Elimination of a restriction at the entrance of barrier flighted extruder screw sections.
Barrier flighted screws have been used for many years to increase rates, eliminate flow surging resulting from solid bed breakup, lower melt temperatures, and minimize temperature fluctuations in the extrudate. For most barrier screws, the melt separation flight starts at a location where the melt pool starts to accumulate. However, if the start of melting occurs downstream from this point, compacted solid polymer can be wedged into the start of the barrier entrance. For some barrier screws, this wedging can lead to a restriction in the flow, decreasing the rate of extrusion by up to 30%. This article discusses the phenomenon and shows how to eliminate the problem.
The melt separation concept, developed by Maillefer in the 1950s, has had wide acceptance. Since then, many numerical simulation techniques have been developed to aid in the design, analysis, and prediction of melting, temperatures, pressures, and rates.[2,3]
Most barrier screws have a constant-depth feed section upstream from the barrier section, and metering sections, and possibly, other sections, downstream. At the entrance of the barrier section, a barrier flight is introduced that divides the main channel into two channels: one on the trailing side for the solids and one on the pushing side for the melted polymer. The clearance between the barrier flight and barrel wall is large enough to permit the flow of melted polymer, but small enough to prevent the flow of solids. Material is melted in the solids channel and then dragged over the barrier flight via the relative motion of the barrel with respect to the screw. It is then conveyed to the end of the barrier section via the melt channel. Numerous designs exist for the depths and widths of the two channels and also for the pitch of the barrier flight.
TABLE 1. Extrusion Rates and Performance Before and After Modifications to the Entrance of the Barrier Section for Case 2 Extruder Screw (88.9 mm Diameter, 33:1 L/D).
Before After Modification Modification
Rate, kg/hr 148 148 Screw speed, rpm 91 69 Specific rate, kg/(hr rpm) 1.63 2.14 Extrudate temperature, [degrees] C 232 223 Gel showers Yes No TABLE 2. Extrusion Rates and Performance Before and After Modifications to the Entrance of the Barrier Section for Case 3 Extruder Screw (152.4 mm Diameter, 26:1 L/D).
Before After Modification Modification
Rate, kg/hr 272 272 Screw speed, rpm 64 51 Specific rate, kg/(hr rpm) 4.25 5.33 Extrudate temperature, [degrees] C 235 224 Gel showers Yes No
Ideally, to maintain high rates, a portion of the polymer should be melted by the time it enters the barrier section. The actual fraction of melted polymer depends on the process conditions, the physical properties of the resin, and the cross-sectional areas for both the feed section and the solids channel of the barrier section. In general, the cross-sectional area of the feed section is larger than the area of the solids channel of the barrier section. This occurs with most designs because the solids channel depth and width usually decrease with axial length. Moreover, the width of the feed section is generally greater than that for the solids channel. If enough of the solids are melted at the exit of the feed section (such that the cross-sectional area of the solid bed is less than the area of the solids channel of the barrier section), then the solids will easily flow into the solids channel and the melted polymer will flow across the barrier flight into the melt channel. Most barrier screws function this way, producing a high-quality extrudate at uniform and high rates. Occasionally, not enough of the polymer entering the barrier section is melted - forcing the solid bed into the start of the barrier flight or wedging the bed into the solids channel of the barrier section. This wedging sometimes causes a restriction in the flow. The restriction can reduce the rate of the extruder by up to 30% and cause the melt temperature to increase. The rate loss can lead to void regions in the screw where the polymer will have extremely long residence times and can thus potentially degrade.
The purpose of this article is to present three studies where restrictions at the entrance to the barrier section caused reduced rates, high extrudate temperatures, and degraded resin that contaminated the product. The barrier screws for these cases were designed and manufactured by different companies. Modifications to the screws that eliminated these problems are discussed.
A two-stage extruder, 50.8-mm diameter, typically operated at a rate lower than design. For example, the rate for a high-impact polystyrene (HIPS) resin ranged from 34 kg/hr at 50 rpm to 76 kg/hr at 150 rpm - specific rates of 0.68 and 0.51 kg/(hr rpm), respectively. The calculated drag flow rate of the first-stage meter of the screw - the rate determining section - was 0.68 kg/(hr rpm). However, in a vented system, the output is expected to be above the first-stage drag flow, perhaps by 10% to 30%, because of the positive effect of pressure flow. Because the pressure in the channel at the vent must be zero, and the measured flow rate was less than the drag rate, it was concluded that the first-stage meter section was operating only partially filled. Axial pressures were consistent with this conclusion; the pressure at the exit of the first stage transition was measured at 2.2 MPa at 150 rpm.
The screw, shown in Fig. 1, had a barrier flight in the first stage; lead length of the main flight was 50.8 mm. The barrier flight started from the pushing side of the main flight and increased in pitch, creating the two flow channels. The cross-sectional area of the feed channel was 4.2 [cm.sup.2]. Total cross-sectional area of the solids channel, one-half diameter into the barrier section, was only 3.3 [cm.sup.2] - a reduction in area of 21%.
It was hypothesized that a restriction at the entrance to the barrier section was reducing the rate. To mitigate the restriction, a screw modification was made for the first three diameters of the barrier section. An increase in depth to that of the solids channel at the entrance was tapered into the melt channel over two diameters, Then the barrier flight was removed for the first two diameters and blended in with the melt channel, the shallower of the two channels in this region. The barrier flight for the next diameter was blended into the original undercut. Figure 2 shows cross sections of the barrier section before and after the modification; they are perpendicular to the main flight at two diameters into the section. The modification was made in an attempt to permit some of the solids to pass into the melt conveying channel, to eliminate the wedging of solids against the entrance of the barrier flight by streamlining the flight, and to increase the specific rate of the screw.
After the modification, the specific rate increased by 5% to 24% depending on screw speed, as indicated by Fig. 3. Moreover, the time-averaged pressure in the channels increased after the modification, as indicated by the axial pressure data in Fig. 4. The increased pressure at the middle of the transition section in the first stage (at 14 diameters) indicated better filling of the flights. Before modification, the pressure at this transducer measured zero for a fraction of the time period. After modification, the pressure at this transducer was never zero. Extrudate temperature, however, was essentially unchanged. Although significant improvements in output were made by removing the restriction at the entrance to the barrier section, the full capabilities of this screw were not realized during extrusion of HIPS. Most likely, a different screw design would be required for optimum extrusion of HIPS.
A new single-stage extruder, 88.9 mm in diameter and 33:1 L/D, was installed in a film plant running a low-density polyethylene (LDPE) resin. At the startup of the line, the extruder was operated at 91 rpm to produce the required rate of 148 kg/hr, a specific rate of 1.63 kg/(hr rpm). The temperature of the extrudate was 232 [degrees] C. This extrusion rate was required in order to maintain the downstream take-away equipment at its maximum rate. The extruder appeared to be operating well, except that the specific rate was lower than predicted. That is, the screw was rotated at an rpm that was higher than expected to produce the 148 kg/hr. At 91 rpm, the drag flow rate was estimated at 228 kg/hr. Thus, the line was operating at only 65% of the drag flow rate. After 12 days, the line began to experience intermittent discharges of crosslinked material (gel showers) and carbon specks. In some cases, the gel showers were observed two to three times per day and would last from 1 to 5 minutes. This led to production downtime, required for purging, and numerous customer complaints.
The screw, shown schematically in Fig. 5, had constant-depth meter sections of 5.8 mm. The lead length was 124 mm for the main flight of the barrier section and 88.9 mm for all other sections of the screw. The lead length for the barrier flight was larger than the barrier main flight, because the barrier flight started at the pushing side of the main flight at the entrance to the section, and transitioned to near the trailing side at the end of the section. The width of the feed channel perpendicular to the flight was 75 mm; the widths of the solids and melt channels of the barrier section near the entrance were 75 mm and 24 mm, respectively. Both the width and depth of the solids channel for the barrier section decreased in the flow direction. The barrier flight undercut, or the radial distance between the main flight land and the barrier flight land, was 1.5 mm.
To determine the cause of the reduced rate, high melt temperature, and degraded material, the authors performed a polymer solidification experiment by stopping the rotation of the screw and cooling the polymer in the channels. Examination of the solidified polymer indicated that in the first meter section, about half the channel width on the trailing sides of the flights, for all but the last diameter (approximately 11 diameters), was filled with a dark-colored, partially carbonized LDPE resin. This indicated that these regions were stagnant. The reduced flow rate caused these regions to be partially filled, creating void regions on the trailing side of the channel. Some of the resin adhered to the trailing side of the screw in the void regions and stayed there for extended time periods. This resin eventually degraded into the dark-colored, cross-linked material. Small process variations dislodged some of this material and caused the intermittent gel showers that contaminated the product. Moreover, compacted solids were found wedged in the channel at the entrance to the barrier section. The wedged material was caused by the relatively large width of the entering solid bed being forced into the continually decreasing width of the solids channel of the barrier section.
An estimate of the pressure gradient in the screw channel was calculated to determine independently if the screw channels were full. The pressure gradient that would be needed to reduce the rate from the drag flow value to that measured for this LDPE resin was estimated at 1.6 MPa/diameter. The pressure measured at the extruder discharge was 13 MPa. Thus, about 8 diameters of metering section were required to generate this pressure, suggesting that about 9 diameters in the first metering section were at zero pressure and thus only partially filled. This numerical analysis is very consistent with the experimental results described above. Experimentally, about 11 diameters in the first metering section were partially filled or at zero pressure. These data and computations lead to the conclusion that the barrier section flow was not matched properly to the flow requirements of the metering sections.
Based on this conclusion, the screw was modified in an attempt to remove the restriction at the entrance of the barrier section. The barrier flight was removed for the first 4.5 diameters and tapered up to its full height over the next half diameter. This modification was similar to that for Case 1 and is shown in Fig. 2.
After the screw modification, the 148 kg/hr rate was obtained at a screw speed of about 69 rpm with an extrudate temperature of 223 [degrees] C. Thus, the specific rate increased from 1.63 kg/(hr rpm) before the modification to 2.14 kg/(hr rpm), a specific rate increase of about 30%. At 69 rpm, the drag flow rate was calculated at 173 kg/hr; now, the extruder was operating at about 86% of the drag flow rate. The calculated axial pressure gradient required to maintain the flow of the extruder at the reported flow rate showed that pressures in the screw never decreased to zero; this indicated that the channels were full. No adverse effects were experienced with the reduced melt temperature (8 [degrees] C lower), no unmelted material was observed in the extrudate, and no gel showers have occurred since the modification. A summary of the extrusion performance before and after the modification is shown in Table 1.
A new screw was installed into a single-stage extruder, 152.4 mm diameter and 26:1 L/D, that was running LDPE. At the startup of the line, the extruder operated at 64 rpm to produce the required rate of 272 kg/hr - a specific rate of 4.25 kg/(hr rpm). The temperature of the extrudate was 235 [degrees] C. As in Case 2, intermittent contamination of the product with crosslinked material was observed, causing production losses and customer complaints. At 64 rpm, the drag flow rate was estimated at 381 kg/hr, a specific rate of 5.95 kg/(hr rpm). Thus, the extruder was operating at about 70% of the drag flow rate.
A schematic of the screw is shown in Fig. 6. The lead length was 171.5 mm for the barrier section (both the main flight and barrier flight) and 152.4 mm for all other sections of the screw. The barrier flight started at the pushing side of the main flight and transitioned to its final position after about 1 diameter, creating widths of 48 and 96 mm for the melt conveying and solids conveying channels, respectively. The width of the feed section perpendicular to the flight was 131 mm before the entrance to the barrier section. Thus, the width and cross-sectional area for the solids channel of the barrier section were only 70% of that for the feed section. The barrier flight undercut was 0.8 mm, and the width of the barrier flight was 4.76 mm. The entrance to the barrier section for this screw was modified as in Case 2. The barrier flight was removed down to the root of the melt section for the first 1.7 diameters, and then was transitioned up to the barrier undercut over 0.7 diameter. The screw was reinstalled and studied for performance. With the modification, the screw produced the required 272 kg/hr at 51 rpm, a specific rate of 5.33 kg/(hr rpm) or 90% of the calculated drag flow rate, with an extrudate temperature of 224 [degrees] C. Intermittent contamination of the product with crosslinked material was eliminated. Thus, removing the restriction to the entrance of the barrier section increased the specific rate by 25%, reduced the extrudate temperature by 11 [degrees] C, and eliminated the intermittent discharge of crosslinked material in the extrudate. A summary of the extrusion performance before and after the modification is shown in Table 2.
Restrictions occurring at the entrance to barrier sections can be very complicated to analyze and eliminate. The severity of the restriction depends on the compaction behavior of the resin, thermal conductivity, coefficients of dynamic friction, process conditions, and screw geometry. For example, if a material such as HIPS is extruded at conditions favoring solids conveying, the material can be compacted easily to form an extremely hard bed before melting starts This bed will then be forced into the entry of the barrier section. If the cross-sectional area of the solid bed exiting the feed section is too large to fit into the solids channel of the barrier section, then a restriction in the flow can occur. As the solids are forced into the entrance, the hard solid bed will resist deformation, causing the rate to decrease. As the pressure upstream from the entrance to the barrier section increases, some of the solids will be melted until the cross section of the bed conforms to the cross section of the solids channel entrance.
By modification of the melt channel and barrier flight for 2 to 5 diameters at the entrance, the restriction can be eliminated or minimized. This permits some of the solids to flow into the melt channel and maintains the specified rate.
While numerous screws have been modified without decreasing the performance of the screw that was originally quoted by the manufacturer, modifying the entrance of the barrier section presents two potential concerns. First, allowing some solids to enter the melt channel may permit solids to exit with the extrudate. But because only a small fraction of the solids flows into the melt channel and it occurs early in the extruder, all solids should melt in the extruder. The second concern originates from one of the reasons that barrier flights are placed on screws - that is, to eliminate or minimize solid bed breakup. But because solid bed breakup generally occurs in the later stages of the melting process, removing the barrier flight at the entrance does not appear to increase the occurrence of solid bed breakup.
Barrier flighted sections for plasticating extruders are vital to commercial plastics processors The section permits extrusion at higher rates, reduces the occurrence of solid bed breakup, and improves the quality of the extrudate. The entrance to the barrier section, however, can cause a restriction in the flow path that can reduce the rate of the extruder. This restriction occurs as a result of an abrupt change in the cross section at the exit of the feed section and entrance to the barrier section, and it is accentuated by polymer properties. Modification to the barrier flight and melt channel at the entrance can eliminate this restriction without decreasing the performance advantages of the barrier section.
The authors wish to thank Jeff Zettle for his assistance in collecting the experimental data.
1. C. Maillefer, Swiss Patent 363,149 (1959).
2. B. Elbirli, J.T Lindt, S.R. Gottgetreu, and S.M. Baba, Polym. Eng. Sci., 23, 86 (1983).
3. H. Potente and H. Stenzel, Intern. Polym. Proc., VI, 126 (1991).
4. C. Rauwendaal, Polymer Extrusion, Hanser Publishers, New York (1986).
5. J.R. Powers, M.A. Spalding, and K.S. Hyun, SPE ANTEC Tech. Papers, 40,151 (1994).
6. K.S. Hyun and M.A. Spalding, Polym. Eng. Sci., 30, 571 (1990).
7. M.A. Spalding, D.E. Kirkpatrick, and K.S. Hyun, Polym Eng. Sci, 33, 423 (1993).
|Printer friendly Cite/link Email Feedback|
|Author:||Hyun, Kun S.; Spalding, Mark A.; Powers, Joseph R.|
|Date:||Apr 1, 1996|
|Previous Article:||Blow-film processing boosts polyethylene oxide applications.|
|Next Article:||Numerical simulation of viscoelastic polymer flow - effects of secondary flows on multilayer coextrusion.|