reduced graphene oxide hybrid for

Carbon 123 (2017) 158e167

Contents lists available at ScienceDirect

Carbon journal homepage: www.elsevier.com/locate/carbon

Carbon nanotube/reduced graphene oxide hybrid for simultaneously enhancing the thermal conductivity and mechanical properties of styrene -butadiene rubber Shiqiang Song a, b, Yong Zhang a, * a

School of Chemistry and Chemical Engineering, State Key Laboratory for Metal Matrix Composite Materials, Shanghai Jiao Tong University, Shanghai 200240, PR China Shanghai Key Laboratory of Electrical Insulation and Thermal Aging, Shanghai Jiao Tong University, Shanghai 200240, PR China

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 5 May 2017 Received in revised form 7 July 2017 Accepted 17 July 2017 Available online 18 July 2017

A novel hybrid of polymer functionalized-carbon nanotube @ reduced graphene oxide (PCNT@RGO) was prepared by reversible addition-fragmentation chain transfer polymerization, esterification reaction and reduction process. PCNT@RGO hybrid as a filler can effectively enhance the thermal conductivity and mechanical properties of styrene-butadiene rubber (SBR) vulcanizate. Remarkably, the tensile strength and stress at 200% extension of the SBR vulcanizate reached as high as 8.8 and 7.7 MPa at PCNT@RGO loading of 3 wt%, increasing by approaching 203% and 450% compared with neat SBR. Furthermore, the vulcanizate also exhibited high thermal conductivity (0.45 W/m K), which is 2.0-fold higher than that of SBR. The improvement in properties could be attributed to the synergetic effect of PCNT and RGO in PCNT@RGO hybrid, strong interfacial between PCNT@RGO and SBR matrix, the large constrained regions and continuous filler networks. Such good performances render the SBR/PCNR@RGO composite appealing for the use in various engineering practices. © 2017 Published by Elsevier Ltd.

1. Introduction With the development of high-degree integration and functionalization of materials, the research of materials with excellent comprehensive performance, especially good thermal conductivity has drawn much attention recently [1,2]. Therein, many novel polymer composites with enhanced thermal conductivity and other properties have applied to meet the demand [3,4]. Based on many methods to enhance the thermal conductivity of polymers, the addition of nanofillers was indispensable for achieving high thermal conductivity and other excellent properties [5,6]. Among a variety of fillers, carbon-based fillers were considered as an ideal filler, which can simultaneously enhance the thermal conductivity and mechanical properties of polymers because of their high intrinsic thermal conductivity and strength at nanoscale, such as carbon nanotube (CNT) (1000e4000 W m1 k1 and: over 60 GPa) [7,8], and graphene (2000e6000 W m1 k1 and strength: 63 GPa) [9,10]. For example, it can achieve a strong enhancement of the

* Corresponding author. E-mail address: [email protected] (Y. Zhang). http://dx.doi.org/10.1016/j.carbon.2017.07.057 0008-6223/© 2017 Published by Elsevier Ltd.

mechanical properties and thermal conductivity of polypropylene using an eco-friendly strategy making graphene sheets welldispersed in the polymer matrix [11]. The effect of the three different click chemistry routes used to functionalize graphene with short-chain polyethylene on the mechanical properties and thermal conductivity of graphene-based high density polyethylene nanocomposites was reported [12]. Dean et al. [13] studied the increase in modulus and thermal conductivity of a magnetically aligned carbon nanotube/epoxy nanocomposite. However, except for the type of filler limiting the mechanical properties and thermal conductivity of composites, the interaction of fillers with polymer matrix also has an important influence on that of composites. Many experiments have shown that the poor interface between filler and polymer could result in high thermal interfacial resistance, interfacial phonon scattering and poor dispersion of filler, which reduced the thermal conductivity and mechanical properties [14e16]. Therefore, improving polymer-filler interfacial interaction can increase the overall thermal conductivity and mechanical properties of composites quite substantially. Additionally, creating a continuous filler network was also the key to obtaining high performance composites. It provided two major benefits: (1) the filler network could reduce the thermal interfacial resistance and interfacial phonon scattering,

S. Song, Y. Zhang / Carbon 123 (2017) 158e167

and form efficient thermally conductive pathways. (2) It could also effectively avoid the aggregation of filler, resulting in enhancing the mechanical properties of composites. Wong et al. [17] fabricated epoxy/graphene composites with a vertically aligned and interconnected graphene network, having a dramatic thermal conductivity enhancement of 1231% compared to the pure matrix. Joseph et al. [18] demonstrated that the segregated multiwalled carbon nanotubes networks in natural rubber (NR) latex led to increasing tensile strength, tensile modulus and tear strength by 61%, 75% and 59% respectively, at 0.5 phr MWCNT. Specially, to improve the thermal conductivity and mechanical properties of styrene -butadiene rubber (SBR), which was an important component of tire tread, highlights the significance both in academic and industrial studies, due to its the poor thermal conduction and mechanical properties [4,19]. The higher thermal conductivity of SBR will be beneficial to transfer heat, which generated in the deformation process under dynamic external force, to the external environment. Accordingly, the enhanced mechanical properties and thermal conductivity of SBR will significantly affect the aging resistance and lifetime of tire. Thereby, many researchers have paid much attention on improving these performances of SBR [20,21]. However, lots of fillers should be incorporated to meet the performance requirements, resulting in deterioration stretchability and processing difficulty. Thus, it remains a big challenge to achieve these performances at lower filler loadings. In terms of the above discussion, in the present study, we successfully prepared polymer functionalized-CNT (PCNT) by reacting carboxylated CNT with a polystyrene-block-poly(2-hydroxyethyl methacrylate) (PS-b-PHEMA) copolymer synthesized via reversible addition fragmentation chain transfer polymerization (RAFT), resulting in improving the interaction with SBR matrix. PCNT then reacted with graphene oxide (GO) by esterification reaction and led to PCNT@GO hybrid, which was subjected to reduction to obtain a PCNR@RGO hybrid. PCNT@RGO was incorporated into SBR/tetrahydrofuran (THF) solution to obtain SBR/PCNT@RGO composite. The composite exhibited high thermal conductivity and good mechanical properties. The synergetic effect of PCNT and RGO in PCNT@RGO hybrid, strong interfacial between PCNT@RGO and matrix, the large constrained regions and continuous filler networks were responsible for the concurrently improved mechanical properties and thermal conductivity. 2. Structure and properties characterization Fourier transform infrared spectroscopy (FTIR) spectra were recorded on a Perkin-Elmer Paragon 1000 PC spectrometer. The chemical state of the surface was characterized by X-ray photoelectron spectroscopy (XPS) on a Kratos AXIS Ultra DLD spectrometer using a monochromatic Al Ka radiation (hy ¼ 1486.6 eV). Mn, Mw and PDI (Mw/Mn) were measured with GPC (HLC-8320GPC, TOSOH, Japan), with PMMA as a reference standard and DMF as an eluent. 1H-NMR (400 MHz) spectra were measured using a Bruker spectrometer in CDCl3 at ambient temperature. The exact weight of PS-b-PHEMA in PCNT, the fillers contents of SBR composites, and the thermal performance of GO were measured by thermogravimetric analysis in nitrogen with a TA Instruments model Q2000 at a heating rate 20 C/min. Raman shifts of CNT-COOH, PCNT@GO and PCNT@RGO were measured by a Raman microscopy (DXR, Thermo Scientific, USA) with 532 nm laser excitation. The filler morphologies were observed by transmission electron microscopy (TEM) (JEM-2100, JEOL Co., Japan). Samples were prepared by dropping the PCNT@GO/DMF and PCNT@RGO/DMF dispersion onto the copper grid, and dried at room temperature overnight. To measure the filler dispersion in composites, transmission electron microscope (TEM) images of ultrathin sections

159

prepared by a microtome (UC6, Leica Microsystems, Germany) were obtained by a biology TEM (Tecnai G2, FEI, USA) with an acceleration voltage of 120 kV. To record surface topology and elastic modulus maps, atomic force microscope (AFM, Bruker, German) was used in PeakForce quantitative nanomechanical property mapping (QNM) mode. Under PeakForce QNM mode, the formation of images is based on difference of modulus or adhesion. Mechanical properties were tested at an extension rate of 200 mm/min on a material tester (Instron 4465, Instron Corp, USA) according to ASTM D412. The samples were cut into strips of 30 4 mm2 with a razor blade and five strips were measured for each sample. Thermal conductivities were determined with LFA 447 Nanoflash (NETZSCH, Germany) on cylindrical samples (1e2 mm in thickness, 12.6 mm in diameter). 3. Results and discussion 3.1. Characterization and structure of PCNT@RGO hybrid The preparation procedure of PCNT@RGO hybrid was shown in Fig. 1, where PCNT was obtained by covalent grafting of PS-b-PHEMA onto MWCNT by “grafting to” method. FTIR spectroscopy was used to characterize the esterification reaction. As shown in Fig. 2, the absorption peaks of MWCNT-COOH at 3441 and 1727 cm1 are assigned to the hydroxyl (OH) and carbonyl groups (C¼O) stretching vibration of -COOH units on the surface of MWCNT [22]. However, after the chemical grafting, in the spectrum of PCNT, the new peaks appeared at 3000e2800, 1743, 1580 and 1460 cm1, corresponding to the CH3/ CH2 stretching vibration, -C¼O stretching vibration, and benzene ring skeleton vibration in PS-b-PHEMA chain, respectively [23,24].There is also a carbonyl group stretching vibration at 1656 cm1, which is attributed to the carbonyl group of MWCNT [25]. The observations demonstrate that the PS-b-PHEMA has been successfully covalently grafted to the surface of MWCNT. For PCNT@GO, the intensity of the absorption peak of O-H (3441 cm1) stretching vibration of -COOH decreased and the intensity of the absorption peak of epoxy/ether groups (1258 cm1) increased, demonstrate the esterification reaction has occurred between carboxyl (-COOH of PCNT) and hydroxyl groups (-OH of GO) with the catalysis of DCC and DMAP [26]. As well known, ascorbic acid is utilized as a reducing agent due to its nontoxic and highly reducing ability toward GO [27]. PCNT@GO was then subjected to reduction by ascorbic acid to yield PCNT@RGO. The evidence was also provided by FTIR spectra. As seen from the FTIR spectrum of PCNT@RGO, the peaks of O-H (3441 cm1) and epoxy/ ether groups (1258 cm1) stretching vibration decreased or even disappeared, indicating that ascorbic acid got rid of residual oxygencontaining groups of GO, which contributed to heat transfer of composites. XPS also provided additional evidence for the esterification reaction and the GO reduction by ascorbic acid. High-resolution C1s XPS spectrum of PCNT (Fig. 2) can be fitted into four peaks with the binding energy at 284.8, 285.8, 287.2, and 288.8 eV, corresponding to the C¼C/C-C, C-OH, C¼O, and O-C¼O, respectively [28]. While high intensity of C-OH signal may be derived from the copolymer PS-b-PHEMA, which was successfully coated on MWCNT surface. The deconvoluted C1s XPS spectrum for PCNT@GO (Fig. 2) has four signals, namely C¼C/C-C (284.8 eV), C-OH (285.7 eV), C-O-C (286.7 eV), C¼O (287.5 eV), and O-C¼O (288.5), respectively. Compared with the C1s XPS spectrum of PCNT, the C1s spectrum of PCNT@GO has a new component of C-O-C, and higher peak intensities of C¼O/O-C¼O. The results prove the esterification reaction between carboxyl groups in PCNT and hydroxyl groups in GO. It should be noted that the intensity of C-OH in PCNT@GO is higher than that in PCNT, which may be attributed to a large number of hydroxyl groups on the surface of GO. However, after chemical

160


Fig. 1. Schematic diagram illustrating the preparation process of PCNT@RGO. (A colour version of this figure can be viewed online.)

Fig. 2. (a) FTIR spectra of CNT-COOH, PCNT, PCNT@GO and PCNT@RGO hybrids. (b) Deconvoluted XPS spectra in the C1S region: PCNT, PCNT@GO and PCNT@RGO. (A colour version of this figure can be viewed online.)

reduction by ascorbic acid, the C-OH peak in PCNT@RGO disappeared and the peak intensity of C-O-C (Fig. 2) was significantly decreased. Furthermore, the C/O atom ratio of PCNT@RGO is 9.78, which is higher than that of PCNT@GO (5.14). The results provide a strong support for the GO reduction by ascorbic acid, which is also consistent with the results of FTIR. The structure of PCNT, PCNT@GO, and PCNT@RGO were characterized by XRD, as shown in Fig. 3a. The characteristic (002) and (101) diffraction peaks of PCNT appeared at around 2q ¼ 25.6 and 43.0 [29]. PCNT@GO, and PCNT@RGO hybrids exhibited similar features, demonstrating PCNT existed in hybrids. However, compared with PCNT, PCNT@GO hybrids had a new diffraction peak at 2q ¼ 9.5 , which is contributed to the (002) diffraction peaks of GO [29]. Compared with the peak position of original GO

(2q ¼ 10.5 ) (Fig. S3), the peak position of GO in hybrids shifted down, indicating the PCNT hampered close contact between the graphene basal planes. After the reduction by ascorbic acid, the sharp peak from GO disappeared and a broader diffraction peak shifted back to approximately 25.6 , which is coincided with the peak of (002) of PCNT. The above results indicate the PCNT@GO hybrid was successfully prepared and the GO was reduced to RGO with the assistance of ascorbic acid. It could be beneficial in improving the mechanical and thermal conductivity properties. Fig. 3b shows the Raman spectra of PCNT, PCNT@GO, and PCNT@RGO. The intensity ratio of D and G band (ID/IG) can be used to estimate the amount of defects and disorder in the carbon materials [30]. The higher ID/IG (1.32) could be attributed to the grafting of copolymer to MWCNT. Similarly, the ID/IG rations of


161

Fig. 3. (a) XRD patterns and (b) Raman spectra of PCNT, PCNT@GO-1, PCNT@RGO-1, PCNT@GO-2, and PCNT@RGO-2. (A colour version of this figure can be viewed online.)

PCNT@GO-1 and PCNT@GO-2 are 1.26 and 1.25, respectively, indicating the existence of defects and disordered structure in PCNT@GO hybrids. However all the PCNT@RGO hybrids have a smaller value of the ratio of ID/IG comparison with PCNT@GO hybrids, indicating the successful reduction of GO by ascorbic acid and the few structural defects of PCNT@RGO hybrids [31]. The structures and morphologies of GO, PCNT, PCNT@GO, and PCNT@RGO are clearly shown by AFM and TEM images in Fig. 4. Fig. 4a shows that the thickness of GO sheets was about 1.0 nm, indicating GO sheets mostly existed in form of monolayer in DMF. Additionally, the TEM image of PCNT (Fig. 4b) clearly shows that the smooth surface of MWCNT is coved with 2e3 nm thick PS-bPHEMA. The result demonstrates the successful grafting of the copolymer to MWCNT with the method of “grafting to”. It is also consistent with the results of FTIR and XPS. Fig. 4ced shows that PCNT and GO/RGO are uniformly recombined in nanoscale dimension, and GO nanosheets are connected each other by PCNT, resulting in avoiding stacking of the GO/RGO sheets and reducing its interface thermal resistance. In Fig. 4c (inset), selected area

electron diffraction (SAED) pattern of GO shows only diffraction rings, indicating that GO nanosheets are amorphous. While the well-defined diffraction spots in Fig. 4d (inset) confirm the crystalline structure of the graphene nanosheets [32]. The results indicate GO nanosheets have been successfully reduced by ascorbic acid, which could increase in thermal conductivity. The magnified and detailed schematic drawing of the process of PCNT@GO converting to the PCNT@RGO is shown in Fig. 4e. 3.2. Thermal conductivity of SBR composites As known, CNTs, and graphene have high thermal conductivity. Their use as filler has been regarded as an effective approach to obtain composites with high thermal conductivity. To investigate the effect of fillers on the thermal conductivity, Fig. 5a compares the thermal conductivity of the composites as a function of the PCNT or RGO content. The results demonstrate that the thermal conductivity of composites increases with increasing PCNT or RGO content. The maximum thermal conductivities are 0.26 and 0.29 (W/m$K),

Fig. 4. (a) AFM image of GO sheets deposited on mica and the mean thickness is ~0.92 nm (b) TEM image of PCNT and the thickness of the polymer grafted on the surface of CNT is 2e3 nm. (c-d) TEM images of PCNT@GO-1 and PCNT@RGO-2 with the corresponding SAED pattern (insets). (e) Detailed schematic construction drawing for the reduction process of PCNT@GO. (A colour version of this figure can be viewed online.)

162


Fig. 5. (a) Thermal conductivity of the composites as a function of PCNT or RGO loadings (b) Thermal conductivity of different fillers (3%) in SBR composites. (c) Calculated Thermal conductivity enhancement of the composites per 1 vol% filler loading. (d) Calculated Thermal conductivity enhancement of different fillers (3%) in SBR composites. (A colour version of this figure can be viewed online.)

respectively, at PCNT and RGO content of 3%, increasing by 13.0% and 26.1%. Moreover, it also can found that the thermal conductivity of SBR/PCNT composites is superior to that of the corresponding SBR/RGO composites. The high thermal conductivity of SBR/PCNT composites should be related to good dispersion of filler in SBR matrix. Furthermore, to investigate the synergetic effect of PCNT and RGO, two different ration 1:1 and 2:1 hybrids (PCNT/ RGO) were prepared by keeping the total filler content (3%) unchanged. As shown in Fig. 5b, higher PCNT/RGO ration led to higher thermal conductivity of composites, and 3% PCNT/RGO (2:1) in hybrid had the thermal conductivity of 0.45 (W/m$K), increasing by 96% in comparison with SBR. The results indicate that high PCNT/ RGO ratio is beneficial to the formation of efficient thermal bridges among assembled hybrids and hamper the agglomeration of RGO nanosheets, thereby enhancing the interface with the matrix. To further understand the enhancement of PCNT and RGO on thermal conductivity, it is introduced that the enhancement of thermal conductivity per 1 vol % filler loading (h) [33], which is defined as

h¼

K Km 100% 100 Vf Km

(1)

where Vf is the volume fraction of fillers, and the densities of MWCNT and RGO are 2.06 and 1.06 g/cm3, respectively. [34] K and

Km are the thermal conductivities of the composite and matrix, respectively. As shown in Fig. 5c, h for both SBR/PCNT and SBR/RGO composites increase with increasing PCNT and RGO content. This may be ascribed to formation of thermal conduction network at high PCNT and RGO content. It is also noted that h of SBR/PCNT is higher than that of the corresponding SBR/RGO, which would be accounted from the better dispersion of PCNT than RGO in polymer matrix. To further explore the synergetic effect of PCNT and RGO in PCNT@RGO hybrid, h values of SBR/RGO, SBR/PCNT, SBR/PCNT@RGO1, and SBR/PCNT@RGO-2 are shown in Fig. 5d at a given filler loading of 3%. One can see that SBR/PCNT@RGO-1 and SBR/PCNT@RGO-2 have high h values. As the PCNT/RGO ratio reaches 2:1, SBR/PCNT@RGO-2 composite exhibited the maximum h value (~53.5%) which is much higher than that of SBR/PCNT (~20.6%) and SBR/RGO (~5.7%), demonstrating the synergetic effect of PCNT and RGO for thermal conductivity improvement. This result also exhibits some advantages compared with previously reported h values for polymer composites filled with CNT [35], boron nitride [36], and silicon carbide [37]. Additionally, the dispersion and morphology of fillers in polymer matrix often have a significant influence on the thermal conductivity of nano-carbon filler/polymer composites. The micrographs of SBR, SBR/RGO, SBR/PCNT, and SBR/PCNT@RGO are shown in Fig. S4 and Fig. 6. PCNT in SBR/PCNT composite shows well dispersion because of the improved compatibility between SBR and PCNT by


163

the copolymer coated on the surface of MWCNT (Fig. 6a). However, the TEM image (Fig. 6b) reveals that some RGO nanosheets are stacked together because the oxygen-containing groups on the surface of GO are eliminated, resulting in weak interfacial of RGO nanosheets. As a comparison, PCNT@RGO shows well dispersion and the addition of PCNT can bridge the isolated RGO nanosheets, resulting in the decrease of the interfacial thermal resistance originated from filler/filler interface and filler/SBR interface (Fig. 6c). Thus, the ultimate thermal energy of SBR/PCNT@RGO composites will be transferred in strictly defined 3D channels, proceeding as regular red arrows in the proposed thermal model (Fig. 6ced). 3.3. Mechanical properties of SBR composites Further explore the effect of the dispersion, structure and synergetic effect of PCNT and RGO on the performance of the composites, the representative stress-strain curves of the neat SBR and composites are compared in Fig. 7 and Table 1. It can be seen that RGO, PCNT and PCNT@RGO hybrids give rise to a significant increase in the tensile strength and stress at a given extension for SBR matrix. Notably, SBR/PCNT composites exhibit higher tensile strength and stress at a given extension with respect to the SBR/RGO composites at the same filler content, which is due to the improved interfacial interaction and dispersion of PCNT confirmed by the TEM results (Fig. 6aeb). As the previous reported, the well dispersion fillers in matrix could effectively transfer the load between GO sheet and matrix [38]. Obviously, SBR/PCNT@RGO composites exhibit superior performance with respect to SBR/RGO and SBR/PCNT composites, indicating the synergetic effect of PCNT and RGO on the improvement of the mechanical properties of SBR. In particular, SBR/ PCNT@RGO-2 composite exhibits the highest tensile strength and stress at 200% extension, increasing by about 203% and 450% compared with SBR. Such improvements may be due to the reason that more PCNT in hybrids may hinder the stacking of RGO nanosheets in matrix, resulting in the well dispersion of RGO nanosheets

Fig. 7. Tensile stress-strain curves of SBR, SBR/RGO, SBR/PCNT, and SBR/PCNT@RGO composites with 3% filler content. (A colour version of this figure can be viewed online.)

and the formation of filler network structure. Note that, the toughness of SBR/RGO, SBR/PCNT, SBR/PCNT@RGO-1 and SBR/ PCNT@RGO-2 is 9.4, 10.9, 10.6 and 10.7 MJ/m3, respectively, larger than that of SBR (5.5 MJ/m3). The results that carbon nanofillers strengthen the toughness of composites are consistent with the results of other study [39]. Furthermore, the toughness of SBR/PCNT, SBR/PCNT@RGO-1 and SBR/PCNT@RGO-2 exhibits higher than that of SBR/RGO, which may be attributed to the well dispersion of PCNT and PCNT@RGO in SBR. It has been widely accepted that the efficiency of carbon fillers in enhancing the polymer is significantly dependent on the size and

Fig. 6. TEM images of the SBR composites containing 3% (a) PCNT, (b) RGO, and (c) PCNT@RGO-2, the red lines illustrate the pathways of heat transfer. (d) Schematic diagram of the proposed thermal conduction model of SBR/PCNT@RGO-2 composite. (A colour version of this figure can be viewed online.)

164


Table 1 Mechanical properties of the SBR, SBR/RGO, SBR/PCNT, and SBR/PCNT@RGO composites with 3% filler content. Sample code

SBR

SBR/RGO

SBR/PCNT

SBR/PCNT@RGO-1

SBR/PCNT@RGO-2

s50%(MPa)a s100%(MPa)a s200%(MPa)a

0.64 0.94 1.4 2.9 380 5.5

0.92 1.6 2.8 4.6 370 9.4

1.1 1.8 3.0 5.2 384 10.9

1.7 2.8 5.3 7.3 280 10.6

2.4 4.2 7.7 8.8 230 10.7

Tensile stress(MPa) Elongation at break (%) Toughness (MJ/m3)b a b

s50%, s100%, s200%: stress at 50%, 100% and 200% extension, respectively. Toughness is calculated by integrating the area under the stressestrain curve.

dispersion of fillers. In this study, the PeakForce QNM mode of AFM was introduced to discuss the relationship of the size and dispersion of fillers and the properties of composites. Fig. 8 shows PFQNM morphology images of SBR, SBR/PCNT, SBR/RGO, and SBR/ PCNT@RGO composites with the corresponding height profiles. As shown in Fig. 8a, the morphology image of a SBR sample appears distribution feature alternate with brightness and darkness,

indicating the features of microphase separation existed in SBR matrix [40]. Compared with the SBR, the brighter domains appear in Fig. 8b and exhibit a circular shape with a diameter ranging from 150 to 400 nm, revealing the presence of RGO nanosheets in SBR matrix. However, its depth of 30 nm in its height profile indicates RGO nanosheets are partly stacked together. The results are consistent with the TEM analysis (Fig. 6b). The PF-QNM image of

Fig. 8. PF-QNM morphology images of (a) SBR, (b) SBR/RGO, (c) SBR/PCNT, (d) SBR/PCNT@RGO-1 and (e) SBR/PCNT@RGO-2 with their corresponding profiles, along the line l. (A colour version of this figure can be viewed online.)


165

Fig. 9. PF-QNM morphology image (a) and PF-QNM modulus (b) for SBR/PCNT@RGO-1. (A colour version of this figure can be viewed online.)

Fig. 8c shows many brighter strip domains with widths ranging from 30 to 100 nm, demonstrating that PCNT is well dispersed in the SBR matrix. Furthermore, it is worth noting that the introduction of PCNT into SBR matrix affects the dispersion and morphology of RGO, as shown in Fig. 8dee. The PCNT@RGO hybrids are well dispersed in SBR matrix with uniform size of RGO nanosheets (diameter: ~300 nm) and nanotubes (width: ~40 nm). As expected, the RGO nanosheets are connected by PCNT with high aspect ratio, resulting in the formation of network structure. Therefore, the well dispersion and nanoscale of PCNT@RGO are an important reason for the improvement of the mechanical properties of SBR composites. Besides the size and dispersion of fillers, the constrained

regions, originated from the polymer chains which are immobilized on the surface of fillers, play a role of bridge between polymer and fillers, thus improving the mechanical properties of composites [41]. In our study, AFM was used to analysis the presence of the constrained regions. From the morphology of SBR/PCNT@RGO-1 composite in Fig. 9a, the evidence of nanoscale phase separation can be clearly discernible. The circular and rodlike regions appear as light areas, which represent the presence of RGO and MWCNT, respectively, while the dark regions represent the SBR matrix. However, between two RGO particles, lighter brown shade regions exist, and are easily distinguished from the fillers and matrix regions. These lighter brown shades surrounding the fillers were ascribed to the constrained region fraction induced by nano

Fig. 10. PF-QNM modulus images of (a) SBR, (b) SBR/RGO, (c) SBR/PCNT and (d) SBR/PCNT@RGO-2. (A colour version of this figure can be viewed online.)

166


PCNT@RGO [42,43]. In addition, the spatial distance of the constrained region was 40e80 nm away from the filler surface. This result is consistent with the theoretical distance of the constrained polymer region predicted by Beall [44], i.e. 50e75 nm from the surface of fillers. Based on the different modulus for different phase domains of composites, the PeakForce QNM mode of AFM was used to distinguish different phase domains [45,46]. In Fig. 9b, PCNT@RGO and SBR matrix appear as purple and red domains, showing the modulus of ~40 MPa and ~5 MPa, respectively. The constrained regions appear as green/yellow areas, showing the modulus of 14e24 MPa. The constrained fraction regions have been shown to be stiffer than the matrix because of the lower mobility of polymer chains in this region. As the previous reports [41], the greater the fraction of the constrained region is, the stronger the filler-rubber interfacial interaction will be. It will be beneficial for improving the mechanical properties of SBR. In Figs. 9b and 10d, SBR/ PCNT@RGO has large constrained fraction regions than that of SBR matrix in Fig. 10a, which is accounted for the better mechanical properties. The constrained regions in Fig. 10c are greater than that in Fig. 10b, indicating PCNT leads to higher restrained regions than RGO. An appropriate size and constraint contributions originated from the better dispersion of hybrid fillers could be considered to be responsible for the significant reinforcing effects of filler on SBR composites. 4. Conclusion In summary, a copolymer PS-b-PHEMA was successfully synthesized by RAFT, and then grafted onto the surface of MWCNT by “grafting to” method to obtain a modified filler PCNT. A novel hybrid PCNT@RGO was successfully prepared via esterification reaction and reduction process of GO. The PCNT@RGO hybrid as a filler can effectively enhance the thermal conductivity and mechanical properties of the SBR matrix. Such high thermal conductivity and mechanical properties can be achieved in the composites with low loading of the PCNT@RGO hybrid, which is well related to the synergetic effect of PCNT and RGO in hybrid, strong interfacial between PCNT@RGO and matrix, the large constrained regions and continuous filler networks. As an example, SBR/PCNT@RGO-2 (97/ 3) composite shows a high thermal conductivity of 0.45 W/m K and the tensile strength of 8.8 MPa, which are 2.0 and 3.1-fold higher than that of SBR matrix, respectively. Therefore, the novel hybrid PCNT@RGO can be used as a filler for fabricating high-performance carbon-based SBR nanocomposites. Acknowledgment The work is supported by a National Natural Science Foundation of China (No. 51273109, 51235008). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.carbon.2017.07.057. References [1] H. Chen, V.V. Ginzburg, J. Yang, Y. Yang, W. Liu, Y. Huang, et al., Thermal conductivity of polymer-based composites: fundamentals and applications, Prog. Polym. Sci. 59 (2016) 41e85. [2] W. Feng, M. Qin, Y. Feng, Toward highly thermally conductive all-carbon composites: structure control, Carbon 109 (2016) 575e597. [3] R. Wang, D. Zhuo, Z. Weng, L. Wu, X. Cheng, Y. Zhou, et al., A novel nanosilica/ graphene oxide hybrid and its flame retarding epoxy resin with simultaneously improved mechanical, thermal conductivity, and dielectric properties, J. Mater Chem. A 3 (18) (2015) 9826e9836.

[4] B. Yin, J. Wang, H. Jia, J. He, X. Zhang, Z. Xu, Enhanced mechanical properties and thermal conductivity of styreneebutadiene rubber reinforced with polyvinylpyrrolidone-modified graphene oxide, J. Mater. Sci. 51 (12) (2016) 5724e5737. [5] Z. Pang, X. Gu, Y. Wei, R. Yang, M.S. Dresselhaus, Bottom-up design of threedimensional carbon-honeycomb with superb specific strength and high thermal conductivity, Nano Lett. 17 (1) (2016) 179e185. [6] J. Yang, E. Zhang, X. Li, Y. Zhang, J. Qu, Z.-Z. Yu, Cellulose/graphene aerogel supported phase change composites with high thermal conductivity and good shape stability for thermal energy storage, Carbon 98 (2016) 50e57. [7] M.C. Richard Gulotty, Jagdale Pravin, ATaAA. Balandin, Effects of functionalization on thermal properties of single-wall and multi-wall carbon nanotubepolymer nanocomposites, ACS Nano 7 (6) (2013) 5114e5121. [8] W. Ma, L. Liu, Z. Zhang, R. Yang, G. Liu, T. Zhang, et al., High-strength composite fibers: realizing true potential of carbon nanotubes in polymer matrix through continuous reticulate architecture and molecular level couplings, Nano Lett. 9 (8) (2009) 2855e2861. [9] G. Fugallo, A. Cepellotti, L. Paulatto, M. Lazzeri, N. Marzari, F. Mauri, Thermal conductivity of graphene and graphite: collective excitations and mean free paths, Nano Lett. 14 (11) (2014) 6109e6114. [10] R.D.P. Ji Won Suk, Jinho An, S. Rodney, Ruoff, Mechanical properties of monolayer graphene oxide, ACS Nano 4 (11) (2010) 6557e6564. [11] P. Song, Z. Cao, Y. Cai, L. Zhao, Z. Fang, S. Fu, Fabrication of exfoliated graphene-based polypropylene nanocomposites with enhanced mechanical and thermal properties, Polymer 52 (18) (2011) 4001e4010. [12] M. Castelaín, G. Martínez, C. Marco, G. Ellis, H.J. Salavagione, Effect of clickchemistry approaches for graphene modification on the electrical, thermal, and mechanical properties of polyethylene/graphene nanocomposites, Macromolecules 46 (22) (2013) 8980e8987. [13] M. Abdalla, D. Dean, M. Theodore, J. Fielding, E. Nyairo, G. Price, Magnetically processed carbon nanotube/epoxy nanocomposites: morphology, thermal, and mechanical properties, Polymer 51 (7) (2010) 1614e1620. [14] K. Yang, M. Gu, Y. Guo, X. Pan, G. Mu, Effects of carbon nanotube functionalization on the mechanical and thermal properties of epoxy composites, Carbon 47 (7) (2009) 1723e1737. [15] R.J. Warzoha, A.S. Fleischer, Heat flow at nanoparticle interfaces, Nano Energy 6 (2014) 137e158. [16] KMFSaAA. Balandin, Graphene e based nanocomposites as highly efficient thermal interface materials, Nano Lett. 2 (12) (2012) 861e867. [17] G. Lian, C.-C. Tuan, L. Li, S. Jiao, Q. Wang, K.-S. Moon, et al., Vertically aligned and interconnected graphene networks for high thermal conductivity of epoxy composites with ultralow loading, Chem. Mater. 28 (17) (2016) 6096e6104. [18] N. George, C.S. JC, A. Mathiazhagan, R. Joseph, High performance natural rubber composites with conductive segregated network of multiwalled carbon nanotubes, Compos. Sci. Technol. 116 (2015) 33e40. [19] Z. Tang, L. Zhang, W. Feng, B. Guo, F. Liu, D. Jia, Rational design of graphene surface chemistry for high-performance rubber/graphene composites, Macromolecules 47 (24) (2014) 8663e8673. €ckelhuber, R. Jurk, M. Saphiannikova, J. Fritzsche, H. Lorenz, et [20] A. Das, K.W. Sto al., Modified and unmodified multiwalled carbon nanotubes in high performance solution-styreneebutadiene and butadiene rubber blends, Polymer 49 (24) (2008) 5276e5283. [21] A. Shojaei, M. Fahimian, B. Derakhshandeh, Thermally conductive rubberbased composite friction materials for railroad brakes e thermal conduction characteristics, Compos. Sci. Technol. 67 (13) (2007) 2665e2674. [22] X. Zhao, W. Lin, N. Song, X. Chen, X. Fan, Q. Zhou, Water soluble multi-walled carbon nanotubes prepared via nitroxide-mediated radical polymerization, J. Mater. Chem. 16 (47) (2006) 4619e4625. [23] B.-B. Ke, L.-S. Wan, W.-X. Zhang, Z.-K. Xu, Controlled synthesis of linear and comb-like glycopolymers for preparation of honeycomb-patterned films, Polymer 51 (10) (2010) 2168e2176. [24] J.T.P. Do Kyoung Lee, Dong Kyu Roh, Byoung Ryul Min, Jong Hak Kim, Synthesis of crosslinked polystyrene-b-poly(hydroxyethyl methacrylate)-bPoly(styrene sulfonic acid) triblock copolymer for electrolyte membranes, Macromol. Res. 17 (5) (2009) 325e331. [25] T. Kar, S. Scheiner, A.K. Roy, H.F. Bettinger, Unusual low-vibrational C]O mode of COOH can distinguish between carboxylated zigzag and armchair single-wall carbon nanotubes, J. Phys. Chem. C 116 (49) (2012) 26072e26083. [26] Z. Tang, H. Kang, Z. Shen, B. Guo, L. Zhang, D. Jia, Grafting of polyester onto graphene for electrically and thermally conductive composites, Macromolecules 45 (8) (2012) 3444e3451. [27] M.M. Sk, C.Y. Yue, R.K. Jena, Synthesis of graphene/vitamin C templatecontrolled polyaniline nanotubes composite for high performance supercapacitor electrode, Polymer 55 (3) (2014) 798e805. [28] M. Lian, J. Fan, Z. Shi, S. Zhang, H. Li, J. Yin, Gelatin-assisted fabrication of graphene-based nacre with high strength, toughness, and electrical conductivity, Carbon 89 (2015) 279e289. [29] S. Chen, W. Yeoh, Q. Liu, G. Wang, Chemical-free synthesis of grapheneecarbon nanotube hybrid materials for reversible lithium storage in lithium-ion batteries, Carbon 50 (12) (2012) 4557e4565. [30] J.P. Sijie Wan, Yuchen Li, Han Hu, Lei Jiang, Qunfeng Cheng, Use of synergistic interactions to fabricate strong, tough, and conductive artificial nacre based on graphene oxide and chitosan, ACS Nano 9 (10) (2015) 9830e9836. [31] C. Wu, X. Huang, G. Wang, L. Lv, G. Chen, G. Li, et al., Highly conductive


[32]

[33]

[34]

[35]

[36]

[37]

[38]

nanocomposites with three-dimensional, compactly interconnected graphene networks via a self-assembly process, Adv. Funct. Mater. 23 (4) (2013) 506e513. G. Wang, J. Yang, J. Park, X. Gou, B. Wang, H. Liu, et al., Facile synthesis and characterization of graphene nanosheets, J. Phys. Chem. C 112 (22) (2008) 8192e8195. Y. Yao, X. Zeng, G. Pan, J. Sun, J. Hu, Y. Huang, et al., Interfacial engineering of silicon carbide nanowire/cellulose microcrystal paper toward high thermal conductivity, ACS Appl. Mater. Interfaces 8 (45) (2016) 31248e31255. F.H. Gojny, M.H.G. Wichmann, B. Fiedler, I.A. Kinloch, W. Bauhofer, A.H. Windle, et al., Evaluation and identification of electrical and thermal conduction mechanisms in carbon nanotube/epoxy composites, Polymer 47 (6) (2006) 2036e2045. A. Yu, M.E. Itkis, E. Bekyarova, R.C. Haddon, Effect of single-walled carbon nanotube purity on the thermal conductivity of carbon nanotube-based composites, Appl. Phys. Lett. 89 (13) (2006) 133102. X. Huang, C. Zhi, P. Jiang, D. Golberg, Y. Bando, T. Tanaka, Polyhedral oligosilsesquioxane-modified boron nitride nanotube based epoxy nanocomposites: an ideal dielectric material with high thermal conductivity, Adv. Funct. Mater. 23 (14) (2013) 1824e1831. Y. Hwang, M. Kim, J. Kim, Effect of Al2O3 coverage on SiC particles for electrically insulated polymer composites with high thermal conductivity, RSC Adv. 4 (33) (2014) 17015e17021. J. Liang, Y. Huang, L. Zhang, Y. Wang, Y. Ma, T. Guo, et al., Molecular-Level dispersion of graphene into poly(vinyl alcohol) and effective reinforcement of

167

their nanocomposites, Adv. Funct. Mater. 19 (14) (2009) 2297e2302. [39] J.-H. Han, H. Zhang, M.-J. Chen, G.-R. Wang, Z. Zhang, CNT buckypaper/thermoplastic polyurethane composites with enhanced stiffness, strength and toughness, Compos. Sci. Technol. 103 (2014) 63e71. [40] G.H. Fredrickson, Multicritical phenomena and microphase ordering in random block copolymer melts, Macromolecules 25 (1992) 6341e6354. [41] Y. Lin, S. Liu, J. Peng, L. Liu, The fillererubber interface and reinforcement in styrene butadiene rubber composites with graphene/silica hybrids: a quantitative correlation with the constrained region, Compos. Part A Appl. Sci. Manuf. 86 (2016) 19e30. [42] D. Sikdar, S.M. Pradhan, D.R. Katti, K.S. Katti, B. Mohanty, Altered phase model for polymer clay nanocomposites, Langmuir ACS J. Surf. Colloids 24 (10) (2008) 5599e5607. [43] S.K. Rath, K. Sudarshan, R.S. Bhavsar, U.K. Kharul, P.K. Pujari, M. Patri, et al., Characterizing the nanoclay induced constrained amorphous region in model segmented polyurethaneeurea/clay nanocomposites and its implications on gas barrier properties, Phys. Chem. Chem. Phys. 18 (3) (2016) 1487e1499. [44] G. Beall, New conceptual model for interpreting nanocomposite behavior, Polymereclay Nanocomposites (2000) 267e279. [45] S. Sun, D. Wang, T.P. Russell, L. Zhang, Nanomechanical mapping of a deformed elastomer: visualizing a self-reinforcement mechanism, ACS Macro Lett. 5 (7) (2016) 839e843. [46] S.N. Magonov, V. Elings, M.H. Whangbo, Phase imaging and stiffness in tapping-mode atomic force microscopy, Surf. Sci. 375 (2e3) (1997) L385eL391.