Structure and properties of poly(ethylene oxide)-organo clay nanocomposite prepared via melt mixing.
In the last decade, the preparation and material properties of polymer/clay nanocomposites (PCN) have been a subject of considerable interest because of the property enhancement, which may be realized in these systems (1), (2). The nano-scale size affects the morphology of PCNs and their uniform dispersion can lead to a very large interfacial area. As a consequence, the bulk properties may be overridden by those of the interphase. Recently, the history of the discovery of polymer/clay nanocomposites has been highlighted by Kawasumi (3), and the mechanical and rheological properties of PCNs have been reviewed by others (4), (5). The main attractions of PCN are availability and low cost of clay and the well developed intercalation chemistry, which makes it possible to achieve nanostructure from a micron size filler. The basic principle behind the formation of PCN is that the polymer should intercalate into the galleries of the clay. When the polymer molecules intercalate into the silicate gallery, there is a decrease in entropy which is compensated for, by either the increase in entropy in the organic modifiers used to widen the gallery spacing or the enthalpic term arising from the intercalation of polymer and organic modified clay layers (1), (2), (5).
The strong hydrophilic nature of pristine clay surface results in a high interfacial tension with organic materials making the layered silicate difficult to intercalate and disperse homogeneously in polymer matrices. To overcome this problem, the hydrophilic silicate surface is normally converted to an organophilic one, making the intercalation of many hydrophobic polymers possible (6). Modification of pristine clay with long chain alkyl ammonium cations introduces hydrophobic character into the silicate gallery and increases the d-spacing, thereby facilitates the intercalation of hydrophobic polymers (6), (7). Most PCNs display an intermediate morphology between the intercalated and exfoliated (when the silicate layers are likely delaminated) nanostructure. The nanocomposite structure depends on the nature of polymer (7), (8), organic modifier (1), (8), (9), and processing conditions (9), (10). An effective exploitation of PCN requires an understanding of nanoscale structure-property-processing relationships.
Poly(ethylene oxide) (PEO) is a crystalline thermoplastic polymer with the general structure H[(--O--[CH.sub.2]--[CH.sub.2]).sub.n]--OH, where n is the number of repeat units. The polyether chain of PEO can strongly coordinate with alkali cations (Li+, Na+, Ca +2 etc.) leading to the formation of a solid polymer electrolyte for use in battery, super capacitor, and fuel cell applications (11-13). The main problem with the conventional PEO-based solid polymer electrolytes is the low ionic conductivity (< 10-5 [[ohm].sup.-1]/cm) which is not sufficient for many applications (12), (13). The incorporation of layered silicate is known to increase the conductivity of PEO-based electrolytes as the silicate layer act as anion and hence, the cation can preferentially move. Another important application of PEO is that it can act as a crystallizable switching component in a shape memory polymer system (14-16). However, major problem is that the related systems exhibit poor mechanical properties, which can likely be improved by incorporating layered silicates.
Since Aranda and Ruiz-Hitzky (17) reported the first preparation of PEO/clay nanocomposites (PEOCN) by solution intercalation, large volumes of works have been done in this area (18-23). Preparation of nanocomposites have been demonstrated by using pristine clay (in intercalant aqueous solution), organophilic clay with polar intercalents (18), (19) ([Closite.sup.R] 30B), and with apolar intercalents (20), (21) (Nanochore 1-30). PEO being a hydrophilic polymer and soluble in water, nanocomposites can be made using pristine clays (22), (23). Choi and coworkers (24) correlated the rheological properties with the mesoscopic structure and it was postulated that the molecular weight and interaction strength would affect the mesoscopic structure and the rheological properties of these nanocomposites. The relative mobility of bulk and interphase polymers in intercalated PEOCNs was determined using rheological experiments (24), (25) nuclear magnetic resonance (26) and neutron scattering (27). Chen et al. (28) have reported comparative sorption experiments for molar mass in PCN and the results show that high molar mass fractions of polymer intercalate preferentially into smectite clay during solution intercalation method. Loizou et al. (29) studied the dynamic responses of PEO nanocomposite hydrogel. It was established that low molecular weight PEO inhibits the aggregation of clay particles by the classic steric hindrance, whereas high molecular weight PEO functions as bridge between the particles, particularly at high clay concentration. This leads to the formation of large clusters, and to a smart gel with novel properties (30), (31). New methods are being developed for fabrication of molecularly ordered composites with hierarchical structure by self assembly (32-34).
The drawbacks in the solution method are the requirement of a suitable monomer/solvent or polymer/solvent pairs and the high costs associated with the solvents, their disposal, and their impact on the environment. Melt intercalation does not require the use of a suitable solvent. The direct melt-intercalation process involves annealing a mixture of the polymer and organically modified nanoclay above the softening point of the polymers, under shear. While annealing, the polymer chains diffuse from the bulk polymer melt into the galleries between the silicate layers leading to the formation of nanocomposites. This method has become the mainstream for the fabrication of PCNs in recent years because it is simple, economical, environmentally friendly, and easy to implement in current polymer processing techniques (35), (36).
Most of the works reported so far have utilized solution intercalation method to prepare PEOCN. On the other hand, there are only limited reports on PEOCN prepared via melt intercalation method (37-39). Shen et al. (38) reported the melt intercalation by annealing hydraulic pressed PEO/silicate pellets. Loyens et al. (39) demonstrated the intercalation via melt extrusion process using various types of Cloisite type clays and reported that Cloisite 30B displayed better exfoliation with highly improved storage moduli regardless of the matrix molar mass. We have used melt compounding in a mixer followed by compression molding for the preparation of PEOCN. This method is crucial for technological exploitation as electrodes for battery and super capacitors. However, considering that PEO easily degrades (40), (41), it is very important to standardize the processing parameters. To our knowledge there is no report in the literature on the optimization of processing parameters and rheological properties of melt intercalated PEOCN.
In this work we have prepared PEOCNs using melt mixing method, and studied the rheological properties, mechanical properties, and morphology. Characterization of PEOCNs with polarized optical microscopy (POM), X-ray diffraction (XRD), atomic force microscopy (AFM), and scanning electron microscopy (SEM) are discussed in the present article. The nanocomposite will find application not only in the solid polymer electrolyte but also for the shape memory polymers with improved mechanical properties.
The matrix PEO with weight average molecular weight of 100,000 g/mole was purchased from Acros Organics (Geel, Belgium). The organophilic clay (Cloisite 30B) was obtained from Southern Clay Products (Gonzales, TX). Organoclays are produced by ion exchange reaction, where usually the quaternary ammonium cations replace the sodium cations in between the galleries of the montmorillonite (MMT) type clays (18). The cationic exchange capacity (CEC) was 95 meq/100 g for all MMTs used in the Cloisite[R] series. Cloisite 30B is an MMT modified with methyl tallow bis-2-hydroxyethyl quaternary ammonium cations (MMT-30B) (data provided by the manufacturer).
The nanocomposites were prepared by melt mixing in a laboratory kneader (Type 50 of Brabender, Duisburg, Germany). The temperature of the mixing was kept above 90[degrees]C which ensured proper melting of PEO. The nanocomposites are denoted by considering the weight percentage of the clay (PEOCN3, PEOCN6, and PEOCN9 containing 3 wt%, 6 wt%, and 9 wt% MMT-30, respectively). For a better comparison, the pure PEO sample was also processed in the kneader to set an identical thermal history with that of the PEOCNs. Before mixing, the MMT-30 was dried by keeping in vacuum oven at 70 C for 48 h. A mixing time of 20 min was given at a rotor speed of 30 rpm. In all the cases, the torque stabilized to a constant value during this mixing time. The specimens for various testing were obtained by compression molding using hot press (EP-Stanzteil, Wallenhorst, Germany) at a temperature of 100[degrees]C.
Viscoelastic properties of PEOCNs were studied using a strain controlled rheometer (ARES of Rheometric Scientific, NJ) equipped with 25 mm diameter stainless steel parallel disks. Measurements were performed in oscillatory shear and step shear configurations at different temperatures. Test specimens were prepared by compression molding of the melt intercalated composite from the kneader at 100 [degrees] C for about 5 min in to 2-mm thickness and 25-mm diameter discs. In the linear viscoelastic measurements, the dynamic strain sweep measurements were carried out first to determine the linear region. In oscillatory shear experiments, a sinusoidal shear strain [gamma](t) = [[gamma].sub.0] sin([omega]t + [phi]) was imposed. In the frequency sweep measurements, [[gamma].sub.0] a small constant and frequency-dependent elastic modulus (G') and loss modulus (G") were determined.
Atomic Force Microscopy (AFM)
A Nanoscope III AFM (Digital Instruments, Santa Barbara, CA), equipped with a microfabricated silicon cantilever with an integral tip (Tapping Mode tip) was used to perform AFM imaging on the surfaces of the compression molded specimens, which were formed using two parallel glass plates to get a smooth surface.
X-ray Diffraction (XRD)
Structural information were gathered from XRD experiments in transmission mode using Ni-filtered Cu [K.sub.[alpha]] radiation ([alpha] = 0.1542 nm) by a D500 diffractometer of Siemens (Karlsruhe, Germany). The samples were scanned in step mode (5 s/step, step = 0.05 [degrees]) in the range of 2[theta] up to 12 [degrees]. The XRD spectrum of the MMT-30 powder was also recorded, however, in the reflection mode.
Scanning Electron Microscopy (SEM)
The morphology of PEOCNs was also investigated by a Zeiss Supra 40 VP SEM (Oberkochen, Germany). The compression molded samples were quenched in liquid nitrogen and cryogenically ruptured to obtain cross sections, which were sputter coated with carbon to avoid charging before the SEM observation.
Polarized Optical Microscopy (POM)
The spherulite formation during crystallization of PEO and PEOCNs were observed using a polarized optical microscopy (Leica, Wild Leitz Gmbh, Ulm, Germany) equipped with a hotstage unit (TMS 91, Linkam Scientific-instruments, Waterfield, England). Thin specimens (layer thickness of ~40 [micro]m) were cut using a microtome. Crystallization was observed while the samples were exposed to following temperature scans: heating at a rate of 10[degrees]C/min to 100[degrees]C, holding for 5 min to erase the thermal history effects, and then cooling to 40[degrees]C at a slow cooling rate of 2[degrees]C/min, during which the crystallization took place. The spherulites were viewed between crossed polarizers.
Dynamic Mechanical Analysis (DMA)
Dynamic mechanical analysis (DMA) was performed on PEOCNs. Specimen were cut from the compression molded sheets with dimensions of 20 mm X 8 mm X 1 mm (length X width X thickness) and tested in a Q800 DMTA instrument (TA Instruments, New Castle) operating in tensile testing mode. The test specimen was cooled to--120[degrees]C, allowed to stabilize, and then heated at a rate of 3[degrees]C/min to room temperature. The frequency of oscillation was fixed at 1 Hz. Storage modulus (E'), loss modulus (E"), and mechanical loss factor (tan [delta]) were determined during the test and plotted against the temperature.
The tensile properties of the samples were determined using dumbbell shaped specimens (S3A type) with a Zwick 1474 universal testing machine (Ulm, Germany) according to DIN 53504 test procedure. The length between the jaws was fixed to 40 mm and at least five parallel measurements were done to conclude the mean values. Tests were run at room temperature at a crosshead speed of 1 mm/min and the related modulus and strength values were determined.
RESULTS AND DISCUSSION
Dynamic Viscoelastic Measurements
Owing to the necessity for understanding the nanocomposite structure and the influence of various shear and thermal environments on polymer nanocomposites, the rheological behavior of nanocomposites has received a considerable attention in recent years (5), (9), (24). The degree of polymer-filler interactions and the structure-property relationship in PEO and PEOCNs can be obtained from viscoelastic measurements which entirely depend on the extent of intercalation of polymer into the galleries of the clay. It is well known that high process temperature favors intercalation (9), (10). However, at high temperature under stress condition, PEO is susceptible to thermal and mechanical degradation (40), (41). To optimize the processing temperature, we have measured the viscoelastic properties of PEOCNs with 3 wt% MMT-30 prepared at three different temperatures (90[degrees]C, 100[degrees]C, and 110[degrees]C, Fig. 1). The viscosity and storage modulus data are found to be more or less same at 90[degrees]C and 100[degrees]C. However, there is a notable decrease in the storage modulus and viscosity for the sample mixed at 110[degrees]C. This can be attributed to the degradation of the PEO at high temperature. Even though the mixing temperature was kept as 110[degrees]C, the actual temperature of the PEO melt in the mixing chamber might be higher due to the high shear involved in the mixing. Because the composite is quite stable up to a set mixing temperature 100[degrees]C, the same temperature was used for the processing of all the compositions.
[FIGURE 1 OMITTED]
[tau] = k[[gamma].sup.n] (1)
Figure 2 shows the shear stress vs. shear rate plots at different temperatures for PEOCNs composites. The data from the figure were accurately fitted by a power law equation. where [tau] is the applied shear stress, [gamma] is the shear rate, k is a constant, and n is the power law index. It was noticed that the power law index, n, increases with the increasing temperature. The power law indices obtained from the data for different PEOCNs, with different nanoclay concentrations processed at different temperatures, are summarized in Table 1. For the PEOCNs, the value of n decreases with the increase in the clay content and shows lower temperature dependence at higher clay concentrations, as expected (42). Although the exact mechanism that causes the decrease in n value is not clear, we believe that it is due to the orientation of the silicate layers under shear. With the increasing shear rate, the intercalated chain conformations are also expected to change as the coils align parallel to the flow.
[FIGURE 2 OMITTED]
TABLE 1. Power law index data for PEOCNs at different temperature. Flow behavior index Sample 80[degrees]C 100[degrees]C 120[degrees]C PEOCN3 0.550 0.589 0.622 PEOCN6 0.450 0.472 0.480 PEOCN9 0.408 0.412 0.392
Figure 3 shows the viscosity of PEO and PEOCNs measured at different temperatures. The curves are typical pseudo plastic in nature, that is, when the temperature is kept constant, the shear viscosities decrease with respect to the shear rate. As expected, viscosity decreases with the increasing temperature. The effect of temperature on the shear viscosity is generally explained by the fact that an increase in temperature leads to an internal more thermal motion of the molecules and generate greater free volume in the polymer, which decreases intermolecular or intermolecular resistances. The storage modulus of the PEOCNs is also decreased with the increasing temperature as shown in Fig. 4, which means that the material has a lower relaxation time as the temperature increases (43), (44).
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
It is also evident from Fig. 3 that the viscosities of the nanocomposites at low shear rate are much higher than that of the pure PEO. The enhancement in viscosity is often explained in terms of confinement of polymer chain within the silicate layer (45), (46). The viscosity of the confined polymer melt is always greater than that of the bulk. The higher viscosity of the confined melt is believed to arise from an "immobilized hydrodynamic layer" near the wall having a thickness in the order of radius of gyration of the polymer chain. According Semenov and Doi (45) for shear flow of melts confined between parallel plates separated by a distance h/a ~ [N.sup.1/2], where a is the segment length and N is the number of segments in the polymer chain, the relative zero-shear viscosity of the confined melt ([[eta].sub.0.sup.c]) with respect to the bulk viscosity ([[eta].sub.0.sup.b]) can be given as
[[[eta].sub.0.sup.c]/[[eta].sub.0.sup.b]][approximately equal to] [a/h] [[[lambda].sup.c]/[[lambda].sup.b]][approximately equal to]([[h/a].sup.2]) for [[[lambda].sup.c]/[[lambda].sup.b]][approximately equal to] [(h/a).sup.3] (2)
Here, [[lambda].sup.c] and [[lambda].sup.b] are the segment relaxation times of the confined and bulk polymer chains, respectively. Thus, the scaling model predicts that [[lambda].sup.c] [much greater than] [[lambda].sup.b] and if the separation distance were of the order of radius of gyration of the polymer chain, then the zero-shear viscosity of the confined system would scale as N, which is about 50 to 100 in case of PEO. Hence, the viscosity of the intercalated melt can be several times higher than that of the bulk melt as reflected in our viscosity data measured at low shear rate.
As the shear rate increases, these nanocomposites display more drastic shear-thinning (viscosity decreases with shear rate) behavior with clay loading. A similar trend has been reported by Krishnamoorti et al. (47) for a series of intercalated poly (dimethyldiphenylsiloxane)-layered silicate nanocomposites with different silicate loadings. At high shear rates, the shear viscosity and shear thinning behavior for the nanocomposites are comparable to those of the pure polymer as a result of the preferential orientation of the clay layers or even anisotropic tactoids parallel to the flow direction (48).
To investigate the confinement of polymer chains, we have measured the dynamic rheological behaviors of PEO and PEOCNs at various temperatures and calculated the flow activation energy using Arrhenius equation (49), (50).
[[eta].sub.[gamma]] = A exp([E.sub.[gamma]]/RT) (3)
where [[eta].sub.[gamma]] and [E.sub.[gamma]] are the apparent viscosity and the activation energy, respectively, at a given shear rate. A is a constant, R is the gas constant, and T is the absolute temperature.
Thus, the activation energy ([E.sub.[gamma]]), at a given shear rate may be obtained directly from plots of In [[eta].sub.[gamma]] against 1/T. The calculated activation energies, at different shear rates for PEOCNs, are shown in Table 2. The computed activation energies decrease with increasing shear rate. According to certain established molecular theories for viscous flow, the activation energy may be taken as a measure of the potential energy barrier that is associated with movement of the molecules (51), (52). However, in the case of polymeric melts, the macromolecules that have large volumes move in units of a fixed size, which is rather independent of the total length or size of the molecules. In practice, the viscosity also often depends upon the actual size of the molecules, because they have to be involved in a cooperative movement so that the molecule as a whole may move in the shear field (53). The sizes, or length, of the mobile chain segments are believed to be determined by the flexibility of polymer chains and the environment in the vicinity of the polymer chains. In our case, it means the interactions between the polymer segments themselves and the clay particles. Under the action of a shear stress, the alignment, orientation, and disentanglement of the polymer chain occurs, and this enables the polymer molecules to move more easily. This, in turn, is accompanied with a corresponding decrease of the activation energy with the increasing shear rate.
TABLE 2. The activation energy and effective filler volume fraction data of PEOCN. Activation energy (kJ/mol) Real particle Effective 0.1 rad/ 398 rad/ volume filler volume Sample [s.sup.-1] [s.sup.-1] fraction (%) fraction (%) PEO 12.313 6.2 -- -- PEOCN3 12.994 6.082 2 17 PEOCN6 7.351 4.923 4 22 PEOCN9 5.693 4.172 7 23
It was also observed that the activation energy PEOCN with 3 wt% clay is almost same as that of the PEO but decreases with the increasing clay concentration. The result is not consistent with that observed in case of the polypropylene (PP)-based nanocomposites, where no change (48) or increase (54) in activation energy was reported. Our finding can be explained by considering the polar interaction between the polyether chain and clay layers, which is absent in a PP matrix. As we have discussed in earlier sections, if the enhancement of viscosity is only due to the confinement of polymer molecule then the activation energy should always increase. Lack of difference between the flow activation energy between the pure PEO and the PEOCN means that the solid-like behavior is due to the strong frictional interactions between the clay layers at higher clay concentrations rather than confinement effects (48).
The storage and loss modulus from the dynamic measurements are shown in Figs. 4 and 5. At low frequencies, G" is higher than G' at all clay concentrations and the difference between G' and G" decreases as the frequency increases. G' intersects with G" at the frequency within the transition zone. This crossover frequencies between storage and loss moduli indicates the transition from a liquid-like to a solid-like behavior (55), (56). At low frequencies, the PEO melt behavior is liquid like (G' < G"), while inheriting solid like (G' > G") behavior at higher frequencies. The transition of pure PEO from liquid to solid like behavior occurs at crossover frequency ([[omega].sub.c]) of apparently 83 [s.sup.-1] (Fig. 6a). The [[omega].sub.c] for the nanocomposites was shifted toward a lower frequency with increasing clay content ([[omega].sub.c] is ~ 40 [s.sup.-1] for 3 wt% and [[omega].sub.c] is ~ 24 [s.sup.-1] for 6 wt%) (Fig. 6b and c). However, in the nanocomposites having clay content more than 6 wt% [[omega].sub.c] is shifting towards higher frequency region ([[omega].sub.c] is ~ 90 [s.sup.-1] for 9 wt%) (Fig. 6d). Hyun et al. (57) reported a similar behavior at higher loading for PEO/Cloisite 25A composites prepared by solution intercalation method. On the basis of mesoscopic structure at low clay concentrations, it was suggested that beyond a critical volume fraction, the tactoids and individual clay layers are incapable of rotating freely and are prevented from complete relaxation when subjected to shear. This incomplete relaxation due to the physical jamming or percolation lead to the presence of pseudo-solid like behavior observed in both the intercalated and the exfoliated nanocomposites (58). Thus, it appears that intimate contact between the polymer and the clay platelets alters the relaxation processes of the polymer, leading to the low frequency plateau in the shear moduli and nonNewtonian viscosity behavior with clay loading at the low-shear rate (59).
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
In the case of nanocomposites, the melt rheological properties are also helpful to find out the degree of polymer-filler interactions and the structure-property relationships. This is because rheological material functions are strongly influenced by the structure and the interfacial properties. The viscosity enhancement of the PEOCNs can be attributed to the formation of a filler network between the polymer and nanoclay, which leads to an increase in the effective filler volume fraction. The effective filler volume fraction can be estimated with different methods. The equilibrium immobilized layer thickness of PEO on a flat surface is assumed to be [R.sub.g]. The confined polymer shell thickness [delta] around the nanospheres filler having radius (r) is deduced from the following relationship: (60)
[R.sub.g] = [4[pi]/3] [[(r + [delta]).sup.3] - [r.sup.3]/4[pi][r.sup.2]] (4)
Another method is to estimate the effective filler volume fraction from the plateau elastic modulus ([G.sub.N]). [G.sub.N] can be obtained from Fig. 6 as the elastic modulus G' at the angular frequency corresponding to a local minimum of the loss modulus G". For isotropic particle filled systems, the dependence of the plateau elastic modulus [G.sub.N] ([[empty set].sub.e]) on the effective filler volume fraction [[empty set].sub.e] can be related with the Guth-Smallwood equation (61):
[G.sub.N]([[empty set].sub.e]) = [G.sub.N](0)(1 + 2.5[[empty set].sub.e] + 14.1[[empty set].sub.e.sup.2]) (5)
With [G.sub.N] ([[empty set].sub.e]) and [G.sub.N](0) already known, the effective filler volume fraction [[empty set].sub.e] can be easily calculated at different clay loadings.
The first method is generally used to calculate the volume of immobilized polymer shells around spherical filler particles. So we have used the second method for estimating the effective filler volume fraction and the results are shown in Table 2. The increase in the modulus value in the oscillatory shear results suggests the existence of a filler network. The effective volume fraction is found to be more pronounced at lower concentration of clay in the PEOCNs. At higher clay content, the effective filler volume fraction is found to be remaining more or less the same. It is well known that a better dispersion of clay can be realized and agglomeration can be avoided/minimized only in low organoclay concentrations (3), (8), (10).
Several methods like transmission electron microscopy (TEM), SEM, AFM, XRD, and neutron scattering have been used to elucidate nanocomposite structure to provide the measure of size, shape, and interfacial conformation (4), (10), (25), (27). It was pointed out by some studies that different tools such as TEM, AFM, and XRD may give a different picture of the state of morphology (62). We have used XRD, AFM, and SEM, to characterize the nanocomposite structure.
Figure 7 represents the XRD patterns for organically modified clay (MMT-30) and PEOCN6. The clay shows the [d.sub.100] peak at 2[theta] = 4.04 which corresponds to a d-spacing of 2.18 nm. This indicates that modification of clay with organic ions not only makes the clay surface hydrophobic but results in a 1.18 nm increase in d-spacing (as the d-spacing for untreated clay is about 1 nm). This facilitates the penetration of PEO into the interlayer galleries (6), (7). In case of the composite sample, the basal spacing of clay is increased to 4.32 nm as the diffraction angle shifts from 2[theta] = 4.04 to 2[theta] = 2.06 for the [d.sub.001] peak. Hence, the polymer layer in between the two silicate layers is ~3.14 nm. The result is consistent with the recent SANS studies (27), (32), (33) on the adsorption of PEO chains onto the clay platelets at low polymer and the clay concentrations using contrast variation methods to separate the contribution from the bulk and adsorbed polymer chain. The studies indicate that the thickness of PEO layer is maximum 1.5 nm and the excess polymer in bulk or solution is not directly cross linked to clay particles but cover the polymer/clay aggregates. Hence, the total PEO layer between the two clay platelets is 3 nm. The appearance of a peak in the nanocomposite close to the peak for the virgin organoclay indicates the presence of tactoids (5), (6), (8). Interestingly, the peak representing the tactoid corresponds to lower d-spacing when compared with the virgin clay. This may suggest the occurrence of some sort of conformational change of the organic moiety inside the clay galleries.
[FIGURE 7 OMITTED]
AFM was found to be a powerful tool to investigate the nanocomposite surface on a nanometer scale (5), (32-34). Figure 8 shows AFM images for PEOCN-6 in x-z planes. Cursor profile and height histogram are also shown. The bright phase represents the stiffer layered silicate, whereas the dark background represents the softer PEO matrix. Individual layers can not be seen very clearly by AFM as they can be analyzed by TEM. However, the AFM image indicates that at least part of the layered clay structure has been broken into 40-60 nm stacks. Because one silicate layer with coated PEO represents a size of about 4 nm, the bundles resolved by AFM contain ~10-15 clay layers.
[FIGURE 8 OMITTED]
The fracture surfaces of compression molded samples were analyzed by high resolution SEM. SEM microphotographs of PEOCN samples containing 6 wt% and 9 wt% clay are shown in Fig. 9. Because of high electron density the layered clay represents the white phase and the black part is due to the PEO matrix. Because the samples analyzed are compression molded, focused on the fracture surfaces, the cross sections of the layered silicates are observed, which are appeared as spherical particles. The photographs in Fig. 9 clearly indicate the presence of clay layers with a size varying from 30 nm to 80 nm. The photographs also indicate the presence of agglomeration as observed by XRD. The agglomeration increases when the clay concentration is increased from 6 wt% to 9 wt%.
[FIGURE 9 OMITTED]
It is expected that the incorporation of nanoparticles in a semi-crystalline polymer matrix would substantially affect the crystallization behavior of the polymer. Depending on the polymer/filler interactions, there can be three general behaviors during crystallization namely development of new crystal structures, heterogeneous nucleation by fillers, and polymer amorphization by filler (63). POM was used to compare the crystal morphology between PEO and PEO/clay nanocomposite. Figure 10 compares the POM images of PEO and PEOCN3 which were isothermally crystallized at 40[degrees]C. The morphology of the crystals is shown at the beginning and at the final stage of crystallization. For the PEO alone, it can be clearly seen that the spherulites are similar in size. Before impinging upon one another, they appear circular, suggesting a spherical shape. For the intercalated system, the spherulite formation varies considerably. The spherulites are typically much smaller than those seen in the virgin PEO. A similar behavior was reported by Strawhecker and Manias (64) and also by Ratna et al. (20) for solution intercalated PEOCN systems. It is clear from Fig, 10, the clay particle act as nucleating agent and the PEO crystallized in the heterogeneous nucleation mode. As the number of nucleation sites increased the number of the spherulites also increased and hence, smaller size of spherulites are formed.
[FIGURE 10 OMITTED]
To study the effect of incorporation of clay on the viscoelastic properties of PEOCNs, pure PEO and nanocomposite samples were subjected to DMA. The dynamic storage modulus (E') vs. temperature and loss modulus vs. temperature plots for pure PEO and PEOCNs are shown in Fig. 11. Analyzing the curves, it can be seen that the loss modulus peak temperature (indicative of the glass-to rubber transition) shifts towards higher temperatures. This can be attributed to the confinement of polymer chains as a result of intercalation into the interlayer gallery of the clay (20), (21).
[FIGURE 11 OMITTED]
The storage moduli E' of the nanocomposites are always greater when compared with pure PEO over all the temperatures investigated. This indicates the reinforcing effect of the organoclay. To evaluate the real contribution of clay dispersion, the ratios between the tensile storage modulus of the nanocomposite ([E'.sub.PEOCN) and the tensile storage modulus of the respective PEO matrix ([E'.sub.PEO]) for all the PEOCNs were determined (see Fig. 12). The tensile storage modulus depends strongly on the organoclay loading, the formed nanocomposite, structure, and also on the effective clay aspect ratio (65). The modulus ratio can be theoretically calculated using a modified version of the Halpin-Tsai model (66), as given below:
[FIGURE 12 OMITTED]
[[E'.sub.PEOCN]/[E'.sub.PEO]] = [1 + (1/d) X[V.sub.f]/1 - X[V.sub.f]] (6)
X = [[E'.sub.r]/[E'.sub.PEO] - 1/[E'.sub.r]/[E'.sub.PEO]] + (1/d) (7)
[E'.sub.PEOCN], [E'.sub.PEO], and [E'.sub.r] are, respectively, the tensile modulus of the nanocomposite, the matrix, and the clay reinforcement. [V.sub.r] is the volume fraction of the nanoclay, and l/d is the aspect ratio of the added clay. The experimental value and the values calculated from the model are shown in Fig. 12. It is evident that the experimental values are close to the theoretical value only at lower concentration of clay and deviate significantly at higher concentrations. This can again be explained from the same point discussed earlier that effective nanodisperson is realized only at lower clay concentrations. Recall, that a considerable agglomeration was detected by SEM for the samples with higher clay contents.
The reinforcing effect is also reflected in tensile properties of PEOCNs. When compared with other thermoplastic polymers, there is less report on the mechanical and dynamic mechanical properties of PEO-based materials because they are not used for high strength applications. Recently, Ratna et al. (21) reported that incorporation of ions in PEO for solid polymer electrolyte applications reduces the strength of PEO substantially, and some sort of reinforcement and mechanical analysis are very important for the development of high performance PEO-based solid polymer electrolyte applications (19-22). The tensile properties, both the tensile strength and modulus, of the PEOCNs were enhanced by the addition of nano-clay. However the tensile properties are increased only up to 6 wt% of the clay (see Fig. 13). In the case of solution intercalated nanocomposites, increase in tensile strength has been reported up to 10 wt% of the clay (20).
[FIGURE 13 OMITTED]
This can be explained by the fact that the PEO viscosity is much higher in case of melt-mixing when compared with the solution intercalation method. Also it may be noted that because of high crystallinity of PEO, the film quality is always better in solution cast method when compared with compression molding. The decrease in tensile properties beyond 6 wt% MMT-30 concentration can be explained in terms of XRD and SEM studies, which indicate the presence of agglomeration (tactoids) even in PEOCN6. A further increase of the tactoid content at higher clay contents may result in a further reduction of the mechanical properties.
PEO-layered silicate nanocomposites were made by a simple mixing method using a laboratory kneader and followed by compression molding. Optimum mixing temperature was found to be 100[degrees]C. Characterization of the nanocomposites by XRD, AFM, and SEM showed the formation of intercalated structures along with the clay tactoids. The presence of organoclay in PEO causes a retardation of the crystal growth and results in smaller and irregular spherulites. Dynamic viscoelastic measurements demonstrated a significant increase in the viscosity at low shear rate. In high frequency range, the increase in the viscosity was found to be marginal. As expected, the melts are pseudo plastic in nature and the viscosity decreases with the increasing temperature. The activation energy for the polymer melt flow was found to decrease with shear rate as well as with the addition of clay. Thus we conclude that the viscosity enhancement at low shear rate is not only due to confinement of polymer chains (as evidenced by increase in [T.sub.g] measured by DMA) but also due to the frictional interactions between the anisotropic silicate layers. The crossover frequencies between storage and loss moduli, which indicate the transition from liquid-like to solid-like behavior, are shifted towards lower frequency region for the nanocomposites up to 6 wt% of MMT-30. The effective particle volume fraction was calculated using the Guth-Smallwood equation. A much higher volume fraction was estimated than that of the real particle volume fraction because of the formation of a good networking between the polymer and the filler particles. The elastic modulus measured by DMA was found to be matching with the theoretical values only in case of composites with lower clay concentration (<5 wt%). A modest reinforcing effect was observed from tensile test up to 6 wt% MMT-30.
(1.) T.J. Pinnavaia and G.W. Beall, Polymer-Clay Nanocomposites, Wiley, Chichester, UK (2000).
(2.) M. Alexandre and P. Dubois, Mater. Sci. Eng., 28, 1 (2000).
(3.) M. Kawasumi, J. Polym. Sci. Part A: Polym. Chem., 42, 819 (2004).
(4.) S. Sinha Ray and M. Okamoto, Prog. Polym. Sci., 28, 1539 (2003).
(5.) S. Sinha Ray, J. Ind. Eng. Chem., 12, 811 (2006).
(6.) M. Okamoto, Rapra Rev. Rep., 14, 1 (2003).
(7.) Q.T. Nguyen and D.G. Baird, Adv. Polym. Technol., 25, 270 (2006).
(8.) D. Ratna, N.R. Manoj, R.K. Singh Raman, R. Varley, and G.P. Simon, Polym. Int., 52, 1403 (2003).
(9.) R.A. Vaia, K.D. Jandt, E.J. Kramer, and E.P. Giannelis, Macromolecules, 28, 8080 (1995).
(10.) O. Becker, Y.B. Cheng, R.J. Varley, and G.P. Simon, Macromolecules, 36, 1616 (2003).
(11.) T. Lan, P.D. Kaviratna, and T.J. Pinnavaia, J. Phys. Chem. Solids, 57, 1005 (1996).
(12.) X.Q. Yang, L. Hanson, J. McBreen, and Y. Okamoto, J. Power Sources, 54, 198 (1995).
(13.) R. Mishra and K. Rao, J. Solid State Ionics, 106, 113 (1998).
(14.) P. Aranda, Y. Mosqueda. E. Perez-Cappe, and E. RuizHitzky, J. Polym. Sci. Part B: Polym. Phys., 41, 3249 (2003).
(15.) P. Gould, Mater. Today, 10, 10 (2007).
(16.) C. Liu, H. Qin, and P.T. Mather, J. Mater. Chem., 17, 1543 (2007).
(17.) P. Aranda and E. Ruiz-Hitzky, Chem. Mater., 4, 1395 (1992).
(18.) Z. Shen, G.P. Simon, and Y.B. Cheng, Polymer, 43, 4251 (2002).
(19.) P. Aranda and E. Ruiz-Hitzky, Acta Polym., 45, 59 (1994).
(20.) D. Ratna, S. Divekar, A.B. Samui, B.C. Chakraborty, and A.K. Banthia, Polymer, 47, 4068 (2006).
(21.) D. Ratna, S. Divekar, P. Sivaraman, A.B. Samui, and B.C. Chakraborty, Polym. Int., 56, 900 (2007).
(22.) J. Kwiatkowski and A.K. Whittaker, J. Polym. Sci. Part B: Polym. Phys., 39, 1678 (2001).
(23.) E. Hackett, E. Manias, and E.P. Giannelis, Chem. Mater., 12, 2161 (2000).
(24.) T.H. Kim, L.W. Jang, D.C. Lee, H.J. Choi, and M.S. Jhon, Macromol. Rapid Commun., 23, 191 (2002).
(25.) R. Krishnamoorti and K. Yurekli, Curr. Opin. Colloid Interface Sci., 6, 464 (2001).
(26.) D.K. Yang and D.B. Zax, J. Chem. Phys., 110, 5325 (1991).
(27.) N.M. Malwitz, A. Dundigalla, V. Ferreiro, P.D. Butler, M.C. Henk, and G. Schmidt, Phys. Chem. Chem. Phys., 42, 3102 (2004).
(28.) B. Chen and J.R.G. Evans, J. Phys. Chem. B, 108, 14986 (2004).
(29.) E. Loizou, P. Butler, L. Porcar, and G. Schmidt. Macromolecules, 39, 1614 (2006).
(30.) G. Schmidt, A.I. Nakatani. and C.C. Han, Rheol. Acta, 41, 45 (2002).
(31.) J. Zebrowski. V. Prasad. W. Zhang. L.M. Walker, and D.A. Weitz, Colloids Surf. A, 213, 189 (2003).
(32.) E.A. Stefanescu, A. Dundigalla, V. Ferreiro, E. Loizou, L. Porcar, I. Negulescu, J. Garnoa, and G. Schmidt, Phys. Chem. Chem. Phys., 8, 1739 (2006).
(33.) E. Loizou, P. Butler, L. Porcar, and G. Schmidt, Macromolecules, 38, 2047 (2005).
(34.) E.A. Stefanescu, P.J. Schenxnailder, A. Dundigalla, I.I. Negulescu, and G. Schmidt, Polymer, 47, 7339 (2006).
(35.) A. Dundigalla, S. Lin-Gibson, V. Ferreeiro, M.M. Malwitz, and G. Schmidt, Macromol. Rapid Commun., 26, 143 (2005).
(36.) P.H. Nam, P. Maiti, M. Okamoto, T. Kotaka, N. Hasegawa, and A. Usuki, Polymer, 42, 9633 (2001).
(37.) R.A. Vaia and E.P. Giannelis. Macromolecules, 30, 7990 (1997).
(38.) Z. Shen, G.P. Simon, and Y.B. Cheng, Eur. Polym. J., 39, 1917 (2003).
(39.) W. Loyens, P. Jannasch, and F.H.J. Maurer, Polymer, 46, 903 (2005).
(40.) M.M. Fares, J. Hacaloglu, and S. Suzer, Eur. Polym. J., 30, 845 (1994).
(41.) S.W. Bigger, J. Scheirs, O. Delatycki, and N.C. Billingham, Polym. Int., 26, 181 (1991).
(42.) H.J. Choi, S.G. Kim, Y.H. Hyun, and M.S. Jhon, Macromol. Rapid Commun., 22, 320 (2001).
(43.) H. Cai, A. Ait-Kadi, and J. Brisson, Polymer, 44, 1481 (2003).
(44.) Y.S. Song, Polym. Eng. Sci., 46. 1350 (2006).
(45.) A. Subbotin, A. Semenov, and M. Doi, Phys. Rev. E, 56, 623 (1997).
(46.) R. Khare, J. Pablo, and A.M. Yethiraj, Macromolecules. 29, 7910 (1996).
(47.) R. Krishnamoorti, A. Vaia, and E.P. Giannelis, Chem. Mater., 8, 1728 (1996).
(48.) G. Galgali, C. Ramesh, and A. Lele, Macromolecules, 34. 852 (2001).
(49.) L. Nielson, Polymer Rheology, Marcel Dekker, New York (1997).
(50.) Y. Xiong and E. Kiran, Polymer, 38, 4817 (1997).
(51.) R.A. Vaia, S. Vasudeevan, W. Krawiec, L.G. Scanlon, and E.P. Giannelis, Adv. Mater., 7, 2 (1995).
(52.) F. Cogswell and J. McGowan, Br. Polym. J., 4. 183 (1972).
(53.) B. Briscoe, P. Luckham, and S. Zhu, J. Appl. Polym. Sci., 70, 419 (1998).
(54.) S.Y. Gu, J. Ren, and Q.-F. Wang, J., Appl. Polym. Sci., 91. 2427 (2004).
(55.) R.G. Larson, The Structure and Rheology of Complex Fluids, Oxford University Press, New York (1999).
(56.) A. Lele, M. Mackley, G. Galgali, and C. Ramesh, J. Rheol., 46, 1091 (2002).
(57.) Y.H. Hyun, S.T. Lim, H.J. Choi, and M.S. Jhon. Macromolecules, 34, 8084 (2001).
(58.) J. Ren, A.S. Silva, and R. Krishnamoorti, Macromolecules, 33, 3739 (2000).
(59.) P. Maiti, Langmuir, 19, 5502 (2003).
(60.) Q. Zhang and L.A. Archer, Langmuir, 18, 26 (2002).
(61.) H.M. Smallwood, J. Appl. Phys., 15, 758 (1944).
(62.) O. Becker, R.J. Varley, and G.P. Simon, Polymer, 43, 4365 (2002).
(63). E. Manias, G. Polizos, H. Nakajima, and M.J. Heidecker, Fundamentals of Polymer Nanocomposite Technology in Flame Retardant Polymer Nanocomposites. Wiley. Hoboken, New Jersey, Chapter 2, 31 (2007).
(64.) K.E. Strawhecker and E. Manias, Chem. Mater., 15, 844 (2003).
(65.) T.D. Fornes, P.J. Yoon, H. Keskkula, and D.R. Paul, Polymer, 42, 9929 (2001).
(66.) E.M. Van, PhD Thesis, TUDelft, The Netherlands (2001).
T.N. Abraham, D. Ratna, S. Siengchin, J. Karger-Kocsis
Institut fur Verbundwerkstoffe GmbH (Institute for Composite Materials), Technical University, Kaiserslautern, D-67663 Kaiserslautern, Germany
Correspondence to: D. Ratna: e-mail: firstname.lastname@example.org
D. Ratna is currently at Naval Materials Research Laboratory, Addl.
Ambernath, Thane 421506, India.
Published online in Wiley InterScience (www.interscience.wiley.com).
[c] 2008 Society of Plastics Engineers
|Printer friendly Cite/link Email Feedback|
|Author:||Abraham, T.N.; Ratna, D.; Siengchin, S.; Karger-Kocsis, J.|
|Publication:||Polymer Engineering and Science|
|Article Type:||Technical report|
|Date:||Feb 1, 2009|
|Previous Article:||Pea starch-based composite films with pea hull fibers and pea hull fiber-derived nanowhiskers.|
|Next Article:||Effects of the spin line temperature profile and melt index of poly(propylene) on melt-electrospinning.|