Dynamic mechanical properties of MREs under the coupled effect of magnetic field, normal force, and CIP content Authors: Zhihong Lin, Zhe Gao, Yuedong Huang Published: 2026 ABSTRACT SummaryMagnetorheological elastomers (MREs) are smart materials whose mechanical properties can be regulated by an external magnetic field. In this study, dynamic shear tests were conducted using a rheometer to investigate the effects of carbonyl iron powder (CIP) content, normal force, and magnetic-field strength on the dynamic mechanical properties of MREs. The results show that under the same test conditions, both the storage and loss moduli increase markedly as the CIP content rises from 40% to 70%. This is because more magnetic particles strengthen the internal load-bearing structure and increase the frictional energy dissipation. When the normal force increases from 2 to 15 N, the moduli also increase. This indicates that moderate compression of the particle structure can reduce performance fluctuation. As the magnetic field increases from 0 to 0.8 T, the moduli increase nonlinearly, although the incremental enhancement gradually diminishes above 0.5 T. Overall, higher CIP content enhances magnetic-field regulation, and a suitable combination of normal force and magnetic field improves the stability of the material response.Keywords: magnetorheological elastomer, rheometer, normal force, storage modulus, loss modulus Introduction Magnetorheological elastomer (MRE) is a magneto-sensitive smart material. An external magnetic field can rapidly and reversibly change its modulus1,2,3 and damping properties.4,5,6 As a result, MREs have potential applications in aerospace,7,8,9 vibration reduction,10,11 and biomedical engineering.12,13 They respond quickly, work reversibly, and are cost controllable, making them attractive for active control of material properties. The matrix of an MRE is the load-bearing skeleton. It determines the initial modulus, the field-induced tunable range, and the magnetorheological (MR) effect. Differences among matrices can be compared using rheological testing. Common matrices include silicone rubber,14,15 natural rubber,2,9,16 polyurethane,17 and thermoplastic elastomers.18 Among them, silicone rubber is widely used for its low viscosity, ease of processing, and rapid curing. In addition to the matrix, the internal microstructure of an MRE also affects its macroscopic performance. When a magnetic field is applied during curing, magnetic particles tend to align along the field direction and form an anisotropic structure.19 Without a magnetic field, the particles are randomly distributed, and the material is isotropic.17 Jung et al.20 studied the dynamic shear properties of natural-rubber-based MREs. Their results showed that the MR effect of anisotropic MREs was higher than that of isotropic MREs. The shear storage modulus increased with frequency and magnetic field but decreased with increasing strain. Palacios-Pineda et al.21 prepared MREs containing nano-iron particles. The nanoparticles increased the matrix’s storage modulus and damping by 4% and 63%, respectively. These studies show that the microstructure strongly affects the field-induced response of MREs. After the matrix is selected, additives can further improve MRE performance. Lokander22 found that plasticizers can reduce the zero-field modulus, thereby increasing the relative MR effect. Khairi et al.23 reported that a silicone oil plasticizer reduces matrix viscosity, improves particle alignment, and enhances the MR effect. Carbon black24,25 can improve the basic mechanical properties. Three-dimensional carbon fillers26 can also impart electrical conductivity and strain-sensing capability to MREs. Magnetically active additives27,28 can further enhance magnetic response. Nanoparticles easily agglomerate and often require surface modification. Marcin et al.29 improved the dispersion of Fe3O4 particles in rubber by adding an ionic liquid. Salem et al.30 further studied the effects of different carbonyl iron powders (CIPs) on MRE microstructure and field-induced properties. Therefore, controlling filler content and dispersion is essential for improving MRE performance. In MRE design, both particle-chain orientation and particle morphology influence the field-induced response. Rheological testing helps link the microstructure to macroscopic performance. Tong et al.31 found that flower-like magnetic particles can expand the tunable range of the storage modulus. Afiq et al.32 demonstrated, through rheological testing, that anisotropic structures enhance the modulus and MR effects. Li et al.33 identified an optimal particle volume fraction and built a transverse shear model. Tran et al.34 studied the effect of magnetic saturation on shear stress relaxation and predicted long-term behavior using a fractional-order model. Nishant et al.35 noted that rheological data are important for viscoelastic modeling and vibration-control design of MREs. Previous studies have discussed the rheological behavior and regulation mechanisms of MREs. However, most of them have focused on a single factor. The combined effect of the magnetic field, normal force, and CIP content on the dynamic rheological properties of MREs still needs to be systematically studied. In particular, the evolution of the storage and loss moduli requires further analysis. This limits the understanding and prediction of MRE behavior under coupled working conditions. Therefore, this study uses rheological experiments to investigate MREs with different CIP contents (40%, 50%, 60%, and 70%) under multiple magnetic fields (0, 0.1, 0.2, 0.5, and 0.8 T) and normal forces (2, 5, 10, and 15 N). The storage modulus and loss modulus are analyzed as functions of magnetic field, normal force, and CIP content. The non-uniform regulation of modulus growth by normal force is also discussed. The results provide reference data for performance prediction and structural optimization of MREs under coupled mechanical and magnetic conditions. Results Magnetic field and normal force regulate the dynamic response of 40 wt % CIP MREs An MRE is a smart soft material whose dynamic mechanical properties can be tuned by a magnetic field and mechanical load. CIP is a commonly used magnetic filler. Its content affects particle spacing, particle interactions, and microstructure, thereby influencing the tunable modulus range. In this study, the storage modulus G′, loss modulus G′′, and normal stress were measured simultaneously using an Anton Paar MCR 302 rotational rheometer. The magnetic field strengths were 0, 0.1, 0.2, 0.5, and 0.8 T. The normal forces were 2, 5, 10, and 15 N. Figures 1 and 2 analyze the effects of the magnetic field and normal force on G′ and G′′ of the 40% CIP MRE. Storage modulus of 40% CIP MRE under different magnetic fields (A–E) Storage moduli of 40% CIP MRE under (A) 0 T, (B) 0.1 T, (C) 0.2 T, (D) 0.5 T, and (E) 0.8 T. Data are presented as the mean ± SD (n = 3). Loss modulus of 40% CIP MRE under different magnetic fields (A–E) Loss moduli of 40% CIP MRE under (A) 0 T, (B) 0.1 T, (C) 0.2 T, (D) 0.5 T, and (E) 0.8 T. Data are presented as the mean ± SD (n = 3). As shown in Figure 1, in the absence of a magnetic field, the storage modulus of the 40% CIP MRE increased with increasing normal force. When the normal force was 2 N, the peak storage modulus was about 20,000 Pa. When the normal force increased to 15 N, the peak value increased to about 38,000 Pa. Under all normal forces, the modulus-time curves first increased rapidly and then decayed slowly. After a magnetic field was applied, the curves shifted upward, and the peak storage modulus increased with magnetic field strength. Under 0.8 and 15 N, the peak storage modulus reached about 100,000 Pa. When the magnetic field exceeded 0.5 T, the growth rate of the peak storage modulus slowed. As shown in Figure 2, the loss modulus of the 40% CIP MRE also increased with the increasing normal force. Under a magnetic field of 0.1 T, the peak loss modulus was approximately 9,500 Pa at 2 N and increased to approximately 11,800 Pa at 15 N. Overall, the loss modulus first increased, then remained relatively stable, and finally decreased slowly. With increasing magnetic field strength, the curves for the same normal force shifted upward overall, and the peak value continued to increase. For example, at 0.5 and 15 N, the peak loss modulus approached 21,000 Pa. Under the coupled effect of low normal force and high magnetic field, particle friction and sliding were more pronounced during the initial stage, resulting in a higher initial loss modulus. Subsequently, the relaxation of the particle-chain structure reduced energy dissipation, thereby decreasing the loss modulus. It should be noted that the peak values discussed in the text are mainly used to describe the maximum instantaneous response of the material under specific conditions, while the initial values and the subsequent decreasing trend are used to explain only the evolution characteristics of the curves and are not used as the main basis for quantitative comparison among different conditions. Figures 1and 2 show that both the magnetic field and the normal force increased the storage modulus of MREs. The loss modulus also increased with magnetic field and normal force. However, under low normal force and strong magnetic field, the particle structure was relatively loose, and the initial friction was strong. Therefore, the loss modulus may first increase and then decrease. This indicates that the magnetic field and normal force not only regulate stiffness and energy dissipation separately but also jointly affect the dynamic mechanical behavior of MREs. CIP content and normal force jointly enhance stiffness and damping The 40% CIP results show that the magnetic field and normal force jointly affect the storage modulus of MREs. CIP content also changes particle spacing, the integrity of the oriented structure, and force-chain transmission efficiency. The coupling mechanism among CIP content, normal force, and magnetic field still requires further analysis. Therefore, the storage modulus and loss modulus of samples with different CIP contents were compared. Figures 3, 4, and 5 show the storage modulus results. Figures 6, 7, and 8 show the loss modulus results. Storage modulus of MREs with different CIP contents at 0 T (A–D) Storage moduli of MREs with (A) 40% CIP, (B) 50% CIP, (C) 60% CIP, and (D) 70% CIP. Data are presented as the mean ± SD (n = 3). Storage modulus of MREs with different CIP contents at 0.2 T (A–D) Storage moduli of MREs with (A) 40% CIP, (B) 50% CIP, (C) 60% CIP, and (D) 70% CIP. Data are presented as the mean ± SD (n = 3). Storage modulus of MREs with different CIP contents at 0.8 T (A–D) Storage moduli of MREs with (A) 40% CIP, (B) 50% CIP, (C) 60% CIP, and (D) 70% CIP. Data are presented as the mean ± SD (n = 3). Loss modulus of MREs with different CIP contents at 0 T (A–D) Loss moduli of MREs with (A) 40% CIP, (B) 50% CIP, (C) 60% CIP, and (D) 70% CIP. Data are presented as the mean ± SD (n = 3). Loss modulus of MREs with different CIP contents at 0.2 T (A–D) Loss moduli of MREs with (A) 40% CIP, (B) 50% CIP, (C) 60% CIP, and (D) 70% CIP. Data are presented as the mean ± SD (n = 3). Loss modulus of MREs with different CIP contents at 0.8 T (A–D) Loss moduli of MREs with (A) 40% CIP, (B) 50% CIP, (C) 60% CIP, and (D) 70% CIP. Data are presented as mean ± SD (n = 3). In the absence of a magnetic field, as the normal force increased from 2 to 15 N, the peak storage modulus increased from about 20 to 38 kPa. This peak value was obtained from the averaged curve of three tests. Pre-compression hardening is common in highly filled MREs. Although CIP particles are not strongly magnetized in zero field, normal force compresses the internal particle network and strengthens the particle-matrix interfacial interactions, thereby increasing the modulus. Nam et al.36 confirmed that the storage modulus increases with the pre-compressive strain in the absence of a magnetic field. Vatandoost et al.37 also found that increasing the pre-compressive strain from 6% to 21% significantly increased the storage modulus of MREs. Vatandoost et al.38 further showed that pre-straining directly increased the storage modulus under compression in the absence of a magnetic field. Therefore, the increase in storage modulus with normal force shown in Figure 3A mainly results from normal-force-induced densification of the particle network and enhanced filler-matrix interaction, rather than an experimental error. On the basis of Figures 3 and 5, the storage moduli of 40%, 50%, 60%, and 70% CIP MREs were compared under different magnetic fields and normal forces. In the absence of a magnetic field, the storage modulus increased with the CIP content. This indicated that the filler content is an important factor for controlling the material’s basic stiffness. At this stage, all samples exhibited a typical response—the modulus first increased rapidly and then decayed slowly. After applying a magnetic field, the peak storage modulus of all samples increased markedly. The higher the CIP content, the stronger the field-induced enhancement. Figure 4 corresponds to 0.2 T, and Figure 5 corresponds to 0.8 T. For example, under 0.8 and 15 N, the peak storage modulus of the 70% CIP sample was much higher than that of the 40% CIP sample. This suggests that a high-filler system is more favorable for particle connectivity and force-chain transmission under a magnetic field. At the same magnetic field and CIP content, the storage modulus increased with normal force. This trend was more obvious in high-filler systems. It indicates that particles are denser in high-CIP samples; so, the normal force can more effectively compress the particle structure, thereby improving the efficiency of force-chain transmission. In the absence of a magnetic field, the loss modulus of the MREs increased with the CIP content. The 70% CIP sample was clearly higher than the 40% CIP sample. This means that more magnetic particles increase the internal friction and energy dissipation. When the magnetic field increased to 0.2 T, the overall level and peak value of the loss modulus increased for all CIP contents. This is because the magnetic field promotes particle magnetization and strengthens the oriented structure, thereby increasing particle interaction and frictional dissipation. Under a strong field of 0.8 T, the loss modulus was higher than that under low-field conditions. It should be noted that at 2 N, a strong magnetic field led to high initial dissipation, but the particle structure was insufficiently stable under dynamic loading, and the modulus decreased rapidly thereafter. Figures 6, 7, and 8 show that the CIP content is a basic factor affecting the stiffness and damping of MREs. Increasing the CIP content simultaneously improves both the storage and loss moduli. Normal force can compact the particle structure, further enhancing the modulus response. This effect becomes stronger at higher CIP contents. When a strong magnetic field and a high normal force act together, the material shows a higher modulus and stronger oriented-structure reinforcement, indicating a synergistic enhancement among multiple physical fields. Figures 5 and 8 also show that the modulus growth rate as a function of magnetic field and CIP content is not uniform. Under the same magnetic field and filler content, the modulus growth rate at the low normal force of 2 N is higher than those at 5, 10, and 15 N. This is because the initial structure is looser at low normal force, and the initial modulus is lower, leaving more room for magnetic field enhancement. In addition, the weaker axial constraint allows particles to rotate and align more easily under the magnetic field. At high normal force, the particle structure is already pre-compressed, and the magnetic field has less space to further reconstruct it. Therefore, the modulus growth becomes smoother. These results indicate that low normal force is suitable for achieving a large tuning range of the modulus, while high normal force is suitable for high stiffness and stability. High-CIP MREs show transient softening under weak constraint The results show that under the coupled conditions of low normal force (2–5 N) and high magnetic field (≥0.5 T), the 70% CIP sample exhibited a clear, continuous decay in modulus. The storage modulus continued to decrease throughout the test window and did not stabilize at the end of the curve. This indicates that the material did not reach a steady state under this condition. Therefore, the final or steady-state value was not used to compare the modulus response and MR effect across different conditions to avoid bias. This behavior is related to the structural instability of the high-CIP sample under weak constraint. When the CIP content reached 70%, the fraction of silicone rubber matrix decreased, weakening its anchoring and support for the particle structure. Under low normal force, the axial constraint was insufficient, and the particle structure was not fully compressed. After a high magnetic field was applied, the particles were strongly magnetized, and the initial modulus increased sharply. However, under continuous oscillatory shear, the particle structure could slip, rearrange, or partially break down, reducing the load-bearing capacity and leading to shear softening. When the normal force exceeded 10 N, axial compaction suppressed particle sliding and rearrangement. The modulus decay was reduced, and the curves became more stable. This indicates that a higher normal force helps stabilize the particle structure. After testing, the contact surface remained intact, and no macroscopic damage was observed, suggesting that the behavior was not primarily due to the interfacial slip. The modulus decay observed in this study is a transient response under constant oscillatory shear, rather than being an equilibrium viscoelastic property. Under the conditions of 70% CIP, 2–5 N normal force, and ≥0.5 T magnetic field, the material continued to evolve during the test. This reflects the real structural change of a high-filler system under weak constraint. In contrast, under more stable conditions, such as a normal force of ≥10 N or a magnetic field of ≤0.5 T, the modulus tended to stabilize more readily. Therefore, peak or time-averaged values, rather than final steady-state values, were used for comparison. In summary, this decay is a real transient response of high-CIP MREs under the combined action of low normal force and strong magnetic field. The magnetic field produces nonlinear modulation across the CIP contents Figures 9, 10, 11, and 12 compare the modulus changes of samples with 40%–70% CIP under 2 and 10 N normal forces. The magnetic fields were 0, 0.1, 0.2, 0.5, and 0.8 T. Figures 9 and 10 show the storage modulus. Figures 11 and 12 show the loss modulus. Storage modulus under different magnetic fields at 2 N normal force (A) 40% CIP. (B) 50% CIP. (C) 60% CIP. (D) 70% CIP. Data are presented as mean ± SD (n = 3). Storage modulus under different magnetic fields at 10 N normal force (A) 40% CIP. (B) 50% CIP. (C) 60% CIP. (D) 70% CIP. Data are presented as the mean ± SD (n = 3). Loss modulus under different magnetic fields at 2 N normal force (A) 40% CIP. (B) 50% CIP. (C) 60% CIP. (D) 70% CIP. Data are presented as the mean ± SD (n = 3). Loss modulus under different magnetic fields at 10 N normal force (A) 40% CIP. (B) 50% CIP. (C) 60% CIP. (D) 70% CIP. Data are presented as the mean ± SD (n = 3). Figures 9, 10, 11, and 12 show that both the storage and loss moduli tended to increase with the increasing magnetic field. This trend was more obvious in high-CIP samples. To avoid purely qualitative analysis, three representative magnetic fields—0, 0.5, and 0.8 T—were selected. The time-averaged moduli were calculated at normal forces of 2 and 10 N. The results are shown in Tables 1 and 2. The values in the tables are the averages of 10 sampling points within the test window and were converted from Pa to kPa. Table 1Time-averaged storage modulus G′ of MREs under different CIP contents, normal forces, and magnetic fieldsCIP content/%Fn/N0 T/kPa0.5 T/kPa0.8 T/kPa40218.7864.0075.58401029.6274.8877.9350219.99106.67124.95501037.67131.44132.8360216.58191.89223.12601035.91222.69221.5170221.71439.29501.78701050.42662.21738.05G′ is the average value of all sampling points in the same time window from the averaged curve of three repeated tests; n = 3. Error bars represent ±SD. Time-averaged storage modulus G′ of MREs under different CIP contents, normal forces, and magnetic fields G′ is the average value of all sampling points in the same time window from the averaged curve of three repeated tests; n = 3. Error bars represent ±SD. Table 2Time-averaged loss modulus G″ of MREs under different CIP contents, normal forces, and magnetic fieldsCIP content/%Fn/N0 T/kPa0.5 T/kPa0.8 T/kPa4024.9816.7318.8340108.0618.8918.385024.2322.3125.8850107.5426.6526.266025.3356.6164.04601011.8266.7864.107028.04105.09119.06701017.80154.75155.52G″ is the average value of all sampling points in the same time window from the averaged curve of three repeated tests; n = 3. Error bars represent ±SD. Time-averaged loss modulus G″ of MREs under different CIP contents, normal forces, and magnetic fields G″ is the average value of all sampling points in the same time window from the averaged curve of three repeated tests; n = 3. Error bars represent ±SD. Combining Figures 9 and 10 with Table 1 shows that the magnetic field significantly increased the time-averaged storage modulus. For example, for the 40% CIP sample under 2 N normal force, when the field increased from 0 to 0.8 T, the average G′ increased from 18.78 to 75.58 kPa, about 4.0 times the value at 0 T. This aligns with the upward shift of the curves in Figure 9A, indicating that the external magnetic field improved the elastic energy storage capability. Combining Figures 11 and 12 with Table 2 shows that the loss modulus also increased with the magnetic field. For the 40% CIP sample under 2 N normal force, when the field increased from 0 to 0.8 T, the average G″ increased from 4.98 to 18.83 kPa, about 3.8 times the value at 0 T. This indicates that the external magnetic field improved not only energy storage but also viscous energy dissipation. The enhancement became more obvious as the CIP content increased. Under 2 N normal force, the curves of the high-CIP samples were generally higher than those of the low-CIP samples. Tables 1 and 2 show that, for the 70% CIP sample, G′ increased from 21.71 kPa at 0 T to 501.78 kPa at 0.8 T, about 23.1 times the zero-field value. G′′ increased from 8.04 to 119.06 kPa, about 14.8 times the zero-field value. This shows that a higher CIP content can greatly improve the magnetic field response. The main reason is that more magnetic particles and smaller particle spacing make it easier to form a continuous oriented structure and an effective load-transfer path. Normal force also affected the modulus level. Comparing Figure 9 with Figures 10, 11, and 12 showed that the modulus curves under 10 N were generally higher than those under 2 N. Taking 70% CIP at 0.5 T as an example, when the normal force increased from 2 to 10 N, the average G′ increased from 439.29 to 662.21 kPa, an increase of about 50.7%. The average G′′ increased from 105.09 to 154.75 kPa, an increase of about 47.3%. This indicates that a higher normal force can increase both energy storage and energy dissipation by compacting the particle structure. It should be noted that some high-field curves decreased markedly in the late stage of testing. Therefore, under some conditions, the time-averaged modulus at 0.8 T was close to, or slightly lower than, that at 0.5 T. For example, for 60% CIP at 10 N, G′ decreased slightly from 222.69 kPa at 0.5 T to 221.51 kPa at 0.8 T. For 40% CIP at 10 N, G′′ decreased slightly from 18.89 kPa at 0.5 T to 18.38 kPa at 0.8 T. This indicates that MRE behavior in the high-field region is not a simple linear enhancement but is jointly affected by particle rearrangement, structural relaxation, and normal constraint. To evaluate data reliability, each condition was independently repeated three times. The error bars in Figures 9, 10, 11, and 12 represent the standard deviation (SD), with n = 3. The time-averaged moduli in Tables 1 and 2 were also calculated from the averaged curves of the three tests. The relative standard deviation (RSD) was used to evaluate the error. The results showed that, under stable conditions (normal force ≥ 10 N and magnetic field ≤ 0.5 T), the average RSD of G′ and G′′ was about 5.1%. Under more fluctuating conditions (2 N, magnetic field ≥ 0.5 T, and CIP content ≥ 60%), the average RSD was about 8.3%. This error range is acceptable and indicates good repeatability. In summary, Figures 9, 10, 11, and 12 show the time-dependent changes of G′ and G′′ under different magnetic fields, CIP contents, and normal forces. Tables 1 and 2 provide the corresponding time-averaged moduli. The data in figures and tables support each other and jointly confirm that the magnetic field, CIP content, and normal force affect the dynamic mechanical properties of MREs. Coupling mechanism among magnetic field, normal force, and CIP content To evaluate the field-induced tunability of the MRE, the MR effect was introduced. The absolute modulus reflects the stiffness and energy dissipation of the material. However, it is not suitable for directly comparing the relative magnetic field response under different CIP contents and normal forces. Therefore, the modulus at 0 T, with the same CIP content and normal force, was used as the reference. The MR effects of the storage modulus and loss modulus were calculated separately. The MR effect corresponding to the storage modulus is defined as: The MR effect corresponding to the loss modulus is defined as: where MRG′ and MRG′′ are the MR effects of the storage modulus and loss modulus, respectively; B is the applied magnetic field strength; Fn is the normal force; ϕ is the CIP content; Ḡ′(B, Fn, ϕ) and Ḡ′′(B, Fn, ϕ) are the time-averaged storage modulus and loss modulus under the corresponding conditions; and Ḡ′(0, Fn, ϕ) and Ḡ′′(0, Fn, ϕ) are the corresponding moduli at 0 T under the same normal force and CIP content. As the time-sweep tests were used in this study, the MR effect was calculated using the average modulus within the same test window. This method reduces the influence of local peaks, transient fluctuations, and late-stage decay. It should be noted that the vertical axes in Figures 13 and 14 represent the MR effect, not the modulus itself. The MR effect refers to the relative increase in the modulus compared with the 0 T condition. Therefore, when the modulus at 0 T was low and the modulus under a strong magnetic field increased greatly, the MR effect could reach several hundred or even more than one thousand percent. Variation of MRGʹ with magnetic field strength for MREs with different CIP contents and normal forces (A) Fn = 2 N. (B) Fn = 5 N. (C) Fn = 10 N. (D) Fn = 15 N. MR effects were calculated from the mean values of three repeated tests (n = 3). Variation of MRG″ with magnetic field strength for MREs with different CIP contents and normal forces (A) Fn = 2 N. (B) Fn = 5 N. (C) Fn = 10 N. (D) Fn = 15 N. MR effects were calculated from the mean values of three repeated tests (n = 3). To describe the change in the MR effect in the high-field region, the high-field growth ratio Rsat,x was introduced. Rsat,x compares the increase in the MR effect from 0.5 to 0.8 T with that from 0.2 to 0.5 T. It should be noted that Rsat,x is used to describe only the change in the MR effect in the high-field region. It does not mean that the CIP particles have reached true magnetic saturation. where X represents G′ or G′′. When Rsat,x < 1, the increase in the MR effect from 0.5 to 0.8 T was smaller than that from 0.2 to 0.5 T. This means that the response growth rate decreased in the high-field region. When Rsat,x ≈ 1, the increases in the two field regions were similar. When Rsat,x > 1, the MR effect still showed strong growth in the high-field region. Based on Equations 1 and 2, Figures 13 and 14 show the variations of MRG′ and MRG′′ with magnetic field strength under different CIP contents and normal forces. In each figure, (A)–(D) correspond to Fn = 2, 5, 10, and 15 N, respectively. As shown in Figures 13 and 14, both MRG′ and MRG′′ tended to increase as the magnetic field strength increased. This indicates that the applied magnetic field improves both the energy storage and dissipation capacities of the MREs. The MR effect also increased with the increasing CIP content. This result shows that a higher CIP content strengthens the magnetic interactions between particles and improves the material’s magnetic field response. In the magnetic field range, the MR effect increased rapidly from 0 to 0.5 T. Beyond 0.5 T, the growth rate decreased for some samples. This indicates that the field-induced enhancement weakens in the high-field region and exhibits a plateau-like trend. The normal force also affects the MR effect. Under a low normal force, the modulus at 0 T was relatively low. Therefore, the relative increase induced by the magnetic field was greater, resulting in a stronger MR effect. As the normal force increased, the particle network was further compacted. Both the modulus at 0 T and the modulus under the magnetic field increased. As the reference modulus at 0 T increases, the relative MR effect may decrease. However, a higher normal force can strengthen the contact constraints between particles and between the particles and the matrix. This makes the modulus curves more stable. Overall, the magnetic field response of the MRE is jointly controlled by the magnetic field strength, normal force, and CIP content. The applied magnetic field enhances the magnetic interaction between particles. A higher CIP content increases the number of magnetic particles and improves the connectivity of the particle network. The normal force reduces the interparticle distance through pre-compression and improves structural stability. Therefore, the dynamic mechanical properties of the MREs are not controlled by a single factor. They result from the combined effect of magnetic-field-induced enhancement, normal-force-induced compaction and stabilization, and improved particle connectivity caused by higher CIP content. This mechanism explains the differences in MR effects across conditions and provides guidance for designing MREs with tunable stiffness. Equivalent viscoelastic modeling describes elastic and viscous responses To improve the theoretical depth of the rheological analysis, equivalent viscoelastic parameters were introduced. Previous studies have shown that MREs exhibit typical viscoelastic responses under harmonic shear loading. G′ represents elastic energy storage, and G′′ represents viscous energy dissipation. Constitutive magneto-mechanical models have also been developed to describe the mechanical and rheological behavior of isotropic and anisotropic MREs.39 Different models have been used to describe harmonic shear behavior of MREs, including modified Kelvin-Voigt models and Maxwell/Ramberg-Osgood combined models.40,41 The experiments in this study were primarily conducted at a fixed frequency of 1 Hz, which is insufficient to construct a generalized Maxwell model or a Prony series model, both of which require broad-frequency-sweep data. Therefore, a complete frequency-domain constitutive model was not established. Instead, an equivalent Kelvin-Voigt model, suitable for single-frequency, small-amplitude oscillatory shear, was used to semi-quantitatively describe the viscoelastic response of MREs.40,42 G′ and G″ were obtained in the same way as the MR effect, namely from the average of all sampling points in the same time window of the averaged curve from three repeated tests. Under small-amplitude sinusoidal shear, the stress response of the MREs can be expressed as the sum of elastic and viscous responses: where τ(t) is the shear stress, γ(t) is the shear strain, Geq is the equivalent elastic modulus, and ηeq is the equivalent viscosity. For a sinusoidal strain input: The complex shear modulus of the material can be written as: where G′ is the storage modulus, G′′ is the loss modulus, and ω is the angular frequency. According to the Kelvin-Voigt model, they can be written as: Therefore, the equivalent viscosity can be calculated from the experimentally measured loss modulus: The test frequency in this study was 1 Hz; so, the angular frequency can be calculated as follows: Thus, G″ can be converted to ηeq to evaluate viscous dissipation under different magnetic fields, normal forces, and CIP contents. G′ can be regarded as the equivalent elastic modulus and reflects the combined contributions of silicone-rubber matrix elasticity, support from the CIP-oriented structure, and normal-force compaction. The model indicates that the magnetic field mainly increases G′ by enhancing the oriented structure of magnetic particles. Normal force further increases G′ and ηeq by reducing particle spacing and strengthening contact constraints. As CIP content increases, particle contact and structural connectivity improve, thereby increasing both elastic energy storage and viscous energy dissipation. To describe the time-dependent modulus decay observed under some conditions, a first-order structural relaxation model was further introduced: where G′0 is the initial storage modulus, G′∞ is the long-time stable storage modulus, and τs is the structural relaxation time constant. τs can be used to characterize the stability of the particle structure during dynamic shear. A larger τs means that the structure is more stable and the modulus decays more slowly. A smaller τs means that particles are more likely to slide, rearrange, or break down locally. A previous study also showed that the stress relaxation and structural evolution of MREs can be described by Kelvin-Voigt-type viscoelastic models.34 Therefore, the equivalent Kelvin-Voigt model links G′ and G′′ to equivalent elastic contribution and viscous dissipation, respectively. The first-order structural relaxation model describes the time evolution of the modulus. Although this model is not a complete broadband viscoelastic constitutive model, it can semi-quantitatively characterize the viscoelastic response of MREs under the coupled effect of the magnetic field, normal force, and CIP content based on the available 1 Hz dynamic shear data. Time-dependent modulus decay reflects structural relaxation The experimental results show that G′ and G′′ of the MREs decreased with test time under some conditions. This behavior is mainly caused by matrix viscoelastic relaxation, particle-structure rearrangement, dynamic shear softening, and changes in interfacial dissipation. First, the silicone-rubber matrix is viscoelastic. Under continuous oscillatory shear, the matrix undergoes stress relaxation, and the load-bearing capacity gradually decreases. Second, the oriented structure formed by CIP particles in the magnetic field is not completely stable. During dynamic shear, local particle sliding, rotation, and rearrangement may occur, weakening the original load-bearing path and decreasing G′. This behavior is more pronounced at high CIP content, in a strong magnetic field, and with a low normal force. In this case, the initial particle magnetization is strong, but the axial constraint is insufficient; so, the structure is more likely to become locally unstable during continuous shear. This behavior may also be associated with Payne-effect-like softening in highly filled elastomer composites. In highly filled particle composites, continuous oscillatory shear can weaken or rearrange the particle-particle contact network and the filler-matrix interfacial structure, resulting in a decrease in stiffness. Changes in G″ reflect the combined effect of particle friction, interfacial sliding, and matrix viscous dissipation. As the particle structure gradually relaxes or rearranges, the energy-dissipation path changes, and G′′ may decrease or fluctuate. According to Equation 10, the structural relaxation time constant τs can be used to evaluate particle-structure stability. A high normal force can compact the particle network and slow the modulus decay. Under a low normal force, particle constraint is insufficient, especially under strong magnetic fields and high CIP content. Structural instability is then more likely, and the decrease in modulus becomes more pronounced. In summary, the time-dependent modulus decay of MREs is not an experimental error; it is a real transient response caused by matrix viscoelastic relaxation, dynamic particle rearrangement, changes in interfacial dissipation, and possible Payne-effect-like softening in the highly filled particle network. Discussion The dynamic mechanical properties of MREs are jointly affected by the CIP content, normal force, and magnetic field strength. In this study, MRE samples with 40%, 50%, 60%, and 70% CIP were prepared and tested under 2–15 N normal force and 0–0.8 T magnetic field. The results reveal the coupled regulation of MRE performance by these three factors. The main conclusions are as follows: (1)CIP content is a key factor determining the basic stiffness and loss characteristics of MREs. Without a magnetic field, increasing the CIP content clearly increases the average level and peak values of G′ and G′′. The modulus response of the 70% CIP sample is much stronger than that of the 40% CIP sample. High CIP content leads to denser particle contacts, strengthens the polymer matrix’s support, and increases energy dissipation due to particle friction. Therefore, CIP content is a core parameter for controlling the basic performance of MREs. CIP content is a key factor determining the basic stiffness and loss characteristics of MREs. Without a magnetic field, increasing the CIP content clearly increases the average level and peak values of G′ and G′′. The modulus response of the 70% CIP sample is much stronger than that of the 40% CIP sample. High CIP content leads to denser particle contacts, strengthens the polymer matrix’s support, and increases energy dissipation due to particle friction. Therefore, CIP content is a core parameter for controlling the basic performance of MREs. (2)Normal force has a dual effect on the MRE modulus. The averaged results of three tests show that increasing normal force increases G′ and G′′. A higher normal force (≥10 N) can compact the particle structure, suppress sliding, and improve stiffness and stability. In terms of the modulus growth rate, the magnetic-field-induced increase is faster under the low normal force of 2 N. This is because the initial modulus is lower at a low normal force, and the magnetic-field-tuning space is larger. It does not necessarily mean that the response sensitivity is higher. Therefore, low normal force is suitable for a large tuning range of the modulus, whereas high normal force is suitable for high stiffness and high stability. Normal force has a dual effect on the MRE modulus. The averaged results of three tests show that increasing normal force increases G′ and G′′. A higher normal force (≥10 N) can compact the particle structure, suppress sliding, and improve stiffness and stability. In terms of the modulus growth rate, the magnetic-field-induced increase is faster under the low normal force of 2 N. This is because the initial modulus is lower at a low normal force, and the magnetic-field-tuning space is larger. It does not necessarily mean that the response sensitivity is higher. Therefore, low normal force is suitable for a large tuning range of the modulus, whereas high normal force is suitable for high stiffness and high stability. (3)Magnetic field strength significantly increases the MRE modulus, but the increase gradually slows when the field exceeds 0.5 T. The magnetic field regulation range increases with the CIP content. Under a strong magnetic field of 0.8 T, the modulus enhancement of the 70% CIP sample is much higher than that of the 40% CIP sample. Especially above 0.5 T, the modulus changes nonlinearly with the magnetic field, indicating that this range is important for dynamic tuning of MRE performance. Magnetic field strength significantly increases the MRE modulus, but the increase gradually slows when the field exceeds 0.5 T. The magnetic field regulation range increases with the CIP content. Under a strong magnetic field of 0.8 T, the modulus enhancement of the 70% CIP sample is much higher than that of the 40% CIP sample. Especially above 0.5 T, the modulus changes nonlinearly with the magnetic field, indicating that this range is important for dynamic tuning of MRE performance. Limitations of the study The dynamic shear experiments were mainly conducted at a fixed frequency of 1 Hz and a fixed strain amplitude of 1%. Therefore, the available data support the comparison of coupled field-force-filler trends but are insufficient to build a broadband constitutive model, such as a generalized Maxwell model or a Prony series. The 1% strain amplitude was used as a unified small-amplitude testing condition, but it was not strictly within the linear viscoelastic region for all CIP contents, magnetic fields, and normal forces. The normal force changed the effective rheometer gap, with the representative relative compression increasing to about 3.38% at 15 N. Absolute modulus values under different normal forces may, therefore, include compression-induced geometric effects, although the relative trends remain comparable because the same fixture, loading protocol, and data-processing method were used. For high-CIP samples under low normal force and strong magnetic field, the modulus decay continued within the 120-s test window. Longer-duration tests and in situ structural characterization would help clarify the evolution of particle-chain rearrangement and interfacial dissipation. Resource availability Lead contact Requests for further information and resources should be directed to and will be fulfilled by the lead contact, Zhihong Lin (lin123hongzhi@163.com). Materials availability This study did not generate new unique reagents. The MRE specimens used in this study can be reproduced from commercially available materials using the formulations and preparation procedures described in the STAR Methods. Further information about specimen preparation is available from the lead contact upon reasonable request. Data and code availability •All data reported in this paper will be shared by the lead contact upon reasonable request. This includes the rheological time-sweep and amplitude-sweep data, representative rheometer-gap records, and the numerical data underlying the figures and tables. All data reported in this paper will be shared by the lead contact upon reasonable request. This includes the rheological time-sweep and amplitude-sweep data, representative rheometer-gap records, and the numerical data underlying the figures and tables. •This paper does not report original code. MATLAB R2024a (MathWorks) was used for data processing and figure preparation; no custom software or algorithms were developed for this study. This paper does not report original code. MATLAB R2024a (MathWorks) was used for data processing and figure preparation; no custom software or algorithms were developed for this study. •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon reasonable request. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon reasonable request. Acknowledgments This work was supported by the Fujian Provincial Natural Science Foundation of China (project nos. 2024J01317 and 2026I0034) and the 2025 Science and Technology Special Mission Service Collaboration Industrial Project of Sanming City, Fujian Province (project no. 2025-N-028). The authors thank the School of Mechanical and Electrical Engineering at Sanming University and Xiamen Ocean Vocational College for institutional support during this study. Author contributions Conceptualization, Z,L.; methodology, Z.L. and Z.G.; investigation, Z.G. and Y.H.; formal analysis, Z.G. and Y.H.; data curation, Z.G. and Y.H.; validation, Y.H.; software, Z.G.; visualization, Z.G.; writing – original draft, Z.L. and Z.G.; writing – review & editing, Z.L., Z.G., and Y.H.; supervision, Z.L.; funding acquisition, Z.L.; project administration, Z.L. All authors have reviewed and approved the final version of the manuscript. Declaration of interests The authors declare no competing interests. STAR★Methods Key resources table REAGENT or RESOURCESOURCEIDENTIFIERChemicals, peptides, and recombinant proteinsCarbonyl iron powder (CIP)BASF, GermanyCAS: 7439-89-6Two-component liquid silicone rubberShenzhen Guoyuan Technology, ChinaN/ADOWSIL SH 200 Fluid, 350 cStDow, USACAS: 63148-62-9Experimental models: OtherAnisotropic MRE specimens (40–70 wt% CIP)This paperN/ASoftware and algorithmsMATLAB R2024aMathWorkshttps://www.mathworks.com/products/matlab.htmlOtherMCR 302 rotational rheometerAnton Paar GmbH, AustriaModel: MCR 302DXKDP programmable DC power supplyXiamen Dexing Magnet Tech., ChinaSeries: DXKDPDX-102F handheld Gauss meterXiamen Dexing Magnet Tech., ChinaModel: DX-102FApreo 2 scanning electron microscopeThermo Fisher Scientific, USAModel: Apreo 2Magnetic-field curing and testing systemThis paperN/A Experimental model and study participant details This study did not involve human participants, animals, plants, microorganisms, or cell lines. The experimental specimens were anisotropic magnetorheological elastomers prepared from a two-component liquid silicone rubber system comprising carbonyl iron powder and dimethyl silicone oil. Method details Materials and equipment Carbonyl iron powder (CIP; BASF, Germany) was used as the magnetic filler. The particles had a regular spherical morphology, a density of at least 2.5 g/cm3, and an average particle size of 3.8–5.3 μm. A two-component liquid silicone rubber (density 1.08 g/cm3; Shenzhen Guoyuan Technology Co., Ltd., China) was used as the matrix. DOWSIL SH 200 Fluid (350 cSt; Dow, USA) was used as an additive. The carbonyl iron powder exhibits soft-magnetic behavior and a high magnetic response.43 Previous research reported that MREs prepared with BASF carbonyl iron powder show increasing magnetization with increasing magnetic field and gradually approach saturation near 0.65 T.44 This behavior helps explain the reduced growth rate of the modulus and magnetorheological effect in the high-field range of 0.5–0.8 T in this study. The magnetic field system was powered by a DXKDP series programmable DC power supply, and the magnetic flux density was measured with a DX-102F handheld gauss meter with real-time automatic zeroing (Xiamen Dexing Magnet Tech. Co., Ltd., China). Microstructures were examined using an Apreo 2 scanning electron microscope (Thermo Fisher Scientific, USA). Dynamic rheological testing was performed using an MCR 302 advanced rotational rheometer (Anton Paar GmbH, Austria). Sample preparation Material ratio The total mass of each sample was fixed at 10.5 g. The CIP mass fractions were 40%, 50%, 60%, and 70%. To serve as a release agent, a certain amount of dimethyl silicone oil (350CS) was added to each sample and mixed with silicone rubber and iron powder. The detailed formulation is shown in Table 3. Table 3Material composition of the MRE samplesCIP (g)Silicone rubber (g)Curing agent (g)Dimethyl silicone oil (g)CIP (g)Silicone rubber (g)Curing agent (g)Dimethyl silicone oil (g)4330.56220.552.52.50.571.51.50.5 Material composition of the MRE samples Mixing and vacuum degassing After weighing, the mixture was stirred for 6–8 min at 500–800 rpm. After the magnetic particles were uniformly dispersed and no obvious agglomeration was observed, the mixture was vacuum-degassed at 29 mmHg for 8–10 min. This step removed bubbles and improved the density and uniformity of the cured material. Magnetic-field-induced curing The degassed mixture was poured into a mold. The mold was then placed in a 500 mT uniform magnetic field, as shown in Figure 15. The magnetic-field generation system was powered by a DXKDP series programmable DC power supply, and the field intensity was checked using a DX-102F handheld gauss meter with real-time automatic zeroing. To ensure similarly oriented structures across samples, all samples were placed in the central region of the magnetic field and maintained the same induction direction. The samples were cured at 25 °C for at least 3 h. During curing, the magnetic particles tended to align with the magnetic field, forming an anisotropically oriented structure. The MRE preparation process is shown in Figure 15. Preparation process of the MREs Cutting of cured samples After curing, the MRE sheets were cut into circular specimens 15 mm in diameter and 1 mm thick using a punch die. This size met the rheometer parallel-plate fixture requirements. The geometry of the MRE specimen is illustrated in Figure 16, and the detailed dimensions are summarized in Table 4. Schematic of the MRE specimen Table 4Dimensions of the MRE specimenabcdeSize (mm)66221151 Dimensions of the MRE specimen Dynamic shear testing Dynamic shear tests were performed using an MCR 302 advanced rotational rheometer (Anton Paar GmbH, Austria) with a 20 mm parallel-plate fixture. The test temperature was 25 °C, the frequency was 1 Hz, and the strain amplitude was 1%. To study the effect of normal load, axial contact forces of 2 N, 5 N, 10 N, and 15 N were applied. The external magnetic-field system was powered by a DXKDP series programmable DC power supply, and the magnetic flux density was monitored using a DX-102F handheld gauss meter with real-time automatic zeroing. The direction of the external magnetic field was consistent with the samples’ curing-field direction, and the magnetic-field strengths were 0 T, 0.1 T, 0.2 T, 0.5 T, and 0.8 T. Representative rheometer gap records were further examined to assess the influence of normal force on the effective test geometry. The results show that the testing gap decreased from about 1.036 mm at 2 N to about 1.001 mm at 15 N, and the relative compression increased to about 3.38% at 15 N. This indicates that the applied normal force caused measurable compression of the specimen and changed the effective testing geometry. Therefore, the absolute values of the modulus under different normal forces may be affected by sample compression. However, all samples were tested using the same fixture, loading protocol, and data-processing method. Thus, the relative trends caused by the magnetic field, the normal force, and the CIP content remain comparable. The corresponding testing gaps under different normal forces are summarized in Table 5. Table 5Representative testing gap under different normal forcesNormal force Fn (N)Representative testing gap/mmGap reduction relative to 2 N (μm)Relative compression (%)21.036 ± 0.0080051.028 ± 0.00880.77101.014 ± 0.009222.12151.001 ± 0.010353.38Values were obtained from representative rheometer gap records; relative compression was calculated using the 2 N test gap as the reference. Representative testing gap under different normal forces Values were obtained from representative rheometer gap records; relative compression was calculated using the 2 N test gap as the reference. The 1% strain amplitude was selected for two reasons. First, it kept the test within a small-amplitude oscillatory range, thereby reducing the influence of nonlinear effects on the characterization of the modulus. Second, together with the normal load, it improved the contact stability between the sample and the fixture and reduced the risk of interfacial slip. After testing, the samples were visually inspected. No obvious slip marks, macroscopic cracks, or edge damage were observed, indicating good validity and repeatability of the data. To verify the rationality of the 1% strain amplitude, an amplitude sweep test was also conducted. The test was performed at 25 °C and an angular frequency of ω = 94.2 rad/s, corresponding to about 15 Hz. The shear strain amplitude ranged from 0.00996% to 20.1%. The trends of G′ and G′′ with strain were used to identify the small-amplitude response range and to support the strain selection for subsequent dynamic shear tests. It should be noted that the 1% strain was not strictly defined as being within the linear viscoelastic region (LVR). In this study, it was used as a unified small-amplitude dynamic shear condition. Under this condition, the modulus trends under different magnetic fields, normal forces, and CIP contents are comparable. Normal force may cause slight compression, and minor edge extrusion may occur under extreme conditions. These effects may influence the absolute modulus values, but they do not change the relative trends among the different test conditions. To ensure the reliability of the test results, each test condition was repeated three times. The curves in the following figures are shown as the mean of three repeated tests, and the error bars represent the standard deviation (SD), with n = 3. The detailed test parameters are listed in Table 6. The rheometer test system is shown in Figure 17. Table 6Test parameters of the MRE samplesItemTest parameterFixed dynamic test frequency1 HzFixed dynamic test strain1%Amplitude sweep angular frequency94.2 rad/s, about 15 HzTest temperature25°CAxial force2, 5, 10, and 15 NAmplitude sweep strain range0.00996%–20.1% Test parameters of the MRE samples Anton Paar MCR 302 rotational rheometer test system Figure 18 shows the SEM image of the MRE sample after magnetic-field-induced curing. It can be seen that the CIP particles are not completely randomly distributed in the silicone rubber matrix. Instead, they show a certain degree of orientation along the magnetic-field-induced direction. Some particles are connected with each other and form local chain-like or strip-like particle-rich structures. This indicates that the applied magnetic field during curing can induce directional assembly of magnetic particles along the magnetic field. SEM morphology and CIP particle orientation structure of the magnetic-field-cured MRE sample Scale bars, 100 μm. This oriented particle-chain structure is an important microstructural basis for the magnetic-field response of anisotropic MREs. In an external magnetic field, CIP particles aligned with the field can generate stronger magnetic interactions, thereby enhancing particle-network connectivity and improving load transfer. Therefore, this structure increases the storage and loss moduli of the MRE and provides microstructural evidence for the field-induced enhancement observed in subsequent rheological tests. Rheological characterization The dynamic mechanical properties of MREs are affected by both the external magnetic field and mechanical load. In the dynamic shear mode, the material shows a viscoelastic response. This response can be described by the complex shear modulus G∗. G∗ consists of the storage modulus G′ and the loss modulus G′′, as follows: According to ISO 4664,45 a small sinusoidal shear displacement x(t) is applied to the MRE specimen during dynamic shear testing, and the resulting shear force F(t) is measured. The shear strain γ(t) and shear stress τ(t) are then obtained as: where h is the specimen thickness and A is the shear area. The complex shear modulus G∗ can be calculated from the amplitude ratio of stress to strain. Its absolute value |G∗| is: The loss angle δ is used to describe material damping. It can be calculated from the hysteresis loop formed by the force-displacement curve in one loading cycle. If the loop area is Ah, then sin δ is: where Ah is the area of the force-displacement hysteresis loop. The storage modulus G′ and loss modulus G′′ can be calculated from |G∗| and δ: Data processing and anisotropic MRE design To ensure a consistent comparison among different test conditions, the data selection method for the following curves and quantitative results is first clarified. All curves are plotted using the average values from three repeated tests. The error bars represent the standard deviation (SD), with n = 3. For comparison of modulus values across different conditions, the time-averaged value over the same test window is used as the primary characteristic. The peak value is only used to describe the maximum instantaneous response under a specific condition. The MR effect and equivalent viscoelastic parameters are also calculated from the average values over the same test window. This method reduces the influence of local peaks, late-stage decay, and transient fluctuations. Since modulus decay still occurs under some conditions with high CIP content, low normal force, and strong magnetic field, the final value or steady-state value is not used as the main basis for comparison. Data processing and figure preparation were performed using MATLAB R2024a (MathWorks). Based on this data-processing method, the following storage and loss modulus curves describe the dynamic evolution of MRE under a constant magnetic field and a constant normal force. The x axis in each figure represents the test time. The original test window was 0–120 s. The modulus values were recorded at 31.04 s, 41.07 s, 51.09 s, 61.11 s, 71.14 s, 80.66 s, 90.17 s, 99.69 s, 109.2 s, and 118.2 s. Using the same test window and data-processing method ensures that the results are comparable among different CIP contents, magnetic fields, and normal forces. In this study, anisotropic MREs were prepared by magnetic-field-induced curing. During curing, the CIP particles tended to align with the applied magnetic field, forming chain- or strip-like, particle-rich structures. The SEM image in Figure 18 further shows that particle-chain structures and an oriented morphology are aligned with the magnetic-field-induced direction. Compared with isotropic MREs with randomly dispersed particles, this anisotropic particle structure can enhance magnetic interactions between particles and improve load-transfer capability under an external magnetic field. In the present test configuration, the direction of the applied magnetic field was consistent with the curing-induced direction. Therefore, the magnetic field could further strengthen interactions within the particle-chain structure, thereby affecting the storage and loss moduli of the MRE. The carbonyl iron powder (CIP) content is a key factor affecting the magnetorheological effect of MRE. Previous studies have shown that a higher CIP content can improve the magnetic-field response of MRE. However, excessive filler content may also cause particle aggregation, poor interfacial compatibility, material embrittlement, and reduced flexibility. These problems are unfavorable for flexible applications such as sensors. Therefore, 40 wt% CIP was selected as the basic composition in this study. Samples with 50 wt%, 60 wt%, and 70 wt% CIP were also prepared for comparison. This design was used to systematically study the effects of filler content, magnetic field strength, and normal force on the dynamic mechanical properties of MRE. Quantification and statistical analysis Each test condition was independently repeated three times. Curves are presented as the mean of three repeated tests, and error bars represent the standard deviation (SD), n = 3. Time-averaged moduli were calculated from the same 10 sampling points in the 0–120 s test window. Relative standard deviation (RSD) was used to assess repeatability under both stable and fluctuating test conditions. Data processing and figure preparation were performed using MATLAB R2024a (MathWorks). No inferential statistical tests were performed in this study. References Associated Data Data Availability Statement •All data reported in this paper will be shared by the lead contact upon reasonable request. This includes the rheological time-sweep and amplitude-sweep data, representative rheometer-gap records, and the numerical data underlying the figures and tables. All data reported in this paper will be shared by the lead contact upon reasonable request. This includes the rheological time-sweep and amplitude-sweep data, representative rheometer-gap records, and the numerical data underlying the figures and tables. •This paper does not report original code. MATLAB R2024a (MathWorks) was used for data processing and figure preparation; no custom software or algorithms were developed for this study. This paper does not report original code. MATLAB R2024a (MathWorks) was used for data processing and figure preparation; no custom software or algorithms were developed for this study. •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon reasonable request. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon reasonable request.