FeMn‐MOF@Bamboo Carbon Synergy: Lightweight Composites for Versatile and High‐Efficiency Electromagnetic Protection

ABSTRACT

With the rapid advancement of fifth‐generation communication technology and the widespread adoption of modern electronic devices, the development of lightweight and efficient multifunctional electromagnetic interference (EMI) shielding materials has become essential for addressing increasingly complex electromagnetic (EM) environments. In this study, natural bamboo sheets were used as a skeleton, and ultrathin lightweight FeMn@C/BC composite materials were successfully fabricated through the in situ growth of iron–manganese Prussian blue analogue (FeMn PBA) on their surfaces, followed by hot‐pressing and carbonization treatments. Owing to the abundant EM coupling network and unique hierarchical porous structure provided by the bamboo‐derived carbon skeleton and magnetic FeMn carbon nanocages. The developed composite demonstrates exceptional performance in shielding electromagnetic interference within the X‐band frequency range, attaining a shielding effectiveness value of 46.3 dB with an extremely small thickness of 0.18 mm, with an excellent normalized specific shielding effectiveness (SSE/t) of 9166.7 dB·cm²·g⁻¹. In addition, the material possesses excellent electrothermal conversion performance; with an applied voltage of just 1 V, the device sustains a constant surface temperature of 65.4°C. These results highlight the potential of the FeMn@C/BC composite material for a wide range of applications, including architectural decoration and indoor heating systems.

Owing to the abundant EM coupling network and unique hierarchical porous structure provided by the bamboo‐derived carbon skeleton and magnetic FeMn carbon nanocages. FeMn@C/BC demonstrates exceptional performance in shielding electromagnetic interference within the X‐band frequency range, attaining a shielding effectiveness value of 46.3 dB with an extremely small thickness of 0.18 mm, with an excellent normalized specific shielding effectiveness (SSE/t) of 9166.7 dB·cm²·g⁻¹.

1. Introduction

With the rapid advancement of smart home technologies, electromagnetic (EM) radiation pollution has become a critical concern [1], as high‐frequency EM waves can interfere with electronic devices and pose risks to human health [2]. Consequently, developing high‐performance multifunctional electromagnetic interference (EMI) shielding materials has become an urgent requirement [3]. Beyond conventional shielding functionality [4], modern applications demand integrated thermal management capabilities (e.g., Joule heating) to achieve synergistic active temperature regulation and health protection [5].

As conventional EMI shielding materials, metal‐based materials (such as magnetite and ferrite) exhibit excellent electrical and thermal conductivity [6]. However, the practical application of these materials is severely limited by their high density, poor environmental stability, and difficulty in processing [7, 8]. Conductive polymer composites offer good processability; however, they suffer from inherent challenges, including insufficient SE, low mechanical strength, and susceptibility to deformation, while their synthesis processes often entail environmental pollution [9, 10, 11, 12]. Emerging carbon nanomaterials, including carbon nanotubes, graphene, and MXenes, have recently shown remarkable efficacy for EMI shielding applications. However, these are associated with complex preparation processes, high costs, and environmental issues, conflicting with sustainable development [13, 14, 15]. Biomass materials, owing to their natural three‐dimensional (3D) porous skeleton structure, renewability, and environmental friendliness, have shown distinct potential for use in EMI shielding applications. Their intrinsic skeleton serves as an ideal precursor for carbon‐based materials [16, 17, 18]. However, carbonized biomass‐derived materials that are simply carbonized often suffer from insufficient conductivity and poor structural stability. Researchers have attempted to address these limitations by developing composite strategies that embed conductive functional components, such as silver nanowires, nickel nanoparticles (NPs), polypyrrole, and polyaniline, into biomass skeletons. For instance, Dai et al. deposited nickel NPs onto wood surfaces through electroless plating to obtain nickel‐coated wood foam (NWF), achieving a maximum SE of 43.4 dB in the X‐band. The NWF also demonstrated highly efficient Joule heating, achieving a stable surface temperature of 49.6°C with an applied voltage of only 0.5 V [19]. Similarly, Hao et al. encapsulated nickel NPs in nitrogen‐doped carbon nanotubes grown in situ within carbonized wood, producing a carbon framework that exhibited an EMI SE of 83.24 dB in the X‐band, along with Joule heating performance reaching 90°C at a driving voltage of 2.2 V [20]. However, despite their excellent combined Joule heating and EMI shielding performance, such highly conductive composites rely predominantly on single‐mode conductive‐loss mechanisms, resulting in pronounced reflection characteristics that can cause secondary EM pollution [21, 22]. Therefore, designing biomass‐based composites through the synergistic effect of multiple loss mechanisms has become a crucial research direction, aiming to achieve high‐efficiency EMI shielding while integrating thermal management capabilities.

Metal–organic frameworks (MOFs) possess unique advantages in the field of EMI shielding owing to their high porosity, tunable topological structures, and excellent dielectric/magnetic properties of the materials from which they are derived [23, 24, 25]. Their periodically arranged metal nodes can be uniformly dispersed along the inner walls of the pore channels in biomass templates [26, 27, 28]. The complexation of unsaturated metal ions with hydroxyl and carboxyl functional groups on the biomass leads to the formation of a structurally stable composite. This enables the construction of an interconnected EM network with the biomass skeleton, which promotes EM loss and enhances the shielding performance of biomass‐based EMI shielding materials. For instance, Ma et al. embedded a conductive MOF (c‐MOF) onto a carbonized wood film (CWF) template derived from poplar wood, forming CWF/c‐MOF composites. The CWF/Ni‐HITP composite achieved a total EMI SE of 50.1 dB in the X‐band, while reaching a steady‐state surface temperature of 43°C under a driving voltage of 1 V [18]. Similarly, Rong et al. introduced MIL‐101 into wood through in situ growth, followed by hot‐pressing and carbonization to produce CW/Fe composites. These composites exhibited a thickness‐normalized specific shielding effectiveness (SSE/t) of 2020 dB·cm²·g⁻¹ in the X‐band, with a steady‐state surface temperature of 47°C at a low driving voltage (2 V) [29]. Nevertheless, among the MOF and biomass composites documented in existing literature, the MOFs used (such as ZIF‐67 and MIL‐101) are often converted into metal particles during high‐temperature carbonization, losing their original porous carbon skeleton structure. This tends to result in localized aggregation, which hinders multiple reflection paths of EM waves and significantly limits EM loss capabilities. Consequently, this inevitably leads to a compromise in overall performance, including mechanical properties, EMI shielding efficiency, or Joule heating performance. In addition, the biomass skeletons employed are primarily derived from wood with long growth cycles (20–50 years), making them unsuitable for large‐scale industrial production. Therefore, selecting appropriate materials and designing rational structures are crucial for developing high‐performance, biomass‐based EMI shielding composites. Magnetic Prussian blue analogs (PBAs) are highly promising magnetic porous nanomaterials comprising transition metal ions (Fe^{3 +}/Mn^{2 +}, etc.) and cyanide ligands (CN⁻) coordinated to form a framework [30, 31, 32, 33, 34]. Compared with traditional magnetic particle materials, magnetic PBAs possess unique 3D open pore structures, high specific surface area, tunable component ratios, and excellent thermal stability. They exhibit significant magnetic capabilities and retain their original structure after high‐temperature carbonization. This characteristic is conducive to improving the overall EMI SE of biomass‐based composites. Bamboo, with its fast growth rate (3–5 years), well‐developed pore structure, and high carbon content, is considered an ideal biomass alternative to wood. The abundant hydroxyl groups on the surface of bamboo fibers provide active sites for anchoring magnetic PBAs, while the unique longitudinal vascular bundle structure offers ample space for PBA anchoring, facilitating the in situ growth of magnetic PBAs [35, 36, 37]. Moreover, the highly aligned hierarchical pore channels form a continuous 3D conductive network after carbonization [38, 39]. These advantages, combined with the stable framework of magnetic PBAs, synergistically enhance multiple reflections and interface losses of EM waves, holding great promise for achieving biomass‐based composites with high EMI SE and thermal management capabilities [40, 41]. Nevertheless, research on the synergistic construction of high‐performance, multifunctional bamboo‐based EMI shielding materials using bamboo templates and magnetic PBAs remains scarce.

In this study, pretreated bamboo provides anchoring sites for the in situ growth of FeMn PBA, enabling uniform loading of FeMn PBA onto the bamboo skeleton to form FeMn PBA/B composites [42, 43, 44]. Subsequently, FeMn PBA/B was subjected to hot‐pressing and carbonization to obtain a green, low‐cost, and lightweight FeMn@C/BC material. The results demonstrate that the hierarchical structure of the bamboo skeleton enhances reflection loss of EM waves, while the carbon‐derived FeMn carbon nanocages further extend the scattering path of EM waves within the material and simultaneously improve the overall magnetic loss capability. The dense hierarchical network structure of FeMn@C/BC facilitates electron transport, thereby enhancing transmission loss of EM waves. The synergistic effect between the compact layered conductive network of FeMn@C/BC and the EM coupling network formed by uniformly distributed magnetic FeMn carbon nanocages contributes to outstanding EMI shielding performance, attaining a SE of 46.3 dB at a low thickness (0.18 mm) and an electrical conductivity of 37.3 S cm⁻¹. Moreover, FeMn@C/BC‐900 exhibits a pronounced Joule heating effect, reaching a surface temperature of 65.4°C instantaneously under a driving voltage of 1 V. In addition, the material combines favorable hydrophobicity and thermal stability, further expanding its potential for application in harsh environments, such as high humidity or hot environments. These combined properties endow the FeMn@C/BC material with remarkable multifunctionality, laying a foundation for its application in integrated systems for EM protection and intelligent thermal management.

2. Results and Discussion

The preparation method of the FeMn@C/BC composite is illustrated in Figure 1a. First, bamboo slices were subjected to delignification and oxidation treatments to obtain a bamboo cellulose skeleton structure [45, 46, 47]. Subsequently, the bamboo skeleton was immersed in a metal solution to allow in situ growth of FeMn PBA, followed by hot‐pressing and carbonization to form a lightweight, multifunctional FeMn@C/BC composite. SEM images (Figure 2a–c) reveal the ordered honeycomb‐like 3D porous structure inside the bamboo, with pore sizes ranging from ∼15 to ∼50 µm. As a natural biomass material, bamboo primarily comprises lignin (∼25%), cellulose (∼40%), and hemicellulose (∼20%). Delignification and TEMPO oxidation significantly alter both the compositional content and internal pore structure of bamboo. The delignification treatment increased the cellulose content from 38.5% to 56.2%, while the lignin content decreased from 25.6% to 7.7% (Figure S1a). Following partial lignin removal, the internal pore structure of bamboo became notably more open (Figure S2a,c). Subsequent TEMPO oxidation further exposed additional pores within the bamboo (Figure S2b,d), providing favorable conditions for enhanced loading of FeMn PBA NPs.

FIGURE 1.

(a) Schematics of the preparation of FeMn@C/BC composite. (b) Functional characteristics of the FeMn@C/BC composite.

FIGURE 2.

SEM images of (a) the cross‐section, (b) the radial section, and (c) the tangential section of natural bamboo; (d) SEM image of the cross‐section of FeMn PBA/B; (e, f) Surface SEM images of FeMn PBA/B; (g) EDS mapping of the FeMn PBA/B surface; (h) SEM images of the cross‐section of FeMn@C/BC‐900; (i, j) Surface SEM images of FeMn@C/BC‐900; (k) EDS mapping of the surface of FeMn@C/BC‐900; (l) Low‐resolution transmission electron microscopy (HRTEM) images of FeMn carbon nanocages; (m, n) High‐resolution transmission electron microscopy (HRTEM) images of FeMn carbon nanocages; and (o) Photograph of FeMn@C/BC‐900.

Delignification time and TEMPO oxidation time are critical parameters for preserving the structural integrity of the bamboo‐derived cellulose skeleton. The effects of different delignification times (0, 2, 4, 6 h) and subsequent TEMPO oxidation times (0, 2, 4, 6 h) on the carboxyl group content and structural integrity of the resulting samples (denoted TB‐0, TB‐2, TB‐4, TB‐6) were systematically investigated. The results indicate that the carboxyl group content of the pretreated bamboo increased from 0 to 0.82 mmol/g (Figure S2) with prolonged TEMPO oxidation time. The elevated carboxyl group content provides additional coordination sites for metal ions (Fe^{3 +}), thereby promoting the in situ growth of FeMn PBA. However, extended oxidation time also exacerbates damage to the bamboo skeleton (Figure S3). A delignification time of 4 h and a TEMPO oxidation time of 4 h were ultimately selected to balance the structural integrity and carboxyl group content (Figure S4). This yielded a bamboo‐derived cellulose skeleton as the optimal material for subsequent experiments. The morphology of FeMn PBA synthesized under varying growth times (from 2 to 48 h) was examined using TEM. The synthesized FeMn PBA clearly exhibited a cubic structure with an average size of ∼200 nm (Figure S5), featuring smooth surfaces and good crystalline uniformity. The cross‐sectional SEM image of FeMn PBA/bamboo shown in Figure 2d demonstrates that a dense layered network structure is formed within the bamboo after hot‐pressing. Figure 2e shows that FeMn PBA (Figure 2f) is uniformly distributed on the bamboo skeleton. The EDS analysis presented in Figure 2g shows a homogeneous distribution of C, O, Fe, and Mn elements, with Fe and Mn originating from FeMn PBA. This compact layered network structure is well retained even after carbonization (Figure 2h). From Figure 2i, it can be observed that the FeMn carbon nanocages (Figure 2j) obtained after calcination largely retain their original morphology. As the carbonization temperature increases further, the internal structure of the FeMn carbon nanocages partially collapses (Figure S7). Elements C, O, Fe, and Mn are uniformly distributed on FeMn@C/BC (Figure 2k). Low‐magnification TEM images (Figure 2l) display numerous oval‐shaped features dispersed throughout a gray matrix, exhibiting pronounced dark‐field contrast. This phenomenon is due to the graphitized carbon inhibiting the agglomeration and detachment of nanocages, allowing the FeMn nanocages to be evenly and stably distributed within the carbon matrix. Further observation of the carbon nanocages derived from FeMn PBA after carbonization using high‐resolution TEM reveals lattice fringes with spacings of 0.25 nm, corresponding to the Fe (110) plane, as well as lattice fringes of 0.35 nm corresponding to the C (002) plane wrapping the surface of the FeMn carbon nanocages (Figure 2m). In the SAED pattern presented in Figure 2n, three concentric diffraction rings are distinctly visible and are indexed to the (220), (420), and (440) crystalline planes. FeMn@C/BC is gray‐black and lightweight, as shown in Figure 2o.

FTIR spectral analysis (Figure 3a) of FeMn PBA shows peaks at ∼460, ∼594, and ∼2074 cm⁻¹, which correspond to Mn─N, Fe─C, and C≡N bonds, respectively. For pretreated bamboo, the characteristic peak at 1616 cm⁻¹ is attributed to the asymmetric stretching vibration of the carboxyl groups (─COOH) formed through the selective oxidation of the C6 hydroxyl groups (─OH) on cellulose chains during TEMPO‐mediated oxidation. This shift is caused by the formation of hydrogen bonds and ionic interactions between FeMn PBA and cellulose in the composite. Furthermore, the characteristic carboxyl (─COOH) peak in both FeMn PBA/B and pretreated bamboo exhibits a shift, indicating changes in the interaction between the carboxyl groups and Fe³⁺ ions. These observations confirm the successful loading of FeMn PBA onto the bamboo skeleton.

FIGURE 3.

(a) FTIR spectra of bamboo, pretreated bamboo, FeMn PBA/B, FeMn PBA, and FeMn@C/BC; (b) X‐ray diffraction (XRD) patterns of FeMn PBA, bamboo, pretreated bamboo, FeMn PBA/B, and FeMn@C/BC‐900; (c) Raman spectra of BC‐900, FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000; (d) X‐ray photoelectron spectroscopy (XPS) survey spectrum of FeMn@C/BC‐900; (e) Electrical conductivity of BC‐900, FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000; (f) Hysteresis loops of BC‐900, FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000; (g) N₂ adsorption–desorption isotherms of FeMn@C/BC‐900 and FeMn PBA; (h) Pore size distribution of FeMn@C/BC‐900 and FeMn PBA; and (i) Density and thickness of the samples BC‐900, FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000.

The crystalline phases of the various samples were characterized by x‐ray diffraction, as shown in Figure 3b. For the bamboo sample, peaks at 16.6° and 22.5° correspond to the (101) and (002) crystal planes of cellulose in bamboo, respectively. The presence of peaks at the same positions in the pretreated bamboo indicates that the crystalline structure of cellulose remained unchanged during delignification. According to JCPDS No. 51–1896, the diffraction peaks of FeMn PBA at 17.5°, 24.9°, 29.4°, 35.5°, 39.8°, and 51° correspond to the (200), (220), (222), (400), (420), and (440) crystal planes, respectively, confirming the successful synthesis of FeMn PBA with the expected crystallographic characteristics. The XRD pattern of FeMn PBA/B demonstrates that FeMn PBA was successfully loaded onto the bamboo skeleton. The crystal structure of FeMn@C/BC‐900 was further analyzed. The x‐ray diffraction pattern reveals a peak at 26.5°, which is assigned to the (002) crystallographic plane of graphitic carbon. This observation confirms the formation of a graphite‐like phase within the FeMn@C/BC composite following pyrolysis at this specific temperature. Characteristic peaks at 44.7° and 65.0° correspond to crystalline Fe⁰ and Mn⁰, respectively, indicating the reduction of Fe and Mn during carbonization. In the Raman spectra (Figure 3c), two characteristic peaks of carbon‐based materials can be observed: the D band at ∼1350 cm⁻¹, which is associated with disordered carbon or defects (sp³ C─X bonds, where X can be O, H, etc.), and the G band at ∼1580 cm⁻¹, attributed to graphitic carbon with sp² C═C bonds. As the carbonization temperature increases, the ID/IG ratio of the FeMn@C/BC series gradually decreases from 1.03 to 0.94, indicating higher graphitization at elevated temperatures. Compared with BC‐900, FeMn@C/BC‐900 exhibits a lower ID/IG value, which can be attributed to the catalytic effect of metallic Fe and Mn during carbonization, promoting the graphitization of the bamboo‐derived carbon. As shown in Figure 3d, XPS characterization was performed on FeMn@C/BC‐900 to further investigate its composition and surface chemical states. In the O 1s spectrum (Figure S8a), the peak centered at 530.4 eV is indicative of lattice oxygen from metal oxides. Peaks observed at 531.3 and 532.7 eV are ascribed to oxygen‐based functional groups and defective oxygen sites with low coordination numbers, respectively. In the C 1s spectrum (Figure S8b), three peaks are observed at 284.7, 286.1, and 288.9 eV, corresponding to C─C/C═C, C─O, and C═O bonds, respectively. The highest proportion of C─C/C═C indicates that aromatic carbon (sp² hybridized) is the dominant structural component of the material. This is due to the thermal instability of oxygen‐bound carbon species during high‐temperature carbonization, as carbon atoms bonded to oxygen are generally less stable than aromatic carbon, resulting in a significant reduction of oxygen‐containing carbon species. The N 1s spectrum (Figure S8c) shows the presence of pyridinic‐N (398.3 eV), pyrrolic‐N (399.2 eV), and graphitic‐N (400.6 eV), confirming the successful incorporation of nitrogen into the graphitic carbon structure. The Fe 2p spectrum (Figure S8d) displays two characteristic peaks at 710.8 eV for Fe 2p_3/2 and 724.3 eV for Fe 2p_1/2. In addition, the Fe 2p spectrum indicates the presence of Fe²⁺ (711.1 and 724.3 eV), Fe³⁺ (713.9 and 726.1 eV), and Fe⁰ (707.5 and 720.6 eV), indicating redox reactions of Fe ions during calcination that led to the formation of magnetic metallic iron. Figure S8e presents the high‐resolution Mn 2p XPS spectrum. Two primary peaks are detected at 641.4 and 653.4 eV, which correspond to the Mn 2p_3/2 and Mn 2p_1/2 states, respectively. Additionally, an observed satellite feature positioned at 645.3 eV is consistent with the presence of Mn²⁺. The Mn spectrum (Figure S8f) further indicates the presence of Mn⁰ (639 eV), Mn²⁺ (640.8 eV), Mn³⁺ (641.8 eV), and Mn⁴⁺ (642.8 eV), confirming the formation of magnetic metallic manganese in FeMn@C/BC‐900 during pyrolysis [48, 49, 50]. The presence of FeMn carbon nanocages contributes to the favorable magnetic properties of the FeMn@C/BC composite. The conductivity test results (Figure 3e) show that the electrical conductivity of the FeMn@C/BC materials increases with the carbonization temperature. The conductivities of FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000 are 28.2, 37.3, and 51.5, respectively. This trend can be attributed to the enhanced graphitization degree of the material at higher carbonization temperatures. Notably, the conductivity of FeMn@C/BC‐900 (37.3 S cm⁻¹) is much higher than that of the control sample BC‐900 (31 S cm⁻¹), which does not contain metal loading. This improvement benefits from the catalytic effects of metallic Fe and Mn, which promote the graphitization of the bamboo‐derived carbon skeleton. Enhanced electrical conductivity plays a crucial role in achieving dielectric loss of EM waves by the material. The saturation magnetization (Ms) values for FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000 are 5.06, 5.98, and 6.51 emu/g, respectively (Figure 3f and Figure S9). The increase in Ms with higher carbonization temperatures can be attributed to the enhanced crystallinity of the magnetic metal NPs and stimulated growth of these particles at elevated temperatures, leading to a more ordered arrangement of magnetic domains and consequently higher Ms values [51]. Furthermore, FeMn PBA exhibits a typical type‐I (S‐shaped) N₂ adsorption–desorption isotherm (Figure 3g). Its pore size distribution (Figure 3h) indicates the presence of characteristic micropores in FeMn PBA. Compared with the S‐shaped N₂ adsorption–desorption isotherm of FeMn@C/BC‐900, the corresponding pore size distribution reveals abundant micropores, which likely originate from the micropores retained in the FeMn carbon nanocages after carbonization. The porous structure helps reduce the density of the composite material. As shown in Figure 3i, BC‐900 exhibits a relatively low density of 0.28 g/cm³. The FeMn carbon nanocages derived from FeMn PBA slightly increase the density of FeMn@C/BC‐900 (0.28 g/cm³). Meanwhile, as the carbonization temperature rises, the densities of FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000 gradually decrease from 0.31 to 0.25 g/cm³, which is significantly lower than the density of natural bamboo (0.47 g/cm³). This phenomenon occurs because higher carbonization temperatures accelerate the release of volatile components from the bamboo, which is of great significance for achieving lightweight EMI shielding materials.

Within the X‐band, the S‐parameters measured using a vector network analyzer were used to calculate the total EMI SE of BC‐900, FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000, as shown in Figure 4a. The average EMI SE of pure BC‐900 is only 31.7 dB, while that of FeMn@C/BC‐900 reaches 46.3 dB, indicating that the addition of FeMn carbon nanocages provides additional EM wave loss and improves the absorption‐based shielding performance. As the carbonization temperature increases, the SE values of the FeMn@C/BC series exhibit a gradual upward trend. The observed performance enhancement is presumably due to the increased electrical conductivity and magnetic properties imparted to the composites at elevated temperatures. These improvements concurrently strengthen both magnetic and dielectric dissipation mechanisms, leading to a substantial rise in EMI SE. As verified by impedance matching and attenuation constants (Figure S10), the composite material‐900 demonstrates excellent electromagnetic screen performance. The absorption efficiency (SEA), reflection efficiency (SER), and total shielding efficiency (SET) are critical parameters that reflect the ability of a material to shield EM waves. As illustrated in Figure 4b, for FeMn@C/BC‐900, the SER, SEA, and SET values are 8.5, 37.7, and 46.3 dB, respectively. The SET is predominantly contributed to by SEA, with an SEA/SET ratio as high as 81.56%. For FeMn@C/BC‐1000, the corresponding values are 10 dB (SER), 41.3 dB (SEA), and 51.3 dB (SET), with an SEA/SET ratio of 80.51%. These results indicate that FeMn@C/BC‐900 can more effectively reduce secondary EM pollution caused by wave reflection. The absorption coefficient (A), reflection coefficient (R), and transmission coefficient (T) were further used to evaluate the reflection, absorption, and transmission behaviors of EM waves in the composites. As shown in Figure 4c, the A value of FeMn@C/BC‐900 is nearly twice that of BC‐900. This enhancement can be attributed to the presence of FeMn carbon nanocages, which increase the opportunities for multiple internal reflections of EM waves entering the material, thereby improving the multi‐reflection loss and raising the A value [52]. Moreover, the A value of FeMn@C/BC‐900 is higher than those of FeMn@C/BC‐800 and FeMn@C/BC‐1000, indicating that FeMn@C/BC‐900 can more effectively suppress secondary EM pollution resulting from high EM wave reflection during operation of the shielding material. Therefore, this study concludes that FeMn@C/BC‐900 exhibits superior EMI shielding performance. In addition, the total EMI SE of the FeMn@C/BC composite reaches 99.999% at a low thickness (0.18 mm) (Figure 4e). This performance confirms its high efficiency in attenuating most electromagnetic radiation across the X‐band. The thickness of an EMI shielding material also influences its SE; generally, thicker materials exhibit higher EMI shielding performance. Therefore, the EM SE of FeMn@C/BC‐900 was further investigated with different thicknesses (Figure 4d). Using a vector network analyzer, the total shielding effectiveness (SET) was measured for samples with thicknesses of 0.1, 0.18, and 0.26 mm, yielding values of 37.2 , 46.3 , and 59.6 dB, respectively. The outstanding EMI shielding performance of FeMn@C/BC‐900 enables it to maintain considerable SE even when the thickness is reduced to as low as 0.1 mm.

FIGURE 4.

(a) EMI SE of BC‐900 and the FeMn@C/BC composite series in the X‐band; (b) EMI shielding by reflection (SER), absorption (SEA), and total shielding effectiveness (SET) of BC‐900 and the FeMn@C/BC series in the 8.2–12.4 GHz band; (c) T, A, and R values of BC‐900 and FeMn@C/BC composites in the 8.2–12.4 GHz band; (d) EMI SE of FeMn@C/BC‐900 with different thicknesses in the 8.2–12.4 GHz band; (e) Total EMI shielding effectiveness (SET) and shielding efficiency results of BC‐900, FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000; (f) EMI SE of FeMn@C/BC‐900 composites prepared with different FeMn PBA growth times, measured in the X‐band; (g) Schematic illustrating the EMI shielding mechanism of the FeMn@C/BC‐900 composite; (h) Comparison of the thickness‐normalized specific shielding effectiveness (SSE/t) with other biomass‐based EMI shielding materials (Table S1).

The loading amount of FeMn PBA also influences the SE of the EMI shielding material. Compared with BC‐900, FeMn nanocages formed after carbonization of FeMn PBAs construct a more abundant EM coupling network, thereby enhancing dielectric loss, while the magnetic FeMn carbon nanocages enhance magnetic loss of EM waves [53]. However, excessive FeMn PBA loading can lead to impedance mismatch after carbonization. Consequently, the FeMn@C/BC‐900 with a FeMn PBA growth time of 48 h exhibits lower SE in the X‐band than that with a growth time of 24 h. As the growth time of FeMn PBA increases, the SE of FeMn@C/BC‐900 in the X‐band first rises and then declines (Figure 4f). Therefore, the optimal FeMn PBA growth time was determined to be 24 h. Figure 4g illustrates the EMI shielding mechanism of FeMn@C/BC‐900. Following in situ growth of FeMn PBA, the bamboo skeleton undergoes hot‐pressing and carbonization, transforming its porous structure into a layered conductive network, which provides an effective pathway for EM wave transmission. When incident EM waves reach the surface of the FeMn@C/BC composite, merely a minor component of the incident EM wave is reflected back into the atmospheric environment, with the predominant fraction propagating into the carbon‐based framework originating from bamboo. Inside the composite, EM energy is dissipated through Ohmic loss, caused by the highly conductive carbon skeleton, and effectively attenuated through magnetic loss provided by the magnetic FeMn carbon nanocages. This synergistic effect helps mitigate impedance mismatch and enhances the absorption efficiency of EM waves. Simultaneously, the hierarchical pores of the bamboo skeleton and the unique cubic structure of FeMn carbon nanocages extend the propagation path of EM waves, promoting multiple internal reflections and scattering within the composite, thereby increasing internal reflection loss. The cooperative action of these multiple loss mechanisms endows the material with outstanding EMI shielding performance [54, 55]. The thickness‐normalized specific SE (SSE/t) of FeMn@C/BC‐900 was calculated using the formula provided in the Supporting Information to enable a fair comparison of shielding performance among EMI shielding materials. FeMn@C/BC‐900 exhibits an exceptionally high SSE/t value (9166.7 dB·cm²·g⁻¹), achieving a high shielding efficiency of 99.99% (calculated using the formula given in the Supporting Information) at a low density of 0.28 g/cm³ and a mere thickness of 0.18 mm. This performance surpasses that of most reported biomass‐based EMI shielding materials (Figure 4h), making it a highly competitive candidate for next‐generation EMI shielding materials.

Owing to its excellent electrical conductivity, FeMn@C/BC‐900 allows current to pass through when energized. Under the driving electric field, carbon atoms undergo irregular motion, colliding and rubbing against each other, thereby generating heat through the conversion of electrical energy into thermal energy [56]. This enables efficient “electrical‐to‐thermal” energy conversion, resulting in notable Joule heating performance of FeMn@C/BC‐900. When FeMn@C/BC‐900 was placed on a model house, as shown in Figure 5a, and connected to a power supply, it rapidly heated up to 65.4°C under a driving voltage of 1 V (Figure 5b). The Joule heating performance was further tested under driving voltages of 0.5, 1, 1.5, and 2 V. The results indicate that FeMn@C/BC‐900 maintains good thermal stability under different voltages (Figure 5c). At 0.5 V, the steady‐state surface temperature reached only 39.5°C, while higher driving voltages led to an increase in the steady‐state surface temperature, rising from 39.5°C up to 133°C. Under a driving voltage of only 2 V, the surface temperature of FeMn@C/BC‐900 instantly reached 133°C (Figure 5 d) and remained stable thereafter. Even after 10、20, and 100 heating–cooling cycles (Figure S11), the material continued to exhibit excellent stability in steady‐state surface temperature. A linear fitting between the steady‐state surface temperature (°C) and the square of the driving voltage (U²) yielded a high coefficient of determination R² = 0.98 (Figure 5e), indicating a strong linear relationship consistent with Joule's law. As shown in Figure 5f, FeMn@C/BC‐900 demonstrates excellent voltage‐responsiveness and electro‐thermal conversion capability when the driving voltage was increased stepwise from 0.5 to 2 V and then decreased back to 0.5 V. Figure 5g shows that the steady‐state surface temperature of FeMn@C/BC‐900 remained stable for >7200 s under driving voltages of 1 and 2 V, confirming its reliability. Finally, a comparison with other reported materials in terms of Joule heating performance (Figure 5h) reveals that FeMn@C/BC‐900 can achieve higher steady‐state surface temperatures at lower driving voltages. These outstanding Joule heating properties indicate that FeMn@C/BC‐900 holds great potential as a novel, efficient electric heater for indoor applications.

FIGURE 5.

(a, b) Model house covered with FeMn@C/BC‐900 and corresponding infrared thermal images demonstrating its Joule heating performance; (c) Evolution of the surface saturation temperature of FeMn@C/BC‐900 over time under driving voltages of 0.5, 1, 1.5, and 2 V; (d) Cycling stability of FeMn@C/BC‐900 under a driving voltage of 2 V over 1, 10, and 20 cycles; (e) Experimental data and linear fitting of the surface temperature vs. U²; (f) Surface temperature of FeMn@C/BC‐900 under stepwise voltage changes; (g) Surface temperature stability of FeMn@C/BC‐900 under driving voltages of 1 and 2 V; (h) Comparison of the Joule heating performance of FeMn@C/BC‐900 with other reported materials (Table S2).

Water resistance analysis is a crucial factor in evaluating the suitability of materials for humid environments. Therefore, the contact angles of bamboo, FeMn PBA/B, and FeMn@C/BC‐900 were measured. The surface of bamboo is rich in hydrophilic groups, causing water droplets to spread immediately upon contact. Upon loading with MOFs, the contact angle of FeMn PBA/B increased from 34.58° to 51.63°. The carbonized FeMn@C/BC‐900 surface became hydrophobic, and the change in its contact angle within 0–30 s is shown in Figure 6a. The hydrophobicity of FeMn@C/BC‐900 can be attributed to two main factors: first, the consumption of oxygen‐containing functional groups during carbonization; second, the hot‐pressing and carbonization processes resulted in a smoother surface with reduced roughness and fewer pores, which collectively enhance the hydrophobic properties of FeMn@C/BC‐900. After being stored at 90% relative humidity for 100 h, the electromagnetic shielding performance of FeMn@C/BC‐900 is almost unaffected (Figure S12).

FIGURE 6.

(a) SEM images (surface) and contact angles of bamboo, FeMn PBA/B, and FeMn@C/BC‐900; (b) Thermal conductivity of bamboo, FeMn PBA/B, and FeMn@C/BC‐900; (c) TG curves of BC‐900, FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000; (d) Tensile strength of BC‐900, FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000; and (e) Sample image and dimensions of large‐scale FeMn@C/BC.

The thermal conductivities (TCs) of BC‐900, FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000 were measured. The results, as shown in Figure 6b, indicate that the TC of the FeMn@C/BC composites increases with rising carbonization temperature. In particular, the TC of FeMn@C/BC‐900 reaches 1.3 W·m⁻¹·K⁻¹, which facilitates uniform heat distribution within the composite and helps prevent localized heat accumulation during operation, thereby reducing the risk of fire or equipment damage. TGA was performed under a nitrogen atmosphere to evaluate the thermal degradation behavior of bamboo, FeMn PBA/B, and FeMn@C/BC‐900. As shown in Figure 6c, FeMn@C/BC‐900 retains 67.92% of its mass at 700°C, which is significantly higher than the residual masses of bamboo (15.75%) and FeMn PBA/B (38.08%). The derivative thermogravimetry (DTG) data (Figure S13) confirm that FeMn@C/BC‐900 exhibits the slowest maximum degradation rate, demonstrating enhanced thermal stability after carbonization. Furthermore, good mechanical strength is a key prerequisite for ensuring the reliability of composite materials in practical applications. The tensile strength and flexural strength of the FeMn@C/BC composites and BC‐900 were evaluated (Figure 6d and Figure  S14). With increasing carbonization temperature, both the tensile and flexural strengths of FeMn@C/BC slightly decrease. Nevertheless, FeMn@C/BC‐900 exhibits higher tensile and flexural strengths than those of BC‐900. It can withstand a weight of 500 g without breaking, and also maintain excellent mechanical stability (Figure S15). This improvement is primarily attributed to the strong metal‐coordination interactions between the metal ions (Fe³⁺ and Mn²⁺) in the in situ formed FeMn PBA and oxygen‐containing functional groups (─C═O, ─OH) on the surface of the bamboo cellulose skeleton, which enhance interfacial bonding. Meanwhile, the TEMPO oxidation treatment provides additional active sites for the in situ growth of FeMn PBA, promoting its uniform dispersion in the composite and avoiding stress concentration, and strengthening the crosslinking network among bamboo fibers during hot‐pressing, thereby improving the overall mechanical properties of the FeMn@C/BC composite. These results indicate potential applications for FeMn@C/BC‐900 composites in fields such as interior decoration. A large‐scale FeMn@C/BC‐900 sample with dimensions of 188.2 mm × 41.8 mm was successfully prepared to better simulate real‐world application scenarios (Figure 6e).

A Tesla coil (its circuit diagram is shown in Figure S16a) was used as a high‐frequency EM wave source to visualize the EMI shielding effect of the FeMn@C/BC composite by illuminating a light bulb, as illustrated in Figure 7a. As the light bulb was moved nearer to the active Tesla coil (about 5 cm) in the absence of any intervening material, it became illuminated. Subsequently, a plain bamboo slice was placed between the bulb and the Tesla coil; The bulb continued to emit light due to the absence of EMI shielding properties in the pristine bamboo sample. Finally, when the FeMn@C/BC composite was placed between the bulb and the Tesla coil, the bulb instantly went out. This occurs because the presence of the FeMn@C/BC composite blocks the transmission path of the EM waves emitted by the Tesla coil, preventing the bulb from receiving the EM energy. Similarly, following the setup shown in Figure S16b,c, plain bamboo and FeMn@C/BC composite were separately placed between a smartphone and a wireless charger. When the plain bamboo slice was positioned, the indicator light of the wireless charger remained on, and the phone charged normally, indicating that the EM waves emitted by the charger passed through the bamboo and were received by the phone. By contrast, when the FeMn@C/BC composite was placed between the phone and the charger, the indicator light turned off, and the phone failed to charge (Figure 7b). This result confirms that the FeMn@C/BC composite effectively blocks the energy transmission path between the wireless charger and the phone, preventing normal charging.

FIGURE 7.

(a) Shielding of high‐energy electromagnetic (EM) wave transmission by the FeMn@C/BC composite; (b) Shielding of low‐energy EM wave transmission by the FeMn@C/BC composite; (c) FeMn@C/BC composite material for shielding high‐energy electromagnetic waves emitted by indoor routers; (d) Electric field and magnetic field measurements inside a wooden cabin before and after the installation of a composite roof made of FeMn@C/BC; (e) Schematic of the shielding demonstration model and the corresponding EM radiation intensity cloud map on the X–Y plane; and (f) Shielding of high‐energy EM wave transmission by the FeMn@C/BC composite in a high‐humidity environment.

Subsequently, real‐life EMI shielding scenarios were simulated from a practical application perspective, as shown in Figure 7c. Near an operating indoor router (10 cm), an EM radiation detector measured radiation levels of ∼56 V/m and 0.14 µT at a distance of 3 cm from the router. When the FeMn@C/BC composite was placed between the detector probe and the router antenna, the radiation levels instantly dropped below the safety threshold (10 V/m, 0.01 µT), demonstrating that the composite can effectively shield EM waves in practical applications. The practical value of the FeMn@C/BC composite was further illustrated using the material as a roof to shield a house from external EM waves (Figure 7d). A Tesla coil was employed to simulate high‐power (The Tesla coil was placed 10 cm outside the house), high‐energy EM wave sources, such as transmission towers or signal towers. An EM radiation detector directly exposed to this environment recorded strong EM fields (629 V/m, 0.52 µT). While a conventional building roof offered little protection against such EM waves, a roof made of the FeMn@C/BC composite effectively blocked the external EM radiation, reducing the measured values to 4 V/m and 0.02 µT. For a more detailed analysis of the shielding performance of the material, a simple 3D model of the FeMn@C/BC composite was constructed using the COMSOL software. EM waves were set to incident vertically from one side of the shielding model to the other in a vacuum simulation environment. The experimentally measured EM parameters of FeMn@C/BC were input into the model to simulate the distribution of electric and magnetic field intensities within the computational domain under the shielding effect of the composite [57]. As clearly shown in Figure 7e, the electric field signal is almost completely blocked in the space on the negative y‐axis side of the FeMn@C/BC composite, and the magnetic field is also effectively suppressed, further confirming that the FeMn@C/BC composite can efficiently shield EM waves. In addition, a Tesla coil was used to simulate EM radiation emitted during the operation of an electric baby crib (Figure S16d–f). When the Tesla coil was activated, a small light bulb placed on the crib (unprotected by any shielding material) was lit by the high‐energy EM waves. Upon placing the FeMn@C/BC composite on the crib, the transmission path of the high‐energy EM waves was interrupted, and the bulb extinguished due to a lack of EM energy received. This indicates the significant potential of FeMn@C/BC composites for providing protection against EM radiation. Moreover, since outdoor building materials are required to withstand harsh natural conditions (such as continuous rainfall), it is crucial to verify the stability of FeMn@C/BC composites under demanding operational environments. Therefore, this study simulated its performance in a humid setting (temperature: 22 ± 2°C, relative humidity: 90 ± 2%) (Figure 7f). The results show that even under damp conditions, the material maintains excellent EMI shielding performance (0 V/m, 0 µT). Collectively, these findings indicate that FeMn@C/BC composites have broad application prospects in fields such as construction materials and smart home devices.

3. Conclusions

In summary, a FeMn@C/BC composite possessing both magnetic and conductive properties was successfully fabricated for EMI shielding applications through the in situ growth of FeMn PBA on a pretreated bamboo‐derived skeleton, followed by hot‐pressing and carbonization. The composite integrates multiple energy‐dissipation mechanisms, including dielectric, magnetic, and multi‐reflection losses. The formation of magnetic FeMn carbon nanocages during carbonization improves the electrical conductivity of the FeMn@C/BC composite and facilitates multiple internal reflections of incident electromagnetic waves. These effects collectively enhance the electromagnetic interference shielding effectiveness. As a result, the composite exhibits an outstanding EMI SE of 46.3 dB at a small thickness of 0.18 mm, together with a high SSE/t of 9166.7 dB·cm²·g⁻¹. This performance exceeds that of numerous previously reported biomass‐based EMI shielding materials and demonstrates notable superiority among bamboo‐derived shielding systems. Furthermore, the FeMn@C/BC composite exhibits outstanding Joule heating performance, confirmed by reaching a steady‐state surface temperature of 65.4°C under a driving voltage of only 1 V. This study provides an innovative strategy for designing high‐performance EMI shielding materials. Combined with its favorable hydrophobicity and mechanical properties, FeMn@C/BC holds considerable application potential and broad prospects in fields such as construction materials and smart home devices.

4. Experimental Section

4.1. Materials

Naturally grown Moso bamboo (4 – 5‐year‐old, sourced from Fujian, China) was processed into thin slices measuring 40 mm × 20 mm × 0.7 mm (length × width × thickness). Sodium chlorite (NaClO₂, 80%), 2,2,6,6‐tetramethyl‐1‐piperidinyloxy (TEMPO, 98%), sodium hypochlorite (NaClO, 14%), sodium bromide (NaBr, 99%), manganese sulfate monohydrate (MnSO₄·H₂O, 99.99%), potassium ferricyanide (K₃ [Fe(CN)₆], 99%), trisodium citrate dihydrate (C₆H₅O₇Na₃·2H₂O, 98%), and absolute ethanol (99.5%) were purchased from Macklin Inc. Deionized water was supplied by the water purification system in the laboratory.

4.2. Preparation of Pretreated Bamboo

First, bamboo slices were fully immersed in 500 mL of an aqueous sodium chlorite solution with a mass fraction of 2%. The pH of the solution was adjusted to 5.0 using acetic acid, and the mixture was continuously stirred at 80°C for 4 h to partially remove lignin and hemicellulose. Subsequently, the delignified bamboo slices (3 g) were placed in a sodium carbonate–sodium bicarbonate buffer solution (500 mL, pH 10.5) containing TEMPO (0.048 g) and NaBr (0.9 g). A 10% NaClO solution (15 mL) was slowly added dropwise under continuous stirring to initiate the oxidation reaction. Anhydrous ethanol was added to quench the reaction after 2, 4, and 6 h. The resulting products were washed with deionized water until the washings were neutral, followed by freeze drying for 48 h. The final samples obtained were labeled as PB‐2, PB‐4, and PB‐6, respectively.

4.3. Preparation of FeMn PBA/Bamboo Composite Materials

First, 1.8 g K₃[Fe(CN)₆] was completely dissolved in 180 mL deionized water to form solution A. The pretreated bamboo slices were then immersed in solution A and incubated for 3 h to ensure sufficient complexation between the carboxyl groups (─COOH) on the bamboo surface and Fe³⁺ ions. Meanwhile, solution B was prepared by completely dissolving 0.3 g MnSO₄·H₂O and 0.66 g C₆H₅O₇Na₃ in 120 mL deionized water. Solution B was then added to solution A under continuous stirring for 120 s. The resulting mixture was allowed to stand at room temperature (20°C–25°C) for 2, 12, 24, and 48 h. At the end of each period, the samples were washed thoroughly with deionized water and labeled as FeMn PBA/bamboo‐1, FeMn PBA/bamboo‐2, FeMn PBA/bamboo‐3, and FeMn PBA/bamboo‐4, respectively. Subsequently, the obtained samples were placed in a hot press and compressed at 80°C and 3 MPa for 8 h to yield stable FeMn PBA/Bamboo composite materials. In parallel, FeMn PBA was also successfully synthesized in the absence of pretreated bamboo as a control.

4.4. Preparation of FeMn@C/BC Composites

The prepared FeMn PBA/bamboo samples were placed in the upstream section of a quartz tube within a tube furnace. Under a nitrogen atmosphere, the temperature was raised at a heating rate of 5°C/min to 800°C, 900°C, and 1000°C, followed by holding at each temperature for 1 h. The resulting products were designated as FeMn@C/BC‐800, FeMn@C/BC‐900, and FeMn@C/BC‐1000. In parallel, pretreated bamboo was subjected to the same hot‐pressing and drying procedures and then placed in the quartz tube of the tube furnace and calcined at 900°C to obtain BC‐900.

4.5. Characterization

The morphology of the samples was characterized using transmission electron microscopy (TEM; JEOL JEM‐2100Plus and HITACHI HT7700, Japan). In addition, scanning electron microscopy (SEM; HITACHI SU8600, Japan) and energy‐dispersive x‐ray spectroscopy (EDS) were employed to characterize the sample's surface structure and determine its spatial elemental composition. X‐ray photoelectron spectroscopy (XPS) was employed to analyze the surface chemical composition and determine the oxidation states of constituent elements (XPS; Thermo Scientific K‐Alpha, USA). In situ Fourier‐transform infrared (FTIR) spectra were recorded on an FTIR spectrometer (Thermo Fisher Scientific Nicolet iS20, USA). The chemical composition was analyzed using Raman spectroscopy (Labram HR800 Evo, France), with a laser wavelength of 532 nm. X‐ray diffraction (XRD; Bruker D8 Advance, Germany) was performed at 40 kV with a minimum scanning speed of 2°/min over a 2θ range of 10°–70° to analyze the chemical phase composition. Thermogravimetric analysis (TGA; PerkinElmer STA 8000, USA) was conducted to record the thermal weight‐loss behavior of the samples under a nitrogen atmosphere. The surface wettability was evaluated by determining the contact angle with a goniometer (WCAS; JY‐82C, Dingsheng, China). Electrical conductivity was determined using a four‐point probe system (ST2258C, Suzhou Jingge Electronics Co., Ltd., China), with a probe spacing of 1 mm. Magnetic properties were analyzed using a vibrating sample magnetometer (Lake Shore 7404, USA). Temperature variations were recorded using an infrared thermal imaging camera (IRT; testo 869, Germany). Nitrogen adsorption–desorption isotherms and pore‐size distribution were obtained using a fully automatic specific surface area and porosity analyzer (BET; Micromeritics ASAP 2460, USA). Thermal conductivity was measured using a thermal constant analyzer (Hot Disk TPS2500S, Sweden). For mechanical testing, samples were processed into specific dimensions (30 mm × 10 mm × 0.2 mm) and evaluated using a universal testing machine (AG‐XPIUS, Japan). EM radiation intensity in demonstration scenarios was measured using a portable EMW radiation tester (LZT‐1150, LZT, China). For EMI shielding performance, samples were cut into small pieces (22.9 mm × 10.2 mm) to fit into a waveguide cavity. EMI SE in the X‐band was measured using a vector network analyzer (VNA; E5071C, Agilent, USA). Tesla coil (JT‐56‐222, input voltage: 12 V DC, operating frequency: approximately 1.5 MHz). incandescent bulb (E14, 220 V/5 W). wireless charger (A2100, 7.5 W output, 127.7 kHz operating frequency). commercial Wi‐Fi router (HUAWEI WS7206, 2.4 GHz band). electromagnetic radiation meter (LZT‐1150, LZT, China). wooden cabin model (dimensions: 300 mm × 200 mm × 200 mm).

Author Contributions

X.S. conducted the investigation and prepared the original draft of the manuscript. R.Z. and Y.C. were responsible for data curation. Y.G. and Z.C. contributed to the methodology. Y.P. handled visualization. J.W. contributed to conceptualization and writing, including review and editing of the manuscript. L.Y. performed validation, while Y.C. supervised the overall study.

Conflicts of Interest

The authors declare no conflict of interest.

Supporting information

Supporting File 1: smll73920‐sup‐0001‐SuppMat.docx.

Supporting File 2: smll73920‐sup‐0002‐MovieS1.mp4.

Supporting File 3: smll73920‐sup‐0003‐MovieS2.mp4.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.