《材料导论》课程教学资源（文献资料）Rheological and mechanical properties of nylon 6 nanocomposites submitted to reprocessing with single and twin screw extruders.pdf_大学文库

Rheological and mechanical properties of nylon 6 nanocomposites submitted to reprocessing with single and twin screw extruders G.M. Russo, V. Nicolais, L. Di Maio, S. Montesano, L. Incarnato* Department of Chemical and Food Engineering, University of Salerno, Via Ponte don Melillo, 84084 Fisciano (SA), Italy Received 1 March 2007; received in revised form 29 May 2007; accepted 24 June 2007 Available online 17 July 2007 Abstract The effect of multiple extrusions on nanostructure and properties of nylon 6 nanocomposites was investigated. Nanocomposites at different silicate loadings were produced by melt compounding and submitted to further reprocessing by using single and twin screw extruders. Rheological, morphological and mechanical analyses were carried out on as-produced and reprocessed samples in order to explore the influence of the number and the type of extrusion cycles on silicate nanodispersion. Rheological measurements, correlated to TEM analyses, were used to probe the nanoscale arrangement developed with the reprocessing as well as the thermo-mechanical degradation involving both the neat matrix and the organoclay. The results have shown that the reprocessing by single screw extruder can modify the initial morphology since the re-agglomeration of the silicate layers can occur. On the other hand, a better nanodispersion was observed in the hybrids reprocessed by twin screw extruder. This was attributed to the additional mechanical stresses able to realizing a dispersive mixing that contributes to avoid re-agglomeration phenomena. The high shear stresses produced with twin screw geometry determined also a significant degradation of neat matrix, principally based on chain scission mechanism. A strong correlation between the rheological behaviour and mechanical properties was observed and all as-produced and reprocessed hybrids showed a substantial enhancement in tensile modulus with the adding of silicate. However, the entity of performance enhancements displayed by the reprocessed hybrids was found to be highly dependent on the degradation of both organoclay and polymer matrix as well as the silicate amount, the number and the type reprocessing. 2007 Elsevier Ltd. All rights reserved. Keywords: Nylon 6 nanocomposites; Melt compounding; Reprocessing; Degradation; Single screw extrusion; Twin screw extrusion 1. Introduction An innovative strategy to enhance the properties of nylon 6 can be offered by reinforcing the polyamide with nanoscopic layered silicates. The technological relevance of polymer layered silicate nanocomposites is testified by the numerous patents issued over the last few years that highlight significant increases in structural and functional properties with the addition of very low silicate content, usually less than 5 wt% [1e4]. This feature appears extremely attractive with regard to plastics reprocessing, because nanocomposites can be easily extruded or moulded with the conventional processing equipments, leading to the same performances of traditional composites with the advantages of reduced weight and costs. Knowledge of microstructural changes arising from reprocessing of nanocomposites is important for evaluating the economical and potential environmental impact of this relatively new class of materials. Although a significant scientific activity has occurred with regard to the use of polyamides as polymer matrices [5e7], a systematic study on the effects of the reprocessing on nanodispersion and the end properties of polyamide based nanocomposites has not been fully performed. The few works on recyclability of polymer nanocomposites principally concern polyolefins, PET and polyamide 12 matrices and indicate decreases in mechanical properties with the extrusion cycles, that are consistent with degradation [8e10]. * Corresponding author. Tel.: þ39 089 964144; fax: þ39 089 964057. E-mail address: lincarnato@unisa.it (L. Incarnato). 0141-3910/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymdegradstab.2007.06.010 Polymer Degradation and Stability 92 (2007) 1925e1933 www.elsevier.com/locate/polydegstab

It is worth pointing out that the degradation of a nanocomposite system is extremely complex to analyze because it depends not only on the characteristics of the neat components but also on the possible interactions between polymer and organoclay and between their degradation products [11]. For most polymers, the main problem arising during reprocessing operations is the reduction of the molecular weight due to the breaking of molecular bonds [12e14]. In the case of polyamides the changed molecular weight after the reprocessing is not easy to predict because different degradation mechanisms can occur like main-chain scission or cross-linking [15e17]. The final molecular weight and its distribution will depend on the initial one and on which degradation mechanism results predominant. Moreover, the number of reprocessing operations and the levels of mechanical stresses which the polymer undergoes also affect the extent of degradation and the final properties [18e20]. With regard to nanocomposite systems, the presence of the organoclay inside the polyamide matrix contributes to degradation either because the organic part of the silicate could itself be involved in the degradation phenomena, either because the clay naturally tends to absorb water and hence to promote the chain scission of the polyamide [11,21]. From these considerations it clearly appears that multiple extrusions of polyamide nanocomposites can provoke strong modifications of the structural parameters of the neat matrix (as molecular weight and crystallinity) and of the nanoscale arrangement of the silicate, strongly sensitive to shear levels in the extruder [5,22,23]. The aim of the present work is to establish the correlation existing between reprocessing, nanomorphology and mechanical properties of nylon 6 based nanocomposites when submitted to multiple extrusions. Hybrids with different silicate loadings were produced by melt compounding with a twin screw extruder in order to achieve the higher degrees of silicate exfoliation. All the samples were submitted to further reprocessing by using both single and twin screw extruder. Rheological, morphological and mechanical analyses were carried out on asproduced and reprocessed samples in order to explore the effect of different parameters, such as number of extrusions and type of extruder, on nanodispersion and properties of the recycled systems. 2. Experimental 2.1. Materials The polymer used in the production of nanocomposite hybrids was a nylon 6 standard (PA6 F34L, IV ¼ 3.4 dl/g) supplied by Caffaro SpA (Italy). The nanofiller was an organically modified silicate named Cloisite 30B by Southern Clay Products, Inc., consisting in a layered sodium montmorillonite organically modified by methyl, tallow, bis-2-hydroxyethyl, quaternary ammonium chloride (90 meq/100 g clay), with an interlayer basal spacing d001 ¼ 18.5 A˚ . 2.2. Melt processing and reprocessing Nanocomposites at 3 and 6 wt% of silicate were prepared by melt compounding using a laboratory counter-rotating, non-intermeshing twin screw extruder with L ¼ 500 mm and L/D ¼ 20 and a rectangular die of 1 mm 40 mm. A screw speed of 60 rpm was used and a temperature profile of 265e 260e255e245 C from hopper to die was imposed. Prior the processing, the materials were dried in a vacuum oven at 90 C for 18 h obtaining a moisture level below 0.2 wt% in order to avoid bubble formation and polymer degradation during processing [24]. The mechanical recycling of the samples was performed by submitting all the samples to a further two extrusion stages using either twin or single screw extruder, as illustrated in Fig. 1. The use of the twin screw extruder in the first stage was fundamental in realizing an homogeneous dispersion and high levels of silicate exfoliation. The twin screw extruder and the conditions used in reprocessing were the same of the first extrusion. The single screw extruder was a Brabender Do-Corder 330 with L ¼ 400 mm and L/D ¼ 20 and a rectangular die of 1 mm 40 mm which was operated with the same temperature profile and a screw speed of 50 rpm. A coloured marker pigment was added to the raw material as a tracer to evaluate the mean residence time in two extruders that were about 65 and 45 s for twin and single screw extruders, respectively. 2.3. Characterization Rheological measurements in oscillatory mode were performed with a Rheometric Scientific rotational rheometer, ARES (Advanced Scientific Expansion System) in an angular Melt compounding with twin screw extrusion 2º twin screw extrusion 2º single screw extrusion 3º twin screw extrusion 3º single screw extrusion Neat nylon 6; 3 and 6 wt nanocomposites Fig. 1. Scheme of single and twin screw reprocessing of nylon 6 and its nanocomposites at 3 and 6 wt% of silicate. 1926 G.M. Russo et al. / Polymer Degradation and Stability 92 (2007) 1925e1933

GM.Russo Polymer Degradation and 9 07)1925-1933 1927 frequency range from 0.1 to 100 rad/s at the temperature of demonstrated by our previous studies and many others from surements in order to ensure the linear viscoelasticity of sely to other exr erimental techniaues that nr ohe a ver the dynamic tests.All rheological experiments were conducted curring with reprocessing being very sensitive to the nanoscale e ndirectoWiCaptredonsectig to th 2a.b th of ultra-thin specimens with a Leica Utracut UCT microtome es of the neat nylor 6 submitted to three extrusion cvcles are compared.re Mechanical tests in tensile mode were performed according ively.for single and twin screw reprocessing.It can be noted h ASTM rom Fig.2a sing with sing erties was evaluated on the basis of olymer beeause the viscosit curves relate to the thr 10 measures per sample sion cycles are similar toeach other although a slight increas nt samples was calcu values is erved after the cond extrusion.Ar the 6.ple second extrusion is also observed.This behaviour could be at- quence radation eads to 3.Results and discussion with single screw extruder,the third processing perfor The mechanical recycling of the samples was performed in ith twin s Fig.1.The to lead to redu e al we necessity to generate substantial dispersive mixing in order to erved.This pehaviour could be attributed to the main-chain uniformly disperse them nd re: screw geo extruders with the aim to investigate the effect of diffe vith respect to single screw extruder 127 281 Both the cross stress fields on nanostructure and end-performances of the linking and the chain scission of nylon 6macromolecules thus sample em to be inve duning the repro the role of th itial of the tu the higher nanoeomposites.heological analyses in oscillatory mode Fis.hows the viscosity curves and the relative TEMim were performed on as-produced and reprocessed samples.As ages of 3 wt nanocomposite sample submitted to the 1E404 1E+04 n(Pas) (a) single screw reprocessing (b) twin screw reprocessing 1E403 1.E403 1E402 10 0.1 10 (rad/s) (rad/s) Fig.2.Complex viscosity of the eat nylon 6 sub ssing:(b)twin

frequency range from 0.1 to 100 rad/s at the temperature of 255 C using 25 mm diameter parallel plates. The deformation of 1% of strain amplitude was determined from strain sweep measurements in order to ensure the linear viscoelasticity of the dynamic tests. All rheological experiments were conducted under a nitrogen atmosphere to prevent oxidative degradation of the specimens, and prior to the tests, the samples were dried for 16 h in vacuum at 85 C. Transmission electron microscopy (TEM) analyses were conducted using a Philips EM 208 at different magnification levels. The images were captured on sections normal to the extrusion direction which had been prepared by microtoming of ultra-thin specimens with a Leica Utracut UCT microtome. Mechanical tests in tensile mode were performed according the ASTM D-638 procedure using a Dynamometer (Instron 5566) with a cross speed of 50 mm/min. The average value of the mechanical properties was evaluated on the basis of 10 measures per sample. Crystallinity degree, Xc, of the different samples was calculated by the ratio of heat of fusion of the sample on heat of fusion of the purely crystalline nylon 6, i.e. 240 J/g [25]. The heat of fusion was evaluated with a Mettler differential scanning calorimeter (DSC30) imposing a heating from 0 to 260 C at 10 C/min. 3. Results and discussion The mechanical recycling of the samples was performed in accordance with the scheme reported in Fig. 1. The choice of twin screw design for the first reprocessing was dictated by the necessity to generate substantial dispersive mixing in order to break up silicate agglomerates and uniformly disperse them inside the polymer matrix [5,7,26,27]. The further reprocessing was instead realized by using both twin and single screw extruders with the aim to investigate the effect of different stress fields on nanostructure and end-performances of the reprocessed samples. In order to study the role of the number and type of reprocessing operations on the initial morphology of polyamide 6 nanocomposites, rheological analyses in oscillatory mode were performed on as-produced and reprocessed samples. As demonstrated by our previous studies and many others from literature [1,7,23], the rheological response of a nanocomposite is strictly correlated with the nanostructure in solid state and, conversely to other experimental techniques that probe a very small sample volume (X-ray diffraction of even TEM micrography), is representative of the global bulk nanostructure. In particular, for the present case study, rheology could appear a powerful tool to discriminate the structural modifications occurring with reprocessing being very sensitive to the nanoscale arrangement of the silicate layers as well as to the viscoelastic characteristics of the neat polymer matrix. In Fig. 2a,b the complex viscosity curves of the neat nylon 6 submitted to three extrusion cycles are compared, respectively, for single and twin screw reprocessing. It can be noted from Fig. 2a that the reprocessing with single screw extruder does not significantly affect the viscoelastic behaviour of the polymer because the viscosity curves related to the three extrusion cycles are similar to each other although a slight increase in viscosity values is observed after the second extrusion. Analyzing the curves of nylon 6 reprocessed by twin screw extrusion (Fig. 2b) an increase in the complex viscosity after the second extrusion is also observed. This behaviour could be attributed to cross-linking phenomena that occur as a consequence of degradation of nylon 6 which leads to an increasing of flow resistance, in agreement with the studies of Lozano-Gonzales et al. [18]. Conversely to what happens with single screw extruder, the third processing performed with twin screw extruder seems to lead to a reduction of the molecular weight of nylon 6 since a strong decrease in viscosity accompanied by a more limited Newtonian plateau is observed. This behaviour could be attributed to the main-chain scission mechanism promoted by the higher shear stresses and residence time realized with twin screw geometry which determine more relevant structural and chemical modifications with respect to single screw extruder [27,28]. Both the crosslinking and the chain scission of nylon 6 macromolecules thus seem to be involved during the reprocessing with the twin screw extruder and the final structure will depend on which of the two opposite phenomena has the higher probability. Fig. 3 shows the viscosity curves and the relative TEM images of 3 wt% nanocomposite sample submitted to the (a) (b) 1.E+02 1.E+03 1.E+04 0.1 1 10 100 ω (rad/s) η* (Pa s) η* (Pa s) as extruded 2 extrusion 3 extrusion as extruded 2 extrusion 3 extrusion single screw reprocessing 1.E+02 1.E+03 1.E+04 0.1 1 10 100 ω (rad/s) twin screw reprocessing Fig. 2. Complex viscosity curves of the neat nylon 6 submitted to three extrusion cycles: (a) single screw reprocessing; (b) twin screw reprocessing. G.M. Russo et al. / Polymer Degradation and Stability 92 (2007) 1925e1933 1927

reprocessing cycles performed with the single screw extruder. It is possible to observe that the complex viscosity of 3 wt% hybrid after the second extrusion is similar to those of the as-extruded sample while it decreases after the third, although it remains higher than the neat matrix. With regard to the third extrusion, it has to be pointed out that the reduced viscosity of the 3 wt% hybrid with respect to the as-extruded sample cannot be associated to the degradation of the neat matrix because, as discussed before, the neat matrix seems to be not involved in significant degradative phenomena when reprocessed with single screw extruder (see Fig. 2a). The decreased viscosity of 3% hybrid after the third extrusion cycle could be more reasonably related to a change of silicate nanoscale arrangement, as shown by a much less uniform nanodispersion of the silicate from the corresponding TEM images in Fig. 3. In this regard, it should be taken into account that the nanofiller used is constituted by an organic modifier (32 wt%) which can undergo thermal degradation due to the processing temperatures and residence times inside the apparatus during the reprocessing. In Fig. 4 the TGA curve of Cloisite 30B is reported with the derivative of weight loss in order to better appreciate the critical degradation temperature of the organoclay. It is possible to observe that thermal degradation of Cloisite 30B is a complex process evolving a temperature range between 200 and 1000 C, according to many literature reports [29e32]. After a minor weight loss above 50 C, assigned to a loss of physically adsorbed water, the degradation of Cloisite 30B is a multi-step process, involving the decomposition of both free and confined organic modifiers, the production of gaseous species, the dehydroxylation of the aluminosilicate lattice and the evolution of products associated with organic carbonaceous residue. Since the first significant weight loss occurs at 268 C and the processing temperatures are in the range of 250e260 C (see Section 2), it clearly appears that the thermal stability of the organoclay is sensitive to processing temperatures and the extent of its decomposition will depend on the specifics of timee temperature history. Therefore, a multiple extrusion process could affect the thermal stability of the alkyl-ammonium surfactant even leading to a collapse of the entire silicate structure [31,32]. This is confirmed by TEM image of 3% hybrid after the third extrusion cycle that shows bigger silicate stacks if compared to the as-extruded sample (Fig. 3). Due to the additional processing temperatures and residence times inside the apparatus, the third extrusion cycle seems hence to promote a sort of ‘‘re-aggregation’’ of silicate layers. 1.E+02 1.E+03 1.E+04 0.1 1 10 100 ω (rad/s) η*(Pa s) as extruded 2 extrusion 3 extrusion single screw reprocessing Fig. 3. Complex viscosity curves of 3 wt% nanocomposite reprocessed with single screw extruder. Micrographs of the hybrids at first and third extrusion are related to the appropriate curves. 70 80 90 100 0 200 400 600 800 1000 1200 T (ºC) residual mass ( ) 0 0.5 1 1.5 2 deriv.wt (g/ºC) weight loss derivative 268.56 ºC 382.89 ºC 573.95 ºC 801.72 ºC 1001.05 ºC Fig. 4. TGA curves and loss weight derivate of pristine silicate used in this study. 1928 G.M. Russo et al. / Polymer Degradation and Stability 92 (2007) 1925e1933

These behaviours are confirmed by the analysis of the viscoelastic response of the 6 wt% nanocomposite, reported in Fig. 5. In particular, the flow curve of the second cycle extruded sample is overlapped with those of the as-extruded, while it decreases in the case of the three times extruded sample. The major presence of silicate inside the 6% nanocomposite emphasizes the effects of the thermal degradation of the organoclay on the final nanostructure of the hybrid: the 6 wt% reprocessed nanocomposite exhibits in fact a much less uniform nanodispersion when compared to as-extruded sample as observed by the TEM images in Fig. 5. On the basis of our results and on the data reported in literature [6,28,33], it clearly appears that the delamination of a nanocomposite hybrid reprocessed with a single screw extruder cannot be improved but even reduced by silicate re-agglomeration. In order to investigate the effect of extruder type on the structural modifications induced by reprocessing, the viscosity curves and the TEM images of 3 wt% nanocomposite at different extrusion cycles performed with twin screw extruder are compared in Fig. 6. The second twin screw extrusion leads to a significant increase of complex viscosity. For instance, at the frequency of 1 rad/s the complex viscosity of as-extruded 3% hybrid is 1760 Pa*s, whilst the complex viscosity of the sample submitted to a second extrusion is 3108 Pa*s showing an increment of 76%. Moreover, it should be noted that the twin screw reprocessing determines an evident change in flow curve shape since the reprocessed sample exhibits a shear thinning behaviour at high frequencies when compared with the as-extruded sample. Many studies on viscoelastic response of filled polymers submitted to shear flow have associated the shear thinning behaviour to the presence of a structural network formed by the dispersed particles [1,23,33,34]. The more pronounced shear thinning behaviour displayed by twin screw extruded sample suggests a more exfoliated structure with respect to the as-produced sample, as highlighted by TEM image in Fig. 6 which shows the presence of more high aspect ratio particles for the reprocessed sample. It is worthy to note that several theoretical and experimental studies have demonstrated that a sufficient stress able to shear the silicate stacks into smaller platelets must be produced during the melt compounding of polymer silicate systems to obtain satisfactory degrees of silicate exfoliation [23,33]. Therefore, the finer silicate dispersion inside the polymer matrix observed with twin screw reprocessing could be related to the additional shear stresses that contribute to realize a dispersive mixing and to avoid the re-agglomeration phenomena. As concerned the third twin screw extrusion, the flow curve of the 3 wt% hybrid is characterized by a significant reduction of viscosity with values at low frequencies similar to those of as-extruded sample (Fig. 6). Even if the organoclay is involved in thermal degradation during the reprocessing, the last feature cannot be only due to the re-aggregation of the silicate layers because the TEM image of 3 wt% hybrid submitted to three twin screw extrusions does not show significant nanoscale modifications with respect to the same sample submitted to two extrusions (see Fig. 6). The drop of viscosity with the third extrusion can be more reasonably associated to the chain scission phenomena involving the neat matrix that, as shown and discussed in Fig. 2b, appear evident only after the third extrusion cycle performed with twin screw extruder. On the basis of such results, two principle factors seem to match during the reprocessing of our systems with twin screw extruder. The first is represented by the higher levels of shear stresses and residence time produced by the twin screw extruder with respect to the single extruder that improve the nanoscale arrangement of the silicate layers and give a pronounced shear thinning behaviour to the correspondent flow curve. The second contribution is related to the thermo-mechanical degradation of the neat matrix that tends to reduce the viscosity values leaving almost unaltered the shape of the flow curve. 1.E+02 1.E+03 1.E+04 0.1 1 10 100 ω (rad/s) η* (Pa s) as extruded 2 extrusion 3 extrusion single screw reprocessing Fig. 5. Complex viscosity curves of 6 wt% nanocomposite reprocessed with single screw extruder. Micrographs of the as-extruded and three times extruded are related to the appropriate curves. G.M. Russo et al. / Polymer Degradation and Stability 92 (2007) 1925e1933 1929

extrusion, although the degradation of the polymer matrix occurs. A study conducted by Dabrowski has demonstrated that the presence of nanoclay inside nylon 6 matrix slows down the rate of thermal degradation of the neat polymer [35]. On the basis of this observation, it could be also possible that the larger amount of silicate in the 6 wt% nanocomposite inhibits the degradation of the matrix more than the 3% hybrid. In order to correlate the nanostructure developed during the reprocessing with the final performances of the different hybrids, mechanical tests in tensile mode were performed on as-produces and reprocessed samples. In Table 1 the principle mechanical properties and the crystallinity degrees of the samples submitted to the reprocessing with single screw extruder are summarized. The mechanical properties of the neat matrix seem to be not significantly affected by single screw reprocessing. Analyzing the properties of nanocomposite samples, it clearly appears that the presence of silicate leads to a substantial enhancement in tensile modulus at any extrusion cycle, suggesting that the reinforcement with a nanolayered silicate may represent an interesting solution to improve the stiffness of a recycled polymer. Moreover all the hybrids, as-produced and reprocessed, show a strong reduction in the elongation at break with respect to the correspondent neat matrix, likely related to the reinforcement contribute from the nanofiller [25,36,37]. Table 2 reports the mechanical properties and crystallinity degrees of the samples submitted to reprocessing with twin screw extruder. Conversely to what happens with single screw extruder, the crystallinity of the neat matrix seems to decrease with twin screw reprocessing, accompanied by a worsening in stiffness at the third extrusion. Moreover, at any twin screw reprocessing cycle, the presence of the silicate determines a significant improvement in stiffness, in particular for the 6% nanocomposite sample. In order to highlight the role of extruder type on the mechanical performances of the reprocessed hybrids, the tensile moduli of 3 and 6 wt% nanocomposite are compared in Fig. 8a,b as a function of single and twin screw extrusion cycles. It is possible to note from Fig. 8a that the stiffness of the 3% hybrid after the third twin screw extrusion results lower than single screw reprocessed nanocomposite, reasonably due to a decrease in mechanical properties of the neat matrix. Moreover the tensile modulus of 3% hybrid reprocessed with twin screw increases after the second extrusion and decreases after the third extrusion, similarly to the correspondent complex viscosity curves reported in Fig. 6. From these results it clearly appears that the final performance of twin screw reprocessed nanocomposites is related to two opposite phenomena that are the improved silicate nanodispersion and the thermalemechanical degradation of the matrix. Both the effects are promoted by the additional dispersive shear stresses produced in twin screw extruder. It is interesting to observe from Fig. 8b that, for any extrusion cycles, the stiffness of 6 wt% hybrid reprocessed with twin screw extruder results always higher than the corresponding single screw reprocessed sample, although the evident degradation of the polymer matrix promoted by twin screw geometry. Moreover, the stiffness of such a hybrid increases with increasing the number of twin screw reprocessing, again in strong agreement with the viscoelastic response reported in Fig. 7. The major presence of silicate inside the 6% nanocomposite allows to emphasize the effects produced by dispersive mixing on the nanoscale arrangement. As shown by TEM images in Fig. 7, the 6 wt% reprocessed hybrid results in fact characterized by the presence of more exfoliated regions with respect to the as-extruded sample, so that the final mechanical properties will be more sensitive to the better silicate nanodispersion rather than to polymer matrix degradation. This interpretation is confirmed by the behaviour of the elongation at break in function of single and twin screw number of extrusion, reported for 3 and 6 wt% nanocomposites in Fig. 9a,b, respectively. Both the nanocomposites reprocessed with single screw extruder show a strong reduction of the elongation at break with respect to as-extruded hybrid, while those reprocessed with twin screw extruder exhibit an enhanced ductility with extrusion cycles, particularly evident for the 6% nanocomposite after the second extrusion (Fig. 9b). The reduced elongation at break displayed by the single screw extruder reprocessed hybrids suggests the presence of bigger sized silicate stacks, confirmed by TEM analyses, which Table 1 Tensile modulus (E ), relative modulus (E/Em), stress at break (sb), elongation at break (3b) and crystallinity degree (Xc) for nylon 6 nanocomposites submitted to single screw extruder reprocessing Extrusion cycles % Silicate E (MPa) Relative modulus sb (MPa) 3b (%) Xc (%) 1 0 720 1 37 260 29 3 950 1.31 42 160 30 6 1030 1.43 41 80 30 2 0 790 1 38 180 27 3 1100 1.26 39 90 28 6 1040 1.31 43 70 29 3 0 750 1 39 200 30 3 1040 1.38 47 90 30 6 1170 1.56 44 40 30 Em represents the tensile modulus of neat matrix. Table 2 Tensile modulus (E ), relative modulus (E/Em), stress at break (sb), elongation at break (3b) and crystallinty degree (Xc) for nylon 6 nanocomposites submitted to twin screw extruder reprocessing Extrusion cycles % Silicate E (MPa) Relative modulus sb (MPa) 3b (%) Xc (%) 1 0 720 1 37 260 29 3 950 1.31 43 160 30 6 1030 1.43 41 80 30 2 0 800 1 42 200 25 3 1000 1.25 43 170 29 6 1100 1.37 42 200 31 3 0 650 1 34 220 26 3 730 1.12 34 200 26 6 1200 1.84 43 180 28 Em represents the tensile modulus of neat matrix. G.M. Russo et al. / Polymer Degradation and Stability 92 (2007) 1925e1933 1931

behave as stress concentrators. This is in accordance with many studies on mechanical properties of nylon 6 based nanocomposites which ascribe the typical brittle fracture displayed by the hybrids to the presence of unexfoliated aggregates inside the polymer matrix [6,7]. The ductility of the twin screw reprocessed hybrids is instead a clear indication that the twin screw reprocessing provides a finer nanodispersion which improves polymereclay adhesion and facilitates the mechanical stress transfer to the reinforcement phase. 4. Conclusions In this work nylon 6 nanocomposites at different silicate loadings (3 and 6 wt% of organo-modified montmorillonite) were produced by melt compounding with a twin screw extruder and submitted to further reprocessing by using single and twin screw extruder. Rheological, morphological and mechanical analyses have been carried out on as-produced and reprocessed samples. The results have shown that the reprocessing can modify the morphology of the nanocomposite hybrids affecting the degree of silicate exfoliation and the mechanical properties. The entity of structural and performance modifications results highly dependent on matrix degradation, silicate amount, number and type of reprocessing. In particular, the rheological and TEM analyses have shown that the reprocessing with single screw extruder leads to a sort of re-agglomeration of the silicate layers, as evidenced by the bigger stack dimensions of the reprocessed hybrids, reasonably related to the thermal degradation of the organoclay, sensitive to the processing temperature. By using a twin screw extruder the reprocessing resulted instead efficacious in improving the silicate exfoliation due to the additional mechanical stresses able to realizing a dispersive mixing and to avoid re-agglomeration phenomena. Compared to the as-extruded samples, TEM images of twin screw reprocessed (a) 0 50 100 150 200 250 1 3 extrusion cycles εb ( ) εb ( ) single screw twin screw (b) 0 50 100 150 200 250 2 3 extrusion cycles single screw twin screw 1 2 Fig. 9. Elongation at break at varying single and twin screw extrusion cycles: (a) 3 wt% nanocomposite; (b) 6 wt% nanocomposite. (a) 0 200 400 600 800 1000 1200 1400 123 extrusion cycles E (MPa) single screw twin screw (b) 0 200 400 600 800 1000 1200 1400 123 extrusion cycles E (MPa) single screw twin screw Fig. 8. Tensile modulus behaviour of nylon 6 nanocomposites at varying single and twin screw extrusion cycles: (a) 3 wt% nanocomposite; (b) 6 wt% nanocomposite. 1932 G.M. Russo et al. / Polymer Degradation and Stability 92 (2007) 1925e1933

hybrids displayed in fact a more uniform nanoscale dispersion, characterized also by increases in complex viscosity and shear thinning behaviour. However, the twin screw reprocessing could imply degradation phenomena of the neat matrix resulting in worsening of mechanical properties. The mechanical tests have shown a substantial enhancement in tensile modulus with the addition of silicate, suggesting that the reinforcement with a nanolayered silicate may represent an interesting solution to improve the stiffness of a recycled polymer. The better nanodispersion promoted by twin screw reprocessing was emphasized also by the behaviour of the elongation at break in function of number of extrusion. All the hybrids reprocessed with twin screw extruder exhibited an enhanced ductility with respect to as-extruded, clear indication of a finer nanodispersion and improved polymereclay adhesion provided by twin screw reprocessing. The hybrids reprocessed with single screw extruder showed instead a strong reduction of the elongation at break with respect to as-extruded hybrid, due to the presence of bigger sized silicate stacks creating discontinuities in stress transfer. References [1] Pinnavaia TJ, Beall GW, editors. Polymer-layered silicate nanocomposites. NY: John Wiley & Sons Ltd.; 2001. [2] Giannelis EP, Krishnamoorti R, Manias E. Adv Polym Sci 1999;138:107. [3] LeBaron PC, Wang Z, Pinnavaia T. J Appl Clay Sci 1999;15:11. [4] Maiti P, Okamoto M. Macromol Mater Eng 2003;288:440e5. [5] Dennis R, Hunter DL, Chang D, Kim S, White JL, Cho JW, et al. Polymer 2001;42:9513e22. [6] Fornes TD, Yoon PJ, Keskkula H, Paul DR. Polymer 2001;42:9929e40. [7] Russo GM, Simon GP, Incarnato L. Macromolecules 2006;39:3855. [8] Thompson MR, Yeung KK. Polym Degrad Stab 2006;91:2396e407. [9] Pegoretti A, Kolarik J, Peroni C, Migliaresi C. Polymer 2004;45:2751e9. [10] Lew CY, Murphy WR, McNally GM. SPE Annu Tech Conf 2004;50:299e308. [11] Fornes TD, Yoon PJ, Paul DR. Polymer 2003;44:7545e56. [12] Brandup J, editor. Recycling and recovery of plastics. Hanser; 1996. [13] Mod Plast January 1997;76e8. [14] Nguyen KT, Weber L, Hebert L. ANTEC’93 1993;3432. [15] Osborn KR, Jenkins WA. Plastic films: technology and packaging applications. Lancaster: Technomic Publishing Company, Inc.; 1992. [16] Zilberman M, Siegman A, Narkis M. Polymer 1995;36:5065e7. [17] Siciliano A, Severgnini D, Seves A, Pedrelli T, Vicini LJ. Appl Polym Sci 1996;60:1757e64. [18] Lozano-Gonzales MaJ, Rodriguez-Hernandez MaT, Gonzales-De Los Santos EA, Villapando-Olmos J. J Appl Polym Sci 2000;76:851e8. [19] Schiebisch J, Ehrenstein GW. ANTEC’95 1995;3757. [20] Marek B, Lerch EJ. Soc Dye Color 1965;81:481. [21] Alexandre M, Dubois P. Mater Sci Eng 2000;28:1e63. [22] Incarnato L, Scarfato P, Russo GM, Di Maio L, Iannelli P, Acierno D. Polymer 2003;44:4625e43. [23] Incarnato L, Scarfato P, Scatteia L, Acierno D. Polymer 2004;45:3487e96. [24] Khanna YP, Han PK. Polym Eng Sci 1996;36:1745. [25] Fornes TD, Paul DR. Polymer 2003;44:3945e61. [26] Ferry JD. Viscoelastic properties of polymers. 3rd ed. NY: John Wiley & Sons Ltd.; 1980. [27] Utracki LA. Two-phase polymer systems. New York: Oxford Univ. Press; 1991. [28] Rauwendaal C. Ed. Mixing in polymer processing. New York; 1991. [29] Xie W, Gao Z, Pan W-P, Hunter D, Singh A, Vaia R. Chem Mater 2001;13:2979e90. [30] Davis RD, Gilman JW, VanderHart DL. Polym Degrad Stab 2003;79:111e21. [31] Monticelli O, Musina Z, Frache A, Bellucci F, Camino G, Russo S. Polym Degrad Stab 2007;92:370e8. [32] Shah RK, Paul DR. Polymer 2006;47:4075e84. [33] Tanoue S, Utracki LA, Garcia-Rejon A, Tatibouet J, Kamal MR. Polym Eng Sci 2005;827e37. [34] Wagener R, Reisinger TJG. Polymer 2003;44:7513e8. [35] Dabrowski F, Bourbigot S, Delobel R, Le Bras M. Eur Polym J 2000;36:273e84. [36] Wu Z, Zhou C, Zhou N. Polym Test 2002;21:479e83. [37] Liu X, Wu Q, Berglund LA, Qi Z. Macromol Mater Eng 2002; 287:515e22. G.M. Russo et al. / Polymer Degradation and Stability 92 (2007) 1925e1933 1933