Printer Friendly

Visual Analysis of Plastication of Long-Glass Fiber-Reinforced Resins Using a Glass-Insert Heating Cylinder.


Many automobile and mechanical parts are manufactured from resins, which are mechanically strengthened by adding fillers (such as glass or carbon fibers) during the injection molding. The mechanical strength of the molded products is affected considerably by the fiber length. Relatively long-fiber pellets yield the desired longer average fiber lengths, but some of the fibers in these pellets are inevitably broken in the molding process (both in the heating cylinder and in the plastication mold). This breakage hinders the production of molded products with consistent fiber lengths. The melting conditions of long-fiber-reinforced pellets are expected to be different from those of ordinary rice-like pellets. For this reason, elucidating the fiber breakage process during injection molding is important for improving the strength of fiber-reinforced resin materials. Fiber breakage is conspicuous during the screw-fed plastication process in the heating cylinder. The solid pellets agglomerate to form a solid bed (SB) through the solid conveying process in the cylinder. Thereafter, the surface of the pellet in the SB along the cylinder wall surface softens and melts to form a melt film. Exposed fibers from the pellets in the SB are subjected to strong shearing stress from the cylinder wall surface, which severely breaks them. Fiber breakage is conventionally analyzed by removing the screw once the resin has cooled (the so-called "screw pulling-out method") and then cutting resin samples from each region of the screw [1-8]. Alternatively, resins ejected from the nozzle are analyzed under varying plastication conditions or screw shapes [6, 9-12]. Screws with deep channels and low-compression ratios are regarded as being effective for suppressing fiber breakage. In other studies, researchers have theoretically predicted the fiber breakage distribution caused by shearing at each part of the screw [6, 12-16]. However, the plastication process of long-glass-fiber-reinforced resins during the solid conveying process, until the resins completely melt inside the heating cylinder, remains unknown because appropriate experimental methods have not yet been developed.

To visualize the inside of the heating cylinder, Zhu and Chen developed a method that visualizes extrusion molding through elliptical windows [17]. Using this equipment, they analyzed the movements of solid pellets during the plastication process of the extruder [18]. Wong and Liu studied the color mixing characteristics of polymers in single-screw extruder that visualize at eight places, four on each side, along the entire barrel length [19]. However, no similar technology currently exists for injection molding machines. Yokoi et al. proposed a glass-insertion method and developed a high-pressure, high-temperature visualization cylinder that inserts into injection molding machines [20]. This cylinder has been widely employed for analyzing nonreinforced resins [21-23]. When applied to short-glass fiber-reinforced resins in recent years, this device avoided opacification, abrasion on the inner surface of the glass-viewing windows, as well as breakage under high pressure. The plastication of resins containing up to 30 wt% glass fiber can also be viewed through this cylinder [24].

This study presents the first visual analysis of the plastication of 50 wt% long-glass fiber-reinforced polypropylene resin pellets. Experimental Devices and Methods section outlines the experiments conducted in this study, whereas Experimental Results and Discussion section analyzes the results. We tested the high-pressure resistance of the viewing window heating cylinder and clarified the continuous extrusion of the long pellets. After observing the feed, compression, and metering zones using full-flight screws with the same screw channel depth in the metering zone but different screw compression ratios, we established a model of the plastication process. Finally, we discussed the correlation between the plastication model and the fiber length distribution in the resin injected through the nozzle.


Visualization and Image Analysis Methods

The so-called "screw pulling-out method" is a method that is always used for investigating the inside of the heating cylinder. However, stopping the melting procedure during the natural cooling process of the cylinder and freezing the plastication process into a cooled sample are difficult tasks, and the unmelted pellets fall off while drawing the screw. In particular, freezing the sequence process is difficult for long pellets that are aligned and arranged. So, the application of dynamic visualization experiment is indispensable for understanding the phenomena in the cylinder. For dynamic visualization method, the glass extrapolation method is generally used to attach a glass block from the outside of the cylinder with a flange. Since the flange portion is large and cracks are more likely to be generated in the concave corner portion of the convex shaped glass block, it cannot be applied when the resin pressure is high. The glass interpolation method overcomes these problems by forming a key-like groove on the inner wall surface of the cylinder and loading a long glass inside. The glass insert minimizes the glass surface area in this cylinder, which is divided into three parts in the axial direction. Each part includes a long quartz-glass block with a camera-viewing window [20]. The resin moves at high-speed through the screw channel. Considering the heat conduction environment, we can conclude that for resin materials with high thermal insulation properties, the glass surface (which comprises only 5.3% of the inner wall surface area) exerts a near-negligible effect on the plastication. By virtue of these innovations, the cylinder can withstand high heat and high pressure and widens the observation area along the axial direction [20, 22]. The plastication of unreinforced resin has already been observed in countless visualization experiments and video images have been released for the public.

In this experiment, the same visualization cylinder as that in the previous report was used [24]. The present study employs a heating cylinder with three observation windows, each of which can observe the feed, compression, and metering zones of the screw. Each observation window is further divided into two parts. Beginning at the hopper side, the cylinder sequentially observes feed zones A and B, compression zones C and D, and metering zones E and F. While recording a video of the plastication process, the pressure is measured by a resin pressure sensor installed at 90 [degrees] upstream of the centerline along the width of each window.

As the observation window is only 6 mm wide, the high-speed video camera cannot visualize the whole plastication process. To resolve this problem, we constructed an extended image from a time series of snapshots captured at the observation window similar to that constructed in the previous report [21]. During the plastication process, the screw and melting conditions in the screw channel are rotated and observed, respectively. Connecting these views expands the visualization range and the subsequent expanded image is a subset of the entire image in the observation window. The center part (0.26-1.29 mm) was cut out in a slit shape (thin-frame part) from the observation image of the thick-frame part (6 mm). The center parts were then laminated in a time series (where the laminating direction represents the time axis) and synthesized. The image was created from bottom-to-top and then rotated by 90[degrees] to the right. The sloping white band in the image corresponds to the back part of the flight. In the expanded image, the horizontal axis denotes the time axis, while the vertical axis shows the axial distance of the screw from the hopper center. In every expanded image, the top and the bottom of the vertical axis denote the nozzle and hopper sides, respectively.

Experimental Method and Plastication Conditions

The following tests were conducted in an injection molding machine (Si 80 V: screw diameter D = 36 mm, manufactured by Toyo Machinery & Metal Co. Ltd., Hyogo Prefecture, Japan), equipped with the prototype visualization cylinder and an HSV500C3 high-speed video camera (NAC Image Technology Inc., Tokyo, Japan) [21]. Figure 1 is a schematic of the screw and cylinder measurements. The pressure fluctuation was measured in the observation areas of compression zones C and D (located 393.5 and 483.5 mm from the hopper center, respectively; see Fig. 1). The pressure sensor was an NP400 (Dynisco Co., Ltd., Kanagawa Prefecture, Japan) with a sensor diameter of 8 mm. The extended images were created from the observation window video using the image processing system Time-im (Library Co., Ltd., Tokyo, Japan).

The channel depth of the metering zone was fixed at 2 mm and the compression ratio of the full-flight screws was varied as 1.8 (FF1), 2.2 (FF2), and 2.75 (FF3). The dimensions of the three screws are listed in Table 1, while the molding conditions (shown in Table 2) simulate the conditions of the actual production site as far as possible. Continuous plastication experiments were performed in ascending order at 30, 60, 90, 120, and 150 rpm after stabilizing the melting process. The resin was long-glass fiber-reinforced polypropylene (L-5050P; Prime Polymer Co., Ltd., Tokyo, Japan) with a glass fiber diameter and length of 16 [micro]m and 8 mm, respectively, and a fiber content of 50 wt%. The pellets were cold cut to an average diameter of 3 mm and an average length of 8 mm. The rotational rates of screws FF1, FF2, and FF3 (kg/(rpm x h)) were 0.09, 0.16, and 0.18. The plastication capacity of the glass fiber-reinforced polypropylene (GFPP) resin as functions of rotational speed of FF1, FF2, and FF3 was also measured. The plastication capacity linearly increased with rotational speed, as well as with compression ratio (FF3 > FF2 > FF1). For example, at 150 rpm, the plastication capacity (g/s) is 3.55 for FF1, 5.73 for FF2, and 7.15 for FF3.

Measurement of Fiber Length

The fiber length after decomposition in an electric furnace (F-120-SP; Tokyo Glass Co., Ltd., Tokyo, Japan) was measured using a laser microscope (LEXT OLS 4000; Olympus Corporation, Tokyo, Japan). After heating for 3 h at 600[degrees]C in the electric furnace, the resin in the sample was completely turned to ash and only the glass fiber remained. The extracted glass fiber was defibrated in water and the fiber length was measured using a laser microscope. The measurement accuracy of the fiber length distribution was greatly affected by the total number of fibers [25]. To ensure the required measurement accuracy and experimental efficiency, we measured the lengths of 800-900 fibers in each sample. The fiber length distribution was computed as follows:

[L.sub.w] = [SIGMA] [n.sub.i] [L.sup.2]/[SIGMA] [n.sub.i] L, (1)

where L is the measured fiber length, [n.sub.i] is the fiber number, and is the weight-averaged fiber length. Very short fibers influence the number-average fiber length. Therefore, we used the weight-averaged fiber length, which is not easily influenced by very short fibers.


Visual Analysis of Plastication Process with Standard Full-Flight Screw (FF2)

Feed Zone. Simple observations of the GFPP through the window cannot clarify the melted and unmelted parts. Therefore, as reported previously [24], the images in this study were extended for a proper visual analysis.

Figure 2a shows an extended image of feed zone A observed at the highest rotational speed (150 rpm). In this feed zone, the solid material is conveyed along a direction parallel to the flight, while the shape of the long pellets is clearly distinguished. Owing to the shallow channel depth, the pellets cannot fill the channels sufficiently at any rotational speed and thus are interspersed with cavities and voids. The resulting state is similar to the starved-fed state (and is hereafter called the starved state). Because there is sufficient space in the screw channel, we can observe the rotations, falls, and other behaviors of the pellets. If the screw channel and pellets are shallow and long, respectively, the rotational space for changing the direction of the pellets from the hopper entrance into the screw channel is insufficient. In such situations, the long pellets intertwine and obstruct the entry of new pellets into the downstream channel. At high rotational speeds (Fig. 2b), the behaviors in feed zone B are identical to those in feed zone A. In contrast, at low rotational speeds (Fig. 2c), the pellets gather and fill the channel near the entrance of the compression zone. While filling the channels, some of the pellets were observed to melt, forming white streaks of extracted fibers.

Compression Zone. Figure 3 is an extended image of the compression zone. The simultaneous pressure readings are recorded along the straight line on each image. In the compression zone C (Fig. 3a) following the feed zone, the interior of the screw channel is uneven and the white streaks overlap in the horizontal direction (the direction of screw rotation) and the periodicity of these overlaps accords during the screw rotational rate of 30 rpm.

At rotational rates of 60 rpm or higher, an SB was formed along the trailing flight (see Fig. 3b recorded at 120 rpm), and a melt pool

(MP) (black band) was observed near the pushing flight. As large amounts of unmelted pellets remained in the SB, the pressure waveform was messy and the pellets in the SB were oriented at a fixed angle to the flight direction.

In compression zone D at high rotational speed ([greater than or equal to] 90 rpm or higher; see Fig. 3c), the width of the SB near the trailing flight decreased as the melting progressed, forming a wide uniform melting zone with white horizontal streaks superimposed along the pushing flight. As the resins were melted and conveyed, the pressure increased and the pressure waveform became smoother. A melt film (MF) was generated between the SB and barrel surface, while the dark area between the pellets spread to the screw channel near the pushing flight. In addition, many white streaks were superimposed along the lateral stripes. These streaks are signatures of the long fibers in the molten resins, which were dragged in the direction of the screw rotation (the horizontal direction in the figure). These streaks were generated when the melted pellets formed the MF between the SB and the barrel surface. As they moved, they slid into the gap between the cylinder surface (glass surface) and the SB. Along the trailing flight, the SB, which contained irregularly long pellets and developed a bent shape, moved under the generation of vigorous breakup behavior (BUP). The split form of the BUP was unstable, with repeated irregular unevenness at the MP side. Additionally, increasing the screw rotational speed increased the SB width and decreased the MP ratio. The originally formed SB was easily separated because the cohesive force between the pellets was low. When the pellets forming the SB aligned vertically to the flight direction, the joint interface between the pellets was aligned parallel to the pellets, and the cohesive force was weakened in the direction perpendicular to the interface (flight direction). Consequently, the cohesive force of the SB became anisotropic. When the component force in the flight direction acting on the SB (in which the pellets are aligned) increases, the SB is easily broken and the occurrence frequency of BUP increases. When compared with the BUP phenomenon, at 90 rpm, the SB became soft and deformed easily. Since the long-glass fibers maintain the cohesive force of the SB, they developed ductile fractures when divided in the BUP zone. However, at high rotational speed (150 rpm), as the screw channel became shallower, the SB interfered with the screw surface and the cylinder wall surface and BUP resulted from the brittle fracture behavior. We suggest that because BUP generation relates to the cohesive force and strength of the SB, it also relates to the fiber length in extruded resins.

Metering Zone. No SB appeared in zone E at rotational speeds of 30, 60, and 90 rpm. The whole surface was even, indicating that the resin had completely melted. At higher rotational speeds (120 and 150 rpm), the long shapes of some of the unmelted resin pellets (BUP fragments) were clearly visible, confirming that the fibers in these pellets had reached the melt zone without breaking. Therefore, they presumably remained intact within the injected resin. These unmelted pellets are partly responsible for the distribution of long (nearly 8 mm) pristine fibers in injected resins. In metering zone F (the zone nearest to the reservoir), the SB fragments at 150 rpm remained in the reservoir and the melting was incomplete.

Visual Analysis of Plastication Process by Full-Flight Screws with Different Compression Ratios

This subsection discusses the effect of screw compression ratio on the plastication process. For this purpose, we compared the plastication processes using the full-flight screws FF1 and FF3 (with compression ratios of 1.8 and 2.75, respectively), as well as a standard full-flight screw with a compression ratio of 2.2.

Plastication Process of the Low-Compression Ratio Screw (FF1). The plastication process of FF1 was observed from the feed zone to the metering zone (Fig. 4). This screw, along with a low-compression ratio, easily induced the starved state, as well as significant changes in the plastication process.

Feed zone. As observed for FF2, unfilled areas appeared in feed zone A of the FF1 plastication process. These unfilled areas exhibited a shallow channel depth at all the rotational speeds and were present at much higher percentages than in the FF2 process. In case of high rotation speed (120 rpm), the pellets were constantly deposited, while their distribution changed intermittently along the pushing flights. The cut long pellets were not observed, contradicting the general assumption that long pellets are sandwiched between the flight and the comers of the hopper hole at the start of feeding under the hopper and are eventually cut by a shallow channel screw. In the feed zone B, intermittently formed resin masses (groups of aggregated pellets) crossed over the channel. Fewer pellets were fed into the screw channel at the hopper throat in the FF1 process than in the FF2 process. This result is expected, because the channel is 0.8 mm shallower in the low-compression-ratio screw FF1 than in the standard screw FF2. Therefore, the number of pellets fed from the hopper decreased and the feeding became more difficult. In addition, increasing the rotational speed increased the percentage of unfilled areas in the channel, resulting in a starved state.

Compression zone. At 60 rpm in compression zone C (Fig. 4a), the melting resin was distributed through approximately half of the screw channel, giving rise to dark and bright areas in the images of the molten resin. The molten resin was uniquely shaped, with the bright areas surrounding the dark areas. The dark area is considered to be the mass of fibers and the molten resin defoamed from the around glass fiber after the resin melted and was pressurized sufficiently. On the other hand, the bright area is a melting resin mass whose outer periphery is featherlike, containing foamed bubbles around the glass fibers. The outer periphery of the bright areas was featherlike, containing foamed bubbles around the glass fibers; this shape confirms that the glass fibers were drawn out. The featherlike fibers were substantially smoothed in compression zone D (Fig. 4b), suggesting that they had re-entered the resin as the melting progressed. The dark area tended to decrease with increasing rotational speed; at 150 rpm (Fig. 4c) the resin mass thinned and divided in compression zone C, leaving signatures of the pellet shape. The pressure readings confirmed an almost constant pressure in the SB area after 1.5 s, as well as a pressure increase when the resins flowed from zone C to the metering zone. In addition, an increase in pressure near the pushing flight is observed at other rotational speeds. When the long-fiber pellets received heat from the cylinder in a sufficient space, they clumped after melting. It is presumed that this setup suppressed the fiber breakage and properly dispersed the fibers before injecting them through the nozzle.

Metering zone. In the metering zone E that follows the compression zone, the resin completely melted at low rotational speeds ([less than or equal to] 60 rpm); this was also observed in the FF2 plastication process. At 90 rpm and higher (refer Fig. 4d), the bright areas show that the fibers were defibrated and expanded, which created the unfilled areas, while the dark areas show that the mixed fibers and melted resin were unsegregated in most regions. At all rotational speeds, the bright and dark areas were observed in the same region to diagonally distribute in the same zone across the screw channel. The screw surface was occasionally seen at the outer boundary zones (unfilled areas) of the bright areas.

Plastication Process of the High-Compression Ratio Screw (FF3). Figures 5 and 6 are composed of extended images of the feed and compression zones in the FF3 plastication process, respectively. Owing to its deep channel, the screw with the high-compression ratio yielded a very different plastication process from that of FF2.

Feed zone. Figure 5 shows extended images of the feed zone in the FF3 process. In feed zone A, all screw channels were filled with pellets. The elongated pellets were aligned and filled orthogonally to the flight direction in the screw channels. Please recall that the extended images were created from the time series; therefore, the pellet angle is not directly measurable from Fig. 5. By setting the horizontal direction to 0 we directly measured the pellet orientation angle a in the counterclockwise direction from the window observation. The pellet angle in feed zone A at 90, 120, and 150 rpm was measured using the method and the results are plotted in Fig. 7. At all rotational speeds, the pellets were tilted approximately 30 [degrees] orthogonal to the flight. The orthogonal filling of the pellets into the channels might be explained by the close packing of the pellets in the channel, as well as the rolling and sliding movements. At this time, owing to the deepened screw channel, which tilts toward the trailing flight, the pellet movement will most likely be concentrated toward the trailing flight. This suggests that the flight slightly increases the inclination angle in the orthogonal direction (by approximately 10 0 in the present case). Next, let us focus on the interesting features in Fig. 5a. The feature labeled K1 in the extended image repeats in synchrony with the screw rotation (K2, K3). These features moved at uniform speed inside the screw, but their positions along the width of the channel were unchanged by the screw rotation. At low rotational speeds, a molten area in feed zone B was confirmed early in the heating cylinder. For instance, the resins processed at 30 rpm were quickly melted (Fig. 5b), while the pellets formed a vague wide and dark area along the pushing flight in feed zone B.

Compression zone. Figure 6 is an extended image of the compression zone of FF3. At 90 and 120 rpm, a clear BUP phenomenon appeared in compression zone C. During the 90 rpm process, different tilting angles were observed in compression zone D (lines marked M1-M2 and N1-N2 in Fig. 6a). The large N1-N2 tilt can largely be explained by the BUP occurring at 2.5 s, which increases the speed of the SB fragments flowing with the melted resin and relaxes the constraint conditions on the SB. At 150 rpm, the BUP phenomenon was clarified in the compression zone D (see Fig. 6(b)). A characteristic phenomenon, in which the melt flowed with the BUP fragment toward the pushing flight, was also observed in this zone (labeled L in Fig. 6b). This phenomenon might have been caused by the shallow screw channel, which compressed the SB between the screw and barrel surfaces. Consequently, as the screw rotated, the moving BUP fragment was forced in the direction of the pushing flight.

Metering zone. In metering zone E, the resin completely melted at low rotational speeds ([less than or equal to] 60 rpm), as observed in the FF2 process. At high rotational speeds ([greater than or equal to] 90 rpm), the melting was incomplete, as evidenced by the residual SB fragments.

Melting Process Model

Based on the above results, we constructed a model of the melting process (Fig. 8). When high-compression ratio screws are employed, most of the unmelted long pellets move rigidly and parallel to the vertical direction of the flight in the feed and compression zones, resulting in a compressed and integrated SB. Some of the pellets melt at the cylinder wall, leaking melted resin into the space between the pellets. The pellets bind together to form an SB. The fibers are oriented orthogonal to the flight in this SB, whose cohesive strength is rather weak. Increasing the rotational speeds increases the force component along the flight direction acting on the SB, generating an intense BUP phenomenon and forming an MF at the SB surface. The long fibers in the MF are drawn in the rotational direction, then dragged and swept away by the screw flight, thus forming a MP. The softened SB and BUP fragments are also dragged and flow spirally into the MP due to spiral flow of in-channel MP. In contrast, when low-compression ratio screws are employed, the shallow channel in the feed zone decelerates the feeding speed of the long pellets into the screw channel, creating unfilled starved state areas in the cylinder. Unfilled areas also appear in the compression zone and are observed in the metering zone at high speeds. When the temperature rises during the solid conveyance of the starved state, the fiber bundle in the pellet expands in a foamy state, while the periphery of the pellets becomes cottony to promote the fibrillation. When the resin enters the subsequent channel, its pressure increases and it mixes and clumps with the fibers to create molten resin, completing the plastication. These results demonstrate that the starved state significantly changes the plastication process.

Evaluation of Fiber Breakage

As the FF1, FF2, and FF3 processes show the same tendencies at all rotational speeds, we discuss only the results at 120 rpm. We measured the fiber length distributions in resin samples produced from the nozzle using screws FF1, FF2, and FF3 rotating at 120 rpm. The results are shown in Fig. 9. Long fibers (~8 mm, equaling the pellet length) were obtained in the FF3 and FF2 processes. Given that the SB forms when the long pellets are aligned, the BUP should easily form before the SB has sufficiently melted. For all screws, the fiber length distribution was concentrated in the range of 0.3-0.6 mm. The FF1 process yielded far fewer extremely short fibers (0.6 mm or less) than that of the FF2 process, as well as more fibers in the range of 0.6-2.4 mm. On the other hand, FF3 yielded many fibers in the range of 0-0.3 mm, and fewer fibers in the range of 0.3-1.2 mm.

Figure 10 relates the weight-averaged fiber lengths to the screw torques at different rotational speeds of each screw. For all screws, the fibers were more easily broken as the rotational speed increased. At high rotational speeds, the shear rate (amount of deformation per unit time) of the melting resin increased. Because the moving speed of the resin also increased, the cool viscous resin, which included the unmelted pellets, was plasticized and conveyed, increasing the torque. As the rotational speed increases, the viscosity of the molten resin also increases. The fibers were entangled in the metering resin and, as the shear rate increased more rapidly than the number of entangled fibers in the metering resin, the fibers were thought to be subjected to high stress originating from their intertwined portions. Consequendy, increasing the shearing frequency increased the likelihood of breaking the fibers.

Comparing the fiber length-torque relationships of the different screws, we inferred that increasing the compression ratio increased both the amount of resin fed from the hopper and the torque at a given rotational speed. When the torque increases, the shear stress on the glass fibers is strengthened and the fibers are more easily broken. These findings suggest that to maintain long fibers, we should reduce the amount of resin to the starved state. Reducing the resin to the starved state enables sufficient heating of the resin, promoting stable melting at low torques. Correlating these results with the aforementioned melting model, we observed that all the 8-mm fibers broke when the FF1 screw was used, while the fiber length distribution was concentrated in the range of 0.6-2.4 mm, indicating a good dispersibility. However, when the FF2 and FF3 screws were used, we assumed that the BUP fragments contained 8-mm long fibers did not disperse when injected; therefore, these processes are assumed to be unsuitable for increasing the strength of the molded products.

Conventionally, long-glass-fiber-reinforced resins have been fabricated using deep channel screws with low-compression ratios. Here, we confirmed the effect of screws with low-compression ratios. Furthermore, the feed zone must contain deep channels for stably transporting the long-fiber pellets, while avoiding unfilled areas. However, we observed that shallow channels in the feed zone cause the starved state, effectively suppressing the fiber breakage and improving the dispersion. As the strength of the molded products is influenced by both the length and dispersibility of the fibers, a more comprehensive evaluation is required.


In this study, the plastication process of long GFPP was observed through a visualization cylinder and the influence of different screw compression ratios was clarified at the same channel depth of the metering zone. The following conclusions can be drawn from our findings.

1. The plastication process of long-GFPP (GF50 wt%) was visualized using three full-flight screws having the same channel depth of the metering zone but different compression ratios. SB, MP, and BUP phenomena were clearly observed in the feed and compression zones. The plastication process strongly depended on the compression ratio.

2. In the compression zone, at a compression ratio of 2.2 (screw FF2), and in the feed zone, at a compression ratio of 2.75 (screw FF3), the long pellets were aligned at approximately 10[degrees] from the orthogonal direction to the flight inclination angle, decreasing the cohesion of the SB. At high-screw rotational speeds ([greater than or equal to] 120 rpm), some long pellets were retained in the BUP fragment of the SB in the metering zone, resulting in some unbroken, pellet length fibers.

3. The plastication process of the high-compression ratio screw (2.75 in screw FF3) was similar to that of the standard screw (FF2); although the melting process in the former was obviously delayed. When using a low-compression ratio screw (1.8 in screw FF1), the feed rate of pellets into the screw channels was decreased by the shallow channel depth in the feed zone, resulting in the starved state. Wide unfilled areas and featherlike edges were uniquely observed in the compression zone, and even in the metering zone, at high rotational speeds. Owing to the shallow channel depth in the feed zone, the pellets were heated in the starved state during the feeding process and the fiber bundle expanded from the pellet in a foamed state, promoting featherlike fibrillations. Therefore, it was presumed that fiber breakage was suppressed in the subsequent melting process in the compression zone.

4. Increasing the compression ratio increased the screw torque and the resin pressure, thereby reducing the fiber length of the extruded resins. FF2 and FF3, with similar melting processes, showed similar fiber length distributions. The low-compression ratio screw FF1 with a different melting process and form of fiber length distribution yielded dramatically longer weight-averaged fiber lengths than that of FF2 and FF3.


The authors would like to express their gratitude to Prime Polymer Co., Ltd. for providing the resins, and members of this study, which was conducted as part of the 2013 and 2014 "Ultimate Injection Molding Project" funded by the Foundation for the Promotion of Industrial Science.


[1.] R. von Turkovich and L. Erwin, Polym. Eng. Sci., 23(13), 743 (1983).

[2.] B. Franzen, C. Klason, J. Kubat, and T. Kitano, Composites, 20 (1), 65 (1989).

[3.] V.B. Gupta, R.K. Mittal, P.K. Sharma, G. Mennig, and J. Wolters, Polym. Compos., 10(1), 8 (1989).

[4.] V.B. Gupta, R.K. Mittal, P.K. Sharma, G. Mennig, and J. Wolters, Polym. Compos., 10(1), 16 (1989).

[5.] P. Ren and G. Dai, Fibers Polym., 15(7), 1507 (2014).

[6.] C.-T. Huang, H.-C. Tseng, J. Vlcek, and R.-Y. Chang, 10P Conf. Ser.: Mater. Sci. Eng., 87, 012023 (2015).

[7.] H. Chen, M. Cieslinski, and D.G. Baird, MP Conf. Proc., 1664, 060004 (2015).

[8.] A. Inoue, K. Morita, T. Tanaka, Y. Arao, and Y. Sawada, J. Compos. Mater., 49(1), 75 (2015).

[9.] E. Lafranche, P. Krawczak, J.P. Ciolczyk, and J. Maugey, Express Polymer Letters, 1(7), 456 (2007).

[10.] H. Zhuang, P. Ren, Y. Zong, and G.C. Dai, Express Polym. Lett., 2(8), 560 (2008).

[11.] G. Jin, X. Lin, G. Tian, S. Zhang, M. Wang, and X. Wang, J. Reinf. Plast. Compos., 35(2), 111 (2016).

[12.] C.-T. Huang and H.-C. Tseng, Polym. Eng. Sci., 58, 452 (2017).

[13.] S.-Y. Fu and B. Lauke, Compos. Sci. Technol., 56(10), 1179 (1996).

[14.] J.L Thomason and M.A. Vlug, Compos. Part A, 27(6), 477 (1996).

[15.] J. Wang, C. Geng, F. Luo, Y. Liu, K. Wang, Q. Fu, and B. He, Mater. Sci. Eng. A, 528(7-8), 3169 (2011).

[16.] T. Huan-Chang, R.-Y. Chang, and C.-H. Hsu, J. Rheol., 57, 1597 (2013).

[17.] F. Zhu and L. Chen, Polym. Eng. Sci., 31(15), 1113 (1991).

[18.] S. Fang, L. Chen, and F. Zhu, Polym. Eng. Sci., 31(15), 1117 (1991).

[19.] A. C.-Y. Wong and T. Liu, "Screw Configuration Effects on the Color Mixing Characteristics of Polymer in Sngle-Screw Extrusion," in ANTEC'98, Vol. 1, 284-287. Available at: 4&dq=Screw+Configuration+Effects+on+the+Color+Mixing+Cha racteristics+of+Polymer+in+Single-Screw+Extrusion&hl=ja&sa=X&ved=0ahUKEwjm-_Lz5NbhAhUiG6YKHRB4C5UQ6AEIKT AA#v=onepage&q=Screw%20Configuration%20Effects%20on% 20the%20Color%20Mixing%20Characteristics%20of%20Polymer %20in%20Single-Screw%20Extrusion&f=false.

[20.] H. Yokoi, Int. J. Electr. Mack, 34(75), 6 (2000).

[21.] H. Yokoi and M. Tatsuno, Seikei-Kakou, 11(11), 922 (1999).

[22.] M. Tatsuno, Y. Takahashi, and H. Yokoi, Seikei-Kakou, 13(2), 125 (2001).

[23.] M. Tatsuno and H. Yokoi, Seisan Kenkyu, 54(3), 167 (2002).

[24.] S. Ma, K. Shibata, and H. Yokoi, Seikei-Kakou, 30(1), 37 (2018).

[25.] S.-Y. Fu, Y.-W. Mai, E.C.-Y. Ching, and R.K.Y. Li, Compos. Part A, 33(11), 1549 (2002).

Sai Ma (ID), (1) Kazuyuki Shibata, (2) Hidetoshi Yokoi (1)

(1) Institute of Industrial Science, The University of Tokyo, 4-6-1, Komaba, Meguro-ku, Tokyo, 153-8505, Japan

(2) Toyo Machinery & Metal Co., Ltd, 523-1, Fukusato, Futamicho, Akashi City, Hyogo, 674-0091, Japan

Correspondence to: S. Ma; e-mail:

DOI 10.1002/pen.25113

Published online in Wiley Online Library (

Caption: FIG. 1. Schematic of the screw and cylinder axial coordinates.

Caption: FIG. 2. Extended images of the screw channels (FF2; feed zone).

Caption: FIG. 3. Extended images of the screw channels (FF2, compression zone). The upper graph in each panel plots the simultaneously recoded pressure readings.

Caption: FIG. 4. Extended images of the screw channels (FF1, compression and metering zones) and the simultaneously recorded pressure variations (upper plots in each panel).

Caption: FIG. 5. Extended images of the screw channels (FF3, feed zone).

Caption: FIG. 6. Extended images of the screw channels (FF3; compression zone) and the simultaneously recorded pressure variations (upper plots in each panel).

Caption: FIG. 7. Binned pellet angle distributions at different screw rotational speeds (FF3, feed zone).

Caption: FIG. 8. Proposed model of the melting process.

Caption: FIG. 9. Distributions of glass fiber lengths under the conditions of FF1, FF2, and FF3.

Caption: FIG. 10. Relationship between torque and weight-averaged fiber lengths.
TABLE 1. Screw dimensions.

Appellation                             FF1   FF2    FF3

Screw diameter (mm)                            36
Effective length (mm)                         720
Screw pitch (mm)                               36
Flight width (mm)                              4
Screw clearance (mm)                          0.06
Channel depth (mm)      Metering zone         2.0
                          Feed zone     3.6   4.4    5.5
Compression ratio                       1.8   2.2    2.75

TABLE 2. Experimental conditions.

                                     230 (nozzle), 230 (metering zone),
                                          230 (compression zone),
Cylinder temperature ([degrees]C)    220 (first half of the feed zone),
                                    210 (second half of the feed zone),
                                           40 (under the hopper)
Back pressure (MPa)                                  15
Drying temperature ([degrees]C)                      80
Screw rotational speed (rpm)                30, 60, 90, 120, 150
COPYRIGHT 2019 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ma, Sai; Shibata, Kazuyuki; Yokoi, Hidetoshi
Publication:Polymer Engineering and Science
Article Type:Report
Geographic Code:9JAPA
Date:Jun 1, 2019
Previous Article:Effect of Recycling and Injection Parameters on Gloss Properties of Smooth Colored Polypropylene Parts: Contribution of Surface and Skin Layer.
Next Article:Thermoplastic Encapsulations of a Sensor Platform by High-Temperature Injection Molding up to 360[degrees]C.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters