Effect of amino-functionalization on the interfacial adhesion of multi-walled carbon nanotubes/epoxy nanocomposites

Yunfeng Luo,Yan Zhao,Jizhe Cai,Yuexin Duan,Shanyi Du

DOI: https://doi.org/10.1016/j.matdes.2011.04.033

IF: 9.417

2012-01-01

Materials & Design

Abstract:Two types of multi-walled carbon nanotubes (MWCNTs) functionalized with different amino-organics, dicyanodiamide and phenylbiguanide, respectively, were achieved in this paper. The physico-chemical properties of MWCNTs before and after amino group modification were characterized by thermogravimetric analysis (TGA), Raman spectroscopy and inverse gas chromatography (IGC). The results showed that amino-functionalization changed evidently the surface properties of MWCNTs, such as the dispersive surface energy (decreased from 122.95 mJ/m 2 to 18.65 mJ/m 2 and 25.69 mJ/m 2 , respectively) and specific surface energy (decreased from 8.84 mJ/m 2 to 0.56 mJ/m 2 and 4.60 mJ/m 2 , respectively) for two functionalized MWCNTs. And then, the interfacial adhesion states of the functionalized MWCNTs/epoxy nanocomposites were investigated using scanning electron microscope (SEM) and dynamic mechanical analysis (DMA). The results also indicated that the dispersion of MWCNTs in epoxy resin and the interfacial adhesion of MWCNTs/epoxy nanocomposites were both strongly dependent on the surface physico-chemical properties of functionalized MWCNTs, and the effect of MWCNTs functionalized by phenylbiguanide with slightly higher polarity was better. Keywords A. Nanomaterials C. Surface treatment I. Adhesion 1 Introduction Since the carbon nanotubes (CNTs) were reported by Iijima in 1991 [1] , they have been attractive as ideal filler in polymer composites for their unique structural, mechanical, electrical, and thermal properties [2–5] . Especially, as structural materials, CNTs-reinforced polymer nanocomposites exhibit a brilliant potential in improving the mechanical properties, comparing with the traditional polymer composite materials [6] . In recent years, CNTs had been applied in different matrices to investigate their dispersion and reinforcement effects in different polymer materials, such as polypropylene [7] , polyethylene [8,9] , epoxy [10–12] , polyvinyl alcohol [13] , polycarbonate [14] and other polymers [15,16] . However, poor dispersing and interfacial bonding conditions of CNTs in matrix polymer due to inert and non-reactive surface, and highly agglomerated state resulting from high Van der Waals attraction, obviously restrain their applications in polymer reinforcement areas [17] . To overcome the obstacles mentioned above, various studies [18–21] had been conducted, such as shear mixing, non-covalent and covalent functionalization, addition of dispersing agent. Covalent functionalization is one of the most promising method, owing to its advantage that it does not only generate chemical bonding between CNTs and polymer matrix, which increases their interfacial adhesion, but also induce better compatibility between CNTs and the polymer matrix. For example, by adding 0.1–1 wt.% MWCNTs grafted with epoxy, Zhou et al. [22] reported 45% increment in tensile strength and 90% improvement in tensile modulus, respectively. Sun et al. [23] successfully grafted PAMAM-0 on the surface of single-walled carbon nanotubes (SWCNTs), which induced enhanced dispersion and adhesion states of CNTs in epoxy resin. Furthermore, many other researchers [24–29] attempted different approaches to graft various organic groups, such as polyamide 6 [24] , polystyrene [26] , amines [27,28] and guanidine [29] , on the surface of CNTs in order to effectively improve the compatibility and adhesion between CNTs and polymer matrix. It is noted that the key issue for obtaining a successful covalent functionalization is to select proper grafted organics with special functional groups which could both improve the compatibility between CNTs and matrix and make CNTs to be reactive with matrix polymer in order to increase interfacial bonding condition. In this paper, two kinds of amino-organics including dicyanodiamide and phenylbiguanide, which have similar structure while different compatibility with epoxy resin, were chosen to be grafted on the surface of MWCNTs. The physico-chemical properties of MWCNTs influenced by amino group modification were characterized by thermogravimetric analysis (TGA), Raman spectroscopy and inverse gas chromatography (IGC). Moreover, the interfacial adhesion states of the functionalized MWCNTs/epoxy nanocomposites were investigated using scanning electron microscope (SEM) and dynamic mechanical analysis (DMA). It was found that both of the dispersion of MWCNTs in epoxy resin and the interfacial adhesion of MWCNTs/epoxy nanocomposites were strongly dependent on the surface physico-chemical properties of functionalized MWCNTs. 2 Experimental 2.1 Materials The MWCNTs (purity > 95%, diameter = 10–30 nm, length = 1–2 μm, special surface area = 40–300 m 2 /g) used in this study were purchased from Shenzhen Nanotechnologies Co., China. Dicyanodiamide and diphenylamine were purchased from Xilong Chemical factory (Santou, China). Phenylbiguanide was synthesized by additive reaction of diphenylamine and dicyanodiamide. The epoxy resin E-51 and curing agent tetraethylenepentamine used in this work were purchased from Tianjin Chemical Reagent Co., Ltd. (China). 2.2 Functionalization of MWCNTs At first, pristine MWCNTs (named as p-MWCNTs) were oxidized by piranha (mixture of sulfuric acid and hydrogen peroxide at the volume ratio of 3:1) in a round bottom flask at 40 °C for 4 h. After that, the resulting dispersion was diluted in deionized water to remove the residual acid, and then filtered. After being dried in vacuum at 60 °C for 12 h, oxidized MWCNTs (named as o-MWCNTs) were obtained. Oxidized MWCNTs, containing carboxylic acid groups on surface, were directly mixed with dicyanodiamide and phenylbiguanide with the help of ultrasonication and stirring at 40 °C for 48 h. After being washed by deionized water for several times to eliminate residual organics, different amino-organics could be successfully grafted onto the MWCNTs surface for the electrical interaction between the carboxylic acid groups on the MWCNTs and amine groups on dicyanodiamide and phenylbiguanide. The MWCNTs functionalized with dicyanodiamide will be named as f 1 -MWCNTs, and the one with phenylbiguanide will be named as f 2 -MWCNTs. The schematic drawing of functionalization is shown in Fig. 1 . 2.3 Preparation of nanocomposites At first, 1 wt.% MWCNTs, including pristine MWCNTs (p-MWCNTs), f 1 -MWCNTs and f 2 -MWCNTs, were added to the epoxy E-51, respectively. Then the mixtures were stirred and followed by sonicating in ultrasonic bath at 60 °C for 2 h to ensure good dispersion. Following this, the mixtures were degassed in vacuum oven at room temperature for 10 min to remove bubbles before adding curing agent. After that, the curing agent was blended into the mixtures, and then mixtures were sonicated in ultrasonic water bath at 30 °C for 10 min and degassed at room temperature for 10 min to eliminate the bubbles. Finally, the specimens were cast in mold and cured at 60 °C for 4 h. 2.4 Characterization Thermogravimetric analysis (TGA) was carried out on a Perkin–Elmer Diamond TG-DTA system to study the thermal stability of the pristine and functionalized MWCNTs, at 5 °C/min from room temperature to 800 °C under a flux nitrogen protection. Raman spectroscopy (Jobin Yvon, HR900) with a 632.8 nm laser line excitation source was used to characterize the changes of covalent attachment of functional groups on the sidewall of MWCNTs. The scanning electron microscope (SEM) HITACHI S4800 was used to observe the morphology of fractured nanocomposites. Dynamic Mechanical Analysis (DMA) of nanocomposites samples with dimensions of 32 mm × 6 mm × 2 mm was carried out on a Rheometric Scientific-DMTA V, using a three-point bend configuration. The tests were run with a frequency of 1 Hz. Thermal scans from room temperature to 200 °C were conducted at a heating rate of 5 °C/min. Inverse gas chromatography (IGC) technique is a powerful technique for investigating the surface characteristics of powders, fibers and thin films. The difference between IGC and conventional gas chromatography (GC) is the solid material packed in the column instead of the injected vapor and the vapors with know physico-chemical properties used as probes, seen in Fig. 2 . In this research, the inverse gas chromatograph analyzer G1530 N from Surface Measurement Systems Ltd. (London, UK) was used to study the surface properties of MWCNTs before and after amino-functionalizations. In the testing, the organic vapors were generated from a liquid reservoir at 373 K, and the flow rates were 10 mL/min. The physico-chemical properties of the probes used in the IGC analysis are supplied in Table 1 . The primary measurement in IGC is the net retention volume, V N , which can be calculated according to the following equation [30] : (1) V N = ( t R - t M ) F j m T 273 K where t R and t M are the retention and dead times measured with specific probe and a non-adsorbing probe (such as methane) respectively, F is the flow rate and T is the column temperature. j is the James–Martin factor for correction of the pressure drop across the packed column between the inlet ( P i ) and outlet ( P o ). (2) j = 3 2 ( P i / P o ) 2 - 1 ( Pi / Po ) 3 - 1 The dispersive component γ S D of the substrates’ surface energy could be calculated by the following equation: (3) RT ln V N = 2 N ( γ S D ) 1 / 2 α ( γ L D ) 1 / 2 + C R is the gas constant, T is the column absolute temperature, α is the surface area of the probe molecule, N is Avogadro’s number, γ L D is the dispersive component of the surface energy of the probe, and the C is the constant. Then, ( γ S D ) 1 / 2 could be obtained from the slope of RT ln V N versus 2 N α ( γ L D ) 1 / 2 . The interactions between polar probes and the substrate involve both dispersive and specific interactions. And the specific interactions include hydrogen bonding, π – π stacking and acid–base interactions, the latter involving much higher energy than the formers. In fact, it is usually assumed that the specific contributions of the adsorption of polar probes are solely from acid–base interactions. The free energy of adsorption, Δ G AB , corresponding to acid–base surface interactions is related to V N by the following equation: (4) - RT ln ( V N / V N ref ) = Δ G AB By applying Good–van Oss equation, γ S - and γ S + values of surface could be calculated by using acidic and basic probe. Then, the specific surface energy, γ S sp , was obtained by using Eq. (6) . (5) Δ G sp = 2 N A a ( ( γ L + · γ S - ) 1 / 2 + ( γ L - · γ S + ) 1 / 2 ) (6) γ S sp = 2 ( γ S + · γ S - ) 1 / 2 3 Results and discussion 3.1 Functionalization of MWCNTs Generally, Raman spectroscopy is a powerful equipment for characterizing the defect on the surface of carbon nanotubes. Therefore, Raman spectroscopy was used to investigate structural evolution of MWCNTs with various modifications. The Raman spectra for the functionalized MWCNTs and pristine MWCNTs are shown in Fig. 3 . It was clearly seen that the two bands around 1575 and 1325 cm −1 in the spectra were assigned to the tangential mode (G-band) for the stretching vibrations of sp 2 carbon atoms and the disorder mode (D-band) induced by sp 3 hybridization, respectively [31,32] . The surface defects of MWCNTs could be evaluated by the ratio of D-band intensity ( I D ) and G-band intensity ( I G ) [33] . The peak intensity ratio ( I D / I G = 1.543) at D-band and G-band for the oxidized MWCNTs (o-MWCNTs) exceeded those of pristine MWCNTs ( I D / I G = 1.244). This result indicates that some of the sp 2 carbon atoms (C C) were converted to sp 3 carbon atoms (C C) at the surface of the MWCNTs after the acid treatment and the carboxylic acid groups were attached on the surface of MWCNTs [34] . Compared with the I D / I G value for p-MWCNTs, the I D / I G ratio of f 1 -MWCNTs ( I D / I G = 1.348) and f 2 -MWCNTs ( I D / I G = 1.477) were both higher than that of the p-MWCNTs, indicating that the functionalization resulted in the structure changes on the surface of MWCNTs. The changes of I D / I G ratio is because of these two aspects: on one hand, the acid treatment destroys the carbon structures on surface of MWCNTs and increases the I D / I G ratio. On the other hand, after functionalization the surface of oxidized MWCNTs is covered by polymer, thus the I D / I G ratio decreases, accordingly the I D / I G ratio of functionalized MWCNT is between o-MWCNT and p-MWCNTs. The difference between I D / I G ratio of f 1 -MWCNT and f 2 -MWCNT may be related to the surface physical and chemical properties of polymer. TGA results of pristine (p-MWCNTs), oxidized (o-MWCNTs) and functionalized MWCNTs are shown in Fig. 4 , which are used to evaluate the weight ratio of the grafted organics on MWCNTs. The weight loss of p-MWCNTs is only 1.04% when the temperature achieved about 800 °C. There is a 5.11% loss of the o-MWCNTs due to pyrolysis of the COOH and OH groups generated during the acid treatment. For f 1 -MWCNTs, the weight loss is approximately 16.72%, which attributes to the decomposition of the dicyanodiamide grafted on the MWCNTs. The decomposition of phenylbiguanide results in 40.20% weight loss for f 2 -MWCNTs. On the basis that the 5.11% weight loss of oxidized MWCNTs is due to decomposition of hydrogenated carbon residues, the actual contents of functional groups are 11.61% and 35.09%, respectively. The molecular weights of dicyanodiamide and phenylbiguanide are 84 and 177, the ratio of which is about 0.475, which is slightly larger than the ratio of the weight loss of f 1 -MWCNTs and f 2 -MWCNTs (0.330). It is demonstrated that functionalization level of f 2 -MWCNTs is slightly higher than one of f 1 -MWCNTs. 3.2 Surface energy analysis The dispersive component of the surface energy of CNTs, γ S D , related to surface density of the atoms, polarizability and ionization energy, is used to investigate the surface condition of CNTs [35] . The dispersive component of the surface energy, γ S D , calculated according to Eq. (3) are presented in Table 2 . The measured γ S D value of p-MWCNTs is about 122.95 mJ/m 2 at 373 K, which is similar to the results reported by others [35,36] . This high surface energy of CNTs is mainly originated from their large surface area, which makes them to obtain a considerable tendency to agglomerate. This is the key obstacle for applying CNTs in reinforcing matrix resin in nanocomposites. The agglomeration of CNTs causes some defects in matrix resin, therefore decreasing the overall performance of the materials. The grafting of amino-organics onto the surface of MWCNTs resulted in the reduction of dispersive surface energy, γ S D . The γ S D values of f 1 -MWCNTs (18.65 mJ/m 2 at 373 K) and f 2 -MWCNTs (25.69 mJ/m 2 at 373 K) are close to those of MWCNTs grafted with poly(acrylic acid) (38 mJ/m 2 at 325.2 K) measured by Zhang et al. [36] . This result showed that the dispersive surface energy of MWCNTs had been evidently changed by functionalization processes since the surface of MWCNTs was covered by the organics. Furthermore, the original large dispersive surface energy of MWCNTs was reduced to a relative small value after the introduction of the amino-organics on their surface, for both f 1 -MWCNTs and f 2 -MWCNTs. The specific free energy, Δ G sp , calculated according to Eq. (5) is shown in Table 2 and Fig. 5 . The Δ G sp values are similar to those measured by Menzel et al. [35] . There is a considerable reduction in value for MWCNTs before and after functionalization measured by different probe molecules. Although the ratios between energy values of different MWCNTs for different probe molecules are not exactly the same due to different interaction between surface of MWCNTs and probe molecules, but the relationship among three kinds of MWCNTs is very obvious that both of the specific free energy values of functionalized MWCNTs are lower than the pristine ones’, reflecting the reduced interaction between functionalized MWCNTs and polar probe molecules. In addition, there is also a distinct difference between two functionalized MWCNTs in the Δ G sp . This difference of specific free energy generates mainly from the grafted organic groups, dicyanodiamide and phenylbiguanide, which have different polarities for their distinct chemical structures, actually, the polarity of phenylbiguanide is slightly higher than dicyanodiamide. According to Eq. (5) and (6) , γ S - , γ S + and γ S sp values were calculated by acidic and basic probes, respectively. As shown in Table 3 . Compared with the p-MWCNTs, the specific surface energy γ S sp for amino-functionalized f 1 -MWCNTs and f 2 -MWCNTs significantly decreased from 8.84 mJ/m 2 (p-MWCNTs) to 0.56 mJ/m 2 (f 1 -MWCNTs) and 4.6 mJ/m 2 (f 2 -MWCNTs), respectively. The similar trend was observed in γ S D and Δ G sp . Both types of surface energy values indicate that the surface energy of amino-functionalized MWCNTs decreased evidently. Decreased dispersive and specific surface energies suggest that the polar groups on the surface of MWCNTs had been replaced by the organic groups with relative lower polarity. The similar result was also obtained by other researchers’ conclusions [35,36] that the surface energy of MWCNTs after grafting functional groups was evidently reduced. Additionally, slightly higher specific surface energy of f 2 -MWCNTs is mainly derived from the greater polarity of phenylbiguanide, for its asymmetric in molecular structure. Meanwhile, the results from IGC are in agreement with the above I D / I G results from Raman spectra. The decreased surface energy after amino-functionalization is advantageous to inhibit CNTs from agglomeration during dispersing in resin [20] and increase the CNTs contents in matrix. 3.3 Morphology characterization To examine the effect of amino-functionalization on MWCNTs’ dispersion in matrix and MWCNTs/epoxy interfacial adhesion state, SEM observation is performed. Fig. 6 is the fractured surfaces of different MWCNTs/epoxy nanocomposites. As indicated in these figures, the MWNCTs before and after amino-functionalization represented distinct difference in dispersion and interfacial adhesion with matrix. As shown in Fig. 6 a and b, the p-MWCNTs dispersed unevenly in epoxy with the presence of agglomeration duo to high surface energy, and poor adhesion between tubes and matrix can be easily observed. Especially, in Fig. 6 b, the p-MWCNTs seem to be separated from the matrix, which shows a poor compatibility of p-MWCNTs with matrix resin, because the inert surface of p-MWCNTs cannot react with the epoxy resin. This poor dispersion and adhesion between carbon nanotubes and matrix not only impair the improvement in the performance of the matrix, but also decrease its overall mechanical properties, for the agglomerated fillers acting as stress concentrators. In Fig. 6 c, the f 1 -MWCNTs functionalized with dicyanodiamide were uniformly dispersed in the epoxy for its low surface energy, compared with the p-MWCNTs. From Fig. 6 d, it can be found that the f 1 -MWCNTs/epoxy interface was not improved evidently, indicating the f 1 -MWCNTs do not obtain a better compatibility with the epoxy after the functionalization process. However, as shown in Fig. 6 e and f, both the dispersion condition and interfacial adhesion with the epoxy resin for f 2 -MWCNTs were improved obviously. There is not clear boundary between f 2 -MWCNTs and matrix in Fig. 6 f, indicating that an excellent adhesion and compatibility between epoxy and f 2 -MWCNTs grafted with phenylbiguanide [37] . 3.4 Dynamical mechanical behavior The DMA technique being used to assess the interfacial adhesion between fiber and matrix was reported in some investigations [38–40] . DMA results of neat epoxy and different MWCNTs/epoxy nanocomposites are shown in Fig. 7 , which exhibit evident changes in dynamic mechanical behavior for pristine and functionalized MWCNTs reinforced epoxy nanocomposites, compared with the neat epoxy sample. As shown in Fig. 7 , the T g of nanocomposites decrease slightly after the addition of pristine and f 1 -MWCNTs, which is similar to other investigation [38,41,42] , because the existence of MWCNTs influenced the normal curing reaction between the curing agent and matrix resin, resultantly decreasing the crosslink density and T g . However, f 2 -MWCNTs/epoxy nanocomposites have a relatively higher T g than pristine and f 1 -MWCNTs/epoxy nanocomposites, indicating that the MWCNTs are totally compatible with the matrix and the reinforcement function has overlapped the negative influence of the sidewall reaction on the thermal stability of the epoxy. Additionally, considering the influence on the storage modulus G′ below T g of the nanocomposites, the change is also considerable. The G′ of p-MWCNTs/epoxy and f 1 -MWCNTs/epoxy have an obvious reduction when comparing with the neat epoxy sample. On the other hand, the modulus of the f 2 -MWCNTs/epoxy is not only higher than the p-MWCNTs/epoxy or f 1 -MWCNTs/epoxy, but also higher than that of pure resin. It is demonstrated that the f 2 -MWCNTs have obtained good dispersion and adhesion states with the matrix resin. The uniform dispersion of MWCNTs in matrix and good adhesion with the matrix make them more effective in bearing and transferring the stress from the matrix, therefore being more useful in stiffening the epoxy resin. Moreover, the decrease in damping peak at the glass transition temperature associates with the improvement of interfaces boding [37,40,42] . The damping peaks of p-MWNCTs/epoxy, f 1 -MWCNTs/epoxy are higher than the values of neat epoxy sample, indicating the poor interfacial bonding between MWCNTs and matrix resin. In contrast, a lower damping peak value of f 2 -MWCNTs means a strong interfacial adhesion between f 2 -MWCNTs and epoxy matrix. Associating with above IGC results, the aggregates of MWCNTs can be obviated and the uniform dispersion of MWCNTs can be achieved through controlling the surface energy. When comparing with the f 1 -MWCNTs, the slightly higher surface energy of f 2 -MWCNTs surface is benefit for the f 2 -MWCNTs surface wetting by the matrix polymer and closer contact with polymer molecules. The reaction occurred between the epoxy resin and the amino functional group on the surface of MWCNTs (shown in Fig. 8 ). Combined with the better compatibility between phenylbiguanide and the epoxy, the interfacial bonding of f 2 -MWCNTs/epoxy is further improved, when compared with that of f 1 -MWCNTs/epoxy. 4 Conclusion Two functionalized MWCNTs with different amino-organics, dicyanodiamide and phenylbiguanide, were obtained respectively. The physico-chemical properties of MWCNTs before and after amino group modification and the MWCNTs/epoxy interfacial adhesion in nanocomposites were characterized. The results showed that amino-functionalization changed evidently the surface energy of MWCNTs, such as the dispersive surface energy decreased from 122.95 mJ/m 2 to 18.65 mJ/m 2 and 25.69 mJ/m 2 for two functionalized MWCNTs, respectively, and the specific surface energy decreased from 8.84 mJ/m 2 to 0.56 mJ/m 2 and 4.60 mJ/m 2 , respectively. The lower surface energy of amino-functionalized MWCNTs is benefit for the uniform dispersion of MWCNTs in the polymer matrix and the better interfacial bonding due to the reaction between the amino group attached on MWCNTs surface and the epoxy resin. In addition, the interfacial adhesion of f 2 -MWCNTs/epoxy nanocomposites is better than that of f 1 -MWCNTs/epoxy nanocomposites, owing to the slightly higher surface energy of functionalized MWCNTs with phenylbiguanide and the improved compatibility between phenylbiguanide and epoxy. Acknowledgment The authors express their sincere gratitude to the Laboratory of Polymer-based Composites, Beihang University, for the financial support of this research. References [1] S. Iijima Helical microtubules of graphitic carbon Nature 354 1991 56 58 [2] M. Terrones Science and technology of the twenty-first century: synthesis, properties and applications of carbon nanotubes Annu Rev Mater Res 33 2003 419 501 [3] J. Hone M.C. Llaguno M.J. Biercuk A.T. Johnson B. Batlogg Z. Benes J.E. Fischer Thermal properties of carbon nanotubes and nanotube-based materials Appl Phys A: Mater Sci Proc 74 2002 339 343 [4] A. Thess R. Lee P. Nikolaev H. Dai P. Petit J. Robert C. Xu Y.H. Lee S.G. Kim A.G. Rinzler D.T. Colbert G.E. Scuseria D. Tomanek J.E. Fischer R.E. Smalley Crystalline ropes of metallic carbon nanotubes Science 273 1996 483 487 [5] M.F. Yu B.S. Files S. Arepalli R.S. Ruoff Tensile loading of ropes of single wall carbon nanotubes and their mechanical properties Phys Rec Lett 84 2000 5552 5555 [6] S. Wang R. Liang B. Wang C. Zhang High strength and multifunctional fabric of single-walled carbon nanotubes Adv Mater 19 2007 1257 1261 [7] K. Prashantha J. Soulestin M.F. Lacrampe P. Krawczak G. Dupin M. Claes Masterbatch-based multi-walled carbon nanotube filled polypropylene nanocomposites: assessment of rheological and mechanical properties Compos Sci Technol 69 2009 1756 1763 [8] N. Xie Q.J. Jiao C.G. Zang C.L. Wang Y.Y. Liu Study on dispersion and electrical property of multi-walled carbon nanotubes/low-density polyethylene nanocomposites Mater Des 31 2010 1676 1683 [9] M.L. Shofner V.N. Khabashesku E.V. Barrera Processing and mechanical properties of fluorinated single-wall carbon nanotube-polyethylene composites Chem Mater 18 2006 906 913 [10] A. Montazeri N. Montazeri Viscoelastic and mechanical properties of multiwalled carbon nanotube/epoxy composites with different nanotube content Mater Des 32 2011 2301 2307 [11] E.T. Thostenson Z. Ren T.W. Chou Advances in the science and technology of carbon nanotubes and their composites: a review Compos Sci Technol 61 2001 1899 1912 [12] A. Allaoui S. Bai H.M. Cheng J.B. Bai Mechanical and electrical properties of a MWNT/epoxy composite Compos Sci Technol 62 2002 1993 1998 [13] X.F. Zhang T. Liu T.V. Sreekumar S. Kumar V.C. Moore R.H. Hauge R.E. Smalley Polyvinyl alcohol/SWNT composite film Nano Lett 3 2003 1285 1288 [14] A. Eitan F.T. Fisher R. Andrews L.C. Brinson L.S. Schadler Reinforcement mechanisms in MWCNTS-filled polycarbonate Compos Sci Technol 66 2006 1162 1173 [15] A. Hirsch Functionalization of single-walled carbon nanotubes Ang Chem Int Ed 41 2002 1853 1859 [16] K.C. Etika M.A. Cox J.C. Grunlan Tailored dispersion of carbon nanotubes in water with pH-repsonsive polymers Polymer 51 2010 1761 1770 [17] M. Zheng A. Jagota E.D. Semke B.A. Diner R.S. Mclean S.R. Lustig R.E. Richardson N.G. Tassi DNA-assisted dispersion and separation of carbon nanotubes Nat Mater 2 2003 338 342 [18] G.L. Hwang Y.T. Shieh K.C. Hwang Efficient load transfer to polymer-grafted multiwalled carbon nanotubes in polymer composites Adv Funct Mater 14 2004 487 491 [19] H. Kong C. Gao D. Yan Controlled functionalization of multi-walled carbon nanotubes by in situ atom transfer radical polymerization J Am Chem Soc 126 2004 412 413 [20] J.F. Shen W.S. Huang L.P. Wu Y.Z. Hu M.X. Ye The reinforcement role of different amino-functionalized multi-walled carbon nanotubes in epoxy nanocomposites Compos Sci Technol 67 2007 3041 3050 [21] M. Wang K.P. Pramoda S.H. Goh Enhancement of interfacial adhesion and dynamic mechanical properties of poly(methyl methacrylate)/multiwalled carbon nanotube composites with amine-terminated poly(ethylene oxide) Carbon 44 2006 613 617 [22] W. Zhou Z.J. Du Y.X. Liu X.P. Yang H.Q. Li C. Zhang Functionalization of MWNTs using polyacryloyl chloride and the properties of CNT-epoxy matrix nanocomposites Compos Sci Technol 68 2008 3259 3264 [23] L.Y. Sun G.L. Warren J.Y.O. Reilly W.N. Everett S.M. Lee D. Davis D. Lagoudas H.J. Sue Mechanical properties of surface-functionalized SWCNT/epoxy composites Carbon 46 2008 320 328 [24] D.G. Yan G.S. Yang A novel approach of in situ grafting polyamide 6 to the surface of multi-walled carbon nanotubes Mater Lett 63 2009 298 300 [25] B. Vigolo V. Mamane F. Valsaque T.N.H. Le J. Thabit J. Ghanbaja L. Aranda Y. Fort E. McRae Evidence of sidewall covalent functionalization of single-walled carbon nanotubes and its advantages for composite processing Carbon 47 2 2009 411 419 [26] G. Mountrichas S. Pispas N. Tagmatarchis Grafting-to approach for the functionalization of carbon nanotubes with polystyrene Mater Sci Eng B 152 2008 40 43 [27] G. Gabriel G. Sauthier J. Fraxedas Preparation and characterisation of single-walled carbon nanotubes functionalised with amines Carbon 44 2006 1891 1897 [28] Lu. Shao Y.P. Bai X. Huang Z.F. Gao L.H. Meng Y.D. Huang J. Ma Multi-walled carbon nanotubes (MWCNTs) functionalized with amino groups by reacting with supercritical ammonia fluids Mater Chem Phys 116 2009 323 326 [29] M.X. Pulikkathara O.V. Kuznetsov V.N. Khabashesku Sidewall covalent functionalization of single wall carbon nanotubes through reactions of fluoronanotubes with urea, guanidine, and thiourea Chem Mater 20 2008 2685 2695 [30] J.R. Conder C.L. Young Physicochemical measurements by gas chromatography 1979 New York John Wiley & Sons [31] N.G. Sahoo Y.G. Jung H.J. Yoo J.W. Cho Effect of functionalized carbon nanotubes on molecular interaction and properties of polyurethane composites Macromol Chem Phys 207 2006 1773 1780 [32] H.C. Kuan C.H.M. Ma W.P. Chang S.M. Yuen H.H. Wu T.M. Lee Synthesis, thermal, mechanical and rheological properties of multiwall carbon nanotube/waterborne polyurethane nanocomposite Compos Sci Technol 65 2005 1703 1710 [33] C.V. Santos A.L.M. Hernandez M.L. Cassou A.A. Castillo V.M. Castano Chemical functionalization of carbon nanotubes through an organosilane Nanotechnology 13 2002 495 498 [34] N.G. Sahoo H.K.F. Cheng J.W. Cai L. Li S.H. Chan J.H. Zhao S.Z. Yu Improvement of mechanical and thermal properties of carbon nanotube composites through nanotube functionalization and processing methods Mater Chem Phys 117 2009 313 320 [35] R. Menzel A. Lee A. Bismarck M.S.P. Shaffer Inverse Gas Chromatography of As-Received and Modified Carbon Nanotubes Langmuir 25 2009 8340 8348 [36] X.L. Zhang D. Yang P. Xu C.C. Wang Q.G. Du Characterizing the surface properties of carbon nanotubes by inverse gas chromatography J Mater Sci 42 2007 7069 7075 [37] Du XD. Study on the mechanism of the synthesis of phenylbiguanide and kinetics of curing epoxy. Master thesis. Wuhan University of Technology. 2007. [38] Y.S. Ko W.C. Forsman T.S. Dziemianowicz Carbon fiber-reinforced composites: effect of fiber surface on polymer properties Polym Eng Sci 22 1982 805 814 [39] P.S. Chua Dynamic mechanical analysis studies of the interphase Polym Compos 8 1987 308 313 [40] J.R. Sarasua J. Pouyet Dynamic mechanical behavior and interphase adhesion of thermoplastic (PEEK PES) short fibre composites J Thermoplast Compos Mater 11 1998 2 21 [41] J.M. Lee S.J. Kim J.W. Kim P.H. Kang Y.C. Nho Y.S. Lee A high resolution XPS study of sidewall functionalized MWCNTs by fluorination J Indus Eng Chem 15 2009 66 71 [42] S. Dong R. Gauvin Application of dynamic mechanical analysis for the study of the interfacial region in carbon fiber/epoxy composite materials Polym Compos 14 1993 414 420