Preliminary dynamic-mechanical analysis of polypropylene/ short carbon fibers composites modified by a succinic anhydride-grafted atactic polypropylene.
In an overall context of sustainability and circular economy [1-3], lower operation costs combined with the high versatility and easy processability of the polyolefins, and more specifically the polypropylene (iPP), become these semicrystalline polymers especially competitive in almost all the organic-based material application markets, nearing nowadays to 200 MT per year. In despite the excellent performance and mechanical properties, better wastes disposal, and recycling opportunities of short carbon fiber iPP-based composites (iPP/SCF), the high prices of SCF made these composites hardly competitive in costs. Nevertheless, the currently available recycled SCF (rSCF), coming from efficient routes of carbon fibers recovering such as the big parts waste disposal sources like aircraft, navigation, road, and railway transport [4-12], should become those iPP/SCF composites competitive in costs always a strong interface between components is achieved.
Indeed, the so-called interface (well defined itself by a finite thickness) refers to the interfacial ply or the dynamic region where the energy, mass, and momentum transfer from the matrix to the fibers take place. This model substitutes the true polymer/fiber interphase, of the very high complexity because of the strong rheology to morphology coupling in the processing of thermoplastics. In fact, in an optimized interface may coexist local chemical bonds, with those others secondary ones, able to yield the required physical interlocking of some polymer chain segments on local areas of the surface reinforcement giving rise to the best composite performance. The true interfacial changes are limited by the finite dimensions of the interface, however, and because of the above-mentioned complex nature, both the effective area and thickness of the interface not only depend on the equilibrium thermodynamics but also are driven by the dynamic thermal and so, flow fields underwent by the heterogeneous material during the processing steps. Thus, the interface is in turn the main responsible of the desired reinforcement effect on the composites and the carbon fiber surfaces exhibit poor interfacial adhesion to polymer molecules because of their intrinsic chemical inertness. Moreover, the recycled carbon fibers (rSCF) are nonuniform in surface because of the different oxidation rates and polar moieties depending on the previous recovering steps (thermal, solution, or pyrolysis treatments) needed to remove the pristine organic matrices [13-20]. Consequently, it means a nonregular surface quality of the rSCF. So, the interfacial modifier must be able to modulate these defects in the interaction of the rSCF and the iPP matrix. Accordingly, previous findings by authors related to the effect of different interfacial modifiers based on grafted atactic polypropylenes (aPP-X, where x is any polar moiety able to modify the nonpolar nature of the matrix), showed that these are able to improve the local secondary interactions with the polar groups on the reinforcement surface, or even by yield primary bonds at the interface [21-26]. Furthermore and because of their amorphous nature, the aPP-based modifiers exhibit lower conformation restrictions, and so, lowers constraints to give rise to strengthen interfaces. On these bases, and for incoming studies on iPP/ rSCF composites , a set of control composites based on both a commercial with a low molecular weight polyamide sizing (SFC) and an injection grade polypropylene were prepared and characterized. Hence, by keeping constant the iPP/SCF ratio on each composite, these were modified by the presence of just a 1.5% of a succinic-grafted atactic polypropylene (aPP-SA/SA). Moreover, it is important mention that the tensile moduli of the pristine and of modified injection molded composites were, respectively, six and ten times above that of the pristine polypropylene, while the break tensile strength values were only over one and a half above that of the iPP matrix. Similarly, the flexural properties also evidenced the interfacial modification efficiency of the above-mentioned interface modifier, which with a high succinic anhydride grafting level (5.5%, w/w), would be strongly competitive in costs because of coming from an industrial residue . Additionally, the aPP-SA/SA may act as a reactive interfacial modifier because of the possibility to form an optimal number of chemical bonds by interaction with the suitable polar moieties bonded at the reinforcement surface. The latter is highly dependent of the optimization of the amount and grafting degree of the interfacial modifier [21-26]. Dynamic Mechanical Analysis (DMA) is one of the most powerful characterization techniques of polymeric materials, in particular for those based on semicrystalline matrices, and mainly because of the high sensitivity of the DMA parameters to changes in the interfacial interaction level between the main components of the composites and specificity of the SCF reinforced composites [28, 29], Therefore, the preliminary study of the efficiency of aPP-SA/SA in iPP/SCF composites in two significant and extreme iPP/SFC ratios (75/15 and 60/40) is the main aim of the present work. Further, the identification of the main response zones in the DMA spectra (and so in the DMA parameters) would be the basis for further optimization studies on iPP/ rSCF composites to be strongly supported by the Box-Wilson's experimental design methodology according the author's experience with other iPP-based heterogeneous materials such as the PP/PA6 system, or the PP/talc and PP/mica composites [21-26],
Materials and Procedures
The starting materials were a polypropylene (iPP), Isplen 050, ([rho] = 0.90 g/[cm.sup.3]; [M.sub.w] = 334,400; [M.sub.n] = 59,500; [T.sub.g] = -13[degrees]C), Repsol-YPF supplied, and short carbon fibers (SCF) Toray 300 PA sizing (4% polyamide sized) from Mitsui (([rho] = 0.78 g/[cm.sup.3]; E = 231 GPa; [sigma] = 3.8 GPa; Diameter = 7 [micro]]).
The interfacial agent used is a succinic anhydride-grafted atactic polypropylene (aPP-SA/SA) with two types of grafts: succinic bridges and side grafted succinic anhydride groups (Fig. 1). This additive (aPP-SA/SA) was obtained by a solution chemical modification of an atactic polypropylene (by-product of industrial polypropylene polymerization processes) supplied by Repsol-YPF. The grafting content is 5.6% w/w (equivalent to 5.6 x [10.sup.-4] mol/g). Additionally, the obtaining and characterization procedures of aPP-SA/SA were fully described elsewhere [21, 26],
This preliminary study has been performed over five different samples with extreme composition. So, a neat iPP, two composites with low SCF content (15%) without and with 1.5% of aPP-SA/SA as interface agent, and the other two with very high SCF content (40%) without and with 1.5% of the interfacial agent. The interfacial agent (aPP-SASA) was incorporated by replacing 1.5% of iPP in the composite in order to maintain constant the polymer/SCF ratio. These materials were first dry blended in the adequate amounts and then compounded by feeding the mixture into a counter-rotating twin-screw extruder Collin ZK-50 working at 20 rpm with a temperature profile from 190[degrees]C up to 220[degrees]C.
Once pelletized the samples, the pellet compounds were compression molded onto 2 mm thick plies in a Dr. Collin press working at 180[degrees]C and 3.3 MPa of pressure for five minutes. Once reached the room temperature after transferring the mold to a water-cooling cartridge under pressure, the press was opened and the ply conditioned at room temperature and 50% RH before being mechanized.
Thermogravimetry analysis (TGA) of the different samples was carried out in TAQ50 thermogravimetric analyzer equipped with an automatic sample feeding. Experimental runs over samples of around 20 mg were undertaken by heating from 30[degrees]C up to 750[degrees]C at a heating rate of 10[degrees]C/min under nitrogen atmosphere (90 mL/min). The later looked for the rightness of the dose of carbon fiber in the composites. In addition, by running the same experimental program, but under oxidizing atmosphere, the absence of SCF agglomerates, by no flashing detection during sample burning, was confirmed (Fig. 2).
Thermal properties of the different materials were dynamically evaluated by Differential Scanning Calorimetry (DSC) by using a Perkin-Elmer DSC 7/Unix calorimeter, indium (Tm= 165.51[degrees]C; [DELTA][H.sub.m] = 28.45 J/g) and zinc ([T.sub.m] = 419.51[degrees]C; [DELTA][H.sub.m] = 108.40 J/g) calibrated. Standardized pans with around 15 mg of sample obtained from the plies were tested under a dynamic melting/crystallization/melting run from 50[degrees]C up to 200[degrees]C, at 10[degrees]C/min, under a nitrogen atmosphere (30 mL/min). After the first heating step, the samples remained 5 min in the molten state beginning then the cooling scan. The test ended after a subsequent second heating scans. After that, the peak temperatures were obtained, and the crystalline content was calculated from the peak areas by considering just the real iPP content in the sample and by using the value of 209 J/g for a hypothetically fully crystalline polypropylene .
DMA tests were carried out over prismatic specimens, 20 x 5 x 2 mm, in the tension mode by using a METTLER DMA861. They were machined from the compression molded plies and tested after a conditioning period of at least 48 h, at room temperature and 50% R. H. DMA parameters: storage (E'). and loss components (E"), of the complex modulus (E*), and the damping or loss factor (tan [delta] = E"/E') were measured within the range of linear viscoelastic behavior of the material under a dynamic force of 12 N at 1 Hz of fixed frequency and 3 pm amplitude. The thermal scan was conducted from -30[degrees]C to 140[degrees]C at a heating rate of 2[degrees]C/min.
RESULTS AND DISCUSSION
The correct dose of the SCF into the iPP/SCF composites was checked by TGA. Figure 2 shows the thermograms for each material under both the two atmospheres evidencing the correct dose of the SFC in the composites, and the complete burning at 750[degrees]C. It is noteworthy to mention that these evolved almost parallel, just 130[degrees]C downward shifted the loss weight start-on under the oxidizing atmosphere. As noted earlier, the absence of SCF agglomerates, by no flashing detection during sample burning was confirmed (Fig. 2).
Table 1 compiles the final loss weight results for the iPP/ SCF composites under inert atmosphere, according the thermograms displayed in Fig. 2, which otherwise let observe the increased thermal stability of the corresponding to the aPP-SA/SA modified composite with respect to the pristine one. The temperature values at the 5% of weight loss ([T.sub.5%]) as compiled in the Table 1 quantify this comment. About the subtle upward shift of each one of the thermograms of the composites incorporating interfacial agent with respect to the pristine ones, it is noteworthy to mention that the thermogram of the 60/40 iPP/ SCF modified composite yields a weight loss at 600[degrees]C, just a 4% above that of the corresponding pristine one. In fact, the latter results coherently with the chemical reaction between the end N[H.sub.2] groups present in the SFC polyamide coating (sizing) and the succinic groups of the aPP-SA/SA. The latter agrees with the previous findings by the authors  and are coherent with the data sheet of the T300-PA sizing SCF, which accounts with a 4 [+ or -] 1% of PA coating at the SCF surface.
Thermal Behavior and Crystalline Content of the iPP/SCF Composites
Figure 3 compiles the DSC dynamic thermograms of the pristine iPP and the iPP/SCF pristine, and aPP-SA/SA modified composites. While, the Table 2 compiles the peak temperatures and the crystalline contents obtained from the corresponding peak areas referred to the iPP matrix amount on each composite. It is interesting to mention the increase of 9[degrees]C and 12[degrees]C in the crystallization temperature for, respectively, the 85/15 and the 60/40 iPP/SCF with respect to the pristine iPP. The latter evidences the well-known nucleation effect of the SCF on the crystallization capability of the iPP matrix, which remains almost constant also for the composites incorporating the interfacial agent.
Indeed, it is noteworthy to mention that the almost symmetrical crystallization peaks from the dynamic cooling scans, and the broadness of almost all the melting peaks for whatever the first or the second heating scans. All these latter show a lower temperature peak arm characterized by a smoothed slope that would be informing about reordering processes to yield more perfect crystals as the temperature approaches to the melting of the samples. Therefore, we find a value of 168.3[degrees]C for the melting temperature (of the first scan) of the pristine iPP (under homogeneous nucleation processes), and 2[degrees]C to 4[degrees]C below that of the iPP/SCF composites (with crystal growing under heterogeneous nucleation), anyhow high enough as to evidence a well developed a crystal morphology. These reordering processes across the iPP amorphous/crystal interface correlate well with the twin-screw extrusion mixing procedure, used to minimize the fiber breakage because of the elongational flow component. Such component favors the extended chain conformation for the polymer matrix on processing, giving rise to a higher amount of tie-molecules or interconnecting segments, than that obtained when mixing takes place on batch mixing chamber (where shear is the main flow component). Furthermore, this extended chain conformation is even favored during the injection molded process. Thus, the tie-molecules fraction results to be a key factor in the mechanical performance of the material once at the solid state, being the responsible for the lost of symmetry in the low temperature arm of the melting peaks .
In addition, it is interesting to observe the very similar peak areas of almost all the crystallization peaks, or even for both the two melting peak series in Fig. 3, with the exception of the corresponding to the unmodified 85/15 iPP/SCF composite. Hence, in this case the first heating scan melting peak is, in fact, the largest. Indeed, and according to the data compiled in Table 2, the crystalline content of this composite (xml) appears as 43.2% above that of the pristine iPP, similarly processed, and 47.3% above that of the same 85/15 iPP/SCF ratio but 1.5% aPP-SA/ SA modified. This high crystalline content of the iPP matrix in the 85/15 PP/SCF pristine composite remains on both its dynamic cooling ([[lambda].sub.c]) and second heating scans ([lambda][m.sub.2]) which show respective crystalline contents 2.5% and 8.8% below that of the as processed composite ([[lambda].sub.ml]). These results agree with the growth of the transcrystalline regions at the SCF surfaces: The latter, because of its low amount, need of very little PP amorphous phase as to be imbibed and then the highest PP supply to the transcrystalline regions growing around each fiber. The effect is two folds, on one hand the maximization of the overall crystalline content of the PP matrix, and on the other, the maximization of the nonfree and highly constrained amorphous phase of polypropylene that is imbibing each SCF at the iPP/SCF interface.
From this view, when the SCF amount is 40% (more than twice and a half higher than of the 85/15 iPP/SCF pristine composite) the inter fiber distance is minimized, and the nonfree amorphous PP phase needed to imbibe the foreign component located at the PP/SCF interface is maximized. It means there are less PP chain segments available to the crystalline and transcrystalline PP domains and, consequently, we found in the 60/40 PP/SCF pristine composite, a crystalline content [[lambda].sub.m1] of the PP matrix of 39.8% lower than that in the 85/15 PP/SCF pristine one, and even 13.8% below that of the pristine iPP as processed.
When one looks the crystalline contents from the first heating scans ([[lambda].sub.ml]) of both the 1.5% aPP-SA/SA modified iPP/SCF composites (in despite their very different SCF content) one finds almost the same crystalline content, but slightly lower (2.8% the 15% SCF, and 3.3% the 40% SCF), than of the pristine PP similarly processed. Thus, the latter lets conclude the interfacial role played by the aPP-SA/SA able to liberate significant amounts of the PP chain segments needed in the pristine composites, to imbibe the SCF at the PP/SC interface. Indeed, and by reducing the transcrystalline growing in the 85/15 composite and by increasing the overall crystalline content of the 60/40 composite, it means the optimization of the materials by approaching each one of them to the crystal/amorphous ratio of the pristine iPP.
Moreover, one may observe the PP crystalline content differences between the first ([[lambda].sub.m1]) and the second one heating scan ([[lambda].sub.m2]) of both the pristine iPP and the composites to check the effect of the processing steps. In fact, it should deal with the significant amount of the interconnecting segments emerging from the elongational flow component during the mixing step, favoring the extended chain conformations. The interconnecting segments fraction is a key control factor in those responses dealing with the dynamic amorphous/crystal interface considered as main responsible of those reordering processes when temperature approaches to the melting temperature. Because of the quantitative response range of the interfacial events cannot overflow the finite dimensions of the interface, on well-processed materials one may expect that the above mentioned differences between both [[lambda].sub.m1] and [[lambda].sub.m2], positive or negatives because of their reversibility , are always below the 20% considered as the limit of these interface restricted events [33, 34]. That is that we find according to the data compiled in Table 2, where the relative differences between [[lambda].sub.m1] and [[lambda].sub.m2], positive or negatives, appear below 13% in absolute value. To confirm these observations just to observe all the values of the crystalline content differences between those melting during the second heating scan ([[lambda].sub.m2]) with respect to those crystallized during the previous cooling scan ([[lambda].sub.c]). Well characterized by broad peaks, Fig. 3, in agreeing with the above mentioned reordering fusion/crystallization processes to give rise to more perfect crystals [31-35], all the materials keep in the usual trend of the iPP to yield lower crystalline amounts from the second heating scan than those obtained from the previous one cooling scan. In such sense, it is to note that both the 1.5% aPP-SA/SA modified composites increase these differences by a 21.5% in the case of the unmodified 85/15 iPP/SCF composite, and by a 42% for the 60/40 one. These findings contrast with previous works by authors working on iPP/talc and PP/mica systems where the particle size always remained constant all along the processing steps, and agree with the role played by the variable size of the reinforcement and then to the interfacial area available during the heterogeneous nucleation processes. Hence, the latter would be a key factor to take in mind in the study and development of the iPP/SCF composites based on recovered SCF.
Dynamic Mechanical Behavior of the PPISCF Composites
Elastic Behavior. Figure 4 shows the evolution of the storage component (E') with the temperature of the different materials. As expected, each one of the E' plots decreases with temperature by a seemingly linearly and successively changing smooth slope while the test runs over the four characteristic relaxation temperature regions of the polypropylene matrix. The first one of these relaxation regions, below -10[degrees]C, with the dissipation modes mainly restricted to atomic vibrations. The second one, from -10[degrees]C up to 40[degrees]C, where the short-range diffusion motions at the chain segment level in the amorphous regions are enabled and the glass-transition takes place. The third and the fourth zones, respectively, appear between 40[degrees]C and 80[degrees]C; and above 80[degrees]C up to 140[degrees]C. The former is where the rubber-elastic transition takes place, sometimes characterized by a plateau in the E" vs. T plots. Thus, the main responsibilities of the dissipation modes at this zone ares the rapid short-range diffusion motions sharply depending on both the molecular weight and the chain entanglement density. Finally, at the fourth zone, the polymer chains may participate in dissipation modes related to the long-range configuration ranges that approach them to flow once over passed the softening state to reach the molten state at higher temperatures. Hence, in both these high temperature zones, the amorphous/crystal interface plays a significant role over the overall relaxation modes being then very sensitive to the material morphology emerging from the processing modes.
So, at a glance, all the E' plots of the iPP/SCF composites appear shifted upward with respect to that of the pristine iPP, being to notice the almost overlapping of the E' plots of the composites once above 40[degrees]C (in the third temperature zone) with the only exception of that 85/15 iPP/SCF, 1.5% aPP-SA/ SA modified one. Thus, the latter means that the differences between the E' values of both the pristine 85/15 and 60/40, iPP/ SCF composites remain observable just up to the glass transition region, that means that are mainly restricted to the "free" amorphous phase on the composites and far away of both the dynamic crystal/amorphous and iPP/SCF interfaces. Accordingly, from such perspective one finds that E' increases more than twice with respect to the pristine iPP for increasing amounts of SCF, and the corresponding downward shifted values when the 1.5% of the aPP-SA/SA is present in the composites. So, the reasons behind the above mentioned overlapping between the E' plots of both the pristine and the modified 60/40 iPP/SCF composites and even the pristine 85/15 one (at the third temperature zone) are related to the PP amorphous/crystal interface. The latter will be better understood over the loss component evolution (E") in Fig. 5. Otherwise, it is coherent with the sharp differences between the crystalline contents of the pristine 85/15 iPP/SCF composite and all the other materials, as previously discussed over the dynamic thermal results.
Viscous Behavior. Figure 5 plots the evolution of the loss modulus (E") with temperature. From this, a very different range between the E" values of the pristine and the 1.5% aPP-pPBMA modified 85/15 iPP/SCF composite with respect to all the other materials, on both the first and the second one temperature zones, are clearly noticeable. Consequently, in order to a better view, we include an insert in the figure. Here, the E" axis has been expanded to display clearly the evolution of E" for the pristine iPP, and both the 60/40 iPP/SCF composites (with and without interfacial agent)
According to the crystalline contents of these materials discussed over these lines, we find an indeed low glass transition peak, clearly defined and almost similar for the three materials (2.6[degrees]C, 2.2[degrees]C, and 2.7[degrees]C) respectively. Otherwise, these values are in full agreement with the so-called amorphous isotactic PP . Hence, the peak areas appear almost similar to informing about a similar fraction of the "free" PP amorphous phase in the three materials able to yield the glass transition process. Conversely, the pristine 85/15 iPP/SCF composite shows an E" plot characterized by a single peak at 27.9[degrees]C and by a continuous decrease with increasing temperature in the third and the fourth temperature zones. Thus, the latter is coherent with the highest crystalline content of this composite, Table 2, and then the lowest amount of overall amorphous phase with its main fraction highly constrained by imbibing the foreign SCF across a dynamic interface. Hence, it should mean, on one hand, a minor fraction of true "free" amorphous phase able to participate in the cooperative short-range motions at the level of the chain segments that the glass transition means. And on the other a significantly larger energy barrier to yield the process. Accordingly, this response region ought to be significantly affected by the presence of an interfacial modifier. In fact, this is just what occurs as it can be deduced from the absence of a similar peak for the 85/15 iPP/SCF, 1.5% aPP-pPBMA modified composite. However, it maintains a very high slope in the first and the second response range temperature zones, that otherwise, recovers a similar E" evolution as all the other materials once in the third and the four temperature intervals.
Furthermore, the insert included in Fig. 5 let appreciate the almost overlapping between the E" plots of both the pristine and the modified 60/40 iPP/SCF composites at the third and the fourth temperature zones. In fact, this should agree with the extreme fiber/matrix packing ratio. In fact, it should maximize the reaction possibilities between the aPP-SA/SA and the PA6 SCF sizing, as deduced from the former, TGA results in Fig. 2. Meanwhile, the almost parallel plots between both the two modified composites on these regions agree with the interfacial action of the aPP-SA/SA modifier, which downward shift its E" values with respect to those of the pristine 85/15 PP/SCF composite giving rise to the relaxation modes of the iPP matrix above the glass transition region. So, at this temperature range, the PP relaxation modes are mainly driven by the amorphous/ crystal interface, single in the pristine PP and two very different in the composites: One of the PP bulk; and two that at the iPP/ SCF interface just placed between the transcrystalline domains and the SCF surface. So, at the third temperature zone, one finds an almost rubbery plateau similarly to the pristine iPP, which otherwise does not evidence any ex transition peak in the E" plot, otherwise well assigned to the amorphous/crystal interface in the PP bulk. Moreover, according to previous findings by the authors on DMA studies of iPP/talc and iPP/mica systems, the plateau region comes from a balanced mini-max evolution between 40[degrees]C and 100[degrees]C corresponding the lowest temperature curve to relaxation modes across the amorphous/ crystal interface far away from the crystalline domains well developed at the PP bulk. The dissipation modes across the amorphous/crystal interface in these latter domains would be the cause of the highest temperature curve, or [alpha] transition peak. The latter agrees with the evolution of the E" plots of iPP and the modified iPP/SCF composites in the third and fourth temperature regions: Besides, it is noteworthy to mention that even the pristine 60/40 iPP/SCF composite (with the largest fiber/matrix packing density) evolves by a similar pattern. Indeed, it shows a small peak beyond 80[degrees]C, once in the fourth temperature region placed at 104[degrees]C, which should correspond to the a transition peak. The aPP-SA/SA modified 60/40 iPP/SCF shows a smoothed and downward shifted [alpha] transition peak placed at 91.5[degrees]C, being both transitions coherent with the above-mentioned packing density for both these two composites.
The above mentioned results are coherent with findings by other authors [36-38] related to the transcrystallinity effects associated to the typical polypropylene morphologies developed around the carbon fiber surfaces by a heterogeneous nucleating process, which in the case of high aspect ratio geometries such as carbon fibers, carbon nanotubes, or mica particles  is characterized by a high nuclei density. So, it means a large amount of highly constrained amorphous regions just placed at the surface carbon fiber/transcrystalline polypropylene interface wrapping each carbon fiber because of the high specific surface that characterizes the short carbon fibers. Then, and by acting as a matrix/reinforcement energy transducer across the iPP/SCF interface, these fractions would be extremely sensitive to the presence of any interfacial modifier: This is just what the E' and E" plots in Figs. 4 and 5 evidence. Meanwhile, the E" plot of the 1.5% aPP-SA/SA modified, 85/15 iPP/SCF composite, shows the effect of the low packing density of the reinforcement fibers in the pristine 85/15 iPP/SCF composite. Indeed and faced with both, the pristine and the modified 60/40 ones, or even the latter once at the fourth temperature zone, does not overlap their E" values with those mentioned, but they evolve close and almost parallel to those of the pristine iPP. Thus, the latter would indicate the remaining of similar dissipation capabilities on this modified composite.
Damping Behavior. The tan [delta] plots displayed in Fig. 6, as the E"/E' ratio, standardize the respective E" and E' variation rates for each material, all along the different relaxation zones. Here, increasing tan [delta] values should indicate increasing variable rates on the viscous responses with respect to those of the elastic ones. The latter explains the pristine iPP tan 5 plot (which faced to that observed in its E' and E" plots) appears above that of the pristine and the modified 60/40 iPP/SCF composites. It agrees with the expected decrease in the viscous character of these composites because of the presence of SCF at the largest packing density as evidenced by almost overlapping between the two tan [delta] plots of the composites at the third temperature region where the PP amorphous/crystal interface domains the material responses. Moreover, and over the insert included at Fig. 6, it is to note the significant increase of the pristine iPP response all along this third region where, furthermore and just around 67[degrees]C, it takes place at the cross point with the decreasing tan [delta] plot of the pristine 85/15 iPP/SCF composite. Coherently, this temperature agrees with the metastable limit of the transition from the conformationally disordered mesophases to the dominant [alpha]-polymorph, which means new dissipation routes available in the mechanical response of the PP matrix , Indeed, all the tan [delta] plots converge above 90[degrees]C and almost overlap up to 100[degrees]C, the [alpha] transition region assigned to dissipation flows across the PP amorphous/crystal interface, showing hereafter increasing values, and almost overlapped those of both the modified composites.
In both the first and the second temperature regions, the tan [delta] plots displayed in Fig. 6, show how the viscous response domains the dynamic mechanical behavior of the composites at the lowest packing density, it means the pristine and the modified one 85/15 iPP/SCF ratio. Both the two composites evidence maximized and seemingly continuous dissipation capabilities restricted to atomic and vibrational motions as they come from the lowest temperature regions , Nevertheless, the pristine composite characterized by a crystalline content of a 30% above any of the other materials, the pristine iPP included, shows a well-defined peak and a high glass transition temperature placed at 29.5[degrees]C evidencing a large energy barrier to yield the process by a small amount of "free" amorphous phase. It is noteworthty to mention that this effect clearly disappears in the modified one composite, showing a linear relationship between tan [delta] values and temperature all along the first and the second temperature zones. A full linear fit gives rise to a r-square value of 0.999 with a negative slope of 4.7 x [10.sup.-3] and an intercept value of 0.14. Nevertheless, between these eight data points, it is not so hard to appreciate two straight lines with subtle different negative slopes, one by fitting the four upper temperature values and the other by fitting the four lowest. Both linear fits remain on r-square values of 0.999, while show respective slopes of 4.5 x [10.sup.-3] the latter and 4.8 x [10.sup.-3] the former with a cross point between both the two straight lines placed at -0.1[degrees]C, otherwise close to the glass transition values of both the 60/40 iPP/SCF pristine and modified composites. Accordingly, it should inform about an almost nondetectable population of "free" amorphous phase able to yield the glass transition peak in the 85/15 PP/SCF 1.5% aPP-SA/SA modified composite. These results, further to evidence the similar interfacial activity of the aPP-SA/SA whatever the SCF amount, should confirm the 1.5% of the interfacial agent as the optimal amount of the interfacial modifier according the interfacial volume available, mainly imposed by the iPP matrix. So, the maximum packing density composite (the modified 60/40 iPP/SCF composite) shows a glass transition temperature close to 3.5[degrees]C, near 3[degrees]C less than that of the pristine composite, otherwise similar to that of the pristine iPP matrix placed at 6[degrees]C but with lower peak areas. The latter implies lower "free" amorphous phase available to participate in the glass transition process because of the need to imbibe the foreign fibers by the amorphous PP placed between the SCF and the transcrystalline PP domains. Moreover, the effect intensifies as the iPP/SCF interfacial interaction level increases on both the modified composites, as it is evidenced by their tan [delta] plots shifted downward with respect to their respective pristine ones, and so, to lower energy barriers to yield the glass transition. Even so, the sharp reduction in the peak area of the 60/40 iPP/SCF composite, and even the disappeared glass transition peak in the modified 85/15 iPP/SCF composite should agree with the existence of primary bonds across the iPP/SCF interface formed between the aPP-SA/SA and the polyamide of the SCF sizing.
The DMA and dynamic thermal results discussed at present work, let conclude the efficiency of an atactic polypropylene with a high succinic anhydride grafting level as an effective interfacial modifier in iPP/SCF composites.
The results obtained appear as very promising and informs about the great complexity of the phenomena involved.
Moreover, these results let consider this interfacial agent (aPP-SA/SA) as an excellent candidate for pre-preg and/or conditioning treatments of both virgin and/or recycled carbon fibers.
This study affords the basis to optimize both the performance and the better composition ratio in rSCF based PP composites by the use of Box-Wilson experimental methodology.
The results discussed at present work were partially supported by both the MAT 2000-1499 and the PIROFIBER (CTM2013-48887-C2-2-R) Research Projects.
[1.] P.M. Subramanian, Res. Cons. Recycl., 28, 253 (2000).
[2.] K. Hamad, M. Kasee, and F. Deri, Polym. Deg. Stab., 98, 2801 (2013).
[3.] A.A. Zorpas and V.J. Inglezakis, Technol. Soc., 34, 55 (2012).
[4.] G. Oliveux, L.O. Dandy, and G.A. Leeke, Prog. Mater. Sci., 72, 61 (2015).
[5.] SJ. Pickering, Compos. A Appl. Sci. Manuf., 37, 1206 (2006).
[6.] Y. Liu and S. Kumar, Polym. Rev., 52, 234 (2012).
[7.] N. Perry, A. Bernard, F. Laroche, and S. Pompidou, CIRP Ann. Manuf. Technol., 61, 151 (2012).
[8.] Y. Yang, R. Boom, B. Irion, D. J. Van Heerden, P. Kuiper, and H. De Wit, 51, Chem. Eng. Process. 53 (2012)
[9.] S. Pimenta, and S.T. Pinho, Waste Manag., 31, 378 (2011).
[10.] M.A. Nahil, and P.T. Williams, J. Anal. Appl. Pyrolysis, 91, 67
[11.] F.A. Lopez, O. Rodriguez, F.J. Alguacil, I. Garcia-Diaz, T.A. Centeno, J.L. Garcia-Fierro, and C. Gonzalez, J. Anal. Appl. Pyrolysis, 104, 675 (2013).
[12.] C. Jiang, S.J. Pickering, E.H. Lester, T.A. Turner, K.H. Wong, and N.A. Warrior, Compos. Sci. Technol, 69, 192 (2009).
[13.] M. Sharma, S. Gao, E. Mader, H. Sharma, L.Y. Wei, and J. Bijwe, Compos. Sci. Technol., 102, 35 (2014).
[14.] S. Pimenta and S.T. Pinho, Compos. Struct., 94, 3669 (2012).
[15.] T.G. Gopakumar and D.J.Y.S. Page, Polym. Eng. Sci., 44, 1162 (2004).
[16.] S.L. Gao and E. Mader, Compos. Part A, 33, 559 (2002).
[17.] G. Jiang, S.J. Pickering, G.S. Walker, K.H. Wong, and C.D. Rudd, Appl. Surf. Sci., 254, 2588 (2008).
[18.] D.H. Kim, B.H. Kim, K.S. Yang, Y.H. Bang, S.R. Kim, and H.K. Im, J. Kor. Chem. Soc., 55, 819 (2011).
[19.] M.A. Montes-Moran, A. Martinez-Alonso, J.M.D. Tascon, M.C. Paiva, and C.A. Bernardo, Carbon, 39, 1057 (2001).
[20.] F. Severini, L. Formaro, M. Pegoraro, and L. Posca, Carbon, 40, 735 (2002).
[21.] J.M. Garcla-Martinez, S. Areso, and E.P. Collar, J. Appl. Polym. Sci., 102, 1182 (2006).
[22.] E.P. Collar, J. Taranco, S. Areso, and J.M. Garcia-Martinez, J. Polym., 620362, (2015).
[23.] J.M. Garcia-Martinez, J. Taranco, S. Areso, and E.P. Collar, J. Appl. Polym. Sci., 132, 42678 (2015).
[24.] J.M. Garcia-Martinez, S. Areso, and E.P. Collar, J. Appl.
Polym. Sci., 114, 551 (2009).
[25.] J.M. Garcia-Martinez, S. Areso, and E.P. Collar, J. Appl. Polym. Sci., 81, 626 (2001).
[26.] J.M. Garcia-Martinez, S. Areso, and E.P. Collar, J. Appl. Polym. Sci., 104, 345 (2007).
[27.] PIROFIBER, CTM2013-48887-C2-2-R, Spanish Research Project 2013-2017, "Carbon Fiber Recycling by pyrolysis to obtain new thermoplastic composites", Unpublished results.
[28.] S. Dong and R. Gauvin, Polym. Compos., 14, 414 (1993).
[29.] B. Wunderlich, Thermal Analysis, Academic Press, San Diego (1990).
[30.] E.J. Clark and J.D. Hoffman, Macromolecules, 17, 878 (1984).
[31.] A. Mehta and B. Wunderlich, Die Makromol. Chem., 175, 977 (1974).
[32.] B. Wunderlich, Prog. Polym. Sci., 28, 383 (2003).
[33.] P.J. Flory and D.Y. Yoon, Macromolecules, 17, 862 (1984).
[34.] P.J. Flory and D.Y. Yoon, Macromolecules, 17, 871 (1984).
[35.] R. Androsch and B. Wunderlich, Macromolecules, 34, 5950 (2001).
[36.] D. Biakiris, Materials, 3, 2884 (2010).
[37.] S. Zhang, M.L. Minus, L. Zhu, C.P. Wong, and S. Kumar, Polymer, 49, 1356 (2008).
[38.] M.V. Jose, D. Dean, J. Tyner, G. Price, and E. Nayro, J. Appl. Polym. Sci., 103, 3844 (2007).
Jesus Maria Garcia-Martinez (iD), Susana Areso, Emilia P. Collar
Grupo de Ingenieria de Polimeros. GIP, Instituto de Ciencia y Tecnologia de Polimeros, ICTP/CSIC, Calle Juan de la Cierva 3, 28006 Madrid, Spain
Correspondence to: J.M. Garcia-Martinez; e-mail: firstname.lastname@example.org
Contract grant sponsor: PIROFIBER; contract grant number: CTM201348887-C2-2-R.
Published online in Wiley Online Library (wileyonlinelibrary.com).
Caption: FIG. 1. Interfacia] agent used in the present study.
Caption: FIG. 2. TGA curves obtained under air and nitrogen atmosphere for each one of the samples studied.
Caption: FIG. 3. DSC curves obtained under nitrogen atmosphere for each one of the samples studied.
Caption: FIG. 4. Evolution of the storage modulus (E') with temperature for the indicated samples.
Caption: FIG. 5. Evolution of the loss modulus (E") with temperature for the indicated samples.
Caption: FIG. 6. Evolution of the loss or damp factor (tan [delta]) with temperature for the indicated samples.
TABLE 1. Temperature at 5% weight loss and loss weight at 600[degrees]C determined from plots in Fig. 2. [T.sub.5%] Loss weight at Sample ([degrees]C) 600[degrees]C (%) iPP 364.72 0.07 iPP/15% SFC 348.20 16.50 iPP/15% SFC/1.5% aPP-SA/SA 359.14 16.02 iPP/40% SFC 385.29 35.20 iPP/40% SFC/1.5% aPP-SA/SA 400.48 39.92 TABLE 2. Thermal parameters determined by DSC for the indicated samples. T[m.sub.1] [DELTA]h[m.sub.1] Sample ([degrees]C) (J/g) iPP 168.3 89.13 iPP/15% SFC 165.9 127.43 iPP/15% SFC/1.5% aPP-SA/SA 164.0 86.55 iPP/40% SFC 164.2 76.66 iPP/40% SFC/1.5% aPP-SA/SA 166.4 86.21 [lambda][m.sub.1] T[c.sub.1] Sample (%) ([degrees]C) iPP 42.6 110.8 iPP/15% SFC 61.0 119.8 iPP/15% SFC/1.5% aPP-SA/SA 41.4 119.6 iPP/40% SFC 36.7 122.1 iPP/40% SFC/1.5% aPP-SA/SA 41.2 122.6 [DELTA]H[m.sub.1] [lambda][m.sub.2] Sample (J/g) (%) iPP 97.06 46.4 iPP/15% SFC 124.47 59.5 iPP/15% SFC/1.5% aPP-SA/SA 95.60 45.7 iPP/40% SFC 93.81 44.9 iPP/40% SFC/1.5% aPP-SA/SA 93.45 44.7 [Tc.sub.2] [DELTA]H[m.sub.2] Sample ([degrees]C (J/g) iPP 162.9 87.52 iPP/15% SFC 164.2 116.31 iPP/15% SFC/1.5% aPP-SA/SA 161.0 88.03 iPP/40% SFC 164.5 80.66 iPP/40% SFC/1.5% aPP-SA/SA 164.7 74.89 [lambda][m.sub.2] Sample (%) iPP 41.8 iPP/15% SFC 55.6 iPP/15% SFC/1.5% aPP-SA/SA 42.1 iPP/40% SFC 38.6 iPP/40% SFC/1.5% aPP-SA/SA 35.8 * [lambda][m.sub.1], [lambda][c.sub.1], and [lambda][m.sub.a] are the crystalline content for the first heating, the cooling and the second heating curves calculated as the ratio between the enthalpy for each peak and the corresponding to a hypothetically fully crystalline polypropylene [[lambda] = ([DELTA]H/209) x 100] (30).
|Printer friendly Cite/link Email Feedback|
|Author:||Garcia-Martinez, Jesus Maria; Areso, Susana; Collar, Emilia P.|
|Publication:||Polymer Engineering and Science|
|Date:||Jul 1, 2017|
|Previous Article:||Probabilistic analysis and experimental results of sound transmission loss of composite panels.|
|Next Article:||Effect of backbone binders on rheological performance of ceramic injection molding feedstocks.|