Functionalization of carbon nanotubes for polymer nanocomposites
Carbon nanotubes (CNTs) are excellent candidates for substituting or complementing conventional nanofillers in the fabrication of multifunctional polymer nanocomposites owing to their unique electronic, thermal and mechanical properties. The critical challenge is to uniformly disperse and exfoliate CNTs into polymers to achieve good matrix/nanotube adhesion that maximizes the transfer of the nanotube properties to the polymer matrix. However, many potential applications have been prevented due to the poor dispersion of CNTs and the weak matrix/CNT interface interaction. Functionalization of CNTs provides an effective means to optimize the physical–mechanical properties of polymer–CNT nanocomposites. In this chapter, we summarize various functionalization techniques, including non-covalent and covalent methodologies, with particular emphasis on the dispersion of CNTs and the interface interaction in polymer–CNT nanocomposites.
Carbon nanotubes (CNTs) are all-carbon macromolecules which consist of one or more graphene sheets seamlessly wrapped into cylinder-shaped tubes, corresponding to single-walled carbon nanotubes (SWNTs) or multi-walled carbon nanotubes (MWNTs), respectively.1−3 The pseudo one-dimensional nanostructures endow them with high surface area, large aspect ratio and unusual physicochemical properties such as excellent electrical and thermal conductivities and mechanical strength.4−6 CNTs have been considered as ideal nano-fillers and functional additives in high-performance polymer-based nanocomposites to improve the matrix properties and enable novel functionalities.7−9 Impressive progress has been made in these directions. For example, adding 1 wt% MWNTs in ultra-high molecular weight polyethylene (UHMMPE) film results in an increase of strain energy density and ductility by 150% and 140%, respectively.10 Elastic stiffness and tensile strength are similarly increased by 42% and 25%, respectively, for polystyrene (PS)−MWNT composites containing 1 wt% MWNTs.11 It is found that epoxy composites show an increase of ~ 120% in thermal conductivity at 1 wt% SWNT loading.12 Yu et al.13 recently prepared novel segregated-network CNT–polymer composites by mixing poly(vinyl acetate) (PVAc) emulsion with a gum arabic-stabilized CNTs aqueous solution. It is found that PVAc–CNT composites show a sharp increase in electrical conductivity, while their thermal conductivity and thermal energy are kept insensitive to filler loading. This work makes possible the preparation of polymer nanocomposites with a higher thermoelectric performance and further development of lightweight, low-cost, and non-toxic composites for future thermo-electric applications.
However, nano-sized CNTs are intrinsically insoluble and have a strong tendency to aggregate in solvents and polymers because of the strong van der Waals forces and electrostatic interactions.14 Therefore, realistic applications of polymer–CNT nanocomposites have been significantly less than their anticipated high potential, due to the poor dispersion and compatibility of CNTs in polymers and the weak interface bonding between polymer and CNTs. Dispersion is a fundamental issue for particulate-reinforced polymer composites.15 It is imperative that homogeneous dispersion of CNTs not only transfers the external load to the nanotubes, but also minimizes the stress concentration sites and obtains a more uniform stress distribution.16−18 The interfacial adhesion between CNTs and polymer is also critical to optimize the mechanical and other functional properties of the composites. It is widely recognized that the efficient load transfer depends on the strong interfacial bonding between the host matrix and the fillers used.16−19 Herein lies the critical challenge to transfer the excellent properties of CNTs. To the host polymer via optimal methodologies during processing of the composites. Generally, improved strategies are to functionalize the CNTs with polymers. Frankland et al.20 used molecular simulation to analyze the shear strength of polymer–SWNT interface, and predicted that it would increase by an order of magnitude owing to the chemical bonding between SWNTs and polymer with ~ 0.3% grafting density. Graff et al.21 improved the dispersibility of SWNTs in epoxy matrices by functionalizing the nanotubes with protein concanavalin. Incorporation of 1 wt% fluorinated-SWNTs in poly(ethylene oxide) showed increases of 145% in tensile modulus and 300% in yield strength.22 Recently, O’Connor et al.23 used Kevlar-functionalized MWNTs as additives to reinforce commercially available polyvinyl acetate, and obtained improvements of 80% in tensile strength and 173% in Young’s modulus at very low loading (0.25–0.5 wt%) of MWNTs.
To date, several reviews have been published on functionalization methods for CNTs,24−27 which conceptually fall into two groups: non-covalent and covalent functionalization. The first group involves non-covalent wrapping or adsorption of functional molecules which can be either small organic molecules such as surfactants or various species of macromolecules. The second group involves chemical bonding between functional molecules and CNTs via versatile reactions onto the graphitic surface of the CNTs. From a materials’ chemistry perspective, both methodologies are equally important to the development and applications of materials that contain CNTs, regardless of the functionalization means of attaching the chemical handles to CNTs. Previous reviews have mainly focused on the methodologies to functionalize CNTs with small organic molecules and polymers.24−27 However, this chapter is a critical review of the most significant results of polymer–CNT composites containing polymer-functionalized CNTs in the past decade, with special emphases on the dispersion of CNTs, polymer–CNT interface bonding and final properties of the resultant nanocomposites.
CNTs are non-covalently functionalized by polymers employing π-π stacking, and electrostatic and hydrophobic interactions. This type of functionalization can be divided into two methods: in-situ polymerization and physical mixing. The former polymerizes the monomer in the presence of pristine or purified CNTs, and can be carried out in the presence and/or absence of solvent. The latter involves direct mixing of the CNTs with a preformed polymer through solution or melt processing. Periodically functionalized CNTs can be achieved by helical wrapping of general polymers, or by forming nanohybrid shish-kebabs of crystalline polymers. During processing, some additives such as surfactants are sometimes used to assist dispersion of CNTs in polymers.
The general procedures of in-situ polymerization are to disperse pristine CNTs or purified CNTs in monomer or mixture of monomer and solvent, followed by polymerizing the monomer to prepare polymer–CNT composites. Clayton et al.28 prepared poly(methyl methacrylate)–SWNT (PMMA–SWNT) composite films by in-situ free radical polymerization that was triggered by heat, ultraviolet (UV) light and gamma radiation, respectively. The resultant composite films upon evaporation of methylene chloride show enhanced transparency with respect to the melt-blended composites due to the improved dispersion of CNTs. Putz et al.29 showed the clear evidence of cohesive interaction between SWNTs and PMMA chains, leading to an increase in the low-temperature elastic modulus of ~ 10% beyond that of pure PMMA at less than 0.1 wt% SWNTs. Recently, Zhang et al.30 obtained PMMA–MWNT composite fibers by infiltration of MMA into aligned MWNT aerogel fibers at room temperature with subsequent in-situ polymerization at 50°C. They found that PMMA fills the space of MWNTs and exhibits local order. When compared to the control PMMA, PMMA–MWNT fibers with 15 wt% MWNTs exhibit a 16-fold and a 49-fold increase in tensile strength and Young’s modulus, respectively.
In-situ polymerization was also applied to prepare polymer–CNT composites by a step-growth mechanism, except for a commonly known chain reaction such as the above free radical polymerization. For instance, poly(p-phenylene benzobisoxazole) (PBO)–SWNT composites were synthesized in the presence of SWNTs in poly(phosphoric acid) (PPA), using typical PBO polymerization conditions, then PBO–SWNT fibers were spun from their liquid crystalline solutions using dry-jet wet spinning.31 The addition of 10 wt% SWNTs increases the tensile strength of PBO fiber by ~ 50% and reduces shrinkage and high-temperature creep. Zhao et al.32 ultrasonically pre-dispersed MWNTs into the polymerizable master solution followed by in-situ hydrolytic polymerization of the caprolactam monomer to achieve the homogeneous dispersion of MWNTs in polyamide 6 (PA6) matrices. They found that adding 0.5 wt% MWNTs has little effect on the molecular weight of PA6, and the tensile strength and storage modulus of PA6–MWNT composites are slightly improved, however, MWNTs clearly influence the crystallization and glass transition behavior.
In some cases, acid oxidation or organic modification has been used to chemically pretreat CNTs before in-situ polymerization of the monomer. Velasco-Santos et al.33 prepared PMMA nanocomposites by in-situ polymerization of MMA in the presence of pristine MWNTs (p-MWNTs) and functionalized-MWNTs (f-MWNTs) containing carboxylic groups, respectively. Both Raman and infrared spectra prove the interaction between PMMA and f-MWNT, which effectively improves the mechanical load transfer with respect to PMMA-p-MWNT. The storage modulus of PMMA − p-MWNT with only 1 wt% f-MWNTs is increased by an outstanding 1,135% at 90°C and the glass transition temperature (Tg) is exceptionally raised by ~ 40°C. In evaluating the force required to pull out each nanotube from the polymer matrix, Barber et al.34 found that the interfacial strength between MWNTs and matrix increases dramatically when the surface of MWNTs is chemically modified. Figure 3.1 shows a transition from pull-out to fracture at a critical embedded length as an individual nanotube is pulled from the polymer matrix. This indicates that a smaller critical length and a larger pull-out force are required when using the chemically modified MWNTs instead of the unmodified tubes. These observations imply that the ability of load transfer of CNTs can be significantly enhanced by using pre-modified rather than pristine CNTs in the preparation of polymer–CNT composites.
3.1 A transition from pull-out to fracture occurs at a critical nanotube embedded length, with chemically modified nanotubes showing a smaller critical length than the unmodified nanotubes. (Reproduced with permission from ref. 34. Copyright 2006 Wiley-VCH Verlag GmbH & Co. KGaA.)
Apart from the aforementioned work, surfactant-assisted processing of CNTs is also used to improve dispersion and interfacial bonding of CNTs with polymer, where surfactants are the third component functioning as dispersants. In general, surfactant-assisted dispersion of CNTs is conducted in an aqueous medium with the aid of ultrasonication. Ultrasonication action offers mechanical energy to weaken the van der Waals forces between CNT bundles to exfoliate CNTs into small ropes or even individual CNTs, while the surfactant organic molecules are physically absorbed onto the external surface of the CNTs, resulting in the formation of a stable CNT colloidal system induced by electrostatic and/or steric repulsion.35−38 Once monomers are added to a homogeneous CNT colloid stabilized by surfactant molecules under ultrasonication, then in-situ polymerization (so-called emulsion polymerization) is carried out to achieve an emulsion-like solution as a result of a mixture of CNTs and polymer particles. Finally, the resultant colloidal solution is dried (prior to demulsifying if needed), to obtain a composite in a solid state, where the dispersion and debundling of CNTs are expected to remain in the polymer matrix. Barraza et al.39 prepared thermo-plastic PS–SWNT and elastomeric poly(styrene-co-isoprene) (PSI)–SWNT composites in the presence of cetyltrimethylammonium bromide (CTAB) as a cationic surfactant. They found that the adsorbed polymer layers on SWNTs are conducive to the dispersion of SWNTs. Raman spectra of PS–SWNT and PSI–SWNT hybrids suggest that the strong interaction between SWNTs and the matrix contributes to a reduced vibrational freedom of polymer chains. Xia et al.40 developed a novel ultrasonically initiated emulsion polymerization approach to prepare PMMA–MWNT and poly(n-butyl acrylate) (PBA)–MWNT composites without any chemical initiator. Based on a combination effect of sonication, such as dispersion, pulverizing, activation and initiation, the aggregation and entanglement of MWNTs in aqueous solution can be broken down, consequently resulting in a uniform dispersion of MWNTs in both PMMA and PBA matrices. Further, incorporation of PBA-encapsulated MWNTs into a nylon 6 matrix effectively improves the interface adhesion; the yield strength and Young’s modulus increase by about 30% and 35% at only 1 wt% loading, respectively.
Some researchers believe that CNTs can partially participate in polymerization to form a strong chemical bonding between CNTs and the matrix. Some suggested that π-bonds in CNTs are initiated by AIBN to generate radicals on the surface of CNTs followed by growing the polymer chains.41−43 Others proposed that some polymer chains may be covalently attached to CNTs by an additional mechanism of macromolecular radicals,44−46 where the CNTs act as terminal agents. In most cases, the formation of chemical linkage between CNTs and polymer is difficult in such polymerization systems. But it is widely recognized that in-situ polymerization is not only a significantly important method that allows better dispersion of CNTs in a polymer matrix, but is a very convenient processing technique to produce polymer nanocomposites with high CNTs loadings. This methodology has shown very good miscibility with most monomers, especially for the preparation of insoluble and thermally unstable polymers.16
Unlike in-situ polymerization of monomer in the presence of CNTs, polymer wrapping directly mixes the pre-synthesized polymer and CNTs by mechanical blending. In this regard, the dispersion and debundling of CNTs in the polymer matrix are achieved by the non-covalent interaction between them such as van der Waals forces, hydrophobic and electrostatic interactions as well as special π-π stacking and CH-π interactions. Researchers used polymer wrapping to fabricate polymer–CNT composites through two methodologies, including solution blending and melt compounding. Solution processing is the most common method that is generally performed by dispersing CNTs in a suitable solvent or polymer solution, followed by mixing and then precipitating from their non-solvent or evaporating the solvent to obtain the polymer–CNT composites. In solution processing, high-power ultrasonication in combination with mechanical agitation is often required to accelerate the deaggregation and exfoliation of nanotube bundles since it is difficult to directly disperse CNTs in solvent using simple stirring. Coleman et al.47 prepared a suspension of poly(vinyl alcohol) (PVA)–MWNT by ultrasonication, and cast them onto a polished Teflon disk followed by evaporation of the solvent at 60°C in an oven. The final free-standing PVA–MWNT films show increases in Young’s modulus, tensile strength and toughness by 3.7, 4.3 and 1.7 times at less than 1 wt% nanotubes compared to the parent PVA. It is believed that the PVA-nanotube interaction is completely non-covalent as a result of the van der Waals forces. A 78% increase in tensile modulus for PVA–SWNT films was reported by Liu et al.48 when adding 0.8 wt% SWNTs modified by multi-hydroxyl groups to the PVA matrix by simple mixing. Recently, Wang et al.49 prepared polyelectrolyte–MWNT core-shell nanocomposites by alternately assembling poly(aniline-co-o-anisidine) and poly(sodium 4-styrenesulfonate) ionomers to the carboxylated–MWNTs. The wrapped polyelectrolyte improves the dispersibility and stability of MWNTs in solution. The conductivity of the nanocomposites increases to 4.2 S/cm by 3 orders of magnitude from 0.004 S/cm of pure polyelectrolyte. These results indicate that the use of surface-modified CNTs can increase the capability of polymer wrapping to enhance the dispersion and interfacial bonding between the components.
CNTs are well known to be π-conjugated, quasi-one-dimensional structures consisting of rolled-up graphene sheets. Hence, conjugated polymer–CNT composites are desirable due to the relatively strong π-π non-covalent interaction between them. Tang et al.50 first reported that CNTs were helically wrapped by poly(phenylacetylene) (PPA) chains to endow them with good solubility in organic solvents. Poly(phenylene vinylene) (PPV) derivates that consist of rigid aromatic phenyl and flexible aliphatic vinyl groups in their main chains, acting as typical conjugated polymers, are more widely studied owing to the great promise of their CNT composites in photoelectric applications.51−54 Curran et al.51 prepared poly(m-phenylenevinylene-co-2,5-dioctoxy-p-phenylenevinylene) (PmPV)–MWNT composites by mixing MWNTs with PmPV in toluene. It is found that the electrical conductivity of PmPV is increased by up to eight orders of magnitude, and MWNTs act as nanometric heat sinks to prevent the build-up of large thermal effects. Considering the luminescent property of PmPV, Star et al.52 used PmPV–CNT composites to fabricate an optoelectronic memory device. The resultant device reveals a photogating effect on charge transport which can rectify or amplify the current flow through the tubes.53 Recently, Tchoul et al.54 prepared PS–SWNT composites by rapid evaporation of chloroform solutions of PmPV-functionalized SWNTs and polystyrene. It is found that PmPV is a good stabilizing agent for dispersion of SWNTs, and the resulting composites show a higher electrical conductivity and lower percolation threshold than those of PS-containing acid-oxidized SWNTs in the absence of PmPV.
Other polymers such as non-conjugated poly(allylamine),55 poly(L-lysine),56 Nafion57 and polyethylenimine,58 poly(o-toluidine),59 pyrene-labeled hydroxypropyl cellulose60 and poly(thiophene)s,61 were successfully applied to functionalize CNTs by polymer wrapping. These results show that such structural polymers can afford well-dispersed CNT solutions, thereby their composites, even with an extremely low polymer concentration in some cases, result in improved mechanical and electrical properties. Recent theoretical simulation62 and experimental observations63, 64 suggest that wrapping or coating is a ‘general phenomenon’ occurring between polymers and CNTs regardless of the preparation methods of solution processing and melt-shear mixing.
Melt processing is an alternative that is especially beneficial to insoluble polymers, and it is mainly conducted with high shear forces at a given temperature that is above the melting point or glass transition temperature of the polymer used. Liu et al.65, 66 prepared MWNT-reinforced PA6 nanocomposites by the simple melt-compounding method. It is found that the elastic modulus and yield strength of PA6–MWNT (2 wt%) composites are increased greatly with respect to those of pure PA6 by about 214% and 162%, respectively, which are due to the uniform dispersion of MWNTs throughout the PA6 matrix and the strong interfacial bonding between nanotubes and matrix. Interestingly, only α-form crystals are observed for PA6–MWNT composites quite different from those observed in PA6–clay nanocomposites. However, in most cases, melt processing inevitably encounters such difficulties as poor dispersibility, low CNT loading in the polymer matrix due to the high melting viscosities, and unexpected polymer degradation, although it is most compatible with current industrial processing techniques.
Cadek et al.67 reported that Young’s modulus of PVA is increased by a factor of 2 at less than 1 wt% MWNTs loading when crystalline polymers wrap around the nanotubes surface. Using non-crystalline polymers towards similar MWNTs, however, results in much lower levels of reinforcement. This finding clearly implies that well-ordered polymer coating is extremely important to effective load transfer of composites. Polymers with high affinity for physisorption or wrapping on CNTs are in principle good candidates to promote enhanced interfacial bonding between the matrix and CNTs. This non-covalent interaction also retains the structural integrity of CNTs.68, 69 Therefore, using crystalline polymers to functionalize CNTs via CNT-induced crystallization seems an ideal method since they possess better mechanical properties than those of non-crystalline polymers. In this regard, Li et al.70 reported some significant results. They first used a polymer solution crystallization technique to achieve periodically decorated CNTs.71−74 In detail, CNTs were first dispersed in a given solvent under ultrasonication followed by mixing with a solution of polymer, and the resulting polymer–CNT dilute solution was then brought to the crystallization temperature (Tc), at which point the polymer crystallizes on the surface of the nanotubes to generate polymer–CNT nanohybrids. Both PE and Nylon 66 were successfully applied to functionalize SWNTs and MWNTs and even vapor-grown carbon nanofibers (CNFs). It is found that CNTs are periodically decorated with polymer lamellar crystals to form nano-hybrid shish-kebab (NHSK) structures (Fig. 3.2). The NHSK structural parameters can be readily adjusted by controlling crystallization conditions; the periodicity of polymer lamellae varies from 20 to 150 nm and the kebabs are ~ 5−10 nm thick along CNTs with a lateral size of ~ 20 nm to micrometers.72 This provides a unique opportunity to control the functionalization degree of CNTs. It is of great interest that the formation conditions of NHSK depend on the structures of CNTs used, probably offering an opportunity for separation of CNTs. Accordingly, the NHSK formation is attributed to a heterogeneous nucleation mechanism of size-dependent soft epitaxy. Such functionalized–MWNTs can be easily dispersed in solution, and can also be used as precursors to prepare polymer–CNT nanocomposites with excellent dispersion.
3.2 TEM images of (a) PE–MWNT-10, (b) PE–MWNT-25 and (c) PE/CNF NHSK structures produced by crystallization of PE in p-xylene for 0.5 h at 103, 97, and 97°C. PE, MWNT, and CNF concentrations are 0.01, 0.002, and 0.002 wt%, respectively. (Reprinted with permission from ref. 72. Copyright 2006 American Chemical Society.)
Periodically functionalized CNTs is an attractive field of research, similar PENHSK structures are achieved by combining the polymerization-filling technique with tandem copolymerization catalysis of ethylene in the presence of metallocene catalyst-treated MWNTs.75, 76 Recently, Fu et al.77−79 reported that fine NHSK superstructures could be directly achieved in injection-molded bars of high-density PE (HDPE) or linear low-density PE (LLDPE)–MWNT composites prepared by a so-called dynamic packing injection molding (DPIM) technique. It is noted that the NHSK structures can confer significant mechanical reinforcement. The oriented HDPE–MWNT composites have their tensile strength increased by 150 and 270% and Young’s modulus enhanced by 130 and 180%, with respect to the oriented pure HDPE and the isotropic composite at the same MWNTs loading (5 wt%), respectively. Both tensile strength and elongation-at-break of LLDPE are also improved by adding LLDPE-decorated MWNTs. These studies may open up a convenient gateway to achieve directly well-defined NHSK structures in simple injection-molding for reinforcement of polymer composites.
Supercritical (SC) fluids have a promising future in materials synthesis and processing due to such excellent properties as near-zero surface tension, high diffusivity, low viscosity and ease of control, and hence they have been widely applied to prepare polymer-based materials.80 Among the supercritical fluids, supercritical carbon dioxide (SC CO2) is the most popular due to its non-flammability, non-toxicity, low cost and high compatibility with many organic solvents. Recently, the supercritical CO2 technique has shown intriguing advantages in the synthesis of CNT-based composites with metal, metal oxide and polymer.81In addition to polymer solution crystallization and DPIM methods as summarized in Section 3.2.3, SC-CO2-induced polymer epitaxy on CNTs has been used to form the NHSK structures. Xu et al.82, 83 employed a SC-CO2 antisolvent-induced polymer epitaxy (SAIPE) method to aid PE epitaxy growth along CNTs form an NHSK structure under different experimental conditions. They found that the periodicity of PE lamellae, lateral size and kebabs’ thickness of NHSK varied by the change of solvent, concentrations of PE and CNTs, and SC-CO2 pressure. Functionalized SWNTs with PVA and PEG are well dispersed in aqueous solutions and have potential applications in biologically relevant systems.83
Dai et al.84 first dispersed MWNTs into an ethanol solution of 2,4-hexadiyne-1,6-diol (HDiD), and then introduced CO2 up to a desired pressure, followed by heating the mixture up to 200°C for polymerization of HDiD to give poly(HDiD)–MWNT composites. It is found that the solvent power of ethanol for HDiD is reduced due to the anti-solvent effect of CO2, resulting in the adsorption of HDiD on the outer surface of CNTs. In contrast, the CO2–ethanol mixture under the experimental conditions has near-zero surface tension and larger diffusivity, leading to the diffusion of HDiD molecules into the interior cavities of CNTs. Thus, it can be deduced that MWNTs were synchronously filled and coated by poly(HDiD) to form the multi-component composites which possess optical properties originated from poly(HDiD). Wang et al.85 directly coated MWNTs with fluorinated graft poly(methyl vinyl ether-alt-maleic anhydride) in SC–CO2 under 100–170 bar at 40°C, resulting in the formation of quasi one-dimensional nano-structures with conducting cores and insulating surfaces with an average thickness of ~ 2 nm. Later, Yue et al.86 prepared PMMA–SWNT composites in SC–CO2 by copolymerizing methyl methacrylate (MMA) monomers with acrylated-modified SWNTs. It is shown that SC–CO2 increases the diffusivity of monomer, and facilitates the growth of tethered PMMA chains near the entanglement area and interstitial space of the nanotube bundles, yielding partial debundling and disentanglement of SWNT bundles and improved dispersion in PMMA.
Covalent grafting of polymer chains to CNTs can be performed by either ‘grafting-from’ or ‘grafting-to’ strategies.87−89 The ‘grafting-from’ technique involves chemical attachment of functional groups to CNTs to form the CNT-supported macro-initiators, followed by surface-initiated polymerization of monomers to yield the tethered polymer chains. In general, initiators are linked to CNTs by employing the derivation reaction of CNT-bound carboxylic acid groups or direct addition of reagents to the surface of CNTs. By comparison, ‘grafting-to’ is concerned with immobilization of preformed functional polymers onto the surface of CNTs that are pristine, purified, oxidized or pre-functionalized.
Xie et al.90 attached allylamine functional molecules to MWNTs by acylation-amidation to generate allyl-modified MWNTs, followed by conventional free radical polymerization of styrene to yield PS-grafted MWNTs (MWNT-g-PS) with a hairy rod nanostructure. It is found that 1 of every 100 carbon atoms in MWNTs is grafted by PS, forming a uniform thin shell (ca. 8~10 nm) on the outer wall. MWNT-g-PS shows enhanced thermal stability and Tg as compared to pure PS, and MWNTs are uniformly dispersed in the polymer matrix. Also, they prepared water-soluble poly(sodium 4-styrenesulfonate-co-acrylic acid)-grafted MWNTs (MWNT-g-PSA) by an analogous method.91 The MWNT-g-PSA modified glassy carbon electrode was found to show strong electron transfer capability, high electrochemical activity and catalytic ability, which could be used in sensitive, selective, rapid and simultaneous monitoring of biomolecules. Further, MWNT-g-PSA incorporated into the mixed ionic membrane of PSA and PVA was used as an electromechanical actuator.92 As expected, MWNT-g-PSA is homogeneously dispersed in the PSA–PVA membrane (Fig. 3.3 (a)), thereby the electric conductivity increases by several orders of magnitude to 1.44×10‒4 from 7.53×10‒13 S/cm of the PSA−PVA membrane. MWNT−g−PSA strengthens the PSA−PVA membrane from 14.4 to 32.7 MPa with a loading up to 20 wt%, and it also enhances the fracture toughness by 34.8%. A steady relationship of time-dependent displacement under a repeated square wave input is clearly visible (Fig. 3.3 (b)). A similar mechanism was employed by Shim et al.93, 94 to prepare MWNT-g-PS in the presence of AIBN as a free radical initiator. Recently, a novel composite of MWNTs and molecularly imprinted polymers (MIPs) was also prepared by copolymerization of methacrylic acid and trimethylolpropane trimethacrylate in the presence of dopamine as a template molecule.95 It is found that the vinyl groups modified on the surface of MWNTs are a key factor for forming the composite of MIPs and MWNTs. Besides, Li et al.96 chemically attached α-alkene groups to SWNTs, and then copolymerized them with ethylene to produce SWNT-g-PE composites in the presence of a metallocene catalyst. The interaction between two components and dispersion of SWNTs in PE matrix are clearly improved, leading to expected results that the thermal stability and mechanical properties (tensile strength, modulus and elongation-at-break) of SWNT-g-PE are higher than those of PE and PE–SWNTs mixture with the same loading of SWNTs.
3.3 (a) SEM image of cryo-fractured MWNT-g-PSA reinforced PSA–PVA membranes at 20 wt% loading. (b) Time-dependent displacement of composite actuator (20 wt% MWNT-g-PSA) under repeated stimulation at an applied square-wave electric potential of (±) 1.5 V with a frequency of 0.25 Hz. (Reprinted with permission from ref. 92. Copyright 2009 American Chemical Society.)
Of all the CLRP techniques, atom transfer radical polymerization (ATRP) was first applied to synthesize polymer-grafted CNTs via a ‘grafting-from’ approach. The key is to introduce ATRP initiators onto the surface of CNTs. Yao et al.97 initially functionalized SWNTs by 1,3-dipolar cycloaddition reaction for attaching phenols, which further reacted with 2-bromoisobutyryl bromide to achieve the SWNT-supported ATRP initiators. These initiators were active in polymerization of methyl methacrylate and tert-butyl acrylate from the surface of SWNTs. Later, Baskaran et al.,98 Kong et al.99 and Qin et al.100 simultaneously and independently reported their results on polymer-grafted CNTs using ATRP, in which either SWNTs or MWNTs were functionalized by ATRP initiators via subsequent acylation-esterification prior to polymer grafting. Homo, block, and copolymer brushes were chemically linked to CNTs at levels up to 90 wt% from the TGA results. The average thickness (or the amount) of polymer grafted onto CNTs varies with the feed ratio of monomer to the CNT-supported macro-initiators and reaction time. Therefore, the ATRP process from the surface of CNTs is still living. As is well known, dispersion of CNTs into solvents before functionalization is difficult even after ultrasonication; however, polymer-grafted CNTs are soluble in given solvents which are compatible with the parent polymers. Moreover, all functionalized-CNTs show enhanced Tg and Td (decomposition temperature) compared to pure polymers due to the high thermal conductivity of CNTs and the constraint effect of grafted polymer chains. Wang et al.101 used ATRP to synthesize MWNT-g-PMMA including 10 wt% MWNTs, and then blended with poly(styrene-co-acrylonitrile) (SAN) by solution processing. Upon addition of MWNT-g-PMMA with an effective MWNT loading of 1 wt%, the storage modulus at 40°C, Young’s modulus, tensile strength, ultimate strain, and toughness of SAN–MWNT-g-PMMA composites are increased by up to 90, 51, 99, 184 and 614%, respectively, compared to pristine SAN. Such simultaneous improvements of a polymer using rigid fillers are rarely observed. These results are due to the uniform dispersion of CNTs and strong CNT–matrix interaction since PMMA is miscible with SAN. Shi et al.102 reported ATRP of n-butyl methacrylate (BMA) from MWNTs to produce MWNT-g-PBMA, in which the relative amount of PBMA was determined by TGA to be ~ 16 wt%. Incorporating MWNT-g-PBMA (0.2 wt% MWNTs) in a PVC matrix leads to increases in storage modulus (at 35°C), Young’s modulus, yield stress, tensile strength, ultimate strain and toughness by 83, 40, 74, 84, 38 and 145%, respectively, compared to PVC. The miscibility between PVC and grafted-PBMA enables the homogeneous dispersion of MWNTs, and also improves the efficiency of load transfer from matrix to nanotubes, as a result of the enhanced mechanical performance.
Polymer chains can grow from the dithioester-modified CNTs in a controlled process by a reversible addition-fragmentation chain transfer (RAFT) mechanism. Similar to ATRP, RAFT agents are anchored onto the surface of CNTs by derivation reaction of the CNT-bound carboxylic groups before surface-initiation polymerization of monomer. Using this method, various polymers such as PS,103, 104 PMMA,105 water-soluble poly(N-(2-hydroxypropyl)-methacrylamide)106 and poly(2-hydroxyethyl methacrylate) (PHEMA)107 and poly(acryl amide) (PAM),108 thermoresponsive poly(N-isopropylacrylamide) (PNIPAM)109, 110 and their block polymers have been successfully attached to the surface of CNTs, resulting in the formation of core-shell nanostructures, with polymer as the brush-like or hairy shell, and CNTs as the hard backbone. The shell thickness or content of grafted polymers on CNTs can be well controlled by changing the reaction parameters of RAFT polymerization. The covalent bonding of polymers to CNTs dramatically improves the solubility and stability of MWNTs in solvents. It is noted that the PNIPAM shell of MWNT-g-PNIPAM shows reversible response to temperature as a result of switching between hydrophilicity and hydrophobicity, which should have potential applications in smart sensors and probes.109 Moreover, using a suitable content of MWNT-g-PS as lubricant additives can effectively improve the anti-wear property of the base lubricant, because the long PS chains prevent MWNTs from flocculating in the base lubricant due to their improved dispersibility.104
In addition to ATRP and RAFT, another important CLRP technique of nitroxide-mediated polymerization (NMP) has been applied to functionalize CNTs with various polymers. Zhao et al.111 prepared poly(4-vinylpyridine)-grafted MWNTs (MWNT-g-P4VP) and poly(sodium 4-styrenesulfonate)-grafted MWNTs (MWNT-g-PSS) by surface-mediated NMP of monomers from the surface of MWNTs that were initially functionalized by TEMPO (2,2,6,6-tetramethyl-piperidinyl-1-oxyl) functional groups. It is found that MWNT-g-P4VP shows good solubility in acidic aqueous solutions, but tends to aggregate in neutral or basic solutions, due to the presence of basic pyridine units. By comparison, MWNT-g-PSS is stable in aqueous solutions throughout the pH range 1 to 14. They also prepared MWNT-g-(PS -b-P4VP) by further polymerization of MWNT-g-PS with 4-vinylpyridine.112 Changing the grafted polymer shell from PS to PS-b-P4VP results in an obvious change of solubility. In the work of Dehonor et al.,113 PS brushes bonded covalently to N-doped MWNTs were also synthesized by a ‘grafting-from’ route using NMP. Datsyuk et al.114 developed a two-step route to prepare block polymers grafted double-walled carbon nanotubes (DWNTs) without chemical pre-treatment. PAA or PS short chains were first polymerized in the presence of NMP initiators to form stable nitroxide radicals around DWNTs, and then these active sites reinitiated polymerization of methyl acrylate (MA) to yield amphiphilic DWNT-g-(PAA-b-PMA) and DWNT-g-(PS-b-PMA), respectively. These grafted block copolymers promote the dispersion of CNTs and improve their surface bonding with various polymer matrices.
As reported above, surface-mediated CLRP techniques not only control the amount and thickness of versatile grafted-polymers on CNTs, but also produce block copolymers with multifunctional components. Such functionalized-CNTs with tailor-made structures and properties are important for improving the dispersibility, compatibility and miscibility of CNTs in polymer matrices and their interface bonding to enhance the effectiveness of load transfer from the matrix to CNTs. Moreover, these cases may open up a path to fabricating novel CNT-based devices and sensors.
Hydroxyl-functionalized CNTs have been widely used as initiating species for ROP of ε-caprolactone115−117 and L-lactide118−120 monomers to achieve biodegradable polymers grafted CNTs and even hyperbranched polymers grafted CNTs.121 The amount of grafted polymers can be controlled by adjusting the feed ratio of monomer to the CNT-supported macro-initiators.115, 119, 121 The grafted poly(ε-caprolactone) (PCL) retains the biodegradability of conventional PCL that can be enzymatically biodegraded, and CNTs retain their tubelike morphologies.115 By adding 1 wt% MWNT-g-PLLA (effective MWNTs: ~ 0.7 wt%), the tensile modulus of PLLA increases from 2,463 to 4,110 MPa, and the strength from 56.4 to 85.6 MPa, resulting in 91 and 52% improvements, respectively.118 The Flory-Huggins interaction parameter (B) is estimated to be 96.6 cal cm−3 for PLLA–MWNT-g-PLLA composite, which suggests that MWNTs are highly compatible with PLLA after functionalization.118 Further, the activation energy of PLLA–MWNT-g−PLLA is higher than that of PLLA–MWNT composites, leading to the higher thermal stability for the former.119 Also, all mechanical properties of PLLA–MWNT-g-PLLA composites are superior to those of PLLA–MWNT composites because of the better dispersion of MWNT-g-PLLA in the PLLA matrix in conjunction with stronger interface adhesion and higher miscibility between MWNT-g-PLLA and PLLA compared to MWNTs and PLLA.118, 119 Figure 3.4 compares SEM images of PLLA–MWNT-g-PLLA and PLLA–MWNT composites containing similar loading (2 wt%) of MWNTs. Most MWNTs aggregate in PLLA–MWNT composites (Fig. 3.4 (a)); however, after functionalization, MWNTs are uniformly embedded in PLLA matrix for PLLA–MWNT-g-PLLA (Fig. 3.4 (b)). These results indicate the potential application of biocompatible polymers functionalized CNTs in bio-nanomaterials, biomedicine, artificial organs and bones.115−120
3.4 SEM images of (a) PLLA–MWNT and (b) PLLA–MWNT-g-PLLA composites at the same MWNT loading of 2 wt%. (Reprinted with permission from ref. 119. Copyright 2007 Elsevier.).
Covalent functionalization of CNTs with polyamide 6 (PA 6) was accomplished by anionic ROP of ε-caprolactam in the presence of ε-caprolactam grafted CNTs as initiators and sodium ε-caprolactamates as catalysts.122, 123 It is found that the CNT-initiated anionic ROP shows high efficiency under low reaction temperature (170°C) and short reaction time (6 h), with the final products having a limited amount of grafted-PA 6 (65 wt%). The functionalized nanotubes are soluble in formic acid and m-cresol, which may prove to be new avenues for homogeneous dispersion of CNTs in PA 6 to form high-quality nanocomposite materials.
Epoxy resin as an important thermosetting material has also been applied to fabricate high-performance nanocomposites with CNTs. Zhu et al.124 pretreated SWNTs through acid oxidation and subsequent fluorination to generate carboxyl and fluorine groups on the surface of SWNTs. Epoxy composites containing 1 wt% SWNTs were then processed by dispersing the functionalized nanotubes in dimethyl formamide (DMF) and mixing with epoxy monomer and diamine hardener. SWNTs are observed to be highly dispersed and well embedded in epoxy composites due to the formation of strong covalent bonds with the defect sites of SWNTs, where epoxy ring-opening esterification with carboxyl and amine curing reaction with fluorine groups proceed. Hence, epoxy–SWNT composites with respect to neat epoxy show a 30% increase in modulus and an 18% increase in tensile strength, respectively. Bae et al.125 and Bekyarova et al.126 also found that surface treatment of CNTs with acid oxidation promotes homogeneous dispersion of CNTs in epoxy resin and provides strong interaction with matrices through chemical bonding of ring-opening esterification, and thus enhances the overall mechanical properties. Tseng et al.127 prepared covalent-integrated epoxy composites using plasma-treated CNTs. Integration of CNTs into a thermosetting epoxy matrix was also achieved by coupling nanotubes with silanes containing epoxy groups and then observing the subsequent reaction with the epoxy monomer and diamine hardener.128 It is found that modifying the CNT surface with siloxanes significantly improves the dispersion of CNTs in epoxy matrix and enhances their interfacial adhesion due to covalent linkage CNTs with epoxy through ring-opening. As a result, composites containing siloxane-CNTs yield better thermal stability, higher flexural modulus and improved strength and fracture resistance than those containing untreated CNTs.
A fully integrated SWNT–epoxy composite was also prepared via chemically attaching amino groups to SWNTs followed by reacting with epoxy resin and curing agent.129 In the SWNT-reinforced epoxy polymer composites, SWNTs are covalently integrated into the epoxy matrix through ring-opening reaction of epoxy groups during epoxy curing, and hence become part of the cross-linked structures rather than separate components. The resulting composites thus show dramatic improvement in their mechanical properties better than those of pure epoxy resin. Young’s modulus increases from 2,026 to 2,650 MPa at just 1 wt% functionalized SWNTs, giving more than 30% improvement. At the same loading, the tensile strength is 25% higher than neat epoxy when compared to untreated SWNT–epoxies. A recent result reported by Liu et al.130 also confirms that epoxy composites containing covalently modified MWNTs possess greater storage modulus relative to the noncovalent system.
Different to the chain reaction mechanism of conventional or living radical polymerization, condensation polymerization belongs in general to the step-growth reaction. This technique has been mainly used to prepare polyesters, polyamides and polyurethanes functionalized CNTs so far. Xie et al.131, 132 developed an in-situ polycondensation method to functionalize MWNTs with hyperbranched poly(urea-urethane) (HPU) and linear photoresponsive polyurethane with azo pendant groups. As shown in Fig. 3.5, MWNTs were pretreated with acyl chloride followed by reacting with diethanolamine to form multiple hydroxyl groups, which were further used as active species for reacting diisocyanates with diethanolamine or diols containing azo-benzene, resulting in the formation of grafted hyperbranched polymers or linear polymers from the surface of MWNTs. The amount or thickness of the polymer shell is adjustable by changing the feed ratio of monomers to MWNTs. It is noted that the modified MWNTs containing azobenzene show reversible photoisomerism behavior in solutions; and the responsive rate constant can be effectively controlled by adjusting the main-chain flexibility of grafted polyurethanes.133 This may play a crucial role in developing novel high-performance optic and photonic nano-devices. A similar methodology was used to attach polyurea, polyurethane134, 135 and polycaprolactone polyurethane136 onto CNTs terminated with hydroxyl or amino groups. As a result, SWNT-g-PU improves the dispersion of SWNTs in PU matrix and strengthens the interfacial interaction between the matrix and SWNTs. Consequently, SWNT-g-PU has a remarkable reinforcing effect over pristine SWNTs. Also, incorporating SWNT-g-PU into the PU matrix improves the phase separation and the degree of crystallization of PCL soft segment.136
3.5 Schematic representation of synthesizing hyperbranched poly(urea-urethane)s-grafted MWNTs by surface-mediated in-situ polycondensation. (Reprinted with permission from ref 132. Copyright 2007 American Chemical Society.)
Recently, Zhou et al.137 first reacted oligo-hydroxyamide (oHA) with acyl chloride modified MWNTs to obtain oligo-hydroxyamide (oHA)-grafted MWNTs (MWNT-oHA), which has a remarkable solubility in polar solvents and a good thermal stability. MWNT-oHA was then covalently incorporated with poly(p-phenylene benzobisoxazole) (PBO) via in-situ condensation reaction. Using a dry-jet wet-spinning technique, continuous MWNT-PBO composite fibers were finally fabricated with different MWNT loadings. As expected, MWNT-PBO fibers relative to pure PBO fibers show overall increases in tensile modulus and strength, thermal stability and electrical conductivity, due to their excellent dispersion and high alignment of MWNTs in PBO matrix and their enhanced interfacial interaction. Baek et al.138−140 have developed a direct Friedel-Crafts acylation for functionalization of CNTs and carbon nanofibers (CNF) in a polyphosphoric acid (PPA)–phosphorous pentoxide (P2O5) medium. Various benzoic acid monomers have been selected to prepare linear or hyperbranced poly(ether-ketone)s (PEKs) grafted MWNTs (MWNT-g-PEK). This method shows a strong driving force to build extremely high-molecular weight PEKs directly from the carboxylic groups instead of the expensive and corrosive acid chloride groups. Accordingly, the proposed strategy could be an ideal chemical functionalization method for CNTs for the fabrication of polymer nanocomposites.
As shown in the above sections, strong interfacial bonding and homogeneous dispersion are critical to take full advantage of the excellent properties of CNTs for the reinforcement of polymer matrices. Hence, researchers attempt at all times to find versatile suitable approaches to functionalize CNTs. Liu et al.141 prepared a styrenic polymer composite containing well-dispersed MWNTs by anionic polymerization of a nanotube-bound p-methylstyrene. Initially, MWNTs were first modified by ligand-exchange reaction of ferrocene, and then monolithiated by tertbutyl-lithium and followed by termination with p-chloromethylstyrene (pMS). The pMS-terminated species on MWNTs were finally functionalized with living polystyryl-lithium anions via anionic polymerization, forming MWNT-g-PS that is soluble in common organic solvents. Based on the anionic mechanism, PS142 and PMMA143 were also chemically grown from the surface of SWNTs.
Zhang et al.144 developed an electrografting method for functionalization of CNTs. In their work, SWNTs and N-succinimidyl acrylate (NSA) were predispersed in room-temperature ionic liquids (RTILs) by a combination of grinding and mixing, yielding a black gel for use as the RTIL-supported 3D SWNTs electrode. Once an electrochemical reaction occurred, poly(N-succin-imidyl acrylate)s were in-situ grafted onto SWNTs. Therefore, large quantities of SWNTs were considerably untangled in RTILs so as to greatly increase the effective area of the electrode. It is thought that the application of the RTIL-supported CNTs electrode can realize the homogeneous electrochemical functionalization of SWNTs in large quantities. Tasis et al.145 reported a ceric ion-induced redox radical polymerization technique for the preparation of watersoluble polyacrylamide grafted CNTs. In this work, carbinol groups, thermally generated from organic peroxides and methanol, were covalently attached to CNTs. Then, these carbinol groups provided initiating sites for surface-mediated polymerization of acrylamide. Recently, Priftis et al.146 functionalized CNTs with PLLA and PCL by a two-step process, whereby CNTs were first modified with titanium alkoxide catalysts through a Diels-Alder cycloaddition, followed by surface initiated titanium-mediated coordination polymerizations of L-lactide and ε-caprolactone monomers.
Ni and Sinnott147 used classical molecular dynamics simulations to predict a pathway to the chemical functionalization of CNTs by coupling either radicals or their fragments to the nanotube walls. Recently, density functional theory calculations were also used by Galano to model the potential ability of CNTs to act as free-radical scavengers.148 Experimentally, Lou et al.149 demonstrated the coupling reaction of sp2−hybridized carbons of CNTs with TEMPO-end-capped poly(2-vinylpyridine) (P2VP) that was pre-synthesized by a nitroxide-mediated polymerization (NMP) method. Figure 3.6 shows the direct covalent attachment of macro-molecular radicals, released by thermolysis of P2VP chains terminated with TEMPO radical-stabilizing group, to the surface of pristine MWNTs. The resultant MWNT-g-P2VP is easily dispersible in organic solvents and acidic water in contrast to neat MWNTs. P2VP chains form a shell around CNTs and their grafing ratio can be adjusted to some extent in the range of 6~12%. The same approach was used for the grafting of PS, PCL, and PCL-b-PS copolymer onto MWNTs as a result of the addition reaction of the parent polymeric radicals.150 It is found that the grafting density of polymers (12~30%) can easily be controlled by tailoring the molecular weight of polymers prior to functionalization. NMP techniques have been used to synthesize a variety of polymers,151 thus, all these polymers should be grafted onto CNTs by a radical addition mechanism.
3.6 Heating of TEMPO-end-capped P2VP chains causes TEMPO groups to dissociate, resulting in the formation of MWNT-g-P2VP by macromolecular radical coupling. (Reproduced with permission from ref. 149. Copyright 2004 Wiley-VCH Verlag GmbH & Co. KGaA.)
Wu et al.152 found that bromine-terminated PS synthesized by ATRP was also directly reacted with MWNTs under ATRP conditions using CuBr/2,2′-bi-pyridine as catalysts, yielding MWNT-g-PS by a macromolecular radical coupling mechanism. Grafting PMMA onto MWNTs was performed by emulsion polymerization of MMA monomers in the presence of nanotubes in micelles.153 Incorporation of MWNT-g-PMMA (39 wt% PMMA) into commercial PMMA leads to an improved nanotube-matrix bonding and uniform nanotube dispersion in the matrix. The applied tensile load on the PMMA–MWNT-g-PMMA composites is effectively transferred to MWNT-g-PMMA. The storage modulus at 20°C is greatly enhanced by 1,100% from ~ 2.5 GPa for pure PMMA to ~ 31 GPa for the PMMA composites containing 20 wt% MWNT-g-PMMA.
Mcintosh et al.154 prepared PP–SWNT composites by melt compounding with and without benzoyl peroxide (BPO) initiators. During the high shear and high temperature phase of processing, free radicals upon decomposition of BPO can initiate PP chains to generate polymer radicals, which further allow covalent linkage with SWNTs. The resulting fibers made from PP–SWNTs show improved interfacial bonding and some alignment with BPO present serving as a free radical initiator. Accordingly, their mechanical properties increase by 82.9, 89.8, 72.3, and 173.1% in tensile strength, and by 69.2, 99.7, 137.2, and 133.7% in elastic modulus, respectively, corresponding to SWNTs loadings of 2.5, 5, 7.5, and 10 wt%. These mechanical properties are highly superior to those of pure PP fibers and PP–SWNT fibers without BPO under comparable SWNTs loading. It is suggested that free radicals scavenge protons from the PP matrix, providing active sites for direct covalent bonding of SWNTs and their ropes. In contrast to conventional matrix and reinforcing filler type of composite, such functionalized composites could be viewed as a material consisting of macromolecules, which are supported by a strong interface between SWNTs and surrounding matrix via covalent linkage. This type of fully integrated composites with CNTs will result in more efficient load transfer and provide better mechanical reinforcement.
Cycloaddition of azides by thermolysis to strained-double-bonds of fullerenes has been widely used in solution to synthesize polymeric fullerenes.155 Since CNTs contain a similar double bond in their backbone, the same principle should be applicable for functionalization of CNTs. In recent years, nitrene cycloaddition of azides has been effectively employed to functionalize CNTs by thermolysis in solution156 and photolysis under UV irradiation.157 Qin et al.158 reported the grafting of PS chains to SWNTs by a two-step process, where PS-Br obtained by ATRP was converted to PS-N3 with sodium azide in DMF, followed by performing cycloaddition of PS-N3 to SWNTs in 1,2-dichlorobenzene (DCB) at 130°C in a nitrogen atmosphere for 60 h. The resultant SWNT-g-PS contains 85 wt% PS and 15 wt% SWNTs, corresponding to 1 PS chain per 48 SWNT carbons. SWNT-g-PS can be diffused into individual SWNT and very small bundles in solvents. Li et al.159 employed disubstituted polyacetylenes that were terminated by halogen atoms at the ends of their alkyl pendants to react with sodium azide so as to yield azido-functionalized polymers. Polyacetylene chains were then attached to the SWNT side-walls in high-boiling-point solvent of DCB by a cycloaddition reaction of azido groups with the strained double bonds. The resultant SWNT-g-polyacetylenes can be soluble in common solvents, emit intense visible lights and strongly attenuate the power of harsh laser pulses.
In contrast to the direct attachment of azido-functionalized polymers to CNTs, covalent functionalization of alkyne-decorated SWNTs with well-defined, azido-end-capped PS chains was accomplished by the Cu(I)-catalyzed [3+2] Huisgen cycloaddition,160 in which alkyne-decorated SWNTs were obtained by the reaction of pristine SWNTs with p-aminophenyl propargyl ether using a solventfree diazotization and coupling procedure. This type of Huisgen cyclo-addition was highly efficient to prepare soluble polymer-nanotube conjugates, even at relatively low temperatures (60°C) and short reaction times (24 hours). The final SWNT-g-PS has a grafting density of 1 PS chain per 200~700 SWNT carbons, corresponding to roughly 45 wt% PS within them by TGA measurement. These materials exhibit high dispersibility, stability and solubility in organic solvents of polymers.
It is well known that various functional groups, such as carboxyl, hydroxyl, ketone and ester groups, are frequently formed on the surface of CNTs during oxidation treatment by strong acids for the purpose of purification, cutting and functionalization. Among them, CNT-bound carboxyl groups have attracted more attention due to their relatively higher reactivity and intrinsic multifunctionality towards chemical reactions. Chen et al.161 prepared soluble CNTs for the first time by an acylation-amidation process, where SWNT-bound carboxyl groups were treated with acyl chloride followed by reacting with octadecylamine (ODA) to achieve SWNT-g-ODA. Poly(m-aminobenzene sulfonic acid) (PABS)-grafted SWNTs (SWNT-g-PABS) was obtained by the same research team using a combination of acylation and amidation of acid-oxidized SWNTs.162−164 SWNT-g-PABS containing 17 wt% nanotubes is quite soluble in water and gives several orders of magnitude increase in electrical conductivity over pure PABS.162 Compared to the purified SWNTs, devices fabricated with SWNT-g-PABS show more than twice the higher resistance change upon exposure to NH3 due to the improved sensor performance.163 Further, SWNT-g-PAB S is used as a supporting scaffold for the growth of artificial bone material. It is found that hydroxyapatite (HA) can nucleate and crystallize on the surface of SWNT-g-PABS, and its film can be mineralized resulting in the well-aligned plate-shaped HA crystals.164 Recently, Zhang et al.165 prepared interlinked MWNTs with polyamines via reaction of the acyl-activated MWNTs with amino-functionalized alternating polyketones. The amount of polymer bonded to MWNT is estimated to be ~ 40 wt% by TGA analysis. Though cross-linked MWNTs are insoluble in any solvent, they can be melt-blended into low-density polyethylene (LDPE). However, the final LDPE composites containing functionalized-MWNTs show comparable mechanical properties to those obtained by simple blending of ‘uncross-linked’ nanotubes with LDPE. Thus, the enhancement of mechanical properties of LDPE matrix is not realized, probably due to the poor dispersion of cross-linked fillers when using melt blending. In the recent work of Ma et al.,166 flame-retardant poly(diaminodiphenyl methane spirocyclic pentaerythritol bisphosphonate) (PDSPB) was grafted onto the surface of acyl-activated MWNTs to obtain MWNT-g-PDSPB, and then integrated into ABS resin via melt blending to produce ABS–MWNT-g-PDSPB nanocomposites. MWNT-g-PDSPB (65 wt% grafting) was soluble and stable in polar solvents. MWNT-g-PDSPB can form structural networks at very low loading (0.2 wt%), lower than unmodified-MWNTs (1.0 wt%) as indicated by their linear viscoelastic behavior; better flame retardance is also observed for ABS–MWNT-g-PDSPB composites. These findings are due to better dispersion of MWNT-g-PDSPB relative to unmodified-MWNTs in ABS matrix.
Similarly, oxidized-CNTs can react with hydroxyl-terminated polymers via an acylation-esterification process. Here, carboxyl groups attached to CNTs are converted to acyl chlorides by treatment with thionyl chloride. The acyl-activated CNTs are then susceptible to reaction with the hydroxyl groups of polymers to produce polymer-grafted CNTs via ester linkages. By using this process, PEG has been used to prepare water-soluble SWNTs.167 Sun et al.168 and Tao et al.169 reported the functionalization of CNTs with various lipophilic and hydrophilic dendrons under the same reaction conditions. The experimental results suggest that there are no significant differences between amidation and esterification reactions regardless of SWNTs or MWNTs.168 In the work of Hill and co-workers,170 poly(styrene-co-p-(4-(4′-vinylphenyl)-3-oxabutanol)) (PSV) with medium molecular weight (Mw= 5,500 g/mol) was used to functionalize SWNTs and MWNTs by this acylation-esterification process. The resulting PSV-functionalized CNTs are soluble in common solvents, and are further applied to fabricate PS–CNT composite films by means of a wet-casting method. The homogeneous dispersion of PSV-functionalized CNTs in a PS matrix promises that CNTs solubilization will enable the preparation of desirable polymeric composites.
Unlike the two-step process of esterification or amidation towards acyl-activated-CNTs, carboxylic acid-functionalized CNTs after oxidization can directly react with amino- and/or hydroxyl-terminated polymers by one-step condensation to produce polymer-grafted CNTs via covalent amide or ester linkages. Sun et al. grafted poly(propionylethylenimine-co-ethylenimine) (PPEI-EI)171 and PEG172 onto oxidized-SWNTs by [1-ethyl-3-(dimethylamino) propyl] carbodiimide hydrochloride (EDC) activated condensation in an aqueous KH2PO4 buffer solution. The nanotube-bound carboxylic acids were also targeted for esterification reactions with pendant hydroxyl groups in the derivatized polyimide using N,N'- dicyclohexylcarbodiimide (DCC) activated condensation in organic solvent.173 Resonance Raman measurements indicate good dispersion of the functionalized samples as reflected by the overwhelming luminescence interference since such interference cannot be found in a simple polymer-nanotube mixture. The same approach was applied to graft PVA to SWNTs and MWNTs in the presence of DCC in highly polar dimethyl sulphoxide (DMSO).174 The resultant PVA-grafted nanotubes are soluble in highly polar solvents such as DMSO and water, thereby allow the intimate mixing of nanotubes with polymer matrix by wetcasting. The PVA composite films containg PVA-grafted nanotubes are of high optical quality without any observable phase separation. CNTs within the films are as well dispersed as in solution (Fig. 3.7), indicating that functionalization of CNTs by the matrix polymer is effective in the homogeneous dispersion of nanotube to fabricate high-quality nano-composites.
3.7 TEM image of PVA–MWNT-g-PVA films from cross-sectional microtome. (Reprinted with permission from ref. 174. Copyright 2003 American Chemical Society.).
In addition to the functionalization of CNTs by carbodiimide-activated condensation, some investigators have found that the surface carboxyl groups of oxidized CNTs can react directly with polymers containg amino- and/or hydroxyl groups through thermally induced condensation at high temperature in a one-step process. In the work of Ge et al.,175 the carboxylic acid-functionalized MWNTs (MWNT-COOH) were initially dispersed in a solution of aromatic polyetherimide with the aid of sonication, followed by removing the solvents from suspension. Finally, interfacial grafting reaction occurred in between MWNTs and the terminal amino groups of polyetherimide at 150–160°C for 96 h under N2 protection without any catalysts or force fields. The tensile strength and modulus of pure polyetherimide film were ~ 121 MPa and ~ 2.9 GPa with an elongation-at-break of 16%. However, the tensile strength and modulus of polyetherimide composite films grafted with 0.14, 25, 38 wt% MWNT–COOH increased to ~ 131, 186, 194 MPa and ~ 3.1, 4.1, 4.4 GPa, respectively, with corresponding elongations-at-break of 10, 8 and 7%. It is suggested that the covalent interfaces between nanotubes and polyetherimide chains play an important role in improving the mechanical performance.
Gao et al176, 177 reported a way to fabricate PA6–SWNT composites by ring-opening polymerization of caprolactam monomers using 6-aminocaproic acid as an initiator in the presence of oxidized SWNTs (SWNT-COOH). As shown in Fig. 3.8, PA6 chains can be chemically linked to SWNTs through condensation reactions between carboxyl groups around SWNTs and the terminal amino groups of PA6 at 250°C, where caprolactam also acts as an excellent dispersant for SWNT-COOH. Continuous fibers are then drawn from PA6–SWNT composites using a custom-made spinneret assembly. SWNTs are uniformly diffused into PA6 matrix with different SWNTs loading (Fig. 3.9), and show improved mechanical properties for PA6 nanocomposites. As compared with pure PA6, incorporating 1.5 wt% oxidized-SWNTs (3~4% carboxyl groups) into PA6 matrix results in increases from 40.9 MPa to 75.1 MPa in tensile strength, from 440 MPa to 1,200 MPa in Young’s modulus, corresponding to enhancements of 2.7 and 1.9 times, respectively.176
3.8 Synthesis of PA6–SWNT composites by ring-opening polymerization of caprolactam using 6-aminocaproci acid as an initiator in the presence of oxidized SWNTs. (Reprinted with permission from ref. 176. Copyright 2005 American Chemical Society.)
3.9 SEM cross-sectional fracture images of PA6–SWNT composite fibers prepared with SWNT loading of (a) 0.2; (b) 0.5; (c) 1.0 and (d) 1.5 wt%. The oxidized-SWNTs contain 3–4% carboxylic-acid sites, and the bright regions in these images are attributed to SWNTs as a result of high conductivity. (Reprinted with permission from ref. 176. Copyright 2005 American Chemical Society).
The mechanical properties of PA6–SWNT composite fibers are highly dependent on the density of the carboxylic acid functional groups attached to SWNTs.177 At similar SWNT loading (0.5 wt%), the composite fibers fabricated using SWNTs with 0, 4.2, 6.0 and 6.8% carboxyl groups show tensile strengths of 69.1, 83.4, 88.6, and 98.5 MPa, respectively, which are clearly stronger than the pure PA6 fiber (40.9 MPa). It can be concluded that the tensile strength of PA6–SWNT fibers increases with the coverage density of carboxyl groups attached to SWNTs. The same tendency for Young’s modulus increase is also observed for the composite fibers concerning the concentration of carboxylic acid groups. For example, PA6–SWNT fiber prepared with SWNTs containing 6.8% carboxyl groups shows a Young’s modulus of 1,500 MPa, an increase of ~ 2.6 times in comparison with that of the composite fiber from unmodified SWNTs (515 MPa). However, the break strain of the composite fibers is nearly independent of the amount of the carboxyl groups and lower than that of pure PA6 fiber. Moreover, it is found that replacement of SWNT–COOH by SWNT–CONH2 in the PA6–SWNT fibers entices grafting longer PA chains to SWNTs as a result of changes in condensation reaction, thereby resulting in different composite morphology and mechanical performance.177 At the same nanotube loading (0.5 wt%), PA6–SWNT fiber containing SWNT–CONH2 has a higher tensile strength and break strain, but a lower Young’s modulus relative to the fiber fabricated using SWNT–COOH under comparable coverage density of functional groups (4.2%). These results indicate that the fiber reinforced by SWNT–COOH is stiffer, while that containing SWNT–CONH2 is tougher (measured by area under the stress–strain curve).
The work reported by Gao et al.176, 177 clearly proves again that polymers attached to CNTs can effectively enhance CNT–matrix interfacial bonding and improve their compatibility; thereby allowing homogeneous dispersion of CNTs in the polymer matrix and resulting in an increased ability of load transfer from the matrix to nanotubes. Such polymer nanocomposites with CNTs are expected to possess excellent mechanical performance, thermal stability and electrical properties. Besides, both coverage density and the nature of functional groups bound onto CNTs influence the linkage reaction between CNTs and matrix, and hence the CNT–matrix interfacial interaction, composite morphology and final material properties.
Polymer–CNT nanocomposites possess desirable combined properties that cannot be found in individual components. From a structural viewpoint, there are three main system requirements for the effective improvement of polymers with CNTs. First, smaller diameters of CNTs (or CNT bundles) are required to produce a higher packing density. Second, a homogeneous dispersion of CNTs in the polymer matrix is necessary to avoid unreinforced regions and thereby obtain a uniform stress distribution throughout the composite. Finally, a strong matrix–nanotube interfacial bonding is desirable to effectively transfer the load from the matrix to CNTs. Apart from the first point, the latter two points can be achieved by maximizing the matrix–CNT physical contact or by constructing chemical linkages between the nanotubes and matrix. Non-covalent functionalization of CNTs is mainly based on the physical interactions of van der Waals forces or π-π stacking between two components, which are chiefly controlled by thermodynamics. This method offers a possibility to exfoliate the nanotubes into smaller bundles and even individual CNTs without causing any obvious structural damages to CNTs in solution and polymer matrix. However, the interface adhesion between the wrapping molecule and nanotube is weak, based on physical forces, thus the efficiency of load transfer is low in the composites. In contrast, covalent functionalization involves direct chemical attachment of functional groups to the π-conjugated skeleton of CNTs by virtue of grafting-from or grafting-to strategies. The grafted chains can be imagined as being entangled with the matrix polymer chains acting as bridges between CNTs and the matrix, and become an integral part of the polymer matrix. Therefore, CNTs with covalently attached polymers are superior in improving the material performance of polymers. However, the chemically functionalized CNTs inevitably result in the loss of their intrinsic structures, even if the evidence collected implies that CNTs can tolerate a limited number of structural defects before CNT-based materials lose their electronic and mechanical properties. Hence, how to functionalize CNTs without damaging their intrinsic structures is a great challenge for future research.
The authors acknowledge financial support from the Outstanding Youth Fund of the National Natural Science Foundation of China (50825301), the National Natural Science Foundation of China (20804014, 51073050) and the Program for Excellent Middle-Aged and Young Talents of Hubei Provincial Department of Education (Q20091005).
20. Frankland, S.J.V., Caglar, A., Brenner, D.W., Griebel, M. Molecular simulation of the influence of chemical cross-links on the shear strength of carbon nanotube-polymer interfaces. J. Phys. Chem. B. 2002; 106:3046–3048.
21. Graff, R.A., Swanson, J.P., Barone, P.W., Baik, S., Heller, D.A., Strano, M.S. Achieving individual-nanotube dispersion at high loading in single-walled carbon nanotube composites. Adv. Mater. 2005; 17:980–984.
22. Geng, H., Rosen, R., Zheng, B., Shimoda, H., Fleming, L., Liu, J., Zhao, Q. Fabrication and properties of composites of poly(ethylene oxide) and functionalized carbon nanotubes. Adv. Mater. 2002; 14:1387–1390.
28. Clayton, L.M., Sikder, A.K., Kumar, A., Cinke, M., Meyyappan, M., Gerasimov, T.G., Harmon, J.P. Transparent poly(methyl methacrylate)/single-walled carbon nanotube (PMMA/SWNT) composite films with increased dielectric constants. Adv. Funct. Mater. 2005; 15:101–106.
29. Putz, K.W., Mitchell, C.A., Krishnamoorti, R., Green, P.F. Elastic modulus of singlewalled carbon nanotube/poly(methyl methacrylate) nanocomposites. J. Polym. Sci. Part B: Polym. Phys. 2004; 42:2286–2293.
31. Kumar, S., Dang, T.D., Arnold, F.E., Bhattacharyya, A.R., Min, B.G., Zhang, X., Vaia, R.A., Park, C., Adams, W.W., Hauge, R.H., Smalley, R.E., Ramesh, S., Willis, P.A. Synthesis, structure, and properties of PBO/SWNT composites. Macromolecules. 2002; 35:9039–9043.
32. Zhao, C., Hu, G., Justice, R., Schaefer, D.W., Zhang, S., Yang, M., Han, C.C. Synthesis and characterization of multi-walled carbon nanotubes reinforced polyamide 6 via in-situ polymerization. Polymer. 2005; 46:5125–5132.
33. Velasco-Santos, C., Martínez-Hernández, A.L., Fisher, F.T., Ruoff, R., Castaño, V.M. Improvement of thermal and mechanical properties of carbon nanotube composites through chemical functionalization. Chem. Mater. 2003; 15:4470–4475.
35. Wang, H., Zhou, W., Ho, D.L., Winey, K.I., Fischer, J.E., Glinka, C.J., Hobbie, E.K. Dispersing single-walled carbon nanotubes with surfactants: a small angle neutron scattering study. Nano Lett. 2004; 4:1789–1793.
37. Grossiord, N., van der Schoot, P., Meuldijk, J., Koning, C.E. Determination of the surface coverage of exfoliated carbon nanotubes by surfactant molecules in aqueous solution. Langmuir. 2007; 23:3646–3653.
42. Park, S.J., Cho, M.S., Lim, S.T., Choi, H.J., Jhon, M.S. Synthesis and dispersion characteristics of multi-walled carbon nanotube composites with poly(methyl methacrylate) prepared by in-situ bulk polymerization. Macromol. Rapid Commun. 2003; 24:1070–1073.
44. Qin, S., Qin, D., Ford, W.T., Herrera, J.E., Resasco, D.E., Bachilo, S.M., Weisman, R.B. Solubilization and purification of single-wall carbon nanotubes in water by in situ radical polymerization of sodium 4-styrenesulfonate. Macromolecules. 2004; 37:3965–3967.
45. Qin, S., Qin, D., Ford, W.T., Herrera, J.E., Resasco, D.E. Grafting of poly(4-vinylpyridine) to single-walled carbon nanotubes and assembly of multilayer films. Macromolecules. 2004; 37:9963–9967.
47. Coleman, J.N., Cadek, M., Blake, R., Nicolosi, V., Ryan, K.P., Belton, C., Fonseca, A., Nagy, J.B., Gun’ko, Y.K., Blau, W.J. High performance nanotube-reinforced plastics: understanding the mechanism of strength increase. Adv. Funct. Mater. 2004; 14:791–798.
49. Wang, F., Wang, G., Yang, S., Li, C. Layer-by-layer assembly of aqueous dispersible, highly conductive poly(aniline-co-o-anisidine)/poly(sodium 4-styrenesulfonate)/MWNTs core-shell nanocomposites. Langmuir. 2008; 24:5825–5831.
51. Curran, S.A., Ajayan, P.M., Blau, W.J., Carroll, D.L., Coleman, J.N., Dalton, A.B., Davey, A.P., Drury, A., McCarthy, B. A composite from poly(m-phenylenevinylene-co-2,5-dioctoxy-p-phenylenevinylene) and carbon nanotubes: a novel material for molecular optoelectronics. Adv. Mater. 1998; 10:1091–1093.
53. Steuerman, D.W., Star, A., Narizzano, R., Choi, H., Ries, R.S., Nicolini, C., Stoddart, J.F., Heath, J.R. Interactions between conjugated polymers and single-walled carbon nanotubes. J. Phys. Chem. B. 2002; 106:3124–3130.
54. Tchoul, M.N., Ford, W.T., Ha, M.L.P., Chavez-Sumarriva, I., Grady, B.P., Lolli, G., Resasco, D.E., Arepalli, S. Composites of single-walled carbon nanotubes and polystyrene: preparation and electrical conductivity. Chem. Mater. 2008; 20:3120–3126.
56. Kim, J.B., Premkumar, T., Giani, O., Robin, J.J., Schue, F., Geckeler, K.E. A mechanochemical approach to nanocomposites using single-wall carbon nanotubes and poly(L-lysine). Macromol. Rapid Commun. 2007; 28:767–771.
59. Bavastrello, V., Carrara, S., Ram, M.K., Nicolini, C. Optical and electrochemical properties of poly(o-toluidine) multiwalled carbon nanotubes composite Langmuir-Schaefer films. Langmuir. 2004; 20:969–973.
61. Kim, K.K., Yoon, S.M., Choi, J.Y., Lee, J., Kim, B.K., Kim, J.M., Lee, J.H., Paik, U., Park, M.H., Yang, C.W., An, K.H., Chung, Y., Lee, Y.H. Design of dispersants for the dispersion of carbon nanotubes in an organic solvent. Adv. Funct. Mater. 2007; 17:1775–1783.
63. Yi, W., Malkovskiy, A., Chu, Q., Sokolov, A.P., Colon, M.L., Meador, M., Pang, Y. Wrapping of single-walled carbon nanotubes by a π-conjugated polymer: the role of polymer conformation-controlled size selectivity. J. Phys. Chem. B. 2008; 112:12263–12269.
67. Cadek, M., Coleman, J.N., Barron, V., Hedicke, K., Blau, W.J. Morphological and mechanical properties of carbon-nanotube-reinforced semicrystalline and amorphous polymer composites. Appl. Phys. Lett. 2002; 81:5123–5125.
74. Li, L., Li, C.Y., Ni, C., Rong, L., Hsiao, B. Structure and crystallization behavior of Nylon 66/multi-walled carbon nanotube nanocomposites at low carbon nanotube contents. Polymer. 2007; 48:3452–3460.
75. Bonduel, D., Bredeau, S., Alexandre, M., Monteverde, F., Dubois, P. Supported metallocene catalysis as an efficient tool for the preparation of polyethylene/carbon nanotube nanocomposites: effect of the catalytic system on the coating morphology. J. Mater. Chem. 2007; 17:2359–2366.
76. Toti, A., Giambastiani, G., Bianchini, C., Meli, A., Bredeau, S., Dubois, Ph., Bonduel, D., Claes, M. Tandem action of early-late transition metal catalysts for the surface coating of multiwalled carbon nanotubes with linear low-density polyethylene. Chem. Mater. 2008; 20:3092–3098.
77. Wang, K., Chen, F., Zhang, Q., Fu, Q. Shish-kebab of polyolefin by ‘melt manipulation’ strategy in injection-molding: a convenience pathway from fundament to application. Polymer. 2008; 49:4745–4755.
78. Liang, S., Wang, K., Chen, D., Zhang, Q., Du, R., Fu, Q. Shear enhanced interfacial interaction between carbon nanotubes and polyethylene and formation of nanohybrid shish-kebabs. Polymer. 2008; 49:4925–4929.
79. Yang, J., Wang, C., Wang, K., Zhang, Q., Chen, F., Du, R., Fu, Q. Direct formation of nanohybrid shish-kebab in the injection molded bar of polyethylene/multiwalled carbon nanotubes composite. Macromolecules. 2009; 42:7016–7023.
83. Zhang, F., Zhang, H., Zhang, Z., Chen, Z., Xu, Q. Modification of carbon nanotubes: water-soluble polymers nanocrystal wrapping to periodic patterning with assistance of supercritical CO2. Macromolecules. 2008; 41:4519–4523.
84. Dai, X., Liu, Z., Han, B., Sun, Z., Wang, Y., Xu, J., Guo, X., Zhao, N., Chen, J. Carbon nanotube/poly(2,4-hexadiyne-1,6-diol) nanocomposites prepared with the aid of supercritical CO2. Chem. Commun. 2004; 2190–2191.
90. Yang, Y.K., Xie, X.L., Wu, J.G., Mai, Y.-W. Synthesis and self-assembly of polystyrene-grafted multiwalled carbon nanotubes with a hairy-rod nanostructure. J. Polym. Sci., Part A: Polym. Chem. 2006; 44:3869–3881.
91. Du, F.P., Wu, K.B., Yang, Y.K., Liu, L., Gan, T., Xie, X.L. Synthesis and electrochemical probing of water-soluble poly(sodium 4-styrenesulfonate-co-acrylic acid)-grafted multiwalled carbon nanotubes. Nanotechnology. 2008; 19:085716.
93. Ha, J.U., Kim, M., Lee, J., Choe, S., Cheong, I.W., Shim, S.E. A novel synthesis of polymer brush on multiwall carbon nanotubes bearing terminal monomeric unit. J. Polym. Sci, Part A: Polym. Chem. 2006; 44:6394–6401.
94. Kim, M., Hong, C.K., Choe, S., Shim, S.E. Synthesis of polystyrene brush on multiwalled carbon nanotubes treated with kmno4 in the presence of a phase-transfer catalyst. J. Polym. Sci, Part A: Polym. Chem. 2007; 45:4413–4420.
96. Li, S., Chen, H., Bi, W., Zhou, J., Wang, Y., Li, J., Cheng, W., Li, M., Li, L., Tang, T. Synthesis and characterization of polyethylene chains grafted onto the sidewalls of single-walled carbon nanotubes via copolymerization. J. Polym. Sci., Part. A: Polym. Chem. 2007; 45:5459–5469.
97. Yao, Z., Braidy, N., Botton, G.A., Adronov, A. Polymerization from the surface of single-walled carbon nanotubes-reparation and characterization of nanocomposites. J. Am. Chem. Soc. 2003; 125:16015–16024.
100. Qin, S., Qin, D., Ford, W.T., Resasco, D.E., Herrera, J.E. Polymer brushes on single-walled carbon nanotubes by atom transfer radical polymerization of n-butyl methacrylate. J. Am. Chem. Soc. 2004; 126:170–176.
101. Wang, M., Pramoda, K.P., Goh, S.H. Enhancement of the mechanical properties of poly(styrene-co-acrylonitrile) with poly(methyl methacrylate)-grafted multiwalled carbon nanotubes. Polymer. 2005; 46:11510–11516.
102. Shi, J.H., Yang, B.X., Pramoda, K.P., Goh, S.H. Enhancement of the mechanical performance of poly(vinyl chloride) using poly(n-butyl methacrylate)-grafted multi-walled carbon nanotubes. Nanotechnology. 2007; 18:375704.
103. Cui, J., Wang, W.P., You, Y.Z., Liu, C., Wang, P. Functionalization of multiwalled carbon nanotubes by reversible addition fragmentation chain-transfer polymerization. Polymer. 2004; 45:8717–8721.
104. Pei, X.W., Liu, W.M., Hao, J.C. Functionalization of multiwalled carbon nanotube via surface reversible addition fragmentation chain transfer polymerization and as lubricant additives. J. Polym. Sci, Part A: Polym. Chem. 2008; 46:3014–3023.
105. Xu, G., Wu, W.T., Wang, Y., Pang, W., Zhu, Q., Wang, P., You, Y. Constructing polymer brushes on multiwalled carbon nanotubes by in situ reversible addition fragmentation chain transfer polymerization. Polymer. 2006; 47:5909–5918.
106. Hong, C.Y., You, Y.Z., Pan, C.Y. Functionalized multi-walled carbon nanotubes with poly(N-(2-hydroxypropyl)methacrylamide) by RAFT polymerization. J. Polym. Sci., Part A: Polym. Chem. 2006; 44:2419–2427.
109. Hong, C.Y., You, Y.Z., Pan, C.Y. Synthesis of water-soluble multiwalled carbon nanotubes with grafted temperature-responsive shells by surface raft polymerization. Chem. Mater. 2005; 17:2247–2254.
110. Xu, G., Wu, W.T., Wang, Y., Pang, W., Wang, P., Zhu, Q., Lu, F. Synthesis and characterization of water-soluble multiwalled carbon nanotubes grafted by a thermoresponsive polymer. Nanotechnology. 2006; 17:2458–2465.
112. Zhao, X.D., Fan, X.H., Chen, X.F., Chai, C.P., Zhou, Q.F. Surface modification of multiwalled carbon nanotubes via nitroxide-mediated radical polymerization. J. Polym. Sci., Part A: Polym. Chem. 2006; 44:4656–4667.
113. Dehonor, M., Masenelli-Varlot, K., González-Montiel, A., Gauthier, C., Cavaillé, J.Y., Terronesa, H., Terrones, M. Nanotube brushes: polystyrene grafted covalently on CNx nanotubes by nitroxide-mediated radical polymerization. Chem. Commun. 2005; 5349–5351.
114. Datsyuk, V., Guerret-Piécourt, C., Dagréou, S. L Billon, J.C. Dupin, E. Flahaut, A. Peigney, C. Laurent, Double walled carbon nanotube/polymer composites via in-situ nitroxide mediated polymerisation of amphiphilic block copolymers. Carbon. 2005; 43:873–876.
116. Buffa, F., Hu, H., Resasco, D.E. Side-wall functionalization of single-walled carbon nanotubes with 4-hydroxymethylaniline followed by polymerization of ε-caprolactone. Macromolecules. 2005; 38:8258–8263.
117. Yang, Y.K., Tsui, C.P., Tang, C.Y., Qiu, S.Q., Zhao, Q., Cheng, X.J., Sun, Z.G., Li, R.K.Y., Xie, X.L. Functionalization of carbon nanotubes with biodegradable supramolecular polypseudorotaxanes from grafted-poly(ε-caprolactone) and α-cyclodextrins. Eur. Polym. J. 2010; 46:145–155.
121. Xu, Y., Gao, C., Kong, H., Yan, D., Jin, Y.Z., Watts, P.C.P. Growing multihydroxyl hyperbranched polymers on the surfaces of carbon nanotubes by in situ ring-opening polymerization. Macromolecules. 2004; 37:8846–8853.
122. Qu, L., Veca, L.M., Lin, Y., Kitaygorodskiy, A., Chen, B., McCall, A.M., Connell, J.W., Sun, Y.P. Soluble Nylon-functionalized carbon nanotubes from anionic ring-opening polymerization from nanotube surface. Macromolecules. 2005; 38:10328–10331.
124. Zhu, J., Kim, J.D., Peng, H., Margrave, J.L., Khabashesku, V.N., Barrera, E.V. Improving the dispersion and integration of single-walled carbon nanotubes in epoxy composites through functionalization. Nano. Lett. 2003; 3:1107–1113.
125. Bae, J., Jang, J., Yoon, S.H. Cure behavior of the liquid-crystalline epoxy/carbon nanotube system and the effect of surface treatment of carbon fillers on cure reaction. Macromol. Chem. Phys. 2002; 203:2196–2204.
126. Bekyarova, E., Thostenson, E.T., Yu, A., Itkis, M.E., Fakhrutdinov, D., Chou, T.W., Haddon, R.C. Functionalized single-walled carbon nanotubes for carbon fiber-epoxy composites. J. Phys. Chem. C. 2007; 111:17865–17871.
129. Zhu, J., Peng, H., Rodriguez-Macias, F., Margrave, J.L., Khabashesku, V.N., Iman, A.M., Lozano, K., Barrera, E.V. Reinforcing epoxy polymer composites through covalent integration of functionalized nanotubes. Adv. Funct. Mater. 2004; 14:643–648.
130. Liu, L., Etika, K.C., Liao, K.S., Hess, L.A., Bergbreiter, D.E., Grunlan, J.C. Comparison of covalently and noncovalently functionalized carbon nanotubes in epoxy. Macromol. Rapid Commun. 2009; 30:627–632.
131. Yang, Y.K., Xie, X.L., Wu, J.G., Yang, Z.F., Wang, X.T., Mai, Y.-W. Multiwalled carbon nanotubes functionalized by hyperbranched poly(urea-urethane)s by a one-pot polycondensation. Macromol. Rapid Commun. 2006; 27:1695–1701.
132. Yang, Y.K., Xie, X.L., Yang, Z.F., Wang, X.T., Cui, W., Yang, J.Y., Mai, Y.-W. Controlled synthesis and novel solution rheology of hyperbranched poly(urea-urethane)-functionalized multiwalled carbon nanotubes. Macromolecules. 2007; 40:5858–5867.
133. Yang, Y.K., Wang, X.T., Liu, L., Xie, X.L., Yang, Z.F., Li, R.K.Y., Mai, Y.-W. Structures and photoresponsive behaviors of multiwalled carbon nanotubes grafted by polyurethanes containing azobenzene side-chains. J. Phys. Chem. C. 2007; 111:11231–11239.
134. Gao, C., Jin, Y.Z., Kong, H., Whitby, R.L.D., Acquah, S.F.A., Chen, G.Y., Qian, H., Hartschuh, A., Silva, S.R.P., Henley, S., Fearon, P., Kroto, H.W., Walton, D.R.M. Polyureafunctionalized multiwalled carbon nanotubes: synthesis, morphology, and Raman spectroscopy. J. Phys. Chem. B. 2005; 109:11925–11932.
135. Chen, X., Chen, X., Lin, M., Zhong, W., Chen, X., Chen, Z. Functionalized multi-walled carbon nanotubes prepared by in-situ polycondensation of polyurethane. Macromol. Chem. Phys. 2007; 208:964–972.
137. Zhou, C., Wang, S., Zhang, Y., Zhuang, Q., Han, Z. In situ preparation and continuous fiber spinning of poly(p-phenylene benzobisoxazole) composites with oligo-hydroxyamide-functionalized multi-walled carbon nanotubes. Polymer. 2008; 49:2520–2530.
138. Oh, S.J., Lee, H.J., Keum, D.K., Lee, S.W., Wang, D.H., Park, S.Y., Tan, L.S., Baek, J.B. Multiwalled carbon nanotubes and nanofibers grafted with polyetherketones in mild and viscous polymeric acid. Polymer. 2006; 47:1132–1140.
139. Choi, J.Y., Oh, S.J., Lee, H.J., Wang, D.H., Tan, L.S., Baek, J.B. In-situ grafting of hyperbranched poly(ether ketone)s onto multiwalled carbon nanotubes via the A3 + B2 approach. Macromolecules. 2007; 40:4474–4480.
140. Jeon, I.Y., Tan, L.S., Baek, J.B. Nanocomposites derived from in situ grafting of linear and hyperbranched poly(ether-ketone)s containing flexible oxyethylene spacers onto the surface of multiwalled carbon nanotubes. J. Polym. Sci., Part A: Polym. Chem. 2008; 46:3471–3481.
141. Liu, I.C., Huang, H.M., Chang, C.Y., Tsai, H.C., Hsu, C.H., Tsiang, R.C.C. Preparing a styrenic polymer composite containing well-dispersed carbon nanotubes: anionic polymerization of a nanotube-bound p-methylstyrene. Macromolecules. 2004; 37:283–287.
142. Viswanathan, G., Chakrapani, N., Yang, H., Wei, B., Chung, H., Cho, K., Ryu, C.Y., Ajayan, P.M. Single-step in situ synthesis of polymer-grafted single-wall nanotube composites. J. Am. Chem. Soc. 2003; 125:9258–9259.
143. Liang, F., Beach, J.M., Kobashi, K., Sadana, A.K., Vega-Cantu, Y.I., Tour, J.M., Billups, W.E. In-situ polymerization initiated by single-walled carbon nanotube salts. Chem. Mater. 2006; 18:4764–4767.
144. Zhang, Y., Shen, Y., Li, J., Niu, L., Dong, S., Ivaska, A. Electrochemical functionalization of single-walled carbon nanotubes in large quantities at a room-temperature ionic liquid supported three-dimensional network electrode. Langmuir. 2005; 21:4797–4800.
146. Priftis, D., Petzetakis, N., Sakellariou, G., Pitsikalis, M., Baskaran, D., Mays, J.W., Hadjichristidis, N. Surface-initiated titanium-mediated coordination polymerization from catalyst-functionalized single and multiwalled carbon nanotubes. Macromolecules. 2009; 42:3340–3346.
149. Lou, X., Detrebleur, C., Pagnoulle, C., Jérôme, R., Bocharova, V., Kiriy, A., Stamm, M. Surface modification of multiwalled carbon nanotubes by poly(2-vinylpyridine): dispersion, selective deposition, and decoration of the nanotubes. Adv. Mater. 2004; 16:2123–2127.
152. Wu, H.X., Tong, R., Qiu, X.Q., Yang, H.F., Lin, Y.H., Cai, R.F., Qian, S.X. Functionalization of multiwalled carbon nanotubes with polystyrene under atom transfer radical polymerization conditions. Carbon. 2007; 45:152–159.
154. McIntosh, D., Khabashesku, V.N., Barrera, E.V. Benzoyl peroxide initiated in situ functionalization, processing, and mechanical properties of single-walled carbon nanotube-polypropylene composite fibers. J. Phys. Chem. C. 2007; 111:1592–1600.
158. Qin, S., Qin, D., Ford, W.T., Resasco, D.E., Herrera, J.E. Functionalization of single-walled carbon nanotubes with polystyrene via grafting to and grafting from methods. Macromolecules. 2004; 37:752–757.
159. Li, Z., Dong, Y., Hussler, M., Lam, J.W.Y., Dong, Y.P., Wu, L., Wong, K.S., Tang, B.Z. Synthesis of, light emission from, and optical power limiting in soluble single-walled carbon nanotubes functionalized by disubstituted polyacetylenes. J. Phys. Chem. B. 2006; 110:2302–2309.
163. Bekyarova, E., Davis, M., Burch, T., Itkis, M.E., Zhao, B., Sunshine, S., Haddon, R.C. Chemically functionalized single-walled carbon nanotubes as ammonia sensors. J. Phys. Chem. B. 2004; 108:19717–19720.
164. Zhao, B., Hu, H., Mandal, S.K., Haddon, R.C. A bone mimic based on the self-assembly of hydroxyapatite on chemically functionalized single-walled carbon nanotubes. Chem. Mater. 2005; 17:3235–3241.
165. Zhang, Y., Broekhuis, A.A., Stuart, M.C.A., Fernandez, T., Landaluce, P., Fausti, D., Rudolf, P., Picchioni, F. Cross-linking of multiwalled carbon nanotubes with polymeric amines. Macromolecules. 2008; 41:6141–6146.
166. Ma, H.Y., Tong, L.F., Xu, Z.B., Fang, Z.P. Functionalizing carbon nanotubes by grafting on intumescent flame retardant: nanocomposite synthesis, morphology, rheology, and flammability. Adv. Funct. Mater. 2008; 18:414–421.
168. Sun, Y.P., Huang, W., Lin, Y., Fu, K., Kitaygorodskiy, A., Riddle, L.A., Yu, Y.J., Carroll, D.L. Soluble dendron-functionalized carbon nanotubes: preparation, characterization, and properties. Chem. Mater. 2001; 13:2864–2869.
169. Tao, L., Chen, G., Mantovani, G., Yorkb, S., Haddleton, D.M. Modification of multi-wall carbon nanotube surfaces with poly(amidoamine) dendrons: synthesis and metal templating. Chem. Commun. 2006; 4949–4951.
171. Lin, Y., Hill, D.E., Bentley, J., Allard, L.F., Sun, Y.P. Characterization of functionalized single-walled carbon nanotubes at individual nanotube-thin bundle level. J. Phys. Chem. B. 2003; 107:10453–10457.
172. Huang, W., Fernando, S., Allard, L.F., Sun, Y.P. Solubilization of single-walled carbon nanotubes with diamine-terminated oligomeric poly(ethylene glycol) in different functionalization reactions. Nano. Lett. 2003; 3:565–568.
174. Lin, Y., Zhou, B., Fernando, K.A.S., Liu, P., Allard, L.F., Sun, Y.P. Polymeric carbon nanocomposites from carbon nanotubes functionalized with matrix polymer. Macromolecules. 2003; 36:7199–7204.
175. Ge, J.J., Zhang, D., Li, Q., Hou, H., Graham, M.J., Dai, L., Harris, F.W., Cheng, S.Z.D. Multiwalled carbon nanotubes with chemically grafted polyetherimides. J. Am. Chem. Soc. 2005; 127:9984–9985.
177. Gao, J., Zhao, B., Itkis, M.E., Bekyarova, E., Hu, H., Kranak, V., Yu, A., Haddon, R.C. Chemical engineering of the single-walled carbon nanotube-nylon 6 interface. J. Am. Chem. Soc. 2006; 128:7492–7496.