Microcellular Injection Molding

By Jingyi Xu and Lih-Sheng (Tom) Turng

This book presents the most important aspects of microcellular injection molding with applications for science and industry. The book includes: experimental rheology and pressure-volume-temperature (PVT) data for different gas materials at real injection molding conditions, new mathematical models, micrographs of rheological and thermodynamic phenomena, and the morphologies of microcellular foam made by injection molding. Further, the author proposes two stages of processing for microcellular injection molding, along with a methodology of systematic analysis for process optimization. This gives critical guidelines for quality and quantity analyses for processing and equipment design.

• Language: English
• Publisher: Wiley
• Release date: Jan 6, 2011
• ISBN: 9781118057872

Book preview

INTRODUCTION

1.1 HISTORY OF MICROCELLULAR PLASTICS

Historically, microcellular plastics are not new: They existed more or less in the thin transition layer of structural foams. It can be found partially in sections with thin thickness, as well in the high shearing zone of structural foam parts. However, as an idea to develop microcellular plastics, Dr. Nam Suh and his students at the Massachusetts Institute of Technology invented micro­cellular processing in the early 1980s. This technology proposes two goals: One is to reduce the material, and another is to promote the material toughness by tiny spherical cells that act as crack arrestors by blunting the crack tip [1]. Furthermore, the rigidity of the material in resisting the buckling of the cell walls has been improved through the formation of spherical closed cells. Concentrated research and development efforts of microcellular foams began in the late 1980s, with a focus on the batch process and the topics mentioned above.

The microcellular batch processing technology was invented at the Massachusetts Institute of Technology (MIT) from 1980 to 1984 [1], and the first U.S. patent on microcellular technology was issued in 1984 [2]. Jonathan Colton showed a heterogeneous nucleation mechanism from the effects of additives in the polymers at certain levels of solubility [3]. Jonathan Colton also investigated the methodology of foaming for semicrystalline polymers such as polypropylene (PP) [4]. The gas can be dissolved into the amorphous structure because raising the temperature beyond its melting point eliminates the crystalline phase of PP. This heterogeneous nucleation is now dominating today’s industry processing. On the other hand, the crystalline material, such as PP, has been used for microcellular foam by Jonathan’s method in the industry practice now. Chul Park and Dan Baldwin studied the continuous extrusion of microcellular foam. Chul Park investigated both (a) the dissolution of gas at the acceptable production rate and (b) the application of a rapid pressure drop nozzle as the nucleation device [5]. Dan Baldwin studied the microcellular structure in both crystalline and amorphous materials [6]. Sung Cha investigated the application of supercritical fluid, such as CO2, to dissolve the gas faster and to create more cells [7, 8]. With supercritical fluid, the cell density was increased from 10⁹ cells/cm³ to 10¹⁵ cells/cm³. Vipin Kumar also used thermoforming supersaturated plastic sheets to study the issues of shaping three-dimensional parts [9]. Sung Cha also found that the large volume of gas in polymers decreases significantly with the glass transition temperature of plastics. Therefore, simultaneous room temperature foaming is possible. All of these pioneer contributions are fundamental to microcellular foam technologies. Through many people’s creative research, this technology has completed the laboratory stage and transitioned to industry application.

The commercial application of microcellular technology began in 1995 by Axiomatics Corp., which was later renamed Trexel Inc. Trexel continued to develop microcellular technology through extrusion first. Then, the first injection molding machine with plunger for injection and extruding screw for plasticizing and gas dosing was developed in Trexel Inc. with the help from Engel Canada in mid-1997. After successful microcellular injection molding trials were carried out in this plunger-plus-extruder injection molding machine, the first reciprocating screw injection microcellular molding machine was built by Trexel and Engel together in 1998 [10]. This machine marks the milestone of the commercialization of microcellular injection molding and is now the most popular microcellular injection molding machine in the world. Trexel also modified a Uniloy Milacron machine to the first microcellular blow-molding machine in 2000.

One important term, supercritical fluid, is abbreviated as SCF. SCF is the name of the state condition of a gas when the gas is above both its critical pressure and critical temperature; this is discussed in more detail in Chapter 2. It is critical to use SCF to describe a gas if the gas is at a supercritical state. Otherwise, use the general term, gas, if the gas is at any condition from normal atmospheric to supercritical state. Unless otherwise specified, the term of SCF and gas will be used with the conditions above in the entire book.

The injection molding aspect of microcellular foam processing has developed the fastest. The main developed technologies of microcellular injection molding are listed in Table 1.1. The most popular trade name for this technology is MuCell® and is licensed by Trexel Inc. since 2000 (MuCell® is a Registered Trademark of Trexel Inc., Woburn, Massachusetts). Several other injection molding companies and research groups in the world were developing this technology prior to Trexel’s announcement of MuCell®. However, they did not finish the commercialization of their technologies for real applications. The MuCell® technology uses a reciprocating screw as the SCF dosing element, and the SCF is injected into the reciprocating screw through the barrel. It makes full use of the shearing and mixing functions of the screw to quickly finish the SCF dosing and to maintain the minimum dosing pressure in the barrel and screw for the possible continuing process of microcellular injection molding. In addition, two other trade names of this technology were found later on: (a) Optifoam® licensed by Sulzer Chemtech [11] and (b) Ergocell® licensed by Demag (now Sumitomo-Demag in 2008) [12]. Optifoam® is a microcellular technology that uses a nozzle as the SCF dosing element. It is a revolutionary change to the traditional SCF dosing method, which adds gas into the barrel. This unique, innovative idea has a special nozzle sleeve made of sintered metal with many ports to let gas go through as tiny droplets. On the other hand, the melt flow through the nozzle is divided into a thin film between the nozzle channel and the sintered metal sleeve. As a result, the gas can diffuse into the melt in a short amount of time. The gas-rich melt is then further mixed in a static blender channel that is located in the downstream of the nozzle dosing sleeve. The advantage of this technology is that the regular injection screw and barrel do not need to be changed. The regular injection molding machine in existence can be easily changed to use the Optifoam® process. However, only some of these applications have been successful [11]. At K2001, Demag Ergotech introduced its Ergocell® cellular foam system [12]. Ergocell® technology has reached an agreement with Trexel to have their customers pay a reduced price to the MuCell® license when using Ergocell® technology legally. The Ergocell® system is essentially an assembly of an accumulator, a mixer, a gas supply, and a special injection system that is mechanically integrated between the end of the barrel and the mold to put gas into the polymer and create the foam upon injection into the mold. A special assembly needs to be created for each screw diameter. Additional hydraulic pumps and motor capacity must be added to operate the mixer and accumulator injection system. The system only uses carbon dioxide as the blowing agent.

TABLE 1.1 Main Developed Microcellular Injection Molding Technologies

The latest developing foam technology from IKV is the ProFoam® process [13]. It is a new and cheap means of physically foaming injection molding technology. The gas, either carbon dioxide or nitrogen, as the blowing agent is directly added into the hopper and diffuses into the polymer during the normal plasticizing process. The plasticizing unit of the molding machine is sealed off in the feeding section of screw for gas adding at pressure, but feeding of pellets of material occurs at normal conditions without pressure. With this ProFoam® process the part can reduce up to 30% weight via the foaming.

Trexel continues to develop and support the microcellular injection molding process worldwide. There are already over 300 MuCell® injection microcellular molding machines in the world. Through the efforts of many more organizations, more and more advances are being made for the microcellular injection molding process. These organizations include not only original equipment manufacturers (OEMs) licensed from Trexel but also numerous unlicensed organizations, such as universities, and university/industry consortia. All of them are contributing to further advances in microcellular technology.

1.2 ADVANTAGES AND APPLICATIONS OF MICROCELLULAR PLASTICS

The microscopic cell size and large number of cells in microcellular material can reduce material consumption as well as improve the molding thermodynamics, which results in a quicker cycle time. Additionally, the process is a low-pressure molding process and produces stress-free and less warped injection molding products. The major differences between conventional foam and microcellular foam are cell density and cell size. The typical conventional polystyrene foam will have an average cell size of about 250 microns, and a typical cell density in the range of 10⁴–10⁵ cells/cm³. Microcellular plastic is ideally defined with a uniform cell size of about 10 µm and with a cell density as high as 10⁹ cells/cm³ [1]. It is possible to make this kind of microstructure cell density with microcellular injection molding if material and processing are controlled very well. The scanning electron microscope (SEM) morphology of glass-fiber-filled PBT is an excellent example of microcellular injection molding that almost matches the ideal definition of microcellular plastics made by batch process. It is made by using 30% glass fiber and reinforced polybutylene terephthalate (PBT) with a 15% weight reduction (see Chapter 3, Figure 3.12). The cell density is about 8 × 10⁸ cells/cm³, with an average of 15 µm of uniform cell distribution. However, this microstructure is not always the result of microcellular injection molding. The SEM picture in Figure 1.1 is a more typical microcellular unfilled polystyrene foam made by injection molding that has an average of 25 microns, and has a cell density of about 8.1 × 10⁷ cells/cm³. The microstructures of industrial parts from microcellular injection molding are characterized by an average cell size on the order of 100 µm, although the real cell size can be varied from 3 µm to 100 µm. However, the cell structure of the microcellular part with microcellular injection molding might not necessarily be defined as the cell density of 10⁹ cells/cm³. The microstructure of ABS has a cell density of about 10⁶ cells/cm³, and it definitely shows a microcellular structure with an average cell size of about 45 µm. The comparisons of average cell sizes between microcellular foam and conventional foam are summarized in Table 1.2. The data in Table 1.2 show that the minimum cell size of conventional foam is about the same size as the maximum cell size of microcellular foam; the maximum cell size of conventional foam is about twice as large as the maximum cell size of microcellular foam. Usually the cell density of the conventional foam is about 10² to 10⁶ cells/cm³. However, the cell density of the microcellular foam is 10⁶ cells/cm³ or higher.

TABLE 1.2 Comparisons Among Conventional Foam, Microcellular Foam, and Regular Solid

c01t00629xjc01t0072aigc01t0082agx

aNA, not available.

Figure 1.1 Morphology of polystyrene microcellular foam (white bar indicates 100 µm). Average cell size: 25 µm. Cell density: 8.1 × 10⁷ cells/cm³.

c01f001

The cell size in the foam mainly determines the property differences between conventional foam and microcellular foam. Table 1.1 shows the comparisons among injection molding parts made by conventional foam, micro­cellular foam, and regular solid. It is clear that microcellular foam has more advantages than conventional foam. Microcellular foam overcomes the major disadvantages of conventional foam, such as a long cycle time and a thick wall. The most important advantages of microcellular foam can be summarized as follows:

The main advantage of structural foam molding (one of the conventional foams) is to increase stiffness without increasing the weight of the component. Microcellular foam can be made for this target as well, by redesigning thin wall structures and by creating a nice cell structure to save material (weight reduction by a thin wall) and cost (shorter cycle time).

The microcellular process can be used for thin-wall solid parts that are difficult to make full mold filling from flow restrictions, which results in either clamp tonnage shortage or injection pressure limit.

Microcellular technology allows mold filling without foaming because the gas-rich melt reduces viscosity significantly.

The microcellular process almost eliminates all dimension stability problems, such as sink mark, flatness defects, warp, and residual stress after molding due to the elimination of pack and hold phases during molding.

The microcellular process dramatically reduces cycle time if the part is designed properly.

Microcellular processing equipment can be designed to save more energy since the peak of injection pressure is not necessary and also saves up to 50% of clamp tonnage.

The disadvantages of microcellular foam are the same as conventional foam, such as poor surface finish, strictly balanced runner system for multi­cavity mold, nontransparent application only, and complicated processing technology.

1.3 PATENTS AND PUBLICATIONS COVERING MICROCELLULAR INJECTION MOLDING TECHNOLOGY

There have been many patents issued for microcellular injection molding since 1998. The major patents, directly or indirectly related to microcellular injection molding technology, are listed here:

Pierick, D. E., et al., International Patent Application WO 98 31 521 A2 (1998)

Park, C. B., et al., U.S. Patent No. 5,866,053 (1999)

Pierick, D. E., et al., International Patent Application WO 00 26 005 A1 (2000)

Xu, J., International Patent Application WO 00 59 702 A1 (2000)

Michaeli, W., et al., German Patent DE 19 853 021 A1, (2000)

Anderson, J. R., et al., International Patent Application WO 01 89 794 A1 (2001)

Xu, J., U.S. Patent No. 6,322,347 (2001)

Burnham, T. B., et al., U.S. Patent No. 6,284,810 (2001)

Anderson, J. R., et al., U.S. Patent No. 6,376,059 (2002)

Gruber, H., et al., U.S. Patent Application No. 0,056,935 A1 (2002)

Pierick, D. E., et al., International Patent Application WO 02 090 085 A1 (2002)

Kim, R. Y., et al., International Patent Application WO 02 081 556 A1 (2002)

Vadala, J. P., et al., International Patent Application WO 02 026 484 A1 (2002)

Kishbaugh, L. A., et al., International Patent Application WO 02 026 485 A1 (2002)

Kishbaugh, L. A., et al., International Patent Application WO 02 072 927 A1 (2002)

Xu, J., U.S. Patent No. 6,579,910 B2 (2003)

Anderson, J. R., et al., U.S. Patent No. 6,593,384 (2003)

Dwivedi, R. K., U.S. Patent No. 6,759,004 (2004)

Cardona, J. C., et al, U.S. Patent No. 6,926,507 (2005)

Anderson, G., et al., U.S. Patent No. 7,172,333 (2007)

Xu, J., U.S. Patent No. 7,267,534 (2007)

Xu, J., et al., U.S. Patent No. 7,318,713 (2008)

Kishbaugh, L.A., et al., U.S. Patent No. 7,364,788 B2 (2008)

Xu, J., et al., U.S. Patent No. 7,615,170 B2 (2009)

There are many publications regarding the technology behind microcellular injection molding. They cover both the fundamentals and real practices in industry. However, it is well known a huge gap exists in fundamentals and realities. Hopefully, this comprehensive coverage in the book will help bridge this gap and will enable readers to apply the concepts in a straightforward manner.

1.4 OUTLINES OF THE BOOK

This book presents the microcellular history and a specific short history of microcellular injection molding in Chapter 1. Then, in Chapters 2 and 3, the fundamental knowledge of microcellular injection molding is covered. With the understanding of the principles of microcellular processing, a review of materials and details of design for microcellular injection molding are well discussed in Chapters 4 and 5. Moreover, injection molding makes the foaming process more complex. Therefore, both theory and experiments are needed for good analyses of microcellular process. Chapter 6 uses the fundamental guidelines in previous chapters to analyze the specific processing procedures one by one with a combination of theory and empirical data. Some comparisons among different gas-entrained processes, such as gas assistant, micro­cellular extrusion, microcellular blow molding, and structural foam molding are discussed in Chapter 6. It is also important to know the differences between regular injection molding and microcellular injection molding, which is discussed briefly in Chapter 6. To realize the processing requirements in Chapter 6, the equipment designing rules are introduced in Chapter 7. It will generate further insight on both the future development and the efficient operation. After understanding normal microcellular injection molding, more specialized microcellular injection molding processes are discussed in Chapter 8. All commercialized special processes and most developing special processes are covered in this chapter. In addition, the modeling of microcellular injection molding is also presented in Chapter 9. Some PVT data and rheology data of the gas-laden polymer melt are given in Chapter 9. The necessary postprocesses and basic test procedures are briefly introduced in Chapter 10. Finally, application in the market is covered in Chapter 11, and cost analyses are presented in Chapter 12.

REFERENCES

1. Suh, N. P. Innovation in Polymer Processing, edited by James F. Stevenson, Hanser/Gardner Publications, Cincinnati, 1996, Chapter 3, pp. 93–149.

2. Martine-Vvedensky, J. E., Suh, N. P., and Waldman, F. A. U.S. Patent No. 4,473,665 (1984).

3. Colton, J. S. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, MA, 1985.

4. Colton, J. S., and Suh, N. P. U.S. Patent No. 4,922,082 (1990).

5. Park, C. B. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, MA, 1993.

6. Baldwin, D. F. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, MA, 1994.

7. Cha, S. W. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, MA, 1994.

8. Cha, S. W., Suh, N. P., Baldwin, D. F., and Park, C. B. U.S. Patent No. 5,158,986 (1992).

9. Kumar, V. Ph.D. Thesis, Massachusetts Institute of Technology, Cambridge, MA, 1988.

10. Xu, J., and Pierick, D. J. Injection Molding Technol. 5, 152–159 (2001).

11. Pfannschmidt, O., and Michaeli, W. SPE ANTEC, Tech. Papers, 2100–2103 (1999).

12. Witzler, S., Injection Molding Mag. December, 80 (2001).

13. Defosse, M. Modern Plastics Worldwide December, 14–15 (2009).

2

BASICS OF MICROCELLULAR INJECTION MOLDING

The fundamental theory for microcellular injection molding has been developed for a decade and is still one of major research topics in the plastics industry. The basics of microcellular plastics introduced in this chapter will serve as the general guidelines to both fundamental research and technology development for the microcellular injection molding.

2.1 BASIC PROCEDURES OF MICROCELLULAR INJECTION MOLDING

Typical analyses of gases as the supercritical fluids (SCF) focus on the solubility and dissolution capability in different plastics. To make a nice mixture of gas–polymer solution is a real challenge in the industrial plasticizing unit. After this solution is ready, the nucleation will be the next key technology for success of microcellular injection molding. Finally, how to control the cell growth and distribution throughout the molding part becomes the value of the microcellular part. If the injection time is too long, the bubble collapse and coalescence in the flow front may occur. However, the injection time usually is very short for microcellular injection molding. Therefore, the bubble collapse and coalescence will not be considered as the normal defects in injection molding. To summarize the key issues for successful microcellular injection molding, there are four basic steps of microcellular injection molding: SCF mixing and dissolution in the melt of polymer; nucleation of cells; cell growth; and shaping in the mold [1–3].

The concept of the continuous process was successfully tried in an extruding process [4]. It basically needs to create the gas–polymer solution first in the barrel. The gas at the supercritical fluid state is metered and injected into the barrel and then dissolved into molten polymer, as shown in Figure 2.1a. The gas is pressurized up to 20.7 MPa before flowing into the barrel. As the gas flows into molten polymer, it forms big gas droplet in the molten polymer since the flow of the gas is briefly interrupted every time the screw flight wipes over the barrel. The size of the gas droplet in the molten polymer is determined by five major factors: gas pressure and molten polymer pressure; gas flow rate; viscosity of molten polymer; wiping frequency of the flights (screw rotation speed); and diameter of orifice in the gas injector. Then, the large gas droplet is elongated in the barrel through the shear deformation induced by the screw rotation. The elongated gas droplet will be broken up forming many small gas droplets above a critical value of the Weber number We, which is a ratio of shear force to the surface force (refer to Chapter 7). These gas droplets may be stabilized in the screw channels to form bubbles in the molten polymer matrix. These bubbles, in turn, undergo elongation with additional shear deformation that increases the area-to-volume ratio of each gas bubble. Then, the gas in the bubble diffuses quickly into the molten polymer due to the increased polymer–gas interfacial area and decreased striation thickness of polymer between the gas bubbles.

Figure 2.1 Schematic of the gas–polymer solution. (a) Gas injected into polymer melt. (b) Mixed gas-polymer solution.

c02f001

However, the shearing rates are varied from the different layers in the channel of screw so that the bubble sizes are different from top (inside diameter of barrel) to bottom (root diameter in the screw) in the flow channel of the screw. The screw mixing section must be designed with the mixing elements to alter the positions of bubbles from top to bottom and vice versa, which will be discussed in Chapter 7. Eventually, the gas droplets must be small and uniformly distributed in the molten polymer matrix, as shown in Figure 2.1b. It may be defined as gas–polymer mixture ready for nucleation. Ideally, the final gas–polymer solution should become the so-called single-phase solution [3, 5]. In other words, there are no separate phases such as gas phase and polymer melt phase. However, the real practice in the injection molding machine can create the excellent gas–polymer solution with tiny bubbles in the molten polymer. The single-phase solution of gas and molten polymer may never truly form in such short recovery time in the plasticizing screw [1], and even shorter mixing time with new technology of injecting gas through the nozzle during injection [6, 7]. Therefore, the single-phase solution may be defined in this book as the gas–polymer solution with a uniformity distribution of many tiny bubbles, which has been proven to be a good mixture of gas–polymer solution ready for the next step of microcellular processing in the most current technologies of microcellular injection molding processes [1–7].

Then, the gas–polymer solution needs to be induced by a rapid thermodynamic instability of this mixture of gas and molten polymer for cell nucleation. The thermodynamic instability is generated by either a pressure drop or a temperature change with high rate. Practically, a quick change in the pressure is much easier than a fast change in the temperature in a very short period. Therefore, a very high pressure drop rate occurs in either the nozzle orifice or the valve gate, where the narrow orifice causes a high pressure drop rate up to 1 GPa/sec or higher.

Once enough nuclei are created, the nucleated gas–melt mixture is still kept warm for cell growth in the center layer of the part when the skin begins to cool down. In addition, the short shot of injection leaves enough space for cell growth. On the other hand, enough gas is available to provide the necessary gas supply around the nuclei, which is growing further to form a stable cell.

Finally, the part in the mold not only conforms to the shape of the mold but also builds up the skin-cell structure. The cell growth in the part results in a perfect microcellular part. The cells retain their shape and size during the cooling, and also the residual gas pressure inside the cells pushes the part, thereby expanding the cells to overcome the shrinkage of polymer. Then, the expanding of cells inside the part helps the part to contact the cold wall of mold to copy the mold shape exactly and to form the solid skin quickly.

2.2 SUPERCRITICAL FLUIDS (SCF)

The gas must be dissolved into molten polymer in the limited time so that the best condition of the gas state to satisfy the rate-limiting process is the gas at the supercritical state. Figure 2.2 shows the gas-phase diagram. The shadowed area represents the region of supercritical state of the gas where the gas is at a liquid-like state, the so-called supercritical fluid (SCF). This SCF region is located beyond both critical pressure Pcr and critical temperature Tcr. Therefore, the processing setup parameters for microcellular processing must have high pressure and temperature to produce the gases used for microcellular processing at the supercritical state. There are also two critical points in this diagram that must be set up accordingly during the process. One is the critical pressure and another is critical temperature. The microcellular process usually works with the processing conditions above both critical pressure and temperature to quickly diffuse the gas into molten polymer.

Figure 2.2 Diagram of material phases.

(Courtesy of Trexel Inc.)

c02f002

The supercritical fluid is neither gas nor liquid in a certain temperature and pressure regimen higher than the critical pressure and critical temperature. In this state the gas will have both gas-like and liquid-like properties. The gas-like property of SCF is the low viscosity similar to the air viscosity even if the viscosity of SCF may increase several times higher than the original one at gas phase. On the other hand, SCF has a liquid-like property—that is, a heavy liquid density compared to gas density. For example, the density of carbon dioxide (CO2) is 0.001 g/cm³ at the gas phase, but 0.7 g/cm³ at SCF state that is close the density of liquid CO2 0.8 g/cm³. Both properties are important to precisely meter the SCF weight percentage to match the molten polymer output rate. In addition, both properties are necessary for possible mixing of SCF into molten polymers.

The list of critical point data for various fluids that can be used for blowing agents is presented in Table 2.1 [4, 8]. Although many of them can be used as supercritical fluids in microcellular processing, their shortcomings may restrict their applications for microcellular process. Nitrogen (N2) has a low solubility but can make very small cells. This is the main reason why N2 gas is widely used as one of the common physical blowing agents. Carbon dioxide (CO2) at a critical state can enhance the solubility and diffusion rate. The great percentage of CO2 can be added into the polymer and can be quickly degassed after the molding. Therefore, CO2 is also a common physical blowing agent for microcellular processing. Water is corrosive and has very low solubility in the molten polymer. Argon (Ar) is expensive and also has relatively low solubility. Lots of other organic liquids are used for blowing agents, but they are abandoned because of environmental consequences also because they are hazardous. Both CO2 and N2 gases are the most common blowing agents used for microcellular processing. They are environmentally benign blowing agents and are inexpensive to obtain from the air. In addition, they are not ozone-depleting and are a viable alternative to other volatile blowing agents.

TABLE 2.1 Critical Points of Pure Components of Potential Blowing Agents [4, 8]

There are huge viscosity differences between SCF and all molten poly­mers. Typically, molten polymers normally exhibit high viscosity in the range of 10–1,000,000 poise (g/cm · sec). However, the gaseous blowing agent will normally exhibit viscosity in the very low range of 0.00005–0.05 poise (g/cm · sec) [9]. Hence, the relatively very low resistance for the blowing agent flows easily in the gas injector and helps to clean up the high-viscosity material in the flow lines of the gas injector. In addition, a low-viscosity liquid such as SCF makes the mixing between SCF and molten polymer easier. However, every time the injector of low-viscosity SCF is opened for injecting gas into high-viscosity material, it is kind of a surge, and an initial large gas pocket may form in the gas dosing position. It must be controlled by the small pressure difference between gas pressure and melt pressure at dosing position in the barrel.

2.3 GAS SOLUBILITY AND DIFFUSION CAPABILITY IN POLYMER MELT

There are many gases available to be used as blowing agents. As the common physical blowing agent sources, some gases have been used for many professionals in industry for a long time, such as nitrogen, CO2 and even air. Some exotic gases such as argon (Ar), helium (He), and hydrogen (H2) have been tested in the laboratory by some researchers, but none of them are used as commercialized applications. Water (H2O) is also a possible blowing agent source that was used for both practices and researches. However, for all foaming industries, including microcellular injection molding, only CO2 and nitrogen gases are by far the most widely used physical blowing agents. Therefore, nitrogen and CO2 will be discussed in more detail in this book. In addition, argon and helium gases will be introduced for their solubility and diffusion capability since recent studies include them as potential new sources for microcellular processing. Some basic physical data for CO2 and N2 gases are listed in Table 2.2 since only these two gases are used now in the microcellular foam industry. The data in Table 2.2 are useful for calculating the gas flow rate to match the processing requirements.

TABLE 2.2 Properties of Popular Blowing Agents

Source: Throne [12], with permission of Sherwood Publishers.

2.3.1 Gas Solubility in Polymer Melt

All gases can dissolve in all liquids to some extent. The measure of a gas dissolving potential in a liquid is defined as solubility. Solubility is measured in standard gas volumetric uptake per unit weight of liquid, such as (cm³ (STP)/g polymer).

2.3.1.1 General Relationship Between Solubility and Processing Conditions in the Gas–Polymer System.

Within the plasticizing stage of injection molding equipment or within an extruder, the pressure and temperature are high and the solubility of the gas blowing agent is high; at this point, the polymer is saturated with the gas blowing agent. In the gate of a mold, the pressure drops quickly, and the gas blowing agent becomes supersaturated within the polymer; at this point, the gas blowing agent will begin to precipitate out in the form of gas, thereby foaming the polymer. If the drop in blowing agent solubility is sufficiently large and sufficiently fast, then conditions will exist for homo­geneous nucleation of cells, and a large number of evenly distributed microscopic cells will form and grow uniformly.

The conditions required for homogeneous nucleation are best illustrated by plotting the solubility of a blowing agent as a function of pressure in a typical polymer system (see Figure 2.3). As the general trend the gases solubility in the molten polymer will increase with the increasing pressure and will decrease with the increasing temperature, as shown in Figure 2.3 [4, 8]. In most gas–polymer systems, the solubility increases almost linearly with the melt pressure [10]. Usually, the pressure and temperature have opposite tendencies to promote the gas solubility in the polymer. The processing parameters can be set up for high solubility based on clear trends shown in Figure 2.3. The results in Figure 2.3 indicate that a higher solubility of a blowing agent in a polymer is obtained by either increasing the processing pressure or decreasing the processing temperature. Sometimes, to maximize the solubility of gas in molten polymer, both pressure increasing and temperature decreasing can be used simultaneously. For the general rule of processing, both high pressure and low temperature will increase the gas solubility in the molten polymer.

Figure 2.3 Solubility of gas in a molten polymer. (a) Solubility versus pressure [4]. (b) Solubility versus temperature: solid line for most materials, dashed line for some reverse solubility materials.

c02f003

The experimental data from Sato and others verify the trends of solubility in Figure 2.3a and the solid line in Figure 2.3b. For example, the same trend of solubility changes with pressure and temperature as shown in Figure 2.3 is verified by the CO2–PP system in Table 2.3, CO2–HDPE system in Table 2.4, and N2–PS system in Table 2.5, respectively [10]. It is obvious that the solubility of CO2 in PP and HDPE and the solubility of N2 in PS will increase with the increasing pressure and will decrease with the increasing temperature.

TABLE 2.3 Solubility of Carbon Dioxide in Unfilled Polypropylene Materiala

aPP with x998_TimesTen-Italic_8n_000100 w = 4.51 × 10⁵, x998_TimesTen-Italic_8n_000100 w/ x998_TimesTen-Italic_8n_000100 n = 7.05, Tm = 431 K.

Source: Sato et al. [10], with permission of Elsevier Publishers.

TABLE 2.4 Solubility of Carbon Dioxide in Unfilled High-Density Polyethylene Materiala

aHDPE with x998_TimesTen-Italic_8n_000100 w = 1.11 × 10⁵, x998_TimesTen-Italic_8n_000100 w/ x998_TimesTen-Italic_8n_000100 n = 13.6, Tm = 402 K.

Source: Sato et al. [10], with permission of Elsevier Publishers.

TABLE 2.5 Solubility of Nitrogen in Unfilled Polystyrene Materiala

aPS with x998_TimesTen-Italic_8n_000100 w = 1.87 × 10⁵, x998_TimesTen-Italic_8n_000100 w/ x998_TimesTen-Italic_8n_000100 n = 2.67, Tm = 373.6 K.

Source: Sato et al. [10], with permission of Elsevier Publishers.

However, there are some exceptions for the temperature effect on the gas solubility in some polymers (see dashed line in Figure 2.3b, represented as reverse solubility). The solubility of gases, like other solubility, can increase or decrease with temperature, as determined by two contributions:

Energy is absorbed to open a pocket in the solvent. Solvent molecules attract each other. Pulling them apart to make a cavity will require energy, and heat is absorbed in this step for most solvents.

Energy is released when a gas molecule is popped into the pocket. Intermolecular attractions between the gas molecule and the surrounding solvent molecules lower its energy, and heat is released. The stronger the attractions, the greater the amount of heat released.

There is usually net absorption of heat when gases are dissolved in organic solvent, such as molten polymer, because the pocket-making contribution is bigger. Le Chatelier’s principle predicts that when heat is absorbed from the dissolution process, it will be favored at higher temperature. Solubility is expected to increase when temperature rises, like the dashed line of the profile in Figure 2.3b [10].

For example, the solubility changes with pressure for both N2–PP and N2–HDPE with the same trend in Figure 2.3a, and the detailed data are listed in Table 2.6, and 2.7. However, the solubility changes with temperature show the exception that matches the dashed line profile in Figure 2.3b. It is named reverse solubility since it is not a common phenomena. Both N2–P and N2–HDPE systems show the reverse solubility in Tables 2.6 and 2.7. It is more obvious for PP with N2 gas at high pressure. However, the HDPE will have a small increase in solubility with the increasing temperature at the same pressure. As a brief comparison, CO2 in PP and HDPE do not have this reverse solubility trend for the temperature effects. Therefore, to know the solubility changes, both plastics and gases need to be checked, and the right choices for the processing conditions need to made accordingly.

TABLE 2.6 Solubility of Nitrogen in Unfilled Polypropylene Material

Source: Sato et al. [10], with permission of Elsevier Publishers.

TABLE 2.7 Solubility of Nitrogen in Unfilled High-Density Polyethylene Material

Source: Sato et al. [10], with permission of Elsevier Publishers.

Although the conclusions above are from the experiments, the general trends in Figure 2.3 are still good standards for the industry. In most cases, there will be more mechanical heat generated from higher processing pressure, so the processing temperature will be automatically increased when the processing pressure is increased. Based on the trend of solubility varied with pressure and temperature in Figure 2.3, if the processing tem­perature and pressure increase simultaneously, then the final solubility will be unknown. However, the experiences in real practices show that the blowing agent solubility in a polymer increases more quickly with the increasing pressure whereas the blowing agent solubility in a polymer decreases with the increasing temperature from mechanical heat. Therefore, the solubility in a polymer will be increased with the increasing pressure in most cases.

2.3.1.2 Gas Concentration Calculation.

To estimate the gas concentration in certain molten polymers, Henry’s law provides an equation. At thermodynamic equilibrium, the external pressure and the gas concentration are related to each other [4, 11]:

(2.1) c02e001

where C is gas concentration (cm³ (STP)/g polymer), H is Henry’s law constant (cm³ (STP)/g atm), Pm is molten polymer pressure (also means gas pressure), and Tpoly is molten polymer temperature (K).

At low melt pressure and low gas concentrations, H is constant. At high pressure, H depends on both pressure and temperature. It is well known that the temperature dependence follows the Arrhenius-type rate equation [4]. For most gases in polymers, the amount of gas dissolved in a polymer is linearly dependent on the imposed gas pressure at a given temperature [12]. Then, Henry’s law constant is temperature-dependent according to

(2.2) c02e002

where H0 is the preexponential constant for Henry’s law constant Rg is the gas constant.

Some experimental values for Henry’s law constant for several polymer–gas combinations can be found in reference 12.

Some of the heats of solution and energies of activation for solution are listed in Table 2.8 [12]. An empirical relationship for the Henry’s law constant has been developed [12]:

(2.3) c02e003

where Tcr is the critical temperature of gas.

TABLE 2.8 Diffusivities and Diffusional Energy of Activation (kcal/mol)

c02t0242awc

aCalculated value.

Source: Throne [12], with permission of Sherwood Publishers.

There are more empirical formulae for different gas–polymer systems listed in Chapter 9. They are used for simplified simulation model for gas solubility, concentration, nucleation, and cell growth.

This correlation has been proposed for the solubility of gases in any amorphous polymer [12]. The solubility of semicrystalline plastics is a function of the extent of crystallinity, Xc:

(2.4) c02e004

where Xc is the extent of crystallinity in semicrystalline material and Xa is the solubility of the gas in an amorphous portion of the semicrystalline material.

The measure for the rate at which molecules move through solids or liquids is diffusivity. Diffusion coefficients are related to the specific molecule moving through a specific liquid at a specific temperature. They are measured as unit area per unit time and are listed in Table 2.8 [12].

2.3.1.3 Nitrogen Gas (N2).

Nitrogen (N2) gas is an inexpensive, nonflammable, nontoxic permanent gas. It can be easily made from the air and is chemically inert, which results in an environmentally safe blowing agent to replace some ozone depletion chemical blowing agents. The gas state of nitrogen is available at 13.8 MPa (2000 psi) to 20.7 MPa (3000 psi) as compressed gas in the steel cylinder. Therefore, the pressure of N2 gas in the vendor’s tank already remains higher than the critical pressure. The liquid state of N2 is stored in the dewars as a cryogenic liquid at about −196 °C. For a heavily N2 flow rate application the cryogenic N2 is preferred. However, the N2 vapor needs to be boiled off from liquid state during the real usage, and the temperature will be near room temperature prior to metering and injection into the machine since the gas temperature needs to be back to critical state as well before injecting into the barrel. In other words, the N2 blowing agent is used only in the gas state in the delivery equipment except in the storage equipment. Overall, N2 is preferred in many, if not most, technical applications because it results in a more consistent and uniform micro­cellular part.

Polyolefin resins typically require significantly higher N2 levels to achieve good cell structure than do most other materials. These materials are also more likely to have significant cell structure variation from the gate to the end of fill. This situation will be aggravated by increased maximum wall thickness, greater than about 3.0 mm. It should be expected that the final nitrogen levels when running unfilled HDPE or unfilled PP will be 1% or higher. In fact, N2 gas levels as high as 2% already ran with these materials successfully with high processing pressure.

Some experimental values for Henry’s law constant for several polymer–nitrogen gas combinations are as follows [12]:

As the data show above, the dosage of N2 gas for polyethylene, or poly­propylene, is more than double that of the N2 gas dosage for polystyrene, which is the result from the laboratory at ideal conditions. The weight gain percentage of N2 gas in different plastics at real processing conditions, 200 °C and 27.6 MPa, are estimated as follows [4, 13–15]:

The real gas dosage used in industry is very close to the ideal data; this verifies that the ideal data represent good guidelines for the real process. Also, for the heats of solution and the energies of activation, the data for N2 in some polymers are as follows [12]:

2.3.1.4 Carbon Dioxide Gas (CO2).

Carbon dioxide (CO2) can be useful for a number of special cases when gas diffusion, or viscosity, is the primary challenge. It is similar to N2 for the usage of an ideal foaming agent. It is also an inexpensive, chemically inert, environmentally acceptable, and intriguing physical blowing agent. However, CO2 gas has some inherent handling problems, such as a relatively low critical point of 31 °C and 7.29 MPa (1027 psi). Therefore, CO2 will be a vapor above the critical point. It may be used in the delivery system with either gas state or liquid state.

Some experimental values for Henry’s law constant for several polymer–carbon dioxide gas combinations [12] are as follows:

Also, for the heats of solution and energy of activation, the data for carbon dioxide in different polymers are as follows [12]:

The weight gain percentages of carbon dioxide gas in different plastics at real processing conditions, 200 °C and 27.6 MPa, are estimated as follows [4, 13–15]:

As the general trend the gas dosage of CO2 in polyethylene, or polypropylene, is just slightly higher than the gas dosage of CO2 in polystyrene. On the other hand, total weight percentage of CO2 in the same plastic material will be much higher than the N2 gas dosage, which is about 3–4 times higher with the exception of PMMA. However, the experimental data in the laboratory of Trexel Inc. also show that an acrylic has the CO2 solubility of (a) 4.25 weight percent at 177 °C (350 °F) and 12.4 MPa and (b) 5.15 weight percent at 177 °C (350 °F) and 18.2 MPa.

Li et al. [16] at the University of Toronto reported the solubility of CO2 in PP at different temperature and pressure. They also matched the trends of the solubility of CO2 in polymer: high at high pressure and low temperature.

Figure 2.4 [17] shows the CO2 gas absorption weight percentage at different saturation percentages and different melt temperatures. The solubility of CO2 in polystyrene (PS) has been plotted as a function of pressure and temperature. As shown, the solubility of CO2 increases with increasing pressure, but it decreases with increasing temperature. It verifies the trends shown in Figure 2.3 regarding the solubility of CO2 in PS material. In addition, one more test result in Figure 2.4 is the solubility of CO2 in PS under the shearing. It is obvious that the shearing helps to increase the solubility of CO2 in PS material.

Figure 2.4 Gas solubility in PS melt (1 MPa=145 psi) at different pressure, temperature, and with shear [17],

(Courtesy of Trexel Inc.)

c02f004

2.3.1.5 Argon Gas (Ar).

Some experimental values for Henry’s law constant for several polymer–argon gas combinations [12] are as follows:

Based on the test performed by Wong et al. [18], the solubility of Ar gas in PP copolymer is the highest compared to N2 and He inert gases tested in their paper.

2.3.1.6 Helium Gas (He).

Wong and others also tested He gas in the PP copolymer melt. It has the lowest solubility compared to N2 and Ar gases. Some experimental values for Henry’s law constant for several polymer–helium gas combinations [12] are as follows:

2.3.1.7 Filled Materials.

Chen et al. [19] reported gas absorption percentages with different filled and unfilled polymer systems: (a) HDPE with/without talc and (b) rigid PVC with/without calcium carbonate. The HDPE was Equistar LP5403. The filler was Talc LG445, with a normal size of about 5 µm. The Rigid PVC uses Geon pipe-grade resin with a K value of 67, and filler is calcium carbonate with a nominal diameter of about 3 µm. Both talc and calcium carbonate were coated with surfactant before compounding in a twin screw. CO2 gas is the only one used in this test. A foaming process simulator has been built to study the gas absorption, and it can be pressurized up to 34.5 MPa and heated up to 232 °C. A rotor applies shear to the polymer melt in the pressurized chamber to investigate the shear effects on the gas absorption [20, 21].

The micropore theory is adapted by many researchers. For a porous surface, it not only offers surface energy but also is the residence for the gas molecules in the cavities that may cause cavitations locally. Figure 2.5 is the schematic of micropore model. The hypothesis of this model is that gas accumulation occurs at micropore [19]. The size of micropore is proportional to the size of filler. This explains why the filled material increases the solubility.

Figure 2.5 Micropore model for filled material [19].

(Reproduced with copyright permission of Society of Plastics Engineers.)

c02f005

Table 2.9 shows the effect of filler level on the CO2 gas absorption in high-density polyethylene (HDPE) at different melt temperatures: 149 °C and 177 °C at fixed pressure 18.6 MPa [19]. The high filler level will help to increase the gas solubility in HDPE. In addition, the high temperature causes the gas solubility decrease in the HDPE melt. This trend verifies the general trend of effects of temperature on the solubility in Figure 2.3.

TABLE 2.9 Solubility of CO2 (Weight Percentage) in HDPE with/without Talc

Source: Chen et al. [19], with permission of Society of Plastics Engineers.

The data in Table 2.10 are the results of the effect of filler level on the gas absorption in rigid PVC (RPVC) at different melt temperatures: 121 °C, 149 °C, and 177 °C at fixed pressure 18.6 MPa [19]. It is clear that the filled materials absorb more gas than the unfilled materials (see the data of filler lever = 0 in Table 2.9), and the gas absorption increases with increasing filler level. However, the dependence of gas absorption on the filler level is not linear. The trend for filled RPVC to absorb the CO2 matches the trend in Figure 2.3 with high pressure and low temperature for high solubility.

TABLE 2.10 Solubility of CO2 (Weight Percentage) in Rigid PVC Calcium Carbonate

c02t0292axi

Source: Chen et al. [19], with permission of Society of Plastics Engineers.

The CO2 gas absorption at different gas pressures is shown in Figure 2.6 for filled and unfilled HDPE samples [19]. As expected, the gas absorption is basically a linear function of the gas saturation pressure. It also shows the significant difference of gas absorption at the same pressure between filled and unfilled materials. It also verifies that filled material gains more gas than unfilled material does at different pressures. The results in Figure 2.6 explain why more cells are created with filled material. The fillers do not absorb gas; the polymer–filler interface is the only place that absorbs extra gas. The conclusions regarding to the sources of gas accumulation are listed below [19]:

Preexisting microgaps between polymer and fillers after compounding.

Convex areas on filler surface where higher interface energy is required for the polymer to fill in. There is a tendency for the polymer to be replaced by the gas after being melted.

If polymer–filler bonding is not strong enough, there is a tendency for the interface to be separated by the gas because the total surface energy of polymer and filler is smaller than the interfacial energy of the polymer–filler combination.

Figure 2.6 CO2 gas absorption as a function of pressure [19]. HDPE with/without filler.

(Reproduced with copyright permission of Society of Plastics Engineers.)

c02f006

Chen et al. [19] also discussed the results of gas absorption with and without shearing. Table 2.11 shows that gas absorption is different between the tests conditions with and without shearing. In addition, much higher gas absorption was observed in all the tests with shearing. It explains why there are some good results of gas dosing in high shear rate (high rotation speed of screw).

TABLE 2.11 CO2 Gas Absorption (Weight Percentage) at Different Filler Level and Shear, HDPE with/without Filler

Source: Wong et al. [18], with permission of Society of Plastics Engineers.

With all materials, the addition of fillers can improve the efficiency of the nitrogen added to the polymer. The most common filler with polypropylene is talc. As talc levels approach 20% or more, the N2 gas level will be in the range of 0.5% to 0.75%. Compared to talc and other fillers, glass fiber is a more efficient filler to reduce the gas level with good microcellular structure. The N2 gas level can be decreased to about 0.5% with glass-fiber-filled material.

Park and his group studied talc-enhanced PS foaming with CO2 gas as a blowing agent. At low weight percent of CO2 (2.1%) the onset time of cell nucleation decreases and the cell density increases with the higher talc content [22]. However, at higher CO2 content up to 4.0 weight percent, cell density is almost invariant with increase of talc content [22]. This conclusion is consistent with the research results on the extrusion foaming of the PS–CO2 system [23]. On the other hand, both the onset time of cell nucleation and the cell density are virtually unaffected by the mean size of talc particles [22]. Furthermore, Park found that the increasing CO2 content weakened the effect of the bubble expansion on the promotion of the cell nucleation since high CO2 gas may reduce the viscosity and the elasticity of the polymer–gas solution. It then suppresses the induction of the negative pressure around the talc particles, and it results in no promotion of cell generation around the expanding bubbles [22].

The most important observations for the talc-filled material foaming is that the generation of new cells propagated outward in the radial direction as the nucleated bubbles grew, and new cells grow even more with the increase of talc content at 2.1 weight percent of CO2 [22]. Park proposed a series of hypotheses for this observation [22]:

With addition of talc particle, the free energy barrier to initiate cell nucleation is reduced.

The rugged surface of talc particles may serve as the sites to trap CO2 as preexisting nuclei at the PS–talc interface. As the pressure drops, the critical radius of cell nucleation also decreases continually until it is smaller than this preexisted nuclei. Then, those preexisting nuclei will be activated and will start to grow.

Some cells grow and push the surrounding polymer gas solution outside the growing cells. As a result, local stretching of the polymer–gas solution may generate a negative pressure in some sections at the surface of the talc to promote the nucleation of new cells around the growing bubbles.

2.3.1.8 Comparison among Different Inert Gases.

There are a few published papers that shows the results of different inert gases as blowing agents in different materials. Chen et al. [24] did some tests with CO2, N2, and Ar gases in filled HDPE and PVC materials. Table 2.12 shows the gas absorption percentages of three different gases in HDPE with different filler levels. At the same temperature and pressure the gas absorption percentage for all gases in HDPE increases with the filler level increasing. However, only CO2 shows an obvious increase in the gas absorption with the filler level increasing. A similar result is from the RPVC test, which presents the following trend: The gas absorptions of all three gases increasing with the filler level become higher. However, the rate of RPVC solubility changing with different filler levels and different temperature is not as obvious as that of HDPE. There are other test results in Figure 2.7 showing that the shortest saturation time (which is the time for the horizontal line of the saturation pressure in the figure) among all three different inert gases tested for solubility is for CO2—that is, only about 35 minutes. The saturation time is about 60 minutes required for both N2 and Ar gases.

TABLE 2.12 Solubility of CO2, Ar, and N2 (Weight Percentage) in HDPE with Different Filler Percentages, 270 °C

c02t0322ayf

Source: Chen et al. [20], with permission of Society of Plastics Engineers.

Figure 2.7 Saturation time for 5% talc-filled HDPE with different gases.

(Courtesy of Trexel Inc.)

c02f007

The conclusion is that the solubility in polymer is generally much lower with N2 and Ar compared to CO2; viscosity reduction is also lower with N2 and Ar; cell density is similar at high saturation pressure, but much lower with N2 or Ar at low pressures; the cell size is smaller and density is higher with N2 and Ar; N2 and Ar are good candidates for high-density foams. However, CO2 can reduce the viscosity significantly because it has greater blowing agent concentration in the molten polymer than do other inert gases. It results in a greater reduction in density as well.

The acrylic material has the largest difference between the solubility for CO2 gas and N2 gas. At 177 °C of temperature, and 12.4 MPa of pressure the solubility of CO2 gas in acrylic is 4.25 weight percent. However, at 177 °C of temperature and 13.2 MPa of pressure, the solubility of N2 gas in acrylic is only 0.34 weight percent.

On the other hand, the weight percentage of gas to be added into the molten polymer for practical process is much less than the solubility measured in the batch process. The data in Table 2.13 show the gas dosing percentage differences among different materials with typical batch process and continue process. It is because the batch process may take many minutes (see Figure 2.7), or even hours, to saturate the material whereas the continue process must finish the gas dosing in the molten polymer in less than 1 min. However, the continue process accelerates the gas diffusion process by the high shearing, which is not easy to do in the batch process. Chen et al. [20, 21] proved that the shearing can speed up the gas diffusion process. The processing pressure and temperature also promote the gas solubility, which is discussed in Chapter 6 for processing and in Chapter 7 for equipment designing.

TABLE 2.13 Comparison of Laboratory Result and Industry Gas Percentage in the Polymer

aNA, not available.

The data in Table 2.13 provide typical operating levels for N2 in various materials. While it is always best to use the