Morphological Study of HDPE/Clay Hybrids Synthesized by an Alternative Compatibilization Path

The control of the factors that influence the internal structure of polymeric materials (crystallinity, miscibility between phases, processing, etc.), is of paramount importance because it is closely related to the outcome of their final properties (Durmus et al., 2007; Sinha Ray & Okamoto, 2003; Franck, n.d.). Therefore, the morphological analysis is essential in order to know the structure-property relationship (Williams & Carter, 2009).

report that polymer structure variations are manifested by changes in the crystallization temperature (T c ) or in the glass transition temperature (T g ) which can be detected by DSC measurements.
On the other hand, rheological tests extensively are used in polymer characterization (Zhang Q. et al., 2004) and they bring important information about processing optimization (C.Wan et al., 2004;Shah & Paul, 2004).Particularly, the structural characterization of filled polymer through low-amplitude oscillatory experiments allows determining the linear viscoelastic properties that describe the material behavior in a condition where the structure remains practically unchanged.These properties are sensitive to the interactions that take place among phases, and to the degree of miscibility reached (Okamoto, 2005;Bretas & D'Avila, 2000;Cassagnau, 2008), due to the fact that the applied strain deformations are small enough to affect the material properties.
In this chapter, several hybrids based on high density polyethylene (HDPE) and montmorillonite are studied.Due to the differences between the hydrophilic natures of these materials, a clay surface modification using difunctional silanes was chosen in order to generate oligomers with hydrophobic lateral groups around the lamellas.
Furthermore, to determine if the final materials behaviors were affected by alterations in the chemical nature of the interphase, alkyl groups were alternated with aryl groups.
On the other hand, transport properties are analyzed aiming to characterize the outcome of the materials at the macroscopic level.In this way, it would be able to know how the use of the structural analysis via TEM can help to understand these behaviors and how, if combined with the two techniques mentioned above, allow a better analysis of polymer composites.
Finally, the results made possible a better understanding of the barrier properties, and some quantitative relations with the structure of the synthesized materials were found.

Materials
In polymer/clay systems, the structure at nanometric level (intercalation/exfoliation) depends on the matrix polarity, being more feasible to get better miscibility in polar matrixes.Consequently, compatibilization is required in order to obtain a polyolefin nanocomposite.This implies using copolymers or clay modification by trifunctional silanes or organic ions rendering the nanolamellas more hydrophobic (Durmus et al., 2007;Okamoto, 2005;Malucelli, 2007; L. Wang & J. Sheng, 2005).
Polymer additives were synthesized by treating an organophilic clay (oC) with difunctional silanes, aiming to obtain surface oligomers with side groups compatible to the high density polyethylene matrix, while their backbones share the same clay chemistry.This procedure is proposed as an alternative path for the obtaining of polyolefin/clay nanocomposites and it is based on the methods reported by (X.Zheng & Wilkie, 2003;J. Zhang et al., 2005;C. Zhao et al., 2004).
The molecular models of the employed silanes are described on Figure 1: From these monomers, by means of an in situ hydrolysis, four clays with their respective oligosiloxanes were obtained, named as: oC/DM, oC/MP, oC/DP Cl y oC/DP OCH 3 .Subsequently, they were mixed with HDPE by melting process thus each composite contained ~3%wt clay.The preparation conditions and the results from X-ray diffraction, Fourier transform infrared spectroscopy and non-oxidative thermal degradation of the clays and the polymeric materials can be found in Monasterio et al. (2010Monasterio et al. ( , 2011) ) and Monasterio & Destéfanis (2011).
To systematize the analysis of the HDPE/clay materials, the collected data was split in two groups according with the criteria described on In the following sections it must be taken into account that the prepared PE composites contain equal clay percentage, and since it is the same polymer matrix and the same inorganic multilayered material, the changes detected by the above mentioned techniques will be associated to the effects obtained by the modifying species present in the different interphases generated on each clay.
To perform a more detailed analysis of the oligomers effects in the interphase, the hybrids with sC and oC are included.

Structural characterization via TEM
Before incorporating the additives (clays) into the HDPE, they were analyzed by TEM.Each of the samples was sonicated in ethanol.The resulting suspension of the product in the solvent was settled for approximately 20 minutes and then, a drop of the dilute suspension was placed onto a coated copper grid.The solvent evaporated from the grid, leaving a thin-layer of each of the products on the grid.Figure 2 shows the photographs of three of the used clays.In general, particulate materials of different sizes were visualized.In addition, tonality changes were observed which are related to the mass-thickness contrast, pointing out differences in the particle thickness or the composition.The thinner particles (light grey) were associated with clay sheets.
The clays containing oligosiloxanes tend to form clusters.In all cases, agglomerates superior to 1 μm were observed but they presented different appearances.In Figure 2a), it has a black color while similar sized agglomerates in the other samples are surrounded by lighter grey particles.This can be the result of higher clay dispersion caused by the in situ generation of oligomers into the organophilic clay.These results are in agreement with previous scanning electron microscopy studies of these additives (Monasterio et al., 2010).
Thin sections (~60nm) cut by means of a cryoultramicrotome were obtained for the HDPE/clay hybrids morphological analysis.Cryoultramicrotomy is a standard method often used with low T g polymers (e.g.HDPE T g =-125°C) (Kaufman & Falcetta, 1977), because it provides thin sections that are free from artefacts.Hence it is possible to carry out a correct interpretation of the actual composite morphology (Michler, 2008).The visualization was performed at high acceleration voltage (100kV), which improves the resolution and reduces the scattering cross-section to protect the specimens which are sensitive to electronic radiation (Williams & Carter, 2009).It must be mentioned that SEM photographs of these composites (Monasterio et al., 2011;Monasterio & Destéfanis, 2011) revealed that clays (with or without surface treatment) achieved a good distribution inside   Dispersed particles of micrometric size were observed in PE-sC and PE-oC.While oligosiloxanes modified clays exhibited a major stacks disaggregation.Taking into account group I (Table 1), PE-oC/MP presents the best dispersion/distribution (particles thickness < 200nm) while the rest of the composites show agglomerations.For group II, where the treated clay composition is basically the same, PE-oC/DP OCH 3 exhibited better dispersion of the filler (particles thickness < 500 nm).
On the other side, X-ray diffractograms reported before (Monasterio et al., 2011;Monasterio & Destéfanis, 2011) showed intercalation/exfoliation for PE-oC/MP and PE-oC/DP OCH 3 , and they were coextensive with that observed in TEM photographs.Although PE-oC/DM and PE-oC/DP Cl , had better delamination than PE-sC y PE-oC, they exhibited 'stacking recovery'.
In conclusion, the decreasing dispersion order was PE-oC/MP, PE-oC/DP OCH 3 , PE-oC/DM, PE-oC/DP Cl , PE-oC, PE-sC.Specially, the morphology obtained in PE-oC/MP is similar to that reported by TEM images by Zhang et al. for PE and PP nanocomposites (J. Zhang et al., 2005).Consequently, the proposed oligosiloxane treatment could be considered as an alternative for compatibilization.

Structural characterization via rheological analysis
Rheological behaviors in oscillatory regime were determined in a rheometer TA AR2000 of parallel plates (D=25 mm, h= 1 mm) at 200°C in N 2 atmosphere.Before measuring, each sample underwent a relaxation at an angular frequency of 5 rad/s.Results were plotted using the 'frequency domain phasor representation' (Figure 4).The locus of a phasor is known as the 'digital fingerprint' of a material.This plot summarizes the values of storage (G') and loss (G") moduli and the phase angle ( ).
From this representation, an increase of clay-polyolefin compatibility was inferred.It can also be noticed that PE was above the other values, whereas PE with sodium clay values were below the other hybrids.PE-oC behavior is near of that described by PE-sC while materials containing oligosiloxanes presented a behavior closing to pure PE.
On the other hand, not only were noticed changes for the materials with different reactive leaving groups (group II) but also for those with different side groups (group I).At low frequencies, composites containing oligomers presented lower G' and G'' values than PE.Phase angle were between 30 and 60° which denotes the composites viscoelastic nature.
Figure 5 illustrates the curve of viscosity as a function of the frequency where a pseudoplastic behavior for pure HDPE and hybrids was observed.This is due to the fact that the input shear energy tends to line up random oriented molecules or particles and to disaggregate any large clay stacks, thereby reducing the overall hydrodynamic drag, which in turn reduces the dissipation of energy in the fluid and the viscosity (Punnarak et al., 2006;D. Chen et al., 2005; TA Instruments, (RN-9B)).For hybrids filled with silane treated clays, oligomers would diminish the average molecular weight affecting the viscosity (Billmeyer, 1988).Among siloxane treated hybrids, it is observed that at low frequencies the highest viscosity is for PE-oC/MP, followed by PE-oC/DM and finally by PE-oC/DP OCH 3 and PE-oC/DP Cl .Probably this is a consequence of the effects caused by oligomeric species obtained during the in situ hydrolysis of each monomer introduced into oC.The Cole-Cole viscosity representation (Figure 6) was also analyzed to know the miscibility among the hybrids components.The curves proximity to a semicircular shape indicates higher miscibility and the deviations respond to a component interaction reduction (Ahmed et al., 2010).In this way, it was detected that both sC and oC did not reach a good compatibility with the polymer matrix, while the other siloxane treated clays composites showed miscibility with PE.
A deep structural analysis can be performed by using the van Gurp-Palmen plot which provides information about the length and amount of branches in a sample, and the molecular weight distribution (Lohse et al., 2002;Schlatter et al., 2005).These results are in agreement with the GPC results (Monasterio et al., 2010(Monasterio et al., , 2011) ) (Table 2) and they explain the above mentioned viscosity decrease (Figure 5).Interestingly in group II, dimers were synthesized and the chemical nature of the interaction was similar.However, different behaviors were detected.In one of the cited papers, it was made a comparative analysis between PE-oC/DP Cl and PE-oC/DP OCH 3 behaviors.From it, it was possible to conclude that the differences between the mechanisms during the synthesis of the oligosiloxane surface modified clays affected their further interactions when being added to PE. Basically, the difference between both additives lies in the degree of coating achieved on clay platelets.

www.intechopen.com
Morphological Study of HDPE/Clay Hybrids Synthesized by an Alternative Compatibilization Path 299 Further analyses were carried out and their values are shown in Table 2.Even though the above graphics give an idea of the information of each material behavior, the  Franco et al., 2006).PDI: polydispersity index of the oligomers extracted from the oligosiloxane clays before their incorporation to PE.Values obtained by GPC measures).
The crossover point modulus (G cop ) indicates the transition from a solid-like behavior to a liquid-like behavior.Its value can provide the qualitative information about the average molecular weight (M w ), the degree of branching and the molecular mass distribution (Yang, Q. et al., 2009;Ahmed et al., 2010).Reminding that the matrix is the same in the analyzed materials (HDPE), it can be observed that G cop has had a significant change according to the surface treatment applied on each clay.When polydipersion of the modifying oligomers increase, the G cop values decrease.The low values presented by PE-sC and PE-oC are associated to poor interfacial adhesion.
Group I hybrids presented similar values among them and a more elastic (liquid-like) behavior than PE, but in group II it is evident that PE-oC/DP OCH 3 , in spite of its similarity with PE-oC/DP Cl , it shows a value even higher than neat PE, which it make expect differences in the internal structures of these materials.
It was also analyzed the effect of the clays on the plateau modulus (G N 0 ).This value is defined by the Equation 1 (Rose et al., 2007): where: b: coefficient equal to 1, according to Ferry, and equal to 4/5, according to Doi and Edwards, R: is the ideal gas constant, T: is the absolute temperature, ρ: is the polymer density, and M e : is the entanglement molecular weight.
So G N 0 bring information about the entanglement spacing of a polymer melt and the entanglement molar mass.In the case of filled materials, this value could be associated with temporary crosslink between the surface of the particles and the polymer (Gacitua E. et al., 2005).In PE-sC the clay has not any organic species as compatibilizer, so the lower value it might be because no interaction takes place between its surface and the long HDPE chains.
It is evident that the presence of intergallery organic ions in oC is not enough to improve the interaction with the matrix.Not only does the G N 0 not increase, but even more it is inferior to PE-sC value.This behavior could be due to oC tendency to easily delaminate (compared to sC), which would increase the amount of low compatibility particle dispersed in the matrix.
The treatments with silanes had different effects, being PE-oC/DP OCH 3 and PE-oC/MP the composites with higher values taking PE-oC as reference, whereas PE-oC/DM and PE-oC/DP Cl had the lowest values.Evidently, these behaviors are not directly affected by the chemical nature of the side groups.It would be more appropriate to take into account the coating generated by the dissimilar species over the clays to explain these results.Therefore one can think in an oligomer network generated around the sheets (or stack of sheets).
In oC/MP there are trimers and longer chains that can lead to an intricate coating, just like oC/DP OCH 3 which has a higher proportion of dimers outside of clay galleries (compared to oC/DP Cl ) .S o , i n s i d e H D P E m a t r i x i s m o r e p r o b a b l e t h a t t h e s e n e t w o r k s g i v e r i s e t o temporary anchoring that would improve its interaction.In the case of oC/DP Cl , the low dimers concentrarion on the surface would prevent the network formation, and although oC/DM has longer oligomers (mostly pentamers), the rigidity and spatial complexity of the -CH 3 group is smaller than the phenyl group, which would explain that this clay does not have such capability to 'entangled' the PE macromolecules.This would also explain the PE-oC/DM and PE-oC/DP Cl 'stacking recovery' tendency mentioned above.
Nevertheless, the G N 0 values (Table 2) were below the pure PE value, evidencing that despite of the improvements got at the interphase there would be more free volume.
G' values at the terminal zone give information about matrix-filler interaction (Wang, K.H. et al., 2002).Due to PE-sC hybrid did not have any compatibilization it is expected to show the lowest value, as in the system coexist hydrophilic and hydrophobic components.
The Na + exchange with hexadecyltrimethylammonium ion provided a slight improvement in the interface interaction.Only after the second treatment of the clays in order to generate oligosiloxanes, it was possible that these additives increase their slopes, which is associated to an increase in compatibility.Moreover, the G' slope for PE-oC/DP OCH 3 presented a value near to the pure polymer, which is in agreement with the behaviors observed in Figure 6 and which would indicate an improvement in polymer-clay interactions.

Structural characterization by DSC
Semicrystalline polymers like HDPE give information about the behaviors of the amorphous and crystalline phases through T g (glass transition temperature) and T m (melt temperature) or ΔH m (melting enthalpy), respectively (Kaufman & Falcetta, 1977).In the analyzed materials, measurements were carried out in a temperature range above the T g (-50 to 150°C).

www.intechopen.com
Morphological Study of HDPE/Clay Hybrids Synthesized by an Alternative Compatibilization Path 301 The determinations were performed on a DSC Q200 equipment from TA Instruments (coupled with a refrigerated cooling system), equipped with modulated DSC software, which allows the application of a sinusoidal heating programme superimposed to the normal linear temperature ramp (Verdonck et al., 1999; TA Instruments (TS-25)).
( ) where: : is the Total Heat Flow due to the linear heating rate.C p : is the Heat Capacity of the Reversing Heat Flow Component.
: is the measured heating rate, which has both a linear and sinusoidal (modulated) component.f(T,t): is the Kinetic Component of the Total heat flow and is calculated from the difference between the Total signal and Heat Capacity Component.(Non-reversing component) : is the Reversing Heat Flow Component of the Total Heat Flow.
This analysis (Equation 2) allows the deconvolution of the total heat flow signal into 'reversing' and 'non-reversing' contributions.Total Heat Flow due to the underlying heating rate it is equivalent to standard DSC signal at the same average heating rate (De Meuter et al., 1999), and the results obtained from these curves are described in Table 3.Compared with pure polymer, hybrids did not presented significant changes for T m , T onset and degrees of crystallinity, which are in agreement with that reported by other authors in HDPE/clay systems (Gupta & Bhattacharya, 2008).However, regarding the degree of crystallinity it is noticed that clay addition (with or without oligomers), had a slight plasticizing effect as it affected the chain regularity, inhibiting crystallization.
To have a better insight of composites structure, each component of the Total Heat Flow was analyzed.The advantage of modulated signals is the possibility of separate overlapping phenomena (overlapping transitions such as melting/recrystallization in semi-crystalline materials).This complementary information enables a more detailed study of complicated material systems (Verdonck et al., 1999;Coleman & Craig, 1996).Special care must be taken in the selection of modulation technique as it will affect data interpretation (Schawe, 1995), here it was adopted the model proposed by TA Instruments (Thomas, 2005).
It must be pointed out that the melting process is partially retrieved in the reversing heat flow as well as in the non-reversing heat flow signal, and moreover the fraction of melting in both of these signals depends on the experimental conditions (Coleman & Craig, 1996;Verdonck et al., 1999).For that reason, experimental conditions (heating rate and period) for the hybrids studied were chosen guaranteeing enough modulation cycles during the transition of interest.
In order to avoid degradative processes overlapping during result interpretation, TG-DTG analyses over the assessed temperature range were analyzed (Monasterio et al., 2011;Monasterio & Destéfanis, 2011) and no degradation was observed.
Signal deconvolution shows that the main contribution corresponds to the non-reversing component, which exhibited a similar behavior to that of Total Heat Flow signal, even the maxima from both curves coincided.Regarding reversing signals (Figure 8), the curve corresponding to pure polymer showed a small exothermic peak above T m and an endothermic peak below the T m .They can be associated to simultaneous meltingcrystallization occurring during heating, until complete melting.
Fig. 8. PE and PE/clay hybrids 'reversing' components.Heating rate was 1°C/min and was modulated by a sinusoid with amplitude ±0,16°C and period 60 s.Nitrogen was used as a purge gas.Samples thermal history was erased by a first heating-cooling cycle.
On the other hand, despite their low clay percentage, composites showed changes compared with the curve above mentioned.Evidently, there are alterations into molecular ordering linked to the different features provided by the interphases that lead to different heat absorption-release mechanisms in the zone near to 130°C.In general it was observed a decrease in the endotherm below to 130°C and in most cases it was seen an endothermic peak above this temperature.PE-oC/MP (group I) and PE-oC/DP OCH 3 (group II) behaviors must be underlined because they are similar to PE behavior, and for that reason they were represented by dashed lines.
Based on this, it would be expected that values in a Table 3 would be closer for these hybrids.Nonetheless, values did not match with the tendencies described by reversing curves, as crystallinity degree estimation involves information provided by conventional DSC curves.In this way, it is observed that crystallinity does not entirely reflect the material information related to the variation of the interphase characteristics.
Comparing Table 3 with the behaviors registered in Figure 8, a better interpretation of results can be carried out.PE-oC and PE-sC presented similar crystallinity values, thus alkylammonium chains are not enough to produce significant changes.Meanwhile, it is observed that composites containing oligomers with methyl side groups (similar to HDPE) presented a closer value to pure polymer.Moreover, the highest crystallinity of PE-oC/MP can be explained by the phenyl side group which increases free volume and according to the spatial configuration acquired, it could give to HDPE less spatial restrictions.
Finally, PE-oC/DP Cl and PE-oC/DP OCH 3 had low flexibility oligomeric species, which would explain the crystallinity decrease because it would be a hindrance to achieve oligomer-HDPE interfacial ordering.The highest value of PE-oC/DP OCH 3 can be the result of the already mentioned difference in the clay coating, allowing a better ordering with HDPE chains at the interphase.
A better understanding of oligomeric species mobility can be performed by reversing heat capacity (Rev Cp) curves (Figure 9) as there are associated to structural changes (Coleman & Craig, 1996).As it was pointed out, DRX analysis of these polymers (Monasterio et al., 2011;Monasterio & Destéfanis, 2011) corroborated the dispersion exhibited in Figure 3 (TEM) showing exfoliation for PE-oC/MP and intercalation for PE-oC/DP OCH 3 .Both composites show similar behaviors to that of HDPE, perhaps the cause is that oligomeric species provide tridimensional arrangements that permit macromolecules to achieve highly free movements near the dispersed particles interphase.For the remaining hybrids, dispersions and interphase features were different.
Therefore, the peaks above 132°C for PE-oC, PE-oC/DM and PE-oC/DP Cl would be the result of clay confined species that only can manifest themselves after complete HDPE melting.

Transport properties
Until now, due to alterations at PE-clay interphase, changes were detected in the structure of the studied hybrids and therefore, it will be analyzed if such changes have led to any effect on barrier properties.There are several papers where these macroscopic properties could be modified by handling polymer morphological features such as orientation and degree of crystallinity, as both are linked to chain packing efficiency (Koros, 1990).
Bearing in mind that HDPE is a semicrystalline material, it must be outstood that transport through these polymeric systems are usually analyzed on the basis of a simple two phase model, consisting of a dispersed impermeable crystalline phase in a permeable amorphous matrix (Koros, 1990).In addition, a more complex scenario is given by filled composites as they have clay, which effect will depend on the aspect ratio provided according to the degree of tactoids delamination.
Other factors that influence these properties are the size and chemical nature of penetrating species (Koros, 1990).For the following experiences, apolar molecules of different sizes were employed.

Gas permeation
Gas permeation measurements were performed on standard permeation equipment.Permeability values of CH 4 and CO 2 were determined at a pressure of 4 bar and a temperature of 30°C.From a simplified model proposed by Nielsen (1967) (Equation 3), it is observed that the permeability ratios described in Table 4 are inversely proportional to the tortuosity factor. where: P F : filled polymer permeability, P U : unfilled polymer permeability, Φ P : volume fraction of the polymer, and τ: tortuosity factor Considering that the volumetric fraction of the polymer on each system is the same, the possible causes that give rise to these behaviors were analyzed.
At first sight, it is noticed that transport properties improvements were not according to the order of dispersion described on TEM analyses.
PE-oC hybrid showed similar permeability rate for both gases, indicating that the phenomenon was indifferent to the penetrant gas characteristics.Evidently, this is a consequence of free volumes generated by poor interfacial adhesion.PE-oC/MP also showed similar ratios for both gases, but permeabilities values were near to pure PE values, which points out that interphase species developed a more tortuous path than the former clay.
The CH 4 permeability ratios of PE-sC, PE-oC/DP Cl and PE-oC/DP OCH 3 systems were similar, despite of the differences at the PE/clay interphase.Among them, they can be mentioned the clay aspect ratio and the oligomeric species coating clays.In Figure 3, PE-sC presents an aspect ratio higher than fillers from group II, which after different treatments (cationic exchange and in situ oligosiloxane synthesis), exhibited a lower aspect ratio.Consequently, their efficiency as barrier material has been reduced.However, at the group II hybrids interphase, there are dimers that evidently equate the effects of tortuosity until reaching PE-sC permeability value.
CO 2 permeability ratios for PE-sC, PE-oC/DP Cl and PE-oC/DP OCH 3 were alike, too.But permeability reduction of CO 2 was more significant than that observed to CH 4 transport.These results are probably associated with differences in gas molecule sizes or with the possible interactions that can take place between CO 2 and the organic groups at the filler surface.
About PE-oC/DM, the CH 4 permeation had a similar value to that on PE and PE-oC/MP.Nevertheless, this hybrid at CO 2 permeation exhibited an improvement in the barrier properties.Therefore, in spite of having the same interphase characteristics (free volume and oligomers), gas molecules had presented different behaviors, probably due to reasons already cited on the above paragraph.

Pervaporation
A conventional pervaporation system was used for membrane performance testing.The permeation cell consisted of two sections separated by the membrane.A porous disk was www.intechopen.com The Transmission Electron Microscope 306 placed to support the membrane on the low pressure side of the membrane cell.The solvent vapor permeated through the membrane because of the very low pressure (vacuum) on the permeate side.The permeate was condensed in a trap cooled in liquid nitrogen.Measurements were carried out at 50, 60 and 70°C.
First, an analysis of PE-filler interaction was performed, and for that task values at 50°C were employed (Figure 10), as it is considered that the interactions effect at the interface will predominate over the temperature effect.PE-oC/DP Cl showed similar values to that of unmodified polymer.While PE-sC and PE-oC/DM exhibited better barrier properties.The remaining composites showed permeabilities above PE.In particular, PE-oC/MP and PE-oC/DP OCH 3 , which in TEM micrographs presented higher dispersion degree, showed the highest permeabilities.These results reveal that the inorganic filler dispersion is not the only factor that affects solvent transport through a polymeric membrane, but also interphase species that interact with cyclohexane molecules (Nielsen, 1967).
This analysis was extended by assessing the composite behaviors at different temperatures.
Results are displayed in Figure 10.In order to compare the temperature effects, dashed lines taking PE values as references were pointed out.By doing so, it is possible to remark that, for each temperature, permeabilities do not increase at the same proportion for each system.This evidences different temperature sensitivities and these behaviors are closely linked to Activation Energy of Permeation (E P ).To complement this information, solubility values were determined at different temperatures which allowed calculating the Heat of Solution (ΔH S ).Finally, the Activation Energy of Difusion (E D ) is obtained by Equation 4 (Koros, 1990): With the values summarized in All hybrids showed an exothermic behavior for cyclohexane solubilization.By means of ΔH S it was observed that PE/filler systems containing oligosiloxane treated clays with side group -CH 3 had responses similar to PE.However, pristine clay (sC) and oC/DP Cl clay, which contain low concentration of oligomeric species at the surface, presented a similar behavior when interacting with the solvent.The remaining hybrids exhibited intermediate values.
The highest E D value corresponds to the polymer with sodium clay, probably due to the tortuosity effect already mentioned in gas permeation.Hybrids with high degree of clay dispersion (PE-oC/MP y PE-oC/DP OCH 3 ) presented the lowest Ed values.This tendency is associated with that proposed on reversing Cp analysis (Figure 9), where the behaviors observed where related to the mobility at the interphase, which according to the curves would be higher for PE-oC/MP and PE-oC/DP OCH 3 .

Discussion
In view of these results it can be stated that transport properties were affected by the structural changes on each material.According to observations in polymer-clay composites, barrier properties diminish when clay dispersion increases (Sinha Ray & Okamoto, 2003).However, although some of the hybrids herein studied presented good dispersion, they do not always present transport properties improvements.
On the other side, through DSC and rheological analysis it was noticed a plasticizing effect that was detected by the low crystallinity percentage and the decrease in viscosity, respectively.In the papers mentioned above (X.Zheng & Wilkie, 2003;J. Zhang et al., 2005;C. Zhao et al., 2004), the treatments carried out on clay also give rise to a plasticizing effect on the polymeric matrices.Consequently, it would be expected that hybrids barrier properties were below than HDPE values.Nonetheless, behaviors were observed which did not follow this tendency and that required the consideration of the incidence of other factors for the composites analysis.
Hybrids presented similar behaviors for gas permeation, but these tendencies were not the same for cyclohexane.Specifically, PE-oC/MP and PE-oC/DP OCH 3 seem to have reached better polymer-filler compatibility according to Cole-Cole plot (Figure 6), polar coordenates representation (Figure 4), the van Gurp-Palmen plot (Figure 7) and DSC curves (Figure 8), where those hybrids exhibited similar behaviors to that of pure PE, and both composites showed better dispersion/distribution on TEM images.Nevertheless, regarding gas transport, PE-oC/DP OCH 3 exhibited improvements in barrier properties, while PE-oC/MP had values near to PE.But for C 6 H 12 transport both hybrids showed poor barrier properties, compared to unmodified polymer.
These responses can be explained by taking into account ´tortuosity' (τ) and 'chain immobilization' ( ) factors, which in spite of being defined for pure polymers, will be applied to the composites herein studied.τ value considers flow hindrance provoked by the increased effective length path as well as variations in the cross sectional area of the transport regions.While indirectly reflects the amount of increase in activation energy of diffusion caused by 'anchoring' effect from crystalline phase dispersed in amorphous phase (Koros, 1990).The diffusing species size will affect these factors in different proportions.As the sizes of the  , 2006), it would be expected that for small molecules (e.g.CH 4 ) predominates the tortuosity effect, and while the molecule size increases the chain immobilization factor will gain importance.
Due to the hybrids contain several components, τ and will also experiment changes related to the size and shape of oligomer side groups, and to the oligomer-oligomer and oligomer-PE interactions.According to chemical nature and spatial arrangement (or oligomer flexibility), they will affect the hybrids responses in different ways.In HDPE, crystalline domains had a 'crosslinking' effect (Koros, 1990) which prevents chain mobility on the amorphous phase.
When PE is modified with clays, although the inorganic filler inhibit chain mobility, the resulting holes from the low adhesion or the structural differences of oligomeric species would promote the amorphous phase.All factors here mentioned would explain the reason why in spite of being the same PE/clay hybrid, the transport properties behaviors in front of different molecules are not the same.
With the data obtained from each material internal structure, an analysis of the possible relations among them it will be performed.Several models have been proposed to explain and predict polymers transport properties (Koros, 1990).Nevertheless, permeability through filled polymers is very complex and the possibility of developing a model that contemplates all factors is remote (Nielsen, 1967).Thereby, they are proposed correlation that could be used in further researches.
In the hybrids containing silane modified clays, dependence between CO 2 permeability and oligosiloxane polydispersions it was found (Figure 11).Unlike CH 4 and C 6 H 12 permeabilities which would indicate that these molecules transport through these hybrids does not depend on PDI, or that this factor is uncovered by others.
For cyclohexane transport, a correlation between the crystallinity percentage and the Enthalpy of Solubility was observed.Figure 12 exhibits the mentioned correlation.Some authors (Koros, 1990) have established a link between crystallinity and solubility.Changes in the degree of crystallinity for a semicryslalline material modifies solubility because both values depend on the cohesion achieved by the species on the system (Kaufman & Falcetta, 1977), and by means of these results it was observed that the fillers used affected the molecular interactions and they were manifested on the solvent solubilization process.
Finally, the possible relations among transport properties and rheological parameters were evaluated.There were not relations describing cyclohexane behavior.However, gas permeabilities offered interesting responses for hybrids containing oC clay, as it is shown in Figure 13.Values of G' slope at the terminal zones were more related to CH 4 permeability than CO 2 permeability.
Figure 14 shows the relations detected between relative permeabilities and G N 0 /G cop .In the paper of García-Franco et al. (2006), it was observed that G N 0 /G cop value allows splitting the PE rheological response from copolymer composition and tends to a constant value that can be implicated with the matrix behavior in front of species of the same chemical nature.From the proposed correlations, it could be said that slope and y intercepts provide information about gas transport through the matrix and dispersed inorganic filler.G N 0 /G cop would bring information about structural changes caused by compatibilizing species (changes at the interphase).As relative permeabilities are related with the tortuosity factor, a power law relationship (Equation 5) (Koros, 1990) was taken as reference aiming to explain the results.In this expression, the crystallinity degree (related to Φ) would be describing the structural changes in the system.But, as it was noticed by modulated DSC, this value does not entirely describe the hybrids global behavior.Therefore, it is proposed the adimensional value G N 0 /G cop , as each value involves several features of the systems.Taking into account that chain immobilization factor ( ) can improve the interpretation of permeation in semicrystalline polymers, the following expression is suggested (Equation 6): Certainly this equation is very simply considering the amount of factors above mentioned and it must be assessed in other system to prove their true validity.

Conclusions
By means of the characterizations obtained from TEM, rheology and DSC, changes in the original internal structure of HDPE were registered and this is due to the modifications caused on PE-clay intephases.
Although Transmission Electron Microscopy is a powerful technique in the detection of internal structures of filled polymers, when it was used with rheological and DSC analyses, it allow to improve the interpretation of the factors influencing the final behavior at a macroscopic level.Especially, when studying the barrier properties, as their components get into contact with the new species that diffuse through the system.
Particularly, for the hybrids from this work (groups I and II), the modulated DSC analyses and the dynamic oscillatory shear measurements lead to detect a plasticizing effect that was not perceived by TEM photographs.
Therefore, barrier properties of a system with a high degree of clay dispersion (corresponding to high filler-polymer compatibility) depend on a factor that it is not directly seen by TEM and it is the nanostructural behavior at the interphase (chain arrangement and chemical affinity between matrix and the species used as compatibilizers).This was evidenced by the group II composites (PE-oC/DP Cl and PE-oC/DP OCH 3 ), since both interphases contain oligomers of the same length and chemistry.However, differences during silane treatment affected further matrix interactions (macromolecules arrangement) and this led to differences in the responses obtained from the applied techniques.
In addition, the use of MDSC technique allowed to distinguish changes that were not detected by the conventional DSC signal.
Finally, taking into account that clay dispersion into apolar matrices (polyolefins) has been considered a critical problem, the proposed compatibilization path offers an alternative to traditional procedures.With the advantage that, by changing oligosiloxane side groups, it is possible to modify interphase behavior using low filler percentage to achieve tailor-made materials with different bulk properties.

www.intechopen.com
Morphological Study of HDPE/Clay Hybrids Synthesized by an Alternative Compatibilization Path 313

Fig. 2 .
Fig. 2. TEM images of a) oC, b) oC/DM y c) oC/DP OCH 3 performed on a JEOL100 CX II (10000×).The whiter ovals are part of the defects on the support films.

Figure 3
Figure 3 is showing that modification of the chemical nature of siloxane oligomers used as compatibilizer induced differences in the hybrids internal structure.

Fig. 4 .
Fig. 4. Phasor representation of pure PE and PE/clay composites at ω=0.1,1,10,100 rad/s.Unlike other works describing an increase in viscosity due to the addition of filler into the polymeric matrix(Cassagnau, 2008), the materials modified with clays showed lower values than that corresponding to HDPE.In the case of PE-sC and PE-oC, one of the factors that c o u l d g i v e r i s e t o t h i s b e h a v i o r w o u l d b e the diminution among PE macromolecules interactions due to the presence of clay tactoids.

Fig. 9 .
Fig. 9. Reversing Cp comparison of the studied composites in the range of 110 to 140°C.

Fig. 14 .
Fig. 14.Correlations founded between transport properties and rheological parameters for both gases.

Table 1 .
Polymeric systems studied in this chapter.

Table 2 .
Table data offer specific quantitative results of each system which later will be associated with transport properties.Rheological characteristics of HDPE and HDPE/clay hybrids.( * Extrapolated value.G N 0 : Plateau modulus calculated from extrapolation on van Gurp-Palmen plot (García-

Table 3 .
Results from DSC non-isothermal analysis.

Table 4
shows higher CO 2 permeability compared to CH 4 permeability, due to higher condensability of CO 2 .For unfilled PE, the CO 2 /CH 4 permeability rate is near to four, similar to that reported byStancell (Tobolsky & Mark, 1971).

Table 4 .
Transport properties for PE and PE/clay hybrids.(P PE : Permeability of pure HDPE) Morphological Study of HDPE/Clay Hybrids Synthesized by an Alternative Compatibilization Path 305 www.intechopen.com

Table 5 .
Table 5 it is possible to split the information hidden in E P .Values of activation parameter, E D , E P and ΔH S .