Morphology and Mechanical Properties of Polyamide 6/Polystyrene Blends Prepared by Diffusion and Subsequent Polymerization of Styrene in Polyamide 6 Pellets

Polyamide 6 (PA6)/polystyrene (PS) blend pellets were prepared by diffusion and subsequent polymerization of styrene in commercial PA6 pellets and processed into cuboid-shaped bars by injection molding. The average PS particle size in the bars was 240 nm, slightly higher than that in the blend pellets (120 nm), showing that only limited coalescence occurs during injection molding. The mechanical properties of PA6/PS bars were investigated by tensile, flexural, and notched impact tests. A 41% increase in notched impact strength was found without decreasing the modulus, tensile, and flexural strengths and elongation to break, when compared with those of neat PA6 bars. These good mechanical properties were attributed to the small PS particle sizes, and the good interfacial adhesion between PS particles and the PA6 matrix resulted from the occurrence of PS grafting onto PA6 during the preparation of the blend pellets and injection molding. The water sorption and water-induced dimensional changes in PA6/PS bars were significantly less than those of neat PA6 bars because of the presence of the hydrophobic PS phase. This work reveals that the PA6/PS quasi-nanoblend pellets are useful stock for plastic production.


Introduction
Polymer nanoblends are an important type of nanomaterial. They show superior properties compared to polymer microblends prepared by conventional melt blending [1][2][3][4][5][6][7][8][9][10][11]. Therefore, reducing the size of the dispersed phase of polymer blends is a very interesting topic in polymer science. Diffusion and subsequent polymerization of a monomer (or monomer mixture) in commercial polymer pellets is a method that has been explored in recent years by our research group for preparing polymer blend pellets with fine particle sizes down to the quasi-nanoscale [12][13][14][15][16][17][18][19]. It involves diffusion of a monomer (typically styrene) into the amorphous regions of semicrystalline polymer pellets and subsequent diffusion of an initiator (typically benzoyl peroxide) along the diffusion paths of the monomer. The particle size of the blend pellets is limited because of the confined polymerization. The blend pellets can be used directly in plastic production without the need for pre-compounding, and only a limited degree of particle coalescence occurs during processing, consequently improving the mechanical properties. So far, polypropylene (PP)/polystyrene (PS) [17], polyethylene (PE)/PS [14] and ethylene-α-octene copolymer (POE)/PS [13] blend pellets have been prepared with this method and processed into dumbbell-shaped bars or rods by melt processing. Although PS particles are rigid and are generally considered to be non-compatible with PP, PE, and POE, all the blends have a good overall performance in regard to stiffness, strength, ductility, and toughness because of their small (PA6), A 1 /A 2 , was determined and the PS content was obtained (7.3%) using a previously-reported calibration curve [27].
The percentage of PS grafting is defined as the weight ratio of grafted PS to PA6 and was measured as follows: PA6/PS pellets were hot-pressed into thin slices and thoroughly extracted with chloroform to remove the ungrafted PS and then analyzed by ATR-FTIR. The ratio of A 1 /A 2 was determined, and the percentage of PS grafting was obtained (2.0%) using the calibration curve.
The diametrical distribution of PS in the blend pellets and the PS distribution along the width direction of the blend bars were investigated using Microscopic Fourier Transform Infrared Spectroscopy (Micro-FTIR) (model Nicolet 6700, Thermo Electron Corporation, Madison, WI, USA), with a microsampling area of 100 × 100 µm 2 and a step length of 100 µm. For sample preparation, thin slices were cut through the center of a blend pellet and the cross section of a bar. A 1 /A 3 , defined as the ratio of the peak areas of absorption at 758 cm −1 and 1122 cm −1 , which are the characteristic peaks of PS and PA6, respectively, was taken as a measure for the PS/PA6 ratio at the measured position.
A JSM-7401 field emission scanning electron microscope (FESEM) (JEOL, Tokyo, Japan) at an accelerating voltage of 3 kV was used to investigate the morphology of the PS/PA6 blend pellets and bars. The pellet was cut into two symmetrical parts and stirred in aqueous trichloroacetic acid (15%) for 30 min to remove the amorphous phase of PA6 of the cross section, washed ultrasonically in deionized water for 5 min, and dried in an oven. The PA6/PS bars were cryofractured in liquid nitrogen. All surfaces were gold sputtered before FESEM observation.
Differential scanning calorimetry (DSC) experiments were conducted with a TA Instrument (model Q100, TA Instrument, Newcastle, DE, USA). Samples of about 6 mg were placed in a standard sealed aluminum pan and then heated from 40 to 240 • C at a constant rate of 10 • C/min under a constant flow of nitrogen (50 mL/min). The melting temperature (T m ) was defined as the maximum of the melting peak, and the enthalpy of melting (∆H m ) was calculated as the integral of the melting peak, corrected by the weight percent of PA6 in the blend.
Water sorption and dimensional changes of the bars were carried out by comparing wet and dry samples. The wet and dry samples refer to those stored in a container conditioned at 70% relative humidity (RH) for 5 days and those dried in a vacuum oven at 110 • C for 24 h, respectively. The weights of the bars before and after sorption of water were recorded. Water sorption was defined as the weight ratio of absorbed water/dry sample. The lengths, widths and thicknesses of the bars were measured by Vernier calipers.
Tensile property testing was performed on a GT-TCS 2000 tensile tester (Gotech, Taichung, Taiwan) with a crosshead speed of 10 mm/min, with a distance of 3 cm between fixtures. Both dry and wet samples were tested. Data from at least three parallel measurements were averaged. The values of the Young's modulus were used for the slopes of the stress-strain curves at an elongation range of 3-7%.

Distribution of PS and Morphology of PA6/PS Pellets and Molded Bars
PA6 and PS are non-compatible. Further, the maximum PS content in PA6/PS blend pellets prepared by the method of diffusion and subsequent polymerization of styrene in commercial PA6 pellets in the previously reported small-scale procedure was only 6.7% [26]. Thus, in this work, the conditions for obtaining the maximum PS content were used to carry out large-scale preparation (40 times) in order to obtain sufficient PA6/PS blend pellets for injection molding and evaluating the mechanical properties of this unique blend. However, the apparent concentrations of PA6 pellets in each step, i.e., the ratios of pellet/medium, were 2.7 times those used in the previously reported small-scale procedure. Although the medium water in the water-sorption step and the medium styrene in the diffusion step were also the diffusants, the variation in the pellet/medium ratio should not have any effect on the PS content, PS distribution, or blend morphology because only a tiny amount of the medium was able to diffuse into the pellets in each step.
The distribution of PS in both blend pellets and bars was investigated by Micro-FTIR for A 1 /A 3 (i.e., the ratio of the peak areas of absorption at 758 cm −1 , a characteristic peak of PS, and 1122 cm −1 , a characteristic peak of PA6), as a measure of the PS/PA6 ratio at each measured position. As shown in Figure 1a, the PS distribution along the diameter direction of a pellet was almost homogeneous and the average value of A 1 /A 3 was slightly higher than that in the small-scale preparation [26], indicating a slightly higher PS content. ATR-FTIR was used to measure the PS content using a previously-reported calibration curve [27] and the value was 7.3%, slightly higher than that of the small-scale procedure, and basically in agreement with the Micro-FTIR results. As shown in Figure 1b, the distribution of PS along the width direction of a molded bar was basically homogeneous with the A 1 /A 3 values of around 1.5, which is practically the same as that of the pellets.
Materials 2018, 11, x; doi: FOR PEER REVIEW www.mdpi.com/journal/materials characteristic peak of PA6), as a measure of the PS/PA6 ratio at each measured position. As shown in Figure 1a, the PS distribution along the diameter direction of a pellet was almost homogeneous and the average value of A1/A3 was slightly higher than that in the small-scale preparation [26], indicating a slightly higher PS content. ATR-FTIR was used to measure the PS content using a previously-reported calibration curve [27] and the value was 7.3%, slightly higher than that of the small-scale procedure, and basically in agreement with the Micro-FTIR results. As shown in Figure  1b, the distribution of PS along the width direction of a molded bar was basically homogeneous with the A1/A3 values of around 1.5, which is practically the same as that of the pellets. FESEM was used to investigate the phase morphology of both the blend pellets and the bars and the resulting micrographs are shown in Figure 2. The cross section of a pellet was observed after etching the amorphous PA6 portions with aqueous trichloracetic acid for better observation, since PS particles could not be observed clearly without removing the amorphous PA6. As shown in Figure  2a,b, a typical sea-island structure was observed and spherical PS particles were evenly distributed in the blend pellet. The average particle size was 130 nm, slightly higher than that in the small-scale procedure (103 nm). As shown in Figure 2c,d, the sizes of the PS particles were still small in the molded bar, but the particle size distribution was wider than that in the blend pellet. Almost all the PS particles were of nano or submicron sizes. The biggest particle size was around 700 nm and the average particle size was 240 nm, indicating that only slight coalescence of PS particles takes place during injection molding. The small PS particle size should be beneficial for the improvement of mechanical properties.  FESEM was used to investigate the phase morphology of both the blend pellets and the bars and the resulting micrographs are shown in Figure 2. The cross section of a pellet was observed after etching the amorphous PA6 portions with aqueous trichloracetic acid for better observation, since PS particles could not be observed clearly without removing the amorphous PA6. As shown in Figure 2a,b, a typical sea-island structure was observed and spherical PS particles were evenly distributed in the blend pellet. The average particle size was 130 nm, slightly higher than that in the small-scale procedure (103 nm). As shown in Figure 2c,d, the sizes of the PS particles were still small in the molded bar, but the particle size distribution was wider than that in the blend pellet. Almost all the PS particles were of nano or submicron sizes. The biggest particle size was around 700 nm and the average particle size was 240 nm, indicating that only slight coalescence of PS particles takes place during injection molding. The small PS particle size should be beneficial for the improvement of mechanical properties.

Crystallization Behavior of PA6 in PA6/PS Pellets and Molded Bars
DSC was used to investigate the crystallization behavior of PA6 in PA6/PS pellets and molded bars. Typical DSC heating curves are shown in Figure 3, and the melting temperature ( ) and enthalpy of melting (∆ ) data are listed in Table 1. The and ∆ of PA6 in the blend pellets were almost unchanged compared to those of neat PA6 pellets, confirming that the diffusion and subsequent polymerization of styrene only occurs in the amorphous regions of PA6 pellets, which is the basic principle of the method of diffusion and subsequent polymerization of a monomer (or monomer mixture) in polymer pellets.
The and ∆ of PA6 in the injection-molded PA6/PS bars were also similar to those of neat PA6 bars, indicating that the presence of PS in the blend has almost no effect on the crystallization behavior of PA6, and the degree of crystallinity is not the reason for the good overall mechanical properties of PA6/PS blend bars.

Crystallization Behavior of PA6 in PA6/PS Pellets and Molded Bars
DSC was used to investigate the crystallization behavior of PA6 in PA6/PS pellets and molded bars. Typical DSC heating curves are shown in Figure 3, and the melting temperature (T m ) and enthalpy of melting (∆H m ) data are listed in Table 1. The T m and ∆H m of PA6 in the blend pellets were almost unchanged compared to those of neat PA6 pellets, confirming that the diffusion and subsequent polymerization of styrene only occurs in the amorphous regions of PA6 pellets, which is the basic principle of the method of diffusion and subsequent polymerization of a monomer (or monomer mixture) in polymer pellets.
The T m and ∆H m of PA6 in the injection-molded PA6/PS bars were also similar to those of neat PA6 bars, indicating that the presence of PS in the blend has almost no effect on the crystallization behavior of PA6, and the degree of crystallinity is not the reason for the good overall mechanical properties of PA6/PS blend bars.

Water Sorption and Dimensional Changes of Injection Molded PA6/PS Blend Bars
The resistance of injection-molded PA6/PS bars to water was assessed and compared with that of neat PA6 bars by placing the bars in 70% RH for 5 days. The water sorption of PA6/PS bars decreased to 2.23%, equivalent to 2.41% when referred to the PA6 fraction. Considering possible experimental error, the latter value is closed to water sorption of neat PA6 bars (2.54%), indicating that the decrease in water sorption of PA6/PS bars is solely due to incorporation of the hydrophobic PS fraction. This reveals that the homogeneous incorporation of PS is less efficient in preventing the sorption of water than the PS-rich shell structure reported earlier, in which a PS content of 1.2% can significantly lower water sorption [27].
The dimensional expansion in the three directions of the bars upon sorption of water was measured, and the results are shown in Figure 4. Clearly, PA6/PS bars show better dimensional stability than neat PA6 bars in all three directions because the decreased water sorption resulted from the presence of hydrophobic PS. The volume expansion of PA6/PS bars caused by water sorption was 2.41%, significantly lower than that of neat PA6 bars (3.16%).

Mechanical Properties of Injection Molded PA6/PS Blend Bars
Tensile, flexural, and notched impact tests were used to evaluate the mechanical properties of PA6/PS bars injection-molded from PA6/PS blend pellets, and the results are summarized in Table 2.

Water Sorption and Dimensional Changes of Injection Molded PA6/PS Blend Bars
The resistance of injection-molded PA6/PS bars to water was assessed and compared with that of neat PA6 bars by placing the bars in 70% RH for 5 days. The water sorption of PA6/PS bars decreased to 2.23%, equivalent to 2.41% when referred to the PA6 fraction. Considering possible experimental error, the latter value is closed to water sorption of neat PA6 bars (2.54%), indicating that the decrease in water sorption of PA6/PS bars is solely due to incorporation of the hydrophobic PS fraction. This reveals that the homogeneous incorporation of PS is less efficient in preventing the sorption of water than the PS-rich shell structure reported earlier, in which a PS content of 1.2% can significantly lower water sorption [27].
The dimensional expansion in the three directions of the bars upon sorption of water was measured, and the results are shown in Figure 4. Clearly, PA6/PS bars show better dimensional stability than neat PA6 bars in all three directions because the decreased water sorption resulted from the presence of hydrophobic PS. The volume expansion of PA6/PS bars caused by water sorption was 2.41%, significantly lower than that of neat PA6 bars (3.16%).

Water Sorption and Dimensional Changes of Injection Molded PA6/PS Blend Bars
The resistance of injection-molded PA6/PS bars to water was assessed and compared with that of neat PA6 bars by placing the bars in 70% RH for 5 days. The water sorption of PA6/PS bars decreased to 2.23%, equivalent to 2.41% when referred to the PA6 fraction. Considering possible experimental error, the latter value is closed to water sorption of neat PA6 bars (2.54%), indicating that the decrease in water sorption of PA6/PS bars is solely due to incorporation of the hydrophobic PS fraction. This reveals that the homogeneous incorporation of PS is less efficient in preventing the sorption of water than the PS-rich shell structure reported earlier, in which a PS content of 1.2% can significantly lower water sorption [27].
The dimensional expansion in the three directions of the bars upon sorption of water was measured, and the results are shown in Figure 4. Clearly, PA6/PS bars show better dimensional stability than neat PA6 bars in all three directions because the decreased water sorption resulted from the presence of hydrophobic PS. The volume expansion of PA6/PS bars caused by water sorption was 2.41%, significantly lower than that of neat PA6 bars (3.16%).

Mechanical Properties of Injection Molded PA6/PS Blend Bars
Tensile, flexural, and notched impact tests were used to evaluate the mechanical properties of PA6/PS bars injection-molded from PA6/PS blend pellets, and the results are summarized in Table 2.

Mechanical Properties of Injection Molded PA6/PS Blend Bars
Tensile, flexural, and notched impact tests were used to evaluate the mechanical properties of PA6/PS bars injection-molded from PA6/PS blend pellets, and the results are summarized in Table 2. The tensile tests were carried out on both dry and wet samples, while the flexural and notch impact tests were only carried out on dry samples.
The typical stress-strain curves of PA6/PS bars along with those of neat PA6 bars are shown in Figure 5a. The tensile behaviors of PA6/PS bars were very similar to those of neat PA6 bars in both dry and wet states, indicating that the presence of immiscible PS particles does not cause any deterioration of the tensile properties of PA6 even in the wet state, although no compatibilizer was used. The good ductility of wet PA6/PS bars suggests that the PA6/PS interface is not damaged upon sorption of water because good interfacial adhesion resulted from the small particle size and the occurrence of PS grafting onto PA6 during the preparation of the blend pellets and injection molding. The necking region of the elongated PA6/PS wet bars was observed by FESEM along the stretching direction, and the micrograph is shown in Figure 5b. PS particles were not stretched and did not act as stress concentrators because of their small size and good interfacial adhesion [17]. For the same reason, the flexural strength was also retained ( Table 2).
Also, as seen in Table 2, the notched impact strength of PA6/PS bars had a remarkable increase of 41.8% compared to that of neat PA6, achieving efficient toughening, although PS is brittle. This occurs because the interfacial adhesion is good enough to ensure stress transfer from the matrix to the particles, and more importantly, the large number of small PS particles generates a huge interfacial area that can effectively dissipate impact energy. The tensile tests were carried out on both dry and wet samples, while the flexural and notch impact tests were only carried out on dry samples. The typical stress-strain curves of PA6/PS bars along with those of neat PA6 bars are shown in Figure 5a. The tensile behaviors of PA6/PS bars were very similar to those of neat PA6 bars in both dry and wet states, indicating that the presence of immiscible PS particles does not cause any deterioration of the tensile properties of PA6 even in the wet state, although no compatibilizer was used. The good ductility of wet PA6/PS bars suggests that the PA6/PS interface is not damaged upon sorption of water because good interfacial adhesion resulted from the small particle size and the occurrence of PS grafting onto PA6 during the preparation of the blend pellets and injection molding. The necking region of the elongated PA6/PS wet bars was observed by FESEM along the stretching direction, and the micrograph is shown in Figure 5b. PS particles were not stretched and did not act as stress concentrators because of their small size and good interfacial adhesion [17]. For the same reason, the flexural strength was also retained (Table 2).
Also, as seen in Table 2, the notched impact strength of PA6/PS bars had a remarkable increase of 41.8% compared to that of neat PA6, achieving efficient toughening, although PS is brittle. This occurs because the interfacial adhesion is good enough to ensure stress transfer from the matrix to the particles, and more importantly, the large number of small PS particles generates a huge interfacial area that can effectively dissipate impact energy.
It has been reported that a PA6/PS (9/1, v/v) blend prepared in a twin-screw extruder at a screw speed of 100 rpm with a draw-down ratio of 1 can significantly reduce the tensile modulus and strength compared to neat PA6 [31]. The PA6/PS (9/1, w/w) bars injection molded from a compound prepared in a twin-screw extruder at 500 rpm have a slightly decreased tensile strength and almost unchanged toughness probably because of the much smaller PS particle size (1.5 μm) [38]. Even in the presence of 5 wt % of ethene-propene elastomer, the toughness of PA6/PS/EPR (9/5/5, w/w/w) only increases by 18% compared to neat PA6 [34]. The remarkable 41.8% increase in toughness with a maintained stiffness, strength, and ductility achieved in the PA6/PS (100/7.3, w/w) blend prepared in this work clearly demonstrates the advantage of fine PS particle size and good interfacial adhesion.   It has been reported that a PA6/PS (9/1, v/v) blend prepared in a twin-screw extruder at a screw speed of 100 rpm with a draw-down ratio of 1 can significantly reduce the tensile modulus and strength compared to neat PA6 [31]. The PA6/PS (9/1, w/w) bars injection molded from a compound prepared in a twin-screw extruder at 500 rpm have a slightly decreased tensile strength and almost unchanged toughness probably because of the much smaller PS particle size (1.5 µm) [38]. Even in the presence of 5 wt % of ethene-propene elastomer, the toughness of PA6/PS/EPR (9/5/5, w/w/w) only increases by 18% compared to neat PA6 [34]. The remarkable 41.8% increase in toughness with a maintained stiffness, strength, and ductility achieved in the PA6/PS (100/7.3, w/w) blend prepared in this work clearly demonstrates the advantage of fine PS particle size and good interfacial adhesion.

Conclusions
PA6/PS bars, injection molded from PA6/PS blend pellets, prepared by diffusion and subsequent polymerization of styrene in commercial PA6 pellets were shown to have significantly increased toughness with maintained stiffness, strength, and ductility compared to neat PA6 bars, because only limited coalescence occurs during injection molding. The small particle sizes and good interfacial adhesion between PS particles and the PA6 matrix resulted from the occurrence of PS grafting onto PA6 during the preparation of the blend pellets, and injection molding was able to ensure effective stress transfer from the matrix to particles and dissipate impact energy. PA6/PS bars were also shown to have an improved antiwater ability and dimensional stability due to the presence of hydrophobic PS.
This work proves that the PA6/PS quasi-nanoblend pellets prepared by diffusion and subsequent polymerization of styrene in polyamide 6 pellets are useful stock for plastic processing, and further confirms that diffusion and subsequent polymerization of a monomer (or monomer mixture) in polymer pellets is a good method for preparing polymer blend pellets in preparation for low-shear manufacturing processes such as injection molding.