Printer Friendly

Microcellular injection-compression molding (MICM): a novel technology for effectively improving cellular structure of polystyrene foams.


Foamed polymer parts produced by microcellular injection molding (MIM) technology are increasingly used in many industries, such as auto, home appliances, and packaging, due to its advantages including light part weight, material saving, shorter molding cycle time, lowered part shrinkage, and superior part dimensional stability (1-4). Despite a number of benefits with the MIM, it is by no means a panacea. Achieving sound parts with the microcellular structure is highly challenging due to intrinsic characteristics of the injection molding process (5). Moreover, there exist differences of shearing deformation and temperature in both thickness and mold filling directions during injection molding (6). This usually results in a nonuniform cellular structure within the MIM parts, which may bring the decrease in their mechanical properties (17), (8). For this reason, some researchers have focused on improving the cellular structure by varying the processing parameters (9-12) such as the dosage of supercritical fluid (SCF), shot size, injection speed, and melt temperature, by adding the fillers (nanoclay or fiber) into the matrix materials (13-15), by using the polymer blends (16), (17), or by combining the MIM with other technologies (18), (19) such as precision mold opening and gas counter pressure.

The purpose of this work is to find a novel and feasible method for improving the cellular structure of the MIM parts. Injection-compression molding (ICM), which can be realized on most injection molding machines, has received increased attention because of its advantages over conventional injection molding (CIM), including lowered molding pressure, reduced part residual stress, decreased part density variation, and increased part dimensional accuracy (20). In the ICM process, the melt is injected into a partially opened mold, i.e., an enlarged cavity, and then compressed to the final thickness using a clamping mechanism. The introduction of the mold compression makes the melt pressure distribution inside the mold cavity quite uniform and the melt flow characteristics different from that in CIM, which may induce different structure and properties of molded parts. For instance, a five-layer structure of fiber orientation is found through the thickness in the ICM short-fiber-reinforced polypropylene parts (21), which is different from that in the CIM fiber-reinforced parts. The backflow during ICM can redistribute the melt in the mold cavity and more importantly distribute the final part thickness (22). Moreover, compared to micro-injection molding, micro-ICM can give superior micro-feature replication performance (23-25). Considering the aforementioned effects brought by ICM, the mold compression was tried to introduce into the MIM process as a novel technology (MICM) to prepare foamed samples in this work. For the sake of contrastive analysis, samples were also prepared under the corresponding molding conditions using the standard MIM technology (without compression). The cellular structures of both MICM and MIM samples were analyzed. Finally, a development mechanism of the cellular structure is proposed for the MICM process.


Equipment and Material

The experimental equipment was consisted of SCF supply system (S 11-TR-35A, Trexel, USA), an 80-ton injection-molding machine (KM8OSP180CX, Krauss-Maffei, Germany), an injection-compression mold, and an acquisition system for capturing the video images in the mold cavity and monitoring the cavity pressure during molding.

The aforementioned mold and acquisition system are schematically shown in Fig. 1. A tempering glass block, LED light source, and a mirror were mounted in the fixed mold half to build a visualized window (as shown in Fig. la) for reflecting the whole cavity image. A high speed video camera (A6221, Basler, Germany) was employed to capture the molding process, the video images of which were collected with an USB data acquisition card and saved in a personal computer (PC). For preparing the foamed samples and monitoring the cavity pressure simultaneously, the glass block was replaced by a metal block, and a piezoelectric pressure sensor (6190BA, Kistler, Japan) was located at 20 mm from the gate (as shown in Fig. lb). For each molding cycle investigated, online pressure data were collected from the sensor using a data acquisition system (DataFlow, Kistler, Japan) and also stored in the PC. For both capturing the video images and monitoring the cavity pressure, the time when two halves of the mold were just contacted was denoted as time "zero." For the sake of contrastive analysis, experiments were also carried out under corresponding molding conditions using the standard MIM technology (without compression).

The polystyrene (PS), grade PG33 (Zhenjiang Chimei Chemical) with a melt index of 7.9 g/10 min (200[degrees]C and 5 kg), was used in this work. Industrial nitrogen ([N.sub.2]) gas with a purity of over 99% was used as a physical foaming agent.

Sample Preparation and Characterization

The PS was dried for 4 h at 80[degrees]C under vacuum to remove moisture before molding. The molding conditions used in this work are listed in Table 1. The last three compression parameters were set only for the MICM. The rectangular foamed samples (as shown in Fig. 2) with the dimension of 120 X 20 X h m[m.sup.3] were molded using both MICM and MIM. Here, h represents the sample thickness. For the MICM samples, the h was changed by adjusting the compression force in the present work. The samples with the thicknesses of 5, 4, and 3 mm were molded under compression forces of 60, 62, and 64 kN, respectively. For the MIM samples, the shim block (as shown in Fig. 1) was adjusted to the appropriate thickness to change the h.

TABLE 1. Molding conditions used in this work.

Parameter                                      Value

Shot size                            11.7 c[m.sup.3]
Screw speed                        80 r [min.sup.-1]
Injection speed                    150 mm [s.sup.-1]
Melt temperature                       225[degrees]C
Packing pressure                               0 MPa
Cooling time                                    20 s
Back pressure                                 19 MPa
Mold temperature                        20[degrees]C
Sc-[N.sub.2] delivery pressure              20.4 MPa
Weight percentage of Sc-[N.sub.2]               0.7%
Sc-[N.sub.2] dosage time                          3s
Compression speed                    2 mm [s.sup.-1]
Compression distance                            5 mm
Compression force                      60, 62, 64 kN

After a steady molding process was reached, the foamed rectangular samples were collected. The foamed samples were immersed in liquid nitrogen for 15 min and fractured to produce clean and intact surfaces with minimum plastic deformation. The fractured surfaces were coated with gold by using a sputter coater and then examined by utilizing a scanning electron microscope (SEM, Nova NanoSEM 430, FEI) at an acceleration voltage of 8 kV. Three positions (P1, P2, and P3, as illustrated in Fig. 2) along the sample axis were chosen for investigating the cellular structure. The SEM observation area (A) is shown in Fig. 2. The SEM micrographs were analyzed using Image-Pro Plus software to obtain the single cell diameter and number for ellipsoidal cells, and the aspect ratio for striation-shaped cells. The cell density ([N.sub.F], cells/c[m.sup.3]) for ellipsoidal cells was calculated with the following equation (26):

[N.sub.F] = [(n/A).sup.3/2] (1)

where n is the number of cells observed in a SEM micrograph with an area of A in c[m.sup.2]. The mean cell diameter (d, [mu]m) and the standard deviation of cell diameter distribution ([sigma], [mu]m) were obtained through the following method. First, the distribution histogram of cell diameter was drawn. Then a distributed curve was fit on the basis of Gaussian equation, because a typical cell diameter distribution usually obeys to a Gaussian distribution (27), (28):


where F is the relative frequency, i.e., the ratio of the number of cells in a certain diameter range to the total number of counted cells; d is the single cell diameter calculated using the aforementioned software; [F.sub.o] and B are Gaussian parameters.

The specimen (B, as illustrated in Fig. 2) with a dimension of 50 x 10 x 4 [mm.sup.3] for dynamic mechanical thermal analysis (DMTA) testing was cut from the foamed samples. The testing was carried out using a dynamical rheometer (Bohlin Gemini 200, Malvem Instruments, UK) in an oscillatory mode with a "solids fixtures" geometry (Rectangular torsion). The testing temperature ranged from 40 to 130[degrees]C at a heating rate of 2[degrees]C [min.sup.-1]. The used frequency and strain were 1 Hz and 0.001%, respectively.


Images in Mold Cavity and Cavity Pressure Profiles During MICM and MIM

Figure 3 shows the video images in the mold cavity taken at a sequent time during molding of both MICM and MIM samples with a thickness of 4 mm. As can be seen in Fig. 3a, the polymer melt/Sc-[N.sub.2] mixture is injected into the mold cavity from 2.0 to 2.2 s during MICM. Subsequent to a short delay, the melt is compressed from 2.6 to 3.5 s, followed by the cooling stage. The melt appears bright gray in the injection stage. Then, its color gradually changes to light gray at the end of the compression stage, and finally almost does not change any more in the cooling stage. In terms of the color change in the compression stage, it can be speculated that the mold compression can further change the cellular structure formed in the injection stage, which will be analyzed later in detail. As demonstrated in Fig. 3b, the melt/ Sc-[N.sub.2] mixture fills the cavity within only 0.2 s and almost maintains gray in the entire process of the MIM.

Figure 4 illustrates the collected cavity pressure profiles at measured position during molding of the MICM and MIM samples with thicknesses of 3, 4, and 5 mm. As can be clearly observed, the cavity pressure reaches a maximum at the end of the compression and injection stages for the MICM and MIM processes, respectively. It is worth noting that compared with the MIM process, the MICM process can lower the maximum cavity pressure by about 18.6, 29.3, and 55.6% for 3, 4, and 5 mm-thick samples, respectively. So the clamping force can be further reduced when using the MICM, especially for molding thicker foamed parts.

Typical Cellular Structure of MICM and MIM Samples

Both 4-mm-thick MICM and MIM samples were selected to investigate their typical cellular structure. The SEM micrographs across half the thickness at three positions along the sample axis (P1, P2, and P3, as shown in Fig. 2) are displayed in the middle row in Figs. 5 and 6. As can be clearly seen, two types of cells, i.e., the irregular striation-shaped and ellipsoidal cells, exist on the fractured surfaces of the foamed samples. Correspondingly, the outer and inner zones are divided across the foamed samples. Higher shear rates occur towards the mold wall, which stretches the cells in the outer zone. The thicknesses of the two zones are also shown in Figs. 5 and 6. As can be observed, the outer zone thickness increases with the distance from the sprue in a nearly linear manner for both MICM and MIM samples. Interestingly, the outer zone thicknesses at all three positions in the MICM sample is decreased by about 25% compared with those in the MIM sample. This may be resulted from lower shear stresses imposing on the melt during MICM. Young (29) reported that the shear stress in the ICM is only half of that in the CIM.

To more clearly observe the cellular structures in the outer and inner zones, the magnified micrographs taken in the selected areas of the two zones are shown in Figs. 5 and 6. As for the outer zone, mean aspect ratio, which is listed in Table 2, was used to characterize the deformed degree of cells. As illustrated in Table 2 and the magnified micrographs in Figs. 5 and 6, the outer zone of the MICM sample exhibit more obviously deformed cells and a little more uniform cellular shape and size distribution at all three positions compared with that of the MIM sample.

TABLE 2. Mean aspect ratio of cells in outer zone of
4-mm-thick foamed samples.

                         MICM               MIM

Cell parameter       P1    P2    P3    P1    P2    P3

Mean aspect ratio  2.67  3.58  3.78  2.30  2.82  2.67

In the inner zone, in general, more uniform cellular structure appears at all three positions, especially at P1, of the MICM sample than the MIM one. It is clearly seen from Fig. 6a that some large and elongated cells exist in the inner zone of P1 in the MIM sample. This may be attributed to the fact that no packing pressure is set in this work. So higher measured cavity pressure at the end of melt filling (shown in Fig. 4b) causes the melt near the sprue to flow back into the runner system. Whereas the measured pressure at the end of melt filling during MICM is very low, as shown in Fig. 4a. This may be the reason why using the MICM can eliminate the aforementioned phenomenon, as shown in Fig. 5a. The mean cell diameter and cell density, which are shown in Fig. 7, are used to characterize the cellular structure in the inner zone. As can be clearly seen, the mean cell diameter increases in a nearly linear manner and the cell density decreases obviously with the distance from the sprue for the MICM sample. This is attributed to the fact that in relatively slow melt filling stage of the MICM, the cells in the melt front area undergo increased growth due to longer growing time (30) and lower pressure than those near the sprue. Whereas the cells at P2 and P3 of the MIM sample show similar cell size and cell density, which results from similar cell growth time in the fast melt filling stage of the MIM. Figure 8 compares the standard deviation of the cell diameter distribution in the inner zone of both MICM and MIM samples. As can be seen, more uniformly distributed cells appear in the MICM sample.]

It is well known that the cellular structure is a key factor affecting the mechanical properties of the microcellular parts. So the DMTA testing was performed on both MICM and MIM samples. Here, the storage modulus curves are given in Fig. 9 for 4-mm-thick samples. As can be clearly seen, the MICM sample exhibits higher storage modulus in the temperatures lower than about 95[degrees]C, which is in the glassy state of the PS. This is resulted from more uniform cellular structure in both outer and inner zones of the MICM sample.

Development Mechanism of Cellular Structure in Compression Stage of MICM

As mentioned above, the mold compression plays an important role in the development of final cellular structure in the MICM samples. However, it is difficult to acquire the cellular structure at different times in the compression stage during MICM. Adjusting the compression force can change the thickness of the MICM samples. Therefore, additional MICM samples with the thicknesses of 5 and 3 mm were molded at the compression forces of 60 and 64 kN, respectively. It is supposed that the cellular structure in the 5, 4, and 3-mm-thick samples is developed sequentially. Consequently, based on the cellular structure in the samples with the three different thicknesses, a cellular development mechanism in the compression stage can be proposed.

Here, the cellular structure at P2 in the 5 and 3-mm-thick MICM samples was given and analyzed. Figure 10 displays the SEM micrographs across half the thickness at P2 in both samples. In Fig. 10, the outer and inner zones are also divided in a similar manner to that in Fig. 5. Combining Fig. 10 with Fig. 5b demonstrates that by decreasing the sample thickness from 5 to 4 and finally to 3 mm, the cells in the outer zone change from ellipsoidal to elongated shape, and almost disappear finally. The mean cell size and cell density in the inner zone of the three samples are shown in Fig. 11. As can be observed, the mean cell diameter decreases and the cell density increases significantly with the decrease in the sample thickness. From the forgoing, a development mechanism of the cellular structure in the compression stage during MICM is proposed, as schematically shown in Fig. 12. As the mold compression progresses, the area with almost no cells increases, and the cell size in the inner zone becomes smaller and more uniform. The former can be briefly explained by the classical nucleation theory. As shown in Fig. 4a, mold compression results in an increase in the measured cavity pressure. This raises the polymer melt pressure around the cells and so increases the critical radius of the cell. The cells with radii smaller than this critical value tend to collapse and the gases inside the cells diffuse back into the melt and dissolve in it (31), which results in higher concentrations of the dissolved gases in the melt. Moreover, the decrease of the number of cells can be used to explain the aforementioned color change during MICM (as shown in Fig. 3a). The reduced cell number in thickness direction makes the light from LED pass through the sample easily (32).

In the purpose of further interpreting the aforementioned cellular development mechanism, MuCell[R] Analysis Module in Autodesk Moldlow Insight 2010, a commercial finite-element (FE) software, was implemented to simulate the pressures inside the cells corresponding to the time at the end of the compression during MICM. Some parameters used in the simulation are listed in Table 3, and other parameters (injection speed, melt temperature, packing pressure, and mold temperature) were the same as those used in the experiments (shown in Table 1). For the 5, 4, and 3-mm-thick samples, the simulated pressures inside the cells at several positions in thickness direction are illustrated in Fig. 13. Each data point on the plots is an average of six element results acquired near position P2. As can be observed, the pressures inside the cells, especially those near the outer surface, increase along with the compression. The concentration of the dissolved gases in the melt c can be obtained by Henry's law (33):

TABLE 3. Some parameters used in simulation.

                                                  Weight reduction

Initial bubble        Nucleation    Initial gas  3 mm   4 mm   5 mm
radius                   density  concentration

1.0 x [10.sup.   2 X [10.sup.11]           0.7%  3.0%  27.1%   41.6%
-6] m           cells [m.sup.-3]

c(t) = [k.sub.h][P.sub.cell] (t) (3)

where [k.sub.h] is the Henry' law constant and Pcell--is the pressure inside the cell. Equation. 3 shows that c is proportional to [P.sub.cell]. Therefore, rapidly increasing [P.sub.cell] toward the mold wall leads to higher concentrations of the dissolved gases in the melt there, which is agreed with the aforementioned analyzed result.

As previously stated, the mold compression has less effect on the cell collapse in the inner zone than in the outer zone of the sample, but it can obviously influence the cell size and density in the former. This may be explained by the change of the gas density inside the cell pg, which is calculated using the ideal gas law (33):

[P.sub.cell](t) = 1000[R.sub.g][T.sub.g][p.sub.g](t)/[M.sub.w] (4)

where [R.sub.g] is the universal gas constant, [T.sub.g] is the gas temperature, and [M.sub.w] is the molecular weight of the gas. Both [R.sub.g] and [M.sub.w] are constants for a given gas. The melt temperature and so the [T.sub.g] decrease with the time prolonging along with the compression. Hence, considering the aforementioned change of the simulated [P.sub.cell], it can be deduced from [E.sub.q]. 4 that the pg increases with the compression going on, which means that large cells may be compressed into small cells. Moreover, increased melt strength, which is resulted from decreasing the melt temperature and increasing the melt pressure, may keep the sizes of these small cells. To sum up, with the compression progressing, the collapse of some small cells and especially the decrease of the number of large cells result in narrower cell size distribution and more uniform cellular structure in the inner zone of the MICM sample.

From the forgoing, the mold compression has a significant effect on the foaming area and the shape, numbers and sizes of the cells across the part thickness, and so MICM can serve as a new technology to effectively improve the cellular structure of foamed parts.


In the present work, microcellular foamed PS samples with three different thicknesses were prepared using both MICM and MIM. Melt filling process and final cellular structure were investigated. It was found that the melt within the mold cavity shows a color change from bright gray to light gray due to the transformation of the cellular structure induced by mold compression during MICM. The cavity pressure curves measured during MICM exhibit two peaks and the maximum one is always lower than the single peak measured during MIM. Taking the 4-mm-thick sample as example, compared with MIM sample, the MICM sample possesses an improved cellular structure, which can be summarized as follows. Outer zone thicknesses at three different positions along the sample axis are decreased by about 25%. In the outer zone, the cells show a little better uniformity in respect of their shape and size distribution; in the inner zone, more uniform cellular structure with smaller sizes at three positions, especially near the sprue, appears. Finally, a cellular development mechanism in the compression stage during MICM was proposed. With the compression going on, the area with almost no cells across the sample thickness increases; the collapse of some small cells and especially the decrease of the number of large cells result in narrower cell size distribution and more uniform cellular structure in the inner zone of the MICM samples. This work may give insight into better control of the cellular structure of foamed parts.

Correspondence to: Han-Xiong Huang; e-mail: Contract grant sponsor: National Natural Science Foundation of China; contract grant number: 11172105; contract grant sponsor: Guangdong Provincial Natural Science Foundation; contract grant number: S2011010002085.

DOI 10.1002/pen.23566

Published online in Wiley Online Library (

[c]2013 Society of Plastics Engineers


(1.) A. Kramschuster, R. Cavitt, D. Ermer, Z.B. Chen, and L.S. Tumg, Polym. Eng. Sci., 45, 1408 (2005).

(2.) J.D. Yoon, S.W. Cha, T.H. Chong, and W.H. Young, Polym. Plast. Technol. Eng., 46, 815 (2007).

(3.) R. Zafar, K.S. Lee, H.B. Kim, B.J. icon, and S.W. Cha, Polym. Plast. Technol. Eng., 47, 1187 (2008).

(4.) A.K. Bledzki, H. Kirschling, M. Rohleder, and A. Chare, J. Cell. Plast., 48, 301 (2012).

(5.) A.H. Behravesh and M. Rajabpour, Cell. Polym., 25, 85 (2006).

(6.) J. Xu, Microcellular Injection Molding, Wiley, New Jersey, 62 (2010).

(7.) W. Gong, J.C. Gao, M. Jiang, L. He, J. Yu, and J.H. Zhu, J. Appl. Polym. Sci., 122, 2907 (2011).

(8.) E.H. Tejeda, C.Z. Sahagun, R. Gonzalez-Nunez, and D. Rodrigue, J. Cell. Plast., 41, 417 (2005).

(9.) M.J. Yuan, L.S. Turng, S.Q. Gong, D. Caulfield, C. Hunt, and R. Spindler, Polym. Eng. Sci., 44, 673 (2004).

(10.) H.X. Huang and J.K. Wang, Polym. Test., 27, 513 (2008).

(11.) S.J.A. Rizvi and N. Bhatnagar, Inter. Polym. Process., 24, 399 (2009).

(12.) N.B. Wu, H.J. Haugen, and E. Wintermantel, J. Cell. Plast., 48, 141 (2011).

(13.) M.J. Yuan, A. Winardi, S.Q. Gong, and L.S. Turng, Part. Eng. Sci., 45, 773 (2005).

(14.) S.S. Hwang, S.P. Liu, P.P. Hsu, J.M. Yeh, K.C. Chang, and Y.Z. Lai, Inter. Commun. Heat. Mass., 37, 1036 (2010).

(15.) S.S. Hwang, S.P. Liu, P.P. Hsu, J.M. Yeh, J.P. Yang, K.C. Chang, and S.N. Chu, Inter. Commun. Heat. Mass., 38, 1219 (2011).

(16.) J.J. Lee, L.S. Turng, and A. Kramschuster, Polym. Plast. Technol. Eng., 49, 1339 (2010).

(17.) A. Javadi, A.J. Kramschuster, S. Pilla, J.J. Lee, S.Q. Gong, and L.S. Turng, Polym. Eng. Sci., 50, 1440 (2010).

(18.) A.K. Bledzki, M. Rohleder, H. Kirschling, and A. Chate, J. Cell. Plast., 46, 415 (2010).

(19.) A.K. Bledzki, J. Kuhn, H. Kirschling, and W. Pitscheneder, Cell. Polym., 27, 91 (2008).

(20.) H.X. Huang, K. Li, and S. Li, Polym. Plast. Technol. Eng., 48, 64 (2009).

(21.) C. Yang, H.X. Huang, and K. Li, Polym. Compos., 31, 1899 (2010).

(22.) W.S. Guan and H.X. Huang, Inter. Commun. Heat. Mass., 39, 792 (2012).

(23.) Y.K. Shen, Plast. Rubber. Compos., 36, 77 (2007).

(24.) C.C.A. Chen and S.C. Kao, Key Eng. Mater., 364-366, 1211 (2008).

(25.) M. Rohde, A. Derdouri, and M.R. Kamal, Inter. Pulpit. Process., 24, 288 (2009).

(26.) R. Gosselin and D. Rodrigue, Polym. Test., 24, 1027 (2005).

(27.) E. Reverchon and S. Cardea, J. Supercrit. Fluid., 40, 144 (2007).

(28.) S.K. Lim, J.J. Lee, S.G. Jang, S.I. Lee, K.H. Lee, H.J. Choi, and I.J. Chin, Polym. Eng. Sci., 51, 1316 (2011).

(29.) W.B. Young, Inter. Polym. Process., 15, 416 (2000).

(30.) Y. Moon, K.S. Lee, and S.W. Cha, J. Mech. Sci. Technol., 23, 3349 (2009).

(31.) S.N. Leung, A. Wong, Q.P. Guo, C.B. Park, and J.H. Zong, Chem. Eng. Sci., 64, 4899 (2009).

(32.) J.H. Seo, S.W. Cha, and H.B. Kim, Polym. Plast. Technol. Eng., 48, 351 (2009).

(33.) S. Han, P. Kennedy, R. Zheng, J. Xu, and L. Kishbaugh, J. Cell. Plast., 39, 475 (2003).

Han-Xiong Huang, Jia-Dong Tian, Wei-Sheng Guan

Lab for Micro Molding and Polymer Rheology, the Key Laboratory of Polymer Processing Engineering of the Ministry of Education, South China University of Technology, Guangzhou 510640, People's Republic of China
COPYRIGHT 2014 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Huang, Han-Xiong; Tian, Jia-Dong; Guan, Wei-Sheng
Publication:Polymer Engineering and Science
Article Type:Report
Geographic Code:9CHIN
Date:Feb 1, 2014
Previous Article:Structures and properties of injection-molded biodegradable polylactic acid nanocomposites prepared with untreated and treated multiwalled carbon...
Next Article:Mechanical reinforcement of polyethylene using n-alkyl group-functionalized multiwalled carbon nanotubes: effect of alkyl group carbon chain length...

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters