喜歡這套資料就充值下載吧。。。資源目錄里展示的都可在線預(yù)覽哦。。。下載后都有，，請放心下載，，文件全都包含在內(nèi)，圖紙為CAD格式可編輯，【有疑問咨詢QQ:414951605 或 1304139763】 1Microsystem Technologies 10 (2004) 531535 _ Springer-Verlag 2004DOI 10.1007/s00542-004-0387-2Replication of microlens arrays by injection moldingB.-K. Lee, D. S. Kim, T. H. KwonB.-K. Lee, D. S. Kim, T. H. Kwon (&)Department of Mechanical Engineering,Pohang University of Science and Technology (POSTECH),San 31, Hyoja-Dong, Nam-Gu, Pohang, 790-784, Koreae-mail: thkwonpostech.ac.krAbstractAbstractAbstractAbstractInjection molding could be used as a mass production technology for microlens arrays. It is of importance, and thusof our concern in the present study, to understand the injection molding processing condition effects on the replicability ofmicrolens array profile. Extensive experiments were performed by varyingprocessing conditions such as flow rate, packingpressure and packing time for three different polymeric materials (PS, PMMA and PC). The nickel mold insert of microlensarrays was made by electroplating a microstructure master fabricated by a modified LIGA process. Effects of processingconditions on the replicability were investigated with the help of the surface profile measurements. Experimental resultsshowed that a packing pressure and a flow rate significantly affects a final surface profile of the injection molded product.Atomic force microscope measurement indicated that the averaged surface roughness value of injection molded microlensarrays is smaller than that of mold insert and is comparable with that of fine optical components in practical use.1 1 1 1IntroductionIntroductionIntroductionIntroductionMicrooptical products such as microlenses or microlens arrays have been used widely in various fields of microoptics,optical data storages, bio-medical applications, display devices and so on. Microlenses and microlens arrays are essentialelements not only for the practical applications but also for the fundamental studies in the microoptics. There have been severalfabrication methods for microlenses or microlens arryas such as a modified LIGA process 1, photoresist reflow process 2,UV laser illumination 3, etc. And the replication techniques, such as injection molding, compression molding 4 and hotembossing 5, are getting more important for a mass production of microoptical products due to the cost-effectiveness. Aslong as the injection molding can replicate subtle microstructures well,itis surely the most cost-effective method in the massproduction stage dueto its excellent reproducibility and productivity.In this regard,itis of utmost importance to check the injection moldability and to determine the molding processingcondition window for proper injection molding of microstructures. In this study, we investigated the effects of processingconditions on the replication of microlens arrays by the injection molding. The microlens arrays were fabricated by a modifiedLIGA process, which was previously reported in 6, 7. Injection molding experiments were performed with an electroplatednickel mold insert so as to investigate the effects of some processing conditions. The surface profiles of molded microlensarrays were measured, and were used to analyze effects of processing conditions. Finally, a surface roughness of microlensarrays was measured by an atomic force microscope (AFM).2 2 2 2MoldMoldMoldMold insertinsertinsertinsert fabricationfabricationfabricationfabricationMicrolens arrays having several different diameters were fabricated on a PMMA sheet by a modified LIGA process 6.This modified LIGA process is composed of an X-ray irradiation on the PMMA sheet and a subsequent thermal treatment. TheX-ray irradiation causes the decrease of molecular weight of PMMA, which in turn decreases the glass transition temperature2and consequently causes a net volume increase during the thermal cycle resulting in a swollen microlens 7. The shapes ofmicrolenses fabricated by the modified LIGA process can be predicted by a method suggested in 7.The microlens arrays used in the experiments were composed of 500m -(a 2 2 array), 300m -(2 2) and 200m (5 5)diameter arrays, and their heights were 20.81, 17.21 and 8.06m, respectively. Using the microlens arrays fabricated by themodified LIGA process as a master, a metallic mold insert was fabricated by a nickel electroplating for the injection molding.Typical materials used in a microfabrication process, such as silicon, photoresists or polymeric materials, cannot be directlyused as the mold or the mold insert due to their weak strength or thermal properties. It is desirable to use metallic materialswhich have appropriate mechanical and thermal properties to endure both a high pressure and a large temperature variationduring the replication process. Therefore, a metallic mold insert is being used rather than the PMMA master on silicon waferfor mass production with such replication techniques. Otherwise special techniques should be adopted as a replication method,e.g. a low pressure injection molding 8.The size of final electroplated mold insert was 30 30 3 mm. The electroplated nickel mold insert having microlensarrays is shown in Fig. 1.Fig.1.Moldinsert fabricated by a nickelelectroplating(a) Real view of the moldinsert (b) SEM image of 200mdiameter microlens array (c) SEM imageof 300mdiameter microlens array3 3 3 3InjectionInjectionInjectionInjection moldingmoldingmoldingmolding experimentsexperimentsexperimentsexperimentsA conventional injection molding machine (Allrounders 220 M, Arburg) was used in the experiments. A mold base for theinjection molding was designed to fix the electroplated nickel mold insert firmly with the help of a frametype bolster plate (Fig.2). Shape of aperture of the bolster plate (in this study, a rectangular one) defines the outer geometry of the molded part onwhich the profiles of microlens arrays are to be transcribed. The mold base itself has delivery systems such as sprue, runnerand gate which lead the molten polymer to the cavity formed by the bolster plate, the mold insert and amoving mold surface.The mold base was designed such that mold insert replacement is simple and easy. Of course, one may introduce anappropriate bolster plate with a specific aperture shape.Fig. 2. Mold base and mold insert used inthe injection molding experiment 3The injection molding experiments were carried out with three general polymeric materials PS (615APR, Dow Chemical),PMMA (IF870, LG MMA) and PC (Lexan 141R, GE Plastics). These materials are quite commonly used for opticalapplications. They have different refractive indices (1.600, 1.490 and 1.586 for PS, PMMA and PC, respectively), giving rise todifferent optical properties in final products, e.g. different foci with the same geometry.The injectionmolding experiments were performed for seven processing conditions by changing flow rate, packing pressureand packing time for each polymeric material. Furthermore, same experiments were repeated three times for checking thereproducibility. It may be mentioned that the mold temperature effect was not considered in this study since the temperatureeffect is relatively less important for these microlens arrays due to their large radius of curvature than other microstructures ofhigh aspect ratio. For high aspect ratio microstructures, we are currently investigating the temperature effect more closely andplan to report separately in the future. Therefore, flow rate, packing pressure and packing time were varied to investigate theireffects more thoroughly with the mold temperature unchanged in this study. Table 1 shows the detailed processing conditionsfor three polymeric materials. Other processing conditions were kept unchanged during the experiment. The mold temperatureswere set to 80, 70 and 60 _C for PC, PMMA and PS, respectively.It might be mentioned that we carried out the experiments without a vacuum condition in the mold cavity considering thatthe large radius of curvature of the microlens arrays in the present study will not entrap air in the microlens cavity during thefilling stage.Table 1. Detailed processing conditions used in the injection molding experimentsCaseFlow rate (cc/sec)Packing time (sec)Packing pressure(MPa)112.05.010.0212.05.015.0312.05.020.0PS412.02.010.0512.010.010.0618.05.010.0724.05.010.0PMMA16.010.010.026.010.015.036.010.020.046.05.010.05676.09.012.015.010.010.010.010.010.0PC16.05.05.026.05.010.0356.06.09.05.010.015.05.065.05.0712.05.05.04 4 4 4ResultsResultsResultsResults andandandand discussiondiscussiondiscussiondiscussionBefore detailed discussion of the experimental results, it might be helpful to summarize why flow rate, packing4pressure and packing time (which were chosen as processing conditions to be varied in this study) affectthereplication quality. As far as the flow rate is concerned, there may exist an optimal flow rate in the sense thattoo small flow rate makes too much cooling before a complete filling and thus possibly results in so-called shortshot phenomena whereas too high flow rate increases pressure fields which is undesirable.The packing stage is generally required to compensate for the volume shrinkage of hot molten polymer whencooled down, so that enough material should flow into a mold cavity during this stage to control the dimensionalaccuracy. The higher the packing pressure, the longer the packing time, more material tends to flow in. However,too much packing pressure sometimes may cause uneven distribution of density, thereby resulting in poor opticalquality. And too long packing time does not help at all since gate will be frozen and prevent material from flowinginto the cavity.Inthis regard, one needs to investigate the effects of packing pressure and packing time.4.14.14.14.1SurfaceSurfaceSurfaceSurface profilesprofilesprofilesprofilesFigure 3 shows typical scanning electron microscope (SEM) images of the injection molded microlens arrays for differentdiameters for PMMA (a) and different materials (b). Cross-sectional surface profiles of the mold insert and all the injectionmolded microlens arrays were measured by a 3D profile measuring system (NH-3N, Mitaka).Fig. 3. SEM images of theinjection molded microlensarrays and microlenses (a)Injection molded microlensarrays (PMMA) (b) Injectionmolded microlenses of 300mdiameterfordifferentmaterialsAs a measure of replicability, we have defined a relative deviation of profile as theheight difference between the moldedone and the corresponding mold insert for each microlens divided by the mold insert one. The computed relative deviations forall the microlenses are listed in Table 2.Diameter( m)Relative deviation (%)1234567PS200300500-7.625.862.38-7.592.03-0.382.082.860.51-5.611.47-8.6660161.47-11.444.291.47-5.731.95PMMA2003005007.205.77-0.661.315.60-1.62-3.886.453.98-5.805.952.80-0.975.95-0.72-8.536.68-0.904.86-2.62-0.72PC20030050023.026.20-0.9316.054.965.0916.872.66-1.8619.664.531.8833.974.786.9618.671.792.43-2.944.15-1.555It may be mentioned that the moldability of polymeric materials affects the replicability. Therefore, the overall relativedeviation differs for three polymeric materials used in this study. It may be noted that PC is the most difficult material forinjection molding amongst the three polymers. The largest relative deviation can be found in PC for the smallest diameter case,as expected. In that specific case, the largest value is corresponding to the low flow rate and low packing pressure. Packingtime in this case does not significantly affect the deviation. The relative deviation for PS and PMMA with the smallestdiameter is far better than PC case.Table 2 indicates that the larger the diameter, the smaller the relative deviation. The larger diameter microlens is, of course,easier to be filled than smaller diameter during the filling stage and packing stage. Microlenses of larger diameters weregenerally replicated well regardless of processing conditions and regardless of materials. The best replicability is found for thecase of PS with 500m diameter. Generally, PS has a good moldability in comparison with PMMA and PC.It may be mentioned that some negative values of relative deviation were observed mostly in the smallest diameter case forPS and PMMA according to Table 2. In these cases, however, the absolute deviation is an order of 0.1m in height, which iswithin the measurement error of the system. Therefore, the negative values could be ignored in interpreting the experimentaldata of replicability.Surface profiles of microlens of 300m diameter are shown in Figs. 4 and 5 for PC and PMMA, respectively. As shownin Fig. 4, the higher packing pressure or the higher flow rate results in the better replication of microlens for the case of PC, asmentioned above. Packing time has little effect on the replication for these cases. For the case of PMMA, the packing pressureand packing time have insignificant effect as shown in Fig. 5; however, flow rate has the similar effect to PC. It might bereminded that packing time does not affect the replicabilityifa gate is frozen since frozen gate prevents material from flowinginto the cavity. Therefore, the effect of packing time disappears after a certain time depending on the processing conditions.Fig.4ac(leftside).Surfaceprofiles of microlens (PCwith diameter (/) of 300m). a effect of packingpressure, b effect of flowrate,ceffectofpackingtimeFig.5ac.(rightside)Surfaceprofilesofmicrolens(PMMA with diameter(/)of 300m). a effect ofpacking pressure, b effectof flow rate,c effectofpacking time64.24.24.24.2SurfaceSurfaceSurfaceSurface roughnessroughnessroughnessroughnessAveraged surface roughness, Ra, values of 300m diameter microlenses and the mold insert were measured by an atomicforce microscope (Bioscope AFM, Digital Instruments). The measurements were performed around the top of each microlensand the measuring area was 5m 5m. Figure 6 shows AFM images and measured Ra values of microlenses. PMMAreplicas of microlens have the lowest Ra value, 1.606 nm. It may be noted that AFM measurement indicated that Ra value ofinjection molded microlens arrays is smaller than the corresponding one of the mold insert. The reason for the improvedsurface roughness in the replicated microlens arrays is not clear at this moment, but might be attributed to the reflow caused bysurface tension during a cooling process. It may be further noted that the Ra value of injection molded microlens arrays iscomparable with that of fine optical components in practical use.Fig. 6. AFM images and averagedsurface roughness, Ra, values of themold insert and injection molded 300mdiametermicrolenses.a Nickelmold insert, b PS, c PMMA, d PC4.34.34.34.3FocalFocalFocalFocal lengthlengthlengthlengthThe focal length of lenses can be calculated by a wellknown equation as follows:112111(1)()nfRRwhere f, nl, R1 and R2 are focal length, refractive index of lens material, two principal radii of curvature, respectively.Forinstance, focal lengths of the molded microlenses were approximately calculated as 1.065 mm (with R10.624 mm and R2￥) for 200m diameter microlens, 1.130 mm (with R1= 0.662 mm and R2=) for 300 m microlens and 2.580 mm (withR1=1.512 mm and R2=) for 500 m microlens according to Eq. (1). These calculations were based on an assumption thatmicrolenses are replicated with PC (nl= 1.586) and have the identical shape of the mold insert. It might be mentioned that thegeometry of themolded microlens might be inversely deduced from an experimental measurement of the focal length.5 5 5 5ConclusionConclusionConclusionConclusionThe replication of microlens arrays was carried out by the injection molding process with the nickel mold insert which was7electroplated from the microlens arrays master fabricated via a modified LIGA process.The effects of processing conditions were investigated through extensive experiments conducted with various processingconditions. The results showed that the higher packing pressure or the higher flow rate is, the better replicability is achieved. Incomparison, the packing time was found to have little effect on the replication of microlens arrays.The injection molded microlens arrays had a smaller averaged surface roughness values than the mold insert, which might beattributed to the reflow induced by surface tension during the cooling stage. And PMMA replicas of microlens arrays had thebest surface quality (i.e. the lowest roughness value of Ra =1.606 nm). The surface roughness of injection molded microlensarrays is comparable with that of fine optical components in practical use. In this regard, injection molding might be a usefulmanufacturing tool for mass production of microlensarrays.ReferencesReferencesReferencesReferences1. Ruther P; Gerlach B; Gottert J; Ilie M; Muller A; Omann C (1997) Fabrication and characterization of microlensesrealized by a modified LIGA process. Pure Appl Opt 6: 6436532. Popovic ZD; Sprague RA; Neville Connell GA (1988) Technique for monolithic fabrication of microlens array. Appl Opt27:128112843. Beinhorn F; Ihlemann J; Luther K; Troe J (1999) Micro-lens arrays generated by UV laser irradiation of doped PMMA.Appl Phys A68: 7097134. Moon S; Lee N; Kang S (2003) Fabrication of a microlens array using micro-compression molding with an electroformedmold insert. J Micromech Microeng 13: 981035. Ong NS; Koh YH; Fu YQ (2002) Microlens array produced using hot embossing process. Microelectron Eng 60: 3653796. Lee S-K; Lee K-C; Lee SS (2002) A simple method for microlens fabrication by the modified LIGA process. J MicromechMicroeng 12: 3343407. Kim DS; Yang SS; Lee S-K; Kwon TH; Lee SS (2003) Physical modeling and analysis of microlens formation fabricated bya modified LIGA process. J Micromech Microeng 13: 5235318. Bauer W; Knitter R; Emde A; Bartelt G; Gohring D; Hansjosten E (2002) Replication techniques for ceramicmicrocomponents with high aspect ratio. Microsyst Technol 7: 85 90