Cobalt(II) chloride hexahydrate (CoCl₂·6H₂O, reagent grade, 99.9%, molecular weight 291.03), ferric chloride hexahydrate (FeCl₃·6H₂O, analytically pure, 99%, molecular weight 270.30), acrylamide (AM, analytically pure, 99.0%, molecular weight 71.08), ammonium persulfate (APS, analytically pure, 98%, molecular weight 228.20), N,N′-methylenebisacrylamide (MBA, 99%, molecular weight 154.17), sodium hydroxide (NaOH, analytically pure, 96.0%, molecular weight 40.00) were all purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. Aniline (AN, ≥ 99.5%, molecular weight 93.13), hydrochloric acid (HCl, 36.5-38.0 wt.%, molecular weight 36.46) and ethylene glycol (EG, analytically pure, ≥ 98%, molecular weight 62.07) were purchased from Chongqing Chuandong Chemical Co., Ltd. Glycerol (Gly, analytically pure, ≥ 99.0%, molecular weight 92.09) was purchased from Xilong Scientific Co., Ltd.

Take a certain mass of CoCl₂·6H₂O and FeCl₃·6H₂O and dissolve them sequentially in 45 mL of ethylene glycol to form solution a. Then dissolve NaOH in 50 mL of ethylene glycol to form solution b. Slowly add a to b while stirring magnetically in a 50 °C water bath for 2 h; Next, react it in a reactor at 200 °C for 8 h. After centrifugation, washing, and drying, CoFe₂O₄ magnetic particles are obtained.

Disperse CoFe₂O₄ nanoparticles in 75 mL of deionized water, then add 0.1 mL of AN and stir for 20 min. Adjust the pH to 1 using HCl, then add the initiator APS. The mixture was stirred at 0-4 °C for 6 h, followed by centrifugation, washing, and drying to obtain PANI@CoFe₂O₄ conductive magnetic nanoparticles.

A specified amount of PANI@CoFe₂O₄ was dispersed in 100 mL of deionized water and uniformly homogenized by mechanical stirring. Subsequently, 8 g of acrylamide (AM), 1.9 mL of glycerol (Gly), and 0.08 g of N,N′-methylenebisacrylamide (MBA) were sequentially introduced into the PANI@CoFe₂O₄ dispersion and thoroughly mixed. The polymerization initiator, ammonium persulfate (APS, 0.195 g), was then added to the mixture, and the resulting solution was transferred into a custom-designed mold for gelation. The obtained hydrogel was designated as X-PG/P@CoFe, where PG denotes the PAM/Gly network, and X indicates the weight percentage (%) of PANI@CoFe₂O₄ filler incorporated into the hydrogel.

Sample microstructures were examined by scanning electron microscopy (SEM, ZEISS Sigma 300, Carl Zeiss AG). Chemical composition was analyzed via Fourier transform infrared spectroscopy (FT-IR, Nicolet iS50, Thermo Fisher Scientific) and X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha, Thermo Fisher Scientific), and crystal structures were determined by X-ray diffraction (XRD, Rigaku Ultima IV, Suzhou Dehe Scientific Instrument Co., Ltd). Thermal behavior of the hydrogels was assessed using differential scanning calorimetry (DSC, Q10, TA Instruments), while compressive properties were measured with an electronic universal testing machine (GB/T 1041-2008, CMT6104，Mettler Industrial Systems (China) Co., Ltd.). Water absorption (Q) and retention (W) capacities were quantified according to Equations (1) and (2)^[11].

(1) $$ Q=\frac{m_{t}-m_{0}}{m_{0}} \times 100 \% $$

(2) $$ W=\frac{w_{t}}{w_{0}} \times 100 \% $$

Here, m_t denotes the weight of the sample in its original state, m₀ denotes the weight of the sample in a dry state; w_t denotes the mass measured after exposure to air for t h, and w₀ denotes the mass immediately upon reaching swelling equilibrium. Using a flexible strain sensor for sensitivity performance testing, the sensitivity coefficient (GF) is defined as the relative change in resistance [expressed as (R-R₀)/R₀, where R₀ and R represent the initial and real-time resistance] divided by the strain (ε). The electrical conductivity of the samples was determined using a standard four-probe resistance tester (FT-331). Using a vector network analyzer (N5225B, Keysight, Malaysia), the scattering parameters (S11, S12, S21, and S22), along with the complex permittivity and complex susceptibility used for analyzing EMI shielding characteristics. EMI shielding performance, including total SET, reflected SER, and absorbed SEA, as well as the coefficients for reflectance (R), absorptance (A), and transmittance (T), is calculated using the following Equations (3) to (8)^[36,37].

(3) $$ R=\left|S_{11}\right|^{2} $$

(4) $$ T=\left|S_{21}\right|^{2} $$

(5) $$ A=1-T-R $$

(6) $$ S E_{R}=10 \log \frac{1}{T} $$

(7) $$ S E_{A}=10 \log \frac{1-R}{T} $$

(8) $$ S E_{T}=S E_{R}+S E_{A} $$

Normality was assessed using the Shapiro-Wilk test prior to parametric analysis. Given the small sample size (n = 3), the data were considered approximately normally distributed. Descriptive statistics were calculated for all datasets. Quantitative data are expressed as mean ± standard deviation (SD). Comparisons between two groups were performed using an unpaired two-tailed t-test. Statistical significance was defined as P < 0.05. All statistical calculations were performed using Origin 2021.

As illustrated in Figure 1A, ethylene glycol promoted the nucleation and crystal growth of CoFe₂O₄ and facilitated the formation of the spinel phase. The abundant surface metal-active sites of CoFe₂O₄ can coordinate with the amino nitrogen of aniline, thereby enriching aniline at the particle interface and favoring subsequent interfacial oxidative polymerization. Upon initiation by APS, aniline polymerized to form PANI, which coated the CoFe₂O₄ surface to yield the PANI@CoFe₂O₄ composite. Subsequently, as shown in Figure 1B, the composite was introduced into the Gly/PAM precursor solution and underwent in situ gelation, resulting in the formation of the PANI@CoFe₂O₄ hydrogel.

As shown in Figure 2A, the CoFe₂O₄ synthesized in ethylene glycol exhibits a relatively uniform spherical morphology. After compositing with PANI in the aqueous phase [Figure 2B and C], the CoFe₂O₄ particles were clearly encapsulated by a PANI coating layer. EDS elemental mapping and energy-dispersive spectroscopy analysis [Supplementary Figure 1] further confirmed the homogeneous distribution of C, N, Fe, and Co throughout the composite. Figure 2D-F shows the scanning electron microscopy image of the PG hydrogel, which exhibits a well-defined three-dimensional porous structure. When CoFe₂O₄ is introduced into the PG hydrogel, the pore structure of the hydrogel is disrupted [Figure 2E]. This phenomenon may be attributed to the aggregation and deposition of CoFe₂O₄ particles, which interact strongly with the polymer matrix. Such interactions lead to a tighter packing of the polymer chains, thereby reducing the size of the pore structures^[38,39]. In contrast, when PANI@CoFe₂O₄ is incorporated, the pore structure becomes more abundant [Figure 2F]. The FTIR spectrum [Figure 2G] shows that the in situ polymerized PANI under acidic conditions exhibits a stretching vibration of the C=C bond at 1,604 cm^-1, a stretching vibration of the C-N bond at 1,498 cm^-1, and bending vibrations of the C-H bonds in the aromatic ring at 1,365 cm^-1 and 775 cm^-1, as well as out-of-plane bending vibrations^[40,41]. For CoFe₂O₄, the stretching vibration peaks of the Fe-O bonds appear at 419 cm^-1 and 1,070 cm^-1, while the stretching vibration of the Co-O bond is observed at 596 cm^-1[42-44]. For the PANI@CoFe₂O₄ composite material, the characteristic peaks at 1,070 cm^-1 and 596 cm^-1 remain, indicating that the Fe-O and Co-O bonds are still present in the composite. Additionally, the peaks corresponding to PANI can also be observed, indicating the successful integration of PANI with CoFe₂O₄. Meanwhile, at the same time, as shown in Figure 2H, the diffraction peaks corresponding to each crystal face in the spinel structure perfectly match those in the standard card (No. 22-1086)^[45], confirming the successful synthesis of the CoFe₂O₄ spinel structure. PG organohydroges were prepared using glycerol (Gly) and acrylamide (AM) as matrix components. This process involves the free-radical polymerization of AM at 60 °C to form PAM^[46]. As shown in Supplementary Figure 2, PANI reduces the aggregation or sedimentation of CoFe₂O₄, enabling its uniform dispersion within the hydrogel, which helps promote the formation and stability of the hydrogel pore structure. In addition, FT-IR analysis of the PG hydrogel [Figure 2I] showed characteristic bands at 2,934, 1,415, and 1,037 cm^-1, corresponding to C-H stretching, O-H bending, and C-O stretching vibrations of glycerol, respectively^[47], confirming the successful construction of the PG hydrogel.

XPS analysis [Figure 3A] confirmed that spinel CoFe₂O₄ mainly consists of Co, Fe, and O, with no detectable impurities other than adventitious carbon. The high-resolution XPS spectra of Co 2p, Fe 2p, and O 1s are shown in Figure 3B-D. In Figure 3B, the two principal peaks at 780.5 and 796.6 eV are assigned to Co 2p_3/2 and Co 2p_1/2, respectively. The peaks at 780.2 and 796.1 eV correspond to Co⁺, whereas those at 778.8 and 794.2 eV are attributed to Co^2[48,49]. The spectrum of Fe2p [Figure 3C] shows characteristic peaks at binding energies of 711.0 eV and 724.7 eV, corresponding to Fe 2p_3/2 and Fe 2p_1/2. The peaks at 710.9 eV and 724.1 eV are attributed to Fe³⁺, while the peaks at 713.7 eV and 726.7 eV correspond to Fe^2+[48,50]. The O 1s XPS spectrum exhibits three characteristic peaks [Figure 3D]. The peak at 530.5 eV corresponds to the metal-oxygen bonds, while the peak at 531.6 eV is assigned to defect oxygen. The peak at 533.0 eV arises from adsorbed water molecules on the surface of CoFe₂O₄^[50,51]. When coated with PANI, as shown in Figure 3E, it is evident that the PANI@CoFe₂O₄ composite nanoparticles contain an additional N element. The N 1s spectrum [Figure 3F] shows three peaks at binding energies of 399.2 eV, 399.7 eV, and 400.6 eV, which are attributed to imine nitrogen (=N-), amine nitrogen (-NH-), and protonated nitrogen (-N⁺-)^[52]. The binding energies of Co, Fe, and O in PANI@CoFe₂O₄ do not exhibit significant shifts [Figure 3G-I], suggesting that no new covalent bonds were formed between PANI and CoFe₂O₄ and that the two components were mainly integrated through interfacial interactions.

The melting and crystallization behaviors of the hydrogels were investigated by differential scanning calorimetry. As shown in Figure 4A and B, the incorporation of functional fillers induced slight variations in the melting and crystallization behaviors of the hydrogel, which can be attributed to their influence on the microcrystalline domains and chain packing of the polymer network. In terms of mechanical performance, PANI@CoFe₂O₄ acted as a multifunctional reinforcing filler, enhancing the mechanical properties of the organohydrogel through multiple interactions and structural effects. As shown in Figure 4C and D, the mechanical properties of the hydrogel were improved after incorporation of functional fillers. Among the samples investigated, the compressive strength (679.8 ± 13.60, n = 3) and modulus (1,258.9 ± 25.18, n = 3) of 4-PG/PANI hydrogel were significantly higher than those of PG hydrogel (376.3 ± 7.53 and 813.6 ± 16.27, n = 3, unpaired t-test, P < 0.0001, conformed to a normal distribution). Although both PANI and CoFe₂O₄ contributed to mechanical reinforcement, their strengthening mechanisms were different. The N- and O-containing functional groups on the PANI chains formed reversible physical crosslinking sites with the organohydrogel matrix, which helped maintain the integrity of the network under small deformation and partially dissociated or slid under large deformation to dissipate energy^[19,53], thereby improving the elongation at break and toughness. In contrast, CoFe₂O₄ possesses a rigid structure with a mechanical modulus much higher than that of the soft polymer network. When uniformly dispersed within the hydrogel network, it can effectively bear and distribute external forces, thereby reducing local stress concentrations^[54]. However, CoFe₂O₄ exhibits poor dispersibility within the hydrogel. Therefore, we introduced PANI onto its surface to overcome this limitation. When 4% PANI@CoFe₂O₄ filler is incorporated, the organohydrogel 4-PG/P@CoFe achieves a compressive strength (644.7 ± 25.89, n = 3) and modulus (1,076 ± 31.53, n = 3) that are significantly higher than those of PG hydrogel (376.3 ± 7.53 and 813.6 ± 16.27, n = 3, unpaired t-test, P < 0.0001, conformed to a normal distribution).The hydrogel’s 3D cross-linked network structure not only endows it with excellent mechanical properties, such as toughness, elasticity, and fatigue resistance, but also ensures its outstanding water absorption and water retention capabilities. The storage of water molecules within its three-dimensional porous structure generally determines the stability of the hydrogel. Accordingly, the water retention capacity and water absorption performance of the glycerol-water binary system was quantitatively evaluated. Figure 4E demonstrates that the incorporation of glycerol exerts a significant enhancing effect on water retention capability. Under the same conditions, the hydrogel without glycerol exhibits a faster water loss rate than the glycerol-containing hydrogel. After 15 days of incubation under room temperature conditions, the water retention rate increases from 63.4 ± 0.75% to 77.1 ± 0.91%. Regarding the water absorption performance [Figure 4F], it can be observed that the hydrogel without glycerol exhibits a faster water uptake rate, indicating that water molecules permeate more rapidly in the glycerol-free system. Figure 4G shows a schematic diagram of the molecular chains moving relative to each other when the hydrogel is loaded. This further demonstrates that glycerol hinders the diffusion of water molecules. Therefore, the PG hydrogel can maintain its initial state and excellent performance over a longer period. Compared with conventional hydrogels, this improvement ensures the long-term preservation of structural integrity and functional performance. The surface of the PG/P@CoFe hydrogel contains abundant polar functional groups, enabling strong interfacial adhesion to various substrates, including biological tissues, inorganic glass, and plastics [Figure 4H]. Such robust adhesion provides a solid foundation for the integration of wearable sensors.

In general, EMI shielding originates from reflection, absorption, and multiple internal reflections, which are associated with the responses of mobile charge carriers, electric dipoles, and internal interfaces/surfaces^[55,56]. Moreover, the porous structure not only reduces the density of the material but also provides a large surface area and abundant accessible active sites. These active sites facilitate the scattering or multiple reflections of electromagnetic waves (EMWs), thereby promoting energy dissipation^[57,58]. Meanwhile, the three-dimensional network structure increases the probability that the incident electromagnetic waves undergo multiple reflections and attenuation. As shown in Figure 5A-D, the hydrogels exhibit efficient and stable EMI shielding performance in the X-band, with the EMI shielding effectiveness increasing as the frequency rises. This variation can be attributed to several factors. (1) Enhanced dipole absorption: Water molecules possess a large permanent dipole moment, and under an alternating electromagnetic field, they undergo continuous orientational polarization. This process leads to stronger attenuation of interfering electromagnetic waves within the hydrogel, thereby dissipating the electromagnetic energy in the form of heat^[57]. (2) Increased polarization relaxation: At higher frequencies, the electromagnetic field oscillates more rapidly, causing more frequent polarization relaxation of electric dipoles, which in turn enhances conduction loss^[59]. More intense reflection and scattering: The shorter wavelengths associated with high-frequency electromagnetic waves interact more readily with the electric dipoles of water molecules and are more effectively reflected and scattered within the porous structure of the hydrogel, thereby generating greater energy dissipation^[60]. As can be seen from Supplementary Figure 3, the electromagnetic shielding performance of hydrogels filled with the composite material formed by PANI and CoFe₂O₄ is superior to that filled with the two components separately. The reason is that PANI coating can reduce the sedimentation of CoFe₂O₄ at the bottom of the hydrogel. Although water has a strong ability to dissipate electromagnetic waves, the pure PG hydrogel cannot achieve a satisfactory SE value due to the absence of conductive and magnetic components. This indicates the necessity of incorporating conductive and magnetic components into the PG hydrogel to achieve high electromagnetic shielding performance. As shown in Figure 5E, the electrical conductivity of the hydrogel changes significantly after the incorporation of the composite material. The electrical conductivity (0.55 ± 0.01, n = 3) and compressive performance (50.27 ± 2.51, n = 3) of 4-PG/P@CoFe hydrogel were significantly higher than those of PG hydrogel (0.0648 ± 0.00389 and 17.51 ± 1.05, n = 3, unpaired t-test, P < 0.0001, conformed to a normal distribution). Although the PG/PANI hydrogel exhibits higher electrical conductivity than the 2-PG/P@CoFe hydrogel, its shielding performance relies primarily on free-electron reflection. Due to the absence of dielectric polarization, interfacial polarization, and magnetic loss-related absorption mechanisms, its overall shielding effectiveness remains limited^[61]. In contrast, the 4-PG/P@CoFe hydrogel exhibits a moderately reduced conductivity while significantly enhancing impedance matching, dielectric loss, magnetic loss, and multiple scattering. This results in an absorption-dominated shielding mechanism, leading to electromagnetic shielding performance far superior to that of the 4-PG/PANI hydrogel. The present results suggest that simply increasing conductivity does not necessarily improve EMI shielding performance. A possible reason is that excessively high conductivity may deteriorate impedance matching, thereby increasing surface reflection^[62]. As shown in Figure 5F, by analyzing the R, A and T coefficients, we can gain additional understanding of the shielding mechanism. It can be observed that the 4-PG/PANI hydrogel exhibits slightly higher R values than the other samples due to its relatively strong dielectric properties, which lead to increased impedance mismatch. More importantly, the T values of all hydrogels are close to zero, indicating their high efficiency in resisting electromagnetic interference^[63,64]. Meanwhile, the A values are consistently higher than the R values, indicating that absorption is the dominant EMI shielding mechanism^[65]. The schematic illustration of the electromagnetic shielding mechanism [Figure 5G] further demonstrates the EMI shielding process of the hydrogel. Table 1 Comparison of the electromagnetic shielding performance of this work with other multifunctional hydrogels^{[5,11,27,66-72]}.

To meet the demands of more complex environments, we further evaluated the infrared stealth performance of the hydrogels. Hydrogels containing different fillers were placed on an 80 °C heating plate for 20 min, and their surface temperatures were recorded at various time intervals using an infrared thermal imaging camera. As shown in Figure 6A, the surface temperature of the hydrogel gradually increases with time. After 15 min, the equilibrium radiation temperature of the PG hydrogel stabilizes at approximately 45 °C, which is about 35 °C lower than the heating plate. This indicates that hydrogel exhibits a noticeable thermal-insulation effect. This phenomenon may be attributed to the following factors. (1) The presence of glycerol in the hydrogel slows down the evaporation rate of water and promotes the formation of a dense and stable hydrogen-bonding network within the three-dimensional gel matrix. This significantly hinders heat transfer during conduction, thereby imparting the material with a low thermal conductivity^[65,73]. (2) The three-dimensional microscopic network structure of the hydrogel suppresses convective heat transfer between the material and the surrounding air^[73]. It is the synergistic effect of these factors that effectively suppresses heat conduction and convection, thereby slowing the temperature rise and enabling excellent infrared stealth performance. However, it is noteworthy that after the incorporation of fillers [Figure 6B-E], the heating rate of the hydrogels increases, and the surface reaches its equilibrium radiation temperature within approximately 10 min. This is because of the contact or close packing between fillers form thermal bridges, creating continuous heat-transfer pathways that allow heat to penetrate the material more easily. However, because the heat-source intensity and surface heat-dissipation conditions remain unchanged, the system ultimately reaches thermal equilibrium at the same heat-flux balance point. As a result, the equilibrium radiation temperature shows only minor variation.

To better simulate the practical application scenario, the hydrogel was applied to the palm. As shown in Figure 6F, the infrared signal of the bare palm exhibited a distinct color contrast relative to the surrounding background. In contrast, the region covered by the hydrogel blended well with the environment and displayed a thermal signature similar to that of the surroundings, demonstrating the excellent infrared stealth capability of the hydrogel. Figure 6G and H respectively show the change in hydrogel surface temperature over time and the infrared stealth mechanism.

As shown in Supplementary Figure 4, the gauge factor (GF) of 4-PG/P@CoFe exhibits a value of 2.88 over a 0%-100% strain range, demonstrating excellent linearity (R² = 0.997). This linear response is attributed to the PANI@CoFe₂O₄ nanoparticles, which, as conductive materials embedded within the organohydrogen network, effectively maintain the continuity of the conductive pathways under applied strain. The 4-PG/P@CoFe hydrogel, with its excellent mechanical properties, electrical conductivity, and adhesion, is highly suitable for real-time sensing and monitoring of human motion and health conditions. As shown in Figure 7, the 4-PG/P@CoFe hydrogel exhibits good linearity and sensitivity, enabling real-time detection of various strains and resistance signals generated under different motion states. It can capture subtle facial expressions in other areas, such as frowning [Figure 7A] and changes in mouth shape [Figure 7B]. When fixed on the throat [Figure 7C], the vibrations of the Adam’s apple can be detected as we shout the words ‘Bing Dun Dun’, resulting in a clear and regular waveform output. Due to the Poisson effect of the hydrogel, different bending amplitudes generate distinct and regular electrical signals, as varying degrees of stretching induce different levels of network contraction and resistance change^[74]. The signal intensity correlates with the magnitude of movement, enabling precise motion detection. Furthermore, the hydrogel was used to track joint motion. Figure 7D-F shows the electrical signals obtained from finger, wrist, elbow, and knee bending at various angles. These results further confirm the rapid response capability and signal stability of the 4-PG/P@CoFe hydrogel sensor under dynamic, real-time monitoring conditions. The outstanding sensing performance of the 4-PG/P@CoFe hydrogel provides a solid functional foundation for its potential applications in human motion monitoring, health management, and gesture recognition.