彎管接頭注射模具設(shè)計【塑料注塑模具1模2腔液壓缸滑塊側(cè)抽芯含proe三維及22張CAD圖帶開題報告-獨家】.zip,塑料注塑模具1模2腔液壓缸滑塊側(cè)抽芯含proe三維及22張CAD圖帶開題報告-獨家,彎管,接頭,注射,模具設(shè)計,塑料,注塑,模具,液壓缸,滑塊側(cè)抽芯含,proe,三維,22,CAD,開題,報告,獨家 Microsyst Technol (2008) 14:1507–1514 DOI 10.1007/s00542-007-0533-8 TECHNICAL PAPER Micro injection molding for mass production using LIGA mold inserts Takanori Katoh ? Ryuichi Tokuno ? Yanping Zhang ? Masahiro Abe ? Katsumi Akita ? Masaharu Akamatsu Received: 13 July 2007 / Accepted: 16 December 2007 / Published online: 8 January 2008 。 Springer-Verlag 2007 1 3 Abstract Micro molding is one of key technologies for mass production of polymer micro parts and structures with high aspect ratios. The authors developed a commercially available micro injection molding technology for high aspect ratio microstructures (HARMs) with LIGA-made mold inserts and pressurized CO2 gasses. The test inserts made of nickel with the smallest surface details of 5 lm with structural height of 15 lm were fabricated by using LIGA technology. High surface quality in terms of low surface roughness of the mold inserts allowed using for injection molding. Compared to standard inserts no draft, which is required to provide a proper demolding, was formed in the inserts. To meet higher economic efficiency and cost reduction, a fully electrical injection molding machine of higher accuracy has been applied with dissolv- ing CO2 gasses into molten resin. The gasses acts as plasticizer and improves the flowability of the resin. Simultaneously, pressurizing the cavity with the gasses allows high replication to be obtained. Micro injection molding, using polycarbonate as polymer resins, with the aspect ratio of two was achieved in the area of 28 9 55 mm2 at the cycle time of 40 s with CO2 gasses, in contrast to the case of the aspect ratio of 0.1 without the gasses. 1 Introduction For recent growing interests in polymer micro parts and structures with high aspect ratios for microsystems and T. Katoh (&) · R. Tokuno · Y. Zhang · M. Abe · K. Akita · M. Akamatsu Sumitomo Heavy Industries Ltd., 2-1-1 Yatocho, Nishitokyo, Tokyo 188-8585, Japan e-mail: tkn_kato@shi.co.jp microelectronics, the economic production technologies have been desired. There are also a lot of applications in cellular mobile-phone system, automotive industries, medical engineering, just to mention a few field. Micro molding as a key technology for mass production of such polymeric micro parts and structures have been made efforts to overcome difficulties of good replication with shorter cycle time for improving productivity (Michaeli and Rogalla 1996; Piotter et al. 1999). Hot embossing including nanoimprinting and reaction injection molding have been developing for higher transcription of micro/ nano structures, in spite of longer cycle time (Heckele et al. 1998; Morton et al. 2006; Datta and Goettert 2007). On the other hand, the critical minimal dimensions which can be replicated in good transcription are mainly determined by the aspect ratio. However, injection molding is the most cost effective technologies for mass production, micro parts molded by the technology are limited that the aspect ratio is small, because of lower flowability of resins for plastic products. For aspect ratio less than one, the minimal structural details go down to the submicron scale as the case of mass production for CD and DVD by injection molding. In high-aspect-ratio microfabrication technologies, the best-known technique certainly is LIGA (Lithographie, Galvanoformung, Abformung) process (Becher et al. 1986). By means of deep-etch X-ray lithography and electroforming, tools with minute structural details in micron range, largest tool heights of up to several mm, and lateral wall roughness in the nm range (Ra = 70 nm typi- cally) can be produced (Guckel 1996). Plastic molding of high aspect ratio micro structures (HARMs) using LIGA made mold inserts has been investigated for past decade (Ruprecht 1995; Despa 1999; Piotter 2002; Yorita 2004). Some work lead to the commercial application of certain 1508 Microsyst Technol (2008) 14:1507–1514 Fig. 1 Compact SR-ring, ‘‘AURORA-2S’’ installed in SHI’s Tanashi works parts by dedicated micro molding machine with special equipment different from the industrial-based plastic molding technologies (Haverkamp et al. 1999; Wallabe et al. 2001). This paper provides the results of high aspect ratio micro injection molding combined with the commercially avail- able our injection molding machine with pressurized CO2 gasses into molten resin which can be improved the flow- ability of the resin and LIGA-made mold inserts fabricated by our compact synchrotron radiation (SR) source, AUR- ORA-2S (Fig. 1). 2 Fabrication of LIGA mold inserts We fabricated LIGA mold inserts with high aspect ratios using X-ray from a compact (footprints: 5 9 10 m2) SR source AURORA-2S built by Sumitomo Heavy Industries, Ltd. (SHI). This is one of SHI’s ‘‘home-made’’ compact electron storage rings, optimized for micro- and nano- fabrication (Hori and Takayama 1995; Zhang and Katoh 1996). The source is installed at SHI’s facility and operates with electron energy of 700 MeV and a routine stored current of 500 mA (a lifetime of 15 h). Several compact beamlines (e.g., BL11 and BL13) less than 5 m long for micro-fabrication already developed were adop- ted to carry out the X-ray deep lithography. Detailed descriptions for BL11 have been written elsewhere (Hi- rose 2000). The BL13 consists of (a) a front end connected to the SR source, an ultrahigh vacuum part (base pressure of 2 9 10-9 Torr), (b) an intermediate part with some filters (50 lm of beryllium, 100 9 10 mm2), also an ultrahigh vacuum part, and (c) an exposure chamber separated by a beryllium window (300 lm, 100 9 10 mm2). The spectrum of this beamline has a critical wavelength of 0.4 nm in the wavelength range below 0.7 nm (i.e., between 2 and 5 keV of X-ray energy). The photon flux per electron beam current on the resist surface was about 3.2 9 1010 photons/s mA mm2. The size of the SR X-ray at the resist surface was 100 9 7 mm2. The exposure chamber was purged with helium gas at 1 atm during lithography in order to prevent attenuation of the X-ray by N2 or O2 gases and to prevent damaging either the mask, which consists of several micron-thick gold absorber and a certain membrane (e.g., polyimide, SiC, and SiN), or the resist by heat load. We used commercially available sheets of PMMA as a resist, which thickness were 2.0 mm. The PMMA was exposed with the mask for mold inserts at an optimum dose of X- ray by vertically scanning at a speed of 1 mm/s. The exposed PMMA was developed with a GG developer (60 vol% 2-(2-butoxy-ethoxy) ethanol, 20% tetra-hydro-1, 4- oxazine, 5 vol% 2-amino-ethanol-1 and 15 vol% water) at 36。C. Successively, stopper liquid (80 vol% 2-(2-butoxy- ethoxy) ethanol, and 20 vol% water) was used at the same temperature for 10 min, followed by DI-water rinsing at 36。C for another 10 min. After fabricating the PMMA as templates for electroforming, high-strength and low-stress nickel electroforming was performed to produce LIGA mold inserts. The insert made of nickel had the size of 42 9 75 mm2 and the thickness of 0.5 mm (Fig. 2). The smallest surface details of the mold inserts was 5 lm with structural height of 15 lm, that is, aspect ratio of three. Many kinds of patterns were included in the inserts (e.g., line and space, dot, grid, cross, fluid channel, grating, optical waveguide and so on). Both convex and concave micro patterns were designed at line symmetry (a broken line in Fig. 2). It was demonstrated by SEM that the surface quality of the mold inserts was extremely fine and also demonstrated in a functional test of molded products that wear of the LIGA mold inserts did not occur even after more than 2,000 molding processes. As described later, in contrast to mold inserts used in conventional process with oblique side wall (typically 1–5。) for an easy demolding, the walls of the LIGA molds are without any inclination. Due to the extremely small roughness of the side walls (less than Ra = 70 nm), the demolding from mold inserts was very smooth as unexpected. 3 Experimental setup and procedure for injection molding Figure 3 shows a schematic of our injection molding sys- tem with commercially available AMOTEC (Asahi Molding Technology with CO2) equipments which consists of the dedicated injection molding machine with a heating cylinder, CO2 supplying system to the barrel (AMOTEC-1) and the cavity (AMOTEC-2) and the gaseous-sealing mold. Microsyst Technol (2008) 14:1507–1514 1509 Fig. 2 LIGA mold insert including various kinds of patterns (left) and SEM images (right). A broken line shows the position of a line of symmetry AMOTEC is a novel high-added technology involves dis- solving CO2 into molten resin (Yamaki et al. 2001; Shimoide et al. 2002; Akamatsu et al. 2003). Asahi Kasei Corporation holds the patents and we (SHI) have a license agreement on that. The CO2 dissolved into the spaces between the molecules of molten resin inside the barrel with a heating cylinder and acts as a plasticizer that improves the flowability of the resin. Using this property of CO2 gasses, good transcription can be achieved when molding products require ultra-fine replication, thin-walled replication and resin that are difficult to mold. Simulta- neously, pressurizing the cavity with CO2 allows high replication to be obtained, due to the flowability of the dissolved CO2 layer which does not form the solidified layer around the flow front of the melted resin. After molding, the CO2 whose contents is less than several per- cent of the total amount of the resin, evaporates and dose not change the properties of the resin. Last but not least, since the glass transition temperature (Tg) of the resin decreases as the CO2 pressure increases, molding can be performed with both the mold and the resin at lower tem- peratures. Less time is required for cooling so that the cycle time can be greatly shortened. Injection molding experiments were performed using a fully electric injection molding machine (Sumitomo  SE75DU, clamping force: 75 tons, maximum injection speed: 400 mm/s, injection unit: C160S, screw diameter: 22 mm) equipped with CO2 supply devices (MAC-100, Asahikasei Engineering Co., Ltd.). The mold platens were designed to hold two LIGA inserts, two cavities (size: 28 9 55 mm2, thickness: 1 mm) and cold runners with fan gates at the end. The material used for mold products was polycarbonate (PC) (AD5503, Teijin Chemicals) thermoplastic resin. The weight of the resin per shot was 5.7 g and the weight of the product (PC-plates of 25 9 55 mm2, 1 mmt) was 1.8 g. The polymer was injected into mold cavities at a pressure ranging from 150 to 200 MPa. The inside the mold and the heating cylinder were pressurized with CO2 at 2–6 MPa. The melt tem- perature in the feeding zone was maintained at 300。C. Temperature of the mold and the sprue were controlled by heaters and maintained at 130 and 80。C, respectively. The cycle time of the molding process was 40 s, which was included the cooling time for 24 s after filling stage of the resin into the cavity. The products molded during continuous production were used as evaluation samples. For evaluation of tran- scription, the replicated heights of the fine patterns of the samples were measured with laser microscope (VK8150, Keyence). The method and patterns for measurements are Fig. 3 Experimental setup for micro injection molding using AMOTEC with CO2 1510 Microsyst Technol (2008) 14:1507–1514 illustrated in Fig. 4. There were five convex test-cross patterns (see Fig. 4a) and five concave test-cross patterns in each sample along the flow direction of the resin. The test- cross patterns had the minimum features of 5 lm near the center. Since it is more difficult to fill the resin into the concave patterns of inserts normally, the convex test-cross patterns were mainly used for evaluation of replicated heights. 4 Results and discussion Figure 5 shows the typical results of the molding with CO2 (AMOTEC) and without CO2 (conventional). The LIGA inserts are also presented in the Fig. 5c. In the case of AMOTEC, good transcription with sharp and defined edges of like ‘‘SHI’’ characters with the micro mold-cavities at the corners can be clearly seen. The shape of molded products is almost the same as reverse images of the LIGA inserts. On the other hand, the result of the conventional molding shows poor transcription with round edges of the characters which dimensions are several tens of microns. These results actually show that the improved flow ability of the resin is effective to achieve good replication of fine patterns. The differences of the replicated height at the different CO2 conditions during molding were investigated (Fig. 6). For the conventional case, the replicated height of the convex test-cross pattern of minimum width of 5 lm was  Fig. 5 SEM images of the molded parts by the molding with CO2 (a) and conventional molding (b), respectively. (c) is a part of the insert formed by LIGA technology only 1 lm. For the CO2 case, two conditions were tested. One is the case that inside the mold and the heating cyl- inder were pressurized with CO2 at 2 and 4 MPa. The other one is the case that the mold and the heating cylinder were pressurized with CO2 at 4 and 6 MPa. Replicated heights could be improved drastically with CO2. Correspondingly, these results show that the replicated height increases as increasing the pressure inside the mold and the cylinder. Following results were corrected with the same condition of Fig. 6c. This condition was basic one in our experiments. The replicated height-to-width ratios of molded micro- structures were used to measure the quality of molding results. The measured results for the test-cross patterns, which have a minimum width of 5 lm, are shown in Fig. 7. The figure shows the comparison of replicated height and shape between molding with CO2 gasses and conventional molding during the mass production, respectively. Hori- zontal and vertical axis of the Fig. 7 shows the lateral scan range and replicated height measured by the laser micro- scope. The microscopic images of the molded products are shown in the same figure as insets. In the case of CO2 gases, the replicated height of the test-cross pattern went Fig. 4 Evaluation methods for replicated height; test-cross (a), dots (b) and L/S (c), respectively over 10 lm as described above. In the transcription, aspect ratio more than two could be achieved for the mass- Microsyst Technol (2008) 14:1507–1514 1511 Fig. 8 Position dependence of replicated height of the test-cross pattern at the flow-end, center and gate side of the cavity, respectively Fig. 6 Differences of replicated height at different conditions. Conventional without CO2 (a), inside the mold and the heating cylinder were pressurized with CO2 at 2 and 4 MPa (b), 4 and 6 MPa (c), respectively Fig. 7 Comparison of replicated height between molding with CO2 gasses (a) and conventional molding (b), respectively. Insets show the microscopic images of (a) and (b), respectively production molding. The shapes of the side walls of the molded test-cross pattern were almost perpendicular. The shape of the top of this pattern was somehow rough since the filling of the resin into the cavity with the test-cross pattern of 15 lm depth might not be sufficient. On the other hand, the replicated height of the conventional molding without CO2 gasses was less than 1 lm (0.4 lm, typically). The shape of the replicated pattern looked like a pancake without clear edges. Correspondingly, in the microscopic image of the molded result of the conventional molding, the replicated edges of the test-cross pattern could not be clearly seen. Moreover the effects of the pressurized CO2 for the molding process were studied. First of all, the position dependence of the replicated height whole transcription area of the insert was investigated. The replicated heights  of the test-cross pattern at near side of the gate, center of the cavity and near side of the flow-end of the resin were measured both the CO2 case and the conventional case. These results are shown in Fig. 8. In the case of conven- tional molding, the replicated height at the near side of the gate was relatively higher than that of the height at the near side of the flow-end. That might be understand the solidi- fied layer of the resin could be easily grown at the flow-end of the resin far from the gate at where the resin was still maintained at higher temperature. In the case of AMOTEC, good replication whole cavity from the gate side to the flow-end can be achieved due to the improved flow ability of the resin even on the flow-end by the pressurized CO2. The replicated heights are almost the same at the all rep- lication area in the cavity. That is also one of advantages of pressurized CO2. Secondly, we checked flow direction dependence of replicated height of the test-cross patterns (Fig. 9). An, Bn and Cn (n; 1,2) denote the measured positions of the rep- licated height in the test-cross pattern, corresponding to the position indicated in the inset of the test-cross pattern in this figure. Flow direction of the resin is from C2 to A2, as also shown in the inset. Figure 9a shows the results of around the center position with about 5 lm width of test- cross pattern both conventional and AMOTEC cases. Fig- ure 9b shows the results of outside position with about 20 lm width of the test-cross pattern. The effectiveness of AMOTEC is obvious at a glance. The transcription for conventional molding became better as broadening pattern width. Moreover, replicated heights at the position (A1 and A2) against the resin-flow were a little bit higher than that at the position (C1 and C2) along the flow direction. On the other hand, the transcription for AMOTEC was almost uniform over the test-cross pattern. There were not noticeable differences of replicated height for flow direc- tion between center and out side position. Furthermore, we investigated effects of pressurized CO2 for the replication of fine patterns on the 5–100 lm size of 1512 Microsyst Technol (2008) 14:1507–1514 Fig. 9 Flow direction dependence of replicated height of test-cross pattern, around center with the width of about 5 lm (a) and out side with the width of about 20 lm (b). The position and direction of the resin-flow are denoted in the inset, respectively round, square and striped pattern (lines and spaces). While there were no differences of replicated height more than 50 lm patterns between conventional and AMOTEC, there were great differences for less than 10 lm patterns. For example, the replicated heights for 10 lm square-pattern were less than 1 lm by conventional molding and about 5 lm by AMOTEC molding, respectively (Fig. 10). In the case of striped pattern, the replicated heights were less than 5 lm by conventional molding and more than 10 lm by AMOTEC molding, respectively. These results show that advantageous effects of pressurized CO2 appear when the molded patterns go down 10 lm of scale. However we have achieved high-aspect-ratio micro injection molding on the basis of mass production and found the advantages of AMOTEC, some defects which might be concerned with the process of filling or demold- ing could be observed during our experiments. Figure 10 shows the SEM images of replicated patterns for 10 lm square mold at both conventional (a) and AMOTEC (b), respectively. It is clear that the replicated height of AMOTEC case in Fig. 10 is much higher than that of conventional case as described above. By checking over th