Preparation of rGO@Fe₃O₄ nanocomposite and its application to enhance the thermal conductivity of epoxy resin

Abstract

In this study, we report a simple method to improve the thermal conductivity of epoxy resin by using new magnetic composites as fillers. The rGO@Fe₃O₄ nanocomposite has been prepared by a solvothermal method, and its morphology and chemical structure were characterized and analyzed by various characterization methods. Afterwards, the rGO@Fe₃O₄/EP composite material was obtained in an external magnetic field, in which the rGO@Fe₃O₄ is uniformly dispersed in the epoxy resin matrix, arranged along the direction of the magnetic field. In addition, the orientation of rGO@Fe₃O₄ increases with the magnetic field intensity. After doping 30% (wt) rGO@Fe₃O₄ into epoxy resin and curing under a 500 Gs magnetic field, the rGO@Fe₃O₄/EP composite material is anisotropic and has a higher thermal conductivity (increased by 196.60%) parallel to the direction of the magnetic field compared to a pure ring oxygen resin.

---

1. Introduction

In recent years, graphene/polymer composite materials have attracted much research interest. Graphene is a two-dimensional material with a honeycomb lattice structure formed by a single layer of carbon atoms through sp² hybridization, and its thickness is only 0.335 nm which is the size of one carbon atom.^1,2 Due to its unique microstructure, graphene has many excellent physical properties, such as the highest intrinsic strength (∼130 GPa),³ the highest carrier mobility (up to 200 000 cm² V⁻¹ s⁻¹),⁴ and the highest room temperature thermal conductivity (∼5000 W m⁻¹ K⁻¹, in-plane thermal conductivity), etc.⁵ Compared with conventional composite materials or pure polymer materials, adding a small amount of graphene often makes the composite material show a huge improvement in mechanical,⁶ electrochemical,⁷ or thermal properties.⁸ At present, there are many studies on graphene composite materials, such as adding graphene to epoxy resin to improve the thermal conductivity of epoxy resin, and adding graphene to metallic copper to improve the thermal conductivity of copper.^9–11 The results show that the thermal conductivity of the composites prepared by adding graphene directly into the matrix does not increase significantly. First of all, the arrangement of graphene in the matrix is extremely inconsistent, and the heat transfer direction along each layer of graphene is inconsistent. In addition, the thermal conductivity of graphene in the vertical laminar direction is only 30–50 W m⁻¹ K⁻¹, which is quite different from that in the laminar direction. This characteristic will make the heat mainly transfer between graphene lamellae, thus limiting the overall performance of graphene composites.^12,13 Therefore, the arrangement of graphene in the matrix and the low thermal conductivity in the vertical plane direction must be considered when studying graphene composite materials.^14,15

Magnetic nanoparticles are composed of transition metal or alloy nanoparticles embedded in nonmetallic media. They have the dual characteristics of magnetic particles and nanomaterials, and have attracted wide attention in basic research and technical applications. By loading magnetic nanoparticles on the surface of graphene to prepare composite materials, it not only inherits the good properties of graphene, but also gives full play to the advantages of magnetic nanomaterials such as superparamagnetism, high catalytic activity and good biocompatibility,^16–18 which makes the materials have application prospects in many aspects. Cakmak et al.¹⁹ synthesized rGO–Fe₃O₄–TiO₂ ternary nanocomposites and dispersed them into ethylene glycol to prepare nanofluids. It was found that the rGO–Fe₃O₄–TiO₂/EG nanofluids had stable thermal conductivity at different temperatures and could be used for heat transfer applications. Suo et al.²⁰ studied the sensitivity of the composite Fe₃O₄–GO@SiO₂ for the determination of trace metals in environmental water samples, and the test showed that seven heavy metal ions in different environmental water samples could be determined simultaneously, and the detection limit of the target analyte was within the range of 2.023–13.810 ng L⁻¹ under the optimal conditions. Zhang et al.²¹ prepared Fe₃O₄@C/rGO nanocomposite with three-layer structure and studied its microwave absorption performance. The synergistic effect of Fe₃O₄ and carbon materials is beneficial to the composites to obtain good impedance matching characteristics and strong electromagnetic attenuation. The preparation method of magnetic graphene and the selection of magnetic particles are very important to its structure and properties, and the appropriate preparation process should be selected according to the basic properties of the product or the application target. Common synthesis methods include hydrothermal and solvothermal method, chemical graft method, chemical co-precipitation method, microwave-assisted synthesis method, and so on.²² Barai et al.²³ successfully synthesized rGO–Fe₃O₄ nanocomposites through ultrasound-assisted co-precipitation method, and prepared rGO–Fe₃O₄-based nanofluids under the action of ultrasound. When the concentration of rGO–Fe₃O₄ was 0.2 vol%, the thermal conductivity of the nanofluids increased by 83.44% at 40 °C. Sundar et al.²⁴ synthesized rGO–Fe₃O₄–TiO₂ ternary nanocomposites using sol–gel technology, and prepared stable ternary hybrid nanofluids based on ethylene glycol, and studied the effects of various conditions on the physical properties, heat transfer performance and friction coefficient of the nanofluids. Yan et al.²⁵ prepared magnetic GNS–Fe₃O₄ hybrid compounds as filler in epoxy resin by co-precipitation method, which were highly aligned in the epoxy resin through magnetic field action. The results showed that the thermal conductivity of epoxy resin/GNS–Fe₃O₄ composites in parallel magnetic field orientation increased by 139% and 41%, respectively, compared with those of composite epoxy resin with magnetic graphene random distribution and in vertical orientation. After composite with magnetic nanoparticles, graphene is superparamagnetic and can be arranged in the polymer matrix under the action of magnetic field to form an effective heat conduction channel, which is more conducive to the improvement of the thermal conductivity of the matrix.

With the rapid development of science and technology, in certain occasions, higher requirements are put forward for the thermal conductivity of traditional polymer materials. As a commonly used thermosetting resin matrix, epoxy resin has the characteristics of strong bonding ability, excellent electrical insulation, stable chemistry, and low curing shrinkage.²⁶ However, the epoxy resin will be brittle after cross-linking and curing, with poor heat resistance, impact resistance, and crack resistance.²⁷ Therefore, graphene is a good candidate to be used as a reinforcing agent to be filled into the epoxy resin system to improve various properties of the epoxy resin for obtaining an epoxy resin-based composite material with excellent comprehensive properties.

Combining the advantages of the above materials, in this study, we synthesized rGO@Fe₃O₄ nanocomposites in one step through a simple solvothermal method, then blended rGO@Fe₃O₄ as a filler with epoxy resin, and cured in an external magnetic field to obtain rGO@Fe₃O₄/EP composite materials. After that, scanning electron microscopy (SEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), Raman, thermal conductivity, and other characterizations were employed to analyze the structure, micro-morphology, and thermal conductivity of the sample.

2. Experimental

2.1 Chemical materials

Flake graphite (325 mesh) was provided by Unigram Carbon Graphene Materials Co., Ltd (Hebei, China; UCGM). Hexahydrate and ferric chloride (AR, ≥99.0%), ethylene glycol (AR), sodium acetate (NaAc, AR, ≥99.0%), tributyl phosphate (AR, 98.5%), sulfuric acid (AR, 95–98%), phosphoric acid (AR, 85%), hydrogen peroxide (30%), potassium permanganate (AR, ≥99.5%) were purchased from Aladdin Chemistry Co. Ltd (Shanghai, China). Polyethylene glycol (M_w = 2000) and epoxy resin (E-51) were purchased from Macklin Biochemical Co., Ltd (Shanghai, China). Polyamide resin (V125 resin, curing agent of EP with amino value of 340 mg KOH per g) was purchased from Danbao Resin Co., Ltd (Anhui, China).

2.2 Preparation of rGO@Fe₃O₄ and rGO@Fe₃O₄/EP composites

Graphene oxide (GO) was synthesized from flake graphite by an improved Hummers method.²⁸ Using the synthesized GO and FeCl₃·6H₂O as raw materials, rGO@Fe₃O₄ nanocomposites were prepared by a solvothermal method which can complete the deposition of Fe₃O₄ nanoparticles on graphene sheets while reducing GO. Firstly, 60 mg GO was added to 30 ml of ethylene glycol and dispersed by ultrasonic for 0.5 hours to form a stable GO solution. Then, 1.2 g FeCl₃·6H₂O, 3.6 g NaAc, and 1 g polyethylene glycol were added into the GO solution in sequence. After that, the solution was stirred for 0.5 hours with ultrasonic vibration. Hence, the chemicals can be fully dissolved faster. The mixed solution was transferred to a polytetrafluoro-ethylene reactor and kept at 200 °C for 12 hours. When the solution was cooled down to room temperature, the resulting product was collected by a magnet adsorption method. Finally, the collected product was thoroughly washed with deionized water and absolute ethanol for several times, dried in a vacuum drying oven at 60 °C for 12 hours, and ground into powder to obtain rGO@Fe₃O₄ nanocomposite materials.

As shown in Fig. 1, the preparation process of the rGO@Fe₃O₄/EP composite is described as follows. Firstly, an appropriate amount of rGO@Fe₃O₄ was weighed according to 5% of the mass of epoxy resin, added into absolute ethanol, and ultrasonically dispersed for 0.5 hours to form a stable dispersion. Then, the rGO@Fe₃O₄/ethanol dispersion was added into the epoxy resin at 80 °C, stirred ultrasonically for 10 minutes, and then heated at 80 °C for 2 hours to remove the solvent. Tributyl phosphate and polyamide resin were added into the above mixed liquid as accelerators. The liquid was stirred evenly and kept in a vacuum oven to remove air bubbles. Finally, the liquid was cured under a specific magnetic field (0 Gs, 100 Gs, 500 Gs), and the curing process were carried out at room temperature for 24 h, at 100 °C for 1 hour, and at 150 °C for 2 hours sequentially.

Fig. 1 Synthesis of rGO@Fe₃O₄ and preparation of rGO@Fe₃O₄/EP.

2.3 Characterization

The morphology and microstructure of the GO and rGO@Fe₃O₄ samples were characterized by SEM (Zeiss Merlin Compact, Carl Zeiss AG, Germany). XPS (ESCALAB 250XI, Thermo Fisher Scientific Inc., USA) was used to determine the elemental composition and the functional group information of the samples, in which Al Kα was used as the X-ray source and carbon C 1s as the corrected peak position (E_{C 1s} = 284.6 eV). FT-IR (Nicolet 6700, Thermo Fisher Scientific Inc., USA) was used to determine the chemical functional group information of the samples. The measurement range was 500–4000 cm⁻¹, and the resolution was 2 cm⁻¹. The crystal structure of the samples was analyzed by an XRD (Rigaku D/MAX-1200 diffractometer, Tokyo, Japan) operating at 40 kV and 100 mA. The Cu-Kα radiation (wavelength λ = 0.15406 nm) was employed to obtain the diffraction patterns in a 2θ range of 5°–80° with a scanning speed of 5° min⁻¹. The internal structure of the sample was characterized and analyzed by a DXR Microscope Raman spectrometer (DXR, Thermo Fisher Scientific Inc., Waltham, MA, USA) with a laser wavelength of 532 nm and a scanning range of 100–3500 cm⁻¹.

2.4 Performance test

The magnetic property of the sample was analyzed by a vibrating sample magnetometer (VSM, PPMS-9, Quantum Design Inc., USA). The thermogravimetric analysis (TGA) test was performed using a thermal analyzer (SDT Q600, TA Instruments, USA) from 30 °C to 800 °C with a heating rate of 10 °C min⁻¹ under a continuous nitrogen flow. The laser thermal conductivity meter (NETZSCH LFA 467, Germany) was used to measure the in-plane thermal diffusivity. The thermal conductivity (λ) is calculated as follows:

λ = αρc_p

where α and ρ are the thermal diffusivity and density of the composite material, which are measured at room temperature; c_p is the specific heat of the composite material, which can be measured from −10 °C to 50 °C with a heating rate of 10 °C min⁻¹ using a differential thermal scanning calorimeter (DSC Q2000, TA Instruments, USA) by a sapphire method.

3. Results and discussion

3.1 Characterization of GO, rGO, and rGO@Fe₃O₄ nanocomposites

As shown in Fig. 2, the morphology and microstructure of the prepared GO and rGO@Fe₃O₄ nanocomposites can be revealed by the SEM images. In Fig. 2a and b, GO presents a typical corrugated sheet structure with a smooth surface and a thin sheet thickness. In Fig. 2c and d, Fe₃O₄ nanoparticles are uniformly and densely attached to the rGO sheet, acting as a barrier and reducing the force between the rGO sheets.

Fig. 2 (a and b) SEM images of GO; (c and d) SEM images of rGO@Fe₃O₄ nanocomposite.

Raman spectroscopy was used to further characterize the structural characteristics of GO and rGO@Fe₃O₄ nanocomposites (Fig. 3a and b). In the Raman spectrum of GO, there are obvious peaks at 1351.80 cm⁻¹ and 1607.32 cm⁻¹, corresponding to the D band and G band, respectively. The defects in the structure can be determined by the intensity ratio of I_D to I_G. The I_D/I_G peak intensity ratios of GO and rGO@Fe₃O₄ are 0.88 and 1.57, respectively. Compared with GO, the G band of rGO@Fe₃O₄ is shifted closer to graphene (1580.4 cm⁻¹),²⁹ indicating that GO was successfully reduced to rGO. In addition, there are peaks at 216.71 cm⁻¹, 277.56 cm⁻¹, 385.60 cm⁻¹, and 582.30 cm⁻¹ in the Raman spectrum of rGO@Fe₃O₄ nanocomposites, which are typical peaks of Fe₃O₄.³⁰ The results of Raman spectroscopy proved that rGO@Fe₃O₄ nanocomposites were successfully synthesized.

Fig. 3 Raman spectra of GO (a) and rGO@Fe₃O₄ (b).

Fig. 4 shows the infrared spectra of GO, rGO, Fe₃O₄, and rGO@Fe₃O₄ nanocomposites. The infrared spectrum of GO has obvious characteristic peaks at 3401.82 cm⁻¹, 1743.33 cm⁻¹, 1238.08 cm⁻¹, and 1060.66 cm⁻¹, corresponding to –OH, C=O, C–O, and C–O–C functional groups, respectively. The peak at 1629.55 cm⁻¹ is due to the skeleton vibration of aromatic C=C in GO. It shows that there are a lot of oxygen-containing functional groups on the surface of GO. In the infrared spectra of rGO and rGO@Fe₃O₄ nanocomposites, several characteristic peaks of GO are significantly weakened or disappeared, indicating that most of the oxygen-containing groups have been reduced and removed. Compared with rGO, rGO@Fe₃O₄ nanocomposite has a characteristic peak at 563.11 cm⁻¹, which is caused by the stretching vibration of the Fe–O bond. The above-mentioned FTIR analysis shows that the GO in the rGO@Fe₃O₄ nanocomposite has been reduced, and it is confirmed that Fe₃O₄ nanoparticles are present on the reduced graphene sheet.

Fig. 4 FTIR spectrum of GO, rGO, Fe₃O₄ and rGO@Fe₃O₄.

XPS technology was used to analyze the chemical composition and functional groups of GO and rGO@Fe₃O₄ nanocomposites. For GO, the C 1s spectrum (Fig. 5b) corresponds to four carbon component peaks with different binding energies: C–C (284.06 eV), C=C (284.57 eV), C–O (285.95 eV), and C=O (288.97 eV), indicating that GO is rich in oxygen functional groups.³¹ In the C 1s spectrum of the rGO@Fe₃O₄ nanocomposite (Fig. 5c), the relative intensity of the oxygen-containing functional groups decreases significantly, indicating that GO is effectively deoxidized and reduced to rGO. Fig. 5d shows the Fe 2p spectrum of the composite material, there are two deconvolution peaks at 723.4 eV and 709.8 eV, corresponding to the spin orbit peaks of Fe 2p_1/2 and Fe 2p_3/2 of Fe₃O₄, respectively.³² It is proved that mixed oxides of Fe(III) and Fe(II) are formed in the composite. There are three split peaks in the O 1s spectrum of the composite material (Fig. 5e), which may be caused by the chemical bonds of different oxygen elements in Fe₃O₄ (O–Fe) and rGO (O–C). The XPS results showed that rGO@Fe₃O₄ nanocomposite was successfully synthesized, and Fe₃O₄ can be combined with rGO well.

Fig. 5 (a) XPS spectra with survey scan of GO and rGO@Fe₃O₄ nanocomposites; (b and c) C 1s XPS spectra of GO and rGO@Fe₃O₄ nanocomposites, (d and e) Fe 2p XPS spectra of rGO@Fe₃O₄ nanocomposites.

Fig. 6a shows the XRD patterns of GO, rGO and rGO@Fe₃O₄ composites. GO has a sharp diffraction peak at 2θ = 10.6°, corresponding to the (100) crystal plane of GO. According to the calculation of Bragg equation,³³ the crystal plane spacing is about 0.87 nm. It shows that the carbon structure has lattice distortion, which is mainly because a large number of functional groups are inserted into the graphite lamellar in the oxidation process, which changes the lump structure of the graphite and makes the GO of the single atomic layer stripped off, thus increasing the specific surface area and adsorption site of the material. rGO has a (002) diffraction peak at 2θ = 23.36°, and the crystal plane spacing is 0.38 nm, which is much smaller than that of GO. It is due to the reduction of oxygen-containing functional groups and the corresponding reduction of layer spacing after solvothermal reduction of GO. It can be seen from the figure that the main peaks of the XRD spectra of rGO@Fe₃O₄ composites and Fe₃O₄ nanoparticles are in the same position. The diffraction peaks at 2θ = 18.3°, 30.14°, 35.48°, 43.08°, 53.54°, 56.94° and 62.64° correspond to the crystal planes of (111), (220), (311), (400), (422), (511) and (440), respectively. It is consistent with the standard XRD data of cubic inverse spinel structure of magnetite (JCPDS No. 19-0629),^34,35 indicating the presence of magnetite Fe₃O₄ in the composite material. According to Sherle's formula,³⁶ the grain size of Fe₃O₄ nanoparticles in rGO@Fe₃O₄ was calculated to be about 71.70 nm. As can be seen from Fig. 6b, the rGO@Fe₃O₄ composite material has a low and broad diffraction peak at 2θ = 24.22°, which indicates that GO has become a smaller and thinner rGO, and that the rGO@Fe₃O₄ composite material has been successfully prepared. However, compared with the characteristic peak of pure rGO, the diffraction peak of rGO in rGO@Fe₃O₄ is relatively weak. This is because a large number of Fe₃O₄ nanoparticles are adsorbed on the surface of graphene, which effectively impedes the interlamellar accumulation of graphene and makes it difficult to form a graphene-like layered accumulation structure, resulting in an unobtrusive diffraction peak.³⁷

Fig. 6 (a) XRD patterns of GO, rGO, Fe₃O₄ and rGO@Fe₃O₄ composites; (b) magnified XRD pattern of rGO@Fe₃O₄ composites.

The magnetic properties of GO, Fe₃O₄, and rGO@Fe₃O₄ nanocomposites were measured by a vibrating sample magnetometer. The hysteresis loops of Fe₃O₄ and rGO@Fe₃O₄ nanocomposites (Fig. 7) both show an S shape, and there is no obvious coercivity, indicating that both of them have superparamagnetic properties.³⁸ Although the saturation magnetization of rGO@Fe₃O₄ nanocomposite (66.39 emu g⁻¹) is slightly lower than that of Fe₃O₄ (74.39 emu g⁻¹), there is still enough saturation magnetization to achieve magnetic adsorption. The magnetic saturation intensity of rGO@Fe₃O₄/EP composite is small (2.55 emu g⁻¹), which may be due to the small doping amount of magnetic graphene.

Fig. 7 VSM patterns of rGO, Fe₃O₄, rGO@Fe₃O₄ nanocomposites and rGO@Fe₃O₄/EP composites.

3.2 Performance analysis of rGO@Fe₃O₄/EP nanocomposites

Fig. 8 shows the SEM images of the fracture surface of the rGO@Fe₃O₄/epoxy resin composite. After being oriented by a magnetic field, rGO@Fe₃O₄ exhibits an orderly arranged microstructure in the epoxy resin. This microstructure is conducive to the formation of heat transmission paths and the conduction of interface heat.

Fig. 8 Cross-sectional SEM image of rGO@Fe₃O₄/EP composite material (a) without magnetic field induction; (b) parallel to the direction of the magnetic field.

The thermal conductivity of pure epoxy resin and rGO@Fe₃O₄/EP composite materials was measured with a laser thermal conductivity meter. Compared with the pure epoxy resin (0.206 W m⁻¹ K⁻¹), the thermal conductivity of the composite (0.224 W m⁻¹ K⁻¹) increases by only 8.43% (Fig. 9) when the composite was cured without external magnetic field. After being oriented by an external magnetic field, the composite material exhibits excellent anisotropic thermal conductivity. When oriented under a 100 Gs magnetic field, the thermal conductivity of the in-plane parallel magnetic field direction is 0.360 W m⁻¹ K⁻¹. When the magnetic field intensity is 500 Gs, the thermal conductivity of the in-plane parallel magnetic field direction is as high as 0.449 W m⁻¹ K⁻¹. It can be seen that the thermal conductivity will increase slightly with increasing the magnetic field. Compared with pure epoxy resin, the thermal conductivity of in-plane parallel magnetic field increases by 117.60% while that of vertical magnetic field (0.274 W m⁻¹ K⁻¹) only increases by 33.01% at 500 Gs, which indicates that the composite has anisotropic thermal conductivity. When the magnetic field induces the directional arrangement of rGO@Fe₃O₄ in the epoxy resin matrix, it can form a heat conduction path to facilitate the directional heat transfer, and effectively reduce the phonon scattering and thermal boundary resistance, so that the thermal conductivity along the magnetic field can be greatly improved.³⁹ It can be seen from Table 1 that the thermal conductivity of rGO@Fe₃O₄/EP composites increases with the increase of rGO@Fe₃O₄ content. When 30% rGO@Fe₃O₄ is doped as filler, the thermal conductivity is as high as 0.611 W m⁻¹ K⁻¹, an increase of 196.60% compared with pure epoxy resin.

Fig. 9 Thermal conductivity diagram of EP, rGO@Fe₃O₄/EP composite.

Table 1 Thermal conductivity of rGO@Fe₃O₄/EP composites with different rGO@Fe₃O₄ contents and magnetic field

Doping amount of rGO@Fe3O4/%	ρ/(g cm^−3)	Cp/(J g^−1 K^−1)	α/(mm^2 s^−1)	λ/(W m^−1 K^−1)
0	—	—	—	0.206
5%	—	—	—	0.224
5% (100 Gs∥)	1.153	1.882	0.217	0.360
5% (100 Gs⊥)	1.133	1.882	0.178	0.302
5% (500 Gs∥)	0.965	1.882	0.247	0.449
5% (500 Gs⊥)	0.950	1.882	0.153	0.274
15% (500 Gs∥)	0.920	2.322	0.253	0.541
30% (500 Gs∥)	0.934	2.353	0.278	0.611

---

Fig. 10 shows the thermal performance of EP, rGO@Fe₃O₄, and rGO@Fe₃O₄/EP composites. It can be seen from the TG curves that the mass loss of rGO@Fe₃O₄ is very small, which is mainly attributed to the vaporization of the water absorbed on the surface of the material and part decomposition of carbonaceous materials. The remaining rGO and Fe₃O₄ will not undergo oxidation reaction under the protective effect of nitrogen, and the structure is stable and does not change. Pure epoxy and rGO@Fe₃O₄/EP composites begin to lose weight at about 150 °C, and have been basically completely decompose at 600 °C, the coke rate is 2.17% and 7.22%, respectively. The improvement of coke yield of rGO@Fe₃O₄/EP composites is due to the existence of stable rGO and Fe₃O₄ nanoparticles. Compared with pure epoxy, the T_max of rGO@Fe₃O₄/EP composite increases by 5.03 °C, and the maximum mass loss rate (R_max) is reduced by 24.16%. In addition, the heat resistance index (T_HRI) is often used to evaluate the thermal stability of the cured resin. As can be seen from Table 2, the heat resistance index (159.96 °C) of the rGO@Fe₃O₄/EP composites is slightly lower than that of pure epoxy resin (170.87 °C), which may be due to the enhancement of heat conduction leading to rapid transfer of external heat to the interior. In general, the TGA results show that rGO@Fe₃O₄/EP composites have excellent thermal stability.

Fig. 10 (a) TG and (b) DTG curves of EP, rGO@Fe₃O₄ nanocomposites, and rGO@Fe₃O₄ composite.

Table 2 The detailed thermal decomposition parameters of EP, rGO@Fe₃O₄ and the composite

Samples	T5% (°C)	T30% (°C)	Tmax (°C)	Rmax (% min^−1)	THRI^a (°C)	Mass residual at 800 °C (%)
a The heat-resistance index of the sample is calculated by the following equation: THRI = 0.49 × [T5% + 0.6 × (T30% − T5%)].^40–43
EP	318.83	368.63	374.98	−12.13	170.87	2.17
rGO@Fe3O4	360.05	—	—	−0.15	—	90.08
rGO@Fe3O4/EP	261.01	370.09	379.95	−9.77	159.96	7.22

---

4. Conclusions

The magnetic rGO@Fe₃O₄ was prepared in one step by a solvothermal method. Fe₃O₄ nanoparticles were deposited on the surface of rGO to form a synergistic effect, which could make the rGO sheets align along the magnetic field direction. The rGO@Fe₃O₄/EP nanocomposite was prepared by applying an external magnetic field during the curing process. In the epoxy resin matrix, rGO@Fe₃O₄ was arranged in an orderly manner in the direction of the magnetic field, so that the thermal conductivity of the EP nanocomposite in the direction of the magnetic field was significantly enhanced, which made the nanocomposite material show obvious anisotropic thermal conductivity. This work provided some research foundation and application value for improving the thermal conductivity and anisotropy of epoxy resin.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. R. R. Haering, Can. J. Phys., 1958, 36, 352–362.
2. X. A. Li, B. L. Wang and Z. R. Liu, Mater. Rep., 2012, 26, 61–65.
3. C. Lee, X. Wei, J. W. Kysar and J. Hone, Science, 2008, 321, 85–388.
4. S. V. Morozov, K. S. Novoselov, M. I. Katsnelson, F. Schedin, D. C. Elias, J. A. Jaszczak and A. K. Geim, Phys. Rev. Lett., 2007, 100, 016602.
5. A. A. Balandin, S. Ghosh, W. Bao, I. Calizo, F. Miao and C. N. Lau, Nano Lett., 2008, 8, 902.
6. D. G. Papageorgiou, I. A. Kinloch and R. J. Young, Prog. Mater. Sci., 2017, 90, 75–127.
7. D. Verma, P. C. Gope, A. Shandilya and A. Gupta, Trans. Indian Inst. Met., 2014, 67, 803–816.
8. R. Bauld, W. Choi, P. Bazylewski, R. Divigalpitiya and G. Fanchini, J. Mater. Chem. C, 2018, 6, 2901–2914.
9. J. Kim, B. S. Yim, J. M. Kim and J. Kim, Microelectron. Reliab., 2012, 52, 595–602.
10. S. H. Song, K. H. Park, B. H. Kim and Y. W. Choi, Adv. Mater., 2013, 25, 732–737.
11. Y. J. Wan, L. C. Tang, L. X. Gong, Y. B. Li, L. B. Wu, J. X. Jiang and G. Q. Lai, Carbon, 2014, 69, 467–480.
12. L. C. Tang, Y. J. Wan, D. Yan, Y. B. Pei and L. Zhao, Carbon, 2013, 60, 16–27.
13. Y. F. Li, D. Datta, S. Li, Z. Li and V. B. Shenoy, Comput. Mater. Sci., 2014, 93, 68–73.
14. B. W. Zhang, B. W. Li and C. S. Xie, Mater. Rep., 2007, 21, 171–178.
15. S. X. Yang, W. C. Jiao, Z. M. Chu and C. W. Zhang, Fiber Reinf. Plast./Compos., 2019, 92–100.
16. A. Akbarzadeh, M. Samiei and S. Davaran, Nanoscale Res. Lett., 2012, 7, 144.
17. J. E. Lee, B. Tsedenbal, B. H. Koo and S. H. Huh, Korean J. Mater. Res., 2020, 30, 589–594.
18. E. Rozhina, A. Danilushkina, F. Akhatova, R. Fakhrullin, A. Rozhin and S. Batasheva, J. Biotechnol., 2021, 325, 25–34.
19. N. K. Cakmak, Z. Said, S. Sundar and A. K. Tiwar, Powder Technol., 2020, 372, 235–245.
20. L. Z. Suo, X. Y. Dong, X. Gao, J. F. Xu, X. M. Lu and L. S. Zhao, Microchem. J., 2019, 149, 104039.
21. H. X. Zhang, Z. R. Jia, A. L. Feng, Z. H. Zhou and L. Chen, Composites, Part B, 2020, 199, 108261.
22. R. Greta, Y. Zhou, K. Roman and F. Conrad, Angew. Chem., Int. Ed., 2011, 50, 826–859.
23. D. P. Barai, B. A. Bhanvase and V. K. Saharan, Ind. Eng. Chem. Res., 2019, 58, 8349–8369.
24. S. L. Sundar, K. V. V. C. Mouli, Z. Said and A. C. M. Sousa, J. Therm. Sci. Eng. Appl., 2021, 13, 1–45.
25. H. Y. Yan, Y. X. Tang, W. Long and Y. F. Li, J. Mater. Sci., 2014, 49, 5256–5264.
26. Y. N. Xie, H. Lei and Q. Shi, Eng. Plast. Appl., 2018, 46, 143–147.
27. X. B. Hong, K. Xie, Y. Pan and J. Y. Xiao, Materials Reports, 2005, 19, 44–48.
28. E. L. K. Chng and M. Pumera, Chem.–Eur. J., 2013, 19, 8227–8235.
29. B. Shen, D. Lu, W. Zhai and W. Zhang, J. Mater. Chem. C, 2012, 1, 50–53.
30. H. W. Wang, T. Liang, D. M. Xu, K. Liao, R. Wang and C. J. Yan, Nanoscale, 2018, 10, 17814–17823.
31. J. Chen, B. W. Yao, C. Li and G. Q. Shi, Carbon, 2013, 64, 225–229.
32. B. J. Tan, K. J. Klabunde and P. M. A. Sherwood, Chem. Mater., 2002, 2, 186–191.
33. H. Cole, J. Appl. Crystallogr., 1970, 3, 405–406.
34. M. Zong, Y. Huang, Y. Zhao, P. B. Liu, Y. Wang and Q. F. Wang, Mater. Lett., 2013, 106, 22–25.
35. Y. Yang, M. Li, Y. P. Wu, T. Wang, J. Ding, B. Y. Zong, Z. H. Yang and J. M. Xue, Nanoscale, 2016, 8, 15989–15998.
36. A. L. Patterson, Phys. Rev., 1939, 56, 978–982.
37. M. Khalaj, A. Sedghi, H. N. Miankushki and S. Z. Golkhatmi, Energy, 2019, 188, 116088.
38. B. J. Li, H. Q. Cao, J. Shao and J. H. Warner, J. Mater. Chem., 2011, 21, 5069–5075.
39. K. M. F. Shahil and A. A. Balandin, Nano Lett., 2012, 12, 861–867.
40. R. S. Lehrle and R. J. Williams, Macromolecules, 1994, 27, 3782–3789.
41. J. Deng, X. Q. Liu, C. Li and J. Zhu, RSC Adv., 2015, 5, 15930–15939.
42. L. Tang, J. Dang, M. K. He and J. Y. Li, Compos. Sci. Technol., 2018, 169, 120–126.
43. Y. Tang, W. C. Dong, L. Tang, Y. K. Zhang, J. Kong and J. W. Gu, Compos. Commun., 2018, 8, 36–41.

---