Effect of glass bead content and diameter on shrinkage and warpage of injection-molded PA6.
One of the most important properties of parts produced with injection molding--being the most widespread technique for processing polymers--is their dimensional accuracy. It can be easily seen that a deviation from the accepted tolerance range of the part can lead to the dismissal of the complete series produced. This attribute of polymeric materials to change their volume during--and after--processing is called shrinkage. According to Jansen et al. (1), there are three types of shrinkages: in-mold shrinkage, which happens during processing, as-molded shrinkage (or mold shrinkage taking place after the mold is opened), and postmold shrinkage (long time shrinkage caused by physical aging and recrystallization of the part) (2), (3). In most studies, according to the relevant standards (DIN 16901, ISO 294-4), shrinkage was measured at 24 or 48 h after production (1), (4-6). Some researchers adopted measurement points at 168 h (7), (8) or later to examine the post shrinkage of their products.
The uneven shrinkage and warpage can be attributed to numerous parameters from among which the most significant are the followings: differential cross-flow and parallel to flow shrinkages, shrinkages caused by uneven cooling, anisotropic material properties, and shrinkages caused by thermal stresses (9), (10). Numerous investigations have been carried out to examine the effecting factors of shrinkage in unfilled thermoplastics (1), (3), (5). Depending on the material used--amorphous or semi-crystalline--the values of flow directional and cross-flow shrinkages differ. In most cases, the studies pointed out that cross-flow shrinkage is larger than the flow directional shrinkage, an exception was Thomas' and Caffery's work (11) where this proportion turned over near the gate using polypropylene (PP). Shelesh-Nezad and Taghizadeh (6) learned that the shrinkage difference in flow and cross-flow directions of pure PP was 14% (1.64% shrinkage in the flow direction when compared with 1.41% in the cross-flow direction) and the addition of talc mineral filler (30 wt%) induced an isotropic shrinkage in the molded part. According to Shay et al. (12), significant difference concerning the two shrinkages was not found using amorphous materials. With semi-crystalline materials, Postawa and Koszkul (3) had observed an 8--10% difference between shrinkages of the two mentioned directions. Almost all research groups agreed--examining a given geometry--on that the holding pressure had the most significant effect on the shrinkage of the part (1), (5), (7). By increasing the holding pressure, the decrease of the shrinkage was observed. According to Postawa and Koszkul, the second most important processing parameter was injection temperature the change of which modified the melt viscosity (3). By increasing the temperature, a better pressure distribution along the flow path was noticed, causing a reduction in shrinkage. With longer holding times, shrinkage values were reduced until the gate freeze-off was reached.
In the case of reinforced or filled injection moldings, several studies dealt with different fiber reinforcements. These studies concluded that the reinforcement decreased the shrinkage in the flow direction in a higher degree than in the cross-flow direction causing the so-called warpage of the part. According to Jansen et al., a typical value for fiber reinforced shrinkage in the flow direction was 0.2%, whereas the reduction of cross-flow shrinkage was reduced negligible (1). DePolo and Baird (13) found that 6 mm long pregenerated thermotropic liquid crystalline polymer (TLCP) microfibrils (from 0 to 40 wt%) significantly increased (from 3% to 18%) the warpage of their PP parts while reducing the value of the flow directional shrinkage by 1% and the cross-flow shrinkage by 0.3%. With the application of fillers instead of fibers, a smaller degree of shrinkage reduction in the flow direction was experienced, but the difference between the flow directional and cross-flow shrinkages was significantly reduced. Mamat et al. (5)--using unfilled and 40 wt% calcium carbonate ([CaCO.sub.3])-filled PP--experienced a reduction (from 1.8% to 1.6%) in the cross-flow shrinkage and in the difference between the two shrinkages of the part.
Several studies dealt with glass beads, however, the majority of them examined the effect on the rheological (14-16) and mechanical properties (17-19) and some on the morphology developed (20-22). Both studies agreed on that the glass bead content raises the viscosity of the melt; moreover, the viscosity is in direct proportion with the glass head content. This effect is increased by decreasing the shear rate. In most cases with increasing the bead content, a rise in the modulus of elasticity, the decrease of the elongation at break, and the fall of tensile stress is experienced. The bead diameter does not have a significant effect on the mentioned mechanical properties of the material. Some investigations were carried out to examine the effect of solid glass beads on the shrinkage of injection molded thermoplastics (18). But these studies were dealing mostly with bead content, leaving the effect of bead size out from the examination range. Huang et al. (18) for example examined the effect of different bead content (0, 5, 20, and 25 wt%) on the shrinkage of micro-injection-molded products and seen that it was inverse proportional to the shrinkage of their parts; the application of 25 wt% glass bead reduced the shrinkage of the part from an initial value of 2.8% to 1.3%.
In spite of the numerous studies dealing with glass beads, relatively few examined the effect of fillers on the shrinkage properties of the materials and none of them examined the effect of bead size. According to the previous studies in the field of segregation (23-25), an effect of the bead size and content on the shrinkage of the part was expected. In this way, the objective of this article is to present a simple technique to evaluate the shrinkage and warpage of injection-molded products. For this purpose, solid glass bead-filled PA6 with varying bead diameters and contents were used.
PA6 matrix filled with solid glass beads of different diameter (11, 85, 156, 203 [micro]m) and content (10, 20, 30, 40 wt%) were examined in this study. The matrix (Lanxess, Durethan B30S) and the beads (Potters Industries Inc., Spheriglass[R] with coupling agent) were mixed on a Brabender Plasticorder 814402 type twin-screw extruder with constant screw revolutions per minute. A Brabender pelletizer was used to produce pellets from the extrudate. The glass bead contents of each material was verified by determining the ash content where the maximum deviation was 3.7 wt% from the nominal values. From these materials, 80 X 80 X 2 mm specimens (Fig. 1) were injection molded on an Arburg 320C 600-250 injection molding machine with the following injection molding parameters (Table 1).
[FIGURE 1 OMITTED]
TABLE 1. The injection molding parameters. Melt temperature Mold temperature Volumetric flow rate 280[degrees]C 80[degrees]C 50 [cm.sup.3]/s
During the injection molding, a uniform filling pattern was adopted. Five measurement locations were marked on each specimen where the appropriate dimensions (Fig. 1) were measured with a Mitutoyo digital caliper--with two decimal digit precision--at 1 min, 1 h, 4 h, 24 h, and 168 h after demolding. To prevent moisture absorption, the specimens were stored under room temperature and were subjected to predried silica gel between the measurements. From the measured dimensions of the specimens and the nominal dimensions of the cavity--measured at room temperature--the linear shrinkages were determined according to (1):
[S.sub.i] = [[[L.sub.M] * [L.sub.i]]/[L.sub.M]] * 100[%], (1)
where [S.sub.i] [%] is the shrinkage in percentage, [L.sub.M] [mm] is the dimension of the cavity, and [L.sub.i] [mm] is the relating measured dimension of the specimen.
The average flow directional shrinkage at the edge of the specimen is as follows:
[S.sub.LE] = [[[S.sub.L1] + [S.sub.L2]]/2][%], (2)
where [S.sub.LE] [%] is the average flow directional shrinkage at the edge of the specimen, [S.sub.L1] [%] and [S.sub.L2] [%] are the flow directional shrinkages at the two sides in flow direction of the specimen.
The mean cross-flow direction and flow directional shrinkages are as follows:
[S.sub.W] = [[[S.sub.WG] + [S.sub.WE]]/2][%], (3)
[S.sub.L] = [[[S.sub.L0] + [S.sub.LE]]/2][%], (4)
where [S.sub.W] [%] is the average cross-flow direction shrinkage, [S.sub.L] [%] is the average flow directional shrinkage, [S.sub.WG] [%] is the cross-flow direction shrinkage near the gate, [S.sub.WE] [%] is the cross-flow direction shrinkage at the end of the flow, [S.sub.L0] [%] is the flow directional shrinkage in the center in flow direction of the specimen, and [S.sub.LE] [%] is the mean flow directional shrinkage at the edge of the specimen.
According to Kovacs and Tabi, three deformation factors were used to describe the warpage caused by inhomogeneous shrinkage (26). These deformation factors were defined using the measured shrinkages of the specimens shown on Fig. 1:
Deformation factor defined by the cross-flow direction and the flow directional shrinkages (Figs. 1 and 2a):
[FIGURE 2 OMITTED]
DFWL = [[S.sub.W]/[S.sub.L]][-], (5)
Deformation factor defined by the pressure drop caused shrinkage differences (Figs. 1 and 2b):
DFP = [[S.sub.WG]/[S.sub.WE]][-], (6)
Deformation factor defined by the flow directional shrinkage differences (Figs. 1 and 2c):
DFL = [[S.sub.L0]/[S.sub.LE]][-], (7)
The ideal value of these factors is 1, when the shrinkages in the numerator and in the denominator are equal.
RESULTS AND DISCUSSION
As a result of the measurements, it can be seen that regardless of bead size and content the curves for different shrinkages preserved their characteristics. The raising bead content decreased the shrinkages. Furthermore, the differences between the shrinkages measured in different directions became smaller, moving the corresponding deformation factor values closer to 1. At low bead content levels, the cross-flow direction shrinkage at the end of the flow ([S.sub.WE]) had the highest values, probably because of the pressure drop along the melt flow. At high bead content levels and large bead sizes, the two cross-flow direction shrinkages change places and the cross-flow direction shrinkage near the gate becomes higher. This phenomenon can be explained by segregation (25), (27), (28). With higher filler content and larger filler size, the local filler content rises at the end of the flow path and decreases near the gates, thereby changing the viscosity of the melt too (Fig. 3).
[FIGURE 3 OMITTED]
By creating an inhomogeneous viscosity along the flow path, the pressure distribution was also changed. The bead content acts in the opposite direction as the pressure drop and decrease the shrinkage at the end of the flow path but decrease it with a smaller degree near the gates. It can be seen that at larger bead sizes and higher bead contents the values for the different kinds of shrinkages were lower than at smaller sizes and contents (Figs. 4 and 5).
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
It can be seen that the bead size had no effect on the shrinkage near the gate but affected the shrinkage at the end of the flow. It can be explained by the change in the flow ability that affects the pressure distribution of the material (Figs. 6 and 7).
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
Time Dependent Shrinkage
Flow directional shrinkages have shown a large deviation in time, as a function of glass bead content (Fig. 8).
[FIGURE 8 OMITTED]
Based on these measurements, a simple formula has been proposed to describe the relation between the changes of flow directional shrinkage in time, glass bead size, and content:
[S.sub.L](t) = [A.sub.SL] * ln(t) + ([S.sub.Lt0] - [x/[[C.sub.SL] * db - [C.sub.SL0]]])[%] (8)
where [S.sub.L](t) [%] is the flow directional shrinkage in the function of time, [A.sub.SL] [-] and [C.sub.SL] [-] are material dependent parameters, [C.sub.[SL0]] [-] is a constant for the unfilled material, x [wt%] is the glass bead content, [d.sub.b] [[mu]m] is the glass bead diameter, and [S.sub.Lt0] [%] is the flow directional shrinkage of the unfilled material measured 1 h after the injection molding.
The trends show a good relation with the measurements. It can be seen on the figure that the application of larger beads decreased the shrinkage values consequently further increasing the distance between the curves with higher glass bead content (Fig. 9).
[FIGURE 9 OMITTED]
The relation's (8) validity was proven only in the range of investigation, meaning a glass bead content range of 10-40 wt%, a duration of 0.016-168 h and a glass bead size range of 11-203 [micro]m. Table 2 contains the assessed values of the parameters featured in Eq. 8. By statistically evaluating the suitability of the data, the value of the correlation coefficient was ([R.sup.2]) 0.97.
TABLE 2. Parameters of glass bead filled PA6 (Durethan B30S). [SL.sub.10] [A.sub.SL] [C.sub.SL] [C.sub.SL0] 1.3 0.0075 -0.0765 96.2
Regarding to injection molded products, it is more important that uniform shrinkage is achieved, namely the deformation of the product is reduced. The previously introduced deformation factors are appropriate to characterize these warpage properties. The deformation factor defined by the flow directional shrinkage differences indicates the changes of the shrinkages along the How path. It was concluded that neither the glass bead content nor the size has any effect on the value of this factor, so the shrinkages ([L.sub.0], [L.sub.1], and [L.sub.2]) reacted to a similar degree to the changes of the main parameters (Fig. 10).
[FIGURE 10 OMITTED]
The deformation factor, defined by the cross-flow direction and the flow directional shrinkages (DFWL, Eq. 5) characterizes the deformations derived from the melt flow. The value of this factor, which can easily be raised by the application of glass fibers, can be considerably lowered by the use of glass beads. It can be seen on the following figure that with raising glass bead content and decreasing bead size the factor's value can be moved to the direction of its ideal value (Fig. 11).
[FIGURE 11 OMITTED]
The pressure dependence of shrinkage is a minor research area, although it has an important role in the deformation of injection molded products. The previously introduced deformation factor defined by the pressure drop caused shrinkage differences is the index-number for characterizing this attribute. The increase of DFP was experienced with the raising of bead content and diameter (Fig. 12). This relation was described with the following equation:
[FIGURE 12 OMITTED]
DFP = ([m.sub.b] * [d.sub.b] + [m.sub.0]) * x + [DFP.sub.0][-] (9)
where DFP [-] is the deformation factor defined by the pressure drop caused shrinkage differences, [m.sub.b] [-] and [m.sub.0] [-] are material dependent parameters, x [wt%] is the glass bead content, [d.sub.b] [[mu]m] is the glass bead diameter, and [DFP.sub.0] [-] is the deformation factor defined by the pressure drop caused shrinkage of the unfilled system.
The relation's (9) validity was proven only in the range of investigation meaning a glass bead content range of 10-40 wt%, a duration of 0.016-168 h and a glass bead size range of 11-203 [micro]m. Table 3 contains the assessed values of the parameters featured in Eq. 9. By statistically evaluating the suitability of the data, the value of the correlation coefficient was ([R.sup.2]) 0.94.
TABLE 3. Parameters of glass bead filled PA6 (Durethan B30S). [DFP.sub.0] [m.sub.b] [m.sub.0] 0.85 2,53 X [10.sup.-5] 2.79 X [10.sup.-3]
It can be seen that DFP can reach its optimum at a given bead content in the function of glass bead diameter (Fig. 13).
[FIGURE 13 OMITTED]
The ideal glass bead content resulting in the optimal value of deformation can be determined by a function incorporating applied bead diameter and a deformation factor, defined by the pressure drop caused shrinkage differences (DFP = 1) according to Eq. 10:
[x.sub.id] = [[1 - [DFP.sub.0]]/[[m.sub.b] * [d.sub.b] + [m.sub.0]]][wt%], (10)
where [x.sub.id] [wt%] is the ideal glass bead content, [m.sub.b] [-] and [m.sub.0] [-] are material dependent parameters, [d.sub.b] [[mu]m] is the glass bead diameter, and [DFP.sub.0] [-] is the deformation factor defined by the shrinkage of the unfilled material caused by the pressure drop.
In this article, the shrinkage and warpage properties of injection molded glass bead filled PA6 systems has been examined. The effect of altering glass bead size and content were investigated. Raising the glass bead content and diameter resulted in a reduction in the levels and variations of shrinkage taking place. It was concluded that the bead size did not have any effect on the near-gate cross-flow direction shrinkage but had a remarkable effect on the shrinkage at the end of the flow. This phenomenon is assumed to be caused by segregation and the inhomogeneous viscosity distribution caused by segregation. The relation between the time dependence of flow directional shrinkage and bead size and diameter was also described.
By determining the deformation factors derived from the measured shrinkages, it was pointed out that the deformation factor (DFL, Eq. 7) defined by the flow directional shrinkage differences was not affected by bead content or size (Fig. 10). It was demonstrated that with raising the glass bead content and decreasing bead size the DFWL factor's value could be moved to the direction of its ideal value (Fig. 11). The increase of DFP was experienced with the rise of both the glass bead content and size (Fig. 12). An equation was proposed to determine the ideal bead content in the function of applied bead diameter for reaching the optimal value of deformation factor defined by the pressure drop caused shrinkage differences.
The authors would like to thank Arburg Ltd. for the injection molding machine.
NOMENCLATURE [A.sub.SL] [--] material dependent parameter Ca[CO.sub.3] calcium carbonate [C.sub.SL] [--] material dependent parameter [C.sub.SL0] [--] constant for the unfilled material [d.sub.b] [[mu]m] glass bead diameter DFL [--] deformation factor defined by the flow directional shrinkage differences DFP [--] deformation factor defined by the pressure drop caused shrinkage differences [DFP.sub.0] [--] deformation factor defined by the pressure drop caused shrinkage of the unfilled system DFWL [--] deformation factor defined by the crossflow direction and the flow directional shrinkages GB glass bead [L.sub.i] [mm] measured dimension of the specimen [L.sub.M] [mm] dimension of the cavity [m.sub.b] [--] material dependent parameter [m.sub.0] [--] material dependent parameter PA6 polyamide 6 PP polypropylene [S.sub.i] [%] linear shrinkage of the part [S.sub.L] [%] average flow directional shrinkage [S.sub.LE] [%] average flow directional shrinkage at the edge of the specimen [S.sub.L0] [%] flow directional shrinkage in the center of the specimen [S.sub.L1],[S.sub.L2] [%] flow directional shrinkage at the two sides of the specimen [S.sub.[Lto]] [%] flow directional shrinkage of the unfilled material measured 1 hour after the injection molding [S.sub.L](t) [%] flow directional shrinkage in the function of time [S.sub.W] [%] average cross-flow direction shrinkage [S.sub.WE] [%] cross-flow direction shrinkage at the end of the flow [S.sub.WG] [%] cross-flow direction shrinkage near the gate TLCP thermotropic liquid crystalline polymer x [wt%] glass bead content [x.sub.id] [wt%] ideal glass bead content
(1.) K.M.B, Jansen, D.J.V. Dijk, and M.H. Husselman, Polym. Eng. Sci., 38, 838 (1998).
(2.) K.M.B. Jansen, G. Titomanilo, and R. Pantani, Int. Polym. Proc., 12, 4 (1997).
(3.) P. Postawa and J. Koszkul, J. Mater. Process. Tech., 162, 109 (2005).
(4.) B.H. Min, J. Mater. Process. Tech., 136, 1 (2003).
(5.) A. Mamat, T.F. Trochu, and B. Sanschagrin, Polym. Eng. Sci., 35, 1511 (1995).
(6.) K. Shelesh-Nezhad and A. Taghizadeh, Polym. Eng. Sci., 47, 2124(2007).
(7.) S.J. Liao, D.Y. Chang, H.J. Chen, L.S. Tsou, J.R. Ho, H.T. Yau, W.H. Hsieh, J.T. Wang, and Y.C. Su, Polym. Eng. Sci., 44, 917 (2004).
(8.) J.G. Kovacs, Ph.D. Thesis, Design and Simulation of Injection Molded Products, Budapest University of Technology and Economics (BME), Budapest (2007).
(9.) R.A. Malloy, Plastic Part Design for Injection Molding, Carl Hanser Verlag, Munich (1994).
(10.) J.M. Fischer, Handbook of Molded Part Shrinkage and Warpage, Plastics Design Library, Norwich (2003).
(11.) R. Thomas and N. Mccaffery, SPE ANTEC., 89, 371 (1989).
(12.) R.M. Shay, P.H. Foss, and C.C. Mentzer, SPE RETEC., 1, 496 (1996).
(13.) W.S. Depolo and D.G. Baird, Polym. Compos., 27, 239 (2006).
(14.) J.-Z. Liang and R.K.Y. Li, J. Appl. Polym. Sci., 73, 1451 (1999).
(15.) J.-Z. Liang, R.K.Y. Li, and S.C. Tjong, Polym. Test., 19, 213 (2000).
(16.) J.-Z. Liang, Macromol. Mater. Eng., 286, 714 (2001).
(17.) J.-Z. Liang, Macromol. Mater. Eng., 287, 588 (2002).
(18.) C.K. Huang, S.W. Chen, and C.T. Yang, Polym. Eng. Sci., 45, 1471 (2005).
(19.) S. Hashemi, Express Polym. Lett., 2, 474 (2008).
(20.) R.K.Y. Li, J.-Z. Liang, and S.C. Tjong, J. Mater. Process Tech., 79, 59 (1998).
(21.) J.-Z. Liang, R.K.Y. Li, and S.C. Tjong, Polym. Test., 16, 529 (1998).
(22.) H. Unal, Mater. Des., 25, 483 (2004).
(23.) S.O. Ogadhoh and T.D. Papathanasiou, Compos. Part A-Appl. S., 27, 57 (1996).
(24.) J.-Z. Liang and R.K.Y. Li, J. Reinf. Plast. Comp., 20, 630 (2001).
(25.) S.C. Jana, Polym. Eng. Sci., 43, 570 (2003).
(26.) J.G. Kovacs and T. Tabi, Proceedings of the Fifth Conference on Mechanical Engineering, Budapest, 5, 5 (2006).
(27.) E. Lafranche, P. Krawczak, J.-P. Ciolczyk, and J. Maugey, Adv. Polym. Tech., 24, 114 (2005).
(28.) E. Lafranche, P. Krawczak, J.-P. Ciolczyk, and J. Maugey, Express Polym. Lett., 1, 456 (2007).
J.G. Kovacs, B. Solymossy
Department of Polymer Engineering, Budapest University of Technology and Economics, H-1111 Budapest, Muegyetem rkp. 3., Hungary
Correspondence to: J.G. Kovacs: e-mail: email@example.com
Contract grant sponsor: The Janos Bolyai Research Scholarship of the Hungarian Academy of Sciences and the Hungarian-French Intergovernmental Science and Technology Cooperation Programme; contract grant number: TeT FR-5/2007; contract grant sponsor: The Jedlik Anyos Programme of Ministry of Economy and Transport of Hungary: contract grant, number: ENGPL_07.
Published online in Wiley InterScience (www.interscience.wiley.com).