**6. Challenges in MWCNT polymer composites fabrication and possible solutions**

Although these fabrication methods helped to enhance the properties of CNT reinforced composites over neat polymer but there are several key challenges that hinders the excellent CNT properties to be fruitful in polymer composite formation.

#### **6.1. Dispersion**

form master batches that were thereafter diluted with pure polymer to form lower mass fraction samples. Compression molding was used to form composite films. A similar combi‐ nation of shear mixing and compression molding is studied by many other groups dis‐ cussed elsewhere [16]. Also Meincke et al. [5] mixed polyamide-6, ABS and CVD-MWCNT

Tang et al. [26] used both compression and twin-screw extrusion to form CNT/polyethylene composites. Although melt-processing technique has advantages of speed and simplicity, it is not much effective in breaking of agglomeration of CNTs and their dispersion. Bhatta‐ charyya et al. [27] made 1 wt% CNT/polypropylene (PP) nanocomposites by melt mixing, but found that melt mixing alone did not provide uniform nanotube dispersion. Niu et el. [28] studied both methods to prepare polyvinylidene fluoride (PVDF)-CNT nanocomposites to study electrical properties and found it better in composites formed by solution casting.

In addition to solvent casting and melt mixing the other method which combines nanotubes with high molecular weight polymers is in-situ polymerization starting with CNTs and mono‐ mers. In-situ polymerization has advantages over other composite fabrication methods. A stronger interface can be obtained because it is easier to get intimate interactions between the polymer and nanotube during the growth stage than afterwards [29, 30].The most common in situ polymerization methods involve epoxy in which the monomer resins and hardeners are combined with CNTs prior to polymerizing [31]. Pande and coworkers [32] performed the insitu polymerization of MWCNT/ PMMA composites for the enhancement in flexural strength and modulus of composites. Li et al. [33] reported the fabrication and characterization of CNT/ polyaniline (PANI) composites. Xiao and Zhou [34]deposited polypyrrolre (PP) and poly(3 methylthiophene) (PMet) on the surface of MWCNTs by in situ polymerization. Saini et al. [35] reported fabrication process of highly conducting polyaniline (PANI)–(MWCNT) nanocompo‐ sites by in-situ polymerization. This material was used in polystyrene for the fabrication of MWCNT-PANI-PS blend for microwave absorption [36]. Moniruzzaman [17] reported many other studies of in-situ polymerization of CNTs with different polymers. Generally, in situ poly‐ merization can be used for the fabrication of almost any polymer composites containing CNT that can be non-covalently or covalently bound to polymer matrix. This technique enables the grafting of large polymer molecules onto the walls of CNT. This technique is particularly impor‐ tant for the preparation of insoluble and thermally unstable polymers, which cannot be process‐

Some studies have been also carried out using combined methods, such as solvent casting in conjunction with sonication, followed by melt mixing. Haggenmueller et al. [37] observed considerable nanotube dispersion in CNT-polymer nanocomposites using combination of solvent casting and melt mixing. Pande et al. [32, 38] also prepared MWCNT bulk compo‐ sites with PMMA and PS using a two-step method of solvent casting followed by compres‐ sion molding and obtained better electrical and mechanical properties. Singh et al. [39] also prepared MWCNT-LDPE composites using solvent casting followed by compression moulding and obtained better electrical conductivity. Jindal et al. [29, 40] prepared

C and used injection molding to make nanocomposites.

in a twin screw extruder at 260o

204 Syntheses and Applications of Carbon Nanotubes and Their Composites

*5.1.3. In-situ polymerization*

ed by solution or melt processing.

Disperion of nanoscale filler in a matrix is the key challenge for the formation of nanocomposite. Dispersion involves separation and then stabilization of CNTs in a medium. The methods de‐ scribed above for the nanocomposites fabrication require CNTs to be well dispersed either in solvent or in polymer for maximizing their contact surface area with polymer matrix. As CNTs have diameters on nanoscale the entanglement during growth and the substantial van der Waals interaction between them forces to agglomerate into bundles. The ability of bundle for‐ mation of CNTs with its inert chemical structure makes these high aspect ratio fibers dissolving in common solvents to form solution quite impossible. The SEM of MWCNTs synthesized by CVD technique seems to be highly entangled and the dimensions of nanotube bundles is hun‐ dreds of micrometer. This shows several thousands of MWCNTs in one bundle as shown in Fig‐ ure 5a. These bundles exhibits inferior mechanical and electrical properties as compared to individual nanotube because of slippage of nanotubes inside bundles and lower aspect ratio as compared to individual nanotube. The aggregated bundles tend to act as defect sites which ad‐ versely affect mechanical and electrical properties of nanocomposites. Effective separation re‐ quires the overcoming of the inter-tube van der Waal attraction, which is anomalously strong in CNT case. To achieve large fractions of individual CNT several methods have been employed. The most effective methods are by attaching several functional sites on the surface of CNTs through some chemical treatment or by surrounding the nanotubes with dispersing agents such as surfactant. Thereafter the difficulty of dispersion can be overcome by mechanical/physical means such as ultrasonication, high shear mixing or melt blending. Another obstacle in dispers‐ ing the CNTs is the presence of various impurities including amorphous carbon, spherical full‐ erenes and other metal catalyst particles. These impurities are responsible for the poor properties of CNTs reinforced composites [45].

#### **6.2. Adhesion between CNTs and polymer**

The second key challenge is in creating a good interface between nanotubes and the poly‐ mer matrix. From the research on microfiber based polymer composites over the past few decades, it is well established that the structure and properties of filler-matrix interface plays a major role in determining the structural integrity and mechanical performance of composite materials. CNTs have atomically smooth non-reactive surfaces and as such there is a lack of interfacial bonding between the CNT and the polymer chains that limits load transfer. Hence the benefits of high mechanical properties of CNTs are not utilized properly. The first experimental study focusing on interfacial interaction between MWCNT and poly‐ mer was carried out by Cooper et al. [46]. They investigated the detachment of MWCNTs from an epoxy matrix using a pullout test for individual MWCNT and observed the interfa‐ cial shear stress varied from 35-376 MPa. This variation is attributed to difference in struc‐ ture and morphology of CNTs.

tween CNTs. It is also vital to stabilize the dispersion to prevent reagglomeration of the CNTs. Chemical functionalization can prevent reagglomeration of CNTs also. Sen et al. [47] carried out chemical functionalization to form ester functionalized CNTs and found that it is an effective approach to exfoliate the CNTs bundles and improve their processibility with polymer matrix. Georgakilas et al. [48] observed that CNT covalently functionalized with pyrrolidone by 1,3-dipolar cycloaddition of azomethine ylides show a solubility of 50 mg/mL in chloroform, even without sonication whereas the pristine CNT is completely in‐ soluble in this solvent. Liang et al. [49] performed reductive alkylation of CNTs using lithi‐ um and alkyl halides in liquid ammonia for sidewall functionalization of CNTs and observed their extensive debundling by inspection of HRTEM images. Kinloch et al. [50] studied the rheological behavior of oxidized CNTs and found that the composites filled with functionalized CNTs had better dispersion. It has been observed by the researchers that amine modified CNTs is very important for the enhancement in the mechanical properties with epoxy. Garg et al. [31] shows the reaction mechanism for the formation of acid func‐ tionalized and amine functionalized CNT and their interaction with the epoxy resin as Fig‐ ure 8a and b respectively. Two different types of functional groups were attached on the CNT surface. In the first case MWCNTs were refluxed for 48 h in HNO3 (400 ml, 60% con‐ centration) to achieve reasonable surface oxidation of the tubes. The mixture was then fil‐ tered and the residue (treated material) was washed several times with distilled water till washings were neutral to pH paper. The treated MWCNT were dried in oven before use. In a second step the oxidized nanotubes were dispersed in benzene by stirring, and then re‐ fluxed with excess SOCl2 along with a few drops of DMF used as catalyst for chlorination of MWCNT surfaces. After the acyl chlorination, SOCl2 and DMF were removed through re‐ peatedly washing by tetrahydrofuran (THF). 100 ml of triethylene tetra-amine (TETA) was added to react with acyl chlorinated MWCNT at 100 °C for 24 h reflux until no HCl gas evolved. After cooling to room temperature, MWCNTs were washed with deionized water 5 times to remove excess TETA. Finally, the black solid was dried at room temperature over‐ night in vacuum and named as amine modified CNTs. These functionalized CNTs were characterized by FTIR, TGA and HRTEM and clearly showed the presence of these types of functional group. Gojny et al. [51]achieved surface modified MWCNTs by refluxing of oxi‐ dized MWCNTs with multifunctional amines and observed from TEM images that these were completely covered by epoxy matrix that confirmed the bonding between them. Sin‐ nott [52]has provided an in depth review of chemical functionalization of CNTs where the chemical bonds are used to tailor the interactions between nanotubes and polymers or sol‐ vent. The chemical functionalization of CNTs has also been accomplished through irradia‐ tion with electrons or ions [53]. In this manner one may hope to improve the binding of

Carbon Nanotubes and Their Composites http://dx.doi.org/10.5772/52897 207

CNTs by interdigitation of active sites on its sides into polymer matrix.

provides bonding sites to the polymer matrix.

Covalent bond also benefits phonon transferring between nanotubes and polymer matrix, which is a key factor for improving thermal conductivity of the nanocomposites. To ensure the adhesion between polymer and nanotubes various surfactant and chemical modification procedures have been adopted to modify the surface of otherwise inert surface of CNTs that

There are three main mechanisms for load transfer from matrix to filler. The first is weak van der Waal interaction between filler and polymer. Using small size filler and close con‐ tact at the interface can increase it. The large specific surface area of CNTs is advantageous for bonding with matrix in a composite, but is a major cause for agglomeration of CNTs. Therefore, uniformally dispersed individual nanotubes in matrix is helpful. The second mechanism of load transfer is micromechanical interlocking which is difficult in CNTs nano‐ composites due to their atomically smooth surface. Although local non uniformity along length of CNTs i.e. varying diameter and bends due to non hexagonal defects contributes to this micromechanical interlocking. This interlocking can increase by using long CNTs to block the movement of polymer chains. The contribution of this mechanism may reach satu‐ ration at low CNT content. The third and best mechanism for better adhesion and hence load transfer between CNTs and polymer is covalent or ionic bonding between them. The chemical bonding between CNTs and polymer can be created and enhanced by the surface treatment such as oxidation of CNTs with acids or other chemicals. This mode of mecha‐ nism have much importance as it provides strong interaction between polymer and CNT and hence efficiently transfers the load from polymer matrix to nanotubes necessary for en‐ hanced mechanical response in high-performance polymers.

#### **6.3. Chemical functionalization of CNTs**

The best route to achieve individual CNT to ensure better dispersion is chemical modifica‐ tions of CNT surface. The chemical functionalization involves the attachment of chemical bonds to CNT surface or on end caps. Nanotube functionalization typically starts with oxi‐ datative conditions, commonly by refluxing in nitric or sulfuric acid or combination of both to attach carboxylic acid moieties to the defect sites. The end caps of nanotubes have extra strain energy because of their high degree of curvature with pentagons and heptagonal car‐ bon atoms are most vulnerable to reaction with acid. The side walls also containing defects like pentagon-heptagon pairs, sp3 hybridized defects and vacancies in nanotube lattice and are easily supplemented by oxidative damage and can be stabilized by formation of func‐ tional groups mainly carboxylic acid and hydroxide group. These acid moieties and hydrox‐ ide groups can be further replaced to more reactive groups like –COCl or –CORNH2. The addition of these functional groups on CNTs possesses intermolecular repulsion between functional groups on surface that overcomes the otherwise weak van der Waal attraction be‐ tween CNTs. It is also vital to stabilize the dispersion to prevent reagglomeration of the CNTs. Chemical functionalization can prevent reagglomeration of CNTs also. Sen et al. [47] carried out chemical functionalization to form ester functionalized CNTs and found that it is an effective approach to exfoliate the CNTs bundles and improve their processibility with polymer matrix. Georgakilas et al. [48] observed that CNT covalently functionalized with pyrrolidone by 1,3-dipolar cycloaddition of azomethine ylides show a solubility of 50 mg/mL in chloroform, even without sonication whereas the pristine CNT is completely in‐ soluble in this solvent. Liang et al. [49] performed reductive alkylation of CNTs using lithi‐ um and alkyl halides in liquid ammonia for sidewall functionalization of CNTs and observed their extensive debundling by inspection of HRTEM images. Kinloch et al. [50] studied the rheological behavior of oxidized CNTs and found that the composites filled with functionalized CNTs had better dispersion. It has been observed by the researchers that amine modified CNTs is very important for the enhancement in the mechanical properties with epoxy. Garg et al. [31] shows the reaction mechanism for the formation of acid func‐ tionalized and amine functionalized CNT and their interaction with the epoxy resin as Fig‐ ure 8a and b respectively. Two different types of functional groups were attached on the CNT surface. In the first case MWCNTs were refluxed for 48 h in HNO3 (400 ml, 60% con‐ centration) to achieve reasonable surface oxidation of the tubes. The mixture was then fil‐ tered and the residue (treated material) was washed several times with distilled water till washings were neutral to pH paper. The treated MWCNT were dried in oven before use. In a second step the oxidized nanotubes were dispersed in benzene by stirring, and then re‐ fluxed with excess SOCl2 along with a few drops of DMF used as catalyst for chlorination of MWCNT surfaces. After the acyl chlorination, SOCl2 and DMF were removed through re‐ peatedly washing by tetrahydrofuran (THF). 100 ml of triethylene tetra-amine (TETA) was added to react with acyl chlorinated MWCNT at 100 °C for 24 h reflux until no HCl gas evolved. After cooling to room temperature, MWCNTs were washed with deionized water 5 times to remove excess TETA. Finally, the black solid was dried at room temperature over‐ night in vacuum and named as amine modified CNTs. These functionalized CNTs were characterized by FTIR, TGA and HRTEM and clearly showed the presence of these types of functional group. Gojny et al. [51]achieved surface modified MWCNTs by refluxing of oxi‐ dized MWCNTs with multifunctional amines and observed from TEM images that these were completely covered by epoxy matrix that confirmed the bonding between them. Sin‐ nott [52]has provided an in depth review of chemical functionalization of CNTs where the chemical bonds are used to tailor the interactions between nanotubes and polymers or sol‐ vent. The chemical functionalization of CNTs has also been accomplished through irradia‐ tion with electrons or ions [53]. In this manner one may hope to improve the binding of CNTs by interdigitation of active sites on its sides into polymer matrix.

decades, it is well established that the structure and properties of filler-matrix interface plays a major role in determining the structural integrity and mechanical performance of composite materials. CNTs have atomically smooth non-reactive surfaces and as such there is a lack of interfacial bonding between the CNT and the polymer chains that limits load transfer. Hence the benefits of high mechanical properties of CNTs are not utilized properly. The first experimental study focusing on interfacial interaction between MWCNT and poly‐ mer was carried out by Cooper et al. [46]. They investigated the detachment of MWCNTs from an epoxy matrix using a pullout test for individual MWCNT and observed the interfa‐ cial shear stress varied from 35-376 MPa. This variation is attributed to difference in struc‐

There are three main mechanisms for load transfer from matrix to filler. The first is weak van der Waal interaction between filler and polymer. Using small size filler and close con‐ tact at the interface can increase it. The large specific surface area of CNTs is advantageous for bonding with matrix in a composite, but is a major cause for agglomeration of CNTs. Therefore, uniformally dispersed individual nanotubes in matrix is helpful. The second mechanism of load transfer is micromechanical interlocking which is difficult in CNTs nano‐ composites due to their atomically smooth surface. Although local non uniformity along length of CNTs i.e. varying diameter and bends due to non hexagonal defects contributes to this micromechanical interlocking. This interlocking can increase by using long CNTs to block the movement of polymer chains. The contribution of this mechanism may reach satu‐ ration at low CNT content. The third and best mechanism for better adhesion and hence load transfer between CNTs and polymer is covalent or ionic bonding between them. The chemical bonding between CNTs and polymer can be created and enhanced by the surface treatment such as oxidation of CNTs with acids or other chemicals. This mode of mecha‐ nism have much importance as it provides strong interaction between polymer and CNT and hence efficiently transfers the load from polymer matrix to nanotubes necessary for en‐

The best route to achieve individual CNT to ensure better dispersion is chemical modifica‐ tions of CNT surface. The chemical functionalization involves the attachment of chemical bonds to CNT surface or on end caps. Nanotube functionalization typically starts with oxi‐ datative conditions, commonly by refluxing in nitric or sulfuric acid or combination of both to attach carboxylic acid moieties to the defect sites. The end caps of nanotubes have extra strain energy because of their high degree of curvature with pentagons and heptagonal car‐ bon atoms are most vulnerable to reaction with acid. The side walls also containing defects

are easily supplemented by oxidative damage and can be stabilized by formation of func‐ tional groups mainly carboxylic acid and hydroxide group. These acid moieties and hydrox‐ ide groups can be further replaced to more reactive groups like –COCl or –CORNH2. The addition of these functional groups on CNTs possesses intermolecular repulsion between functional groups on surface that overcomes the otherwise weak van der Waal attraction be‐

hybridized defects and vacancies in nanotube lattice and

ture and morphology of CNTs.

206 Syntheses and Applications of Carbon Nanotubes and Their Composites

hanced mechanical response in high-performance polymers.

**6.3. Chemical functionalization of CNTs**

like pentagon-heptagon pairs, sp3

Covalent bond also benefits phonon transferring between nanotubes and polymer matrix, which is a key factor for improving thermal conductivity of the nanocomposites. To ensure the adhesion between polymer and nanotubes various surfactant and chemical modification procedures have been adopted to modify the surface of otherwise inert surface of CNTs that provides bonding sites to the polymer matrix.

So the surface modification of CNTs is the crucial factor that decides the effective dispersion and improves the interactions between CNTs and matrix. However there are certain draw‐ backs of using chemically functionalized CNTs. Chemical functionalization normally em‐ ploys harsh techniques resulting in tube fragmentation and also disrupts the bonding between graphene sheets and thereby reduces the properties of CNTs. Studies revealed that different chemical treatments may decrease the maximum buckling force of nanotubes by 15% [16]. Also the chemical functionalized CNTs significantly decrease the electrical con‐ ductivity of CNTs nanocomposites due to unbalance polarization effect, shortening of length and physical structure defects during acidic treatment [54]. But it is still necessary for increased dispersion and strengthens the interfacial bonding of CNTs with polymer matrix

Carbon Nanotubes and Their Composites http://dx.doi.org/10.5772/52897 209

The solubility or dispersion of CNTs in certain specified solvents or polymers can also in‐ crease by non covalent association which is more fragile. The non-ionic surfactant such as sodium dodecylbenzene sulfonate (SDS) or polyoxyethylene-8-lauryl (PoEL) has two seg‐ ments. The hydrophobic segment of surfactant shows strong interactions with carbon of CNTs via van der Waal force and the hydrophilic segment shows hydrogen bonding with solvent or polymer used for dispersion. Islam et al. [55] reported that ~ 65% CNT bundles exfoliated into individual nanotubes even with a very low of 20 mg (CNT)/ml of water con‐ taining SDS as surfactant. Barrau et al. [56] used palmitic acid as surfactant to disperse CNTs into epoxy resin and observed that electrical percolation threshold decreases indicating bet‐ ter CNT dispersion. Gong et al. [57] added PoEL as surfactant in CNT/epoxy composite to assist the dispersion of CNTs. The improvement in dispersion in chitosan with nitric acid treated CNTs was also reported by Ozarkar et al. [58] and the stability of dispersion pre‐ pared by using functionalized CNTs was observed to be better. However, CNTs treated with different surfactants are wrapped in it and hence contacts between CNTs decreases thereby the transport properties (electrical and thermal conductivities) of CNTs/polymer

Dispersion of high loading of CNTs in any polymer is very difficult due to the formation of agglomerates by the conventional techniques. To maximize the improvment in properties, higher loading of CNTs is preferred [59]. However, polymer composites synthesized by us‐ ing the conventional methods generally have low CNT contents. It has been observed that beyond 1 wt.-% of loading, CNTs tend to agglomerate [60] resulting in poor mechanical properties of the composites. It is therefore important to develop a technique to incorporate higher CNT loading in the polymer matrices without sacrificing their mechanical properties. Recently, several methods have been developed for fabricating CNT/polymer composites with high CNT loadings. One such technique is mechanical densification technique where vertically aligned CNTs were densified by the capillary induced wetting with epoxy resin [61]. By this technique dimensions of sample preparation are limited. In another technique, a filtration system was used to impregnate the epoxy resin into CNT bucky paper [62, 63]. However, it was very difficult to completely impregnate the bucky paper with epoxy resin.

that is more important in structural applications.

nanocomposites are adversely affected.

**6.4. Dispersion of high loading of CNTs in polymer matrix**

**Figure 8.** Reaction mechanism for (a) acid modified-MWCNT (b) amine modified-MWCNT with epoxy resin (Reprinted with permission from Springer(31))

So the surface modification of CNTs is the crucial factor that decides the effective dispersion and improves the interactions between CNTs and matrix. However there are certain draw‐ backs of using chemically functionalized CNTs. Chemical functionalization normally em‐ ploys harsh techniques resulting in tube fragmentation and also disrupts the bonding between graphene sheets and thereby reduces the properties of CNTs. Studies revealed that different chemical treatments may decrease the maximum buckling force of nanotubes by 15% [16]. Also the chemical functionalized CNTs significantly decrease the electrical con‐ ductivity of CNTs nanocomposites due to unbalance polarization effect, shortening of length and physical structure defects during acidic treatment [54]. But it is still necessary for increased dispersion and strengthens the interfacial bonding of CNTs with polymer matrix that is more important in structural applications.

The solubility or dispersion of CNTs in certain specified solvents or polymers can also in‐ crease by non covalent association which is more fragile. The non-ionic surfactant such as sodium dodecylbenzene sulfonate (SDS) or polyoxyethylene-8-lauryl (PoEL) has two seg‐ ments. The hydrophobic segment of surfactant shows strong interactions with carbon of CNTs via van der Waal force and the hydrophilic segment shows hydrogen bonding with solvent or polymer used for dispersion. Islam et al. [55] reported that ~ 65% CNT bundles exfoliated into individual nanotubes even with a very low of 20 mg (CNT)/ml of water con‐ taining SDS as surfactant. Barrau et al. [56] used palmitic acid as surfactant to disperse CNTs into epoxy resin and observed that electrical percolation threshold decreases indicating bet‐ ter CNT dispersion. Gong et al. [57] added PoEL as surfactant in CNT/epoxy composite to assist the dispersion of CNTs. The improvement in dispersion in chitosan with nitric acid treated CNTs was also reported by Ozarkar et al. [58] and the stability of dispersion pre‐ pared by using functionalized CNTs was observed to be better. However, CNTs treated with different surfactants are wrapped in it and hence contacts between CNTs decreases thereby the transport properties (electrical and thermal conductivities) of CNTs/polymer nanocomposites are adversely affected.

#### **6.4. Dispersion of high loading of CNTs in polymer matrix**

**Figure 8.** Reaction mechanism for (a) acid modified-MWCNT (b) amine modified-MWCNT with epoxy resin (Reprinted

with permission from Springer(31))

208 Syntheses and Applications of Carbon Nanotubes and Their Composites

Dispersion of high loading of CNTs in any polymer is very difficult due to the formation of agglomerates by the conventional techniques. To maximize the improvment in properties, higher loading of CNTs is preferred [59]. However, polymer composites synthesized by us‐ ing the conventional methods generally have low CNT contents. It has been observed that beyond 1 wt.-% of loading, CNTs tend to agglomerate [60] resulting in poor mechanical properties of the composites. It is therefore important to develop a technique to incorporate higher CNT loading in the polymer matrices without sacrificing their mechanical properties. Recently, several methods have been developed for fabricating CNT/polymer composites with high CNT loadings. One such technique is mechanical densification technique where vertically aligned CNTs were densified by the capillary induced wetting with epoxy resin [61]. By this technique dimensions of sample preparation are limited. In another technique, a filtration system was used to impregnate the epoxy resin into CNT bucky paper [62, 63]. However, it was very difficult to completely impregnate the bucky paper with epoxy resin. Recently, Feng et al., [64] reported a mixed curing assisted layer by layer method to synthe‐ size MWCNT/epoxy composite film with a high CNT loading from ~15 to ~36 wt.-%. The mixed-curing-agent consists of two types of agents, one of which is responsible for the parti‐ al initial curing at room temperature to avoid agglomeration of the CNTs, and the other for complete curing of epoxy resin at high temperature to synthesize epoxy composite films with good CNT dispersion. In another study by Feng et al. [65] upto ~39.1 wt. % SWCNTepoxy composites were fabricated using same mixed curing layer by layer method and their mechanical properties were enhanced significantly. Bradford et al. [66] reported a method to quickly produce macroscopic CNT composites with a high volume fraction upto 27% of mil‐ limeter long, well aligned CNTs. Specifically, they used the novel method, shear pressing, to process tall, vertically aligned CNT arrays into dense aligned CNT preforms, which are sub‐ sequently processed into composites. In another study by Ogasawara et al. [67] aligned MWCNT/epoxy composites were processed using a hot-melt prepreg method. Vertically aligned ultra-long CNT arrays (forest) were converted to horizontally aligned CNT sheets by pulling them out. An aligned CNT/epoxy prepreg was fabricated using hot-melting with B-stage cured epoxy resin film. The final composites contains 21.4 vol% of MWCNTs.

duced by cutting with a diamond knife, however no quantitative mechanical measurements were made. The first true study for tensile and compression properties of CNT polymer composites was carried by Schadler et al. [75] with epoxy. On addition of 5-wt% MWCNTs the tensile modulus increased from 3.1 GPa to 3.71 GPa. and compression modulus in‐ creased from 3.63 to 4.5 GPa. However, no significant increases in toughness values were observed. Bai et al. [76] observed doubling of Young's modulus from 1.2 to 2.4 GPa and sig‐ nificant increase in strength from 30 to 41 MPa on addition of 1 wt.% MWCNTs. Also excel‐ lent matrix–nanotube adhesion was confirmed by the observation of nanotube breakage during fracture surface studies. Zhou et al. [77] reported steady increase of flexural modulus in CNT-epoxy composite with higher CNT weight percentage and found an improvement of 11.7% in modulus with 0.4 wt% loading of CNTs and 28% enhancement in flexural strength with 0.3 wt% loading. Garg et al. [31] reported an increment of 155 % in flexural strength of epoxy with addition of merely 0.3% amine functionalized MWCNTs and an increment of 38% in flexural modulus. Mathur et al. [13] reported an increment of 158% in flexural strength of phenolic with addition of 5 vol% of MWCNT. Colemann et al. [16] reviewed the mechanical properties of a large number of CNT reinforced polymer (thermoplastic and thermosetting) composites fabricated by various methods and reported enhancement in me‐ chanical properties. Du et al. [78] studied the experimental results for mechanical perform‐ ance of CNTs nanocomposites carried out by different research groups and observed that the gains are modest and far below the simplest theoretical estimates. Haggenmueller [79] applied the Halpen Tsai composite theory to CNT nanocomposites and observed that the ex‐ perimental elastic modulus is smaller than predicted by more than one order. It is attributed to the lack of perfect load transfer from nanotubes to matrix due to non uniform dispersion and small interfacial interaction. Although chemical functionalization of CNTs has sorted out those problems to an extent yet the best results have to be achieved. Also aspect ratio is other source of uncertainty in mechanical properties. Defects on the CNT surface also ex‐ pected to influence the mechanical properties significantly. The methods of handling nano‐ tubes, including acid treatments and sonication for long time are known to shorten nanotubes results in decreasing aspect ratio and are detrimental to mechanical properties. The mechanical properties of CNT based composites increased upto a certain loading of CNTs and beyond which it starts decreasing. This may be because of increase in viscosity of polymers at higher CNTs loading and also cause some surface of CNTs not to be completely covered by polymers matrix due to the large specific surface area of CNTs.Therefore, few studies have been carried out to disperse high loading of CNTs in polymer matrices with improved mechanical properties. Bradford et al. [66] reported 400 MPa (tensile strength) and 22.3 GPa (tensile modulus) for 27vol% of MWCNT–epoxy composites. Feng et al. [80] also reported 183% and 408% improvement in tensile strength and tensile modulus respec‐ tively at 39.1 wt% SWCNTs loading compared with those of the neat epoxy. Ogasawara et al. [67] found 50.6 GPa and 183 MPa modulus and ultimate tensile strength respectively of CNT (21.4 vol.%)/epoxy composite. These values were 19 and 2.9 times those of the epoxy

Carbon Nanotubes and Their Composites http://dx.doi.org/10.5772/52897 211

Andrews et al. [80] prepared aligned CNT/pitch composites and found the significant im‐ provement in the electrical and mechanical properties especially due to orientation of CNTs.

resin respectively.

#### **6.5. Alignment of CNTs in polymer matrix**

The other key challenge is to understand the effect of nanotube alignment on nanocompo‐ sites properties because the nanotubes have asymmetric structure and properties. Like other one-dimensional fiber fillers CNTs displays highest properties in the oriented reinforced di‐ rection and the mechanical, electrical, magnetic and optical performance of its composites are linked directly to their alignment in the matrix. So to take the full advantage of excellent properties of CNTs these should be aligned in a particular direction. For example, the align‐ ment of CNT increases the elastic modulus and electrical conductivity of nanocomposites along the nanotube alignment direction.

Several methods like application of electric field during composite formation and carbon arc discharge [68], composite slicing [69], film rubbing [70], chemical vapor deposition [71, 72], mechanical stretching of CNT-polymer composites [73] and magnetic orientation [74] have been reported for aligning nanotubes in composites. Electrospinning is also an effective method for the alignment of CNTs in polymer matrix.
