2.1 Structural evolution
Bimetallic MOF and its derivatives were synthesized through co-precipitation, followed by pyrolysis [24]. The observed structure of the leaf-like Co/Zn-ZIF (expressed as CZ-Z; sample names are listed in Table S1) can be attributed to the asymmetric coordination behaviors induced by the distinct ligand characteristics of nitrate versus acetate anions (Fig. 1a and Fig. S1). Nevertheless, the CZ-Z-derived composites (labeled as CZ) display structural collapse into stacked circular-like sheets (Fig. 1d and Fig. S2). The abnormal structural disinheritance might stem from the asymmetric coordination environment within the MOF precursor, triggering crystallographic lattice distortions and architectural destabilization [25]. Meanwhile, the Co3ZnC intermetallic compound (PDF#29-0524) and metallic Co (PDF#15-0806), as identified by X-ray diffraction (XRD), inherently possess high surface energy characteristics that promote Ostwald ripening processes (Fig. 1c and Note S1) [26, 27]. This results in particle migration and agglomeration, ultimately leading to the catastrophic failure of the carbon scaffold's load-bearing architecture. Hence, to suppress the metallic particle aggregation while preserving the structural integrity of the precursor, the strategic incorporation of a third metallic Fe element through targeted inducing to achieve precise modulation of the phase and structure proves essential.
XRD patterns of FCZ-Zx precursors demonstrate almost identical peak positions and intensities compared to CZ-Z (Fig. S3). However, the gradual lightening of the powder coloration observed in macroscopic samples confirms the successful integration of Fe3+ into the framework and reveals alterations in the coordination environment within the system (Fig. S4a). As presented in Fig. 1e, introducing 0.6 mmol Fe3+ into the bimetallic MOF precursor results in the partial retention of leaf-like morphological features (called FCZ1), indicating that Fe3+ incorporation effectively alleviates structural collapse. Specifically, Fe3+ with inherently higher charge density competitively coordinates with 2-methylimidazole against Co2+/Zn2+ counterparts, establishing a reinforced Fe-Co-Zn hybrid coordination node [28]. Additionally, Fe nanoparticles (NPs) embedded within the carbon matrix can mitigate the migration and aggregation of metallic particles. These synergistic effects drive a topological transformation in the bimetallic MOF.
Notably, the diffraction peak intensity corresponding to the Co3ZnC phase exhibits marked attenuation, suggesting that Fe3+ incorporation suppresses the formation of the Co3ZnC phase (Fig. S4b). This phenomenon may be rationalized by the inherently superior standard reduction potential (E°) of Fe3+ (Fe3+/Fe0 = −0.037 V) compared to Co2+ (Co2+/Co0 = −0.277 V) and Zn2+ (Zn2+/Zn0 = −0.762 V) [29-31]. The preferential reduction of Fe3+ to metallic Fe dominates the initial reduction sequence, rapidly depleting available reducing agents within the system. This competitive consumption consequently imposes kinetic limitations for subsequent reduction processes of Co2+ and Zn2+, hindering the Co-Zn alloying pathway. Besides, the E° of Co2+ is higher than that of Zn2+, and the establishment of this directional reduction gradient contributes to the preferential generation of Fe-Co solid solutions (Fig. 1b). Conversely, the Zn reduction reaction is significantly inhibited, leading to the growth of amorphous ZnO.
The pyrolyzed composites fully inherit the precursor framework after increasing the Fe3+ concentration to 1.2 mmol (marked as FCZ2), as depicted in Fig. 1f. Moreover, the overall structure displays a self-assembly behavioral tendency, forming a three-dimensional interconnected network through oriented growth along two orthogonal axes of the leaf-like architecture. The observed morphology may originate from magnetic interactions and anisotropic stress effects, collectively driving the spontaneous cross-assembly phenomena [32]. With a further elevation of the Fe3+ level (labeled as FCZ3), the magnetic properties demonstrate a corresponding enhancement, accompanied by a self-assembly process initiated among multiple adjacent leaf-like architectures. Ultimately, this leads to the formation of thermodynamically favorable configurations through multistage structural integration (Fig. 1g). Concurrently, the Co3ZnC content progressively diminishes, signifying the efficacy of Fe3+ incorporation in modulating the phase distribution within the bimetallic system.
Interestingly, when introduced Fe3+ reaches 2.4 mmol (denoted as FCZ4), a well-defined morphological evolution is observed (Fig. 1h and Fig. S5). The self-assembled leaf-like architectures initially undergo mutual crop and segmentation processes, splitting into triangular boomerang-shaped units. Afterward, epitaxial growth occurs preferentially at the junction points of these units, driving further structural reorganization until a developed triangular-like architecture is achieved. The separation procedure may be attributed to the supersaturated doping of Fe3+, leading to the precipitation of excess metallic Fe nanoparticles, which catalyze the oxidative etching of the carbon substrate through high-temperature metal-carbon interfacial reactions [33]. This process could induce localized dissolution at the leaf-like edges of the carbon framework, generating triangular-like notch defects. Then, Fe NPs mediate the anisotropic carbon deposition, which is preferentially localized at triangular notches, facilitating the nucleation of branching substructures that propagate the development of pseudo-triangular architectures [34]. Eventually, the reduced interfacial energy at triangular edges activates a "self-healing" mechanism that thermodynamically stabilizes the system, enabling structural optimization toward geometrically regular configurations. Notably, XRD analysis reveals no detectable diffraction peaks related to Co3ZnC, only featuring characteristics of amorphous carbon and broadened metallic Co diffraction peaks. The result firmly suggests that the reduction of Fe3+ fully consumes the available reductant within the system, thereby effectively obstructing the formation pathway of Co3ZnC.
Incorporating 3.0 mmol Fe3+ induces enhanced magnetic interaction (recorded as FCZ5), combined with carbon-mediated bridging effects, collectively motivate the stacking and reconfiguring of the triangular-like architecture (Fig. 1i). The emergence of a novel Fe3ZnC0.5 (PDF#29-0741) phase is identified in the XRD patterns, accompanied by distinct diffraction peaks corresponding to the metallic Fe (PDF#06-0696) phase [14, 35]. This phenomenon could be explained by the saturation of Fe dissolution within the Fe-Co solid solution matrix. Once the solubility limit is exceeded, excess Fe atoms facilitate localized Fe-Zn alloying pathways. Therefore, a Fe3+ concentration of 2.4 mmol represents the critical threshold for phase transition within the system. When the Fe3+ concentration exceeds 3.0 mmol (noted as FCZ6), the diffraction peak intensities related to Fe3ZnC0.5 and Fe phases display a pronounced enhancement (Fig. S6). Simultaneously, the FCZ6 evolves and integrates at the structural articulation, ultimately attaining a coherent quasi-octahedral configuration (Fig. S7). Furthermore, systematic investigations into the effects of adding Fe3+ to monometallic MOF systems were also conducted (Fig. S8-S13 and Note S2). These results indicate that the Fe3+-dominated ternary coordination competition model and directed reduction hierarchy realize a precise regulation of the phase and structure of bimetallic MOF-derived materials (Fig. 1j). This Fe3+-driven modulation operates through targeted tuning of kinetic pathways and thermodynamic equilibria, demonstrating broad applicability across polymetallic MOF systems, thus offering the potential to attain customizable MOF-derived EMW absorbers.
Transmission electron microscopy (TEM) characterization further illustrates the structural evolution governed by the gradient Fe3+ content (Fig. 1k and Fig. S14). The magnified TEM images unlock more explicit morphological information, showing that metal NPs of varying sizes are randomly dispersed across the amorphous carbon matrix (Fig. S15). The high-resolution transmission electron microscopy (HRTEM) images confirm the existence of metal NPs anchored on amorphous carbon (Fig. S16). Meanwhile, localized regions featuring metal particles encapsulated by amorphous/graphitic carbon are also observed. HRTEM analysis reveals distinct lattice fringes with interplanar spacings of 0.204 nm and 0.202 nm, characteristic of the (111) and (110) crystallographic planes in metallic Co and Fe phases, respectively. Additionally, observed spacings of 0.216 nm and 0.219 nm are unambiguously assigned to the (111) plane family of the Co3ZnC phase and the newly formed Fe3ZnC0.5 phase. These crystal signatures provide conclusive evidence for the effective substitution of Fe3+ into the crystal lattice and the subsequent phase transition. The elements C, N, O, Co, Zn, and Fe are uniformly distributed throughout the composites, as illustrated by the results of the energy dispersive spectrometer (EDS) elemental mapping (Fig. 1l and Fig. S17-S23).
2.2 Phase transition
To elucidate the underlying mechanisms governing phase transition following Fe3+ incorporation, X-ray photoelectron spectroscopy (XPS) was employed to investigate the elemental composition and chemical states. The survey XPS spectra of CZ exhibit the presence of C, N, O, Co, and Zn elements, while the FCZx displays an extra Fe element (Fig. S24). The high-resolution C 1s spectrum can be deconvoluted into C-C/C=C (284.8 eV), C-N (286.1 eV), C-O (287.9 eV), and C=O (289.5 eV), respectively (Fig. 2a) [36]. Three characteristic peaks located at 398.7 eV, 400.8 eV, and 403.0 eV are also identified in the high-resolution N 1s spectrum, which are ascribed to pyridinic N, pyrrolic N, and graphitic N, respectively (Fig. 2b) [37]. These findings collectively indicate the successful doping of heteroatom N into the carbon lattice [38].
Notably, the high-resolution Zn 2p spectrum of CZ and FCZx can be distinctly separated into two peaks at 1022.1 eV and 1045.1 eV, which correspond to Zn 2p3/2 and Zn 2p1/2, respectively (Fig. 2d) [24]. Quantitative analysis of elemental composition via inductively coupled plasma optical emission spectrometry (ICP-OES) reveals that the Zn species concentration remained at dynamic equilibrium (Fig. 2g and Table S2-S3). This observation, along with the persistent absence of characteristic Zn-related diffraction peaks in XRD patterns and undetectable crystallographic signatures assigned to ZnO in HRTEM, strongly reveals that Zn predominantly exists as amorphous ZnO within the composites. Concurrently, the ZnO content increases with elevated Fe3+ introduction levels, directly correlated with the progressive diminishment of the Co3ZnC phase. The crystal structures and electronic properties of monometallic MZ and FZ materials provide additional evidence to corroborate this proposed argument (Fig. S25).
The high-resolution O 1s spectrum demonstrates the presence of lattice oxygen (533.5 eV) and surface-adsorbed oxygen (535.8 eV) species, as well as prominent spectral features of oxygen vacancies (531.6 eV), occupying approximately 80% of the integrated oxygen signal (Fig. 2c and Fig. S24f) [39]. These are likely to trigger enhanced dipole polarization originating from a high density of lattice defects, thus promoting further dissipation of EMW. The Co0 (778.5 eV) species and multivalent Co are recognized in the high-resolution Co 2p spectrum (Fig. 2e) [38]. Similarly, the high-resolution Fe 2p spectra of FCZx upon introducing Fe3+ exhibit Fe0 (707.5 eV) species and multivalent Fe (Fig. 2f) [40]. The surface-sensitive nature of XPS analysis suggests that the observed higher oxidation states of Co or Fe are primarily attributed to surface oxidation or atmospheric exposure effects. The bulk phase mainly retains its metallic state and solid solution characteristics. Consistent findings are further validated in the monometallic MC and FC systems (Fig. S26). ICP research also illustrates a progressive enrichment of Fe species within the composites, offering solid proof for successfully incorporating Fe3+ into the bimetallic MOF system.
The abundant metallic Co creates favorable conditions for Fe3+ to substitute its lattice sites, ultimately resulting in the generation of Fe-Co solid solution through atomic-scale substitutional doping. The following results confirm the formation of Fe-Co solid solution: i) The slight angular shift of the Co phase observed at 44.2°, ascribed to the (111) crystal plane, combined with the pronounced diffraction peak broadening characteristic, provides proof of Fe atomic incorporation into the Co lattice (Fig. 1c). ii) HRTEM analysis reveals multiple indistinguishable interplanar spacing values intermediate between pure Co (0.204 nm) and pure Fe (0.202 nm), offering direct crystallographic evidence for the generation of a disordered solid solution (Fig. S16). iii) Thermal field modulation studies demonstrate that the ordered Co7Fe3 (PDF#50-0795) intermetallic phase formation initiates at temperatures exceeding 900 °C, which corroborates the growth of the Fe-Co solid solution at low temperatures (Fig. S27 and Note S3) [41]. iv) The saturation magnetization (Ms) tested via a vibrating sample magnetometer (VSM) gradually enhances from an initial value of 13.33 emu g-1 to 20.48 emu g-1 at maximum Fe addition concentration, indicating a magnetic behavior that aligns with the fundamental principle of magnetic moment superposition in Fe-Co solid solution (Fig. 2h, Fig. S28, and Table S4) [42]. v) Density functional theory (DFT) calculations show that the Fe-Co solid solution exhibits improved thermodynamic stability relative to Co3ZnC and Fe3ZnC0.5, as evidenced by its minimum formation energy (Fig. 2j, Fig. S29-S35, Table S5, and Note S4). Besides, the hybridization between Fe 3d and Co 3d orbitals induces intensified spin polarization and enhanced exchange interactions, resulting in an increased population of unoccupied states in the spin-down channel. This configuration results in progressive improvement of energy band density with elevated Fe incorporation concentrations (Fig. 2k, l). Concurrently, the orbital hybridization shifts the spin-down-dominated antibonding states closer to the Fermi level, significantly raising the projected density of states (PDOS) of Fe. In contrast, the partial transfer of Co's d-orbital electrons into the Fe-Co hybridized states slightly reduces the PDOS contribution from Co (Fig. 2m, n). Interfacial charge density gradients at Fe-Co hybrid interfaces generate strong local dipoles, facilitating polarization relaxation and promoting dielectric loss capacity.
Electron paramagnetic resonance (EPR) spectroscopy was employed to qualitatively characterize the spatial distribution of unpaired electron density (paramagnetic substance). The high signal-to-noise ratio peaks observed in the narrow spectral region fail to provide discernible information (Fig. S36). Nevertheless, the change in the intensity of these signals likely reflects the concentration variation of paramagnetic centers and unpaired electrons. The broad EPR signals centered at g ≈ 2.15 suggest the presence of paramagnetic centers, probably attributed to high-spin Fe3+ species or unpaired electrons in Fe-Co solid solutions, instead of high-spin Co2+ (g ≈ 2.3~2.5) or metal vacancies (g ≈ 1.96) [43, 44]. The signal intensity initially increases with the Fe3+ content, reaching its peak at 2.4 mmol. This growth suggests a rising concentration of paramagnetic Fe3+ or Fe-Co solid solutions. However, a decline occurs at FCZ5, probably due to a partial Fe3+ reduction to metallic Fe0 or a decrease in Fe-Co solid solution content caused by the generation of Fe3ZnC0.5 and Fe phases.
The thermal stability of the composites was investigated by thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC). The decomposition of CZ in an air atmosphere begins at approximately 327 °C, with a sharp mass reduction above 548 °C, culminating in a loss of roughly 40.89% of its mass at 950 °C (Fig. S37). In comparison, FCZ4 exhibits an elevated decomposition temperature of 345 °C and a final mass loss of about 38.30%. This implies that multiple phases raise the temperature at which carbon decomposes by facilitating the generation of a highly eutectic mixture [45]. N2 adsorption-desorption isotherms were utilized to assess the specific surface area and pore size distribution of the composites. All samples demonstrate type IV isotherms with a distinct hysteresis loop, signifying the existence of mesoporous features (Fig. S38) [46]. The specific surface area of the composites progressively reduces with the incorporation of Fe3+, varying from an initial value of 223.3 m2 g-1 to 47.4 m2 g-1. At the same time, the average pore size continuously rises, ranging from 3.62 nm to 18.60 nm (Fig. 2i). The hierarchical self-assembly process and subsequent structural reconfiguration of the leaf-like architectures lead to a decrease in specific surface area while promoting the generation of a higher density of mesoporous structures. These mesopores are conducive to multiple internal reflections of incident EMW within the material matrix, extending their propagation pathway and favoring enhanced attenuation efficiency of EMW.
2.3 Outstanding EMW Absorption Performances
The composites with tunable phases and structures were ingeniously applied to absorb EMW at gigahertz frequencies. The reflection loss (RL) intensity and effective absorption bandwidth (EAB, the frequency range for RL ≤−10 dB) were employed to assess the EMW capacity of the composites with a mass filler loading of 50%. As illustrated in Fig. 3a and Fig. S39-S40, CZ exhibits relatively general absorption properties at high frequency, achieving a minimum RL (RLmin) of −50.99 dB at a thickness of 2.97 mm and a maximum EAB of 0.48 GHz at a thickness of 1.50 mm. Nevertheless, CZ demonstrates remarkable EMW attenuation performance in the low-to-mid frequency range, showing an absorption bandwidth of 4.32 GHz spanning from 5.60 to 9.92 GHz. It is noteworthy that the remarkably enhanced comprehensive performance of the bimetallic MOF-derived composites compared to their monometallic counterparts may be ascribed to the multicomponent competitive synergies (Fig. S41). As the leaf-like morphology gradually occurs, FCZ1 presents an improvement in RL and EAB, with RLmin of −61.67 [email protected] mm and a 5.20 [email protected] mm wide EAB, covering 12.80-18.00 GHz (Fig. 3b). Then, despite the reduction in RLmin (−22.26 dB) of the FCZ2, its EAB elevates to 5.60 GHz at a matched thickness of 1.83 mm (Fig. 3c). The absorption band progressively shifts from the low-to-mid frequency range toward the higher frequency region. Consequently, the effective bandwidth in the low-to-mid frequency range gradually narrows, while the EAB in the high-frequency range demonstrates continuous broadening. When increasing the level of Fe3+, the absorption capacity of FCZ3 is ameliorated with an EAB of 5.84 [email protected] mm (Fig. 3d). This suggests that the hierarchical self-assembly architecture is critical in facilitating enhanced geometric scattering and optimized interfacial charge polarization mechanisms, thereby boosting the electromagnetic response.
Among all the synthesized absorbers, FCZ4 with diverse microstructures possesses the most exceptional EMW absorption performance, featuring an ultra-wide EAB of up to 6.08 GHz only at 2.03 mm thickness, completely covering the Ku-band (Fig. 3e). At this moment, the geometric scattering effects and electromagnetic component compatibility achieve an optimal equilibrium, thus realizing maximum EMW efficiency. Nevertheless, the structural reconfiguration and ordered rearrangement of FCZ5 induce simplified geometric scattering pathways and compromised impedance matching, resulting in a substantial decline of its EMW absorption capabilities, with an EAB of 5.60 [email protected] mm (Fig. 3f). As the Fe3+ concentration is promoted further and architectural features fade, this diminishing tendency becomes more prominent. Hence, the FCZ6 demonstrates an RLmin of −17.71 dB and an EAB of 5.20 GHz (Fig. S42). Moreover, the monometallic MOF system reveals enhanced performance upon introduction of Fe3+, validating the efficacy and broad applicability of this strategic Fe3+ incorporation approach in advanced absorber design engineering (Fig. S43). Notably, optimizing the FCZ4 filler loading enables a prominently strengthened reflection loss capability, with RLmin values reaching −84.41 dB and −66.96 dB, as depicted in Fig. 3g, h and Fig. S44. To explicitly reflect the excellent EMW absorption property exhibited by FCZ4, the performance of other recently reported MOF-derived absorbers is summarized for comparison. As displayed in Fig. 3i, j and Table S6, the minimum RL and maximum EAB of FCZ4 exceed those of other absorbers, along with its advantages in controlled structure and modifiable component, FCZ4 emerges as a promising candidate for customizable high-performance MOF-derived absorbers.
2.4 Distinctive Electromagnetic Response Behaviors
The exceptionally high-frequency EMW absorption property is linked to its distinct electromagnetic parameters. Thus, the related complex permittivity (εʹ, εʹʹ) and complex permeability (μʹ, μʹʹ) are first analyzed. The real parts of the complex permittivity and complex permeability, which stand for the storage abilities for electrical polarization and magnetic polarization, respectively, are denoted by the terms εʹ and μʹ [47]. Conversely, the imaginary parts of the complex permittivity (εʹʹ) and complex permeability (μʹʹ) express the capability of electric dipole and magnetic dipole moments to dissipate EMW as they realign within an alternating electromagnetic field [27]. Besides, the tangent values of complex permittivity (tanδε = εʹʹ/εʹ) and complex permeability (tanδμ = μʹʹ/μʹ) represent the relationship between the ratio of loss and stored energy [48]. As illustrated in Fig. 4a, b, the εʹ and εʹʹ values of CZ exhibit marked fluctuations within the ranges of 6.81 to 12.96 and 1.65 to 6.63, respectively. Meanwhile, FCZ1 enhances both parameters following the incorporation of Fe3+. Then, despite the continued increase in Fe3+ concentration leading to a decrease in the εʹ value (from 7.19 to 14.05), FCZ2 displays a significant improvement in εʹʹ value (from 5.13 to 8.62). As the Fe3+ level progressively rises, the composites show a gradual enhancement in both εʹ and εʹʹ values alongside reduced curve fluctuations, with FCZ5 achieving the highest εʹ (from 7.41 to 16.20) and εʹʹ (from 3.71 to 9.84) values among all samples.
However, the μʹ and μʹʹ values fluctuate around 1 and 0, respectively (Fig. S45). This behavior may also be attributed to the intrinsic resonance frequency shift caused by nanoscale magnetic metallic particles, insufficient magnetic coupling interactions, and negligible eddy current generation resulting from the amorphous carbon encapsulation layer. These factors prevent the establishment of frequency-specific magnetically responsive characteristics within the system [49]. The considerably higher tanδε values than tanδμ values in the composites reveal that the EMW dissipation capability originates primarily from dielectric loss rather than magnetic loss contributions (Fig. 4c). Additionally, FCZ4 demonstrates the highest tanδε value throughout the frequency band, which accounts for its exceptional EMW absorption performance among all samples. Instead, despite possessing the highest values of εʹ and εʹʹ, the formation of new phases elevates electrical conductivity (σ), which induces impedance mismatch and compromises EMW attenuation efficiency, thereby reducing the tanδε value of FCZ5 (Fig. 4d and Fig. S46).
The Co3ZnC and metallic Co phases, with superb electrical properties, are responsible for the highest σ of the CZ sample. Nevertheless, the gradual reduction in σ of the composites with the rise in Fe3+ amount reaches a minimum at FCZ3. The result is primarily driven by the synergistic effects of Fe-Co solid solution formation with enhanced electrical resistivity relative to metallic Co and the progressive depletion of Co3ZnC phase content. Subsequently, the highly conductive Fe3ZnC0.5 and Fe phases progressively nucleate and grow, while the structural evolution involving reorganization facilitates a partial restoration of the conductive network, thereby contributing to the recovery of the electrical conductivity. Therefore, the conductivity loss (εcʹʹ), which is directly proportional to the material conductivity (εcʹʹ = σ/ωε0), also exhibits a sharp initial decline followed by a slow climb (Fig. 4e-g) [38]. The polarization loss (εpʹʹ), which additionally contributes to the dielectric loss (εpʹʹ = εʹʹ-εcʹʹ), shows an opposite trend. In brief, introducing Fe3+ induces a substantial alteration in the dielectric response by shifting the primary energy dissipation mechanism from conductive loss to polarization-controlled processes.
Intriguingly, two prominent polarization relaxation peaks are observed in the εʹʹ-curve for the CZ. These characteristic peaks demonstrate a continuous reduction in intensity with increasing Fe3+ concentration. The relaxation feature eventually diminished to baseline levels upon reaching the FCZ3. The Cole-Cole plots, based on the Debye theory, were utilized to elaborate on this unique polarization behavior. Individual semicircles correspond to specific relaxation processes induced by interfacial polarization or dipole reorientation in the Cole-Cole plots, and the trailing straight lines signify electronic conduction losses arising from charge carrier migration [50]. The CZ plot presents two conspicuous semicircles accompanied by linear tails, indicating the concurrent presence of multiple polarization processes coupled with conduction loss (Fig. 4i) [51]. The semicircular features progressively decline in prominence as Fe3+ levels rise, while the linear components reveal enhanced definition, which seems to imply a gradual transition from polarization-dominated to conductivity-dominated dielectric response. The result, however, contradicts the conclusions from the quantitative studies mentioned above.
Consequently, we posit that the anomalous dielectric behavior is inextricably linked to the compositional variations. Specifically, in the bimetallic CZ system before Fe3+ incorporation, dipole polarization mainly originates from lattice defects in amorphous ZnO, while interfacial polarization is predominantly governed by the heterointerfaces at Co3ZnC/C and Co/C boundaries. These independent polarization mechanisms collectively give rise to the formation of two well-separated relaxation peaks. Upon adding Fe3+, the progressive increase of Fe-Co solid solution content facilitates the generation of extensive and sophisticated heterogeneous interfacial structures, considerably amplifying the interfacial polarization response. The incremental raising of the amorphous ZnO level also induces higher concentrations of lattice defects, thereby boosting the defect-induced polarization effect and further enhancing dipole polarization relaxation. Hence, the intensified synergistic coupling within the multipolarization mechanism may lead to the progressive overlap of relaxation peaks into a single, broad dielectric response, culminating in the gradual smoothing of the curve.
The polarization relaxation parameters could be obtained from fitting analysis using a modified Havriliak-Negami model, thus quantitatively accounting for the observed concurrent reduction in relaxation peak intensity and elevation of polarization loss (Fig. S47 and Table S7) [52]. As depicted in Fig. 4q, the parameter α, which governs the breadth of the relaxation time distribution, demonstrates a consistent downward trend. This suggests broadening the relaxation time distribution, thus diminishing the distinct peak shape in the εʹʹ curve. Besides, the gradual increase of the relaxation time (τ) results in the characteristic relaxation frequency shift to a lower frequency. When this frequency progressively falls outside the detectable frequency range (i.e., less than 2.0 GHz), it manifests as the "disappearance" of the relaxation peak phenomenon. Notably, even in the vanishing of the relaxation peak, the substantial enhancement in dielectric strength (Δε = εs-ε∞) sustains the overall polarization loss at elevated levels, which corroborates the earlier findings derived from experimental conductivity measurements. These results deviate from the expected behavior of the classical relaxation model, implying the presence of intricate multiscale polarization coupling mechanisms within the material. In summary, the anomalous dielectric characteristics can be ascribed to the coupling effect of multi-polarization mechanisms primarily governed by the Fe-Co solid solution and amorphous ZnO phases.
The D and G bands are critical features for characterizing carbon-based materials in Raman spectroscopy [37]. The D band, appearing near 1350 cm-1, is associated with structural defects and arises from the breathing mode of sp3-hybridized carbon atoms [14]. The G band, typically observed around 1580 cm-1, corresponds to the in-plane vibrational mode of sp2-hybridized carbon atoms and reflects the crystallinity of the graphitic structure [53]. The intensity ratio of these bands (ID/IG) is a quantitative indicator of defect density within the material. As depicted in Fig. 4h, the ID/IG ratio exhibits a progressive increase with incremental Fe3+ concentration, reaching peak values at FCZ3 and FCZ4 before demonstrating a subsequent decline. This trend suggests that the defect density within the system initially rises and then decreases, causing corresponding variations in the defect-induced polarization loss. Moreover, photoluminescence (PL) spectroscopy was employed to characterize the concentration of defects within the materials, including vacancies, interstitials, and dislocations. These crystal defects act as nonradiative recombination centers, dissipating excitation energy through phonon interactions, which results in the intensity of the PL spectra being inversely proportional to the level of the defects [36]. The observed tendency of spectral intensity displays an initial reduction followed by an elevation with gradual Fe3+ incorporation (Fig. S48). This provides compelling evidence that polarization relaxation triggered by high-concentration defects progressively emerges as a key factor governing the system's dielectric behavior.
HRTEM analysis of FCZ4 reveals the existence of multiple homogeneous interfaces composed of the Co phase, along with heterogeneous interfaces consisting of both Fe and Co phases embedded within an amorphous carbon matrix (Fig. 4j, k). These interfaces promote substantial electron aggregation at interfacial boundaries, generating robust interfacial polarization phenomena [54]. Meanwhile, several stress concentration points are intuitively observed, derived from geometrical phase analysis (GPA), and this inhomogeneous stress fraction, due to an unbalanced charge distribution at the interface, could be a defect-induced polarization center (Fig. 4n, o) [18]. Similarly, through inverse fast Fourier transform (IFFT) research, well-defined grain boundaries and lattice stacking are resolved, with these lattice deficiencies demonstrating the capacity to act as polarization-active centers owing to their localized structural distortions (Fig. 4p) [55]. In conjunction with N atoms doped into the carbon lattice, which can also serve as polarization-triggered centers, these substantial polarization centers generate pronounced polarization relaxation phenomena, thereby amplifying the dielectric response under alternating electromagnetic fields [38]. Furthermore, the high density of oxygen vacancies, point defects, lattice discontinuities, and dislocations induces local electric field distortions, which yield a dipole polarization response that effectively lowers the polarization relaxation energy barrier, thereby significantly enhancing polarization relaxation loss (Fig. 4l, m) [53]. The synergistic interface and dipole polarization interaction facilitate a distinctive and efficient dielectric response for reinforced EMW attenuation.
Impedance matching (Z) and attenuation constant (AC) constitute two vital performance metrics for evaluating EMW absorption efficiency. The Z ensures minimal interfacial reflection to maximize EMW penetration into the material's interior, while the AC governs the rapid energy conversion of incident waves into thermal or other forms of energy for dissipation. The impedance matching region is defined when the input impedance (Zin) approaches the free space impedance (Z0) with a normalized magnitude ratio |Zin/Z0| ranging between 0.8 and 1.2, as this specific impedance range allows near-complete penetration of EMW into the material's interior [50]. As illustrated in Fig. S49, the impedance characteristics demonstrate progressive deviation from the ideal matching area, with the FCZ4 also failing to attain optimal impedance matching performance. Although the impedance matching declines in FCZ4, the enhanced attenuation capability resulting from the elevated εʹʹ partially compensates for reflection losses, thereby substantially broadening the EAB. By adjusting the FCZ4 filler loading in the composites to modulate the complex permittivity, optimal impedance matching can be realized. This leads to an improved reflection loss intensity while simultaneously narrowing the EAB due to the constraints imposed by the quarter-wavelength matching mechanism (Fig. S50-S52) [56]. Besides, the variation trend of the AC is observed in Fig. S53. The progressive degradation of impedance characteristics prevents FCZ5 from attaining satisfactory absorption performance despite exhibiting the maximum AC value among the investigated materials [57].
Thus, the low doping ratio fulfills excellent impedance matching and strong reflection loss, but is unable to achieve the best EAB. In contrast, the high doping ratio enhances dielectric loss and broadband attenuation capability, thereby boosting EAB at the expense of impedance mismatch, which weakens the absorption peak intensity and induces a high-frequency shift. This requires balanced optimization between these competing mechanisms for optimal EMW performance. Moreover, the thermal field modulation results demonstrate that controlled annealing temperature not only dominates the phase evolution path of the composites but also enables effective regulation of the electromagnetic response, rendering the synthesized materials potential for multifunctional EMW management (Fig. S54-S55). The ingeniously designed gradient Fe3+ introduction achieves the tailoring of dielectric response characteristics through precise phase modulation engineering. This combines hierarchical architectures that enhance electromagnetic impedance compatibility, collectively establishing an intelligent platform for developing high-performance EMW absorbers with tunable electromagnetic properties.
2.5 EMW Absorption Mechanisms
To assess the effectiveness of the synthesized composites in absorbing EMW for practical applications, computational simulations of the far-field radar scattering cross section (RCS) were implemented using the CST electromagnetic simulation platform [53]. As displayed in Fig. 5a and Fig. S56-S57, FCZ4 demonstrates the lowest RCS signals. The analysis of the specific functional relationship between the RCS value and the incident angle exhibits that the FCZ4 is capable of reducing RCS signal strength by up to 10.7 dB m2 compared to the PEC (Fig. 5b) [36]. This indicates that the FCZ4 reveals good environmental adaptability to complex far-field conditions while fulfilling practical application requirements. The electromagnetic power loss density (EPLD) analysis of composites simulated using COMSOL Multiphysics software suggests that progressive geometric structural optimization effectively enhances the scattering effect toward EMW, enabling the FCZ4 to possess superior EMW consumption efficiency (Fig. 5c) [38].
The main loss mechanisms responsible for the extraordinary absorption performance of FCZ4 are summarized as follows: i) The morphological evolution and configuration optimization within FCZ4 synergistically enhance EMW interactions through intensified multiple reflections and scattering [58]. ii) Rational modulation of intrinsic conductivity effectively reduces the reflection loss of EMW on the material surface. The three-dimensional conductive network, composed of metal and carbon components, facilitates the directional migration of free charge carriers in response to alternating electromagnetic fields. This microcurrent generation mechanism promotes efficient electromagnetic energy dissipation via conduction loss through thermal conversion [59]. iii) The dispersed heterogeneous structural units within the N-doped carbon matrix, particularly the Fe-Co solid solution, effectively modulate the carrier migration and separation behavior by constructing a high-density phase boundary/heterogeneous interface network [60]. This multiscale interface engineering induces the accumulation of directional spatial charge, forming a pronounced non-equilibrium charge distribution gradient that generates a strong built-in electric field via interfacial dielectric relaxation mechanisms and amplifies the effects of interfacial polarization [61]. iv) The competitive coordination interactions among ternary transition metal ions, coupled with the concurrent presence of multi-ligand configurations, trigger the development of lattice distortions and defect architectures generated during the pyrolysis of MOF materials. These topological defects, coupled with amorphous ZnO, function as localized polarization centers that induce relaxation losses associated with dipole orientation polarization by establishing asymmetric charge distributions [62]. The synergistic interaction among multifaceted dissipation mechanisms, including geometric scattering effects, conductive losses, interfacial polarization relaxation, and dipole relaxation processes, collectively contributes to the high-efficiency EMW attenuation characteristics in FCZ4 composites (Fig. 5d).