外文翻譯--一種改善微孔注射液型注塑件表面質(zhì)量的新方法 英文版

a r t i c l e i n f o Article history: Received 25 October 2010 Received in revised form 12 January 2011 Accepted 14 January 2011 Available online 21 January 2011 Keywords: Microcellular injection molding Plastic foaming Swirl-free surface a b s t r a c t Microcellular injection molding is the manufacturing method used for producing foamed plastic parts.Microcellular injection molding has many advantages including material, energy, and cost savings as well as enhanced dimensional stability. In spite of these advantages, this technique has been limited by its propensity to create parts with surface defects such as a rough surface or gas flow marks. Methods for improving the surface quality of microcellular plastic parts have been investigated by several researchers. This paper describes a novel method for achieving swirl-free foamed plastic parts using the microcellular injection molding process. By controlling the cell nucleation rate of the polymer/gas solution through material formulation and gas concentration, microcellular injection molded parts free of surface defects were achieved. This paper presents the theoretical background of this approach as well as the experimental results in terms of surface roughness and profile, microstructures, mechanical properties, and dimensional stability. Introduction The commercially available microcellular injection molding process (a.k.a. the MuCell Process) consists of four distinctive steps, namely, gas dissolution, nucleation, cell growth, and shaping 1. In the gas dissolution stage, polymer in the injection barrel is mixed with supercritical fluid (SCF) nitrogen, carbon dioxide, or another type of gas using a special screw which is designed to maximize the mixing and dissolving of the gas in the polymer melt. During injection, a large number of nucleation sites (orders of magnitude higher than conventional foaming processes) are formed by a rapid and substantial pressure drop as the polymer/gas solution is injected into the mold cavity, thus causing the formation of cells (bubbles). During the rest of the injection molding cycle, cells continue to grow to fill and pack out the mold and subsequently compensate for the polymer shrinkage as the material cools inside the mold. The cell growth is driven by the amount and spatial distribution of the dissolved gas. The cell growth is also controlled by processing conditions such as melt pressure and temperature as well as material properties such as melt strength and gas solubility. Finally, the shaping of the part takes place inside the mold until the mold opens allowing the part to be ejected. Since the microcellular injection molding process was invented, there have been numerous studies on process, material, and technical developments aimed at materializing the full process potential. According to previous studies 1-5, microcellular injection molding offers a number of advantages such as cost savings, weight reduction, ease in processing due to low viscosity, and outstanding dimensional accuracy. Due to these advantages, the microcellular injection molding process has been used in many industries such as automotive, electrical goods, and home appliances using a broad range of thermoplastics. Despite these advantages, however, the surface imperfections associated with microcellular injection molded partsdsuch as unique gas flow marks, referred to as swirl marks throughout this paper, and a lack of smoothnessdstill remain one of the main drawbacks surrounding microcellular injection molding. In order to eliminate or reduce these surface imperfections there have been several studies attempted, as reported in Refs. 6-14. Some researchers have focused on temperature modification of the mold surface to improve the surface quality of microcellular injection molded parts 6-8. With polymeric foam, it was found that bubbles forming at the advancing melt front are first stretched by the fountain flow behavior toward the mold surface and subsequently dragged against the mold wall causing swirl marks 9. During the filling stage of polymer melts, keeping the mold wall temperature high enough for bubbles at the mold surface to beeliminated improves the surface quality of microcellular injection molded parts. By controlling the mold temperature rapidly and precisely using mold temperature control units or other kinds of thermal or surface heating devices, microcellular foamed plastics with glossy and swirl-free surfaces can be produced. There have also been efforts to eliminate the swirl marks on microcellular injection molded parts without any mold temperature controller. In particular, it was proposed that inserting an insulator onto the mold wall might help keeping the interface temperature between the mold and the polymer melt high. This technique basically yields the same result as temperature modification of the mold 10. Thermal analysis and experimental results prove that the addition of an insulator layer on the mold can improve the surface quality of microcellular injection parts 11. Another method of producing parts with an improved surface quality leads to a microcellular co-injection molding process 12. In this technique, a proper amount of solid skin material is injected prior to the injection of a foaming core material. This can yield a sandwiched (solid skinefoamed coreesolid skin) structure with a surface finish similar to a conventionally molded component while partially maintaining the advantages of microcellular injection molding. Another approach for improving the surface quality of microcellular injection molded parts is the gas counter pressure process 13,14. In this process, a high-pressure gas is injected into the mold prior to the polymer/gas solution to suppress cell nucleation and bubble growth while the polymer/gas solution is being injected into the mold cavity. Toward the end of injection, counter gas pressure is released and bubbles begin to form within the cavity. Since a majority of the part surface is already solidified, gas flow marks are eliminated. In spite of these efforts to improve the surface quality, there have been difficulties in applying the microcellular injection molding process in industries requiring parts with high surface qualities because these techniques entail additional equipment which results in high costs or maintenance. There have been no reported studies on improving the surface quality of microcellular injection molded parts without any additional equipment or modification to existing equipment. This paper proposes a novel approach to improve the surface quality of microcellular injection molded parts by controlling the cell nucleation rate. In this study, the cell nucleation rate was dramatically lowered or delayed by controlling the degree of supersaturation so that cell nucleation was delayed during the filling stage. After the polymer/gas solution volumetrically filled the mold cavity, intentionally delayed nucleation occurred and bubbles formed in the polymer matrix, except on the surface where the material had already solidified upon touching the mold surface. Theoretical background and experimental results are described in this paper. Microstructure, surface profile, surface roughness,mechanical properties, and dimensional stability are also investigated in this study. 2. Theoretical 2.1. Nucleation theory for polymeric foams In polymeric foams, nucleation refers to the initial stage of the formation of gas bubbles in the polymeregas solution. For nucleation, gas bubbles must overcome the free energy barrier before they can survive and grow to macroscopic size 15. According to classical nucleation theories 16-18, the nucleation rate is controlled by the macroscopic properties and states of the polymer and gas such as solubility, diffusivity, surface tension, gas concentration, temperature, and the degree of super saturation. One representative equation for the nucleation rate of polymeric foams was reported by Colton and Suh 19,20. In addition to the mathematical representation, they also verified their nucleation theory experimentally for a batch foaming process using a high pressure vessel. The nucleation equation for microcellular foams dominated by the classical nucleation theory 16e18 can be expressed as where N is the nucleation rate, f is the frequency of atomic molecular lattice vibration, C is the concentration of gas molecules, k is the Boltzmanns constant, T is the absolute temperature, and is the activation energy barrier for nucleation. According to previous studies 19,20, the nucleation rate of polymeric foams is composed of two components: a homogeneous term and a heterogeneous term. The activation energy for homogeneous nucleation is given by where g is the surface energy of the bubble interface and is assumed to be the gas saturation pressure. More precisely, where Pr is the pressure that is exerted in a high pressure vessel and Pr is the pressure of the supersaturated vapor in the sample 16. That is, DP is the pressure difference between the pressure that is applied to the sample and the pressure of the supersaturated vapor in the sample. When the pressure that saturates the gas in a high pressure vessel is suddenly released to trigger the so-called thermodynamic instability by rendering the sample into the supersaturated state, Pr becomes 1 bardso low compared to Pr that DP can be approximated as Pr. On the other hand, the activation energy for heterogeneous nucleation is affected by a geometric factor that depends on the contact (wetting) angle between the polymer and the particle and can be expressed as (3a) (3b) where f(q) is a geometric factor that is dependent upon the contact angle, , of the interface between the polymer and a second phase, and has values of less than or equal to 1. For a typical wetting angle of around on the interface between a solid particle and the polymer melt, the geometric factor is 2.7X, suggesting that the energy barrier for heterogeneous nucleation can be reduced by three orders of magnitude with the presence of an interface 20,21. 2.2. Nucleation theory for microcellular injection molding In the batch foaming process, the theory of Colton and Suh was verified by their experiments. Due to the large difference between the pressure exerted in a high pressure vessel and the pressure of the supersaturated vapor in the sample, the gas pressure dissolved in the polymer, the in the Gibbs free energy equation, can be approximately assumed to be the saturation gas pressure. The assumption that is the gas saturation pressure is fairly reasonable in a batch foaming process although the can still have an error of about 30-40% due to overestimation as reported in a previous study 15. The nucleation theory by Colton and Suh is a simplified form derived and modified from classic nucleation theories 16-18 and is generally adequate for the batch foaming process. However, there is a need for this theory to be modified in cases of microcellular injection molding and extrusion systems because cannot be directly controlled and measured. To predict nucleation in microcellular injection molding and extrusion processes more precisely, this paper proposes a cell nucleation theory of a different form, which includes a term for the degree of supersaturation because it is a directly controllable factor. To avoid misestimating , and to consider the degree of supersaturation, a more proper activation energy equation for nucleation can be derived from the following equation 16,17 （ 4） where is the radius of a characteristic droplet, and the W. Thomson equation （ 5） where is the pressure of the saturated vapor (i.e., the equilibrium pressure), R is the universal gas constant, M is the molar mass, and is the density. These equations yield （ 6） which can be alternatively expressed as (7) whereis the molecular density of the bulk liquid, and S(=) is defined as the degree of supersaturation. Thus, the activation energy equation (cf. Equation (2) for nucleation in the microcellular injection molding process can be given by （ 8） Hence it can be stated that the activation energy for nucleation is inversely proportional to the square of the natural logarithm of the supersaturation degree. In the microcellular injection molding process, the polymer/gas solution becomes a metastable supersaturation solution when it is injected into the mold cavity. This is because the amount of gas able to be dissolved in the polymer in the presence of a rapid pressure drop is less than the gas amount originally dissolved in polymer melts. In particular, assuming the air in the cavity is properly vented, the pressure at the advancing melt front is at the atmospheric pressure. The solubility of a gas in a polymer at atmospheric pressure and processing temperature can be obtained by an Arrhenius-type expression with regard to temperature 22 (9) where is the solubility of the gas in the polymer at standard temperature and pressure conditions (298 K and 1 atm). The parameter DHs is the molar heat of sorption, and Tmelt is the polymer melt temperature. Thus, the degree of supersaturation is given by (10) where is the gas dosage which can be controlled by the supercritical fluid (SCF) supply system. The heat of sorption, , of various polymer/gas systems at standard temperature has been studied and summarized in many previously published studies. In order to obtain the degree of supersaturation for a polymer/gas solution in the microcellular injection molding process, one has to either measure the solubility of the gas in the polymer at standard temperature and pressure or consult published data on the solubility of the gas in the polymer. Then, the activation energy barrier for nucleation in Equation (8),G*, can be obtained based on the calculated degree of supersaturation and the surface energy of the bubble interface, . Given the activation energy barrier and the frequency factor, f, the nucleation rate (expressed in Equation (1) can then be calculated.The estimate of the surface energy of the bubble interface and the frequency factor is discussed below. In microcellular injection molding, the polymer/gas solution can be treated as a liquid mixture. Thus, the surface energy of the bubble interface, g, can be expressed as 23,24 （ 11） where is the surface energy of the polymer, are the densities, and is the weight fraction of gas. In addition, a frequency factor for a gas molecule, f, in Eq. (1) can be expressed as 24-26 (12) where z is the Zeldovich factor, which accounts for the many clusters that have reached the critical size, , but are still unable to grow to sustainable bubbles. The parameter b is the impingement rate at which gas molecules collide with the wall of a cluster. The parameter can be used as a correction factor and is determined experimentally. Once the nucleation rate as a function of the degree of supersaturation is obtained, one can control the gas (SCF) content in the polymer melt to control or delay the onset of cell nucleation so that no bubble will form at the advancing melt front during the injection filling stage, thus, allowing microcellular parts with solid, swirl-free surface to be injection molded. 3. Experimental 3.1. Materials The material used in this study was an injection molding grade low density polyethylene, LDPE (Chevron Phillips Chemical Company, Texas, USA). It has a melt index of 25 g/10 min and a density of 0.925 g/. To confirm the theory for improving surface quality by controlling the degree of supersaturation, a random copolymer polypropylene (PP)was also used in this study. The PP used in this study was Titanpro SM668 (Titan Chemicals Corp., Malaysia), with a melt flow index of 20 g/10 min and a density of 0.9 g/. Both materials were used as received without any colorant, fillers, or additives. Commercial grade nitrogen was used as a physical blowing agent for the microcellular injection molding trials. 3.2. Microcellular injection molding In this study, an Arburg 320S injection molding machine (Arburg,Germany) was used for both the solid conventional and microcellular injection molding experiments. The supercritical fluid (SCF) supply system used in this study was the S11-TR3 model (Trexel, Woburn,MA, USA). The total gas dosagewas controlled by adjusting the gas injection time, t, and the gas injection flowrate,m_ g. A tensile test mold, which produces tensile test specimens that meet the ASTM D638 Type I standards, was used for this experiment. For injectionmolding of both LDPE and PP tensile test specimens, nozzle and mold temperatures were set at 221 and 25 , respectively. The cycle time was 40 s. An injection speed of 80 cm3/s was employed. In this study, six different gas dosages (concentrations) were used for injection molding of LDPE as shown in Table 1. Also, four different gas dosages were employed for microcellular injection molding of PP. The supercritical fluid was injected into the injection barrel at 140 bar pressure to be mixed with the polymer melts in this experiment. The weight reduction of foamed versus solid plastic partswas targeted at 8 _ 0.5% for each specimen. For the conventional injectionmolding experiment, the shot size of 20.2 and a packing pressure of 800 bars were employed for 6 s. For the microcellular injection molding experiments, the shot size of the polymer melt was 19.2 and the packing stage was eliminated. 3.3. Analysis methods To analyze the surface roughness of the molded tensile bar specimens, a Federal Surfanalyzer 4000 (Federal Product Corporation, RI, USA)was used. The surface roughnesses of conventional and microcellular injection molded parts were evaluated at three locations shown in Fig. 1 and the averaged surface roughness based on measurementsdone at all three locationswas recordedandreported. The cutoff, drive speed, and drive length for the test were 0.75 mm, 2.5 mm/s, and 25 mm, respectively. For each process condition, ten specimens and three points on each specimen were tested. In addition to the surface roughness, swirl marks commonly observed in microcellular injection molded samples can also be clearly revealed by a 3-D surface profiler. Zygo NewView (Zygo Corporation, CT, USA), a non-contact 3-D surface profiler, was employed to examine the surface profile of injection molded parts in this study using a scan distance of 10 mm. A JEOL JSM-6100 scanning electron microscope with an accelerating voltage of 15 kV was employed for observing the microstructures of the foamed parts. To observe the cross section of the microcellular injection molded parts, test specimens were frozen by liquid nitrogen and subsequently fractured. Representative images of each process condition were selected and cell sizes and densities were analyzed. A UTHSCSA Image Tool was employed as the image analysis software to evaluate cell densities and sizes. A MTS Sintech 10GL screw driven machine was used to test the mechanical properties of the molded specimens, including the yield stress, strain at break, modulus, and ultimate stress. Ten specimens for each condition were tested. The tensile test speed was 50.8 mm/min. The schematic of the ASTM tensile bar and locations of the various analyses are shown in Fig. 1. To test the dimensional stability of the injection molded specimens, a dial caliper made by Mitutoyo was used. Dimensions of the mold cavity were first measured and then the injection molded parts were measured and compared with the actual dimensions of the mold cavity. 4. Results and discussion 4.1. Surface profile and roughness measurement As a visual illustration, Fig. 2 shows the representative injection molded low density polyethylene (LDPE) parts. To better reveal the surface quality in the photo, 5wt% colorant was added to the material although the same surface quality was obtained without colorant. As anticipated, the conventional solid injection molded part (Fig. 2(a) has a glossy and flawless surface. On the other hand, the typical microcellular injection molded part (Fig. 2(b) produced with a moderate or high gas concentration has a lusterless surface due to swirl marks. A microcellular plastic part (Fig. 2(c) molded with a carefully controlled gas concentration of 0.173 wt% or less has a high quality surface finish comparable to the conventional solid injection molded part. No swirl marks or defects are observed on the surface. For polypropylene (PP), it was observed that microcellular injection molded parts produced with a gas amount of 0.173 wt% or less also exhibited swirl-free and shiny surfaces. Fig. 3 shows the representative surface profiles of injection molded parts made from LDPE. For this material, microcellular plastic parts molded with a gas amount of 0.173 wt% or lessdi.e., MC1-L, MC2-L and MC3-L (cf. Table 1)dhave a surface quality equivalent to that of conventional injection molded parts. For those parts, any swirl marks on the surface were invisible to the naked eye (cf. Fig. 2). Also, the lack of smoothness and brightness typically found in microcellular injection molded parts were imperceptible to touch or to the human eye. Microcellular injection molded PP specimens show similar results to LDPE and thus are not shown. MC1-P and MC2-P have swirl-free surfaces without any noticeable surface defects. On the other hand, MC3-P and MC4-P, which have a higher gas amount, show swirl marks on their surfaces and have a rough surface like one might associate with typical microcellular injection molded parts. As shown in Fig. 3(a), the surface of the conventional LDPE injection molded part has streaks on it due to either machining marks or normal wear induced by the repeated flow of polymer melt over its surface. Except for these streaks, it has a smooth and even surface profile. Fig. 3(b) shows the representative surface profile of the microcellular LDPE injection molded with a gas concentration of 0.173 wt% or less. It has a smooth and even surface profile, comparable to that of conventional injection molded parts. On the other hand, swirl marks and a rough surface are evident on the surface of microcellular LDPE plastics molded with nitrogen over 0.173 wt% as shown in 2-D and 3-D surface maps of Fig. 3(c). The area shown as black in Fig. 3(c) is the domain in which the surface profiler could not measure the height difference due to an uneven surface that exceeds the pre-set measurement limits (i.e,10 mm). The surface profiles of conventional and microcellular PPinjection molded parts are very similar to their LDPE counterparts. As mentioned previously, surface imperfections of the microcellular injection molded parts, such as swirl marks and a lack of smoothness, are caused by bubbles forming at the advancing melt front during the mold filling stage. These bubbles are transported toward the mold surface by fountain flow behavior and are subsequently stretched and dragged against the mold wall by the incoming polymer melt 6,9. The use of a small amount of gas in the microcellular injection molding process results in a reduction of supersaturation that leads to a great increase in the activation energy needed for nucleation. An elevated activation energy causes a large drop in the nucleation rate, which retards the nucleation and subsequent growth of bubbles so that the probability of bubbles being formed at the advancing melt front, and subsequently being dragged and stretched against the mold wall,dramatically decreases. Table 2 shows the theoretical nucleation rate of microcellular injection molded parts based on the nucleation theory containing the degree of supersaturation presented previously. For the materials and experiments under consideration, constant zb is assumed to be 1.0 _ 1011 s_1 m_2. The solubility of nitrogen in LDPE was obtained from reference 27. The solubility of nitrogen in PP is known to be around 35% more than that of LDPE according to previous studies conducted by Park and his colleagues 28-30, and unpublished research results. Thus, given the same amount of gas, the degree of supersaturation for PP/N2 becomes about 25% lower than that for LDPE/N2.To take into account the effect of temperature on the surface tension, the Guggenheim equation reported in reference 31 was used for the calculation of surface tension with regard to temperature. The density of nitrogen was obtained by Engineering Equation Solver (EES) using the property data of real gases 32. As shown in Table 2, theoretical nucleation rates of MC1-L, MC2-L, MC3-L, MC1-P, and MC2-Pdi.e., microcellular injection molded parts not having swirl marks on the surfacedare significantly small. On the other hand, as the degree of supersaturation increases with a gas dosage beyond 0.173 wt%, the activation energy for nucleation decreases while the nucleation rate increases sharply. Dramatic changes in the nucleation rate and activation energy as a function of gas dosage are plotted in Fig. 4. As the amount of gas increases incrementally, the nucleation rate increases significantly and, at a certain point, numerous cell nuclei form enough bubbles to cause swirl marks on the surface of the microcellular injection molded parts. In Table 2, it is believed that the difference in the nucleation rate and cell size between LDPE and PP is primarily due to differences in the surface tension of polymer melts and solubility of nitrogen in polymers. Fig. 5 illustrates the mechanism of cell nucleation when the polymer/gas solution is injected into the mold cavity. As the polymer/ gas solution is injected into the mold, the pressure drops along the flow direction toward the advancing melt. At this point, the pressure is at atmospheric pressure. As the pressure decreases, the solubility of nitrogen in the polymer decreases and the degree of supersaturation increases so that nucleation starts to occur. Theoretically, the starting point for cell nucleation to occur is at the point where the degree of supersaturation becomes greater than unity. However, the cell nucleation rate remains too small and thus is negligible at this moment. To become an issue for the surface quality, a significantly high cell nucleation ratedwhich can cause swirl marks on the surface of foamed plasticsdshould be approximately around to . This cell nucleation rate range corresponds to a degree of supersaturation between 6.1 and 7.0 for LDPE and between 4.2 and 4.7 for PP. As shown in Fig. 5(a), as the gas amount decreases, the starting point for cells to nucleate is delayed. It is conceivable that, below a certain degree of supersaturation, cell nucleation does not occur before the melt front of the polymer/gas solution touches the mold wall. Once a solid skin layer solidifies, a swirl-free surface is formed. After that, cells start to nucleate and grow in the hot polymer core until the material solidifies within the chilled mold. By controlling the degree of supersaturation, and thus the cell nucleation rate, bubble formation on the surface of the foamed part can be delayed or suppressed. This is the reason why microcellular injection molded LDPE parts with a nitrogen content of 0.173 wt% or less have much smoother surfaces than those with a higher gas dosage (cf. Fig. 6). In Fig. 6, Ra, Rq and Ry are the average, root-meansquare, and maximum peak-to-valley height roughnesses, respectively. MC4-L, MC5-L, and MC6-L have 75e101% rougher surfaces, on average, than conventional injection molded parts. In addition, these microcellular injection molded parts show a wide deviation of roughness which demonstrates the difficulty of controlling surface roughness. On the other hand, samples injection molded with a nitrogen of 0.173 wt% or less have much smoother surfaces than those with moderate or high gas concentrations, although they are still 11e20% rougher than conventional injection molded parts. When the standard deviation of the surface roughness is considered, swirl-free microcellular injection molded parts have fairly comparable surfaces to those of conventional molded parts. Test results for PP show similar patterns to those of LDPE, as shown in Fig. 6(b). 4.2. Microstructures Table 3 and Fig. 7 show the cell diameters and densities of microcellular injection molded parts versus gas amount. The cell diameters and densities were analyzed and calculated based on the SEM images of cell structures shown in Fig. 8 and using imaging analysis software (UTHSCSA Image Tool). As gas amount increases, average cell diameter decreases and cell density increases. It is believed that cells nucleate more and earlier during the filling stage due to the increase in cell nucleation rate with a higher degree of supersaturation. For microcellular injection molded LDPE parts with no swirl marks on their surfaces, the average cell diameter is over 550 mm. On the other hand, microcellular injection molded LDPE parts having swirl marks can have average cell diameters of less than 300 mm and cell densities about 8 times more than their swirl-free counterparts. For PP, microcellular injection molded parts with bubbles of diameters 149 mmand celldensities of27,000 percubic centimeteron averagewere achieved without swirl marks on their surfaces. MC1-P had bigger and sparser bubbles than MC2-P and had a huge cell size distribution. On the other hand, MC3-P and MC4-P had small and denser bubbles inside. Also, bubble sizes of MC3-P and MC4-P were much more regular than the MC1-P specimen. MC2-P had smaller, more even, and denser bubbles thanMC1-P although it had relatively bigger and sparser bubbles than the MC3-P or MC4-P specimen. From these microstructure analyses, it can be concluded that typical microcellular plastics molded with moderate or high gas concentrations have smaller and denser bubbles than those molded with lower gas contents. For swirl-free microcellular injection molded parts, there is a trade-off for finer and denser cell morphology since cell nucleation occurs more slowly and later during the filling stage than typical microcellular injection molding. With lesser number of nuclei, the molded parts tend to have larger and sparser bubbles. While the cells were large (due to a slower nucleation rate) compared to those in typical microcellular injection molded parts, those cells are still in the sub-mm range. Nonetheless, it should be pointed out that typical microcellular injection molded parts normally have a cell size of 100 mm or less and a cell density around 106 cells/. In addition, the cell morphology can affect both the mechanical properties and the dimensional stability. Further research on both will be introduced in the following sections. The effect of gas foaming on crystallization and crystal morphology as well as mathematical modeling and numerical prediction in microcellular injection molding (cf. 33-35) require further studies. 4.3. Mechanical properties Fig. 9 shows the relative mechanical properties of microcellular injection molded parts as compared to conventional injection molded parts. For LDPE, the strain at break of swirl-free microcellular injection molded parts was noticeably lower than the rest of microcellular parts that have smaller, denser, and more uniform bubbles associated with higher gas concentrations and thus a more uniform and longer elongation before fracture. For yield stress, no noticeable change with regard to different gas amount was found. Interestingly, swirl-free microcellular injection molded parts have better moduli and ultimate stresses than typical microcellular injection molded parts. This is because decreasing the blowing agent increases the thickness of the skin layer which is not foamed such that some of the mechanical properties of the microcellular injection molded parts are increased 36-38. The PP specimens behaved similarly to the LDPE specimens for the mechanical properties shown. The moduli of MC1-P and MC2-P were significantly greater than those of MC3-P and MC4-P. Also, injection molded parts with low gas contents had higher ultimate stresses than their counterparts. Strains at break of swirl-free microcellular injection molded parts were much less than those of typical microcellular injection molded specimens. It is evident that the cell morphologies of microcellular plastics strongly affect their mechanical properties. It is also likely that the strain at break is strongly affected by the cell morphology. A microcellular injection molded part with more tiny bubbles and a uniform bubble distribution tends to exhibit greater strain at break than the ones with fewer but larger bubbles. 4.4. Dimensional stability The dimensional stability of injection molded parts versus gas concentrations are shown in Fig. 10. As gas content increases, the shrinkage of microcellular injection molded parts distinctly decreases. Changes in the thickness direction are greater than the flow and transverse directions for both LDPE and PP specimens. If cell nucleation is too low, the benefit of improvement in dimensional stability may not be fully realized. The shrinkage of a microcellular LDPE part molded with 0.173 wt% of nitrogen was closest to a conventional injection molded part among the six different batches of specimens studied. For PP, similar patterns on shrinkage were observed. Due to the smaller and denser bubbles of PP as compared with LDPE specimens, shrinkage in every direction was lower than that of the microcellular injection molded LDPE. 5. Conclusions By reducing the degree of super saturation, the activation energy for cell nucleation will increase, thereby greatly reducing the cell nucleation rate. This retards cell nucleation during the mold filling stage, thus preventing bubble formation on the melt front of the polymer/gas solution, resulting in swirl-free microcellular injection molded parts. This research demonstrates the feasibility of this approach both theoretically and experimentally. The swirl-free microcellular injection molded parts produced in this study had comparable surface qualities to conventional injection molded parts with the added bonus of an 8 percent reduction in weight. Surface imperfections such as lack of brightness, gas flow marks, and rough surfaces which have long been considered unavoidable shortcomings of the microcellular injection molded process were eliminated without any additional equipment or equipment modification, as required by other methods. The microstructure of typical microcellular injection molded parts show smaller and denser bubbles than swirl-free microcellular injection molded parts. It can be seen that, as gas amount increases gradually, injection molded parts possess a reduced bubble size and an increased bubble density while maintaining surface quality, until a certain degree of super saturation or gas dosage is reached. When this threshold is reached, however, the cell nucleation rate is high enough that swirl marks begin to appear on the surface. How precisely to maintain the desirable cell morphology that is, smaller and denser bubbles for a swirl-free surface for microcellular injection molded parts requires further investigation. The strain at break of a swirl-free microcellular injection molded part is noticeably lower than that of a typical microcellular injection molded part. On the other hand, a swirl-free microcellular injection molded part exhibits a higher modulus and ultimate stress than a typical microcellular injection molded part. For dimensional stability, the shrinkage of microcellular i