Recently, graphene has been recognized as the most promising of the 2D nanomaterials due to its excellent electrical, thermal, and mechanical properties. Graphene is receiving much interest in the field of high performance polymer nanocomposites [1-5]. In order to produce high performance graphene nanocomposites, however, it is necessary to be able to produce a large amount of graphene, and the issue of graphene dispersibility must be resolved. In order to produce graphene in large amounts and to use it with composite materials, the method of using strong acid treatment and oxidation of graphite to graphite oxide (GO) and thermal or chemical reducing to produce reduced graphite oxide (RGO) has been widely reported [3,4]. GO produced as such exhibits a strong hydrophilic property; the inter-layer gap is broader than that of pristine graphite so that the material can be easily exfoliated through sonication or long-time stirring. It also has excellent dispersibility [6]. However, many oxygen atoms are found on the surface of a functional group, so that the electrical conductivity is usually low, which is a disadvantage when producing conducting composite materials. For the chemical reduction of GO, hydrazine (N2H4) is a well-known reducing agent [6,7]. However, a reduction process using hydrazine has disadvantages such as toxic gas generation and attachment of nitrogen atoms to the graphene sheet. Another reduction method is the heat treatment technique. Heating of GO at a temperature over 800 C removes the oxygen groups, which improves the electrical conductivity [7].
However, heating to a high temperature can cause large energy consumption.
GO was produced using the modified Hummer’s method [15] and was supplied by Nanosolution Co., Ltd (Korea). Na2S·5H2O was supplied by Yakuri (Japan), and NaOH by Showa (Japan). NMP and 1,4-dichlorobenzene (Sigma-Aldrich Corporation, USA) were used without purification.
Na2S·5H2O (0.2 mol), NMP (100 mL), DI-water (1.2 mol), and NaOH (0.006 mol) were added to a round flask and heated in a silicon oil bath at 190℃ for the dehydration reaction. After the dehydration, 1,4-dichlorobenzene and NMP were mixed with water-free Na2S solution; the resulting solution was put inside a Parr reactor. The reactor was filled with nitrogen. Afterwards, the temperature was increased and pressure was applied to initiate the reaction. The temperature was raised to 265℃ within 1 h; temperature was then maintained for 5 h for polymerization. After 5 h, the reactor was cooled. Once the temperature dropped to room temperature, the polymerized compound was filtered and washed with water, methyl alcohol, and hot DI-water. Afterwards, the compound was dried for over 15 h in a vacuum oven at 100℃. The molecular weight (Mv) of pure PPS was about 6000 (g/mol). In order to produce PPS/RGO composites, GO was first mixed with Na2S·5H2O and then dehydration was conducted. In this experiment, the weight ratio of GO to PPS was varied during the polymerization.
The intrinsic viscosities of pure PPS were obtained using a Brookfield viscometer with a thermosel system and a S18 spindle (DV-II+Pro, Brookfield Co., UK). Sample was dissolved in
1-chloronaphthalene in order to prepare 4 wt% polymer solutions at 210℃. The intrinsic viscosities were calculated using the Solomon-Ciuta relation, as follows.
where
where
The surface functional groups of GO were measured by Fourier transform infrared spectroscopy (IS 10, Sinco, Japan). The KBr disks were prepared by mixing KBr and the sample powder homogeneously. Fig. 1 shows that the functional groups of GO were carboxylic acid, ketone, and so on. Thermal analysis was conducted using a differential scanning calorimeter (DSC) and a thermogravimetric analyzer (TGA). For DSC (Q20, TA Instrument, USA) analysis, the increment of temperature rise of the nitrogen was set at 10℃ /min; temperature was kept at 350℃ for 10 min, and was then dropped at the speed of 10℃ /min to 40℃. TGA (Q50, TA Instrument, USA) was conducted until the temperature of 900℃ was reached, at a speed of 10℃/min in nitrogen. An scanning electron microscope (SEM) image was obtained using a Hitachi S-4700 field emission SEM. Electrical conductivity was measured using a 4-point probe employing the FPP-RS8 of DASOL ENG.
Fig. 2 shows the DSC thermograms of pure PPS and PPS/ RGO nanocomposites. First, it was seen that there are two melting temperatures in the heating process; this might be due to the unstable and small crystals of PPS fabricated using GO. Also,
it was observed that the melting temperature dropped with increasing GO content; this may be caused by interference of the polymerization of PPS on the high molecular weight polymers due to the addition of GO. When Na2S reacts with the epoxy group of GO, the ratio of the
Fig. 3 shows the TGA analysis, revealing the thermal stability of PPS, GO, and PPS/RGO composite materials. Comparing the temperature (T10) with a 10 wt% decrease and the temperature (T20) with a 20 wt% decrease, it can be seen that the thermal stability decreases as the GO content increases (Table 2). This, as seen in the DSC results, appears to be because as GO increases, the polymerization of PPS is restricted and GO is not well reduced during the PPS polymerization, thus decreasing the
[Table 1.] DSC thermograms of PPS/RGO nanocomposites
DSC thermograms of PPS/RGO nanocomposites
[Table 2.] TGA results of PPS/RGO nanocomposites
TGA results of PPS/RGO nanocomposites
thermal stability. If GO is efficiently reduced in the polymerization reaction, the thermal stability should increase with the increase of the GO content. The graphite nanoplatelets produced in the direct exfoliation method have relatively fewer functional groups than do the oxidized GO, so that there is less weight loss incurred by the functional groups. In the case of the PPS/graphite composite material produced as such, the thermal stability increased along with the increase of the amount of graphite [16].
Fig. 4 provides an FE-SEM image of the PPS/RGO composites obtained from pure PPS to 20 wt% of GO composite. The FE-SEM image shows that the GO in the PPS substrate is well dispersed and that the graphene network has become much more
[Table 3.] Electrical conductivities of pure PPS, GO, and PPS/RGO nanocomposites
Electrical conductivities of pure PPS, GO, and PPS/RGO nanocomposites
dense above 10 wt% of GO. The dispersed structure of the GO affects the electrical conductivity and the thermal properties of the PPS/RGO composites.
Table 3 shows the electrical conductivity values of the PPS/ RGO nanocomposite material with different GO contents. The electrical conductivity of GO appears to be higher than the reported conductivity of GO [8]; however, with heat treatment at 300℃, the conductivity drops 10 times. Such different results appear to be caused by the different degree of oxidation during the acid treatment for GO production. For the case of PPS composite materials, it appears that the percolation threshold of GO is between 5-8 wt%. Compared to the value of 5 wt % GO composites, the electrical conductivity of 20 wt% GO composites is improved over 108 times.
It was found that GO is better dispersed and reduced by heat during