Printer Friendly

Mechanical reinforcement of polyethylene using n-alkyl group-functionalized multiwalled carbon nanotubes: effect of alkyl group carbon chain length and density.

INTRODUCTION

As one of the most produced synthetic resins, polyolefins are used in a diverse range of applications. To further enhance their performance or empower them with new properties, various nanoscale fillers have been used to prepare polyolefin nanocomposites (1). Among them, multi-walled carbon nanotubes (MWCNTs), which are rolled-up cylinders of one-atom-thick carbon layers with several to tens of nanometers in diameter and varied length from nanometers to centimeters, are regarded as an excellent candidate due to their unique combination of mechanical and electrical properties. Using carbon nanotubes (CNTs) as nanoscale fillers, a new generation of strong and multifunctional CNT reinforced polymer nanocomposites are being realized (2-8). Particularly, many studies have been focused on using the exceptional mechanical properties of MWCNTs for the mechanical reinforcement of polyethylene (PE) composites (9-16). Upon incorporation of CNTs, PE composites commonly show improved modulus and tensile strength, however, their ductile properties, such as the elongation at break and the toughness, are often reduced (9), (10). This has been attributed to the poor dispersion of CNTs in polymer matrix and inefficient interfacial stress transfer (4). CNTs usually aggregate and form large bundles by themselves due to the strong van der Waals force among them. And the weak interaction between the hydrophobic carbon surface of CNTs and PE strands limit the external stress transfer from PE to CNTs.

Both physical and chemical approaches have been explored to enhance MWCNT dispersion and stress transfer in PE composites. Physical processing methods, such as the high energy ball-milling (16), melt-blending (17), electrostatic spraying (18), and solution-phase mixing coprecipitation 1191, have showed some improvements. Chemical approaches usually display better results as compared with physical processing methods. Polyethylene glycol (PEG) was used as a noncovalent dispersant to improve the dispersion of MWCNTs in PE matrix, however, the weak tensile properties of PEG lead to reduced mechanical properties (20). Covalent nanotube sidewall functionalization not only provides repulsion forces among grafted functional groups to minimize nanotube aggregation but also improves interfacial stress transfer through the interaction between PE matrices and grafted functional groups (21). CNTs grafted with macromolecules, such as PE, by both in situ polymerization (22) and anhydride-amine covalent reaction (12) have been used as reinforce fillers, and the resulted polyolefin composites show improved tensile properties while their ductility is retained or even improved. However, in situ polymerization normally requires a moisture and oxygen free reaction environment; furthermore, anhydride-amine covalent reactions show limited control over the bond formation and the density of grafted molecules. Compared with macromolecules, it is easier to controllably graft small molecules, such as 3-aminopropyltriethoxysilane, 2-hydroxyethyl triphenyl phosphonium, 1,1,1,3,3,3-hexafluoro-2-phenyl-2-propanol, and n-alkyl groups, on MWCNTs (23-26). Among various small molecules, n-alkyl groups are especially attractive because of their simple structures similar to those of PE. Previous studies showed that the modulus of medium-density polyethylene reinforced by I wt% of--[C.sub.11][H.sub.23] group functionalized single walled CNTs was improved by 28.3% (26). The polypropylene composite prepared using dodecyl-grafted MWCNTs showed an increase of 34.4% in its Young's modulus (25).

In order to further develop functionalized MWCNTs as efficient fillers in PE composites, we need answer two important questions. First, what is the optimum length of alkyl groups for efficient mechanical reinforcement; and second, what is the optimum density of grafted alkyl groups on CNT surface. These two effects are likely interplaying together to affect the resulting PE composites. In this work, n-alkyl groups of different lengths were grafted on MWCNTs by amidation reactions. These functionalized MWCNT fillers were incorporated into linear low-density PE (LLDPE) at different loadings by a methanol-assisted coprecipitation method. The modulus, tensile strength, ultimate strain, and toughness of resulting PE composites were measured. A direct correlation between the density and the alkyl chain length of grafted n-alkyl groups and the mechanical reinforcement of PEs by the functionalized MWCNTs was showed. These results provide useful guidelines for developing CNT based nanocomposites.

EXPERIMENTAL

Functionalization of MWCNTs

MWCNTs of -40 nm in diameter (FloTube 9000, CNano Technology) were first purified by refluxing in 37% HCI (Sigma) for 2 h to remove metal residues. Afterwards, the nanotubes were collected by vacuum filtration, and oxidized at 400 C in air flow to remove amorphous carbons. The purified MWCNTs were denoted as "p-CNT". As shown in Fig. I, different numbers of carboxylic groups were added on MWCNTs by refluxing in 68% HNO3 (Merck) for 1-9 h (denoted as "n-CNT", where n is the time of acid treatment). MWCNTs functionalized with carboxylic groups were recovered by filtration using a polyvinyli-dene fluoride (5 pm pore size, Millipore) membrane filter, and then washed by deionized water. NaOH (Merck) aqueous solution (0.2 M) was used to further remove small car- bonaceous fragments adsorbed on nanotube surfaces (27). Then, MWCNT samples were washed by deionized water several times to remove the base before dried in oven. The density of carboxylic groups was determined by a photoluminescence (PL) method (28), (29), described in the Supporting Information. Briefly, 4-(9-anthroyloxy) phenacyl bromide (PB; Invitrogen) was used as a fluorescence probe to react with carboxylic groups on MWCNTs. The change of PL intensity of PB solutions before and after the reaction with MWCNTs was used to calculate the density of carboxylic groups.

bonaceous fragments adsorbed on nanotube surfaces (27). Then, MWCNT samples were washed by deionized water several times to remove the base before dried in oven. The density of carboxylic groups was determined by a photoluminescence (PL) method (28), (29), described in the Supporting Information. Briefly, 4-(9-anthroyloxy) phenacyl bromide (PB; Invitrogen) was used as a fluorescence probe to react with carboxylic groups on MWCNTs. The change of PL intensity of PB solutions before and after the reaction with MWCNTs was used to calculate the density of carboxylic groups.

The alkyl groups were grafted by amidation reaction. The dried 1-CNT or 3-CNT samples were dispersed in dimethylformamide (DMF; Sigma) by bath sonication, and then N,[N.sup.1]-dicyclohexylcarbodiimide (DCC; Sigma) and 4-dimethyla-minopyridine (DMAP; Sigma) were added as a coupling agent and a catalyst, respectively. Afterwards, n-alkyl amines with different carbon chain lengths were added, including n-propylamine (n-[C.sub.3][H.sub.2]--N[h.sub.2], forming C3-3-CNT), n-hexyl-amine (n-[C.sub.6][H.sub.13]--N[H.sub.2], forming C6-3-CNT), n-decylamine (n-[C.sub.10][H.sub.21]--N[H.sub.2], forming Cl0-3-CNT), n-tetradecylamine (n-[C.sub.14][H.sub.29]--N[H.sub.2], forming C14-3-CNT or C14-1-CNT), and n-octadecylamine (n-[C.sub.18][H.sub.37]--N[H.sub.2], forming C18-3-CNT). The reactions were carried out for 8 h. Finally, the functionalized MWCNTs were recovered by filtration, washed with DMF and methanol, and dried in a vacuum oven.

Preparation of MWCNT--PE Composites

LLDPE (melting flow index is 1 g/10 min (190[degrees]C/2.16 kg), [rho] = 0.918 g/c[m.sup.3], Sigma-Aldrich) was used to prepare composites by the methanol-assisted coprecipitate method. MWCNTs were first suspended in 1,2-dichlorobenzene (ODCB; 99%, Sigma, Singapore) by tip sonication (VCX-130, Sonics) at the concentration of 0.5 mg/mL. PE pellets were then dissolved in hot ODCB, and mixed with the desired volume of MWCNT/ODCB suspensions under bath sonication to prepare composites with different MWCNT weight loadings. The hot MWCNT and PE mixtures in ODCB were poured into excessive amount of cold methanol, resulting in quick precipitation of MWCNT--PE composites. The composites were recovered by vacuum filtration, and washed by ether and methanol before they were dried in vacuum oven overnight.

Characterization of MWCNT--PE Composites

The [N.sub.2] adsorption experiments of p-CNT, 1-CNT, and 3-CNT samples were carried out using a surface area analyzer (Autosorb-6B, QuantaChrome), and their specific surface area (SSA) was calculated by the Brunauer--Emmett--Teller method. Before [N.sub.2] adsorption measurement, 1-CNT and 3-CNT samples were annealed in He (Soxal, Alphagaz1) at 1000[degrees]C for 15 min to eliminate the conglomeration of nanotubes due to surface functional groups (30). Fourier-transform infrared (FTIR) spectra of MWCNTs mixed with potassium bromide (99.9%, Alfa) were collected on a FTIR spectrometer (Perkin Elmer Nicolet). Thermal gravimetric analysis (TGA) of MWCNTs was performed on a thermal analyzer (Perkin Elmer Diamond). The heating rate was set to lO[degrees]C/min. Raman spectra of MWCNTs and PE composites were obtained by a Raman microscope (Renishaw inVia) in a backscattering configuration using a 514-nm laser. Scanning electron microscopic images of MWCNTs and composites were recorded on a field-emission scanning electron microscopy (SEM; Joel JSM6701F). The cross-sections of composites were exposed by liquid [N.sub.2] cooling and cracking. Differential scanning calorimeter (DSC) characterization was performed on a DSC instrument (Mettler Toledo DSC822e). At least three specimens were tested for each composite sample. The dried composites were hot pressed in a mold, and then cut into thin strips (50 mm x 5 mm x 0.2 mm). The tensile properties of composites were tested on a mechanic testing machine (Instron 5543) operated at an extension rate of 50 mm/ min. At least five specimens were tested for each composite sample.

RESULT AND DISCUSSION

Purification and Functionalization of MWCNTs

We first examined the physiochemical properties of functionalized MWCNTs. HN[O.sub.3] treatment introduces defect sites and functional groups on nanotube surfaces. Heat treatment in He at 1000[degrees]C was performed to eliminate functional groups, which would hinder the access of N2, resulting in errors in the measured SSA. The SSA of p-CNT was 217.3 [m.sup.2]/g, while that of 1-CNT and 3-CNT increased to 255.7 and 322.6 [m.sup.2]/g, respectively. The increase in surface area can be attributed to the opening of tube ends and the exposure of nanotube inner spaces. The density of functional groups needs to be accurately quantified in order to determine the density of grafted n-alkyl groups. Using the PL method (28), (29), the density of carboxylic groups on MWCNT samples is listed in Table 1. There are small numbers of carboxylic groups on the purified nanotubes, which may be introduced during the HCI purification. The density of carboxylic groups increased more than 20 times after 1 h of HN[O.sub.3] treatment. It reaches 0.657 [mu]mol/mg after 3 h of HN[O.sub.3] treatment. The density would further increase ~50% when the acid treatment time is increased to 9 h. However, the recovery rate of MWCNTs would drop from 91 to 23 wt% when the acid treatment time increases from 1 to 9 h, because small carbonaceous fragments resulting from acid treatment are removed by filtration after NaOH wash, which leads to the loss of MWCNTs (27).

TABLE 1. Specific surface area, density of carboxylic
group, and the recovery rate after HN[O.sub.3] treatment
of MWCNT samples.

                                   Carboxylic
                            group density

Acid                  SSA   [micro]mol/mg  [micro]mol/[m.sup.2]
treatment   ([m.sup.2]/g)
time
(h)/(name)

0/(p-CNT)           217.3  0.013 [+ or -]  0.055 [+ or -] 0.050
                                    0.011

1/(1-CNT)           255.7     0.269 [+ or    1.052[+ or -]0.102
                                  -]0.021

3/(3-CNT)           322.6     0.657 [+ or   2.037 [+ or -]0.121
                                  -]0.039

6/(6-CNT)             N/A      0.825[+ or                   N/A
                                  -]0.036

9/(9-CNT)             N/A     0.983 [+ or                   N/A
                                  -]0.043

Acid        Recovery
treatment   rate (a)
time           (wt%)
(h)/(name)

0/(p-CNT)        N/A

1/(1-CNT)         91

3/(3-CNT)         82

6/(6-CNT)         53

9/(9-CNT)         23

(a) The recovery rate is defined
as the ratio between the dried MWCNTs before
and after undergoing HN[O.sub.3] treatment,
filtration, and NaOH wash.


Grafting of Alkyl Groups

The grafting of alkyl groups was investigated by FTIR spectroscopy and TGA in [N.sub.2]. The infrared (IR) spectra of three typical samples were shown in Fig. 2a. The broad peaks centered around 3400 c[m.sup.-1] on the spectra of p-CNT and 3-CNT samples come from --OH groups on CNT surfaces and/or adsorbed water. The C=C bond stretching leads to a small peak at 1560 c[m.sup.-1]. After acid treatment, the peak emerged at 1721 cm -I can be assigned as the carboxylic C=0 bond stretching, demonstrating that --COOH groups have been created on CNT surfaces (31). After amidation reaction, a few new features were observed on the IR spectrum of the C14-3-CNT sample. First, the peak at 1721 [m.sup.-1] shifts to 1643 c[m.sup.-1], which corresponds to the amide C=0 bond stretching, indicating the formation of amide bonds. Second, aliphatic C-H bond stretching mode at 2849 and 2920 c[m.sup.-1] can be attributed to the alkyl groups grafted on nanotube surfaces. Third, the intensity of the peak around 1500-1600 c[m.sup.-1] increases due to the overlap of the C=C stretching mode with the stretching vibrations of the secondary amide. Overall, FTIR results indicate the successful grafting of alkyl groups on CNT surfaces through amide bonds (32).

Typical TGA and DTG profile of C14-3-CNT sample were shown in Fig. 2b and c. When functionalized MWCNTs are heated in [N.sub.2], the removal of surface functional groups would result in gradual weight loss. Figure S4a in the Supporting Information shows the weight loss profiles of functionalized MWCNTs. The weight loss increases with the extension of oxidation time, which agrees with FTIR results, indicating more surface functional groups are introduced after longer acid treatment time. The typical weight loss and differentiate thermal gravimetric (DTG) profiles of alkyl-grafted MWCNTs (C14-3-CNT) are shown in Fig. 2b and c, respectively. There is a sharp peak centered at 340[degrees]C on the DTG profile, attributing to the removal of grafted alkyl groups by breaking amide bonds. Table 2 lists the quantity of alkyl groups grafted on MWCNTs, calculated by subtracting the weight loss profiles of functionalized MWCNTs from those of the corresponding alkyl group-grafted MWCNTs. Longer alkyl groups show larger weight loss. Moreover, more alkyl groups can be grafted on MWCNTs with higher carboxylic group density, for example, C14-3-CNT (8.37 wt%) versus C14-1-CNT (3.35 wt%). However, the grafting reaction efficiency only shows minor changes. Here, the grafting reaction efficiency is defined as the quantity of grafted alkyl groups divided by the molecular weight of the specific amine and the density of carboxylic groups on MWCNTs. Shorter alkyl amines (C3) have a slightly higher chance (61.78%) to be grafted compared with longer alkyl amines (C18 at 57.51%). This result also suggests that the surface density of carboxylic groups on the 3-CNT sample would not cause the repulsion among grafted alkyl chains, because the reaction efficiency shows insignificant changes.

TABLE 2. Weight loss and reaction
efficiency of n-alkyl amines of
different carbon chain lengths.

                                 The quantity of alkyl    Reaction
                                                groups  efficiency

Alkyl groups  [M.sub.w]  MWCNTs                  (wt%)         (%)
                (g/mol)

Propyl-, C3       59.11   3-CNT                   2.41       61.78

Hexyl-, C6       101.19   3-CNT                   4.09       61.24

Decyl-, C10      157.30   3-CNT                   6.08       58.57

Tetradecyl-,     213.40   l-CNT                   3.35       58.28
C14

                          3-CNT                   8.37       59.43

                          6-CNT                   9.03       51.27

                          9-CNT                  10.16       48.45

Octadecyl-,      269.52   3-CNT                  10.23       57.51
C18


Dispersion of MWCNTs in PE Composites

PE-MWCNT composites were prepared by the methanol-assisted coprecipitation method. SEM and Raman spectroscopy were used to evaluate the dispersion of MWCNTs in PE matrix. Figure 3a and b shows the cross-section of the C14-3-PE nanocomposites with 1 wt% nanotube loading. Few nanotube bundles were observed. The bright white spots (marked by the circles in Fig. 3b) are the broken ends of individual MWCNTs extending outside of the PE matrix, suggesting that MWCNTs are well dispersed in the PE matrix. Figure 3c and d shows Raman spectra of functionalized MWCNTs and PE composites. The peaks at 1100, 1250, and 1420 c[m.sup.-1] can be assigned to the phonon modes of PE (33). The strong peaks at 1320 c[m.sup.-1] come from the defect induced disordered D-band of MWCNTs. The peaks at around 1572 c[m.sup.-1] are the G-band of graphitic carbon materials. Figure 3d shows a clear upshift of 0-band peaks from 1572 to 1593-1594 c[m.sup.-1] when MWCNTs are incorporated in PE. The upshift of 0-band peaks can be attributed to the debundling of nanotubes (10), suggesting that individual nanotubes are dispersed in PE matrix (26). In contrast, the 0-band peak of the PE composite prepared using the 3-CNT samples shows less upshift (9-10 c[m.sup.-1] ), indicating that nanotubes in the 3-CNT sample form large bundles in PE matrix. Last, the 0-band peaks of PE composites from C3-3-CNT and C14-3-CNT are identical, suggesting that MWCNTs grafted with alkyl groups of different lengths all have good dispersion in PE matrix.

Mechanical Properties of PE Composites

The typical stress-strain curves of PE composites are shown in Fig. 4a. The measured yield strength ([[sigma].sub.y]), Young's modulus (n, tensile strength ([sigma]), ultimate strain ([[SIGMA].sub.max]), and toughness are summarized in Table 3. The changes in tensile properties of PE composites related to the neat PE are also shown in Fig. 4b and c. The modulus of PEs composites with 1 wt% p-CNT and 3-CNT increases by 11 and 4%, respectively. However, their ductile performances decrease significantly. Figure 4c shows that their ultimate strain decreases by 66 and 71%, and the toughness reduces by 66 and 72%, respectively. This agrees with previous reports, when pristine CNTs or acid treated CNTs were used as fillers (9), (10). In contrast, when alkyl group-grafted MWCNTs are used as fillers, their yield stress and Young's modulus improve significantly up to 54 and 63% for the C18-3-CNT fillers; at the same time, their ductile properties are preserved or improved by 30,--6, and 33% in terms of the tensile strength, ultimate strain, and toughness, respectively. In previous studies, the ductility of composites can only be retained or improved, when CNTs grafted with polymers are used as fillers (12), (34). In this study, we demonstrated that using MWCNTs grafted with short alkyl groups other than long chain polymers can also achieve similar ductile performance improvements.

TABLE 3. Tensile properties
of PE composites prepared
from different CNT samples at
1 wt% loading, percentage in
the parentless are the
properties change compared
with that of the neat PE.

   Filler     [[sigma].sub.y]   Y (MPa)      [sigma]
   type                 (MPa)                  (MPa)

1  PE          10.15 [+ or -]    202.53  14.51 [+ or
                    0.87 (0%)     [+ or      -] 0.83
                               -] 15.42         (0%)
                                   (0%)

2  p-CNT       11.07 [+ or -]    223.97  12.37 [+ or
                   1.01 (-9%)     [+ or      -] 0.97
                               -] 19.09       (-15%)
                                  (11%)

3  3-CNT       10.78 [+ or -]    210.75  11.97 [+ or
                    0.72 (6%)     [+ or      -] 1.52
                               -] 11.71       (-18%)
                                   (4%)

4  C3-3-CNT    11.95 [+ or -]    248.05  16.57 [+ or
                   0.93 (18%)     [+ or      -] 1.09
                               -] 16.37        (14%)
                                  (22%)

5  C6-3-CNT    12.89 [+ or -]    276.74  17.30 [+ or
                   0.95 (27%)     [+ or           -]
                               -] 16.22    0.96(19%)
                                  (37%)

6  CI0-3-CNT   13.75 [+ or -]    307.36  17.79 [+ or
                   1.28 (35%)     [+ or      -] 1.19
                               -] 21.75        (23%)
                                  (52%)

7  C14-3-CNT   15.15 [+ or -]    320.35  18.46 [+ or
                   0.89 (49%)     [+ or      -] 0.66
                               -] 14.76        (27%)
                                  (58%)

8  C18-3-CNT   15.64 [+ or -]    330.32  18.80 [+ or
                   0.77 (54%)     [+ or      -] 0.94
                               -] 19.61        (30%)
                                  (63%)

   Filler     [[epsilon].sub.max](mm/mm)      Thoughness
   type                                   (J/m[m.sup.3])

1  PE            6.82 [+ or -] 0.19 (0%)  71.41 [+ or -]
                                               6.37 (0%)

2  p-CNT       2.35 [+ or -] 0.41 (-66%)  24.34 [+ or -]
                                             3.21 (-66%)

3  3-CNT       1.95 [+ or -] 0.31 (-71%)  19.81 [+ or -]
                                            3.091 (-72%)

4  C3-3-CNT    5.17 [+ or -] 0.11 (-24%)  63.00 [+ or -]
                                            5.161 (-12%)

5  C6-3-CNT    5.38 [+ or -] 0.12 (-21%)  74.30 [+ or -]
                                               9.28 (4%)

6  CI0-3-CNT   5.59 [+ or -] 0.35 (-18%)  82.05 [+ or -]
                                             10.21 (15%)

7  C14-3-CNT   5.95 [+ or -] 0.22 (-13%)  89.19 [+ or -]
                                             8.831 (25%)

8  C18-3-CNT    6.39 [+ or -] 0.26 (-6%)  95.15 [+ or -]
                                             9.941 (33%)


Optimum Carbon Chain Length of Grafted Alkyl Groups

Comparing Samples 4-8 in Fig. 4 and Table 3, the composites produced from MWCNTs grafted with alkyl groups with longer carbon chains have better mechanical performances than those with shorter carbon chains. Because the same 3-CNT nanotube sample was used in these composites, the performance difference unlikely comes from the aspect ratio changes of nanotubes themselves. Furthermore, all grafted MWCNT samples are well distributed in PE matrix as shown by SEM and Raman spectroscopy (Fig. 3). The nanotube dispersion should not be the reason for the performance variation either. We propose that this is related to the carbon chain length of grafted alkyl groups. To develop MWCNTs grafted with alkyl groups as efficient fillers for PE composites, we need answer the two questions mentioned in the introduction part: what is the optimum length of alkyl groups and what is their optimum density.

To answer the first question, we produced PE composites using MWCNTs grafted with alkyl groups of different carbon chain lengths at 0.25, 0.5, and I wt% loadings. We first evaluated their tensile properties. Coleman et al. (4) have proposed to use the rate of increase of Young's modulus with the volume fraction of the filler dY/d[V.sub.f] to evaluate the mechanical reinforcement efficiency of poly-mer--nanotube composites. We calculated the volume fraction ([V.sub.f]) using the density of PE at 0.918 g/c[m.sup.3] (from supplier) and MWCNTs' density at 1.33 g/c[m.sup.3] (4). The correlation between Young's modulus and the volume fraction is shown in Fig. 5a. Young's modulus increase almost linearly with the increase in volume fraction for all tillers until their volume fraction reaches about 0.00354% (0.5 wt%). The dYld[V.sub.f] values were calculated at the 0.5 wt% filler loading. As shown in Fig. 4b, the dYld[V.sub.f] of composites using p-CNT is 3.71 GPa, similar to previous studies (12). It doubles for the composites using C3-3-CNT, and increases sharply with the extension of the grafted alkyl group carbon chain length. For the composites using C14-3-CNT, it is 17.06 GPa, which is more than 5 times higher than that of the composites using p-CNT. It increased slightly to 17.99 GPa for the composites using C18-3-CNT. Figure 5 shows that alkyl groups with longer carbon chains are more beneficial for the mechanical reinforcement efficiency of PE-nanotube composites.

We further compared the reinforcement efficiency based on the contribution of each carbon atoms in the alkyl chains. The values of dYld[V.sub.f] were divided by the number of the carbon atoms in their respective alkyl chains as an approximation of the "contribution per carbon atom". As shown in Fig. 5b, the maximum value (2.16 GPa) could be reached when the C6-3-CNT was used. This value decreases to 1.06 GPa, when the C18-3-CNT was used. With the increase in the carbon chain length, the modulus of composite increase, however, the reinforcement efficiency per carbon atom would decrease. This suggests the scenario that each carbon atoms in short alkyl chains are important for the stress transfer between nanotubes and PE matrix. However, for alkyl groups longer than C6, the contribution of additional carbon atoms in the chain to the interfacial adhesion with PE would gradually decrease.

PE is a typical semicrystalline polymer. The melting and recrystallization of PE composites might be affected by the interaction between filler molecules and PE matrix. Thus, we also evaluated the carbon chain length of alkyl groups by their influences on the melting and recrystallization of composites. DSC characterization of PE composites is shown in Fig. 6. The degree of crystallization ([X.sub.c]) is calculated by the following equation: [X.sub.c]= ([DELTA]H/[DELTA]H[degrees])X100 %-where [DELTA]H is the enthalpy of PE fusion obtained from its DSC curve. [DELTA]H[degrees] is the fusion enthalpy of pure crystalline PE at 289 J/g (35).

As shown in Fig. 6a, the [T.sub.m] of neat PE and composites are similar. The [T.sub.c], and [X.sub.c] of neat PE are 113.36[degrees]C and 46.64%, respectively. It increases to 115.23[degrees]C and 49.97% for the composite prepared using p-CNT. The increases can be attributed to the fact that nanotubes have low compatibility with PE matrix, and they are potential nucleating sites when PE molecules recrystallize (19). The [T.sub.c] and [X.sub.c] of composites produced using functional-ized MWCNTs decrease with the increase in alkyl group carbon chain length. When the carbon chain is longer than C14, the [T.sub.c], and [X.sub.c] values are very similar to those of neat PE, suggesting much better interaction between the long carbon chain alkyl groups and PE matrix. Based on the results shown in Figs. 5 and 6, also considering the reaction efficiency of grafting reactions (Table 2), we proposed that alkyl groups with the C14--C18 chain could be the optimum length to be grafted on MWCNTs as efficient fillers for PE composites.

Optinnun Density of Grafted Alkyl Groups

Last, we examined the optimum density of the grafted alkyl groups by using MWCNTs functionalized with different number of carboxylic groups. The quantity of C14 alkyl groups was determined by the TGA weight loss after heating to 900[degrees]C in [N.sub.2], as shown in Table 2. The grafted alkyl group density was calculated based on the carboxyl group density (Table 1) and the grafting reaction efficiency (Table 2). Figure 7 shows the Young's modulus and tensile strength of resulting PE composites plotted against the density of alkyl groups on MWCNTs. The mechanical properties increase with the increase in alkyl group density until it reaches about 0.424 [micro]mol/mg, and then drop sharply. The high alkyl group density samples are produced using 9-CNT samples with 9 h of acid treatment. The decrease in their mechanical properties could be attributed to the fact that these highly functionalized MWCNTs have reduced mechanical strength and aspect ratio. Raman analysis shows that they have much more defects with a larger D band. Although higher density of grafted alkyl groups could be beneficial to enhance the interaction between PE matrix and nanotubes, in order to graft more alkyl groups, MWCNTs must have higher density of carboxylic groups. The prolonged acid treatment could add more defects on nanotubes, which serve as potential break points when nanotubes are individualized by sonication or under composite processing (36), (37). Based on the carboxyl group density (Table 1) and the grafting reaction efficiency (Table 2), we estimated that the optimum density of grafted alkyl groups could be around 0.390-0.423 [micro]mol/mg when the C14 alkyl groups are used.

CONCLUSIONS

Carboxylic groups were created on MWCNTs by HN[O.sub.3] treatment. Their density was determined by a PL method varying from 0.013 to 0.983 [micro]mol/mg. Afterwards, n-alkyl groups with carbon chain ranging from C3 to C18 were grafted on MWCNTs by amidation reaction. The grafting reaction efficiency shows small changes between 48.45 and 61.78% among different n-alkyl groups. PE composites were fabricated using the methanol-assisted coprecipitate method. MWCNTs are well dispersed in PE matrix as demonstrated by SEM and Raman analysis. MWCNTs grafted with short n-alkyl groups are effective fillers for mechanical reinforcement of PE composites. The yield stress and Young's modulus of PE composites improve significantly up to 54 and 63% as compared to neat PE, respectively. Among ductile properties, tensile strength increases up to 30%; ultimate strain is preserved; and toughness increases up to 33%. Alkyl group carbon chain length and density strongly affect the mechanical reinforcement of functional ized MWCNTs. By evaluating the increase rate of Young's modules versus the volume fraction of fillers (dY/d[V.sub.f]) and melting and recrystallization properties of PE composites, we found that n-alkyl groups with longer carbon chains are beneficial for the stress transfer between MWCNTs and PE. The contribution of individual carbon atoms in the carbon chains would decrease with the extension of chain length. Alkyl groups with the C14--C18 chains are at the optimum length to be grafted on MWCNTs as efficient fillers for PE composites. Moreover, the optimum density of grafted alkyl groups should be around 0.390-0.423 [micro]mol/mg when the C14 alkyl groups are used.

Additional Supporting Information may be found in the online version of this article.

Correspondence to: Yuan Chen; e-mail: chenyuan@ntu.edu.sg

Contract grant sponsor: National Research Foundation, Singapore; contract grant number: NRF-CRP2-2007-02; contract grant sponsor: Ministry of Education, Singapore; contract grant number: MOE2011-T2-2-062. DOI 10. 002/pen.23579

Published online in Wiley Online Library (wileyonlinelibrary.com). [c] 2013 Society of Plastics Engineers

REFERENCES

(1.) V. Mittal, Advances in Polyolefin Nanocomposites, CRC Press, Boca Raton, FL, USA (2011).

(2.) E.T. Thostenson, Z.F. Ren, and T.W. Chou, Compos. Sci. Technol., 61, 1899 (2001).

(3.) E.T. Thostenson, C. Li, and T.-W. Chou, Compos. Sci. Technol., 65, 491 (2005).

(4.) J.N. Coleman, U. Khan, W.J. Blau, and Y.K. Guniko, Carbon, 44, 1624 (2006).

(5.) M. Moniruzzaman and K.I. Winey, Macromolecules, 39, 5194 (2006).

(6.) Z. Spitalsky, D. Tasis, K. Papagelis, and C. Galiotis, Prog. Polym. Sci, 35, 357 (2010).

(7.) H. Gu, S. Tadakamalla, X. Zhang, Y. Huang, Y. Jiang, H.A. Colorado. Z. Luo, S. Wei, and Z. Guo, J. Mater. Chem. C, 1, 729 (2013).

(8.) Y. Li, J. Zhu, S. Wei, J. Ryu, Q. Wang, L. Sun, and Z. Guo, Macromol. Chem. Phys., 212, 2429 (2011).

(9.) K.Q. Xiao, L.C. Zhang, and I. Zarudi, Compos. Sci. Tech-nol., 67, 177 (2007).

(10.) T. McNally, P. Potschke, P. Halley, M. Murphy. D. Martin, S.E.J. Bell. G.P. Brennan, D. Bein, P. Lernoine, and J.P. Quinn, Polymer, 46, 8222 (2005).

(11.) Y.Z. Bin. M. Kitanaka, D. Zhu, and M. Matsuo, Macromolecules, 36, 6213 (2003).

(12.) B.X. Yang, K.P. Pramoda, G.Q. Xu, and S.H. Goh, Adv. Funct. Mater., 17, 2062 (2007).

(13.) P. Potschke, A.R. Bhattacharyya, and A. Janke, Polymer, 44, 8061 (2003).

(14.) M.O. Lisunova, Y.P. Mamunya, N.I. Lebovka, and A.V. Melezhyk, Eur. Polym. J., 43, 949 (2007).

(15.) S.L. Ruan, P. Gao, X.G. Yang, and T.X. Yu, Polymer, 44, 5643 (2003).

(16.) G. Gorrasi, M. Sarno, A. Di Bartolomeo, D. Sannino, P. Ciambelli, and V. Vittoria, J . Polym. Sci. Part B: Polym. Phys., 45, 597 (2007).

(17.) M.A.L. Manchado, L. Valentini, J. Biagiotti, and J.M. Kenny, Carbon, 43, 1499 (2005).

(18.) S.R. Bakshi, J.E. Tercero, and A. Agarwal, Compos. Part A: Appl. Sci. Manuf., 38, 2493 (2007).

(19.) R. Haggenmueller, J.E. Fischer, and K.I. Winey, Macromolecules, 39, 2964 (2006).

(20.) Y. Zou, Y. Feng, L. Wang, and X. Liu, Carbon, 42, 271 (2004).

(21.) M.E. Mackay, A. Tuteja, P.M. Duxbury, C.J. Hawker, B. Van Horn, Z. Guan, G. Chen, and R.S. Krishnan, Science, 311, 1740 (2006).

(22.) J. Kim, S.M. Hong, S. Kwak, and Y. Seo, Phys. Chem. Chem. Phys., 11, 10851 (2009).

(23.) J.-H. Ko and J.-H. Chang, Polym. Eng. Sci., 49, 2168 (2009).

(24.) J.-11. Lee, J. Kathi, K.Y. Rhee, and J.H. Lee, Polym. Eng. Sci., 50, 1433 (2010).

(25.) J.-H. Ko, C.S. Yoon, and J.-H. Chang, J. Polym. Sci. Part B: Poly'''. Phys., 49, 244 (2011).

(26.) X.P. Merlyn, V.K. Oleksandr, R.G.P. Ivana. W. Xin, and N.K. Valery. Nanotechnology, 20, 195602 (2009).

(27.) Z. Wang, M.D. Shirley, S.T. Meikle, R.L.D. Whitby, and S.V. Mikhalovsky, Carbon, 47, 73 (2009).

(28.) N. Dementev, R. Ronca, and E. Borguet, Appl. Surf Sc., 258, 10185 (2012).

(29.) N. Dementev, X. Feng, and E. Borguet, Langmuir, 25, 7573 (2009).

(30.) L. Lafi, D. Cossement, and R. Chahine, Carbon, 43, 1347 (2005).

(31.) U.J. Kim, C.A. Furtado, X. Liu, G. Chen, and P.C. Eklund, J. Am. Chem. Soc., 127, 15437 (2005).

(32.) T. Ramanathan, F.T. Fisher, R.S. Ruoff, and L.C. Brinson, Chem. Mater., 17, 1290 (2005).

(33.) F.J. Boerio and J.L. Koenig, J. Chem. Phys., 52, 3425 (1970).

(34.) R. Blake, Y.K. Gun'ko, J. Coleman, M. Cadek. A. Fonseca, J.B. Nagy, and W.J. Blau, J. Am. Chem. Soc., 126, 10226 (2004).

(35.) J.D. Hoffman and R.L. Miller, Polymer, 38, 3151 (1997).

(36.) J. Liu, A.G. Riniler, H. Dai, J.H. Hafner, R.K. Bradley, P.J. Boul, A. Lu, T. Iverson, K. Shelimov, C.B. Huffman, F. Rodriguez-Macias, Y.-S. Shon, T.R. Lee, D.T. Colbert, and R.E. Smalley, Science, 280, 1253 (1998).

(37.) I.D. Rosca, F. Watari, M. Uo, and T. Akasaka, Carbon. 43. 3124 (2005).

Li Wei, Wenchao Jiang, Kunli Goh, Yuan Chen

School of Chemical and Biomedical Engineering, Nanyang Technological University, Singapore 637459, Singapore
COPYRIGHT 2014 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Wei, Li; Jiang, Wenchao; Goh, Kunli; Chen, Yuan
Publication:Polymer Engineering and Science
Article Type:Report
Geographic Code:9SING
Date:Feb 1, 2014
Words:5534
Previous Article:Microcellular injection-compression molding (MICM): a novel technology for effectively improving cellular structure of polystyrene foams.
Next Article:Effect of Nanoclay on Physical, Mechanical, and Microbial Degradation of Jute-Reinforced, Soy Milk-Based Nano-Biocomposites.
Topics:

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |