multiwalled carbon nanotube hybrids as reinforcing filler in silicone rubber

Composites: Part A 56 (2014) 290–299

Contents lists available at ScienceDirect

Composites: Part A journal homepage: www.elsevier.com/locate/compositesa

Layered double hydroxide/multiwalled carbon nanotube hybrids as reinforcing filler in silicone rubber B. Pradhan, S.K. Srivastava ⇑ Inorganic Materials and Nanocomposite Laboratory, Department of Chemistry, Indian Institute of Technology, Kharagpur 721302, India

a r t i c l e

i n f o

Article history: Received 12 March 2013 Received in revised form 29 August 2013 Accepted 14 October 2013 Available online 24 October 2013 Keywords: Hybrid Nano-structures Thermal properties Mechanical properties

a b s t r a c t The dispersion of filler and filler–matrix interfacial interaction are crucial factors to improve the properties of nanocomposites. In the present work, the dry grinding of 1D multiwalled carbon nanotube (MWCNT) and 2D layered double hydroxides (Li–Al-LDH, Mg–Al-LDH and Co–Al-LDH) have been used to prepare the corresponding 3D Li–Al-LDH/MWCNT, Mg–Al-LDH/MWCNT and Co–Al-LDH/MWCNT hybrids and characterized. Subsequently, these 3D hybrids are used as nanofiller in the development of silicone rubber (SR) nanocomposites. Tensile strength is found to be significantly improved by 134%, 100% and 125% compared to neat SR in 1 wt.% Mg–Al-LDH/MWCNT, Li–Al-LDH/MWCNT and Co–Al-LDH/MWCNT hybrids filled SR nanocomposites respectively. It is also noted that Mg–Al-LDH/ MWCNT/SR nanocomposites exhibit superior thermal stability and swelling behavior. The best effect of Mg–Al-LDH/MWCNT on SR compared to other hybrids is related to the highest surface area which contributes to nanolevel dispersion and strong interfacial interaction between Mg–Al-LDH/MWCNT and SR matrix. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction In recent years, polymer nanocomposites have received a considerable amount of attention due to their enhanced mechanical, electrical, thermal, flame retardant and other functional properties for better applications compared to the neat polymer [1–3]. Such improvements in properties strongly depend on the nature and properties of filler, dimension, aspect ratio, dispersion of the filler and interfacial interaction between matrix and the filler. Silicone rubber (SR) is one of most important functional polymers with its flexible silicone–oxygen backbones. Though, it exhibits excellent physical, chemical and thermal properties, the low thermal/ electrical conductivity and poor mechanical strength remains one of the bigger challenges for many multifaceted applications. In this regard, one-dimensional (1D) carbon nanotubes (CNT) have been considered as most exciting nanofillers in the development of polymer nanocomposites [4–6] due to their excellent mechanical strength and high electrical and thermal conductivity. Compared with 1D CNT, layered double hydroxides (LDH) are one kind of two-dimensional (2D) nanomaterials which have been also used as nanofiller for fabrication of polymer nanocomposites [7,8]. For both 1D and 2D nanofillers, homogeneous dispersion and strong interactions with the matrices maintaining the intrinsic properties

⇑ Corresponding author. Tel.: +91 03222 283334; fax: +91 3222 255303. E-mail address: [email protected] (S.K. Srivastava). 1359-835X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.compositesa.2013.10.011

of nanofillers are the most important issue. Many studies have been conducted on modification and dispersion of CNT [9–11] and LDH [7,8] in polymeric matrices. However, considering special structure and properties of CNT and LDH, it is very interesting to prepare 3D LDH/CNT by the hybridization of 1D CNT and 2D LDH for their promising applications in the field of hydrogen storage device [12], electro catalyst [13], photo degradation of dye etc [14]. Recently, this kind of hybrids have been prepared by in-situ growth of CNTs on LDH [12,15–17], co-precipitation method [13,14], hydrothermal method [18,19], wet mixing [20] and dry grinding of CNT and LDH [21]. Huang et al. [20] reported the synergistic effect of Co–AlLDH/CNT hybrid filler on the mechanical properties of polyamide 6. Motivated by this, we focused our work on the preparation of 3D hybrids by dry grinding of CNT and three different LDH viz. Li-Al-LDH, Mg-Al-LDH and Co–Al-LDH and their characterization. Subsequently, these 3D hybrid fillers have been used as reinforcing fillers in the development of SR nanocomposites.

2. Experimental 2.1. Materials Commercially available silicone rubber (Baysilone U10, vinyl content 0.05 mmol/g), vinyl terminated, linear polydimethylsiloxane base polymer and Baysilone U crosslinking agent 430, polysiloxane,

B. Pradhan, S.K. Srivastava / Composites: Part A 56 (2014) 290–299

and Pt catalyst complex were supplied by Momentive Performance Materials, Bangalore, India. Ethynyl cyclohexanol (Inhibitor) and carbon nanotube, multiwalled 724769 (carbon > 95%, O.D  L 6– 9 nm  6 lm) were purchased from Sigma-Aldrich. Magnesium nitrate [Mg(NO3)26H2O], aluminum nitrate [Al(NO3)39H2O], cobalt nitrate [Co(NO3)26H2O] and tetrahydrofuran (THF) were purchased from E. Merck, India. Lithium nitrate (LiNO3) and sodium hydroxide (NaOH) was obtained from Loba Chemie Pvt. ltd., Mumbai and Quest Chemicals (Kolkata, India) respectively. 2.2. Synthesis of Li–Al-LDH, Mg–Al-LDH, Co–Al LDH Li–Al-LDH, Mg–Al-LDH and Co–Al-LDH were prepared according to the co-precipitation method [22]. In this method, Mg(NO3)26H2O (0.025 mol, 19.65 g) and Al(NO3)39H2O (0.075 mol, 9.25 g) were first dissolved completely in 100 mL of H2O and then added drop wise to aqueous NaOH solution (pH  8–9) and stirred time to time. The resulting product was subsequently aged at 80 °C for 16 h followed by filtration and washing with distilled water and dried under vacuum at 80 °C/ 24 h. The similar methodology was adopted for the preparation of Li–Al-LDH and Co–Al-LDH. 2.3. Preparation of LDH/MWCNT hybrid LDH/MWCNT hybrids were prepared by simple dry grinding of pristine LDH (Li–Al-LDH, Mg–Al-LDH and Co–Al-LDH) and MWCNT with various weight ratios (6:1, 3:1, 2:1, 1:1, 1:2) in a mortar-pestle for half an hour [21]. The sides of the mortar were also occasionally scraped down with the pestle during grinding in order to ensure the through mixing. 2.4. Preparation of LDH/MWCNT/Silicone rubber composites LDH/MWCNT/silicone rubber composites were prepared by solution intercalation method. A desired amount (0.5, 0.75, 1.0 and 1.5 wt.%, as the case it may be) of different LDH/MWCNT hybrids (1:1) were dispersed separately in THF and mixed with vinyl terminated, linear polydimethylsiloxane base polymer solution by ultrasonication for 1 hour. Then, the appropriate amounts of crosslinker (V430) with a 3:1 mole ratio of hydride (crosslinker) to the vinyl group of base polymer, catalyst and inhibitor were added to the above mixture at room temperature. Finally, the resultant mixture was cast on a Teflon Petri dish for solvent evaporation and then cured at 165 °C for 15 min followed by post curing at 200 °C for 4 h in a hot air oven. This same methodology was adopted for neat SR, MWCNT/SR, Li–Al-LDH/SR, Mg–Al-LDH/SR, and Co–AlLDH/SR composites. 2.5. Characterization Wide angle X-ray diffraction (WAXD) patterns were conducted at room temperature on a PANalytical (PW3040/60), model ‘X’ pert pro with Cu Ka radiation (k = 0.1542 nm) in the range of diffraction angle 2h = 10–70 and 5–60° at a scanning rate of 3°/min. The average crystallite size of LDH and LDH/MWCNT hybrids was calculated using the Scherrer equation [23,24]:

Lhkl ¼

Kk b cos h

where k is 0.9 (constant), Lhkl is the crystallite size, k is the wavelength of the radiation (k = 0.15418 nm), b is the width of the (hkl) peak at half maximum intensity, and h is the Bragg angle. Microstrain (e) and dislocation density (q) were also calculated as follows [25]e ¼ b2h 4cos h, where, b2h is the full width at half maximum and h is the Bragg angle.

291

q ¼ aL15hkle , where the value of a = 3.15  108 A°. Field emission scanning electron microscopy (FESEM) images were conducted on a Carl Zeiss Supra 40 instrument at an accelerating voltage of 20 kV. The samples were coated with Au before FESEM measurements. Transmission electron microscopy (TEM) images of LDH/MWCNT hybrid and its SR composites were recorded using a JEOL 2100 200 KV transmission electron microscope. The corresponding for TEM probes of the LDH/MWCNT hybrids were prepared by mounting the THF suspension of the samples on the copper grid for about 12 h. TEM samples of LDH/ MWCNT filled SR composite were prepared by ultramicrotomy using a diamond knife at 130 °C with a thickness of a section 100 nm. AFM analysis was performed on AFM (Digital Instruments, Nanoscope III), XRD (RigakuMiniflex), by spin-coating of samples in silicon wafer. Brunauer, Emmett, and Teller (BET) analysis is the standard method for determining surface areas from nitrogen adsorption isotherms using the BET surface area analyzer (AUTOSORB1C), supplied by Blue Star India Ltd for Quantachrome Instruments, USA. Ultaviolate–visible (UV–Vis) absorption spectra of MWCNT and LDH/MWCNT hybrids dispersion were recorded on Shimadzu UV-2450 UV–Vis spectrophotometer. For this purpose, 0.2 mg MWCNT and LDH/MWCNT hybrids were separately dispersed in 1 mL distilled water by 5 min sonication. Zeta potential has been measured using Malvern NanoZS. To check the dispersion stability, the hybrids were separately dispersed by ultrasonication in vial consisting 0.8 mg filler in 4 mL of THF and water respectively. Tensile measurements were performed according to ASTMD 412-98 using a Tinius Olsen h10KS universal testing machine at 25 °C with a crosshead speed of 300 mm/min; dumbbell shape specimens with overall length of 100 mm and width of 3 mm. Differential scanning calorimetry (DSC) of neat SR and SR nanocomposites was conducted using Perkin Elmer Pyris differential scanning calorimetric instrument at scan rate of 10 °C/min under nitrogen atmosphere over a range of temperature 150 to +150 °C. Thermo gravimetric analysis (TGA) of the neat SR and SR nanocomposites was performed using Redcroft 870 thermal analyzer, Perkin–Elmer at a heating rate of 10 °C/ min over a temperature range of 50–800 °C under nitrogen atmosphere. The volume fraction and crosslink density of polymer and polymer composite in the swollen network were estimated by following techniques. Previously weighed samples were allowed to swell in toluene to reach equilibrium swelling for 72 h at room temperature. Then the sample masses in the equilibrium swollen state were removed from toluene, blot-dryed with a tissue paper and weighed it. Finally the samples were dried to a constant weight in a vacuum oven for minimum of 24 h. The volume fraction (Vr) of neat polymer and polymer composites were calculated using the Flory–Rehner approach [26]:

Vr ¼

W r q1 r 1 W r q1 r þ W s qs

where qr (Wr) and qs (Ws) are the density (weight) of polymer/composite and solvent respectively. The crosslink density (nc) and toluene uptake (mole%) of the polymer and polymer composite under equilibrium swollen conditions were also calculated from the relationship:

nc ¼

½lnð1  V r Þ þ V r þ vV 2r   1  V 0 V 3r  V2r

where nc, V0, v, Wr and Ws are the crosslink density, molar volume of solvent (Toluene, V0 = 106.8 cm3/mol), Flory-Huggins polymer–solvent

292

B. Pradhan, S.K. Srivastava / Composites: Part A 56 (2014) 290–299

interaction parameter, and weight of the polymer/polymer composites respectively. 3. Results and discussion 3.1. Structural characterization of LDH/MWCNT hybrids Fig. 1. records wide angle X-ray diffraction (WAXD) spectra of pristine LDH (Li–Al-LDH, Mg–Al-LDH, Co–Al-LDH) and LDH/ MWCNT hybrids in 6:1, 3:1, 2:1, 1:1 and 1:2 wt. ratio. Mg-AlLDH and Co-Al-LDH show the presence of the characteristic (0 0 3), (0 0 6), (1 1 0), (1 1 3) reflections [27], whereas diffraction peaks corresponding to (0 0 2), (0 0 4), (0 0 6), (3 3 0), (6 0 0) planes appeared in case of Li–Al-LDH [28,29]. Fig. S1 (under supporting information) shows an intensive and narrow (0 0 2) diffraction peak at about 25.85° (JCPDS No. 41-1487) in MWCNT. Table S1 (under supporting information) records crystallite size, microstrain and dislocation density of LDH and LDH/MWCNT hybrids. Interestingly, the intensity of 00l and 11l reflections of LDH in LDH/MWCNT is significantly reduced. Furthermore, it is also observed that peaks are asymmetric at the lower angle than on the higher angle regions. It is well known that such asymmetric behavior is due to the introduction of layer disorder in LDH in presence of MWCNT [30,31]. Imperfection of crystallite as a result of lattice strain, dislocation and stacking faults are also known to contribute towards the broadening of X-ray diffraction peak of LDH/MWCNT hybrids. Table S1 (under supporting information) also shows that all LDH/ MWCNT (1:1) hybrids exhibit the lowest crystallite sizes, the highest microstrain and dislocation density of LDH in all probability due to the confined effect of MWCNT on the LDH crystallites [19]. Therefore, LDH/MWCNT (1:1) hybrids have further been characterized by different physicochemical techniques and also subsequently used as reinforcing fillers in the development of SR nanocomposites. Fig. 2 shows FESEM images of Li–Al-LDH/MWCNT, Mg–Al-LDH/ MWCNT, Co–Al-LDH/MWCNT hybrids. The homogeneous distribution of MWCNT and LDH as well as the confinement of MWCNT on the surface of Mg–Al-LDH and Co–Al-LDH nanoplatelets is clearly evident in the respective hybrids. On the contrary, the heterogeneous mixture of Li–Al-LDH and MWCNT is formed in Li–Al-LDH/ MWCNT hybrid. These images also show that the bundles of MWCNT tend to bulge out through the layers of LDH (indicated by red circle). This indicates the formation of a 3-dimensional network comprising 2D LDH and 1D MWCNTs similar to MoS2/ MWCNT [32], MWCNT/Graphene [33], WS2-MWCNT [34] and TiS2-MWCNT [35]. TEM analysis of the hybrids in Fig. 3 clearly

shows that MWCNT is well dispersed in presence of Mg–Al-LDH and Co–Al-LDH. On the contrary, the dispersion of MWCNT is poor in case of Li–Al-LDH/MWCNT hybrid. The surface profile analysis of Mg-Al-LDH/MWCNT has also been carried through AFM technique to further ascertain our contention on formation of three-dimensional network. Fig. 4(a) clearly indicates that Mg–Al-LDH is present on the surface with MWCNT protruding in the third dimension (viz. Z-axis). In addition, the distribution of MWCNT appeared to be continuous indicating its homogenous dispersion on the LDH platelets. In addition, the section profile in Fig. 4(b) of the marked cross section shows the about 7 nm variation in the height. This could be attributed to the MWCNT bulging out from the flat surface of LDH. The surface area of pristine LDH and LDH/MWCNT has been determined by nitrogen gas adsorption-desorption isotherm. It is noted that the surface area of pristine LDH follows the order: Li– Al-LDH (33 m2/g) < Co–Al-LDH (45 m2/g) < Mg–Al-LDH (97 m2/g). The maximum surface area of Mg-Al-LDH suggests that it can disperse MWCNT in most effective manner. The surface area of Mg– Al-LDH/MWCNT (188 m2/g) is maximum compared to either Li– Al-LDH/MWCNT (155 m2/g) and Co–Al-LDH/MWCNT (160 m2/g). Therefore, it is anticipated that Mg–Al-LDH/MWCNT could be one of the best reinforcing fillers in SR. Fig. 5 dispaly the UV–Vis spectra of MWCNT and LDH/MWCNT hybrids dispersed in water. The absorbance of MWCNT solutions shows a maximum 271 nm and gradually decreases from UV to near-IR [36]. This is attributed to scattering, especially in the lower wavelength range. Similar results have been reported for UV–Vis absorption spectra of SWCNTs by few workers [37,38]. However, this absorption peak is red shifted in Li–Al-LDH/MWCNT, Mg–AlLDH/MWCNT and Co–Al-LDH/MWCNT hybrids to 273, 278 and 278 nm respectively. It is also noted that the absorbance of MWCNT in Mg–Al-LDH/MWCNT is maximum compared to other Li–Al-LDH/MWCNT and Co–Al-LDH/MWCNT hybrids. According to the UV–Vis analysis, the Mg–Al-LDH is most effective for enhancing the MWCNT dispersion in water. The zeta potential of aqueous suspensions of pristine MWCNT and LDH/MWCNT hybrids has been measured at pH = 7 and follows the order: Mg-Al-LDH/MWCNT (25 mV) > Co–Al-LDH/ MWCNT (19 mV) > Li–Al-LDH/MWCNT (15.75 mV) > MWCNT (13.5 mV). This indicates that the aqueous suspension of Mg– Al-LDH/MWCNT hybrid is most stable compared to other due to the larger surface area of Mg–Al-LDH which contributes to the strong interaction between positively charged Mg–Al-LDH layers and negatively charged MWCNT. As a result, van der Waal interaction between individual MWCNTs is reduced, which could

Fig. 1. WAXD patterns of (A) Li–Al-LDH and Li–Al-LDH/MWCNT hybrids, (B) Mg–Al-LDH and Mg–Al-LDH/MWCNT hybrids, (C) Co–Al-LDH and Co–Al-LDH/MWCNT hybrids.

B. Pradhan, S.K. Srivastava / Composites: Part A 56 (2014) 290–299

293

suspension of Co–Al-LDH/MWCNT is partially aggregated, whereas Mg–Al-LDH/MWCNT remained stable even up to 24 h. Therefore, it may be concluded that the Mg–Al-LDH/MWCNT could be one of the most suitable nanofiller in the preparation of SR composites by solution blending using THF as a solvent. 3.2. Nanostructure of SR composites TEM analysis provides direct evidence about the dispersion of the fillers in SR matrix. Fig. 7 compares the morphology of 1 wt.% loaded Li–Al-LDH/MWCNT, Mg–Al-LDH/MWCNT and Co–Al-LDH/ MWCNT of SR composites. It clearly indicates that the MWCNT remains aggregated in the Li–Al-LDH/MWCNT/SR composite as compact bundles. In case of Co–Al-LDH/MWCNT/SR composite, MWCNT are individually distributed in SR matrix along with few composed tactoids. Interestingly, the Mg–Al-LDH/MWCNT/SR composite shows that the MWCNT are homogeneously distributed in SR matrix. Fig. 8 shows WAXD analysis of SR, Mg–Al-LDH/SR, MWCNT/SR and Mg–Al-LDH/MWCNT/SR composites, whereas the coprresponding WAXD patterns of Li–Al-LDH and Co–Al-LDH loaded SR are provided under supplimentry information as Figs. S2 and S3 respectively. It is observed that neat SR exhibits a broad diffraction peak at 21.31° due to the amorphous regimes of silicone rubber and a strong peak at 12.28° suggesting the inherent crystallinity of PDMS arising from the un-stretched chain conformations [40]. Interestingly, the diffraction peaks of LDH hybrid fillers are absent in the corresponding LDH/MWCNT/SR composites. It is also noted that the full width at half maximum (FWHM) of SR and its LDH/ MWCNT at all filler loadings follows the order: SR > Li–Al-LDH/ MWCNT/SR > Co–Al-LDH/MWCNT/SR > Mg–Al-LDH/MWCNT/SR. This indicates that the SR chains are most ordered in presence of Mg–Al-LDH/MWCNT hybrid filler. This could be attributed to the reduction in the distortions, such as dislocations, point defects etc, in the crystalline grain of polymer segments [41]. 3.3. Mechanical property of SR composites

Fig. 2. FESEM images of Li–Al-LDH/MWCNT (a), Mg–Al-LDH/MWCNT (b) and Co– Al-LDH/MWCNT (c). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

ultimately account for its colloidal stability [39]. Our findings are also in agreement with UV–Vis studies as discussed earlier. Room temperature stability of MWCNT and LDH/MWCNT suspensions has also been investigated in THF by sedimentation experiment and the corresponding images after 1 day are displayed in Fig. 6. It is observed that pristine MWCNT is poorly dispersed in THF due to its instantaneous aggregation. Our experiments also confirmed that Li–Al-LDH/MWCNT suspension is completely sediment at the bottom of the vial after 12 h. The

Fig. 9 represents the stress-strain plots of SR filled with different loadings (0, 0.5, 0.75, 1.0 and 1.5 wt.%) of Mg–Al-LDH/MWCNT composites, whereas the similar images of Li-Al-LDH/MWCNT/SR and Co–Al-LDH/MWCNT/SR are provided under supplementary information as Figs. S4 and S5 respectively. The corresponding mechanical property data of all the composites are provided in Table S2. Though, tensile strength (TS) and elongation at break (EB) of the hybrid filled/SR composites are always higher compared to neat SR, the maximum improvements are observed for 1 wt.% of Mg–Al-LDH/MWCNT, Li–Al-LDH/MWCNT and Co–Al-LDH/MWCNT loaded SR. Such improvements in the mechanical properties of SR could be due to the synergistic effect of 1D MWCNT and 2D LDH fillers. In order to address this issue, the stress-strain behavior of neat SR, Mg–Al-LDH (0.5 wt.%)/SR, MWCNT (0.5 wt.%)/SR and Mg–Al-LDH/MWCNT (1.0 wt.%)/SR has been studied and the findings are displayed in Fig. 10. Further, the corresponding plots of SR, Li-Al-LDH (0.5 wt.%)/SR, MWCNT (0.5 wt.%)/SR, Li–Al-LDH/ MWCNT (1.0 wt.%)/SR and SR, Co–Al-LDH (0.5 wt.%)/SR, MWCNT (0.5 wt.%)/SR, Co–Al-LDH/MWCNT (1.0 wt.%)/SR are provided under supplementary information as Figs. S6 and S7 respectively. It shows that the TS of SR is increased by 46, 10, 21 and 12% for 0.5 wt.% of MWCNT, Li–Al-LDH, Mg–Al-LDH and Co–Al-LDH fillers respectively. However, EB of these composites is always less compared to neat SR. It is also noted that the TS of SR and its 1 wt.% hybrid filled SR composites follow the order: Mg-Al-LDH/MWCNT/SR (0.75 MPa) > Co-Al-LDH/MWCNT/SR (0.72 MPa) > Li-Al-LDH/ MWCNT/SR (0.68 MPa) > SR (0.32 MPa). These findings clearly demonstrate that the tensile strength of SR is improved by about

294

B. Pradhan, S.K. Srivastava / Composites: Part A 56 (2014) 290–299

Fig. 3. TEM images of Li–Al-LDH/MWCNT (a), Mg–Al-LDH/MWCNT (b) and Co–Al-LDH/MWCNT (c).

Fig. 4. (a) Typical 3-D AFM image of Mg–Al-LDH indicating protruding MWCNT out from Mg–Al-LDH surface and (b) Section profile of Mg–Al-LDH/MWCNT with its corresponding height profile. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

B. Pradhan, S.K. Srivastava / Composites: Part A 56 (2014) 290–299

Absorbance

0.12

0.08

c d b a

0.04

0.00 300

400

500

600

Wave length (nm) Fig. 5. UV-vis absorption spectra of the aqueous suspension of MWCNT (a), Li–Al-LDH/MWCNT (b), Mg–Al-LDH/MWCNT (c) and Co–Al-LDH/MWCNT (d). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 6. Digital images showing dispersion of MWCNT (a), Li–Al-LDH/MWCNT (b), Co–Al-LDH/MWCNT (c) and Mg–Al-LDH/MWCNT (d) in THF at room temperature (Photographs recorded after 1 day). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

134%, 125% and 100% in Mg–Al-LDH/MWCNT/SR, Co–Al-LDH/ MWCNT/SR and Li–Al-LDH/MWCNT/SR composites respectively. It is also noted that EB of individually filled SR composites is considerably reduced with respect to neat SR. Interestingly, EB of SR is nearly recovered in case of Li–Al-LDH/MWCNT/SR whereas it is improved by 14% and 11% in Mg–Al-LDH/MWCNT/SR and Co–Al-LDH/ MWCNT/SR respectively. All these findings further confirm our earlier contention that the synergistic effect of 3D hybrid fillers makes a definite contribution in improving the mechanical properties of SR. Furthermore, there is a possibility that the highest specific surface area of Mg–Al-LDH/MWCNT hybrid and its homogeneous dispersion could result in the strong interfacial interaction and efficient stress transfer between SR and Mg–Al-LDH/MWCNT fillers. This could account for the maximum improvement in the mechanical properties Mg–Al-LDH/MWCNT/SR composite. It is also evident that EB of SR gradually increases in presence of LDH/MWCNT hybrids and attains maximum value at 1 wt.% filler loading. This is in all probability due to the entanglement of polymer chain/synergistic effect of chain slippage, platelet orientation of LDH and deformation of the MWCNT. However, TS and EB of SR decreases at higher filler loadings due to the aggregation tendency of the LDH/MWCNT.

295

The improvements in mechanical properties of LDH/MWCNT hybrid filled SR composites have also been correlated on the basis of its fracture behavior. Fig. 11 displays the corresponding FESEM images of tensile fracture surface of 1 wt.% filled Li–Al-LDH/ MWCNT, Mg–Al-LDH/MWCNT and Co–Al-LDH/MWCNT composites of SR. It is apparent from the image of Li–Al-LDH/MWCNT/SR that MWCNT are pulled out from SR matrix suggesting the poor interaction between hybrid and rubber matrix. However, the partial extraction of MWCNT is observed in the corresponding Co–Al-LDH/MWCNT/SR composite. On the contrary, the fracture surface image of Mg–Al-DH/MWCNT/SR composites shows that MWCNT as well as Mg–Al-LDH particles are homogeneously dispersed in SR. In addition, the strong interfacial interaction between Mg–Al-LDH/MWCNT hybrid and SR results in the effective wrapping of SR chains on hybrid filler without any pullout of MWCNT from rubber matrix. 3.4. Thermal property of SR composites DSC plots of neat SR and its 1.0 wt.% Li–Al-LDH/MWCNT, Co–AlLDH/MWCNT and Mg–Al-LDH/MWCNT loaded SR composites are shown in Fig. 12 and the glass transition temperature (Tg), crystalline temperature (Tc) and melting temperature (Tm) and enthalpy of melting (DHm) data are provided in Table S3 under supplementary information. It is observed that Tg, Tc and Tm of hybrid filled SR composites are improved compared to neat SR. The increase in Tg is more likely due to the dispersed hybrid filler restrict the segmental motions of polymer at the polymer–fillers interface. It may be noted that the crystallization peak of SR is not detected due to its transformation to from supercooled to a glassy state without crystallization under the fast cooling [42–45]. However, the further improvement in Tc of the nanocomposites could be attributed to the hybrid filler acting as a nucleating agent of SR [46]. In addition, the enhancement in Tm in all the hybrid filled SR composites could be attributed to the retarded mobility of SR chains due to the interaction between dispersed hybrid filler and SR matrix. Table S3 shows that Tg, Tc, Tm of Mg–Al-LDH/MWCNT/SR composites, when compared to neat SR or its other hybrid filled SR composites, are maximum improved by 5 °C, 8 °C and 5 °C, respectively. Further, we have also carried out the DSC measurement on 0.5 wt.% filled MWCNT, Li–Al-LDH, Mg– Al-LDH and Co–Al-LDH to investigate the synergistic effect of 1D MWCNT and 2D LDH on thermal properties of SR and the corresponding data are summarized in Table S3. It is inferred that Tg, Tc and Tm of all hybrid filled SR composites are relatively higher compared to their respective individually filled SR composites. All these findings clearly demonstrate that LDH (Li–Al-LDH, Mg–Al-LDH, Co– Al-LDH) and MWCNT fillers contribute synergistically to improve the thermal properties of SR. This is in all probability due to the nanolevel dispersion and confined geometry of the hybrid fillers. Alternatively, the entanglement of SR is decreased due to interfacial interaction between the hybrid and SR matrix which consequently improves thermal properties. TGA plots of SR, 1 wt.% loaded pristine as well as hybrid SR composites of MWCNT, Li–Al-LDH, Mg–Al-LDH and Co–Al-LDH have been recorded (Figs. S8–S11 under Supplementary Information) and the corresponding thermal data, such as, degradation temperature corresponding to 10% and 50% wt loss (T10 and T50) and char residue at 800 °C are recorded in Table 1. It is clearly evident that neat SR undergoes two step degradations. The initial weight loss in TG begins at 414 °C due to the depolymerisation of the siloxane chains and is extended up to 585 °C. The final weight loss in TG around at 700 °C corresponds to the complete degradation of polymer backbone. It is noticed that T10 of MWCNT/SR remains virtually unaltered, while T50 is significantly improved to 715 °C, when compared to Mg-Al-LDH/MWCNT/SR nanocomposite. This is more likely due to formation of metal oxides, when LDH

296

B. Pradhan, S.K. Srivastava / Composites: Part A 56 (2014) 290–299

Fig. 7. TEM images of Li-Al-LDH/MWCNT (1%)/SR (a), Mg–Al-LDH/MWCNT (1%)/SR (b) and Co–Al-LDH/MWCNT (1%)/SR (c).

0.8

d

e c b

Stress (MPa)

0.6

0.4

a 0.2

0.0

0

50

100

150

200

Strain (%) Fig. 8. WAXD patterns of SR (a), Mg–Al-LDH (0.5 wt.%)/SR (b), MWCNT(0.5 wt.%)/SR (c), Mg–Al-LDH/MWCNT (0.5 wt.%)/SR (d) and Mg–Al-LDH/MWCNT (1 wt.%)/SR (e).

undergoes decomposition at higher temperature causing the random degradation of SR resulting in sharp decrease in thermal stability [47]. When 10% wt loss is considered as point of comparison, the thermal stability of SR (T10  442 °C) is always inferior compared to its LDH composites. This may be attributed to the evaporation of physically adsorbed as well as intercalated water and the loss of hydroxide groups in LDH. On the contrary, the thermal degradation temperatures of neat SR and LDH composites at 50% wt. loss follow the order: Li–Al-LDH/SR (505 °C) < SR (508 °C) < Mg– Al-LDH/SR (520 °C) < Co–Al-LDH/SR (525 °C). Li–Al-LDH filled SR

Fig. 9. Stress-strain plots neat SR (a), Mg–Al-LDH/MWCNT (0.5 wt.%)/SR (b), Mg–AlLDH/MWCNT (0.75 wt.%)/SR (c), Mg–Al-LDH/MWCNT (1.0 wt.%)/SR (d) and Mg–AlLDH/MWCNT (1.5 wt.%)/SR (e). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

shows almost no improvement in the thermal stability in all probability due to its thermal decomposition product (Li2O) acting as destabilizer for SR [45]. On the contrary, the thermal stability in other LDH composites is improved by 12–17 °C with respect to neat SR. This is in all probability due to the resistance exerted by LDH layers preventing the diffusion of volatile products of SR. These studies also show that T10, T50 and char residue of the corresponding nanocomposites are significantly higher compared

297

B. Pradhan, S.K. Srivastava / Composites: Part A 56 (2014) 290–299

0.8

Table 1 Thermal data of SR and its MWCNT, LDH, and LDH/MWCNT composites.

d

Stress (MPa)

0.6

c b

0.4

a 0.2

Sample

T10 (°C)

T50 (°C)

Residue at 800 (°C)

Neat SR MWCNT (1.0 wt.%)/SR Li–Al-LDH (1 wt.%)/SR Mg–Al-LDH (1 wt.%)/SR Co–Al-LDH (1 wt.%)/SR Li–Al-LDH/MWCNT(1 wt.%)/SR Co–Al-LDH/MWCNT(1 wt.%)/SR Mg–Al-LDH/MWCNT(0.5 wt.%)/SR Mg–Al-LDH/MWCNT(1 wt.%)/SR Mg–Al-LDH/MWCNT(1.5 wt.%)/SR

442 485 427 433 439 429 470 470 485 487

510 715 505 520 530 523 636 602 653 657

0.8 47 9.0 14.0 11.0 15 38 36 38 39

0.0 0

50

100

150

200

250

Strain (%) Fig. 10. Stress-strain plots neat SR (a), Mg–Al-LDH (0.5 wt.%)/SR (b), MWCNT (0.5 wt.%)/SR (c) and Mg–Al-LDH/MWCNT (1.0 wt.%)/SR (d). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

to neat SR or its pristine LDH filled composites. It is also noted that the maximum improvement in T10 and T50 is observed in 1.0 (1.5) wt.% filled Mg–Al-LDH/MWCNT nanocomposite of SR corresponding to 485 (487) and 653 (657) °C respectively. This could be attributed to the hindering effect of the confined geometry originating from MWCNT and LDH together, where MWCNT are inlaid between LDH nanoplatelets forming a sandwiched structure and wrapping around 3D structure of hybrid. T10 and T50 data of 1 wt.% filled LDH/MWCNT/SR, when compared with the composites corresponding individually filled LDH (Li–Al-LDH Mg–AlLDH, Co–Al-LDH) and MWCNT composites of SR confirmed synergistic effect of 1D MWCNT and 2D LDH fillers.

Fig. 12. DSC curves of neat SR (a), Li–Al-LDH/MWCNT (1.0 wt.%)/SR (b), Mg–AlLDH/MWCNT (1.0 wt.%)/SR (c) and Co–Al-LDH/MWCNT (1.0 wt.%)/SR (d). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3.5. Swelling property of SR composites In order to correlate the mechanical property with equilibrium solvent uptake, the swelling experiments have been performed in toluene. Fig. 13 show the variations in crosslink density (nc) and toluene uptake (mol%) of neat SR, 0.5 and 1.0 wt.% MWCNT and LDH filled and 0.5, 0.75, 1.0 and 1.5 wt.% Mg–Al-LDH/MWCNT filled composites of SR. The corresponding data related to Li–AlLDH/MWCNT/SR and Co–Al-LDH/MWCNT/SR composites are also provided in Figs. S12 and S13 under supplementary information. These studies show that the crosslink density gradually increases, whereas the solvent uptake decreases with increasing Li–Al-LDH/ MWCNT, Mg–Al-LDH/MWCNT and Co–Al-LDH/MWCNT contents in SR. However, the highest crosslink density and lowest solvent

up take is observed in Mg–Al-LDH/MWCNT/SR composites. This is probably due to the homogeneous dispersion and strong interfacial interaction between Mg–Al-LDH/MWCNT and SR polymer segments forming a bound polymer to restrict the solvent uptake from surface to bulk. 4. Conclusions LDH/MWCNT hybrids have been prepared by simple dry grinding of different LDH, viz. Li–Al-LDH, Mg–Al-LDH and Co–Al-LDH with MWCNT. Our studies have established that Mg–Al-LDH is most effective in dispersing MWCNT in polar and non polar solvent

Fig. 11. FESEM image of tensile fracture surface of Li–Al-LDH/MWCNT (1%)/SR (a), Mg–Al-LDH/MWCNT (1%)/SR (b) and Co–Al-LDH/MWCNT (1%)/SR (c). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

298

B. Pradhan, S.K. Srivastava / Composites: Part A 56 (2014) 290–299

Fig. 13. Variation of (A) crosslink density and (B) Toluene uptake (mole%) of neat SR (a), SR filled with 0.5 and 1.0 wt.% of Mg–Al-LDH (b and c), SR filled with 0.5 and 1.0 wt.% of MWCNT (d and e) and SR filled with 0.50, 1.0 and 1.5 wt.% Mg–Al-LDH/MWCNT (f, g and h).

due to its cationic charge on the platelet surface and large surface area. Subsequently, the resultant LDH/MWCNT hybrids are used to fabricate SR composites by solution intercalation method. The tensile strength is improved by 134%, 100% and 125% compared to neat SR with 1 wt.% of Mg–Al-LDH/MWCNT, Li–Al-LDH/MWCNT and Co–Al-LDH/MWCNT respectively. Differential scanning calorimetry demonstrates the maximum improvement of glass transition temperature (5 °C), crystalline temperature (8 °C) and melting temperature (5 °C) in Mg–Al-LDH/MWCNT (1 wt.%)/SR composites. Swelling measurements also confirmed that the crosslink density and solvent resistance property is found to be maximum for Mg– Al-LDH/MWCNT/SR composites. The improvement in properties is due to the synergistic effect of MWCNT and LDH on SR matrix. In addition, the excellent synergistic effect of Mg–Al-LDH/MWCNT hybrid is related to the highest surface area which contributes to nanolevel dispersion and strong interfacial interaction between Mg–Al-LDH/MWCNT with SR matrix to improve the properties. Acknowledgments The authors are thankful to CSIR and DRDO, New Delhi, India, for the financial support. Authors also gratefully acknowledge Dr. Anubhav Saxena, Momentive Performance Materials Programs, GE India, Bangalore, for providing the polymer samples. Thanks are also due to Professor S.K. Roy and Ms Susnata Bera (Department of Physics and Meteorology) and Professor Rahul Mitra (Metallurgical and Mateials Engineering Department), Indian Institute of Technology, Kharagpur for AFM and FESEM of the samples respectively.

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.compositesa. 2013.10.011. References [1] Giannelis EP. Polymer layered silicate nanocomposites. Adv Mater 1996;8(1):29–35. [2] Gao Y, Xu X, Gao J, Liang D, Fu Q, Zhang Q, et al. Highly efficient load transfer in polypropylene/vapor grown carbon fiber nanocomposites. Composites Part A 2013;45:35–43. [3] Chieruzzi M, Miliozzi A, Kenny J. Effects of the nanoparticles on the thermal expansion and mechanical properties of unsaturated polyester/clay nanocomposites. Composites Part A 2013;45:44–8. [4] Chua TP, Mariatti M, Azizan A, Rashid AA. Effects of surface-functionalized multi-walled carbon nanotubes on the properties of poly(dimethyl siloxane) nanocomposites. Compos Sci Technol 2010;70(4):671–7.

[5] Jiang MJ, Dang ZM, Xu HP. Enhanced electrical conductivity in chemically modified carbon nanotube/methylvinyl silicone rubber nanocomposite. Eur Polym J 2007;43(12):4924–30. [6] Jiang MJ, Dang ZM, Yao SH, Bai J. Effects of surface modification of carbon nanotubes on the microstructure and electrical properties of carbon nanotubes/rubber nanocomposites. Chem Phys Lett 2008;457(4–6):352–6. [7] Pradhan B, Srivastava SK, Ananthakrishnan R, Saxena A. Preparation and characterization of exfoliated layered double hydroxide/silicone rubber nanocomposites. J Appl Polym Sci 2011;119(1):343–51. [8] Pradhan B, Srivastava SK, Bhowmick AK, Saxena A. Effect of bilayered stearate ion-modified Mg–Al layered double hydroxide on the thermal and mechanical properties of silicone rubber nanocomposites. Polym Int 2012;61(3):458–65. [9] Tchoul MN, Ford WT, Lolli G, Resasco DE, Arepall S. Effect of mild nitric acid oxidation on dispersability, size, and structure of single-walled carbon nanotubes. Chem Mater 2007;19(23):5765–72. [10] Gao C, Muthukrishnan S, Li W, Yuan J, Xu Y, Muller HE. Linear and hyperbranched glycopolymer-functionalized carbon nanotubes: synthesis, kinetics, and characterization. Macromolecules 2007;40(6):1803–15. [11] Williams KA, Veenhuizen TM, Torre G, Eritja R, Dekker C. Nanotechnology: carbon nanotubes with DNA recognition. Nature 2002;420(6917):761. [12] Shaijumon MM, Bejoy N, Ramaprabhu S. Catalytic growth of carbon nanotubes over Ni/Cr hydrotalcite-type anionic clay and their hydrogen storage properties. Appl Surfe Sci 2005;242(1–2):192–8. [13] Wang H, Xiang X, Li F. Facile synthesis and novel electrocatalytic performance of nanostructured Ni–Al layered double hydroxide/carbon nanotube composites. J Mater Chem 2010;20(19):3944–52. [14] Wang H, Xiang X, Li F. Hybrid ZnAl-LDH/CNTs nanocomposites: noncovalent assembly and enhanced photodegradation performance. AIChE J 2010;56(3):768–78. [15] Xiang X, Zhang Lu, Hima HI, Li F, Evans DG. Co-based catalysts from Co/Fe/Al layered double hydroxides for preparation of carbon nanotubes. Appl Clay Sci 2009;42(3-4):405–9. [16] Cao Y, Zhao Y, Li Q, Jiao Q. Catalytic synthesis of nitrogen-doped multi-walled carbon nanotubes using layered double hydroxides as catalyst precursors. J Chem Sci 2009;121(2):225–9. [17] Zhao MQ, Zhang Q, Jia XL, Huang JQ, Zhang YH, Wei F. Hierarchical composites of single/double-walled carbon nanotubes interlinked flakes from direct carbon deposition on layered double hydroxides. Adv Funct Mater 2010;20(4):677–85. [18] Du B, Fang Z. The preparation of layered double hydroxide wrapped carbon nanotubes and their application as a flame retardant for polypropylene. Nanotechnology 2010; 21:315603 (6pp). [19] Xiang X, Bai L, Li F. Formation and catalytic performance of supported ni nanoparticles via self-reduction of hybrid NiAl-LDH/C composites. AIChE J 2010;56(11):2934–45. [20] Huang S, Peng H, Tjiu WW, Yang Z, Zhu H, Tang T, et al. Assembling exfoliated layered double hydroxide (LDH) nanosheet/carbon nanotube (CNT) hybrids via electrostatic force and fabricating nylon nanocomposites. J Phys Chem B 2010;114(50):16766–72. [21] Lan YF, Lin JJ. Observation of carbon nanotube and clay micellelike microstructures with dual dispersion property. J Phys Chem A 2009;113(30):8654–9. [22] Crepaldi EL, Pavan PC, Valim JB. Comparative study of the coprecipitation methods for the preparation of layered double hydroxides. J Braz Chem Soc 2000;11(1):64–70. [23] Patterson AL. The scherrer formula for X-ray particle size determination. Phys Rev 1939;56(10):978–82. [24] Scherrer P, Gottinger Nachrichten (1918);cf. also R Zsigmondy, E’olloidchenzie, 3rd ed.; 1920. p. 394. [25] Kalita PK, Sharma BK, Das HL. Structural characterization of vacuum evaporated ZnSe thin films. Bull Mater Sci 2000;23(4):313–7. [26] Flory PJ, Rehner JH. Statistical mechanics of cross-linked polymer networks I. rubberlike elasticity. J Chem Phys 1943;11(11):521–30.

B. Pradhan, S.K. Srivastava / Composites: Part A 56 (2014) 290–299 [27] Segal SR, Anderson KB, Carrado KA, Marshall CL. Low temperature steam reforming of methanol over layered double hydroxide-derived catalysts. Appl Catal A Genl 2002;231(1–2):215–26. [28] Wang J, Lei Z, Qin H, Zhang L, Li F. Structure and catalytic property of Li–Al metal oxides from layered double hydroxide precursors prepared via a facile solution route. Ind Eng Chem Res 2011;50(12):7120–8. [29] Shumaker JL, Crofcheck C, Tackett SA, Santillan-Jimenez E, Crocker M. Biodiesel production from soybean oil using calcined Li–Al layered double hydroxide catalysts. Catal Lett 2007;115(1–2):56–61. [30] Srivastava SK, Pramanik M, Palit D, Haeuseler H, Ochel M. X-ray diffraction, topographical studies, and thermal behavior of layer-type CdIn2S4-xSex (1.75 6 X 6 2.75). and its lithium intercalation compounds. Chem Mater 2004;16(21):4168–73. [31] Srivastava SK, Pramanik M, Palit D, Mathur BK, Kar AK, Samanta RBK, et al. Electrical, optical, and scanning tunneling microscopic studies on layer type CdIn2S4xSex (1.75 6 X 6 2.75). Chem Mater 2001;13(11):4342–7. [32] Bindumadhavana K, Srivastava SK, Mahanty S. MoS2–MWCNT hybrids as a superior anode in lithium-ion batteries. Chem Commun 2013;49:1823–5. [33] Pradhan Bratati, Srivastava Suneel Kumar. Synergistic effect of three dimensional multi-walled carbon nanotube/graphene nanofiller in enhancing the mechanical and thermal properties of high performance silicone rubber. Polym Inter, 2013 [Ref: PI-13-0160.R1, Manuscript received for minor corrections]. [34] Bindumadhavana K, Srivastava SK, Mahanty S. Tungsten disulfide-multiwalled carbon nanotube hybrid anode for lithium-ion battery. J Nanosci Nanotechnol. doi:10.1166/jnn.2014.8737 [Accepted August 10, 2013]. [35] Bindumadhavana K, Srivastava SK, Mahanty S. TiS2-MWCNT hybrid as high performance anode in lithium-ion battery. J Nanopart Res 2013;15:1950–62. [36] Yu J, Grossiord N, Koning CE, Loos J. Controlling the dispersion of multi-wall carbon nanotubes in aqueous surfactant solution. Carbon 2007;45(3):618–23.

299

[37] Jiang LQ, Gao L, Sun J. Production of aqueous colloidal dispersions of carbon nanotubes. J Colloid Interface Sci 2003;260(1):89–94. [38] Grossiord N, Regev O, Loos J, Meuldijk J, Koning CE. Time-dependent study of the exfoliation process of carbon nanotubes in aqueous dispersions by using UV–Visible spectroscopy. Anal Chem 2005;77(6):5135–9. [39] Vigolo B, Pénicaud A, Coulon C, Sauder C, Pailler R, Journet C, et al. Macroscopic fibers and ribbons of oriented carbon nanotubes. Science 2000;290(5495): 1331–4. [40] Albouy PA. The conformation of poly(dimethylsiloxane) in the crystalline state. Polymer 2000;41(8):3083–6. [41] Qu L, Huang G, Qiang W, Xie Z. Effect of diphenylsiloxane unit content on aggregation structure of poly(dimethylsiloxane-co-diphenylsiloxane). J Polym Sci Part B Poly Phys 2008;46(1):72–9. [42] Dollase T, Wilhelm M, Spiess HW, Yagen Y, Yerushalmi-Rozen R, Gottlie M. Effect of interfaces on the crystallization behavior of PDMS. Interface Sci 2003;11:199–209. [43] Dollase T, Spiess HW, Gottlieb M, Yerushalmi-Rozen R. Crystallization of PDMS the effect of physical and chemical crosslinks. Europhys Lett 2002;60:390–6. [44] Aranguren MI. Crystallization of polydimethylsiloxane: effect of silica filler and curing. Polymer 1998;39:4897–903. [45] Dupont A. Silicones in industrial applications. In: De Jaeger Roger, Gleria Mario, editors. Inorganic Polymers. NOVA Publishers; 2007 [Chapter 2]. [46] Zhang C, Tjiu WW, Liu T, Lui WY, Phang IY, Zhang WD. Dramatically enhanced mechanical performance of nylon-6 magnetic composites with nanostructured hybrid one-dimensional carbon nanotube-two-dimensional clay nanoplatelet heterostructures. J Phys Chem B 2011;115(13):3392–9. [47] Rhein RA. Thermally stable elastomers: a review. Naval Weapons Center 1983;6372:1–39.