Excellent Microwave Absorption Properties Derived from the Synthesis of Hollow Fe3o4@Reduced Graphite Oxide (RGO) Nanocomposites

Magnetic nanoparticles, such as Fe3O4 and Co3O4, play a vital role in the research on advanced microwave absorbing materials, even if problems such as high density and narrow band impedance matching are still unsolved. Herein, the study of lightweight hollow Fe3O4@reduced graphite oxide (RGO) nanocomposites synthesized via the solvothermal method is presented. The microstructure and crystal morphology of the materials were characterized by X-ray diffractometer (XRD), X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), and transmission electron microscopy (TEM) analyses. Single crystalline hollow Fe3O4 spheres were grown onto RGO flakes, leading to the formation of heterojunction, which further influenced the microwave absorption properties. The latter were evaluated by standard microwave characterization in the frequency range of 2–18 GHz. It was found that, for a specific Fe3O4@0.125 g RGO composite, the minimum reflection loss can reach −41.89 dB at 6.7 GHz, while the reflection loss was less than −10 dB from 3.4 GHz to 13.6 GHz for a nanocomposite sample thickness in the range of 1–4 mm. The combination of these two materials thus proved to give remarkable microwave absorption properties, owing to enhanced magnetic losses and favorable impedance matching conditions.


Introduction
With the rapid advancement of science and technology, more and more electrical equipment and information systems have been used in various fields. However, such systems cause electromagnetic (EM) radiation, which is harmful to people's health and affects the normal operation of electronic equipment. Therefore, the research on microwave absorbing materials (MAMs) has attracted extensive attention [1]. As an effective means to improve the survivability and penetration ability of weapon systems, the utilization of MAMs is extremely important for absorbing EM waves, reducing EM radiation, and improving the human living environment [2,3]. Nowadays, exploring new lightweight, reduced-thickness MAMs to meet the requirements of a wide absorption bandwidth, strong attenuation property with a good impedance match is very important [4][5][6].

Characterization and Measurement
An X-ray diffractometer (XRD, D8A Advance, BRUKER) was utilized to characterize the crystal structure of samples. X-ray photoelectron spectroscopy (XPS) data were obtained using a Thermo Fisher 0ESCALAB 250Xi. The morphology, size, and microstructure of the nanocomposites were examined on an FEIXL30 scanning electron microscope (SEM) and a G2F20 transmission electron microscope (TEM) [29]. The electromagnetic parameters were collected by a network analyzer (VNA, N5242A PNA-X, Agilent) through the classical coaxial measurement method by measuring in the frequency span from 10 MHz to 26.5 GHz. For the measurement of electromagnetic wave absorption properties, the samples were dispersed in paraffin homogeneously with a weight (wt%) ratio of 1:1 (sample/paraffin), and then the mixture was pressed into a cylindrical shape ( out φ = 7 mm and in φ = 3 mm) with a thickness of only 3 mm. High requirements were needed for the test system to prepare samples, such as a smooth surface, flat without burrs or scratches, and no gaps between the inner and outer surfaces of the samples.

XRD and XPS Analyses
For the sake of clarification, the XRD patterns of hollow Fe3O4 and hollow Fe3O4@RGO are illustrated in Figure 2. As could be observed, the typical peaks at 29.9°, 35.5°, 42.9°, 53.1°, 56.8°, and 62.5° could be readily indexed to the (220), (311), (400), (422), (511), and (440) planes of hollow Fe3O4; moreover, the face-centered cubic phase structure (JCPDS no. 89-2355) could also be seen [30]. Obviously, all peaks in the hollow Fe3O4@RGO curve are slightly weaker than those in the hollow Fe3O4 curve, which could be ascribed to the graphene content in the composites. In addition, the (311) peak of hollow Fe3O4@RGO shifted toward the lower angle of 35.3°, which is in contrast to that of the hollow Fe3O4 shown in the magnified picture of XRD patterns in Figure 2b. To confirm the phases and structures of the hollow Fe3O4@RGO composites, XPS spectra were measured. Elements, including C, O, and Fe were found to be correlated with various atom contents, as illustrated by the large energy range spectrum (Figure 3a). In Figure 3b, the high-resolution

Characterization and Measurement
An X-ray diffractometer (XRD, D8A Advance, BRUKER) was utilized to characterize the crystal structure of samples. X-ray photoelectron spectroscopy (XPS) data were obtained using a Thermo Fisher 0ESCALAB 250Xi. The morphology, size, and microstructure of the nanocomposites were examined on an FEIXL30 scanning electron microscope (SEM) and a G2F20 transmission electron microscope (TEM) [29]. The electromagnetic parameters were collected by a network analyzer (VNA, N5242A PNA-X, Agilent) through the classical coaxial measurement method by measuring in the frequency span from 10 MHz to 26.5 GHz. For the measurement of electromagnetic wave absorption properties, the samples were dispersed in paraffin homogeneously with a weight (wt%) ratio of 1:1 (sample/paraffin), and then the mixture was pressed into a cylindrical shape (φ out = 7 mm and φ in = 3 mm) with a thickness of only 3 mm. High requirements were needed for the test system to prepare samples, such as a smooth surface, flat without burrs or scratches, and no gaps between the inner and outer surfaces of the samples.

XRD and XPS Analyses
For the sake of clarification, the XRD patterns of hollow Fe 3 O 4 and hollow Fe 3 O 4 @RGO are illustrated in Figure 2. As could be observed, the typical peaks at 29.9 • , 35.5 • , 42.9 • , 53.1 • , 56.8 • , and 62.5 • could be readily indexed to the (220), (311), (400), (422), (511), and (440) planes of hollow Fe 3 O 4 ; moreover, the face-centered cubic phase structure (JCPDS no. 89-2355) could also be seen [30]. Obviously, all peaks in the hollow Fe 3 O 4 @RGO curve are slightly weaker than those in the hollow Fe 3 O 4 curve, which could be ascribed to the graphene content in the composites. In addition, the (311) peak of hollow Fe 3 O 4 @RGO shifted toward the lower angle of 35.3 • , which is in contrast to that of the hollow Fe 3 O 4 shown in the magnified picture of XRD patterns in Figure 2b.

Characterization and Measurement
An X-ray diffractometer (XRD, D8A Advance, BRUKER) was utilized to characterize the crystal structure of samples. X-ray photoelectron spectroscopy (XPS) data were obtained using a Thermo Fisher 0ESCALAB 250Xi. The morphology, size, and microstructure of the nanocomposites were examined on an FEIXL30 scanning electron microscope (SEM) and a G2F20 transmission electron microscope (TEM) [29]. The electromagnetic parameters were collected by a network analyzer (VNA, N5242A PNA-X, Agilent) through the classical coaxial measurement method by measuring in the frequency span from 10 MHz to 26.5 GHz. For the measurement of electromagnetic wave absorption properties, the samples were dispersed in paraffin homogeneously with a weight (wt%) ratio of 1:1 (sample/paraffin), and then the mixture was pressed into a cylindrical shape ( out φ = 7 mm and in φ = 3 mm) with a thickness of only 3 mm. High requirements were needed for the test system to prepare samples, such as a smooth surface, flat without burrs or scratches, and no gaps between the inner and outer surfaces of the samples.

XRD and XPS Analyses
For the sake of clarification, the XRD patterns of hollow Fe3O4 and hollow Fe3O4@RGO are illustrated in Figure 2. As could be observed, the typical peaks at 29.9°, 35.5°, 42.9°, 53.1°, 56.8°, and 62.5° could be readily indexed to the (220), (311), (400), (422), (511), and (440) planes of hollow Fe3O4; moreover, the face-centered cubic phase structure (JCPDS no. 89-2355) could also be seen [30]. Obviously, all peaks in the hollow Fe3O4@RGO curve are slightly weaker than those in the hollow Fe3O4 curve, which could be ascribed to the graphene content in the composites. In addition, the (311) peak of hollow Fe3O4@RGO shifted toward the lower angle of 35.3°, which is in contrast to that of the hollow Fe3O4 shown in the magnified picture of XRD patterns in Figure 2b. To confirm the phases and structures of the hollow Fe3O4@RGO composites, XPS spectra were measured. Elements, including C, O, and Fe were found to be correlated with various atom contents, as illustrated by the large energy range spectrum (Figure 3a). In Figure 3b, the high-resolution To confirm the phases and structures of the hollow Fe 3 O 4 @RGO composites, XPS spectra were measured. Elements, including C, O, and Fe were found to be correlated with various atom contents, as illustrated by the large energy range spectrum (Figure 3a). In Figure 3b, the high-resolution spectrum of Fe could be deconvoluted into two peaks, related to Fe 2p3/2 and Fe 2p1/2, which correspond to the band energies of 710.9 eV and 724.3 eV, respectively, indicating that the mixing ratio of oxides of Fe (II) and Fe (III) is consistent with the reported value of Fe 3 O 4 [31]. Figure 3c shows the O1s peaks at 529.1 eV and 532.1 eV, which could be attributed to the O element in Fe 3 O 4 and RGO, respectively [32]. Moreover, the C1s peak ( Figure 3d) at 285.1 eV could be attributed to the C element in RGO. Hence, the composites are suggested to be composed of Fe 3 O 4 and RGO, revealing that hollow Fe 3 O 4 was successfully grown on the RGO multilayer. Nanomaterials 2019, 9, x FOR PEER REVIEW 4 of 12 spectrum of Fe could be deconvoluted into two peaks, related to Fe 2p3/2 and Fe 2p1/2, which correspond to the band energies of 710.9 eV and 724.3 eV, respectively, indicating that the mixing ratio of oxides of Fe (II) and Fe (III) is consistent with the reported value of Fe3O4 [31]. Figure 3c shows the O1s peaks at 529.1 eV and 532.1 eV, which could be attributed to the O element in Fe3O4 and RGO, respectively [32]. Moreover, the C1s peak ( Figure 3d) at 285.1 eV could be attributed to the C element in RGO. Hence, the composites are suggested to be composed of Fe3O4 and RGO, revealing that hollow Fe3O4 was successfully grown on the RGO multilayer.

Morphology Analysis
At the top of Figure 4, SEM and TEM images of the materials are shown, which contribute to the visualization of the morphology of both the hollow Fe3O4 and hollow Fe3O4@RGO nanocomposites. As presented in Figure 4a, all products have a hollow spherical structure with a uniform size, and the inner hollow structure can be observed in several broken pellets. Clearly, the cavity size is 130 nm, and the corresponding shell thickness is 35 nm, as can be seen in Figure 4d. Figures 4b-d provide representative images of the hollow Fe3O4@RGO, which suggest that the hollow Fe3O4 microspheres are adhered to the flaky RGO. Figures 4e-h display the overall structure of the hollow Fe3O4@RGO. Typically, the white area covered by gray or black areas is indicative of the hollowness of Fe3O4 microspheres, while the sample edge indicates the existence of RGO. In particular, it can be observed from Figure 4h that the RGO had coated the hollow Fe3O4. Notably, the Fe3O4 spheres in the surface of the hollow Fe3O4@RGO composites could improve the interaction between Fe3O4 and RGO. For the hollow Fe3O4@RGO composites, many charge carriers accumulated onto the interfaces, as shown in the inset of Figure 4h, which would lead to increased dielectric loss. Typically, the incident electromagnetic microwaves could be reflected and scattered between the multilayer interfaces, which would give rise to the dissipation of electromagnetic radiation energy [33].

Morphology Analysis
At the top of Figure 4, SEM and TEM images of the materials are shown, which contribute to the visualization of the morphology of both the hollow Fe 3 O 4 and hollow Fe 3 O 4 @RGO nanocomposites. As presented in Figure 4a, all products have a hollow spherical structure with a uniform size, and the inner hollow structure can be observed in several broken pellets. Clearly, the cavity size is 130 nm, and the corresponding shell thickness is 35 nm, as can be seen in Figure 4d.  Figure 4h, which would lead to increased dielectric loss. Typically, the incident electromagnetic microwaves could be reflected and scattered between the multilayer interfaces, which would give rise to the dissipation of electromagnetic radiation energy [33].

Microwave Absorption Properties
Generally, the electromagnetic microwave absorption mechanism is investigated by parameters such as the relative complex dielectric permittivity (εr = ε' − jε″), complex magnetic permeability (μr = μ' − jμ″), and corresponding tangents, as displayed in Figure 5. Among them, ε' and μ' are well known to represent the storage ability of electromagnetic energy, whereas ε″ and μ″ are linked with energy dissipation [34]. Furthermore, both dielectric loss and magnetic loss are responsible for energy attenuation in electromagnetic microwave absorption, which can be characterized by tanδ ε and tanδμ, respectively [35].
As can be observed from Figures 5a,b, the ε' and ε″ values of the hollow Fe 3 O 4 @RGO composites are higher than those of the hollow Fe 3 O 4 , owing to the addition of dielectric loss materials, which is particularly true for the hollow Fe 3 O 4 @0.5 g RGO composites. In detail, the ε' values show a declining trend versus the changing frequency, whereas the ε″ values exhibit a fluctuant variation trend, since the variation peaks appear at 6.5, 10, and 15 GHz across the whole range. Moreover, the permittivity curves indicate that the dielectric properties of the hollow

Microwave Absorption Properties
Generally, the electromagnetic microwave absorption mechanism is investigated by parameters such as the relative complex dielectric permittivity (ε r = ε' -jε"), complex magnetic permeability (µ r = µ' -jµ"), and corresponding tangents, as displayed in Figure 5. Among them, ε' and µ' are well known to represent the storage ability of electromagnetic energy, whereas ε" and µ" are linked with energy dissipation [34]. Furthermore, both dielectric loss and magnetic loss are responsible for energy attenuation in electromagnetic microwave absorption, which can be characterized by tanδ ε and tanδ µ , respectively [35]. Based on the measured relative complex permeability and permittivity, the reflection loss (RL) of samples could be obtained according to the expressions listed below [36]: = (3) Figure 5. Real part of the relative complex permittivity (a); imaginary part of the relative complex permittivity (b); electric loss tangents (c); real part of the relative complex permeability (d); imaginary part of the relative complex permeability (e); and magnetic loss tangents of the prepared composites (f).
As can be observed from Figure 5a,b, the ε' and ε" values of the hollow Fe 3 O 4 @RGO composites are higher than those of the hollow Fe 3 O 4 , owing to the addition of dielectric loss materials, which is particularly true for the hollow Fe 3 O 4 @0.5 g RGO composites. In detail, the ε' values show a declining trend versus the changing frequency, whereas the ε" values exhibit a fluctuant variation trend, since the variation peaks appear at 6.5, 10, and 15 GHz across the whole range. Moreover, the permittivity curves indicate that the dielectric properties of the hollow Fe 3 O 4 are greatly improved with the addition of RGO. Figure 5d Figure 5c,f present the dielectric loss tangent and magnetic loss tangent of the samples, respectively. Notably, the dielectric loss tangent is less than 0.15, except for the hollow Fe 3 O 4 @0.25 g RGO in the range of 13-17 GHz, while the magnetic loss is mostly above 0.2. Obviously, the hollow Fe 3 O 4 @RGO composites possess a higher tanδ µ at 2-18 GHz, which indicates that the magnetic loss contributes to the electromagnetic microwave absorption.
Based on the measured relative complex permeability and permittivity, the reflection loss (RL) of samples could be obtained according to the expressions listed below [36]: where Z in , Z 0 are the normalized input and free space characteristic impedance, respectively; ε 0 , µ 0 are the permittivity and permeability of vacuum, respectively; c is the light velocity; d is the thickness of absorber; and f represents the microwave frequency. Figure 6a- Figure 6c. For the hollow Fe 3 O 4 @0.125 g RGO composites in Figure 6d, the minimum RL was −41.89 dB at 6.7 GHz at 2.5 mm. Moreover, the RL increased with the decrease in RGO, and the peak shifted to a higher frequency. It could be clearly observed when comparing the minimum RL of three samples that the hollow Fe 3 O 4 @0.125 g RGO composites displayed the optimal composite mode with a reduced thickness. More importantly, the top-level microwave absorption performances of the hollow Fe 3 O 4 @0.125 g RGO composites further underlined the strong magnetic loss of the material. All composites exhibited higher RLs compared with those of the single hollow Fe 3 O 4 , which proves that the combination of hollow Fe 3 O 4 and RGO would induce a better performance of microwave absorption, thus demonstrating that RGO could greatly improve the absorption property owing to the dielectric loss. Microwave absorption properties are greatly associated with thickness; typically, the RL of the hollow Fe3O4@RGO is calculated for an absorber thickness of 1-4 mm by Equations (1)-(3), as shown in Figures 7a-c. Remarkably, the bandwidths (RL below −10 dB) cover the range of 2.7-13.8 GHz for thicknesses ranging from 1 to 4 mm within the overall frequency for all the composites. Moreover, it can be noticed that for higher thicknesses the minimum RL shifts towards lower frequencies, while the absorption peaks of the hollow Fe3O4@RGO composites become sharper. The RL reached a maximum of −34.3 dB for the 3-mm thick hollow Fe3O4@0.25 g RGO composite. Noticeably, for lower thicknesses, higher losses occur for lower RGO contents within the composites, while an opposite trend can be established at higher thicknesses. A sort of 'turning point' can be thus conceived at a thickness of 3 mm (green curves in Figure 7) regarding the interaction between microwaves and this typology of RGO-reinforced composites. Specifically, the frequency was found to be partly related to the maximum RL at a certain thickness, as displayed on the quarter-wavelength cancellation model [37]. The matching equation is expressed below: where tm, λ, and fm are the matching thickness of the absorber, the wavelength, and the absorption frequency of the electromagnetic microwave. The calculated matching thickness and the peak frequency are plotted below the RL curves, where the red squares represent the specific thickness in the range of 1-4 mm corresponding to the peak frequency. In addition, the simulation of tm vs frequency for the hollow Fe3O4@RGO composites are also depicted based on the above equation. The red squares appear on the simulation curve, suggesting that the absorber thickness could be tailored to design the microwave absorption materials. Furthermore, better impedance matching maximizes the penetration of electromagnetic waves into the material and reduces the direct reflection on the surface of the material [38]. When the value Microwave absorption properties are greatly associated with thickness; typically, the RL of the hollow Fe 3 O 4 @RGO is calculated for an absorber thickness of 1-4 mm by Equations (1)-(3), as shown in Figure 7a-c. Remarkably, the bandwidths (RL below −10 dB) cover the range of 2.7-13.8 GHz for thicknesses ranging from 1 to 4 mm within the overall frequency for all the composites. Moreover, it can be noticed that for higher thicknesses the minimum RL shifts towards lower frequencies, while the absorption peaks of the hollow Fe 3 O 4 @RGO composites become sharper. The RL reached a maximum of −34.3 dB for the 3-mm thick hollow Fe 3 O 4 @0.25 g RGO composite. Noticeably, for lower thicknesses, higher losses occur for lower RGO contents within the composites, while an opposite trend can be established at higher thicknesses. A sort of 'turning point' can be thus conceived at a thickness of 3 mm (green curves in Figure 7) regarding the interaction between microwaves and this typology of RGO-reinforced composites. Specifically, the frequency was found to be partly related to the maximum RL at a certain thickness, as displayed on the quarter-wavelength cancellation model [37]. The matching equation is expressed below: where t m , λ, and f m are the matching thickness of the absorber, the wavelength, and the absorption frequency of the electromagnetic microwave. The calculated matching thickness and the peak frequency are plotted below the RL curves, where the red squares represent the specific thickness in the range of 1-4 mm corresponding to the peak frequency. In addition, the simulation of t m vs. frequency for the hollow Fe 3 O 4 @RGO composites are also depicted based on the above equation. The red squares appear on the simulation curve, suggesting that the absorber thickness could be tailored to design the microwave absorption materials.
The Z value of hollow Fe3O4@0.125 g RGO composites (Figure 7c) is almost equal to 1. Figure 7 reveals the amount of electromagnetic waves entering into the absorber and the good impedance matching in comparison with hollow Fe3O4@0.5 g RGO and hollow Fe3O4@0.25 g RGO composites. Furthermore, magnetic loss is also closely correlated with natural resonance, eddy current resonance, hysteresis loss, and domain wall displacement. Generally, the eddy current coefficient (C0) is almost a constant in accordance with the following equation when only magnetic loss is caused by the eddy current at the frequency range of 2-18 GHz.
As shown in Figure 8, the C0-f curve presents an obviously declining trend at the frequency range of 2-18 GHz, which means that the attenuation of electromagnetic microwaves is not caused by eddy current resonance. However, the hysteresis loss is always inoperative in the weak field, while the domain wall resonance loss commonly appears in the MHz frequency range [39]. Hence, a conclusion could be drawn that the magnetic loss is mainly produced by natural resonance. Furthermore, better impedance matching maximizes the penetration of electromagnetic waves into the material and reduces the direct reflection on the surface of the material [38]. When the value of the impedance matching characteristics (Z) is equal or close to 1, zero reflection is achieved, which is expressed as the following equation: The Z value of hollow Fe 3 O 4 @0.125 g RGO composites (Figure 7c) is almost equal to 1. Figure 7 reveals the amount of electromagnetic waves entering into the absorber and the good impedance matching in comparison with hollow Fe 3 O 4 @0.5 g RGO and hollow Fe 3 O 4 @0.25 g RGO composites.
Furthermore, magnetic loss is also closely correlated with natural resonance, eddy current resonance, hysteresis loss, and domain wall displacement. Generally, the eddy current coefficient (C 0 ) is almost a constant in accordance with the following equation when only magnetic loss is caused by the eddy current at the frequency range of 2-18 GHz.
As shown in Figure 8, the C 0 -f curve presents an obviously declining trend at the frequency range of 2-18 GHz, which means that the attenuation of electromagnetic microwaves is not caused by eddy current resonance. However, the hysteresis loss is always inoperative in the weak field, while the domain wall resonance loss commonly appears in the MHz frequency range [39]. Hence, a conclusion could be drawn that the magnetic loss is mainly produced by natural resonance.  Table 1 lists some reported microwave absorption composites of the representative Fe3O4 material-based, graphene material-based, and Fe3O4@0.125 g RGO composite prepared in this work. Notably, the hollow Fe3O4@0.125 g RGO composites not only had a wide effective absorption bandwidth, but also displayed a promising negative RL value owing to this special hollow structure. Furthermore, the as-fabricated hollow Fe3O4@RGO properties could enhance the absorption performance and suit the requirements of ideal MAMs. The microwave absorption performance of the hollow Fe3O4@RGO could be ascribed to the following reasons. (a) The outstanding magnetic hollow Fe3O4 sphere would induce a magnetic loss. (b) The main loss mechanism derives from dielectric loss rather than magnetic loss. Normally, dielectric loss is related to electronic dipole polarization and interfacial polarization. Firstly, electron migration, such as Fe 2+ and Fe 3+ ions, would induce dipole polarization in the composites. Secondly, the different neighboring phases, including dielectric constant and conductivity, have a significant effect on interfacial polarization. For the hollow Fe3O4@RGO composites, many charge carriers accumulated onto the interfaces, as shown in the TEM images, which aggravated the dielectric loss. (c) The interaction of dielectric loss and magnetic loss is also an important factor for the improvement of the microwave absorption performance; as a result, the RGO nanosheets modified by magnetic hollow Fe3O4 spheres have excellent absorption abilities.

Conclusions
In summary, hollow Fe3O4@RGO composites were successfully synthesized by the one-step  Table 1 lists some reported microwave absorption composites of the representative Fe 3 O 4 material-based, graphene material-based, and Fe 3 O 4 @0.125 g RGO composite prepared in this work. Notably, the hollow Fe 3 O 4 @0.125 g RGO composites not only had a wide effective absorption bandwidth, but also displayed a promising negative RL value owing to this special hollow structure. Furthermore, the as-fabricated hollow Fe 3 O 4 @RGO properties could enhance the absorption performance and suit the requirements of ideal MAMs. The microwave absorption performance of the hollow Fe 3 O 4 @RGO could be ascribed to the following reasons. (a) The outstanding magnetic hollow Fe 3 O 4 sphere would induce a magnetic loss. (b) The main loss mechanism derives from dielectric loss rather than magnetic loss. Normally, dielectric loss is related to electronic dipole polarization and interfacial polarization. Firstly, electron migration, such as Fe 2+ and Fe 3+ ions, would induce dipole polarization in the composites. Secondly, the different neighboring phases, including dielectric constant and conductivity, have a significant effect on interfacial polarization. For the hollow Fe 3 O 4 @RGO composites, many charge carriers accumulated onto the interfaces, as shown in the TEM images, which aggravated the dielectric loss. (c) The interaction of dielectric loss and magnetic loss is also an important factor for the improvement of the microwave absorption performance; as a result, the RGO nanosheets modified by magnetic hollow Fe 3 O 4 spheres have excellent absorption abilities.

Conclusions
In summary, hollow Fe 3 O 4 @RGO composites were successfully synthesized by the one-step solvothermal reaction route, which exhibited excellent microwave absorption properties in terms of the maximum RL value and the absorption bandwidth in the range of 2-18 GHz through tuning the hollow [Fe 3 O 4 ]/[RGO] ratio. The surfaces of RGO were densely covered with hollow Fe 3 O 4 spheres 130 nm in diameter, with a corresponding shell thickness of~35 nm, as shown by the SEM and TEM images. Specifically, the maximum RL could reach −41.89 dB at 6.7 GHz and the absorption bandwidth below −10 dB was as wide as 4.2 GHz at a thickness of 2.5 mm. The excellent performance of the novel composites could be ascribed to the strong magnetic loss and favorable impedance matching, and such microwave absorbers with strong absorption and a wide frequency band have shown great application potential in military and commercial fields.