# Melt processing effects on the structure and mechanical properties of PA-6/clay nanocomposites.

INTRODUCTIONPolymer/clay nanocomposites are materials composed of a polymer matrix and nanometer-size clay particles. They exhibit significant improvements in tensile modulus and strength and reduced permeability to gases and liquids, as compared with the pure polymer. These property improvements can be realized, while retaining clarity of the polymer without significant increase in density, since the typical clay loading is 2-5%.

Composites that exhibit a change in composition and structure over a nanometer length scale have been shown to afford remarkable property enhancements, relative to conventional microcomposites [1]. While the high aspect ratio of silicate nanolayers is ideal for reinforcement, the nanolayers are not easily dispersed in most polymers, due to their preferred face-to-face stacking in agglomerated tactoids. From the fundamental point of view, the reinforcing effect of nanoparticles is related to the aspect ratio and to the particle-matrix interactions. Dispersion of tactoids into discrete monolayers is further hindered by intrinsic incompatibility of hydrophilic layered silicates and hydrophobic plastics. This necessitates the compatibilization of the clay surface. The purpose of this work has been to evaluate the effects of mixing conditions (two mixing systems), residence time, and interactions between the polymer and clay surface on structural characteristics and mechanical properties of polyamide-6 (PA-6)/clay nanocomposites. Furthermore, comparisons were made between experimental data and the predictions of composite models usually employed to predict mechanical properties.

Polyamide-6 nanocomposites were prepared using two organoclays: Cloisite 30B, which contains polar hydroxyl functional groups, and Cloisite 15A, which has higher gallery spacing but contains no polar or hydrogen bonding groups, and Cloisite [Na.sup.+] which is unmodified sodium montmorillonite (Na-MMT) clay. The nanocomposites were prepared using two twin-screw extrusion systems: one employing conventional mixing and residence time conditions, while the other was modified to achieve longer residence time and higher mixing efficiency.

TECHNICAL BACKGROUND

The Hamaker Constant

The specific interactions at equilibrium between organoclays and polymer melts can be evaluated based on the effective Hamaker constant of the system. Neumann et al. [2] showed, in particle engulfment experiments, that if the effective Hamaker constant of the system is negative, the particles are rejected by the polymer melts. The effective Hamaker constant [A.sub.132] for the system of the pristine clay-1 and the polymer-2 with the organic modifier-3 between them is given below:

[A.sub.132] = ([square root of [A.sub.11]] - [square root of [A.sub.33]])([square root of [A.sub.22]] - [square root of [A.sub.33]]) (1)

where [A.sub.11] is the Hamaker constant of pristine montmorillonite, [A.sub.22] is the Hamaker constant of the polymer, and [A.sub.33] is the Hamaker constant of the organic modifier. If the value of [A.sub.132] is positive, the specific interactions between clay-1 and polymer-2 are favorable and the clay particles are engulfed by the melt. This results in good dispersion and probably leads to exfoliation. If this value is negative, the particles are rejected by the polymer melt and large tactoids are formed. The Hamaker constant values for the organic modifiers were estimated by using the group contribution method described by Vial and Carre [3] and the surface tension data reported by Jasper [4].

A general conclusion based on microparticle-filled polymers is that the strength of composite materials can be maximized, when the interfacial adhesion between the filler surface and the polymer matrix is optimized [5]. The tensile modulus of the composites is more sensitive to the aspect ratio of the filled particles, whereas the tensile strength is more sensitive to the interfacial adhesion. The work of adhesion [W.sub.a] can be quantified to determine interfacial bond strength between the silica surface and the polymer matrix as follows [6]:

[W.sub.a] = [W.sub.a.sup.d] + [W.sub.a.sup.h] (2)

where [W.sub.a.sup.d] represents the work of adhesion due to the dispersion forces and [W.sub.a.sup.h] is the work of adhesion due to the hydrogen bonds. In the case of polymers like polyethylene and polystyrene, which do not have the capacity to form hydrogen bonds with the clay, only dispersion forces are responsible for interfacial adhesion. The work of adhesion between the two neighboring particles from dispersion forces follows the relation:

[W.sub.a.sup.d] = -A/12[pi][d.sup.2] (3)

where A is the Hamaker constant of the system consisting of neighboring particles and d is the separation distance between the two entities. The value for d, the interatomic cut-off distance, is usually taken as 0.165-0.185 nm [1, 2, 7]. In the case of PA-6/clay nanocomposites, the effective Hamaker constant values between the clay platelets and polymer matrix were estimated to calculate the dispersion forces between the two entities. The density of the hydrogen bonds between the clay platelets and the polymer matrix and the bond energy of N--H bonds were used to estimate forces due to hydrogen bonding. Thus, the total work of adhesion between different clay/polymer systems was estimated.

It should be pointed out that the above arguments imply that any reactions that might occur during processing, including degradation of the organic modifier, do not change the values of the Hamaker constant significantly. However, further research is needed to resolve this issue for the systems under consideration.

Models of Composite Modulus

The tensile modulus data of the nanocomposites were fit to models based on composite theory to estimate the aspect ratios of the reinforced particles. These models include the equation proposed by Halpin-Tsai [8]:

[E.sub.c]/[E.sub.m] = [1 + 2p[eta][phi]]/[1 - [eta][phi]] (4)

where [eta] = [[E.sub.r] - 1]/[[E.sub.r] + 2p], [E.sub.r] = [E.sub.f]/[E.sub.m], and [E.sub.c], [E.sub.m], and [E.sub.f] are the Young's moduli of the composite, matrix, and clay platelets, respectively; p is the aspect ratio defined as the ratio of the width to the thickness of filler particle, and [phi] is volume fraction of the filler.

Another composite model employs the modified rule of mixtures (MROM) proposed by Riley [9]:

[E.sub.c] = [phi][E.sub.f]MRF + (1 - [phi])[E.sub.m] (5)

where the modulus reduction factor, MRF = 1 - [[ln(u + 1)]/u], u = [1/p] [square root of ([phi]G/[[E.sub.f](1 - [phi])]]), and G is the shear modulus of the polymer matrix.

Shia and coworkers [10, 11] derived the following formulae for the Young's modulus of composites containing aligned platelets with a perfect interface,

[E.sub.c]/[E.sub.m] = 1/[1 - [[phi]/4][[1/[xi]] + [3/[[xi] + [LAMBDA]]]]]

[xi] = [phi] + [[E.sub.m]/[[E.sub.f] - [E.sub.m]]] + 3(1 - [phi])[[(1 - g)[p.sup.2] - (g/2)]/[[p.sup.2] - 1]] (6)

g = [[pi]/2]p

[LAMBDA] = (1 - [phi]) [[3([p.sup.2] + 0.25)g - 2[p.sup.2]]/[[p.sup.2] - 1]].

The aspect ratio p in this model is defined as the ratio of the thickness to the width of the filler particle. The Young's modulus of the clay platelets was taken as 170 GPa [10, 12], in all the cases.

Pukanszky's Model of Composite Yield Stress

Pukanszky et al. [13-15] estimated composite yield stresses using the following relation:

[[sigma].sub.c] = [[sigma].sub.m][[1 - [phi]]/[1 + 2.5[phi]]]exp(B[phi]) (7)

where [[sigma].sub.c] and [[sigma].sub.m] are composite and matrix yield stresses, [phi] is the volume fraction of the filler and B is a parameter, which can be evaluated from experimental data. The equation consists of two parts. The (1 - [phi])/(1 + 2.5[phi]) component takes into consideration the decrease in the effective load bearing cross section [16], and the exponent describes all other effects, which result in an increase of yield stress. The parameter B reflects the interfacial interaction, including the interlayer thickness, the interfacial strength and the specific surface area of the filler particles. The relationship between parameter B and the above characteristics is given below:

B = (1 + l[[rho].sub.f][S.sub.f])ln[[[sigma].sub.i]/[[sigma].sub.m]] (8)

where l, the thickness of interphase, is proportional to the interfacial adhesion [[gamma].sub.12] and is given by relation l = k[[gamma].sub.12]. The quantities [[rho].sub.f], [S.sub.f], and [[sigma].sub.i] represent the density of filler, the specific surface area of the filler, and the yield stress of the interphase, respectively. It is difficult to provide values for the length l and the yield stress [[sigma].sub.i] of the interphase. In spite of this limitation, the parameter B is useful in quantifying the effects of the above characteristics on the yield stress of the composite. Rong et al. [5] used this model to analyze the interfacial interactions in polypropylene nanocomposites and showed that higher values of parameter B are obtained, when the interfacial adhesion is high. Sumita et al. [17] used this model to describe dynamic mechanical properties of polypropylene composites filled with ultrathin particles. In the present study, the yield stress data of PA-6 nanocomposites were fit to Eq. 7, and parameter B was calculated.

EXPERIMENTAL

Polyamide-6 (PA-6) (Capron B135ZP) was purchased from BASF Corporation. Organoclays Cloisite 30B and Cloisite 15A, as well as untreated clay Cloisite [Na.sup.+], were purchased from Southern Clay Products.

Berstorff (Model ZE 25) co-rotating twin-screw extruder with intermeshing screws was used in the present study. The barrel diameter of the extruder was 25 mm, and the ratio of screw length to screw diameter was 30. The extruder barrel was divided into five zones. Prior to melt processing, the polymer and clay were dried in a vacuum oven at 80[degrees]C for 24 hr. The extruder screw was operated at 200 rpm. The temperature of each zone was controlled individually by an electrical heating band and a cooling fan. The barrel and die temperature settings ranged from 200[degrees]C to 260[degrees]C, and the maximum shear stress employed was of the order of [10.sup.5] N/[m.sup.2]. Sample ribbons were obtained by using a slit die. The polymer melt exiting the die was cooled with air fans located above and below the take-off guide rolls, while being pulled by a horizontal take-off system.

Two experimental systems were employed in the study. They will be referred to as System A and System C. Both systems employed the twin-screw extruder described earlier. In the case of System A, the extrusion was carried out to provide more intensive mixing and longer residence time, compared to system C. For example, the mean residence time calculated for System A at a polymer feed rate of 1.7 kg/hr was 333 sec, while it was 212 sec for System C, for the same polymer feed rate. However, the temperature settings and screw speeds were the same as those indicated above, in both cases.

[FIGURE 1 OMITTED]

The Rotaflex X-ray diffractometer with Cu-K[alpha] radiation ([lambda] = 0.154 nm) manufactured by Rigacu (Tokyo, Japan) was used for X-ray diffraction. For transmission electron microscopy (TEM) analysis, the samples were ultramic-rotomed to a thickness of less than 100 nm with a diamond knife, using Leica Ultracut S microtome at -120[degrees]C. The TEM photomicrographs were obtained with Jeol 2000 FX transmission electron microscope (Peabody, MA) and were recorded with a digital camera. The micrographs were analyzed with a system of cycloid test lines [18] and semiautomated image analysis. A line grid was superimposed on the micrograph. The number of intersections between the test lines and the polymer/clay interface were recorded for four quadrants [19, 20]. The procedure was repeated with 10 independent pictures. The distribution of the number of silicate sheets composing stacks of different heights was generated. By estimating the stack heights and the lengths of the particles, the aspect ratios were calculated.

The tensile properties of the samples were measured as per ASTM D638, using Instron Universal Testing Machine, Model 1123 R equipped with Series IX software. The crosshead speed was 5 mm/min for all the tensile tests. The sample ribbons obtained from the slit die were cut into the desired shape for tensile testing. The tensile strength is expressed in terms of the yield stress.

A Rheometrics Variable-Speed Impact Tester, RIT-8000, (Rheometrics, Union, NJ) was used to evaluate the impact properties of the samples. RIT-8000 is a controlled velocity impact tester, in which an electronically controlled hydraulic system drives a dart-shape probe to impact a flat specimen fixed between sample retaining rings. The force exerted on the probe, as a result of impacting the specimen, is measured by a load cell positioned at the tip of the rod and is plotted against the travel distance. The impact speed used was 1000 in./min for all the tests, and sample thickness was maintained the same for the different specimens compared.

[FIGURE 2 OMITTED]

RESULTS AND DISCUSSION

Compatibility Between Polymer Matrix and Organoclay

Hamaker Constants. Interactions between the clay and the polymer can be evaluated by estimating the effective Hamaker constant of the clay/polymer system. Since the chemical structures of the organic modifiers of Cloisite clays are known, the Hamaker constant values can be calculated using the group contribution method [3, 4]. The Hamaker constant values of the organic modifiers fall in the range of 5.3-6.0 x [10.sup.-20] J. The Hamaker constant values are higher for amides ([approximately equal to] 12 x [10.sup.-20] J), since they have polar functional groups capable of forming hydrogen bonds. The effective Hamaker constant in the case of the poly-amide matrix and organoclay is always positive ([approximately equal to]0.6 x [10.sup.-20] J). Polystyrene-based systems have slightly positive values ([approximately equal to] 0.1 x [10.sup.-20] J). Polyethylene shows values close to zero or negative, and poly(tetrafluoroethylene) (PTFE) always gives negative value. Nonpolar polymer matrices yield negative effective Hamaker constant values, which may lead to poor dispersion and tactoid formation. This suggests that the organoclay will be better dispersed in the polyamide matrix.

Work of Adhesion Between the Polymer Matrix and Clay Platelets

Figure 1 shows a schematic representation of the adhesion between the clay platelets and the polymer matrix. The effective Hamaker constant between the untreated clay platelet and the polymer matrix, [A.sub.12], is given by:

[A.sub.12] = [square root of ([A.sub.11][A.sub.22])] (9)

where [A.sub.11] and [A.sub.22] are the Hamaker constants of the pristine clay and the polymer, respectively. The effective Hamaker constant between the organically modified clay platelets and the polymer, [A.sub.132], is given by Eq. 1 as:

[A.sub.132] = ([square root of [A.sub.33]] - [square root of [A.sub.11]])([square root of [A.sub.33]] - [square root of [A.sub.22]]) (10)

where [A.sub.33] is the Hamaker constant of the organic modifier. The values of the Hamaker constant for different materials are given in the literature [2, 21-24]. The dispersion component of the work of adhesion can be estimated by using these values.

In the case of polyamide-6, the improvement in tensile properties may be attributed to the strong interaction between the matrix, the organic modifier, and the silicate layers, via the formation of hydrogen bonds. The bond energy of N-H hydrogen bonds is in the range 12-20 kJ/mol [25]. The specific surface area of exfoliated MMT clay is 700-786 [m.sup.2]/g [26, 27], and the cation exchange capacity is 0.92-0.95 mequiv./g [26]. Consider the case of polyamide-6 nanocomposites made from organically modified clay. The organic modifier in Cloisite 15A has no functional groups that would form hydrogen bonds. Thus, the work of adhesion arises only from the dispersion forces. On the other hand, Cloisite 30B has two hydroxyl groups per molecule of organic modifier, which can form hydrogen bonds with the polyamide matrix. If the specific surface area of the exfoliated clay is 760 [m.sup.2]/g, the organic modifier associated with it is (0.95/760) mequiv./[m.sup.2]. Since there are two hydroxyl groups per molecule of modifier, the PA-6 matrix associated with this clay via hydrogen bonds will be (2 x 0.95/760) mequiv./[m.sup.2]. The total bond energy of hydrogen bonds in the system will then be (2 x 0.95 x 20/760 x 1000) kJ/[m.sup.2]. This yields 5 x [10.sup.-5] kJ/[m.sup.2] or 5 x [10.sup.-2] J/[m.sup.2].

The work of adhesion from hydrogen bonding can be estimated for pristine clay by a different method. As shown in Fig. 2, the chain length of PA-6 between two hydrogen-bonding sites is 1.67 nm [28]. In a unit square meter, there will be 3.58 x [10.sup.17] hydrogen bonds. Since the bond energy is given as 12-20 kJ/mol, which is for the 6.022 x [10.sup.23] (Avogadro's number) bonds, the bond energy per square meter will be 1.19 x [10.sup.-5] kJ/[m.sup.2] or 1.19 x [10.sup.-2] J/[m.sup.2].

[FIGURE 3 OMITTED]

Table 1 summarizes the calculated values of the effective Hamaker constant, work of adhesion from dispersion forces, work of adhesion from hydrogen bonding, and the total work of adhesion between the clay platelets and the polymer matrix for different clay/polymer systems. These results show that the major contribution to the total work of adhesion for the PA-6/Cloisite 30B system comes from hydrogen bonding, which is absent in the case of the PA-6/Cloisite 15A system. The total work of adhesion is highest for the PA-6/Cloisite [Na.sup.+] system, which is untreated clay. The major contribution to the total work of adhesion in this case comes from dispersion forces. The platelets in the stack of pristine clay particle are close to each other (1 nm) and the adhesion between them is strong. As a result, untreated clay does not undergo exfoliation in the polymer matrix and the specific surface area is low. Therefore, the total adhesive force between the clay and polymer is low. However, if the untreated clay could be exfoliated by some means, it would produce nanocomposites with the highest property enhancements. The implications of this observation, in case of the degradation of the organic modifier during processing, are worthy of further examination.

[FIGURE 4 OMITTED]

[FIGURE 5 OMITTED]

X-ray Diffraction

Figure 3 shows the XRD patterns for PA-6/Cloisite 30B nanocomposite samples produced with System A at different clay concentrations, and Fig. 4 shows the XRD patterns for similar samples produced with System C. It would be expected that at the higher clay concentrations, the number of partially exfoliated and intercalated clay particles will be high. Fully exfoliated clay particles do not contribute to the intensity. System C samples, with 4.5 and 5.0 wt% clay content, show higher X-ray intensities at lower 2[theta] angles than System A samples. Thus, System C samples have more partially exfoliated and intercalated clay particles than System A samples.

Figure 5 shows the XRD patterns for PA-6/Cloisite 15A nanocomposites made with System A and System C for 4.1 wt% inorganic clay content. The gallery spacing increases marginally from 3.33 to 3.68 nm in both cases. The peak at 1.84 nm is a secondary peak of the 3.68 nm primary peak. The increase in gallery spacing by 0.35 nm does not mean that polymer chains are intercalated. Although the peak position is the same for both systems, the intensity of the diffraction peak is lower for the system A sample. If the larger silicate particles are broken into smaller tactoids which are located away form each other (distance more than 8 nm) in the polymer matrix, the X-ray diffraction intensity would be substantially lowered for the same concentration of clay. System A, by virtue of its higher mixing efficiency, tends to produce a larger number of smaller tactoids than System C. This may result in lowering the X-ray diffraction intensity. Organoclay Cloisite 15A has higher initial d-spacing, but it is nonpolar and does not have functional groups capable of forming hydrogen bonds, which results in partially exfoliated structure and tactoid formation. Untreated clay did not show any characteristic X-ray pattern.

Transmission Electron Microscopy

Figures 6 and 7 show TEM micrographs of PA-6/Closite 30B nanocomposites. They exhibit exfoliated structure. The individual silicate sheets appear as straight or slightly curved dark lines. Occasionally, some tactoids of clay are also seen. In System A samples, the clay platelets are more oriented in the flow direction than System C samples.

[FIGURE 6 OMITTED]

The micrographs of PA-6/Cloisite 15A samples are shown in Figs. 8 and 9. This is a partially exfoliated system, and clay tactoids of different sizes are observed. In this case, the System A micrograph exhibits a mixed distribution of individual platelets, small tactoids, and large tactoids. The System C micrograph indicates larger tactoids than in System A samples. The nanocomposites of untreated clay do not show exfoliated structure. Only large clay particles are observed in Figs. 10 and 11. Few clay tactoids or platelets are seen apart from large clay particle.

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

The micrographs of PA-6/Cloisite 30B and PA-6/Cloisite 15A nanocomposites were analyzed by the system of cycloid test lines [19, 20] and semiautomated quantitative image analysis. The aspect ratios of the particles were estimated from the average lengths and average thicknesses obtained in these analyses. Table 2 shows that PA-6/Cloisite 30B (5 wt%) nanocomposites, produced with System A and System C, contain clay particles with similar lengths but different thicknesses. The aspect ratio of the nanoclay particles in the System A sample is higher than that in the System C sample.

[FIGURE 9 OMITTED]

[FIGURE 10 OMITTED]

The analysis of TEM micrographs made it possible to estimate the volume fraction, surface density, specific surface area, and the degree of exfoliation. The surface density is the area generated at the interface between the particles and the matrix per unit volume of composite. It depends upon the clay content in the composite. The specific surface area is the surface area of the particles per unit mass of the particles and is independent of the clay content in the composite. The surface density is the product of the specific surface area, the volume fraction of the silicate particles, and the density of the silicate particles. In this analysis, the volume fraction of the silicate particles was estimated from TEM micrographs, as the ratio of the length of test line falling on the particles to the total length of the test line. The ratio of the specific surface area determined from TEM analysis to the maximum surface area of fully exfoliated montmorillonite clay (786 [m.sup.2]/g) [27] gives an estimate of the extent or degree of exfoliation. The results in Table 3 show that the specific surface area and the degree of exfoliation are higher for the PA-6/Cloisite 30B (5 wt%) nanocomposite prepared using System A than those prepared using System C. The System A specimen shows 94% exfoliation, whereas the System C specimen shows 70%. At lower clay content (3.1 wt%), even the System C sample shows 97% exfoliation. It can be concluded that System A makes it possible to achieve high dispersion at higher levels of clay content in PA-6 nanocomposites, In the case of partially compatible organoclay Cloisite 15A, the X-ray diffraction patterns and TEM micrographs do not show exfoliated structure. The specific surface area and the degree of exfoliation in these samples are low.

[FIGURE 11 OMITTED]

Tensile Properties of PA-6 Nanocomposites

Figure 12 shows plots of tensile modulus of the nanocomposites produced using System A as a function of clay content, for different organoclays. Cloisite 15A produces nanocomposites with higher modulus enhancement at lower clay concentrations. Cloisite 15A contains higher level of organic modifier (43%) and has higher gallery spacing (3.15 nm) than Cloisite 30B (organic content 30%, and gallery spacing 1.85 nm). Therefore, it may be easier to break the large particles of Cloisite 15A into smaller tactoids than those of Cloisite 30B. However, since Cloisite 30B has a greater tendency for producing exfoliated particles, samples incorporating this organoclay continue to show modulus enhancement at higher clay loadings. On the other hand, Cloisite 15A has lesser tendency to produce exfoliated structure. Thus, it is likely that reagglomeration of the tactoids may occur as the concentration of tactoids increases with the increase in clay loading. Therefore, Cloisite 15A samples show a leveling off of tensile modulus above a concentration of 3 wt%. Untreated Cloisite [Na.sup.+] clay yields modest improvement in the tensile modulus with the increase in clay loading, only with System A processing. This could be attributed to the breaking of some large clay particles and their efficient distribution throughout the matrix. The tensile moduli of nanocomposites prepared using System C with different organoclays (Fig. 13) show similar trends, but overall, the moduli are significantly lower than those of System A samples.

[FIGURE 12 OMITTED]

It can be seen from Figs. 14 and 15 that System A is more effective in improving tensile strength of the nanocomposites. Cloisite 15A and Cloisite [Na.sup.+] nanocomposites produced with System C show little improvement in tensile strength. Moreover, Cloisite 30B seems to be the most effective organoclay in improving tensile strength of the polyamide-6 nanocomposites.

[FIGURE 13 OMITTED]

[FIGURE 14 OMITTED]

Figure 16 shows that the ultimate impact force required to break the PA-6/Cloisite 30B nanocomposites produced with System A increases slightly with the addition of the clay. In case of Cloisite 15A and Cloisite [Na.sup.+], the impact strength decreases by the addition of the clay. As seen in Fig. 17, for the System C nanocomposites, the required ultimate impact force decreases with the increase in the clay content for all the clays.

Table 4 shows that the percentage elongation of dry PA-6, extruded using system A as well as System C, does not change significantly with the addition of different clays. The standard deviation in the percentage elongation values between the specimens was 4.

[FIGURE 15 OMITTED]

[FIGURE 16 OMITTED]

Modeling Results

The experimental tensile modulus data for the nanocomposites were fit to the Halpin-Tsai, MROM, and Hui-Shia models. The fitting of data to the model equations yielded [R.sup.2] values ranging from 0.87 to 0.94, except for the PA-6/Cloisite [Na.sup.+] composites produced with System C, in which case [R.sup.2] was 0.5. The aspect ratio values obtained from fitting the different models are given in Table 5. The Hui-Shia equations yielded aspect ratio values closer to those obtained by TEM analysis, and which are between the values obtained using the Halpin-Tsai and MROM equations. The aspect ratios are always higher for System A specimens. The aspect ratios were also higher for Cloisite 30B organoclay, followed by Cloisite 15A and Cloisite [Na.sup.+].

The data on yield stress of polyamide-6 nanocomposites were fit to Eq. 7, and the parameter B was evaluated. The values of [R.sup.2] ranged from 0.85 to 0.97. Table 6 shows the values of parameter B obtained for the different nanocomposite systems. Equation 8 suggests that the parameter B depends upon interfacial adhesion, density of the filler, the specific surface area of the filler, and the yield stress of the interphase. Since PA-6/Cloisite 30B nanocomposites represent the most compatible system, they should have the higher interfacial adhesion. This is reflected in the higher values of the parameter B. System A nanocomposites tend to contain filler particles with higher aspect ratios. Thus, the values of B are higher for System A samples. In the case of untreated clay, the work of adhesion is high, but the aspect ratios are low. The net effect is reflected in lower values of parameter B.

[FIGURE 17 OMITTED]

CONCLUSIONS

The effective Hamaker constant values for the different clay/polymer systems were calculated. The polyamide matrix always yielded positive and higher values, compared to the other polymers considered. This suggests stronger interactions between the clay particles and the polyamide melt, which results in better dispersion and exfoliation of the clay platelets.

The major contribution to the work of adhesion is attributed to hydrogen bonding for Cloisite 30B, whereas it derives from the dispersion component for Cloisite [Na.sup.+]. The total work of adhesion was highest for the untreated clay. It shows that if the untreated clay could be exfoliated in the polymer matrix, it would probably produce nanocomposites with the highest property enhancements.

X-ray diffraction and TEM micrographs show that Cloisite 30B produces exfoliated structures, Cloisite 15A yields partially exfoliated structure, and Cloisite [Na.sup.+] yields phase separated structure. The longer residence time and higher mixing efficiency of System A, coupled with Cloisite 30B, produces nanocomposites with high degrees of exfoliation. TEM results show that processing PA-6 containing 5% Cloisite 30B in System A produces nanocomposites with 94% degree of exfoliation (specific surface area: 760 [m.sup.2]/g). As a result the nanocomposites produced using Cloisite 30B exhibit higher tensile property enhancements, compared to Cloisite 15A and the untreated clay.

The Hui-Shia equations yielded aspect ratio values closer to those obtained by TEM analysis. These values were between the values obtained using the Halpin-Tsai and MROM equations. The fitted values of the aspect ratios are always higher for System A specimens. However, there are generally large differences between experimental TEM-based values of aspect ratios and those obtained by fitting the composite models. As expected, higher interactions between the clay and polymer, as well as System A processing, yield higher values of parameter B in Pukanszky's equation.

ACKNOWLEDGMENTS

DuPont Canada and Nova Chemicals supplied some of the materials employed in the research.

NOMENCLATURE Symbols/abbreviations expansion/definition A Hamaker constant B Parameter in Pukanszky's equation E Young's modulus G Shear modulus MRF Modulus Reduction Factor p Aspect ratio of filler particles PA-6 Polyamide-6 TEM Transmission electron microscopy W Work of adhesion XRD X-ray diffraction [phi] Volume fraction of filler [gamma] Interfacial adhesion [rho] Density [sigma] Yield stress

REFERENCES

1. P.C. LeBaron, Z. Wang, and T.J. Pinnavia, Appl. Clay Sci., 15,11 (1999).

2. A.W. Neumann, S.N. Omenyi, and C.J. van Oss, Colloid Polym. Sci., 257, 413, 737 (1979).

3. J. Vial and A. Carre, Int. J. Adhes. Adhes., 11, 40 (1991).

4. J.J. Jasper, J. Phys. Chem. Ref. Data, 4, 841 (1972).

5. M.Z. Rong, M.Q. Zhang, S.L. Pan, B. Lehmann, and K. Friedrich, Polym. Int., 53, 176 (2004).

6. S.W. Shang, J.W. Williams, and K.-J.M. Soderholms, J. Mater. Sci., 29, 2406 (1994).

7. J.N. Israelachvili, Intermolecular and Surface Forces, 2nd ed., Academic Press, London (1992).

8. J.C. Halpin and J.L. Kardos, Polym. Eng. Sci., 16, 344 (1976).

9. V.R. Riley, Presented at the Polymer Conference Series, University of Utah, Salt Lake City, UT (1976).

10. D. Shia, C.Y. Hui, S.D. Burnside, and E.P. Giannelis, Polym. Compos., 19, 608 (1998).

11. C.Y. Hui and D. Shia, Polym. Eng. Sci., 38, 774 (1998).

12. T.D. Fornes and D.R. Paul, Polymer, 44, 4993 (2003).

13. B. Pukanszky and E. Fekete, Adv. Polym. Sci., 139, 106 (1999).

14. B. Pukanszky and G. Voros, Compos. Interfaces, 1, 411 (1993).

15. B. Pukanszky, B. Turcsanyi, and F. Tudos, in Interfaces in Polymer, Ceramics and Metal Matrix Composites, H. Ishida, editor, Elsevier, New York, 467 (1988).

16. L. Nicolais and M. Narkis, Polym. Eng. Sci., 11, 194 (1971).

17. M. Sumita, H. Tsukihi, K. Miyasaka, and K. Ishikawa, J. Appl. Polym. Sci., 29, 1523 (1984).

18. C.V. Howard and M.G. Reed, Unbiased Stereology: Three Dimensional Measurements in Microscopy, Springer, New York (1998).

19. V. Mollet, M. Eng. Thesis, Chemical Engineering Department, McGill University, Montreal, Canada (2004).

20. N. Borse, M. Kamal, and V. Mollet, SPE ANTEC Tech. Pap., 63, 1417 (2005).

21. A. El Ghzaoui, J. Appl. Phys., 86, 2920 (1999).

22. E. Kissa, Fluorinated Surfactants and Repellents, Marcel Dekker, New York (1994).

23. C.J. Drummond and D.Y.C. Chan, Langmuir, 13, 3890 (1997).

24. V. Medout-Marere, J. Colloid Interface Sci., 228, 434 (2000).

25. J. Jancar, Adv. Polym. Sci., 139, 1 (1999).

26. D.L. Ho, R.M. Briber, and C.J. Glinka, Chem. Mater., 13, 1923 (2001).

27. J.C. Santamarina, K.A. Klein, Y.H. Wang, and E. Prencke, Can. Geotech. J., 39, 223 (2002).

28. S. Sinha Ray and M. Okamoto, Prog. Polym. Sci., 28, 1539 (2003).

Nitin K. Borse, Musa R. Kamal

Chemical Engineering Department, McGill University, Montreal, Quebec H3A 2B2, Canada

Correspondence to: Musa R. Kamal; e-mail: musa.kamal@mcgill.ca

Contract grant sponsors: Natural Sciences and Engineering Research Council of Canada; McGill University.

TABLE 1. Work of adhesion for different polymer/clay systems. Effective Hamaker [W.sub.a.sup.d] Clay/polymer constant, [A.sub.132] ([10.sup.-2]) system ([10.sup.-20]) (J) (J/[m.sup.2]) PA-6/Cloisite 30B 0.35 0.28 PA-6/Cloisite 15A 0.44 0.35 PA-6/Cloisite 9.7 7.7 [Na.sup.+] Total [W.sub.a] Clay/polymer [W.sub.a.sup.h] ([10.sup.-2]) ([10.sup.-2]) system (J/[m.sup.2]) (J/[m.sup.2]) PA-6/Cloisite 30B 5.0 5.28 PA-6/Cloisite 15A -- 0.35 PA-6/Cloisite 1.19 8.89 [Na.sup.+] TABLE 2. Aspect ratios of clay particles in PA-6 nanocomposites [19]. Length Thickness Aspect Sample (nm) (nm) ratio PA-6/Cloisite 30B (5 wt%), System A 140 1.9 74 PA-6/Cloisite 30B (5 wt%), System C 139 4.8 29 PA-6/Cloisite 30B (3.1 wt%), System C 117 2.5 47 TABLE 3. Surface density, specific surface area, and the degree of exfoliation of PA-6 nanocomposites [19]. Volume Surface density Material fraction ([micro][m.sup.-1]) Sodium montmorillonite -- -- PA-6/Cloisite30B (5 wt%), System A 0.021 55 PA-6/Cloisite30B (5 wt%), System C 0.025 48 PA-6/Cloisite30B (3.1 wt%), System C 0.011 29 PA-6/Cloisite15A (2.5 wt%), System A 0.009 8 PA-6/Cloisite15A (2.5 wt%), System C 0.010 11 Specific Degree of surface area exfoliation Material ([m.sup.2]/gm) (%) Sodium montmorillonite 786 (a) 100 PA-6/Cloisite30B (5 wt%), System A 736 94 PA-6/Cloisite30B (5 wt%), System C 549 70 PA-6/Cloisite30B (3.1 wt%), System C 760 97 PA-6/Cloisite15A (2.5 wt%), System A 245 31 PA-6/Cloisite15A (2.5 wt%), System C 292 37 "Ref. [27]. TABLE 4. Percentage elongation-at-break for PA-6 nanocomposites. Sample % Elongation-at-break PA-6 28 PA-6/Cloisite 30B (3.1 wt%), System A 25 PA-6/Cloisite 30B (5.0 wt%), System A 25 PA-6/Cloisite 30B (3.1 wt%), System C 24 PA-6/Cloisite 30B (5.0 wt%), System C 19 PA-6/Cloisite 15A (2.5 wt%), System A 23 PA-6/Cloisite 15A (4.1 wt%), System A 22 PA-6/Cloisite 15A (2.5 wt%), System C 28 PA-6/Cloisite 15A (4.1 wt%), System C 27 PA-6/Cloisite [Na.sup.+] (4.2 wt%), System A 24 PA-6/Cloisite [Na.sup.+] (6.9 wt%), System A 21 PA-6/Cloisite [Na.sup.+] (4.2 wt%), System C 31 PA-6/Cloisite [Na.sup.+] (6.9 wt%), System C 28 TABLE 5. Aspect ratios of the clay fillers for polyamide-6 nanocomposites calculated using different composite models. PA-6/Cloisite PA-6/Cloisite PA-6/Cloisite 30B 15A [Na.sup.+] System System System System System System Model A C A C A C Halpin-Tsai 16.7 10.4 16.1 6.3 3.1 0.4 MROM 130 80 125 49 27 6.7 Hui-Shia 47 29.2 45 17.7 8.9 1.4 TABLE 6. Values of parameter B for different nanocomposites. PA-6/30B PA-6/15A PA-6/[Na.sup.+] System System System System System System A C A C A C Parameter B 11.4 10.4 9.5 4.4 7.2 4.9

Printer friendly Cite/link Email Feedback | |

Author: | Borse, Nitin K.; Kamal, Musa R. |
---|---|

Publication: | Polymer Engineering and Science |

Date: | Aug 1, 2006 |

Words: | 6019 |

Previous Article: | Effect of crystallization on intercalation of clay-polyolefin nanocomposites and their performance. |

Next Article: | Epoxy nanocomposites curing by microwaves. |

Topics: |