«Viscoelastic flow effects in multilayer polymer coextrusion Dooley, J. DOI: 10.6100/IR555718 Published: 01/01/2002 Document Version Publisher’s ...»
One disadvantage in using layered structures to analyze the elastic layer rearrangement in coextruded systems is the fact that the layers are all aligned in one direction. This means that movements of the layers in a horizontal or vertical direction will produce different visual patterns. One way to circumvent this problem would be to develop structures that are not composed of layers but of strands positioned in a specific array. This was done by designing a feedblock that produced a unique coextruded structure consisting of an array of individual strands of one polymer surrounded by a matrix of another polymer (see Figure 5-1 for an illustration of a 49-strand structure).
Figure 5-9 shows a schematic diagram of a multi-strand feedblock (73) that would produce a 6 row by 6 column array of strands. This diagram shows how the polymer “A” is distributed to the strands and then encapsulated by a matrix material “B”. It also shows how material from an additional flow stream “C” may be added to encapsulate the strands with another layer.
Figure 5-10 shows an example of a cross section of a structure that could be produced by this type of feedblock in which an array of 9 rows by 9 columns of strands is produced in which each strand has been individually encapsulated.
This style of feedblock allows tracing of flow paths within a particular geometry by following the distortions of the individual strands from the entry to the exit of the channel which can be used to predict layer interface location(s) in coextruded layered structures. This feedblock not only allows tracing for interface locations for coextrusion flows, but also can be used to show the flow patterns in monolithic viscoelastic polymer flows.
The initial coextrusion experiment was conducted with a 49-strand feedblock (7 rows by 7 columns). This experiment consisted of operating the feedblock with no die attached to ensure that the strands were uniform in diameter and location at the exit of the feedblock before they entered the die. Figure 5-11 shows the strand profile for the polystyrene resin at the end of the feedblock. The 49 strands of polystyrene were pigmented black and the matrix polystyrene pigmented white so that the interfaces could be clearly distinguished between the two. Note that the positions of the strands at the exit of the feedblock clearly show the 7 row by 7 column pattern and the strands are fairly uniform.
Figure 5-11. Forty-nine-strand polystyrene structure near the exit of the feedblock
The next experiment consisted of running the same polystyrene resin through the 49strand feedblock system with an attached die with a square cross-sectional shape. Crosssectional cuts of this sample taken at the entry to the channel (shown on the left in the figure) and at 50 cm downstream from the feedblock (on the right in the figure) are shown in Figure 5-12.
This figure clearly shows that the strand movement is symmetrical within each quadrant as material moves along the channel walls into the corners and then toward the center of the channel along the 45 degree diagonal line. This figure also shows how each individual strand has moved and been deformed as it flowed down the channel. Note that the central point in the original matrix has remained essentially unchanged at the exit since it represents a neutral or stagnant point with respect to the secondary flows.
Figure 5-12. Forty-nine-strand polystyrene structure near the entry and exit of the square channel The strands shown in Figure 5-11 rearrange progressively as the polymer flows down the channel producing a more distorted strand profile as the distance traveled increases. This phenomenon is shown in Figure 5-13. In this figure, the interfaces are shown at several intervals from the entry of the channel to near the exit (viewed from left to right, respectively). This figure allows the tracing of the flow path of any of the individual strands as it flows down the channel. This type of figure gives a much better understanding of the deformations occurring due to secondary flows compared to structures with parallel layers.
Figure 5-13. Progression of a forty-nine-strand polystyrene structure as it flows down a square channel Figure 5-14 shows the forty-nine-strand polystyrene structure near the entry (top) and exit (bottom) of the rectangular channel. This figure gives an indication of just how much deformation actually occurs in the rectangular channel. Note the high degree of elongation of the strands near the top and bottom walls as well as the compression of the strands near the side walls. Even though the overall deformation in this geometry is less than the square geometry, there is still significant deformation for the elastic polystyrene resin.
Figure 5-14. Forty-nine-strand polystyrene structure near the entry and exit of the rectangular channel Figure 5-15 shows the forty-nine-strand polystyrene structure near the entry (left) and exit (right) of the teardrop channel. These images show that the movement of the strands in the teardrop geometry is very large and the flow field appears to cause more deformation in the strands than was evident in the rectangular geometry. It is also interesting to note that the strand that started in the center of the 7 by 7 matrix at the entry of the channel has deformed at the exit of the channel. This differs from the results of the square and rectangular channels since the central strand was not deformed in those geometries. This deformation that occurs to the central strand in the teardrop channel is probably due to the fact that there is not a central neutral or stagnant point in the center of the channel due to the different symmetry that exists. This will be discussed in more detail in the numerical simulation section.
Figure 5-15. Forty-nine-strand polystyrene structure near the entry and exit of the teardrop channel Figure 5-16 shows the forty-nine-strand polystyrene structure near the entry and exit of the circular channel. Note that the location and shape of the strands is essentially the same at the entry and exit of the channel. This once again confirms the fact that even with a very elastic material like polystyrene, no secondary flows are developed in a radially symmetric channel.
This demonstrates the importance of understanding how channel geometry relates to viscoelastic flow effects when designing equipment for polymer processing.
Figures 5-17, 5-18, and 5-19 show a series of cross-sectional cuts made at 7.6 cm intervals down the rectangular manifold channel for the polycarbonate, polyethylene, and polystyrene samples, respectively. Again the cut on the left was made nearest to the feedblock while the cut on the right was near the end of the manifold.
Figure 5-17 shows the progression of a forty-nine-strand polycarbonate structure as it flows down a rectangular manifold channel. Near the entry (on the left), all 49 strands in the matrix are still visible. As the flow progresses down the channel (left to right in the figure), the rows of 7 strands flow fairly uniformly out of the bottom of the channel leaving the rows above intact. This continues until the final cut where only the row that originated at the top of the channel at the entry remains. This is very important because it implies that any small defect in a coextruded structure that is introduced near the top of the die channel at the entry will be magnified by the time it reaches the end of the channel. Comparing the relative heights of the top row of strands near the entry and the exit of this manifold channel shows this phenomenon.
Near the entry, the top row occupies approximately one seventh (~15%) of the channel height while near the end it occupies more than half (50%) of the channel height.
Because the flow out of this manifold is so uniform for the polycarbonate material, this combination of manifold and material would be a good choice for an inverse design of the feedblock. The feedblock channels could be designed to produce a film or sheet with proper layer distribution in a fairly straightforward manner.
Figure 5-17. Progression of a forty-nine-strand polycarbonate structure as it flows down a rectangular manifold channel Figure 5-17 demonstrates how uniform the flow can be down a die manifold channel for an inelastic material like polycarbonate. However, the flow for more elastic materials like polyethylene and polystyrene are not so uniform in this type of manifold channel. Figure 5-18 shows the progression of a forty-nine-strand polyethylene structure as it flows down a rectangular manifold channel. As in the polycarbonate sample, all 49 strands are visible at the entry of the channel. However, as the strands flow down the channel there are significant differences in how the strands flow out of the bottom of the channel. In the polycarbonate structure (Figure 5-17), the tops and bottoms of each row of strands remained essentially flat as they moved down the channel. In the polyethylene structure (Figure 5-18), the ends of the rows begin to exhibit curvature as they flow down the channel. This curvature means that each row will not flow out of the manifold as a group but will be spread out as the material near the center of the channel flows out sooner that that near the walls. This is shown in Figure 5-18 by the next to last sample from the end of the channel. In this cut, there are 9 strands across the width of the channel near the bottom as opposed to 7 strands at the beginning. This means that one row of strands has now overlapped with another showing that they are not flowing out in the same manner as the polycarbonate strands.
Figure 5-18. Progression of a forty-nine-strand polyethylene structure as it flows down a rectangular manifold channel Figure 5-19 shows the progression of a forty-nine-strand polystyrene structure as it flows down a rectangular manifold channel. This figure shows very similar results to the polyethylene results shown in Figure 5-18. However, the distortions in the tops and bottoms of the rows of strands are even more exaggerated than in the polyethylene sample. The reason for the distortion of these strands can be determined by looking at the results shown in Figures 5-7 and 5-8 for the 165-layer polystyrene sample. These figures clearly show the secondary flows in the upper corners of the manifold channel that become magnified as the structure flows down the channel.
These secondary flows would cause the material in the center of the channel to flow downwards more quickly while the material near the walls would be forced upwards and exit the channel later. This phenomenon produces the curvature of the tops and bottoms of the rows of strands in the more elastic materials.
5.4 Concentric ring feedblocks (2 and 13 rings) As was described earlier, the two main styles of coextruded structures are parallel layers and concentric layers. While parallel layers are used in many film and sheet structures, many structures with concentric layers, such as in wire coating, are also used. This section will focus on the effect of secondary flows on coextruded structures composed of concentric layers.
Coextruded structures containing concentric layers were formed using a specially designed feedblock shown schematically in Figure 5-20 (74). This diagram shows the tip of the extruder screw on the right side where the polymer flows into the concentric ring feedblock.
This feedblock will produce a core layer of polymer A encapsulated by a layer of polymer B.
Adding more encapsulating sections to the feedblock will produce coextruded structures with more concentric layers. This concept is very similar to the one used in stackable plate blown film dies as was shown previously in Figure 1-2.
Figure 5-21 shows an example of a coextruded structure composed of concentric layers.
This figure shows a two-layer concentric ring polystyrene structure near the entry of a circular channel. This polystyrene structure is composed of a 30% skin layer and a 70% core layer. The initial series of experiments were conducted with structures that contain two layers.
Figure 5-21. Two-layer concentric ring polystyrene structure near the entry of the circular channel Figure 5-22 shows the same two-layer concentric ring polystyrene structure that was shown in Figure 5-21 but near the exit of the circular channel. As was observed in the parallel layered and strand structures, no significant deformation of the concentric layered structure occurs as it flows down the circular channel. This again demonstrates that no secondary flows occur in a radially symmetric structure.
Figure 5-22. Two-layer concentric ring polystyrene structure near the exit of the circular channel Figure 5-23 shows the two-layer concentric ring polystyrene structure after it has been extruded into the entry of the square channel. This structure also contains a 30% skin layer of polystyrene that has been pigmented black to allow observation of the layer interface.
As the two-layer concentric ring polystyrene structure flows down the square channel, the layers rearrange and produce the structure shown in Figure 5-24. Note that the layers near the walls have become much thinner than in the original structure while material has flowed from the corners toward the center of the channel along the 45 degree diagonals. This result is very similar to that seen previously in the structures with initially parallel layers. This result also follows the trend predicted for secondary flows produced by elastic effects, as was shown in Figure 4-6. This figure demonstrates that the same elastic forces are present in structures composed of concentric rings as are present in structures composed of planar layers.
Figure 5-24. Two-layer concentric ring polystyrene structure near the exit of the square channel Figure 5-25 shows a two-layer concentric ring polystyrene structure near the entry of the teardrop channel. Note that the outer black layer is fairly thick and uniform around the structure.