The paper helps us understand that for applications in soft robotics, the choice of filler isn't just about magnetic strength; it’s about controlling the surface topography Lev.

Lev: They highlight that their simple model has limitations, specifically that it overmagnetizes particles below three times ten nine because it ignores things like intra-particle domain interactions Kai.

Kai: So the main implication is that for real-world design, you need to use a more detailed model when the field isn't already pushing those particles into their saturation state Mira.

Mira: It means understanding the interplay between magnetic interactions and mechanical structure is crucial for predicting how these materials will behave physically Lev.

Conclusion: Kai: So, we're wrapping up this look at "Surface Roughness and Filler Restructuring in Magneto-Active Elastomers: Magnetically Hard versus Magnetically Soft Particles." Basically, the paper shows how whether you use magnetically hard or soft particles really dictates how rough a material's surface ends up being.

Mira: Right, it’s about the difference between those two particle types. The key finding is that the way they restructure under a magnetic field leads to different surface textures for these magneto-active elastomers.

Lev: From my side, it’s interesting because it confirms what we saw in simulations—that the hard particles have this barrier related to breaking those initial parallel chains in the material's plane.

Kai: And that means we can predict a rougher surface when using the hard ones because of that energetic barrier.

Mira: But you have to remember, the soft matrix also gets rougher because it lets those particles move around more freely, which is inversely proportional to how much filler you use in the material.

Lev: I just want to flag something though; their simple model overmagnetizes particles below a certain field strength because it skips some important interactions inside the particle itself.

Kai: That’s a fair caveat, Lev. It shows us that for real hardware, we need to account for those internal forces when designing things like this.

Mira: Exactly. The whole point of this study is that the choice of filler isn't just about getting strong magnetism; it directly controls the surface topography you’re working with in soft robotics.

Lev: So, understanding these distinct pathways—the hard breaking versus the soft clustering—is crucial for designing materials where you need precise surface control.

Kai: It really makes you think about how we engineer these things to achieve a specific feel or function, not just a magnetic response.

University of Vienna, Faculty of Physics · Zoological Institute, University of Kiel · East Bavarian Center for Intelligent Materials (EBACIM), Ostbayerische Technische Hochschule Regensburg

cond-mat.soft

Submitted: 2026-08-14

Updated: 2026-11-13

License: http://creativecommons.org/licenses/by-sa/4.0/

Importance score: 80/100

The gist: Magneto-active elastomers (MAEs) are promising for soft robotics because their shape and mechanical properties can be controlled by an applied magnetic field <ref:2608.14433#pg2>.

Key concepts

Magnetically Hard (MH) Elastomers
These elastomers undergo a specific restructuring process where their initial zero-field chains are broken. They then rotate into alignment with the external magnetic field at high fields. This process is linked to an energetic barrier related to breaking these in-plane chains, which causes an abrupt increase in surface roughness.
Magnetically Soft (MS) Elastomers
MS elastomers gradually build up field-aligned chains by interacting with neighboring particles. Unlike MH types, the dipolar energy for MS systems becomes negative at higher fields, indicating net attractive interactions between the particles. This gradual buildup leads to different surface roughness characteristics.
Dipolar Energy
This energy component measures the magnetic interaction between particles and is crucial for distinguishing MH and MS systems. For MHPs, it shows a sharp increase above a threshold field due to chain rupture. For MSPs, it turns negative at lower fields, signaling attractive forces that drive the particle restructuring process.
Surface Roughness (Rrms/σ)
The surface roughness parameter is analyzed across different magnetic field strengths ($\xi$). The abrupt increase in roughness for MH elastomers is explained by the energy required to break their initial chains. Furthermore, a softer elastic matrix allows particles to move further, resulting in rougher surfaces that are inversely proportional to the particle volume fraction.

Terminology

Summary

Magneto-active elastomers (MAEs) are promising for soft robotics because their shape and mechanical properties can be controlled by an applied magnetic field <ref:2608. p. 0

Where the paper says it

Gorb,2 and Sofia Kantorovich1 1University of Vienna, Faculty of Physics, Kollingasse 14-16, 1090 Vienna, Austria 2Functional Morphology and Biomechanics, Zoological Institute, University of Kiel, Am Botanischen Garten 1-9, Kiel, 24118, Germany 3East Bavarian Center for Intelligent Materials (EBACIM), Ostbayerische Technische Hochschule (OTH) Regensburg, Seybothstr. 2, Regensburg, 93053, Germany (Dated: August 14, 2026) Magneto-active elastomers (MAEs) – composites of magnetic nano-/micro-particles embedded in a soft polymer matrix – are promising for soft robotics, as their shape and mechanical properties can be controlled by an applied magnetic field.

14433#pg2>. The MS model reproduces the magnetization curves and surface roughness of CIP-based MAEs for magnetic fields close to saturation, while maintaining the observed qualitative features at lower field strengths <ref:2608. p. 0
Where the paper says it

We find that both MH and MS elastomers converge to the same high-field state – a labyrinth of bundled, field-aligned chains – but do so through distinct pathways: MH MAEs break their zero-field chains, which lie parallel to the MAE layer plane (in-plane), and rotate them into alignment with an external magnetic field, whereas MS MAEs gradually build up field-aligned chains from neighboring particles. We show that the MS model reproduces the magnetization curves and surface roughness of CIP-based MAEs for magnetic fields close to saturation, while maintaining the observed qualitative features at lower field strengths.

14433#pg3>.

Particle Restructuring Mechanisms

Both magnetically hard (MH) and magnetically soft (MS) elastomers converge to the same high-field state—a labyrinth of bundled, field-aligned chains—but do so through distinct pathways <ref:2608.14433#pg6>. The MH MAEs break their zero-field chains, which lie parallel to the MAE layer plane (in-plane), and rotate them into alignment with an external magnetic field, whereas MS MAEs gradually build up field-aligned chains from neighboring particles <ref:2608. p. 0

Where the paper says it

We employ molecular dynamics to study the differences between thin MAE layers with MS and magnetically hard (MH) filler particles having the same saturation magnetization. We find that both MH and MS elastomers converge to the same high-field state – a labyrinth of bundled, field-aligned chains – but do so through distinct pathways: MH MAEs break their zero-field chains, which lie parallel to the MAE layer plane (in-plane), and rotate them into alignment with an external magnetic field, whereas MS MAEs gradually build up field-aligned chains from neighboring particles.

14433#pg6>.

Energy Analysis of Interactions

The system's evolution is understood by analyzing three per-particle average energies: the elastic energy, the µ-normalized Zeeman energy, and the dipolar energy <ref:2608.14433#pg6>. The dipolar energy highlights a pronounced difference between MHP and MSP, as it is close to −1. p. 5

Where the paper says it

The dipolar energy (Fig. 2 c)) highlights the most pronounced difference between MHP and MSP.

2 × 108 for MHP at ξ = 0 and increases rapidly above a threshold field strength, consistent with the rupture of initially favorable horizontal in-plane chains <ref:2608. p. 5
Where the paper says it

2 × 108 and increases rapidly over a finite range above ξ ≈ 0. 12 × 109 , consistent with the rupture of initially favorable horizontal in-plane chains.

14433#pg7>. For MSP, the dipolar energy becomes positive for 0 < ξ ≤ 0. p. 5
Where the paper says it

In other words, we have long energetically highly advantageous but still imperfect chains. For MSP, the dipolar energy is zero at ξ = 0 (particles are not magnetized), becomes positive (net repulsive interactions) for 0 < ξ ⪅ 0.

3 × 109 depending on particle concentration and turns increasingly negative for ξ ≤ 0. p. 5
Where the paper says it

For MSP, the dipolar energy is zero at ξ = 0 (particles are not magnetized), becomes positive (net repulsive interactions) for 0 < ξ ⪅ 0. 3 × 109 depending on particle concentration, and turns increasingly negative (net attractive interactions) for ξ ⪆ 0.

4 × 109, indicating net attractive interactions <ref:2608. p. 5
Where the paper says it

For MSP, the dipolar energy is zero at ξ = 0 (particles are not magnetized), becomes positive (net repulsive interactions) for 0 < ξ ⪅ 0. 3 × 109 depending on particle concentration, and turns increasingly negative (net attractive interactions) for ξ ⪆ 0.

14433#pg7>.

Magnetization and Saturation Behavior

The saturation magnetization, M∞, is determined solely by the particle volume fraction, saturation magnetization of the material and particle size <ref:2608.14433#pg8>. The normalized z-component of the average magnetic moment for MSP is equal to 1 for all non-zero fields, indicating perfect alignment with the external field at all times <ref:2608. p. 6

Where the paper says it

Finally, the difference in magnetization mechanisms is most clearly seen in Fig. The normalized zcomponent of the average magnetic moment of the MSP is equal to 1 for all non-zero fields, indicating perfect alignment with the external field at all times.

14433#pg8>. In contrast, for MHPs, this quantity increases rapidly over a finite field range, reflecting the progressive reorientation of the particles <ref:2608. p. 6
Where the paper says it

The normalized zcomponent of the average magnetic moment of the MSP is equal to 1 for all non-zero fields, indicating perfect alignment with the external field at all times. In contrast, for the MHPs, this quantity increases rapidly over a finite field range, reflecting the progressive reorientation of the particles.

14433#pg8>.

Surface Topography and Roughness

The surface roughness parameter Rrms/σ can be separated into three regimes based on the non-dimensional field ξ <ref:2608.14433#pg6>. The abrupt increase in Rrms for the MH elastomers is consistent with the energetic barrier related to breaking the in-plane chains <ref:2608. p. 6

Where the paper says it

18 × 109 , the MH elastomer roughness reaches a plateau, and the MS elastomer slowly approaches comparable Rrms values. The abrupt increase in Rrms for the MH elastomers is consistent with the aforementioned energetic barrier related to breaking the in-plane chains.

14433#pg6>. Furthermore, a softer elastic matrix permits larger particle displacements and thus leads to rougher surfaces, showing that roughness is inversely proportional to the particle volume fraction <ref:2608. p. 6
Where the paper says it

As expected and consistently across all data sets, a softer elastic matrix (blue) permits larger particle displacements and thus leads to rougher surfaces. The volume fraction exerts a weak yet systematic influence: MAEs with lower particle loadings exhibit rougher surfaces.

14433#pg9>.

Internal Microstructure Characterization

The coordination number distribution shows that at high fields (orange and yellow columns), the difference in the number of neighbors between MAE with MSPs and MHPs is not significant <ref:2608.14433#pg10>. The bond-order parameters reveal that neighboring chains adopt a locally hexagonal arrangement in the plane perpendicular to the chain direction, reflecting an anisotropic nature where orientational order along the chain direction coexists with hexagonal packing of neighboring chains in the transverse plane <ref:2608. p. 8

Where the paper says it

As a consequence, the bundles develop a local hexagonal coordination in the transverse plane while remaining elongated along the field direction. The coexistence of large ⟨q¯2⟩ with hcplike higher-order BOPs therefore reflects the anisotropic nature of the bundles: orientational order along the chain direction coexists with hexagonal packing of neighboring chains in the perpendicular plane.

14433#pg10>.

Comparison to Experimental Data

The MS model reproduces both magnetization curves and surface roughness of carbonyl iron MAEs for fields close to saturation—B ≤ 250 mT—and preserves qualitative trends down to lower fields <ref:2608.14433#pg14>. The simulation overmagnetizes particles below ξ ≈ 1. p. 2

Where the paper says it

(6) Here, rij , is the vector connecting the centers of particles i and j, rij ≡ ∥rij∥, µi is the dipole moment of i-th particle, and B is the uniform external magnetic flux density. Computational Details To prepare the initial elastomer sample, we first place a substrate of volume-excluding particles at the bottom of the simulation box.

5 × 109, which is attributed to neglecting intra-particle domain interactions and demagnetizing fields in the simple magnetizable dipole model <ref:2608. p. 9
Where the paper says it

For nondimensional fields lower than ξ ≈ 1. 5 × 109 , we observe a systematic deviation: the experimental moments have a slower rise, as the multi-domain carbonyl iron particles have a low effective initial susceptibility due to the domain walls and demagnetizing fields inside the particles; we do not consider such interactions in our simple magnetizable dipole model and this leads to an overmagnetization of the simulated particles at lower fields.

14433#pg13>. The choice of filler particles is therefore a question of the desired actuation profile, such as a threshold-like response or a smooth, proportional control <ref:2608.14433#pg14>.

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. Improved understanding of magnetic particle behavior in soft materials through two distinct pathways: "MH MAEs break their zero-field chains, which lie parallel to the MAE layer plane (in-plane), and rotate them into alignment with an external magnetic field, whereas MS MAEs gradually build up field-aligned chains from neighboring particles." This allows for predicting the resulting microstructure based on particle type.

  2. Ability to predict surface topography changes under varying magnetic fields by linking structural evolution to roughness: The abrupt increase in Rrms for the MH elastomers is consistent with the aforementioned energetic barrier related to breaking the in-plane chains. This enables prediction of localized surface features in soft robotics applications.

  3. Quantitative prediction of magnetization response across different particle types and matrix stiffnesses using non-dimensional parameters: The system can predict that The saturation magnetization, M infinity, is determined solely by the particle volume fraction, saturation magnetization of the material and particle size, providing a direct scaling law for material response.

  4. Accurate modeling of low-field behavior in complex magnetic systems: The improved system can account for experimental limitations by recognizing that the simulation χ0 is considerably smaller (around 50×) than the predicted value for carbonyl iron particles and that the model systematically over-magnetizes particles, because the Langevin response of each particle neglects the intra-particle domain interactions and demagnetizing fields.

  5. Characterization of internal microstructure through bond order parameters: The system can differentiate between ordered and disordered packings by calculating the average of the averaged bond-order parameters (BOPs) and observing that for high fields, neighboring chains adopt a locally hexagonal arrangement in the plane perpendicular to the chain direction.

Abstract

In magneto-active elastomers (MAEs), a magnetic field reshapes the surface and with it adhesion, friction and wetting. Most use magnetically soft carbonyl iron, whose restructuring has lacked a particle-resolved description. We put forward a molecular dynamics model with magnetizable particles and compare thin layers of hard and soft fillers. Both converge to the same high-field state: a labyrinth of bundled field-aligned chains roughening the surface by about one particle diameter. Hard fillers break and rotate in-plane chains above a threshold field, roughening the surface abruptly. Soft fillers build chains particle by particle, roughening it continuously. The model reproduces the magnetization of carbonyl iron MAEs above about 250 mT and the magnitude of their roughness. Measured roughness varies linearly with inverse zero-field Young's modulus: matrix stiffness, not particle number, controls it. At equal saturation magnetization, filler type sets not how rough the surface becomes, but whether it roughens abruptly or continuously.

Sources

Related papers