- Email: [email protected]
Enhanced dispersion of carbon nanotube in silicone rubber assisted by graphene
Polymer 53 (2012) 3378e3385
Contents lists available at SciVerse ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
Enhanced dispersion of carbon nanotube in silicone rubber assisted by graphene Haiqing Hu a, *, Li Zhao a, Jiaqiang Liu a, Yin Liu a, Junmei Cheng a, Jun Luo b, Yongri Liang b, Yong Tao a, Xin Wang a, Jian Zhao a, * a
Key Laboratory of Rubber-Plastics, Ministry of Education/Shandong Provincial Key Laboratory of Rubber-plastics, School of Polymer Science and Engineering, Qingdao University of Science and Technology, Qingdao 266042, China b State Key Laboratory of Polymer Physics and Chemistry, Joint Laboratory of Polymer Science and Materials, Beijing National Laboratory for Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 November 2011 Received in revised form 2 March 2012 Accepted 19 May 2012 Available online 30 May 2012
We report a novel, scalable and inexpensive approach to fully disperse carbon nanotubes in silicone rubber by the addition of graphene. In comparison to graphene, the dispersion of multi-walled carbon nanotubes (MWNTs) in silicone rubber matrix is extremely difficult although both of them possess similar physical structure. The different dispersion behavior of graphene and MWNTs could be contributed to the difference in their interaction with polymer matrix and their geometry. Based on SEM, TEM and XRD analysis, we find that the dispersion of MWNTs in silicone rubber is dramatically improved by the addition of graphene. Graphene acts as a compatilizer since it shows strong interaction with both polymer matrix and MWNTs. This method provides a simple route to enhance the dispersion of carbon nanotubes and improve the electrical property of the polymer composites. The synergic effect of the hybrid materials may not to be limited to the applications in polymer composites. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Graphene Carbon nanotubes Silicone rubber
1. Introduction Since the discovery of carbon nanotubes (CNTs) [1] and graphene [2], they have received increasing interest in various fields such as chemistry, physics, materials science, and electrical engineering [1e16]. Both of the carbon materials, possess unique nanostructures with remarkable physical and mechanical properties [1e8]. They are widely used as nanofillers to improve electrical, thermal and optical properties of polymer composites [9e12]. The prospect of obtaining advanced nanocomposites with multifunctional features and materials used for structures and electrical conductors has attracted researchers’ attention in both academic and industrial communities [13e16], especially as electro statically dissipative materials and aerospace structural materials [17e20]. Although the research on carbon nanotube-based polymer composites has achieved significant progress, two major issues remain challenging for the fabrications of carbon nanotube reinforced polymer composites: (1) the dispersion of CNTs, i.e. the tendency of nanotubes to aggregate during processing; (2) the limited availability of high-quality carbon nanotubes in large quantities and high cost of their production. Graphene, a hexagonal network of carbon atoms, is actually the basic unit of a carbon * Corresponding authors. Tel.: þ86 532 88766117. E-mail addresses: [email protected], [email protected] (H. Hu), zhaojian@ gmail.com (J. Zhao). 0032-3861/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymer.2012.05.039
nanotube. Rolling graphene into a seamless, hollow cylinder forms the carbon nanotube [21,22]. In addition, although graphene and carbon nanotubes are similar in chemical composition, different topology distinguishes carbon nanotubes from graphene and offers them unique properties. One of the major advantages of graphene over carbon nanotubes lies in the precursor of graphenedgraphite, which is natural abundant [23]. It is essential for industrial applications in composite fields, since high yield production and a costeffective and scalable process is possible [24,25]. The investigations on graphene or carbon nanotube-based polymer composites are mainly focused on plastics matrices such as polyamide 6 (PA6) [26], polycarbonate (PC) [26,27], epoxy [28], polyurethane (PU) [29], polystyrene (PS) [30,31], poly (methyl methacrylate) (PMMA) [32,33] or poly (styrene-b-isoprene-bstyrene) (SIS) block copolymer [34]. There are only a few of studies on graphene or CNTs/rubber composites [35e38]. One important reason is that it is difficult for nanoparticles to be dispersed in rubber matrices because of their high viscosity or low surface energy [36]. Conductive rubbers are critically needed in the fields such as ESD (electrostatic discharge) and EMI (electromagnetic interference) shielding materials, electronics packaging, telecommunication antenna, mobile phone parts and frequency shielding coatings for aircraft and electronics [28,35e38]. However, the physical interaction between rubbers and nanoparticles are weaker than that between plastics and nanoparticles. It is of importance to understand the mechanism of the dispersion of nanoparticles in rubbers.
H. Hu et al. / Polymer 53 (2012) 3378e3385
Silicone rubber offers good tolerance to extremely low and high temperatures, being able to operate normally from 55 C to þ300 C. The carbon backbone of organic rubbers makes them susceptible to ozone, UV, heat and other aging factors while silicone rubber can perform well. This feature makes silicone rubber become one of the choices of elastomers in many extreme environments. Silicone rubber also has other superior properties such as the ease of changing its shape and filling the seals in a matrix. Therefore, silicone rubber was selected as the polymer matrix in this paper. One critical issue for the fabrication of nanocomposites is how to uniformly disperse nanofillers in polymer matrix. Many methods including physical and chemical methods were investigated to assist in the dispersion of CNTs [39]. Many reports investigated the synergetic effect of different fillers to make balance between properties and cost. Graphite (G) together with carbon fibers (CF) in polymer composites [40e47] has been studied for many years and excellent electrical conductivity of the composites has been obtained. Carbon nanotubes also have been investigated when blended with other fillers such as clay [48,49], single-walled carbon nanotubes (SWNTs) [50], graphite (G) and carbon black (CB) [51]. Nanoplatelets/carbon black/carbon nanotubes three-filler system in epoxy resin has also been reported [52]. Balberg et al. [53,54] proposed models to explain electrical percolation based on excluded volume theory. Sun et al. [51] extended the excluded volume theory and obtained an equation to predict the amount of fillers used to get the percolation when two fillers were used. It is commonly believed that two (or three) types of fillers can form cosupporting conductive networks in which the fibrous CF filler act as long distance charge transporters and the particulate fillers such as CB serves as an interconnection between the fibers by forming local conductive paths [42,43]. However, the dispersion of nanofillers in a synergetic system is rarely explored. In this work, we report a simple and effective way to disperse carbon nanotubes and graphene in silicone rubber. The excellent dispersion of CNTs can be achieved with the aid of graphene because of strong interaction between graphene and silicone rubber matrix as well as strong interaction between graphene and CNTs. The mechanism of dispersion will be discussed in detail in this paper. 2. Experimental section 2.1. Materials Silicone rubber (Methyl-Vinyl-Silicone, MVQ 110-2) was provided by Jiangsu Hongda Materials Company, China. The molecular weight of the transparent silicone rubber is about 4.50e5.9 105 g/mol. Natural flake graphite (LC50-9999, particle size 40 mm) was purchased from Qingdao Ruisheng Graphite Co. Ltd. China. Carboxylated multi-walled carbon nanotubes (MWNTs, MC5, purity > 95%, OD: 20e30 nm, Length: 10e30 mm, eCOOH content: 1.23 wt%, resistivity: 2.8 U cm) was purchased from Chengdu Institute of Organic Chemistry Co. Ltd. China. Curing agent 2,5-dimethyl-2,5-di(tert-butyl peroxy) hexane (AkzoNobel, Tianjin), is commercially available and used as received. 2.2. Preparation of graphene Graphite oxide (GO) was prepared according to the modified Hummers method first [55]. 5 g of flake graphite (40 mm), 3.75 g of NaNO3, and 230 mL of H2SO4 were mixed at 0 C. 15 g of KMnO4 was slowly added into the mixture within about 1 h, followed by stirring for 2 h in an ice-water bath. After additional stirring at room temperature for 5 days, 500 mL of 5 wt% H2SO4 was added in 1 h. The resultant mixture was further stirred for 2 h at 98 C. Then, the
3379
temperature was lowered to 60 C and 15 mL of H2O2 was added. The mixture was stirred for 2 h at room temperature and then precipitated, followed by being washed 7 times with an aqueous solution of 3 wt% H2SO4/0.5 wt% H2O2 and subsequent filtering. The resultant solid was further washed 3 times with 3 wt% and 0.6 wt% HCl solution. Finally, GO was washed with deionized water until the pH value was close to 7, followed by filtering and drying in a vacuum oven for one week. The dried GO samples were thermally treated at 1050 C for 5 min in a tubular furnace to obtain thermally reduced graphene (TrG). The detailed process and properties of graphene can be obtained elsewhere (see reference 10). 2.3. Preparation of graphene/carbon nanotubes/silicone rubber composites The rubber composites were fabricated as follows: (1) Silicone rubber (MVQ, before crosslinking) was dissolved in THF to obtain homogeneous solution. (2) Graphene (or MWNTs) was dispersed in THF. The mixture was sonicated for 1 h to get stable suspension. (3) The above (1) and (2) suspensions were mixed together and sonicated for 30 min to obtain homogeneous suspension. (4) THF was evaporated at 50 C, and the mixture was dried in vacuum oven. (5) Curing agent was added into the mixture, and then the resulting compound was moved to vulcanizing machine for curing followed by vulcanizing at 160 C for 12 min to obtain the composites. 2.4. Characterizations 2.4.1. X-ray diffraction (XRD) X-ray diffraction measurements (D-MAX 2500/PC, 18 kW rotating anode, Rigaku Corporation) were carried out to characterize the structure of graphene, and the composites with Cu Ka radiation (40 kV, 100 mA, wavelength, 1.5406 Å). X-ray diffraction patterns were obtained at room temperature with a scanning rate of 4 /min and the step size of 0.02 . 2.4.2. Scanning electron microscopy (SEM) Cryogenic fracture surfaces of the specimens were coated by sputtering with gold and then observed with field emission scanning electron microscopy (FESEM, Japan Electronics Co., Ltd., JSM6700F). 2.4.3. Transmission electron microscopy (TEM) Ultrathin specimen (thinner than 100 nm) for TEM observation were cryogenically cut with a diamond knife using an ultra thin microtome (UCT, Leica Ultracut, EMFC7) and collected on 200-mesh copper grids. The exfoliation of graphene nanosheets in silicone rubber was observed with transmission electron microscope (TEM, JEOL, JEM2100). 2.4.4. X-ray photoelectron spectroscopy (XPS) XPS experiments were carried out on a RBD upgraded PHI5000C ESCA system (Perkin Elmer) with Mg Ka radiation (hn ¼ 1253.6 eV) or Al Ka radiation (hn ¼ 1486.6 eV). Binding energies were calibrated by the containment carbon (C1s ¼ 284.6 eV). The data analysis was carried out using RBD AugerScan 3.21 software (RBD Enterprises, USA). 2.4.5. Resistivity analysis PC-68 high resistance meter (Shanghai Baihe Instrument Technology Co., Ltd, China) was used to test the resistance of silicone rubber composites. The resistivity was calculated through the formula:
r v ¼ p*d*R=ln2
3380
H. Hu et al. / Polymer 53 (2012) 3378e3385
where r is resistivity, d is the thickness of the sample, and R is the resistance of the sample. 2.4.6. Thermal diffusivity Thermal diffusivity was measured by NETZSCH LFA 447 at 25 C. Xenon lamp as heating source launched a bunch of pulse to the surface of sample, the increasing temperature was measured by infrared detector. 3. Results and discussion 3.1. Characterization of graphite oxide (GO) and thermally reduced graphene (TrG) FTIR, XRD and XPS were used to characterize graphite oxide (GO) and thermally reduced graphene (TrG). TrG was prepared as reported previously [10]. As shown in Fig. 1(a), the band at
1732 cm1 appeared in the FTIR spectrum of GO, indicating that carbonyl groups (vibration modes of C¼O at 1732 cm1 and CeOeC at 1065 cm1) were anchored on GO during the oxidation of graphite. The oxygen-containing groups of TrG almost disappeared, and the vibration peaks of graphene (w1385, w1446, and w1635 cm1) appeared, indicating that GO was fully reduced to TrG after thermal treatment. XRD (Fig. 1(b)) pattern shows that the interlayer spacing of GO (0.794 nm, 11.13 ) was larger than that of pristine graphite (0.335 nm, 26.6 ) [56] because of the intercalation of functional groups on GO. After thermal reduction of GO, the diffraction peak (11.13 ) of GO disappeared and a broad diffraction peak of about 25 appeared, implying that the distance between the intercalated graphite layers decreased and the layer structure was reconstructed during thermal reduction although remarkable amount of graphene was fully exfoliated by thermal expansion [57].
Fig. 1. The characterization of GO and TrG, (a) FTIR spectra of GO and TrG, (b) XRD profiles of GO and TrG, (c) C1s XPS spectrum of GO, (d) C1s XPS spectrum of TrG, (e) SEM image of GO, (f) TEM image of TrG.
H. Hu et al. / Polymer 53 (2012) 3378e3385
X-ray photoelectron spectroscopy (XPS) was used to probe functional groups on graphite oxide and thermally reduced graphene. C1s XPS spectrum of GO (Fig. 1(c)) shows the existence of oxygencontaining groups (CeO at 286.2 eV, and C]O at 287.8 eV) due to the oxidation of graphite. The peak area for CeC (284.5 eV) in the XPS spectrum of GO accounts for about 64 percent of total peak area, indicating that graphite has been fully oxidized. C1s XPS spectrum of TrG (Fig. 1(d)) shows that the peaks of CeO and C]O have dramatically decreased after 5 min pyrolysis reduction at 1050 C, but a few of C¼O and CeO groups were still retained on graphene, which can not be detected in the FTIR spectrum of TrG. These oxygen-containing groups play an important role in the dispersion of graphene. Fig. 1(e) is the SEM image of GO, and Fig. 1(f) is the TEM image of TrG. It can be seen that the surfaces of GO and TrG are not smooth but wrinkled. The original sp2 bonding of two-dimensional graphene evolved into a sp3 hybrid non-planar bonding, resulting in the disruption of the original plane structure. As a result, a wrinkled morphology was formed. This morphology is very favorable for the dispersion of graphene in rubber, which will be discussed in the following section. 3.2. Dispersion of multi-walled carbon nanotubes (MWNTs) in silicone rubber The agglomeration of carbon nanotubes (labeled by white circle) in silicone rubber/carbon nanotube (5 wt%) were observed by SEM and TEM as shown in Fig. 2 (a) and (b). Although ultrasonic treatment of carbon nanotubes have been carried out during the preparation of MWNTs/silicone rubber composites, MWNTs still can not be fully dispersed in the polymer matrix due to intrinsic strong attractive forces (Van der Waals force) of carbon nanotubes. Fig. 2 (c) shows XRD profiles of MWNTs, MVQ and MVQ/MWNTs (5 wt%) composite. The diffraction peaks of MWNTs were observed at 24.4 (002) and 43.9 (100), which were assigned to interlayer space in the radial direction and in-plane graphitic structure of MWNTs, respectively. The MVQ showed a diffraction peak at 12.1. In the XRD profiles of MVQ/MWNTs composite, the relative intensity and peak position of diffraction peak at 12.1 were almost not changed and the peaks at 24.4 and 43.9 were very week, which indicated that the MWNTs were contained in the nanocomposite and did not influence the structure of MVQ. 3.3. Dispersion of graphene in silicone rubber SEM, TEM and XRD of silicone rubber/graphene (3 wt%) composite were shown in Fig. 3. Fig. 3 (a) and (b) showed SEM images of silicone rubber/graphene composite. It could be clearly
3381
seen that almost all of graphene was well dispersed in the silicone rubber matrix, consistent with the results of TEM observation of silicone rubber/graphene composites (Fig. 3 (c)). The XRD profile of graphene (TrG) (Fig. 3 (d)) exhibited a broad diffraction peak at about 25.4 . When graphene was blended with silicone rubber (1 wt% or 3 wt%), the diffraction peak of TrG in those composites disappeared. It implies that graphene was almost exfoliated by silicone rubber and very well dispersed in the silicone rubber. Graphene and carbon nanotube have similar sp2 chemical structure, but they show dramatically different dispersion behavior in silicone rubber. The main reasons can be found in the following two aspects: The first aspect is that the morphology of the two carbon materials is quite different. Fig. 3 (b) is the amplification of the region of black dashed circle in Fig. 3 (a). We can find that the surface of graphene is not flat. Instead, a worm-like wrinkled morphology of graphene was observed. The original sp2 bonding of two-dimensional plane graphene evolved into a sp3 hybrid non-planar bonding, resulting in the disruption of the original plane structure. As a result, a wrinkled morphology was formed. This worm-like structure can prevent the restacking of graphene layers when mixing with silicone rubber. The ripples of graphene are very beneficial for the formation of occluded rubber in the silicone rubber composites, which limits the movement of molecular chains of silicone rubber. As a result, the strength and hardness of the graphene/silicone rubber composites are improved. Meanwhile, the wrinkled morphology renders the silicone rubber to be better integrated with graphene, which facilitates the dispersion of graphene in silicone rubber. The second aspect is that the interaction between graphene and silicone rubber is much stronger than that between MWNTs and silicone rubber. Schniepp [57] reported that the surface area of graphene is about 800 m2/g as measured by BET surface adsorption. The huge surface area of graphene offers numerous active adsorption sites, which render better interaction between the graphene and silicone rubber. In contrast with MWNTs, the two planes of graphene possess more contact area with silicone rubber. Also, a few of oxygen-containing groups were still retained on graphene (XPS spectrum in Fig. 1(d)). Hydrogen bonds may form when graphene contact with the oxygen-containing groups on the silicone rubber [36]. Thus, an effective interface bonding forms between the graphene and the silicone rubber. The mixing energy in the blend system can be described by eq. (1):
DGM ¼ DHM T DSM
(1)
When graphene or MWNTs is mixed with silicone rubber, the entropy changes little, because conformation entropy change little
Fig. 2. MWNTs agglomerated in silicone rubber/MWNTs (5 wt%), (a) SEM image of MVQ/MWNTs (5 wt%) (10000), (b) TEM image of MVQ/MWNTs (5 wt%) (150000), (c) XRD profiles of MWNTs, MVQ and MVQ/MWNTs (5 wt%) composite.
3382
H. Hu et al. / Polymer 53 (2012) 3378e3385
Fig. 3. SEM, TEM images and XRD profiles of MVQ/TrG (3 wt%), (a) SEM image, (b) SEM image, (c) TEM image, (d) XRD profiles of TrG, MVQ and MVQ/TrG composite.
and entropy change mainly arises from the position change of graphene or MWNTs. In the mixing process, the entropy is positive, which is beneficial for the dispersion of graphene or MWNTs into silicone rubber. The main factor that prevents nanoparticles from being dispersed in silicone rubber or other matrices is the effect of enthalpy. The interaction between the nanoparticles is strong due to the huge surface energy of nanoparticles. There are three competing interactions in the system when nanoparticles are mixed with polymers. The first is the interaction for a polymerpolymer contact (DHp-p), the second is the interaction for a nanoparticleenanoparticle contact (DHn-n), and the third is the interaction for a polymer-nanoparticle contact (DHp-n). The first and second interactions will cause the phase separation of polymer and nanoparticles, which is conducive to the aggregation of nanoparticles. The third interaction will favor the dispersion of nanoparticles into the polymer. The difference in the dispersion of graphene and MWNTs is caused by the interaction between the nanoparticles and polymers. DHp-p and DHn-n are almost the same for graphene and MWNTs as they have the same elements——carbon and the sp2 structure of CeC are similar for both graphene and MWNTs. The difference arises from the DHp-n. Good dispersion of Graphene in polymers implies that the DHp-G (G represents graphene) is much higher than DHp-c (C represents carbon nanotube). Suppose that the interactions among graphene and polymer chains are the sum of Van der Waals forces. The force is weak for only one carbon atom interacting with a silicone atom. There are
numerous interactions in a sample, so the sum is huge enough. Let us compare the difference between graphene and carbon nanotubes. The elements of graphene and carbon nanotubes are the same, meaning that they have the same Van der Waals forces between the carbon elements and other elements of the polymer. The difference lies in the shape of these two nanoparticles. Both sides of a graphene plane can interact with the polymer, but only the outer layer of MWNTs can do. The total Van der Waals forces among graphene and the polymer are doubled as compared to that among the polymer and MWNTs. The total DH of the graphene/ polymer system will be much smaller than that of the MWNTs/ polymer system. As a result, it is easier for graphene to be dispersed in the polymer. 3.4. Graphene-assisted dispersion of MWNTs in silicone rubber When Graphene/carbon nanotubes/silicone rubber were blended together, an interesting phenomenon was observed. Fig. 4 shows the SEM, TEM images and XRD profiles of the system. It can be seen from the SEM and TEM images (Fig. 4 (a)e(e)) that graphene and MWNTs were well dispersed in the silicone rubber. And the graphene are dispersed around carbon nanotubes. Carbon nanotubes and graphene together form a conductive network (as indicated by the black lines in Fig. 4 (a) and (b)) in the silicone rubber matrix. Graphene is well dispersed in the silicone rubber matrix, and the aggregation of carbon nanotube in the matrix was not found in SEM or TEM images.
H. Hu et al. / Polymer 53 (2012) 3378e3385
XRD profiles (Fig. 4(f)) showed that when the graphene was mixed with MWNTs and silicone rubber (silicone rubber/graphene/ MWNTs ¼ 100/1/2.5), the diffraction peak of interlayer structure of TrG (25.4 ) was not found, whereas the diffraction peak of MWNTs (24.4 ) was still observed as showed in Fig. 4(f), indicating that graphene was still well dispersed in the silicone rubber matrix regardless of the presence of MWNTs. The dispersion of graphene in silicone rubber was not influenced by the presence of MWNTs. The dispersion of MWNTs can not be assessed by XRD measurement. However, the SEM and TEM images indicate that both MWNTs and TrG are well dispersed in silicone rubber matrix.
3383
The excellent dispersion of MWNTs and graphene in the blend system may be attributed to the following reasons: The diameter of graphene is in the range of tens of nanometer to several micrometers, while the dimension (length) of MWNT is 10e30 mm. Graphene is smaller in particle size as compared with MWNTs, and it can act as a separator between MWNTs. The force that the sonication provides results in the intercalation of graphene into nanotube bundles. A graphene sheet may “adhere” to many carbon nanotubes due to the strong interaction between carbon nanotubes and graphene, leading to the disruption of carbon nanotube bundles. Once graphene “adheres” to MWNTs, it is not
Fig. 4. Silicone rubber/graphene/MWNTs ¼ 100/1/2.5, (a), (b)SEM images, (c) TEM image, the insert is the enlargement of carbon nanotube in the white rectangle, (d) TEM image, (e) TEM image of silicone rubber/graphene/MWNTs ¼ 100/1/3, (f) XRD profiles of MWNT, TrG, MVQ and MVQ composites. (The arrows point to the MWNTs and graphene are in white circles).
H. Hu et al. / Polymer 53 (2012) 3378e3385
easy for graphene to be “driven out” from MWNTs because graphene possesses the extended p electron system similar to that of carbon nanotubes. There exists a good interaction between graphene and silicone rubber matrix. On one hand, the eOHe on graphene may form hydrogen bonds with the eOe on silicone rubber. On the other hand, there is huge contact surface area between silicone rubber and graphene because graphene has very large specific surface area [57] and each plane of graphene can contact with silicone rubber. Graphene can be dispersed uniformly in silicone rubber as discussed above. Since graphene has good interaction with silicone rubber and carbon nanotubes, graphene plays a role of compatilizer for MWNTs and silicone rubber, which facilitates the dispersion of MWNTs in silicone rubber. Based on the aforementioned analysis, we believe that the most important factor for the excellent dispersion of MWNTs assisted by graphene is the strong interaction between graphene and the matrix as well as the strong interaction between graphene and MWNTs. Graphene has a large surface area and carbon nanotubes have a great aspect ratio. One dimensional and two-dimensional materials cooperate, forming a “conductive bridge” in the silicone rubber as shown in the schematic diagram (Fig. 5). 3.5. Effect of the dispersion on electrical properties of silicone rubber composites With the increase of filler content, the volume resistivity of silicone rubber decreased (Fig. 6). The percolation threshold for graphene is about 2 wt% with volume resistivity of 1 105 U cm. By contrast, the percolation threshold for MWNTs is about 5 wt% with volume resistivity of 4.21 106 U cm. The graphene with twodimensional planar structure is better dispersed than MWNTs, which favors the formation of conductive network. MWNTs tend to aggregate in silicone rubber, leading to larger percolation value. When both graphene and carbon nanotubes are filled in silicone rubber composites, carbon nanotube is well dispersed with the aid of graphene and the conductive network forms. Graphene is distributed around the carbon nanotubes. Carbon nanotubes play as a “conductive bridge” connection between graphene layers and the matrix. At the graphene content of 1 wt% and the MWNT content of 2 wt%, the volume resistivity of the composite was 4.91 109 U cm. At the graphene content of 1 wt% and the MWNT content of 2.5 wt%, the volume resistivity was reduced to 8.0 104 U cm.
Fig. 6. The resistivity of silicone rubber composites.
0.13
Thermal diffusivity (m2 /s)
3384
0.12
0.11
0.10
0.09
MVQ
TrG(1)
MWNTs (5)
TrG/MWNTs (1/3)
Fig. 7. Thermal conductivity of silicone rubber composites.
3.6. Effect of the dispersion on thermal diffusivity of silicone rubber composites Thermal diffusivity of silicone rubber composites was shown in Fig. 7. The thermal diffusivity of pure silicone rubber was 0.105 m2/s, which was increased by the addition of graphene or WMNTs or two fillers together. Thermal diffusivity of silicone rubber composites were 0.113 and 0.119 m2/s respectively by filling graphene (1 wt%) and MWNTs (5 wt%). By contrast, thermal diffusivity of silicone rubber composite was enhanced to 0.123 m2/s by filling 1 wt% graphene and 3 wt% nanotubes together. Graphene has strong interaction with silicone rubber and also breaks up the agglomeration of MWNTs. The remarkable enhancement is due to the synergistic effect of graphene and MWNTs. 4. Conclusion
Fig. 5. Schematic of the synergetic interaction between graphene and carbon nanotubes.
In summary, we studied the dispersion of MWNTs and graphene in silicone rubber matrix. It is difficult to disperse MWNTs in silicone rubber alone, whereas it is much easier for MWNTs to be dispersed in the matrix with the aid of graphene. The strong interaction between graphene and silicone rubber as well as the strong interaction between graphene and MWNTs accounts for the good dispersion of MWNTs. Graphene behaves as a compatilizer. Both physical isolation between graphene and MWNTs and some hydrogen bonding between graphene and MWNTs may form during the dispersion of MWNTs assisted by graphene. Addition of
H. Hu et al. / Polymer 53 (2012) 3378e3385
graphene provides a simple route to disperse carbon nanotube and the synergetic effect facilitates the fabrication of high performance polymer composites. Acknowledgments This project is supported by the NSFC (Grant No.20944004 and 51073082), the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry, MOE Key Laboratory of Molecular Engineering of Polymers (Fudan University), MOE Key Laboratory of Organosilicon Chemistry and Material Technology of Ministry of Education (Hangzhou Normal University) Qingdao Municipal Science and Technology Program, Basic Research Project (12-1-4-3-(7)-jch). References [1] Iijima S. Nature 1991;354:56e8. [2] Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y, Dubonos SV, et al. Science 2004;306:666e9. [3] Geim AK, Novoselov KS. Nat Mater 2007;6:183e91. [4] Song P, Cao Z, Cai Y, Zhao L, Fang Z, Fu S. Polymer 2011;52:4001e10. [5] Soldano C, Mahmood A, Dujardin E. Carbon 2010;48:2127e50. [6] Luong ND, Hippi U, Korhonen JT, Soininen AJ, Ruokolainen J, Johansson L, et al. Polymer 2011;52:5237e42. [7] Rafiee MA, Rafiee J, Wang Z, Song HH, Yu ZZ, Koratkar N. ACS Nano 2009; 3(12):3884e9. [8] Park S, Ruoff RS. Nat Nanotechnology 2009;4Z:217e27. [9] Ajayan PM, Schadler LS, Giannaris C, Rubio A. Adv Mater 2000;10:750e3. [10] Hu H, Liu Y, Wang Q, Zhao J, Liang Y. Mater Lett 2011;65:2582e4. [11] Chen WF, Yan LF. Nanoscale 2010;2:559e63. [12] Sheng K, Bai H, Sun Y, Li C, Shi G. Polymer 2011;52(24):5567e72. [13] Hu HQ, Yu JH, Li YY, Zhao J, Dong HQ. J Biomed Mater Res A 2012;100A(1): 141e8. [14] Allaoui A, Bai S, Cheng HM, Bai JB. Comp Sci.Technol 2002;62(15):1993e8. [15] Dong X, Su CY, Zhang W, Zhao J, Ling Q, Huang W, et al. Phys Chem Chem Phys 2010;12:2164e9. [16] Zhao X, Zhang QH, Chen DJ. Mcromolecules 2010;43:2357e63. [17] Jin Z, Pramoda KP, Xu G, Goh SH. Chem.Phys Lett 2001;337:43e7. [18] Li B, Olson E, Perugini A, Zhong W. Polymer 2011;52:5606e14. [19] Vajtai R, Wei BQ, Zhang ZJ, Jung Y, Ramanath G, Ajayan PM. Smart Mater Struct 2002;11:691e8. [20] Liang JJ, Xu YF, Huang Y, Zhang L, Wang Y, Ma YF, et al. J Phys Chem C 2009; 113:9921e7. [21] Dresselhaus MS, Lin YM, Rabin O, Jorio A, Souza Filho AG, Pimenta MA, et al. Mater Sci Eng 2003;C23:129e40. [22] Kim H, Abdala AA, Macosko CW. Macromolecules 2010;43:6515e30. [23] Kotov NA. Nature 2006;442:254e5.
3385
[24] Park S, Ruoff RS. Nat Nanotechnol 2009;4:217e24. [25] Vaia RA, Wagner HD. Mater Today 2004;7:32e7. [26] Steurer P, Wissert R, Thomann R, Mulhaupt R. Macromol Rapid Commun 2009;30:316e27. [27] Kim H, Macosko CW. Polymer 2009;50:3797e809. [28] Zaman I, ThanhPhan T, Kuan H, Meng Q, La LT, Luong L, et al. Polymer 2011; 52:1603e11. [29] Tkalya E, Ghislandi M, Alekseev A, Koning C, Loos J. J Mater Chem 2010;20: 3035e9. [30] Kim H, Miura Y, Macosko CW. Chem Mater 2010;22:3441e50. [31] Stankovich S, Dikin DA, Dommett GHB, Kohlhaas KM, Zimney EJ, Stach EA, et al. Nature 2006;442:282e6. [32] Das B, Prasad KE, Ramamurty U, Rao CNR. Nanotechnology 2009;20: 125701e5. [33] Layek RK, Samanta S, Chatterjee DP, Nandi AK. Polymer 2010;51:5846e56. [34] Peponi L, Tercjak A, Verdejo R, Lopez-Manchado MA, Mondragon I, Kenny JM. J Phys Chem C 2009;113:17973e8. [35] Verdejo R, Bernal MM, Romasanta LJ, Lopez-Manchado MA. J Mater Chem 2011;21:3301e10. [36] Jiang MJ, Dang ZM, Xu HP. Eur Polym J 2007;43:4924e30. [37] Zhang H, Zheng W, Yan Q, Yang Y, Wang J, Lu Z, et al. Polymer 2010;51: 1191e6. [38] Prud’homme RK, Ozbas B, Aksay IA, Register RA, Adamson DH. International patent, WO 2008/045778 A1; 2008. [39] Moniruzzaman M, Winey KI. Macromolecules 2006;39:5194e205. [40] Ma PC, Liu MY, Zhang H, Wang SQ, Wang R, Wang K, et al. ACS Appl Mat Interfaces 2009;1(5):1090e6. [41] Lee GW, Park M, Kim J, Lee JI, Yoon HG. Composites: Part A 2006;37: 727e34. [42] Clingerman LM, Weber EH, King JA, Schulz KH. Polym Compos 2002;23: 911e24. [43] Thongruang W, Spontak RJ, Balik CM. Polymer 2002;43:2279e86. [44] Drubetski M, Siegmann A, Nakis M. Polym Compos 2005;26:454e64. [45] Lee JH, Kim SK, Kim NH. Scr Mater 2006;55:1119e22. [46] Drubetski M, Siegmann A, Nakis M. J Mater Sci 2007;42:1e8. [47] EI-Bounia NE, Piccione PM. J Polym Eng 2008;28:141e54. [48] Wang Z, Meng XY, Li JZ, Du XH, Li SY, Jiang ZW, et al. J Phys Chem C 2009;113: 8058e64. [49] Tang C, Xiang L, Su J, Wang K, Yang C, Zhang Q, et al. J Phys Chem B 2008;112: 3876e81. [50] Brady-Estevez AS, Schnoor MH, Kang S, Elimelech M. Langmuir 2010;26(24): 19153e8. [51] Sun Y, Bao HD, Guo ZX, Yu J. Macromolecules 2009;42:459e63. [52] Wei T, Song L, Zheng C, Wang K, Yan J, Shao Bo, et al. Mater Lett 2010;64: 2376e9. [53] Balberg I. Phys Rev B 1986;33:3618e20. [54] Celzard A, McRae E, Deleuze C, Dufort M, Furdin G, Mareˆche0 JF. Phys Rev B 1996;53:6209e14. [55] Becerril HA, Mao J, Liu ZF, Stoltenberg RM, Bao ZN, Chen YS. ACS Nano 2008;2: 463e70. [56] Guo HL, Wang XF, Qian QY, Wang FB, Xia XH. J.ACS Nano 2009;3:2653e9. [57] Schniepp HC, Li JL, McAllister MJ, Sai H, Herrera-Alonso M, Adamson DH, et al. Phys Chem B 2006;110(17):8535e9.
Contents lists available at SciVerse ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
Enhanced dispersion of carbon nanotube in silicone rubber assisted by graphene Haiqing Hu a, *, Li Zhao a, Jiaqiang Liu a, Yin Liu a, Junmei Cheng a, Jun Luo b, Yongri Liang b, Yong Tao a, Xin Wang a, Jian Zhao a, * a
Key Laboratory of Rubber-Plastics, Ministry of Education/Shandong Provincial Key Laboratory of Rubber-plastics, School of Polymer Science and Engineering, Qingdao University of Science and Technology, Qingdao 266042, China b State Key Laboratory of Polymer Physics and Chemistry, Joint Laboratory of Polymer Science and Materials, Beijing National Laboratory for Molecular Sciences, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 23 November 2011 Received in revised form 2 March 2012 Accepted 19 May 2012 Available online 30 May 2012
We report a novel, scalable and inexpensive approach to fully disperse carbon nanotubes in silicone rubber by the addition of graphene. In comparison to graphene, the dispersion of multi-walled carbon nanotubes (MWNTs) in silicone rubber matrix is extremely difficult although both of them possess similar physical structure. The different dispersion behavior of graphene and MWNTs could be contributed to the difference in their interaction with polymer matrix and their geometry. Based on SEM, TEM and XRD analysis, we find that the dispersion of MWNTs in silicone rubber is dramatically improved by the addition of graphene. Graphene acts as a compatilizer since it shows strong interaction with both polymer matrix and MWNTs. This method provides a simple route to enhance the dispersion of carbon nanotubes and improve the electrical property of the polymer composites. The synergic effect of the hybrid materials may not to be limited to the applications in polymer composites. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: Graphene Carbon nanotubes Silicone rubber
1. Introduction Since the discovery of carbon nanotubes (CNTs) [1] and graphene [2], they have received increasing interest in various fields such as chemistry, physics, materials science, and electrical engineering [1e16]. Both of the carbon materials, possess unique nanostructures with remarkable physical and mechanical properties [1e8]. They are widely used as nanofillers to improve electrical, thermal and optical properties of polymer composites [9e12]. The prospect of obtaining advanced nanocomposites with multifunctional features and materials used for structures and electrical conductors has attracted researchers’ attention in both academic and industrial communities [13e16], especially as electro statically dissipative materials and aerospace structural materials [17e20]. Although the research on carbon nanotube-based polymer composites has achieved significant progress, two major issues remain challenging for the fabrications of carbon nanotube reinforced polymer composites: (1) the dispersion of CNTs, i.e. the tendency of nanotubes to aggregate during processing; (2) the limited availability of high-quality carbon nanotubes in large quantities and high cost of their production. Graphene, a hexagonal network of carbon atoms, is actually the basic unit of a carbon * Corresponding authors. Tel.: þ86 532 88766117. E-mail addresses: [email protected], [email protected] (H. Hu), zhaojian@ gmail.com (J. Zhao). 0032-3861/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymer.2012.05.039
nanotube. Rolling graphene into a seamless, hollow cylinder forms the carbon nanotube [21,22]. In addition, although graphene and carbon nanotubes are similar in chemical composition, different topology distinguishes carbon nanotubes from graphene and offers them unique properties. One of the major advantages of graphene over carbon nanotubes lies in the precursor of graphenedgraphite, which is natural abundant [23]. It is essential for industrial applications in composite fields, since high yield production and a costeffective and scalable process is possible [24,25]. The investigations on graphene or carbon nanotube-based polymer composites are mainly focused on plastics matrices such as polyamide 6 (PA6) [26], polycarbonate (PC) [26,27], epoxy [28], polyurethane (PU) [29], polystyrene (PS) [30,31], poly (methyl methacrylate) (PMMA) [32,33] or poly (styrene-b-isoprene-bstyrene) (SIS) block copolymer [34]. There are only a few of studies on graphene or CNTs/rubber composites [35e38]. One important reason is that it is difficult for nanoparticles to be dispersed in rubber matrices because of their high viscosity or low surface energy [36]. Conductive rubbers are critically needed in the fields such as ESD (electrostatic discharge) and EMI (electromagnetic interference) shielding materials, electronics packaging, telecommunication antenna, mobile phone parts and frequency shielding coatings for aircraft and electronics [28,35e38]. However, the physical interaction between rubbers and nanoparticles are weaker than that between plastics and nanoparticles. It is of importance to understand the mechanism of the dispersion of nanoparticles in rubbers.
H. Hu et al. / Polymer 53 (2012) 3378e3385
Silicone rubber offers good tolerance to extremely low and high temperatures, being able to operate normally from 55 C to þ300 C. The carbon backbone of organic rubbers makes them susceptible to ozone, UV, heat and other aging factors while silicone rubber can perform well. This feature makes silicone rubber become one of the choices of elastomers in many extreme environments. Silicone rubber also has other superior properties such as the ease of changing its shape and filling the seals in a matrix. Therefore, silicone rubber was selected as the polymer matrix in this paper. One critical issue for the fabrication of nanocomposites is how to uniformly disperse nanofillers in polymer matrix. Many methods including physical and chemical methods were investigated to assist in the dispersion of CNTs [39]. Many reports investigated the synergetic effect of different fillers to make balance between properties and cost. Graphite (G) together with carbon fibers (CF) in polymer composites [40e47] has been studied for many years and excellent electrical conductivity of the composites has been obtained. Carbon nanotubes also have been investigated when blended with other fillers such as clay [48,49], single-walled carbon nanotubes (SWNTs) [50], graphite (G) and carbon black (CB) [51]. Nanoplatelets/carbon black/carbon nanotubes three-filler system in epoxy resin has also been reported [52]. Balberg et al. [53,54] proposed models to explain electrical percolation based on excluded volume theory. Sun et al. [51] extended the excluded volume theory and obtained an equation to predict the amount of fillers used to get the percolation when two fillers were used. It is commonly believed that two (or three) types of fillers can form cosupporting conductive networks in which the fibrous CF filler act as long distance charge transporters and the particulate fillers such as CB serves as an interconnection between the fibers by forming local conductive paths [42,43]. However, the dispersion of nanofillers in a synergetic system is rarely explored. In this work, we report a simple and effective way to disperse carbon nanotubes and graphene in silicone rubber. The excellent dispersion of CNTs can be achieved with the aid of graphene because of strong interaction between graphene and silicone rubber matrix as well as strong interaction between graphene and CNTs. The mechanism of dispersion will be discussed in detail in this paper. 2. Experimental section 2.1. Materials Silicone rubber (Methyl-Vinyl-Silicone, MVQ 110-2) was provided by Jiangsu Hongda Materials Company, China. The molecular weight of the transparent silicone rubber is about 4.50e5.9 105 g/mol. Natural flake graphite (LC50-9999, particle size 40 mm) was purchased from Qingdao Ruisheng Graphite Co. Ltd. China. Carboxylated multi-walled carbon nanotubes (MWNTs, MC5, purity > 95%, OD: 20e30 nm, Length: 10e30 mm, eCOOH content: 1.23 wt%, resistivity: 2.8 U cm) was purchased from Chengdu Institute of Organic Chemistry Co. Ltd. China. Curing agent 2,5-dimethyl-2,5-di(tert-butyl peroxy) hexane (AkzoNobel, Tianjin), is commercially available and used as received. 2.2. Preparation of graphene Graphite oxide (GO) was prepared according to the modified Hummers method first [55]. 5 g of flake graphite (40 mm), 3.75 g of NaNO3, and 230 mL of H2SO4 were mixed at 0 C. 15 g of KMnO4 was slowly added into the mixture within about 1 h, followed by stirring for 2 h in an ice-water bath. After additional stirring at room temperature for 5 days, 500 mL of 5 wt% H2SO4 was added in 1 h. The resultant mixture was further stirred for 2 h at 98 C. Then, the
3379
temperature was lowered to 60 C and 15 mL of H2O2 was added. The mixture was stirred for 2 h at room temperature and then precipitated, followed by being washed 7 times with an aqueous solution of 3 wt% H2SO4/0.5 wt% H2O2 and subsequent filtering. The resultant solid was further washed 3 times with 3 wt% and 0.6 wt% HCl solution. Finally, GO was washed with deionized water until the pH value was close to 7, followed by filtering and drying in a vacuum oven for one week. The dried GO samples were thermally treated at 1050 C for 5 min in a tubular furnace to obtain thermally reduced graphene (TrG). The detailed process and properties of graphene can be obtained elsewhere (see reference 10). 2.3. Preparation of graphene/carbon nanotubes/silicone rubber composites The rubber composites were fabricated as follows: (1) Silicone rubber (MVQ, before crosslinking) was dissolved in THF to obtain homogeneous solution. (2) Graphene (or MWNTs) was dispersed in THF. The mixture was sonicated for 1 h to get stable suspension. (3) The above (1) and (2) suspensions were mixed together and sonicated for 30 min to obtain homogeneous suspension. (4) THF was evaporated at 50 C, and the mixture was dried in vacuum oven. (5) Curing agent was added into the mixture, and then the resulting compound was moved to vulcanizing machine for curing followed by vulcanizing at 160 C for 12 min to obtain the composites. 2.4. Characterizations 2.4.1. X-ray diffraction (XRD) X-ray diffraction measurements (D-MAX 2500/PC, 18 kW rotating anode, Rigaku Corporation) were carried out to characterize the structure of graphene, and the composites with Cu Ka radiation (40 kV, 100 mA, wavelength, 1.5406 Å). X-ray diffraction patterns were obtained at room temperature with a scanning rate of 4 /min and the step size of 0.02 . 2.4.2. Scanning electron microscopy (SEM) Cryogenic fracture surfaces of the specimens were coated by sputtering with gold and then observed with field emission scanning electron microscopy (FESEM, Japan Electronics Co., Ltd., JSM6700F). 2.4.3. Transmission electron microscopy (TEM) Ultrathin specimen (thinner than 100 nm) for TEM observation were cryogenically cut with a diamond knife using an ultra thin microtome (UCT, Leica Ultracut, EMFC7) and collected on 200-mesh copper grids. The exfoliation of graphene nanosheets in silicone rubber was observed with transmission electron microscope (TEM, JEOL, JEM2100). 2.4.4. X-ray photoelectron spectroscopy (XPS) XPS experiments were carried out on a RBD upgraded PHI5000C ESCA system (Perkin Elmer) with Mg Ka radiation (hn ¼ 1253.6 eV) or Al Ka radiation (hn ¼ 1486.6 eV). Binding energies were calibrated by the containment carbon (C1s ¼ 284.6 eV). The data analysis was carried out using RBD AugerScan 3.21 software (RBD Enterprises, USA). 2.4.5. Resistivity analysis PC-68 high resistance meter (Shanghai Baihe Instrument Technology Co., Ltd, China) was used to test the resistance of silicone rubber composites. The resistivity was calculated through the formula:
r v ¼ p*d*R=ln2
3380
H. Hu et al. / Polymer 53 (2012) 3378e3385
where r is resistivity, d is the thickness of the sample, and R is the resistance of the sample. 2.4.6. Thermal diffusivity Thermal diffusivity was measured by NETZSCH LFA 447 at 25 C. Xenon lamp as heating source launched a bunch of pulse to the surface of sample, the increasing temperature was measured by infrared detector. 3. Results and discussion 3.1. Characterization of graphite oxide (GO) and thermally reduced graphene (TrG) FTIR, XRD and XPS were used to characterize graphite oxide (GO) and thermally reduced graphene (TrG). TrG was prepared as reported previously [10]. As shown in Fig. 1(a), the band at
1732 cm1 appeared in the FTIR spectrum of GO, indicating that carbonyl groups (vibration modes of C¼O at 1732 cm1 and CeOeC at 1065 cm1) were anchored on GO during the oxidation of graphite. The oxygen-containing groups of TrG almost disappeared, and the vibration peaks of graphene (w1385, w1446, and w1635 cm1) appeared, indicating that GO was fully reduced to TrG after thermal treatment. XRD (Fig. 1(b)) pattern shows that the interlayer spacing of GO (0.794 nm, 11.13 ) was larger than that of pristine graphite (0.335 nm, 26.6 ) [56] because of the intercalation of functional groups on GO. After thermal reduction of GO, the diffraction peak (11.13 ) of GO disappeared and a broad diffraction peak of about 25 appeared, implying that the distance between the intercalated graphite layers decreased and the layer structure was reconstructed during thermal reduction although remarkable amount of graphene was fully exfoliated by thermal expansion [57].
Fig. 1. The characterization of GO and TrG, (a) FTIR spectra of GO and TrG, (b) XRD profiles of GO and TrG, (c) C1s XPS spectrum of GO, (d) C1s XPS spectrum of TrG, (e) SEM image of GO, (f) TEM image of TrG.
H. Hu et al. / Polymer 53 (2012) 3378e3385
X-ray photoelectron spectroscopy (XPS) was used to probe functional groups on graphite oxide and thermally reduced graphene. C1s XPS spectrum of GO (Fig. 1(c)) shows the existence of oxygencontaining groups (CeO at 286.2 eV, and C]O at 287.8 eV) due to the oxidation of graphite. The peak area for CeC (284.5 eV) in the XPS spectrum of GO accounts for about 64 percent of total peak area, indicating that graphite has been fully oxidized. C1s XPS spectrum of TrG (Fig. 1(d)) shows that the peaks of CeO and C]O have dramatically decreased after 5 min pyrolysis reduction at 1050 C, but a few of C¼O and CeO groups were still retained on graphene, which can not be detected in the FTIR spectrum of TrG. These oxygen-containing groups play an important role in the dispersion of graphene. Fig. 1(e) is the SEM image of GO, and Fig. 1(f) is the TEM image of TrG. It can be seen that the surfaces of GO and TrG are not smooth but wrinkled. The original sp2 bonding of two-dimensional graphene evolved into a sp3 hybrid non-planar bonding, resulting in the disruption of the original plane structure. As a result, a wrinkled morphology was formed. This morphology is very favorable for the dispersion of graphene in rubber, which will be discussed in the following section. 3.2. Dispersion of multi-walled carbon nanotubes (MWNTs) in silicone rubber The agglomeration of carbon nanotubes (labeled by white circle) in silicone rubber/carbon nanotube (5 wt%) were observed by SEM and TEM as shown in Fig. 2 (a) and (b). Although ultrasonic treatment of carbon nanotubes have been carried out during the preparation of MWNTs/silicone rubber composites, MWNTs still can not be fully dispersed in the polymer matrix due to intrinsic strong attractive forces (Van der Waals force) of carbon nanotubes. Fig. 2 (c) shows XRD profiles of MWNTs, MVQ and MVQ/MWNTs (5 wt%) composite. The diffraction peaks of MWNTs were observed at 24.4 (002) and 43.9 (100), which were assigned to interlayer space in the radial direction and in-plane graphitic structure of MWNTs, respectively. The MVQ showed a diffraction peak at 12.1. In the XRD profiles of MVQ/MWNTs composite, the relative intensity and peak position of diffraction peak at 12.1 were almost not changed and the peaks at 24.4 and 43.9 were very week, which indicated that the MWNTs were contained in the nanocomposite and did not influence the structure of MVQ. 3.3. Dispersion of graphene in silicone rubber SEM, TEM and XRD of silicone rubber/graphene (3 wt%) composite were shown in Fig. 3. Fig. 3 (a) and (b) showed SEM images of silicone rubber/graphene composite. It could be clearly
3381
seen that almost all of graphene was well dispersed in the silicone rubber matrix, consistent with the results of TEM observation of silicone rubber/graphene composites (Fig. 3 (c)). The XRD profile of graphene (TrG) (Fig. 3 (d)) exhibited a broad diffraction peak at about 25.4 . When graphene was blended with silicone rubber (1 wt% or 3 wt%), the diffraction peak of TrG in those composites disappeared. It implies that graphene was almost exfoliated by silicone rubber and very well dispersed in the silicone rubber. Graphene and carbon nanotube have similar sp2 chemical structure, but they show dramatically different dispersion behavior in silicone rubber. The main reasons can be found in the following two aspects: The first aspect is that the morphology of the two carbon materials is quite different. Fig. 3 (b) is the amplification of the region of black dashed circle in Fig. 3 (a). We can find that the surface of graphene is not flat. Instead, a worm-like wrinkled morphology of graphene was observed. The original sp2 bonding of two-dimensional plane graphene evolved into a sp3 hybrid non-planar bonding, resulting in the disruption of the original plane structure. As a result, a wrinkled morphology was formed. This worm-like structure can prevent the restacking of graphene layers when mixing with silicone rubber. The ripples of graphene are very beneficial for the formation of occluded rubber in the silicone rubber composites, which limits the movement of molecular chains of silicone rubber. As a result, the strength and hardness of the graphene/silicone rubber composites are improved. Meanwhile, the wrinkled morphology renders the silicone rubber to be better integrated with graphene, which facilitates the dispersion of graphene in silicone rubber. The second aspect is that the interaction between graphene and silicone rubber is much stronger than that between MWNTs and silicone rubber. Schniepp [57] reported that the surface area of graphene is about 800 m2/g as measured by BET surface adsorption. The huge surface area of graphene offers numerous active adsorption sites, which render better interaction between the graphene and silicone rubber. In contrast with MWNTs, the two planes of graphene possess more contact area with silicone rubber. Also, a few of oxygen-containing groups were still retained on graphene (XPS spectrum in Fig. 1(d)). Hydrogen bonds may form when graphene contact with the oxygen-containing groups on the silicone rubber [36]. Thus, an effective interface bonding forms between the graphene and the silicone rubber. The mixing energy in the blend system can be described by eq. (1):
DGM ¼ DHM T DSM
(1)
When graphene or MWNTs is mixed with silicone rubber, the entropy changes little, because conformation entropy change little
Fig. 2. MWNTs agglomerated in silicone rubber/MWNTs (5 wt%), (a) SEM image of MVQ/MWNTs (5 wt%) (10000), (b) TEM image of MVQ/MWNTs (5 wt%) (150000), (c) XRD profiles of MWNTs, MVQ and MVQ/MWNTs (5 wt%) composite.
3382
H. Hu et al. / Polymer 53 (2012) 3378e3385
Fig. 3. SEM, TEM images and XRD profiles of MVQ/TrG (3 wt%), (a) SEM image, (b) SEM image, (c) TEM image, (d) XRD profiles of TrG, MVQ and MVQ/TrG composite.
and entropy change mainly arises from the position change of graphene or MWNTs. In the mixing process, the entropy is positive, which is beneficial for the dispersion of graphene or MWNTs into silicone rubber. The main factor that prevents nanoparticles from being dispersed in silicone rubber or other matrices is the effect of enthalpy. The interaction between the nanoparticles is strong due to the huge surface energy of nanoparticles. There are three competing interactions in the system when nanoparticles are mixed with polymers. The first is the interaction for a polymerpolymer contact (DHp-p), the second is the interaction for a nanoparticleenanoparticle contact (DHn-n), and the third is the interaction for a polymer-nanoparticle contact (DHp-n). The first and second interactions will cause the phase separation of polymer and nanoparticles, which is conducive to the aggregation of nanoparticles. The third interaction will favor the dispersion of nanoparticles into the polymer. The difference in the dispersion of graphene and MWNTs is caused by the interaction between the nanoparticles and polymers. DHp-p and DHn-n are almost the same for graphene and MWNTs as they have the same elements——carbon and the sp2 structure of CeC are similar for both graphene and MWNTs. The difference arises from the DHp-n. Good dispersion of Graphene in polymers implies that the DHp-G (G represents graphene) is much higher than DHp-c (C represents carbon nanotube). Suppose that the interactions among graphene and polymer chains are the sum of Van der Waals forces. The force is weak for only one carbon atom interacting with a silicone atom. There are
numerous interactions in a sample, so the sum is huge enough. Let us compare the difference between graphene and carbon nanotubes. The elements of graphene and carbon nanotubes are the same, meaning that they have the same Van der Waals forces between the carbon elements and other elements of the polymer. The difference lies in the shape of these two nanoparticles. Both sides of a graphene plane can interact with the polymer, but only the outer layer of MWNTs can do. The total Van der Waals forces among graphene and the polymer are doubled as compared to that among the polymer and MWNTs. The total DH of the graphene/ polymer system will be much smaller than that of the MWNTs/ polymer system. As a result, it is easier for graphene to be dispersed in the polymer. 3.4. Graphene-assisted dispersion of MWNTs in silicone rubber When Graphene/carbon nanotubes/silicone rubber were blended together, an interesting phenomenon was observed. Fig. 4 shows the SEM, TEM images and XRD profiles of the system. It can be seen from the SEM and TEM images (Fig. 4 (a)e(e)) that graphene and MWNTs were well dispersed in the silicone rubber. And the graphene are dispersed around carbon nanotubes. Carbon nanotubes and graphene together form a conductive network (as indicated by the black lines in Fig. 4 (a) and (b)) in the silicone rubber matrix. Graphene is well dispersed in the silicone rubber matrix, and the aggregation of carbon nanotube in the matrix was not found in SEM or TEM images.
H. Hu et al. / Polymer 53 (2012) 3378e3385
XRD profiles (Fig. 4(f)) showed that when the graphene was mixed with MWNTs and silicone rubber (silicone rubber/graphene/ MWNTs ¼ 100/1/2.5), the diffraction peak of interlayer structure of TrG (25.4 ) was not found, whereas the diffraction peak of MWNTs (24.4 ) was still observed as showed in Fig. 4(f), indicating that graphene was still well dispersed in the silicone rubber matrix regardless of the presence of MWNTs. The dispersion of graphene in silicone rubber was not influenced by the presence of MWNTs. The dispersion of MWNTs can not be assessed by XRD measurement. However, the SEM and TEM images indicate that both MWNTs and TrG are well dispersed in silicone rubber matrix.
3383
The excellent dispersion of MWNTs and graphene in the blend system may be attributed to the following reasons: The diameter of graphene is in the range of tens of nanometer to several micrometers, while the dimension (length) of MWNT is 10e30 mm. Graphene is smaller in particle size as compared with MWNTs, and it can act as a separator between MWNTs. The force that the sonication provides results in the intercalation of graphene into nanotube bundles. A graphene sheet may “adhere” to many carbon nanotubes due to the strong interaction between carbon nanotubes and graphene, leading to the disruption of carbon nanotube bundles. Once graphene “adheres” to MWNTs, it is not
Fig. 4. Silicone rubber/graphene/MWNTs ¼ 100/1/2.5, (a), (b)SEM images, (c) TEM image, the insert is the enlargement of carbon nanotube in the white rectangle, (d) TEM image, (e) TEM image of silicone rubber/graphene/MWNTs ¼ 100/1/3, (f) XRD profiles of MWNT, TrG, MVQ and MVQ composites. (The arrows point to the MWNTs and graphene are in white circles).
H. Hu et al. / Polymer 53 (2012) 3378e3385
easy for graphene to be “driven out” from MWNTs because graphene possesses the extended p electron system similar to that of carbon nanotubes. There exists a good interaction between graphene and silicone rubber matrix. On one hand, the eOHe on graphene may form hydrogen bonds with the eOe on silicone rubber. On the other hand, there is huge contact surface area between silicone rubber and graphene because graphene has very large specific surface area [57] and each plane of graphene can contact with silicone rubber. Graphene can be dispersed uniformly in silicone rubber as discussed above. Since graphene has good interaction with silicone rubber and carbon nanotubes, graphene plays a role of compatilizer for MWNTs and silicone rubber, which facilitates the dispersion of MWNTs in silicone rubber. Based on the aforementioned analysis, we believe that the most important factor for the excellent dispersion of MWNTs assisted by graphene is the strong interaction between graphene and the matrix as well as the strong interaction between graphene and MWNTs. Graphene has a large surface area and carbon nanotubes have a great aspect ratio. One dimensional and two-dimensional materials cooperate, forming a “conductive bridge” in the silicone rubber as shown in the schematic diagram (Fig. 5). 3.5. Effect of the dispersion on electrical properties of silicone rubber composites With the increase of filler content, the volume resistivity of silicone rubber decreased (Fig. 6). The percolation threshold for graphene is about 2 wt% with volume resistivity of 1 105 U cm. By contrast, the percolation threshold for MWNTs is about 5 wt% with volume resistivity of 4.21 106 U cm. The graphene with twodimensional planar structure is better dispersed than MWNTs, which favors the formation of conductive network. MWNTs tend to aggregate in silicone rubber, leading to larger percolation value. When both graphene and carbon nanotubes are filled in silicone rubber composites, carbon nanotube is well dispersed with the aid of graphene and the conductive network forms. Graphene is distributed around the carbon nanotubes. Carbon nanotubes play as a “conductive bridge” connection between graphene layers and the matrix. At the graphene content of 1 wt% and the MWNT content of 2 wt%, the volume resistivity of the composite was 4.91 109 U cm. At the graphene content of 1 wt% and the MWNT content of 2.5 wt%, the volume resistivity was reduced to 8.0 104 U cm.
Fig. 6. The resistivity of silicone rubber composites.
0.13
Thermal diffusivity (m2 /s)
3384
0.12
0.11
0.10
0.09
MVQ
TrG(1)
MWNTs (5)
TrG/MWNTs (1/3)
Fig. 7. Thermal conductivity of silicone rubber composites.
3.6. Effect of the dispersion on thermal diffusivity of silicone rubber composites Thermal diffusivity of silicone rubber composites was shown in Fig. 7. The thermal diffusivity of pure silicone rubber was 0.105 m2/s, which was increased by the addition of graphene or WMNTs or two fillers together. Thermal diffusivity of silicone rubber composites were 0.113 and 0.119 m2/s respectively by filling graphene (1 wt%) and MWNTs (5 wt%). By contrast, thermal diffusivity of silicone rubber composite was enhanced to 0.123 m2/s by filling 1 wt% graphene and 3 wt% nanotubes together. Graphene has strong interaction with silicone rubber and also breaks up the agglomeration of MWNTs. The remarkable enhancement is due to the synergistic effect of graphene and MWNTs. 4. Conclusion
Fig. 5. Schematic of the synergetic interaction between graphene and carbon nanotubes.
In summary, we studied the dispersion of MWNTs and graphene in silicone rubber matrix. It is difficult to disperse MWNTs in silicone rubber alone, whereas it is much easier for MWNTs to be dispersed in the matrix with the aid of graphene. The strong interaction between graphene and silicone rubber as well as the strong interaction between graphene and MWNTs accounts for the good dispersion of MWNTs. Graphene behaves as a compatilizer. Both physical isolation between graphene and MWNTs and some hydrogen bonding between graphene and MWNTs may form during the dispersion of MWNTs assisted by graphene. Addition of
H. Hu et al. / Polymer 53 (2012) 3378e3385
graphene provides a simple route to disperse carbon nanotube and the synergetic effect facilitates the fabrication of high performance polymer composites. Acknowledgments This project is supported by the NSFC (Grant No.20944004 and 51073082), the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry, MOE Key Laboratory of Molecular Engineering of Polymers (Fudan University), MOE Key Laboratory of Organosilicon Chemistry and Material Technology of Ministry of Education (Hangzhou Normal University) Qingdao Municipal Science and Technology Program, Basic Research Project (12-1-4-3-(7)-jch). References [1] Iijima S. Nature 1991;354:56e8. [2] Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y, Dubonos SV, et al. Science 2004;306:666e9. [3] Geim AK, Novoselov KS. Nat Mater 2007;6:183e91. [4] Song P, Cao Z, Cai Y, Zhao L, Fang Z, Fu S. Polymer 2011;52:4001e10. [5] Soldano C, Mahmood A, Dujardin E. Carbon 2010;48:2127e50. [6] Luong ND, Hippi U, Korhonen JT, Soininen AJ, Ruokolainen J, Johansson L, et al. Polymer 2011;52:5237e42. [7] Rafiee MA, Rafiee J, Wang Z, Song HH, Yu ZZ, Koratkar N. ACS Nano 2009; 3(12):3884e9. [8] Park S, Ruoff RS. Nat Nanotechnology 2009;4Z:217e27. [9] Ajayan PM, Schadler LS, Giannaris C, Rubio A. Adv Mater 2000;10:750e3. [10] Hu H, Liu Y, Wang Q, Zhao J, Liang Y. Mater Lett 2011;65:2582e4. [11] Chen WF, Yan LF. Nanoscale 2010;2:559e63. [12] Sheng K, Bai H, Sun Y, Li C, Shi G. Polymer 2011;52(24):5567e72. [13] Hu HQ, Yu JH, Li YY, Zhao J, Dong HQ. J Biomed Mater Res A 2012;100A(1): 141e8. [14] Allaoui A, Bai S, Cheng HM, Bai JB. Comp Sci.Technol 2002;62(15):1993e8. [15] Dong X, Su CY, Zhang W, Zhao J, Ling Q, Huang W, et al. Phys Chem Chem Phys 2010;12:2164e9. [16] Zhao X, Zhang QH, Chen DJ. Mcromolecules 2010;43:2357e63. [17] Jin Z, Pramoda KP, Xu G, Goh SH. Chem.Phys Lett 2001;337:43e7. [18] Li B, Olson E, Perugini A, Zhong W. Polymer 2011;52:5606e14. [19] Vajtai R, Wei BQ, Zhang ZJ, Jung Y, Ramanath G, Ajayan PM. Smart Mater Struct 2002;11:691e8. [20] Liang JJ, Xu YF, Huang Y, Zhang L, Wang Y, Ma YF, et al. J Phys Chem C 2009; 113:9921e7. [21] Dresselhaus MS, Lin YM, Rabin O, Jorio A, Souza Filho AG, Pimenta MA, et al. Mater Sci Eng 2003;C23:129e40. [22] Kim H, Abdala AA, Macosko CW. Macromolecules 2010;43:6515e30. [23] Kotov NA. Nature 2006;442:254e5.
3385
[24] Park S, Ruoff RS. Nat Nanotechnol 2009;4:217e24. [25] Vaia RA, Wagner HD. Mater Today 2004;7:32e7. [26] Steurer P, Wissert R, Thomann R, Mulhaupt R. Macromol Rapid Commun 2009;30:316e27. [27] Kim H, Macosko CW. Polymer 2009;50:3797e809. [28] Zaman I, ThanhPhan T, Kuan H, Meng Q, La LT, Luong L, et al. Polymer 2011; 52:1603e11. [29] Tkalya E, Ghislandi M, Alekseev A, Koning C, Loos J. J Mater Chem 2010;20: 3035e9. [30] Kim H, Miura Y, Macosko CW. Chem Mater 2010;22:3441e50. [31] Stankovich S, Dikin DA, Dommett GHB, Kohlhaas KM, Zimney EJ, Stach EA, et al. Nature 2006;442:282e6. [32] Das B, Prasad KE, Ramamurty U, Rao CNR. Nanotechnology 2009;20: 125701e5. [33] Layek RK, Samanta S, Chatterjee DP, Nandi AK. Polymer 2010;51:5846e56. [34] Peponi L, Tercjak A, Verdejo R, Lopez-Manchado MA, Mondragon I, Kenny JM. J Phys Chem C 2009;113:17973e8. [35] Verdejo R, Bernal MM, Romasanta LJ, Lopez-Manchado MA. J Mater Chem 2011;21:3301e10. [36] Jiang MJ, Dang ZM, Xu HP. Eur Polym J 2007;43:4924e30. [37] Zhang H, Zheng W, Yan Q, Yang Y, Wang J, Lu Z, et al. Polymer 2010;51: 1191e6. [38] Prud’homme RK, Ozbas B, Aksay IA, Register RA, Adamson DH. International patent, WO 2008/045778 A1; 2008. [39] Moniruzzaman M, Winey KI. Macromolecules 2006;39:5194e205. [40] Ma PC, Liu MY, Zhang H, Wang SQ, Wang R, Wang K, et al. ACS Appl Mat Interfaces 2009;1(5):1090e6. [41] Lee GW, Park M, Kim J, Lee JI, Yoon HG. Composites: Part A 2006;37: 727e34. [42] Clingerman LM, Weber EH, King JA, Schulz KH. Polym Compos 2002;23: 911e24. [43] Thongruang W, Spontak RJ, Balik CM. Polymer 2002;43:2279e86. [44] Drubetski M, Siegmann A, Nakis M. Polym Compos 2005;26:454e64. [45] Lee JH, Kim SK, Kim NH. Scr Mater 2006;55:1119e22. [46] Drubetski M, Siegmann A, Nakis M. J Mater Sci 2007;42:1e8. [47] EI-Bounia NE, Piccione PM. J Polym Eng 2008;28:141e54. [48] Wang Z, Meng XY, Li JZ, Du XH, Li SY, Jiang ZW, et al. J Phys Chem C 2009;113: 8058e64. [49] Tang C, Xiang L, Su J, Wang K, Yang C, Zhang Q, et al. J Phys Chem B 2008;112: 3876e81. [50] Brady-Estevez AS, Schnoor MH, Kang S, Elimelech M. Langmuir 2010;26(24): 19153e8. [51] Sun Y, Bao HD, Guo ZX, Yu J. Macromolecules 2009;42:459e63. [52] Wei T, Song L, Zheng C, Wang K, Yan J, Shao Bo, et al. Mater Lett 2010;64: 2376e9. [53] Balberg I. Phys Rev B 1986;33:3618e20. [54] Celzard A, McRae E, Deleuze C, Dufort M, Furdin G, Mareˆche0 JF. Phys Rev B 1996;53:6209e14. [55] Becerril HA, Mao J, Liu ZF, Stoltenberg RM, Bao ZN, Chen YS. ACS Nano 2008;2: 463e70. [56] Guo HL, Wang XF, Qian QY, Wang FB, Xia XH. J.ACS Nano 2009;3:2653e9. [57] Schniepp HC, Li JL, McAllister MJ, Sai H, Herrera-Alonso M, Adamson DH, et al. Phys Chem B 2006;110(17):8535e9.