Effects of hybrid filler networks of carbon nanotubes and carbon black on fracture resistance of styrene-butadiene rubber composites.
Because of their unique overall performances, elastomers are widely employed in the tire industry and damper products, which are always working under quasi-static and dynamic loading conditions. However, microcracks can be readily initiated (crack nucleation) and then propagated (crack growth) under such conditions, and eventually cause the total failure of rubber products [1, 2]. Many factors including types of rubber matrix, reinforced fillers, environmental conditions, and mechanical loading conditions can influence the failure process [3-5]. Derived from the Griffith energy criterion, the tearing energy theory proposed by Rivlin and Thomas in 1953  has been still widely used to study the failure of elastomeric materials. It revealed that for elastomeric materials, the dynamic crack growth rate was only dependent on the input tearing energy (namely strain energy release rate) and the geometry of crack tip, not on the global specimen geometry [6, 7].
J-integral, first proposed by Rice et al. , was to characterize the fracture behaviors of ductile metals and plastics, without concerning the complex stress or strain field around the crack tip. It is another powerful energetic fracture criterion, which describes a path-independent line integral around the crack tip. However, due to the high nonlinear viscoelasticity of elastomeric materials, J-integral criterion cannot be directly applied for elastomeric materials without modification . Thanks to the tentative applications of J-integral in elastomer fields, nowadays, the J-integral theory has been successfully adopted to evaluate the fracture resistance of rubber materials, and some instructive results were obtained [9-14]. It has been proved that the J-integral and tearing energy gave the equivalent calculation results for single-edge notched tension (SENT) specimens of rubbers . Chow et al.  revealed that the critical J-integral could be considered as an intrinsic material property and a valid fracture criterion for elastomers. Our previous work  combined the J-integral parameters with strain energy density under the same global strain, and successfully predicted the fracture and fatigue resistance of silica/carbon black (CB)/natural rubber (NR) composites. Agnelli et al. [13, 14] declared that the crack tip opening displacement (CTOD) could be seen as an indirect reflection of crack advancement for viscoelastic elastomeric materials. By combining J-integral theory with CTOD concept, Grellmann et al.  evaluated the crack initiation and propagation resistance of elastomers by introducing critical J-value [J.sub.IC] and tearing modulus [T.sub.R].
Due to the incorporation of stiffer filler particles into the soft viscoelastic elastomeric matrix, a strain amplification effect of rubber matrix can be observed . The local stress concentration induced by strain amplification near the crack tip region has a close relationship with the static and dynamic crack propagation behaviors of rubber composites . Nowadays, various methods have been applied to evaluate the local strain field distribution state and its amplification factor. Finite element analysis can succeed in predicting the strain distribution contours and even the fatigue life of rubber products, from the perspective of numerical simulation . Recently, in situ synchrotron techniques, such as wide-angle X-ray diffraction (WAXD) and small-angle X-ray scattering (SAXS), were employed for the evaluation of strain field distribution of crystalline NR composites under deformation [18-20]. The strain field distribution around the crack tip could be indirectly reflected from the crystallinity contours of NR. It opened up a new platform for the evaluation of local strain distribution at the crack tip of crystalline rubber. Certainly, these advanced synchrotron techniques are not suitable for noncrystalline styrene-butadiene rubber (SBR) composites.
Digital image correlation (DIC), an efficient noncontacting image analysis technique and optical displacement measurement method, can characterize the deformation and strain field distribution state at the surface of materials [21, 22]. Since first proposed by Peters et al. . DIC has been widely used in exploring the fracture mechanism of materials thanks to the improvement of algorithms and hardware [24-26]. Sutton et al. precisely measured the CTOD and stress intensity factor of metallic materials by employing DIC techniques [27, 28]. On the basis of Brown's stretched craze theory , Mzabi et al.  proved that for viscoelastic rubber composites, the singularity of strain field distribution near the crack tip region could be also evaluated by DIC. Notably, unlike the above mentioned WAXD and SAXS, DIC could be applicable for all kinds of elastomers. Meanwhile, a new energetic criterion based on the local tearing energy near the crack tip was proposed by Mzabi et al. to predict the fatigue resistance of elastomeric materials. Recently, the strain distribution near the crack tip of the SBR composites filled with different types of CB was investigated using DIC in our previous study , which demonstrated that the larger the CB particle size, the less strain amplification at the crack tip.
Recently, because of their exceptional electrical conductivity, thermal conductivity, and reinforcing efficiency, one-dimensional carbon nanotubes (CNTs) have been regarded as promising nanofillers in rubber industry [31-33]. However, the highly entangled agglomeration of CNTs in rubber composites is still discouraging . Tremendous attention has been attracted in the fields of hybrid filler system, such as spherical CB and fibrous CNTs [35-37], which revealed that partial replacement of CB with small amounts of CNTs induced synergistic effects on the mechanical, thermal, and electrical properties. For the hybrid filler networks, unequal replacement (low amounts of CNTs replacing high amounts of CB) might lead to a better synergistic effect than equal replacement . To the best of our knowledge, however, the fracture resistance of hybrid CNTs and CB filled SBR composites, especially the influence of hybrid filler networks on strain amplification and distribution near the crack tip and its relationship with fracture resistance, are rare and of great interest.
Herein, the objectives of this research were to relate the local strain distribution at the crack tip with the fracture resistance of CNTs/CB/SBR composites, and to discover the proper replacement method of CNTs to CB for a better fracture resistance. In this study, the strain distribution near the crack tip of SENT specimen was measured via DIC, and the strain amplification levels at the crack tips were compared. The quasi-static fracture resistance of CNTs/CB/SBR composites was measured by J-integral testing method. Results revealed that unequal replacement of CB with CNTs (1 phr CNTs replacing 4 phr CB) induced a weaker local strain amplification effect of rubber matrix, leading to a better fracture resistance than the case of equal replacement (1 phr CNTs replacing 1 phr CB). The results were expected to provide a new clue to understand the synergistic effect between CB and CNTs in improving the fracture resistance of elastomeric materials.
Materials and Preparation Procedures
The formulas of SBR composites are detailed in Table 1. SBR (SBR-1502, provided by Organic Synthesis Factory of Jilin Chemical Industrial Co., Ltd., Jilin, China) was filled with different contents of CB (N234, Cabot Chemical Products Co., Ltd., Tianjin, China) and CNTs (Flotube TM 7000, CNano Technology Co., Ltd., Beijing, China) to form hybrid filler networks. The spherical CB N234 has an overage original particle size of ~22 nm, with a specific surface area of ~120 [m.sup.2]/g (BET method). The structure degree of N234 which is characterized by dibutylphthalate absorption is about 125 ml/100g. CNTs are a kind of easily dispersed, highly one-dimensional aligned CNT bundles with few mutual entanglements. It has a purity of 92%, length greater than 50 [micro]m, and average diameter of 6-8 nm. All other ingredients were commercially available industrial products.
CB was partially replaced by CNTs in two different ways: (a) unequal replacement, 1 phr CNTs replacing 4 phr CB, termed u-CNTs-1, u-CNTs-2, and u-CNTs-3, respectively; (b) equal replacement, 1 phr CNTs replacing 1 phr CB, termed e-CNTs-1, e-CNTs-2, and e-CNTs-3, respectively. For both cases, the numbers 1, 2, and 3 meant the CNTs contents. The CNTs/CB/SBR compounds were prepared with a 6-inch two-roll mill by <2767022H_TB001> traditional mechanical compounding method. The vulcanization curves of the compounds (shown in Supporting Information Fig. S1) were measured to determine the optimum curing time ([t.sub.90]) via a moving die rheometer at 150[degrees]C. Each of the SBR compounds was vulcanized in the form of 2 mm thick sheets at 150[degrees]C and 15 MPa for [t.sub.90].
Rubber Process Analysis. The strain amplitude dependence of shear storage modulus G' of SBR. compounds was measured by rubber processing analyzer RPA 2000 (Alpha Technologies Co., Ltd., Akron) at strain range from 0.28% to 400%, 100[degrees]C, and 1 Hz [39, 40].
Mechanical Properties. The mechanical performances including tensile and tear properties of the SBR composites were measured via a CMT-4204 universal testing machine (MTS Systems Co., Ltd., Shanghai, China) with a cross-head speed of 500 mm/min according to the standards ISO 37: 2011 and ISO 34-1: 2010, respectively. Meanwhile, the Shore A hardness was measured according to ISO 868: 2003.
J-Integral Tests. The J-integral tests were conducted to measure the resistance of quasi-static crack initiation and propagation of CNTs/CB/SBR composites. Its detailed testing process has been presented in our previous works [12, 38]. Briefly, SENT (Fig. 1) specimens with working strained length L = 40 mm (initial length of 100 mm), width W = 15 mm, thickness B = 2 mm, and different initial crack a (pre-cut length) in the middle edge were adopted. The SENT specimens with varied pre-cut lengths were stretched with a constant cross-head speed of 10 mm/min. The J-value of specimen with a certain crack length at a certain displacement was calculated according to Eq. 1.
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)
where U is the strain energy at a certain displacement [DELTA], A is the fracture surface area, B and a are the thickness and pre-cut length, respectively.
Before the J-tests, the crack tips of SENT specimens were uniformly coated by silver powers for easy determination of crack initiation point. A Canon EOS 70D camera was put in front of the crack tip to real-time record the variation in CTOD. Then J-integral was correlated with the CTOD and the J-CTOD curve was plotted. The point at CTOD = 0.1 mm was considered as the crack initiation point, and the corresponding J-value was named the critical J-value ([J.sub.IC]), which was connected with the crack initiation resistance. The slope of J-CTOD curve at CTOD ranging from 0.1 mm to 0.5 mm was tearing modulus ([T.sub.R]), which was utilized to characterize the crack propagation resistance of the composites.
DIC Measurements. The strain distribution near the crack tip was characterized by DIC. The lateral side of SENT specimen was sprinkled with silver powders to enable the formation of random texture, as shown in Fig. 2. The specimen was then stretched to global strain [epsilon] = 1.0 with a cross-head speed of 10 mm/min. Meanwhile the images of the crack tip were taken consecutively at an interval of 1 s by a Canon EOS 70D camera equipped with a macro-objective 100/2.8, and the size of each pixel of the digital image reached ~8 [micro]m. The size of region of interest (ROI) near the crack tip was about 5 mm x 5 mm. The LED lights source with luminance of 160 1x were used to maintain constant gray scale of every pixel. The acquired images were successively correlated with the reference one by the procedure programmed in MATLAB and the displacement of each subset inside ROI was deduced.
An example of the vertical displacement contour measured by DIC is presented in Fig. 3. As we can see, most parts of the contours were parallel to each other at the region far away from the crack tip, suggesting that the far-field strain is uniform; whereas the contours near the crack tip converged to the edge of the seam. Along the y axis, we scanned vertical displacement u at different y values. According to the definition of vertical strain [[epsilon].sub.y] = du/dy, the slope of u-y curve could be considered as [[epsilon].sub.y]. We obtained the largest slope at the axis of the crack (y = 0). Similarly, we scanned along x axis, the strongest strain amplification occurred at x = 0 ([[epsilon].sub.y|x=0, y=0]), which proved the ubiquitous existence of the singular zone at the crack tip.
RESULTS AND DISCUSSION
Hybrid Filler Networks of CNTs/CB/SBR Composites
The shear storage modulus G' as a function of dynamic strain of CNTs/CB/SBR composites are presented in Fig. 4a and b. It could be noted that the G' was nonlinearly decreased with the increase in shear strain for all CNTs/CB/SBR composites (Payne effect) , which was mainly caused by the breakdown of filler--filler networks. Generally, the degree of filler networks was characterized by the absolute value of G' in the low strain regime. For u-CNTs composites (unequal replacement), the G' in the low strain regime was only slightly enhanced with increasing CNTs content, suggesting a very limited increase in the degree of hybrid filler networks. However, for e-CNTs composites (equal replacement), the G' under low strain dramatically increased with increasing CNTs content and pronounced Payne effect was observed. Reasons can be ascribed that the one-dimensional CNTs with super-high aspect ratio were much easier to form filler network than spherical CB at the same filler content. That is, at the same filler content, the hybrid filler network of CNTs and CB is much stronger than CB one in Fig. 4b; while the ability of forming filler network of 1 phr CNTs is almost the same as 4 phr CB in Fig. 4a.
The mechanical properties of hybrid CNTs and CB filled SBR composites are shown in Table 2. For the case of unequal replacement, a moderate elevation in stress at 100% and Shore A hardness was observed with the increase in CNTs content; whereas for the case of equal replacement, an obvious increase was observed. Compared with those of CNTs-0, an increment of 59% in stress at 100% and a 6[degrees] increase in hardness were observed for e-CNTs-5. The reasons mainly lie in two aspects: (a) the total filler content in e-CNTs was higher than that in u-CNTs; (b) CNTs could construct the more powerful filler networks than CB at the equal content, that is, fibrous CNTs restricted the mobility of rubber macromolecules more evidently than CB. These are consistent with the results of G'. The tensile strength and tear strength changed little regardless of equal or unequal replacement.
Results of G' and mechanical properties confirmed that unequal replacement exhibited a slight influence on the stiffness and deformability of SBR composites. However, equal replacement made the composites much stiffer and more difficult to deform under external loading conditions. The increased stiffness and decreased deformability resulting from the powerful hybrid filler networks would cause a great influence on the fracture resistance of SBR composites.
Quasi-Static Fracture Resistance by J-Testing
J-testing has been proved to be a valid method to evaluate the quasi-static fracture performance of elastomeric materials . The critical J-value [J.sub.IC] was used to evaluate the crack initiation resistance. [J.sub.IC]s of CNTs/CB/SBR composites with varied pre-cut lengths are given in Fig. 5a and b. We could recognize that [J.sub.IC]s remained constant at the initial crack length a ranging from 2 mm to 7 mm for all the specimens. However, [J.sub.IC] at initial crack length a = 1 mm was much lower than that at other initial crack lengths, which may be due to the experimental accuracy. Therefore, the average [J.sub.IC] at a ranging from 2 mm to 7 mm was chosen to evaluate the crack initiation resistance. It was worth noticing that the u-CNTs exhibited much higher [J.sub.IC] than e-CNTs, indicating the better crack initiation resistance for unequal replacement. Meanwhile, with increasing CNTs content, the crack initiation resistance was highly improved for u-CNTs, however, reduced for e-CNTs.
Plots of J-CTOD curves in the CTOD range from 0.1 mm to 0.5 mm when a was 4 mm are shown in Fig. 5c and d, and their corresponding linear slopes [T.sub.R] and correlation coefficients ([R.sup.2]) are summarized in Table 3. For u-CNTs, the obviously increased [T.sub.R] with increasing CNTs content implied an enhanced crack propagation resistance. However, the crack propagation resistance of e-CNTs was highly reduced, which was reflected from the gradually decreased [T.sub.R].
Here, we confirmed that unequal replacement simultaneously enhanced the crack initiation and propagation resistance, and the fracture resistance was enhanced with increasing CNTs content. However, an antipodal result was obtained for equal replacement. Although CNTs exhibited powerful reinforcement effect on mechanical properties, the quasi-static fracture resistance of filled SBR was not improved by equal replacement. The distribution of strain field near the crack tip had a direct relationship with the fracture resistance . Therefore, different effects of CNTs/CB hybrid filler networks on the strain field distribution near the crack tip were investigated by DIC to reveal the fracture mechanisms.
Strain Amplification and Distribution Near the Crack Tip
The maximum strains at crack tip [[epsilon].sub.y|x = 0, y = 0] for u-CNTs and e-CNTs composites at global strain [epsilon] = 0.3, 0.5, 0.7, 0.9, and 1.0 were measured and presented in Fig. 6a and b, respectively. Results showed that the maximum strains linearly increased with increasing global strain for all SBR composites. Differently, for u-CNTs, the level of strain amplification at the crack tip decreased with increasing CNTs content, whereas the maximum strains of e-CNTs increased with increasing CNTs content. For e-CNTs composites, the stronger strain amplification effect of rubber matrix at the crack tip was mainly induced by the decreased deformability of more powerful hybrid filler networks, which is consistent with the results of composites filled with different particle sizes of CB . The strain amplification level at the crack tip increased with decreasing the CB particle size.
Figure 6c and d, respectively illustrate the vertical strain ([[epsilon].sub.y]) distribution along the crack axis (x direction) of u-CNTs and e-CNTs. The abscissa x = 0 represents the exact point of crack tip and the crack is assumed to propagate from right to left of the axis. Correspondingly, the horizontal size of the strain singular zone can be roughly evaluated. It was worth noticing that u-CNTs exhibited less strain amplification level than the reference composite CNTs-0 along x direction; while at about x > 0.25 mm along x direction, e-CNTs exhibited the same trend: the strain amplification level of e-CNTs is less than that of CNTs-0. Further, the strain amplification areas of u-CNTs, e-CNTs, and CNTs-0 composites were calculated and compared. For reference CNTs-0, strain amplification effect affected the area ~1.2 mm near the crack tip. The amplification area expanded to ~1.5 mm for u-CNTs and reduced to ~1.0 mm for e-CNTs.
Based on the above results of the strain amplification levels near the crack tip and the strain amplification area, it is worth noticing that u-CNTs exhibited decreased strain amplification levels near the crack tip and larger strain amplification area, compared to CNTs-0. These would lead to less stress concentration and more polymer chains involved in homogenizing the stress concentration and dissipating the input local tearing energy, which would highly contribute to enhancing the fracture resistance. Conversely, in the case of e-CNTs, due to the stronger stress concentration at the crack tip resulting from higher strain amplification level, and less amplification area relative to CNTs-0, e-CNTs composites exhibited weaker crack initiation and propagation resistance. Results of strain amplification and distribution near the crack tip well explained the fact that u-CNTs exhibited much higher fracture resistance than e-CNTs.
As a summary, although the one-dimensional CNTs with high anisotropy could be easily oriented along the stretching direction near the crack tip to prevent the crack propagation , the results for equal and unequal replacement were just opposite. Compared with that of equal replacement, the degree of hybrid filler networks was only slightly improved by unequal replacement. Therefore, under the same global strain, u-CNTs would be more deformable than e-CNTs, resulting in a weaker local strain amplification factor and a larger amplification area near the crack tip, all of which contributed to dissipating the input local tearing energy and then enhancing the fracture resistance. However, due to the high local stress concentration and local tearing energy input near the crack tip resulting from powerful hybrid filler networks, e-CNTs composites exhibited weak fracture resistance.
Unequal and equal replacements of CB with CNTs were both used to prepare CNTs/CB/SBR composites. A slightly improved hybrid filler networks and stiffness were observed for unequal replacement, J-integral tests confirmed that with the increase in CNTs content, the fracture resistance gradually improved for unequal replacement. However, equal replacement highly reduced the fracture resistance. For u-CNTs, under a certain global strain, the strain amplification effect of rubber matrix at the crack tip decreased with increasing CNTs content, which would decrease the local tearing energy input and then improve the fracture resistance. The enlarged amplification area of u-CNTs near the crack tip also contributed to the improved fracture resistance by homogenizing the stress concentration. Absolutely opposite trends about the strain field at the crack tip were observed for e-CNTs. This research confirmed that unequal replacement was an effective method to obtain synergy on fracture resistance of SBR composites between CB and CNTs.
[1.] G.J. Lake, Rubber Chem. Technol., 68, 435 (1995).
[2.] T. Zarrin-Ghalami and A. Fatemi, Polym. Eng. Sci., 52, 1795 (2012).
[3.] W.V. Mars and A. Fatemi, Rubber Chem. Technol., 77, 391 (2004).
[4.] K. Legorju-Jago and C. Bathias, Int. J. Fatigue, 24, 85 (2002).
[5.] X. Wu, Y. Wang, J. Liu, S. He, and L. Zhang, Polym. Eng. Sci., 52, 1027 (2012).
[6.] R.S. Rivlin and A.G. Thomas, J. Polym. Sci., 10, 291 (1953).
[7.] G.J. Lake, Rubber Chem. Technol., 76, 567 (2003).
[8.] J.R. Rice and G.F. Rosengren, J. Mech. Phys. Solids, 16, 1 (1968).
[9.] W. Grellmann and K. Reincke, Technical Material Diagnostics--Fracture Mechanics of Filled Elastomer Blends, Springer, Berlin, 227 (2013).
[10.] R.F. Lee and J.A. Donovan, Rubber Chem. Technol., 60, 674 (1987).
[11.] C.L. Chow, J. Wang, and P.N. Tse, Tire Sci. Technol., 16, 44 (1988).
[12.] B. Dong, C. Liu, and Y.P. Wu, Polym. Test., 38, 40 (2014).
[13.] S. Agnelli, F. Baldi, and T. Ricco, Eng. Fract. Mech., 90, 76 (2012).
[14.] S. Agnelli, G. Ramorino, S. Passera, J. Karger-Kocsis, and T. Ricco, Express Polym. Lett., 6, 581 (2012).
[15.] J.A.C. Harwood, L. Mullins, and A.R. Payne, J. Appl. Polym. Sci., 9, 3011 (1965).
[16.] S. Mzabi, D. Berghezan, S. Roux, F. Hild, and C. Creton, J. Polym. Sci. Part B: Polym. Phys., 49, 1518 (2011).
[17.] D.W. Cha and H.Y. Jeong, Int. J. Fatigue, 33, 912 (2011).
[18.] K. Bruning, K. Schneider, S.V. Roth, and G. Heinrich, Polymer, 54, 6200 (2013).
[19.] K. Bruning, K. Schneider, S.V. Roth, and G. Heinrich, Macromolecules, 45, 7914 (2012).
[20.] K. Bruning, K. Schneider, and G. Heinrich, J. Polym. Sci. Polym. Phys., 50, 1728 (2012).
[21.] Y. Han, A.D. Rogalsky, B. Zhao, and H.J. Kwon, Polym. Eng. Sci., 52, 826 (2012).
[22.] Y. Yang, P. Zeng, L. Lei, and Y. Xu, Polym. Eng. Sci., 55, 196 (2015).
[23.] W.H. Peters and W.F. Ranson, Opt. Eng., 21, 427 (1982).
[24.] G. Tao and Z. Xia, Polym. Test., 24, 844 (2005).
[25.] T. Guelon, E. Toussaint, J.B. Le Cam, N. Promma, and M. Grediac, Polym. Test., 28, 715 (2009).
[26.] D. Zhang, C.D. Eggleton, and D.D. Arola, Exp. Mech., 42, 409 (2002).
[27.] M.A. Sutton, J.D. Helm, and M.L. Boone, Int. J. Fract., 109, 285 (2001).
[28.] S.R. McNeill, W.H. Peters, and M.A. Sutton, Eng. Fract. Mech., 28, 101 (1987).
[29.] H.R. Brown, Macromolecules, 24, 2752 (1991).
[30.] C. Liu, B. Dong, L.Q. Zhang, Q. Zheng, and Y.P. Wu, Rubber Chem. Technol, 88, 276 (2015).
[31.] G. Pircheraghi, T. Powell, V.S. Bonab, and I. Manas-Zloczower, Polym. Eng. Sci., 56, 394 (2016).
[32.] C. Kummerlowe, N. Vennemann, E. Yankova, M. Wanitschek, C. GroB, T. Heider, F. Haberkorn, and A. Siebert, Polym. Eng. Sci., 53, 849 (2013).
[33.] Y. Nakaramontri, C. Nakason, C. Kummerlowe, and N. Vennemann, Polym. Eng. Sci., 55, 2500 (2015).
[34.] H. Lorenz, J. Fritzsche, A. Das, K.W. Stockelhuber, R. Jurk, G. Heinrich, and M. Kluppel, Compos. Sci. Technol., 69, 2135 (2009).
[35.] M. Galimberti, M. Coombs, P. Riccio, T. Ricco, S. Passera, S. Pandini, L. Conzatti, A. Ravasio, and I. Tritto, Macromol. Mater. Eng., 298, 241 (2013).
[36.] H. Ismail, A.F. Ramly, and N. Othman, Polym-Plast. Technol., 50, 660 (2011).
[37.] L. Bokobza, M. Rahmani, C. Belin, J.L. Bruneel, and N.E.E.I. Bounia, J. Polym. Sci. Part B: Polym. Phys., 46, 1939 (2008).
[38.] B. Dong, C. Liu, Y. Lu, and Y. Wu, J. Appl. Polym. Sci., 132, 42075 (2015).
[39.] J.H. Ma, Y.X. Wang, L.Q. Zhang, and Y.P. Wu, J. Appl. Polym. Sci., 125, 3484 (2012).
[40.] J.H. Ma, L.Q. Zhang, and Y.P. Wu, J. Macromol. Sci. B, 52, 1128 (2013).
[41.] A.R. Payne, J. Appl. Polym. Sci., 6, 5, (1962).
Bin Dong, (1) Chang Liu, (1) Yonglai Lu, (1,2) Liqun Zhang, (1,2) Youping Wu (1,2)
(1) State Key Laboratory of Organic-Inorganic Composites, College of Materials Science and Engineering, Beijing University of Chemical Technology, Beijing 100029, China
(2) Beijing Engineering Research Center of Advanced Elastomers, College of Materials Science and Engineering, Beijing University of Chemical Technology, Beijing 100029, China
Additional Supporting Information may be found in the online version of this article.
Correspondence to: Y. Wu; e-mail: email@example.com
Contract grant sponsor: State Key Program of National Natural Science of China; contract grant number: 51333004; contract grant sponsor: National Basic Research Program of China; contract grant number: 2011CB932603; contract grant sponsor: National Basic Research Program of China; contract grant number: 2015CB654700 (2015CB654703); contract grant sponsor: Program for New Century Excellent Talents in University; contract grant number: NCET-12-0762.
Published online in Wiley Online Library (wileyonlinelibrary.com).
TABLE 1. Formulae of the SBR composites, phr. (a) Sample CNTs-0 u-CNTs-1 u-CNTs-2 u-CNTs-3 SBR 100 100 100 100 CB 50 46 42 38 CNTs -- 1 2 3 Aromatic oil 10 10 10 10 Zinc oxide 3 3 3 3 Stearic acid 2 2 2 2 4010NA (b) 2 2 2 2 Wax 1.5 1.5 1.5 1.5 DM (c) 1.2 1.2 1.2 1.2 D (d) 0.6 0.6 0.6 0.6 Sulfur 1.5 1.5 1.5 1.5 Sample e-CNTs-1 e-CNTs-2 e-CNTs-3 SBR 100 100 100 CB 49 48 47 CNTs 1 2 3 Aromatic oil 10 10 10 Zinc oxide 3 3 3 Stearic acid 2 2 2 4010NA (b) 2 2 2 Wax 1.5 1.5 1.5 DM (c) 1.2 1.2 1.2 D (d) 0.6 0.6 0.6 Sulfur 1.5 1.5 1.5 (a) phr: parts per hundred parts of rubber. (b) 4010NA: iV-Isopropyl-Ar-phenyl-4-phenylenediamin. (c) DM: 2,2'-dibenzothiazole disulfide. (d) D: 1,3-diphenyl guanidine. TABLE 2. Mechanical properties of CNTs/CB/SBR composites. Sample CNTs-0 u-CNTs-1 u-CNTs-2 u-CNTs-3 Stress at 100%/MPa 1.7 1.9 2.1 2.2 Tensile strength/MPa 23.0 23.2 23.6 23.8 Elongation at break/% 643 512 555 570 Shore A hardness 64 65 66 67 Tear Strength/k[Nm.sup.-1] 53.1 57.2 57.6 58.0 Sample e-CNTs-1 e-CNTs-2 e-CNTs-3 Stress at 100%/MPa 2.3 2.4 2.7 Tensile strength/MPa 24.0 24.2 25.3 Elongation at break/% 655 643 594 Shore A hardness 67 69 70 Tear Strength/k[Nm.sup.-1] 55.7 56.3 57.7 TABLE 3. Tearing modulus ([T.sub.R]) of SBR composites at pre-cut length a = 4mm. Parameter CNTs-0 u-CNTs-1 u-CNTs-2 u-CNTs-3 [T.sub.R]/MPa 12.3 13.9 14.0 16.6 [R.sup.2] 0.97 0.98 0.98 0.98 Parameter e-CNTs-1 e-CNTs-2 e-CNTs-3 [T.sub.R]/MPa 11.8 9.9 9.4 [R.sup.2] 0.96 0.98 0.99
Please note: Some tables or figures were omitted from this article.
|Printer friendly Cite/link Email Feedback|
|Author:||Dong, Bin; Liu, Chang; Lu, Yonglai; Zhang, Liqun; Wu, Youping|
|Publication:||Polymer Engineering and Science|
|Date:||Dec 1, 2016|
|Previous Article:||Toughening of polylactide nanocomposites with an ethylene alkyl acrylate copolymer: effects of the addition of nanoparticles on phase morphology and...|
|Next Article:||Preparation and characterization of chitosan-graft-poly(L-lactic acid) microparticles.|