Graphene/poly(vinylidene fluoride) dielectric composites with polydopamine as interface layers

2.1 Materials

Poly(vinylidene fluoride) (PVDF) with a melt flow index of 10 g/min was purchased from Shanghai 3F New Materials (China) and used as received. Graphite flakes were purchased from Beijing Nanopowder Science & Technology (China). Dopamine hydrochloride was obtained from Sinopharm Chemical Reagents (China). Other chemical reagents were of analytical grade purchased from Sinopharm Chemical Reagents (China) and used as received.

2.2 Preparation of rGO@PDOPA nanosheets

Firstly, graphene oxide (GO) was synthesized through a modified Hummers method from graphite flakes [16]. Afterwards, 100 mg GO was dissolved in 10 mm Tris-HCl (pH 8.5, 200 ml), followed by a sonication of 2 h at room temperature. Then, 400 mg of dopamine hydrochloride (2 mg/ml) was added into the above suspension and stirred for 30 min. The solution’s color changed to dark brown due to pH-induced oxidation. The PDOPA-coated GO (GO@PDOPA) nanosheets were centrifuged and rinsed with distilled water three times. Afterwards, an aqueous solution of hydrazine (0.4 g in 2 ml deionized water) was slowly added to the GO@PDOPA dispersion (1 mg ml^-1, 100 ml), keeping stirring and refluxing for 6 h at 85°C. After multiple washes with distilled water and then drying in vacuum for 24 h at 60°C, rGO@PDOPA nanosheets could be obtained.

2.3 Preparation of rGO@PDOPA/PVDF composites

To fabricate the rGO@PDOPA/PVDF composites, rGO@PDOPA nanosheets were first dispersed in dimethylformamide with the aid of sonication at ambient temperature. Three hundred milligrams of PVDF powder was added to the above well-mixed solution under stirring, and the mixture was stirred vigorously overnight to give a homogeneous dispersion, which was then drop-casted on glass plates. The cast dispersions were first dried at 60°C for 6 h and subsequently peeled off the glass plates for further drying in vacuum at 80°C for 48 h. The solution-cast films were then hot pressed at 180°C and 10 MPa for 10 min to ensure the formation of uniform films around 0.15 mm in thickness and then cooled down to room temperature while maintaining the pressure.

2.4 Characterization

Infrared measurements were made on a Fourier transform infrared spectrometer (FTIR, Thermo Nicolet Nexus, USA) using KBr pellets. Thermogravimetric analysis (TGA) measurements were performed at a rate of 10°C min^-1 to 800°C with a simultaneous thermal analyzer (Netzsch, STA 499C, Germany) under N₂ flow. UV-Vis spectra in the wavelength range of 200–800 nm were recorded using a Shimadzu UV-2550 spectrophotometer (Japan). Atomic force microscopy (AFM) was performed with a commercial AFM (DI Nanoscope IV, USA) equipped with a tapping probe. Transmission electron microscopy (TEM, Joel JEM-2001F, Japan) images were obtained by placing a few drops of the dispersion on a copper grid and evaporating them at room temperature prior to observation. The cryogenic fracture surface morphologies of the rGO@PDOPA/PVDF composite films were examined by scanning electron microscopy (SEM) on an ULTRA PLUS-43-13 scanning electron microscope (Germany). The dielectric properties of the rGO@PDOPA/PVDF composite films were determined using a Hioki 3532-50 LCR meter (Japan) at ambient temperature. The composite films were coated with a silver film on the top and bottom surfaces of the films.

3.1 Structural analysis of rGO@PDOPA hybrid nanosheets

Chemical characterizations of GO@PDOPA and rGO@PDOPA were carried out using FTIR spectroscopy, and the results are shown in Figure 1. The spectrum of PDOPA presents intense absorption peaks around 3426 cm^-1 from catechol -OH groups and 1434 cm^-1 from -NH₂ groups. The peaks at 1646 cm^-1 and ~1560 cm^-1 can be assigned to an amide carbonyl-stretching mode and N-H shearing vibration of the amide group, respectively. For GO@PDOPA, the peak at 1717 cm^-1 represents the stretching vibration of functional groups of C=O. For rGO@PDOPA, the peak from the carbonyl stretch at 1717 cm^-1 disappears, while the peak at 1434 cm^-1 attributed to -NH₂ group of PDOPA still remains after hydrazine reduction. The FTIR analysis validates the successful wrapping of PDOPA onto GO surface and suggests that the reducing agent hydrazine has no significant influence on the surface coating layer of PDOPA.

Figure 1:

FT-IR spectra of GO@PDOPA and rGO@PDOPA.

UV-Vis spectroscopy was utilized to monitor the reduction process of the hybrid nanosheets. As can be seen in Figure 2, the absorption peak of the GO@PDOPA dispersion at 225 nm red shifted to 270 nm after it was converted to rGO@PDOPA, while no visible peak could be detected within this region in the UV-Vis curve of PDOPA, suggesting the restoration of electronic conjugation in the basal plane of graphene upon reduction.

Figure 2:

UV-vis spectra of PDOPA, GO@PDOPA and rGO@PDOPA.

Figure 3A and B show typical topographic images and height profiles of GO and GO@PDOPA nanosheets. A thickness of 1.5–1.7 nm was identified in a typical hybrid planar flake of GO@PDOPA, which is thicker than that of the GO nanoflake (i.e. ~0.9 nm). This denotes a ~0.35 nm thickness of the PDOPA coating layer on each side of the GO sheet in the hybrid nanosheet. Figure 3C illustrates the TEM image of an individual GO@PDOPA nanosheet with rough surface morphology due to PDOPA coating on rGO surfaces.

Figure 3:

AFM images and the corresponding height profiles of (A) GO and (B) GO@PDOPA nanosheets. (C) TEM image of GO@PDOPA nanosheets.

The hybrid nanosheets were analyzed by thermogravimetry. TGA shows that no well-defined decomposition temperature can be found for PDOPA, suggesting that PDOPA has a heterogeneous structure and the presence of different chemical groups. As showed in Figure 4, rGO started to lose weight at around 280°C, while pure PDOPA started to decompose early from 60°C. The rGO@PDOPA also started to lose weight from 60°C in a slow rate. The TGA results reveal that rGO undergoes 19.91% weight loss as the temperature increases to 800°C due to the existence of oxygen groups. By comparison, the weight loss for rGO@PDOPA is larger than that for rGO, which is ascribed to the decomposition of PDOPA. The char yield at 800°C of rGO@PDOPA is about 60.67%, higher than the 47.16% of PDOPA. The composition of rGO@PDOPA is calculated as 58.97 wt% PDOPA and 41.03 wt% rGO according to the TGA results.

Figure 4:

TGA traces of rGO, PDOPA and rGO@PDOPA.

3.2 Dielectric properties of rGO@PDOPA/PVDF composite films

Figure 5A shows the dielectric permittivity of the rGO@PDOPA/PVDF nanocomposite films as a function of frequency with different rGO loadings. The rGO@PDOPA volume ratio data in the composites shown in this paper exclude the volume of PDOPA. As can be seen, no abrupt increases in permittivity are observed below 1.32 vol% of rGO content, suggesting that conductive pathways were not formed in the films. When the rGO content is beyond 1.5 vol%, the electrical conductivity of the composites is apparently higher than those below 1.5 vol% over the whole frequency range (Figure 5B), indicating the threshold percolation of the composites below 1.5 vol%, as previously estimated about 1.32 vol%. It was also observed that the frequency dependence of the electrical conductivity for the composites was relatively weak over the whole frequency range, indicating a relatively good insulating property. The permittivity improved sharply in the low-frequency range as the rGO content approached the percolation threshold, followed by a gradual decrease in the high-frequency range. As summarized in Figure 5A and D, the permittivity increased dramatically and up to more than 279 at 100 Hz when the rGO content reached 1.32 vol%, which is ~30 times larger than that of pure PVDF (about 9).

Figure 5:

Dependence of (A) dielectric permittivity, (B) electrical conductivity and (C) dielectric loss on the frequency for rGO@PDOPA/PVDF composites. (D) Dielectric permittivity and dielectric loss of rGO@PDOPA/PVDF as a function of volume fraction of rGO at 1000 Hz.

The percolation threshold of rGO@PDOPA/PVDF nanocomposites in the present study is much higher than those of PVDF nanocomposites filled with pristine rGOs, according to the results of previous studies on rGO/PVDF composites [11], [21]. This reveals that the PDOPA layers anchored on rGO surfaces can prevent the direct contact of the conductive rGOs and thus make the percolative network of rGOs harder to form. Compared with the rGO/PVDF composites, the permittivity of rGO@PDOPA/PVDF composites at low and modest frequencies (e.g. <10⁵ Hz) is higher near the percolation threshold owing to increased interfacial polarization. The enhancement of the interfacial polarization is mainly attributed to increased interfacial area and improved interfacial bonding between rGOs and PVDF.

As far as the dielectric loss is concerned, the loss tangents for the rGO@PDOPA/PVDF composites displayed a similar dielectric relaxation to that of the pure PVDF below the threshold (Figure 5C). An abrupt increase in loss tangent was observed near the percolation threshold as one important feature of the percolative composites (Figure 5D), while the loss factor maintained a value under 0.5 over the whole frequency range. Near the threshold, the loss tangent of the rGO@PDOPA/PVDF composite is 0.18 at 1000 Hz, which is about 5.5 times lower than that of the rGO/PVDF composite, which is acceptable for applications such as high-charge storage capacitors.

As the rGO content increased beyond the percolation threshold, the dielectric permittivity of the rGO@PDOPA/PVDF composites continued to rise, to an exceptionally high value of about 1100. At this point, the loss tangents of the rGO@PDOPA/PVDF composites maintained below 1 at moderate and high frequencies. The observed depressed loss had been rationalized by the presence of the insulating PDOPA layers, which separated the conductive rGOs and thus reduced conduction loss.

The morphology of the rGO@PDOPA/PVDF composite containing an rGO volume fraction of 1.32%, as shown in Figure 6, indicates that the rGO@PDOPA nanoplates homogeneously dispersed in the PVDF matrix without serious aggregation. This can be ascribed to strong interactions between the functional groups of the PDOPA and the -CF₂- groups of the PVDF [26]. Therefore, the wrapping of PDOPA onto rGO surfaces led to homogeneous dispersion of rGOs and enhanced interfacial polarization effect induced by increased phase interfaces between rGOs and PVDF. Furthermore, it should be noted that the rGO@PDOPA nanosheets tended to lie down inside the PVDF matrix due to the 2D structure of the nanosheets [11], [12]. As a result, the faces of the rGO@PDOPAs are almost parallel to the nanocomposite-film plate. That is, the rGO@PDOPA nanoplates are prone to be parallel to each other, isolated by a polymer layer as a medium between them. This promoted the formation of a larger number of microcapacitors in the nanocomposites [18], [21]. It suggests that the enhancement of the dielectric permittivity can be mainly ascribed to enhanced interfacial polarization and the formation of microcapacitor networks. Therefore, we believe that the non-covalent coating of PDOPA insulating layers on rGO surfaces and the existence of many microcapacitors in the PVDF matrix are beneficial in achieving a high dielectric permittivity, increased percolation threshold and reduced dielectric loss.

Figure 6:

SEM image of fracture surface of rGO@PDOPA/PVDF hybrid film containing 1.32 vol% rGO nanosheets.