The Effect of Injection Molding Temperature on the Morphology and Mechanical Properties of PP/PET Blends and Microfibrillar Composites

Within this research the effect of injection molding temperature on polypropylene (PP)/poly(ethylene terephthalate) (PET) blends and microfibrillar composites was investigated. Injection molding blends (IMBs) and microfibrillar composites (MFCs) of PP/PET have been prepared in a weight ratio 70/30. The samples were processed at three different injection molding temperatures (Tim) (210, 230, 280 °C) and subjected to extensive characterization. The observations from the fracture surfaces of MFCs showed that PET fibers can be achieved by three step processing. The results indicated that Tim has a big influence on morphology of IMBs and MFCs. With increasing the Tim, distinctive variations in particle and fiber diameters were noticed. The differences in mechanical performances were obtained by flexural and impact tests. Establishing relationships between the processing parameters, properties, and morphology of composites is of key importance for the valorization of MFC polymers.


Introduction
Polyolefins are a class of polymers which are quite inexpensive and very easy to process. However, these polyolefins, such as polypropylene (PP) and polyethylene (PE), are often blended with engineering polymers, such as poly(ethylene terephthalate) (PET), polyamide (PA), or polycarbonate (PC) [1][2][3][4][5], which have superior mechanical and thermal properties. By combining both the easy processability of polyolefins and the high properties of engineering polymers a product could be obtained, which satisfies all performance demands [5]. Polypropylene (PP) is a semicrystalline polymer which represents one of the most commonly used polymeric materials. As a cheap polymer, its composites are of great interest, i.e., blends with PET which can improve dyeability and mechanical properties [6][7][8]. Poly(ethylene terephthalate) (PET) is a semicrystalline thermoplastic polymer with high toughness, and good physical and chemical properties. This set of properties makes PET of huge importance to the packaging industry. The main applications of this polymer are the manufacturing of films, fibers, and in the fabrication of bottles for beverages [6,7,[9][10][11][12].
Researchers consider the blending of polymers as an economical and practical method to obtain new materials with adjusted properties, but very frequently different polymers are immiscible in the melt phase [3,[13][14][15][16][17][18]. This immiscibility of polymers causes phase separation and low adhesion between two phases [4], which results in low mechanical properties [6,19]. Today, studying the phase separation of polymer blends is of great interest in determining the performance of new polymeric materials [20].
Two different approaches are often employed to avoid the drawbacks of this immiscibility. The first one consists in the addition of a compatibilizer to these blends. Mostly copolymers are used which can interact with both polymeric building components. When it is added to the mixtures, it will result in smaller dispersion of the second phase and better adhesion between the phases. Excellent mechanical properties of the materials could be achieved. In some cases, the compatibilized blends have obtained the tensile fracture behavior from brittle to ductile [21]. A second option is to transform the blends into microfibrillar composites (MFCs). MFC are an interesting class of composites where a high melting fibrillated polymer reinforces a lower melting matrix polymer by applying a specific processing sequence. They could be utilized in a wide range of applications because of improved mechanical properties such as strength and stiffness [12,15,16,22]. The manufacturing of MFCs consists of three basic steps ( Figure 1): Extrusion-melt blending of the two immiscible polymers which have different melting temperatures T m (mixing); 2.
Hot or cold stretching of the extrudate with a molecular orientation of the two polymers (fibrillation); 3.
Fakirov et al. [22,25] and Evstatiev et al. [26] introduced the concept of microfibrillar composites (MFC) from polymer blends many years ago. These researchers stated that the processing of immiscible polymers in which the dispersed phase forms in situ reinforced fibers is a good way to achieve high properties [3]. Two main factors determine the improvement of mechanical properties: (i) the adhesion between matrix and reinforcement; and (ii) the aspect ratios (length/thickness ratio) of the reinforcing elements [19]. separation of polymer blends is of great interest in determining the performance of new polymeric materials [20]. Two different approaches are often employed to avoid the drawbacks of this immiscibility. The first one consists in the addition of a compatibilizer to these blends. Mostly copolymers are used which can interact with both polymeric building components. When it is added to the mixtures, it will result in smaller dispersion of the second phase and better adhesion between the phases. Excellent mechanical properties of the materials could be achieved. In some cases, the compatibilized blends have obtained the tensile fracture behavior from brittle to ductile [21]. A second option is to transform the blends into microfibrillar composites (MFCs). MFC are an interesting class of composites where a high melting fibrillated polymer reinforces a lower melting matrix polymer by applying a specific processing sequence. They could be utilized in a wide range of applications because of improved mechanical properties such as strength and stiffness [12,15,16,22]. The manufacturing of MFCs consists of three basic steps ( Figure 1): 1. Extrusion-melt blending of the two immiscible polymers which have different melting temperatures Tm (mixing); 2. Hot or cold stretching of the extrudate with a molecular orientation of the two polymers (fibrillation); 3. Injection or compression molding-forming treatment at processing temperature of the lower melting component (isotropization) [3,14,23,24].
Fakirov et al. [22,25] and Evstatiev et al. [26] introduced the concept of microfibrillar composites (MFC) from polymer blends many years ago. These researchers stated that the processing of immiscible polymers in which the dispersed phase forms in situ reinforced fibers is a good way to achieve high properties [3]. Two main factors determine the improvement of mechanical properties: (i) the adhesion between matrix and reinforcement; and (ii) the aspect ratios (length/thickness ratio) of the reinforcing elements [19]. An interesting study by Friedrich et al. [27] hypothesized that, within MFCs, a copolymeric interface can be formed, which could play the role of a self-compatibilizer, meaning that by merely introducing this concept, the composites could already have better properties. This is possible in the case of condensation polymers. The improvement of mechanical properties by the MFC concept has already been shown [3,4,14], although not much attention has been dedicated to the specific influence of the processing temperature during the injection molding step (isotropization).
However, similar research for fiber-reinforced polymeric systems has demonstrated the importance of the injection molding temperatures on the properties of these composites. For natural fibers, this temperature strongly influences the fiber length, fiber diameter and, hence, the final mechanical properties [29]. For the MFC technique, most research focuses on differences in composition [14], influence of draw ratio [13,30], and influence of compatibilizers [15,17,21,28]. Several researchers [13,31] found that the influence of screw speed, barrel temperature, cooling time during molding, and mold temperature are very important for the final material properties. An interesting study by Friedrich et al. [27] hypothesized that, within MFCs, a copolymeric interface can be formed, which could play the role of a self-compatibilizer, meaning that by merely introducing this concept, the composites could already have better properties. This is possible in the case of condensation polymers. The improvement of mechanical properties by the MFC concept has already been shown [3,4,14], although not much attention has been dedicated to the specific influence of the processing temperature during the injection molding step (isotropization).
However, similar research for fiber-reinforced polymeric systems has demonstrated the importance of the injection molding temperatures on the properties of these composites. For natural fibers, this temperature strongly influences the fiber length, fiber diameter and, hence, the final mechanical properties [29]. For the MFC technique, most research focuses on differences in composition [14], influence of draw ratio [13,30], and influence of compatibilizers [15,17,21,28]. Several researchers [13,31] found that the influence of screw speed, barrel temperature, cooling time during molding, and mold temperature are very important for the final material properties.
No studies, however, extensively reported and explained as a main topic the effect of processing parameters on morphology and mechanical properties of PP/PET composition with the focus on the injection molding temperature.
The goal of this research is to investigate the influence of the injection molding temperature on the properties of the (non-stretched) injection molding blends (IMBs) and microfibrillar composites (MFCs), which are expected to originate within the morphology of the PET phase in PP/PET blends and composites. The IMBs were prepared by extrusion and injection molding while, for the preparation of MFCs, an additional cold stretching step was introduced between extrusion and injection molding. The relevant mechanical and thermal properties were investigated by flexural and impact testing and differential scanning calorimetry (DSC), respectively. The thermal degradation of the blends and composites was studied by thermogravimetric analysis (TGA). Further, scanning electron microscopy (SEM) was used to investigate the morphology of the samples and study the influence of different injection molding temperatures.
In this study, we present the results of non-compatibilized blends and composites as we would like to explain how, specifically, the injection molding temperature (T im ) affects the main morphology and mechanical properties made from two virgin polymers, i.e., without additives, compatibilizers, or fillers. The main contribution of this work is that it gives a better insight into the temperature-based morphological development of MFCs during manufacturing. This fundamental insight remains relevant when expanding research towards the incorporation of compatibilizing agents.

Materials
The polypropylene (PP) was purchased from Sabic (Sabic 575P, Bergen op Zoom, The Netherlands) with a melt flow rate (MFR) of 11 g/10 min (2.16 kg, 230 • C), and the used polyethylene terephthalate (PET) was LIGHTER C93 (Equipolymers, Schkopau, Germany), which is a bottle-grade material with an intrinsic viscosity of 0.80 ± 0.02 dL/g. PET was dried in a vacuum oven for 24 h at 60 • C before processing, while PP was used as received.

Preparation of PP/PET (Polypropylene/Polyethylene Terephthalate), IMBs (Injection Molding Blends) and MFCs (Microfibrillar Composites)
In this work, PP was used as a matrix, and PET as a reinforcing element. Materials were prepared in a weight ratio of 70/30 PP/PET. The sample preparation was divided into IMB and MFC preparation, respectively. The IMB samples were prepared in two basic steps ( Figure 2, processing route 1): Extrusion and injection molding. The PP and dried PET pellets were properly dry-mixed in a fixed weight ratio of 70/30 before being added to the hopper. The first step, melt blending, was done at the processing temperature of the component with the higher T m using a twin-screw extruder (Coperion ZSK18, Stuttgart, Germany) with a die opening of 19 mm × 2 mm. The screw speed was set at 120 rpm and the barrel temperatures were placed from 205 to 260 • C. The extrudate has been obtained as a sheet with dimensions of 25 mm × 1 mm, by passing through calander rolls, which were cooled down to~15 • C. Further on, the sheets were shredded and injection molded with a BOY 22S injection molding machine (BOY, Neustadt-Fernthal, Germany). In order to study the effect of injection molding temperature, samples were injection molded at three different T im : 210, 230, and 280 • C.
The manufacturing of MFCs consists of three steps ( Figure 2, processing route 2). The blending extrusion temperatures for both IMB and MFC were maintained equal. The first step is the same: melt blending of PP and PET by extrusion. Then the received cooled extrudate was entered directly into a hot oven (200 • C, 55.5 cm × 60 cm) and stretched above the glass transition temperature of PET at a measured surface temperature of 95 • C by a pair of rolls. The speed of the rolls was adjusted to obtain a draw ratio of four. This step is called fibrillation. Before the final isotropization step, the stretched blend was granulated by using a shredder (Piovan RSP15/30, Piovan, Maria di Sala, Italy), and processed by injection molding at the same temperatures as the IMBs. The mold consisted of rectangular bars with dimensions of 100 mm × 10 mm × 4 mm, suited for both impact testing and three-point bending.

Characterization of PP/PET IMBs and MFCs
Thermogravimetric analysis of the virgin polymers and composites was performed on a STA449 Netzsch device (Selb, Germany) in the temperature interval from 30-600 °C with a heating rate of 10 and 20 mL·min −1 flow of nitrogen gas. The onset degradation temperature was calculated at 5% of mass loss for all samples. Differential scanning calorimetry (DSC) was employed to investigate the crystallization and melting behavior. Measurements were performed in two cycles of heating-cooling in the temperature range between 20-300 °C by a Netzsch DSC 204F1 device (Selb, Germany) under nitrogen atmosphere, the heating/cooling rate was 10 °C min −1 , and the flow of nitrogen gas was 20 mL·min −1 . αc was calculated for the PP phase based on the theoretical enthalpy for 100% crystalline polymer and taking the mass percentage into account (Equation (1)): where ΔHo for PP is 207 J/g [32], and Wf is the weight fraction of the relevant polymer in the PP/PET composition.
To study the morphology of IMBs and MFCs, the specimens were prepared for microscopy by immersing in liquid nitrogen and, consequently, fracturing them. Then, the samples were sputtered with gold by a Bal-Tec SCD005 sputter coater (Bal-Tec, Balzers, Liechtenstein). Micrographs were obtained by scanning electron microscopy (SEM), more specifically, a FEG SEM JEOL JSM-7600F 202 instrument (Tokyo, Japan). The accelerating voltage was 15 kV. The average diameter of the particles and fibrils was determined by Image J software (National Institutes of Health, Bethesda, Maryland). For the calculation, at least 30 measurements were used.
The three-point bending flexural test was done with an Instron 3601 testing machine (Norwood, MA, USA) with a load cell of 2 kN according to standard ISO 178. The sample was placed on two supporting spans with a set distance of 64 mm apart and, from above, a third loading pin was lowered at a constant rate (5 mm·min −1 ) until sample failure. Before the final isotropization step, the stretched blend was granulated by using a shredder (Piovan RSP15/30, Piovan, Maria di Sala, Italy), and processed by injection molding at the same temperatures as the IMBs. The mold consisted of rectangular bars with dimensions of 100 mm × 10 mm × 4 mm, suited for both impact testing and three-point bending.

Characterization of PP/PET IMBs and MFCs
Thermogravimetric analysis of the virgin polymers and composites was performed on a STA449 Netzsch device (Selb, Germany) in the temperature interval from 30-600 • C with a heating rate of 10 and 20 mL·min −1 flow of nitrogen gas. The onset degradation temperature was calculated at 5% of mass loss for all samples. Differential scanning calorimetry (DSC) was employed to investigate the crystallization and melting behavior. Measurements were performed in two cycles of heating-cooling in the temperature range between 20-300 • C by a Netzsch DSC 204F1 device (Selb, Germany) under nitrogen atmosphere, the heating/cooling rate was 10 • C·min −1 , and the flow of nitrogen gas was 20 mL·min −1 . α c was calculated for the PP phase based on the theoretical enthalpy for 100% crystalline polymer and taking the mass percentage into account (Equation (1)): where ∆H o for PP is 207 J/g [32], and w f is the weight fraction of the relevant polymer in the PP/PET composition.
To study the morphology of IMBs and MFCs, the specimens were prepared for microscopy by immersing in liquid nitrogen and, consequently, fracturing them. Then, the samples were sputtered with gold by a Bal-Tec SCD005 sputter coater (Bal-Tec, Balzers, Liechtenstein). Micrographs were obtained by scanning electron microscopy (SEM), more specifically, a FEG SEM JEOL JSM-7600F 202 instrument (Tokyo, Japan). The accelerating voltage was 15 kV. The average diameter of the particles and fibrils was determined by Image J software (National Institutes of Health, Bethesda, MD, USA). For the calculation, at least 30 measurements were used.
The three-point bending flexural test was done with an Instron 3601 testing machine (Norwood, MA, USA) with a load cell of 2 kN according to standard ISO 178. The sample was placed on two supporting spans with a set distance of 64 mm apart and, from above, a third loading pin was lowered at a constant rate (5 mm·min −1 ) until sample failure.
Charpy impact strength was measured using a Tinius Olsen IT 503 Pendulum Impact Tester machine (Ulm, Germany) according to standard ISO 179. The specimens were notched at 2 mm depth and broken by a hammer with an energy value of 2 J. At least five specimens were tested for both flexural and impact tests. Statistical analysis of all results was performed by software package SPSS Statistics 22 (Armonk, NY, USA) (t-independent sample tests, p = 0.05).

Thermal Properties of IMBs and MFCs
Thermogravimetric analyses of the samples were performed to investigate the degradation of PP and PET in IMB and MFC compared with virgin PP and PET. The results are shown in Figure 3a and summarized in Table 1. All samples show a single weight loss upon heating to 600 • C. The degradation step of PP was within the range of 342 to 436 • C with a limited char yield of 0.1% and for PET within 396 to 491 • C with a char yield of 14.5%. Those results are very consistent with literature [9,13,33]. The thermal degradation of PP does not imply chain branching or crosslinking [33], so the process started by thermal scission of C-C chain bonds [34]. As can be expected, if we compare the decompositions of PP and PET, it is obvious that PET has a higher thermal stability than PP. PET's decomposition is initiated by scission of the ester group in the chain, yielding to carboxyl and vinyl ester groups, and eventually 14.5% of char residue remains. As the IMB and MFC contain a high amount of PET (30 wt %) the difference between the PP degradation and the other two is obvious. A delay in the onset temperature for IMB and MFC due the presence of the PET phase can be noticed. The higher temperature onset for MFC indicates the delayed degradation due to the formation of PET microfibrils. From the first derivative (dTG) of thermogravimetric (TG) curves represented in Figure 3b, can be seen that the presence of PET severely affects the degradation behavior of PP in both IMB and MFC. The degradation peaks for IMB and MFC were found at 419.1 and 430.8 • C, respectively. Due to the presence of PET in the fibril shape, the degradation of the blend is strongly affected, as the fibrils are prone to degrade slower than particles. Even after melting, orientation and alignment of fibrils can be maintained, since only temperature is applied during the test (no shear or mixing) so the alignment remains in the molten phase, which results in larger cohesion forces and later onset of degradation. Jayanarayanan et al. [13] elaborated that a high aspect ratio of PET microfibrils (five or eight) can shift the temperature onset of decomposition in MFC. Charpy impact strength was measured using a Tinius Olsen IT 503 Pendulum Impact Tester machine (Ulm, Germany) according to standard ISO 179. The specimens were notched at 2 mm depth and broken by a hammer with an energy value of 2 J. At least five specimens were tested for both flexural and impact tests. Statistical analysis of all results was performed by software package SPSS Statistics 22 (Armonk, NY, USA) (t-independent sample tests, p = 0.05).

Thermal Properties of IMBs and MFCs
Thermogravimetric analyses of the samples were performed to investigate the degradation of PP and PET in IMB and MFC compared with virgin PP and PET. The results are shown in Figure 3a and summarized in Table 1. All samples show a single weight loss upon heating to 600 °C. The degradation step of PP was within the range of 342 to 436 °C with a limited char yield of 0.1% and for PET within 396 to 491 °C with a char yield of 14.5%. Those results are very consistent with literature [9,13,33]. The thermal degradation of PP does not imply chain branching or crosslinking [33], so the process started by thermal scission of C-C chain bonds [34]. As can be expected, if we compare the decompositions of PP and PET, it is obvious that PET has a higher thermal stability than PP. PET's decomposition is initiated by scission of the ester group in the chain, yielding to carboxyl and vinyl ester groups, and eventually 14.5% of char residue remains. As the IMB and MFC contain a high amount of PET (30 wt %) the difference between the PP degradation and the other two is obvious. A delay in the onset temperature for IMB and MFC due the presence of the PET phase can be noticed. The higher temperature onset for MFC indicates the delayed degradation due to the formation of PET microfibrils. From the first derivative (dTG) of thermogravimetric (TG) curves represented in Figure 3b, can be seen that the presence of PET severely affects the degradation behavior of PP in both IMB and MFC. The degradation peaks for IMB and MFC were found at 419.1 and 430.8 °C, respectively. Due to the presence of PET in the fibril shape, the degradation of the blend is strongly affected, as the fibrils are prone to degrade slower than particles. Even after melting, orientation and alignment of fibrils can be maintained, since only temperature is applied during the test (no shear or mixing) so the alignment remains in the molten phase, which results in larger cohesion forces and later onset of degradation. Jayanarayanan et al. [13] elaborated that a high aspect ratio of PET microfibrils (five or eight) can shift the temperature onset of decomposition in MFC.   Differential scanning calorimetry (DSC) was used for studying the melting and crystallization behavior of PP and PET in the IMB and MFC samples. Table 2 shows the results of thermal properties during first heating of all samples. The melting temperature (T m ), crystallization temperatures (T c ), T c onset , and T c endset , which indicate the temperature at the beginning and end of crystallization, heat of fusion ∆H m of the PP phase and the PET phase, absolute value of crystallization enthalpy ∆H c of PP, and percentage of crystallinity (α c ) are displayed.  In Table 2 it can be seen that the melting enthalpy for IMB is lower than that of the pure PP, which results in a lower crystallinity degree in IMB. There is a possibility that mobility of PP chains was reduced due to the presence of the PET phase [6]. However, in the MFC, α c is higher in comparison to both pure PP and IMB, due to the long fibrils. These act as nucleation sites for the trans-crystallization of the PP phase [13], thereby increasing the total amount of crystallinity. In IMB, this nucleation effect is much lower and dominated by the constraining effect of the PP phase, slightly reducing α c . Figure 4a represents the DSC thermograms of the samples after injection molding. The temperature of 171.7 • C corresponds to the melting peak of the PP phase, during which time PET is still a crystalline solid, while the PP is in a high melting state. Analyzing the PET curve, a typical cold crystallization at 130.4 • C and the melting point at 256 • C were found. Comparing the T m for IMB and MFC with pure PP, a decrease in temperature can be found, still with a single melting peak of PP, which confirms the presence of α-crystalline modification. The crystallization temperature of neat PP is around 114 • C, but in IMB the PET phase-shifted T c of PP to the higher levels~120 • C (Figure 4b). PP crystals become imperfect due to the presence of the PET phase, which means a lower spherulite size of PP and a higher number of spherulites. Therefore, the result is a decrease of the melting temperature of PP and an enhanced crystallization. In the case of MFC the same effect is noted, which indicates that PET in situ microfibers had a significant nucleating effect on the crystallization of the PP phase due to the higher surface-to-volume ratio. Similar results have been described by other authors [6,8,13,35,36].

Morphology Development
Morphological control of the dispersed phase is extremely important to achieve good properties for polymer blends and microfibrillar composites [2,4,9,25,37,38]. Figure 5 represents SEM micrographs of PP/PET composition with 30 wt % PET after extrusion, stretching, and injection molding at 210 °C. The unstretched blend shows a typical incompatible morphology (Figure 5a), where the spherical PET particles with a diameter ranging between 1-5 μm are found to be distributed in the PP matrix. Additionally, it can be seen that the PET particles exhibit a uniform dispersion in the PP matrix, with no adhesion between the two phases, which indicates a completely immiscible blend [2,4]. This morphology is very convenient for stretching and making fibers, and later, MFCs.
Figures 5b,c show the morphology after fibrillation and before injection molding. It is clear that both phases are oriented and the average diameter of the fibrils is 1.5 μm. Figure 5d presents the IMB sample injection molded at 210 °C with obvious incompatible blend morphology. Discrete domains of the minor phase dispersed within the continuous phase of the major phase without adhesion between the phases can be seen. After injection molding at 210 °C it can be noticed that the MFC structure obtained after stretching was preserved in the samples (Figure 5e,f). Fibers with diameters of 1-2 μm can be found.

Morphology Development
Morphological control of the dispersed phase is extremely important to achieve good properties for polymer blends and microfibrillar composites [2,4,9,25,37,38]. Figure 5 represents SEM micrographs of PP/PET composition with 30 wt % PET after extrusion, stretching, and injection molding at 210 • C. The unstretched blend shows a typical incompatible morphology (Figure 5a), where the spherical PET particles with a diameter ranging between 1-5 µm are found to be distributed in the PP matrix. Additionally, it can be seen that the PET particles exhibit a uniform dispersion in the PP matrix, with no adhesion between the two phases, which indicates a completely immiscible blend [2,4]. This morphology is very convenient for stretching and making fibers, and later, MFCs.

Morphology Development
Morphological control of the dispersed phase is extremely important to achieve good properties for polymer blends and microfibrillar composites [2,4,9,25,37,38]. Figure 5 represents SEM micrographs of PP/PET composition with 30 wt % PET after extrusion, stretching, and injection molding at 210 °C. The unstretched blend shows a typical incompatible morphology (Figure 5a), where the spherical PET particles with a diameter ranging between 1-5 μm are found to be distributed in the PP matrix. Additionally, it can be seen that the PET particles exhibit a uniform dispersion in the PP matrix, with no adhesion between the two phases, which indicates a completely immiscible blend [2,4]. This morphology is very convenient for stretching and making fibers, and later, MFCs.
Figures 5b,c show the morphology after fibrillation and before injection molding. It is clear that both phases are oriented and the average diameter of the fibrils is 1.5 μm. Figure 5d presents the IMB sample injection molded at 210 °C with obvious incompatible blend morphology. Discrete domains of the minor phase dispersed within the continuous phase of the major phase without adhesion between the phases can be seen. After injection molding at 210 °C it can be noticed that the MFC structure obtained after stretching was preserved in the samples (Figure 5e,f). Fibers with diameters of 1-2 μm can be found.    Figure 5d presents the IMB sample injection molded at 210 • C with obvious incompatible blend morphology. Discrete domains of the minor phase dispersed within the continuous phase of the major phase without adhesion between the phases can be seen. After injection molding at 210 • C it can be noticed that the MFC structure obtained after stretching was preserved in the samples (Figure 5e,f). Fibers with diameters of 1-2 µm can be found.

Influence of T im on Morphology
IMBs and MFCs were injection molded at three different injection molding temperatures of 210, 230, and 280 • C. Figure 6 shows the different morphologies of the IMBs at different T im . All prepared IMBs and MFCs, either low or high T im show typical incompatible morphology. At 230 and 280 • C (Figure 6b,c), some larger elliptical particles occur, which could be mistaken for microfibrils, but contrary to microfibrils they have uneven cross-sections. Several PET particles have made new single daughter particles, ellipsoids which, from the obtained observation, could be mistaken for fibrils, but it is known that blend morphology does not display these structures. This phenomenon is known as a coalescence effect. Under the effect of coalescence, the merging of two or more particles into one new larger particle which is often not spherical in shape is implied [25]. During the injection molding process at higher temperature there is a significant possibility for deformation and coalescence of PET particles and fibrils [13,14,24,25,30,[38][39][40][41][42][43][44][45].

Influence of Tim on Morphology
IMBs and MFCs were injection molded at three different injection molding temperatures of 210, 230, and 280 °C. Figure 6 shows the different morphologies of the IMBs at different Tim. All prepared IMBs and MFCs, either low or high Tim show typical incompatible morphology. At 230 and 280 °C (Figure 6b,c), some larger elliptical particles occur, which could be mistaken for microfibrils, but contrary to microfibrils they have uneven cross-sections. Several PET particles have made new single daughter particles, ellipsoids which, from the obtained observation, could be mistaken for fibrils, but it is known that blend morphology does not display these structures. This phenomenon is known as a coalescence effect. Under the effect of coalescence, the merging of two or more particles into one new larger particle which is often not spherical in shape is implied [25]. During the injection molding process at higher temperature there is a significant possibility for deformation and coalescence of PET particles and fibrils [13,14,24,25,30,[38][39][40][41][42][43][44][45].  Table 3 the average, minimum, and maximum diameters of PET particles and fibers in PP/PET compositions are listed. Comparing the size of PET particles at lower and higher Tim, the particles at lower Tim are smaller (1-2.5 μm), while the particle size at 280 °C can reach diameters up to 6 μm due to the aforementioned coalescence. Higgins et al. [19] explained that is difficult to control the mixture of polymers as they are thermodynamically stable only in limited temperature ranges. As a consequence, the uncompatibilized blends at high processing temperature will show larger-sized domains, as was confirmed by Van Bruggen et al. in a recent study [46].
Micrographs presented in the Figures 7-9 show the specimen fracture of MFC in the parallel direction to injection molding flow. It was hard to measure the exact longitude of fibers from these micrographs as the matrix was not removed, so some fibers are hidden, while the diameters were measured ( Table 3). As mentioned earlier above, coalescence has a significant influence for PET fibers due to low PET viscosity. Hence, a significant possibility is that long fibers deform and, afterwards, stick together. This is more likely in the case of higher processing temperatures, but we cannot exclude that it could happen during lower processing temperatures. The obtained microstructure of MFC made at 210 °C (Figure 7a) clearly indicates the uniformity in fibers, as well their size. On the In Table 3 the average, minimum, and maximum diameters of PET particles and fibers in PP/PET compositions are listed. Comparing the size of PET particles at lower and higher T im , the particles at lower T im are smaller (1-2.5 µm), while the particle size at 280 • C can reach diameters up to 6 µm due to the aforementioned coalescence. Higgins et al. [19] explained that is difficult to control the mixture of polymers as they are thermodynamically stable only in limited temperature ranges. As a consequence, the uncompatibilized blends at high processing temperature will show larger-sized domains, as was confirmed by Van Bruggen et al. in a recent study [46].
Micrographs presented in the Figures 7-9 show the specimen fracture of MFC in the parallel direction to injection molding flow. It was hard to measure the exact longitude of fibers from these micrographs as the matrix was not removed, so some fibers are hidden, while the diameters were measured ( Table 3). As mentioned earlier above, coalescence has a significant influence for PET fibers due to low PET viscosity. Hence, a significant possibility is that long fibers deform and, afterwards, stick together. This is more likely in the case of higher processing temperatures, but we cannot exclude   Körmendy et al. [24] explained that the particles and fibers break up and their coalescence is supported by their amorphous state in the PP matrix. The reason for the larger fiber diameter in the case of MFCs made at 230 °C is probably caused by the high shear stress at elevated temperatures during injection molding. Long fibrils cannot withstand this, so they are prone to deform, break up, and coalesce [13,25]. At elevated temperatures, such as 280 °C, PET fibers are effectively re-melted and yield a different morphology altogether, as can be seen in Figure 9. The obtained microstructure once more resembles the IMB. Definitely, the control of Tim is of great importance during this step. We have seen that there is an extreme possibility that PET particles and fibers are broken in IMBs and MFCs. From our point of view, the long PET fibers made during cold stretching at high temperature are lost, as the PET was melted again at its processing temperature, which is also confirmed for the IMB sample. Similar   Körmendy et al. [24] explained that the particles and fibers break up and their coalescence is supported by their amorphous state in the PP matrix. The reason for the larger fiber diameter in the case of MFCs made at 230 °C is probably caused by the high shear stress at elevated temperatures during injection molding. Long fibrils cannot withstand this, so they are prone to deform, break up, and coalesce [13,25]. At elevated temperatures, such as 280 °C, PET fibers are effectively re-melted and yield a different morphology altogether, as can be seen in Figure 9. The obtained microstructure once more resembles the IMB. Definitely, the control of Tim is of great importance during this step. We have seen that there is an extreme possibility that PET particles and fibers are broken in IMBs and MFCs. From our point of view, the long PET fibers made during cold stretching at high temperature are lost, as the PET was melted again at its processing temperature, which is also confirmed for the IMB sample. Similar observed, as well as some necking through the length of the fibers.  Körmendy et al. [24] explained that the particles and fibers break up and their coalescence is supported by their amorphous state in the PP matrix. The reason for the larger fiber diameter in the case of MFCs made at 230 °C is probably caused by the high shear stress at elevated temperatures during injection molding. Long fibrils cannot withstand this, so they are prone to deform, break up, and coalesce [13,25]. At elevated temperatures, such as 280 °C, PET fibers are effectively re-melted and yield a different morphology altogether, as can be seen in Figure 9. The obtained microstructure once more resembles the IMB. Definitely, the control of Tim is of great importance during this step. We have seen that there is an extreme possibility that PET particles and fibers are broken in IMBs and MFCs. From our point of view, the long PET fibers made during cold stretching at high temperature are lost, as the PET was melted again at its processing temperature, which is also confirmed for the IMB sample. Similar explanations were found in several studies [14,38,47]. Körmendy et al. [24] explained that the particles and fibers break up and their coalescence is supported by their amorphous state in the PP matrix. The reason for the larger fiber diameter in the case of MFCs made at 230 • C is probably caused by the high shear stress at elevated temperatures during injection molding. Long fibrils cannot withstand this, so they are prone to deform, break up, and coalesce [13,25]. At elevated temperatures, such as 280 • C, PET fibers are effectively re-melted and yield a different morphology altogether, as can be seen in Figure 9. The obtained microstructure once more resembles the IMB.
Definitely, the control of T im is of great importance during this step. We have seen that there is an extreme possibility that PET particles and fibers are broken in IMBs and MFCs. From our point of view, the long PET fibers made during cold stretching at high temperature are lost, as the PET was melted again at its processing temperature, which is also confirmed for the IMB sample. Similar explanations were found in several studies [14,38,47].

Development of Mechanical Properties
Generally, it is known that uncompatibilized blends have inferior mechanical properties compared with the virgin polymers [15]. In order to achieve superior properties and improve them in IMBs and MFCs, our focus was to investigate the influence of T im on those properties.
In Table 4 the results obtained for samples made at 210 • C are listed. The impact energy for both PP/PET IMB and MFC was lower than for neat PP and PET. This is due to the incompatibility between PP and PET and the occurrence of phase separation [6,38]. MFC showed a slight increase of 8% in comparison with IMB (p = 0.057) for the impact energy. The reason of the enhanced impact properties in MFC, as explained by some authors [17,29,48], probably lies in the presence of PET fibers which are more resistant to crack propagation. Perkins et al. [48] explained that impact resistance can vary depending on the degree of crystallinity. In our case PET fibers reinforce the individual PP spherulites. As it has been mentioned earlier, the PP crystals become imperfect and smaller, which may increase the impact strength. Both the size and amount of PP spherulites play an important role and, together with the interfacial adhesion between the oriented PET fibers and the PP matrix, will determine the impact strength. Comparing the flexural modulus of IMB and MFC with that of pure PP, a significant difference was found: an increase of 37% (p = 0.000) and 42% (p = 0.000) is noted, respectively. IMB (p = 0.00) showed a significant increase in flexural strength relative to PP (p = 0.000) of 22%, while MFC (p = 0.001) increased a smaller amount, up to 12%.
Due to the high flexural properties of PET, the IMB and MFC have shown an increase in flexural modulus and strength in relation to neat PP. Comparing the flexural modulus of IMB to MFC (p = 0.011), a small, yet significant, improvement was observed, while flexural strength was significantly higher for IMB compared with MFC (p = 0.002). Confirmations are found by other researchers, as well [3,26,35] On the other hand, there is a significant difference between 210 and 280 • C. The differences probably originate from different skin-core structures. Several researchers studied the formation of skin-core structures of IM parts. Yi et al. [49], for one, described how, during injection, the skin layer is exposed to a high stress, strain, and large cooling rate, all of which lead to different morphology than in the core. This skin-core structure is, amongst others, influenced by melt temperature and hold pressure elaborated by Zhou et al. [50]. 0.00) (Figure 10b). Furthermore, the MFC samples made at 210 °C showed a significant improvement for the flexural modulus compared to both 230 and 280 °C (p = 0.00). On the other hand, there are no differences found for flexural strength at different Tim. Comparing the flexural strength of IMB and MFC at 210 °C, IMB showed a superior value. It could be explained that the strain at maximum stress, eFSmax, is higher for IMB (Table 4). As straining is allowed to continue further than for MFC, it makes sense that higher stress levels are also reached. The higher eFSmax for IMB is due to the fact that the spherical PET parts are in a previously unstretched state, as opposed to the PET microfibrils in MFC, which have undergone strain hardening during the stretching. This will lead to a reduced ductility of the microfibrils. The PP impact results between different Tim show the slight variations. This is, again, due to the different skin-core structure and PP crystal morphology, strongly affected by the melt temperature [50]. Perkins et al. [48] reported that the combination of high melting temperature, low mold temperature, and fast injection time will give the optimal impact strength.
The impact energy for the IMB samples at the lowest Tim has the highest value. This is probably due to the smaller particle size. According to the measured diameters listed in Table 3, there is no significant difference in the average particle size for IMB samples at 210 and 230 °C but, looking into the morphology, there was found a more uniform dispersion of the small PET particles in the case of 210 °C, while the sample at 230 °C shows a distribution of the higher particle diameter. Moreover, in Figure 6b, some collapsed elliptical particles are noticeable which can reduce the final mechanical properties.
In the case of the MFC sample, we found a significant difference between 210 and 280 °C (p = 0.022) and, as it can be seen from the graph, there are no differences between 230 and 280 °C (p = 0.55).
However, the impact energy for MFC made at 210 °C is higher than that made at 280 °C, as we expected. There is a strong possibility that long PET microfibrils can act as nucleating agents for the PP, which will lower the PP crystal size and improve the adhesion between the two phases [29]. The MFC made at 280 °C has converted into a blend, as was confirmed by its morphological development, and its value corresponds to that of IMB. Comparing all obtained values for impact strength at lower and higher Tim, IMBs and MFCs at high temperature have shown inferior impact properties ( Figure 11). There is an evident increase in the flexural properties for IMB and MFC injection molded at 210 and 230 • C relative to the samples made at 280 • C. The IMB samples processed at lower T im showed an increase in flexural modulus in comparison with 280 • C (Figure 10a). However, for the flexural strength of IMB was found a significant increase in the case of 210 • C relative to 230 and 280 • C (p = 0.00) (Figure 10b).
Furthermore, the MFC samples made at 210 • C showed a significant improvement for the flexural modulus compared to both 230 and 280 • C (p = 0.00). On the other hand, there are no differences found for flexural strength at different T im . Comparing the flexural strength of IMB and MFC at 210 • C, IMB showed a superior value. It could be explained that the strain at maximum stress, e FSmax , is higher for IMB (Table 4). As straining is allowed to continue further than for MFC, it makes sense that higher stress levels are also reached. The higher e FSmax for IMB is due to the fact that the spherical PET parts are in a previously unstretched state, as opposed to the PET microfibrils in MFC, which have undergone strain hardening during the stretching. This will lead to a reduced ductility of the microfibrils.
The PP impact results between different T im show the slight variations. This is, again, due to the different skin-core structure and PP crystal morphology, strongly affected by the melt temperature [50]. Perkins et al. [48] reported that the combination of high melting temperature, low mold temperature, and fast injection time will give the optimal impact strength.
The impact energy for the IMB samples at the lowest T im has the highest value. This is probably due to the smaller particle size. According to the measured diameters listed in Table 3, there is no significant difference in the average particle size for IMB samples at 210 and 230 • C but, looking into the morphology, there was found a more uniform dispersion of the small PET particles in the case of 210 • C, while the sample at 230 • C shows a distribution of the higher particle diameter. Moreover, in Figure 6b, some collapsed elliptical particles are noticeable which can reduce the final mechanical properties.
In the case of the MFC sample, we found a significant difference between 210 and 280 • C (p = 0.022) and, as it can be seen from the graph, there are no differences between 230 and 280 • C (p = 0.55).
However, the impact energy for MFC made at 210 • C is higher than that made at 280 • C, as we expected. There is a strong possibility that long PET microfibrils can act as nucleating agents for the PP, which will lower the PP crystal size and improve the adhesion between the two phases [29]. The MFC made at 280 • C has converted into a blend, as was confirmed by its morphological development, and its value corresponds to that of IMB. Comparing all obtained values for impact strength at lower and higher T im , IMBs and MFCs at high temperature have shown inferior impact properties ( Figure 11). According to Friedrich et al. [3] and Jayanarayanan et al. [29] processing by injection molding has a great influence on PET particles and fibrils. Summarizing all of the results from Figures 10 and  11, the differences between the samples processed at low and high temperatures are obvious. The samples produced at lower temperatures have shown better properties, but also improvements in properties were shown when dealing with the MFC structure over IMB.
At high Tim, the MFCs lose the morphology obtained by twin-screw extrusion and cold stretching, due to the coalescence of PET fibers, which will affect the mechanical properties. During injection molding the PET fibers are exposed to high shear stresses and relaxation behavior at elevated temperatures, which causes the reduction of their aspect ratio and the decrease in properties. As it was shown, for both IMB and MFC samples made at 280 °C, the lowest properties were obtained, which could be explained by a fact that, at the high temperature, PET particles, as well fibers, were re-melted again, losing the original morphology. In addition, the explanation could lie in the effect of coalescence. At the higher temperature, the PET phase is in a molten state and, according to their microstructures, it is evident that PET fibers (particles) were deformed and broken which, afterwards, during mixing, resulted in their interaction and coalescence (see Section 3.2.2). It is known that reinforcing fibers will only act as an effective stress transfer agent up to a certain critical length, which is related to the adhesion between the matrix and the fibers [29], so they do have a tendency to break into shorter ones. This fiber breakup will affect the mechanical strength, as was shown in the results, and confirmed by a morphology study.
We have shown that Tim strongly affects the mechanical properties of all samples, especially MFC, as they lose their fibril morphology at the highest processing temperature. In terms of maximizing the mechanical properties, these could be obtained by mitigating the immiscibility between PP and PET [17,21,26,28,38,47] or increasing the draw ratio [26,29] (which was deliberately kept low within the current work).
It was, however, not the purpose of this manuscript to maximize mechanical properties, but to investigate the separate effect of Tim on morphology and other properties. A combination of these insights with the aforementioned known strategies to increase mechanical properties would potentially create a synergistic optimizing effect.
In this, it is very important to maintain high adhesion between the two phases and a high aspect ratio of reinforcement [25] to achieve superior mechanical properties. This could be achieved beside a good control of injection molding temperature and other settings during processing by, for example, adding a compatibilizer [7,26,28,38,47], using chemical interactions [3,51] and transcrystallinity phenomenon [3,25,27,52], and controlling the settings during cold stretching and injection molding. According to Friedrich et al. [3] and Jayanarayanan et al. [29] processing by injection molding has a great influence on PET particles and fibrils. Summarizing all of the results from Figures 10 and 11, the differences between the samples processed at low and high temperatures are obvious. The samples produced at lower temperatures have shown better properties, but also improvements in properties were shown when dealing with the MFC structure over IMB.
At high T im , the MFCs lose the morphology obtained by twin-screw extrusion and cold stretching, due to the coalescence of PET fibers, which will affect the mechanical properties. During injection molding the PET fibers are exposed to high shear stresses and relaxation behavior at elevated temperatures, which causes the reduction of their aspect ratio and the decrease in properties. As it was shown, for both IMB and MFC samples made at 280 • C, the lowest properties were obtained, which could be explained by a fact that, at the high temperature, PET particles, as well fibers, were re-melted again, losing the original morphology. In addition, the explanation could lie in the effect of coalescence. At the higher temperature, the PET phase is in a molten state and, according to their microstructures, it is evident that PET fibers (particles) were deformed and broken which, afterwards, during mixing, resulted in their interaction and coalescence (see Section 3.2.2). It is known that reinforcing fibers will only act as an effective stress transfer agent up to a certain critical length, which is related to the adhesion between the matrix and the fibers [29], so they do have a tendency to break into shorter ones. This fiber breakup will affect the mechanical strength, as was shown in the results, and confirmed by a morphology study.
We have shown that T im strongly affects the mechanical properties of all samples, especially MFC, as they lose their fibril morphology at the highest processing temperature. In terms of maximizing the mechanical properties, these could be obtained by mitigating the immiscibility between PP and PET [17,21,26,28,38,47] or increasing the draw ratio [26,29] (which was deliberately kept low within the current work).
It was, however, not the purpose of this manuscript to maximize mechanical properties, but to investigate the separate effect of T im on morphology and other properties. A combination of these insights with the aforementioned known strategies to increase mechanical properties would potentially create a synergistic optimizing effect.
In this, it is very important to maintain high adhesion between the two phases and a high aspect ratio of reinforcement [25] to achieve superior mechanical properties. This could be achieved beside a good control of injection molding temperature and other settings during processing by, for example, adding a compatibilizer [7,26,28,38,47], using chemical interactions [3,51] and transcrystallinity phenomenon [3,25,27,52], and controlling the settings during cold stretching and injection molding.

Conclusions
The influence of injection molding temperature on morphology and properties of IMBs and MFCs based on PP and PET was studied. Unstretched blends and microfibrillar composites of PP/PET (70/30, w/w) were successfully achieved by two (extrusion/injection molding) or three processing steps (extrusion/cold stretching/injection molding), respectively. SEM microscopy confirmed the immiscibility of PP and PET in all of the samples, as well as the existence of highly oriented PET microfibrils after cold stretching. Moreover, after injection molding at low T im the MFC structure with oriented fibrils preserved its original morphology. At the highest T im , PET fibers were re-melted and a new morphology was obtained, much like the corresponding IMB. The phenomenon of coalescence was present in all IMBs and MFCs at high T im as the diameter of particles and fibers was significantly higher, which indicates that the PET phase was effectively re-melted.
The crystallization temperature of PP phase in PP/PET IMBs and MFCs was shifted to higher temperatures around 120 • C, as PP crystals became imperfect due to the presence of the PET phase. The crystallinity degree in the MFC was found to be higher than in both neat PP and IMB, as the PET fibrils act as nucleating agents for the transcrystallization of PP and dominates the constraining effect. Additionally, due to the existence of the PET phase in IMBs and MFCs, the poor thermal stability of PP was improved. Analysis of the degradation peaks via dTG curves confirmed the later temperature onset for the MFC sample, due to the formation of PET fibrils, which are harder to decompose then spheres.
The mechanical tests have confirmed that the morphology and properties are significantly affected by processing temperature. The flexural modulus and strength of the IMBs and MFCs were found to be superior to those of neat PP. MFCs have shown a greater impact strength in relation to IMBs, which confirms that PET fibers are a tough reinforcement for the weak PP/PET blend. Both IMB and MFC at elevated T im (280 • C) have shown inferior properties, since the materials have lost their original morphology and the PET phase was re-melted, accompanied by coalescence. The rather limited improvement in mechanical properties of the MFC is attributed to the low draw ratio and low adhesion between the matrix and reinforcement.
Finally, on the basis of these results, it can be concluded that the control of temperature during processing of MFCs is of huge importance, in the same way that the other known factors, such as control of the size of the dispersed phase, draw ratio, and adhesion between the matrix and reinforcement are.