Available online at ScienceDirect. Physics Procedia 56 (2014 ) Erlangen, Germany

Transcription

1 Available online at ScienceDirect Physics Procedia 56 (2014 ) Manufact turing of conductive circuits for embedding stereolithography by means of conductivee adhesive and laser sintering b E 8 th International Conferencee on Photonic Technologies LANEE 2014 Bernd Niese a, *, Thomas Stichel a, Philipp Amend a, Uwe UrmoneitU a, Stephan Roth a,b, Michael Schmidt a-c a Bayerisches Laserzentrum GmbH (blz), Konrad-Zuse-Str. 2-6, Erlangen, Germany Erlangen Graduate School in Advanced Optical Technologies (SAOT), Friedrich-Alexander-Universität Erlangen-Nürnberg, Paul-Gordan-Str. 6, Erlangen, Germany c Institute of Photonic Technologies (LPT), Friedrich-Alexander-Universität Erlangen-Nürnberg, Konrad-Zuse-Str. K 3-5, Erlangen, Germany Abstract The embedding stereolithography (esla) is an additive, a hybridd process which combines e flexible production of 3D-components with integration of electrical and optical conductive structures andd functional components. This combinationn of several process steps in one manufacturing process implies a high technological potential regarding integration densityy of assemblies. To createe conductive circuits inside and on surface of SLA-parts, manufacturing process of se structures has to be b integrated into SLA-process and should not contain disassembling of parts from SLA-building platform. In this context, production of embedded conductive circuits c by means of dispensing conductive adhesives and laser sintering is a highly promising process. The dispensing can be made during entire SLA-process by nterrupting it. In this way conductive adhesive can be deposit inside part and electrical conductivity of se structures will be achieved by laser sintering in next step. This paper shows fundamental investigationss concerning applicabilityy of conductive adhesive for embedding stereolithography and e laser sinteringg process as well The Authors. Published by Elsevier B.V. This is an open access article under CC BY-NC-ND license 2014 The Authors. Published by Elsevier B.V. ( Selection and blind-review Stereolithography; additive manufacturing; electrical e conductive circuits; selective laser sintering under responsibility off Bayerisches Laserzentrum GmbH. Peer-review under responsibility of Bayerisches Laserzentrum GmbH Keywords: 1. Introduction Due to a minimization of design and an optimization of spatial arrangement concerning layout and part placement in electronic housing, mechatronic devices become increasingly complex concerning ir assemblyy and circuit layouts. This trend is accompanied by variant diversity of mechatronic devices and a decreasing demand in ir quantity (Runge, 2000). Foremost, conventional production technologies of mechatronic devices like two-compone nt injection molding or injection molding in combination with stamping are suitable for a cost effective mass production but not flexible in n case of a fast modification of geometry and circuit layout of a part. In contrast to that, laser direct structuring (LDS) of injection molding parts enables flexible creation of conductive circuits by adding laser additives in molding compound. Beside, rapid r prototyping technologies like stereolithography offer a high potential to produce functional prototypes as well as small series of products in a fast and economicc way withoutt any shape forming tools (Ayers & Hilbrandt, 1998). * Corresponding author. Tel.: ; fax: address:b.niese@blz.org The Authors. Published by Elsevier B.V. This is an open access article under CC BY-NC-ND license ( Peer-review under responsibility of Bayerisches Laserzentrum GmbH doi: /j.phpro

2 Bernd Niese et al. / Physics Procedia 56 ( 2014 ) Fig. 1. Example of an esla module with integrated conductive circuits and electrical components. For production process of mechatronic devices, it is not only sufficient to have a fast and flexible modification of part geometry but also to enable a flexible integration of mechanical and electronic components in part. In this context embedded stereolithography could provide a solution for a direct integration of functional components (see Fig. 1). Because of automatic, layered design of parts it is possible to produce three-dimensional components with a large number of variants combined with a maximum flexibility in quantity. The SLA-process enables creation of parts with a high surface quality and using high temperature resistant resin materials. 2. Embedded stereolithography Stereolithography enables fabrication of a part from a designed CAD model. The digital data of this model has to be placed in virtual building space of stereolithography machine. Furr support is generated automatically by data preprocessing software. This support construction stabilizes polymerized overhanging parts of component. As building operation is performed layer wise, geometrical data of part and support construction has to be sliced in adequate digital layers with a height of typically 0.1 mm. Each of se layers is fabricated by spreading a liquid photo-resin over building platform with a spreading knife. Afterwards radiation of an UV laser ( = 355 nm) is used to cure material on resin surface according to given layer information. Finally building platform is lowered. By repeating this procedure complete component is created layer by layer. The idea of embedded SLA is to extend flexible creation of geometries to electrical and optical functionality. On basis of conventional SLA a new hybrid production technology is investigated, which enables integration and contacting of electronic components in parts during SLA-process by means of stereolithography and laser sintering of conductive adhesive. At first a housing shell with cavities is created by conventional SLA. In se cavities functional components like lead frame assemblies, printed circuit board assemblies and peripheral components e. g. capacitors or optical transceivers can be placed after removing fluid resin of cavities by a vacuum pump. However, cavities cannot be cleaned completely in this process step, so a subsequent laser ablation process of remaining resin is necessary to ensure reproducible and defined placement of components and creation of conductive circuits. In next step, placed functional components are connected with fluid conductive adhesive by a dispensing system. The fluid conductive adhesive is exposed and subsequently sintered by laser radiation ( = 355 nm). After placing components, remaining hollows of cavities are completely filled with resin by coating system of SLA-machine. By doing this entire components are wetted by resin and resin level in cavities is equal to resin level of SLA-bath. Finally upper housing of connected components is fabricated by continuing building process, see Fig. 2. It is important to note that embedded structures as well as components must not overtop current top layer of part. Orwise re is a collision between coating system and integrated components respectively conductive circuits. Using via or plated through-holes to create conductive joints to conductive circuits in upper layers of SLA-part, it is possible to contact components which are placed in a three-dimensional array.

3 338 Bernd Niese et al. / Physics Procedia 56 ( 2014 ) Fig. 2. a) Creating of housing shell and cavities b) Removing resin in cavities by vacuum pump and laser ablation c) Placement of functional components d) Dispensing off conductive adhesive e) Laser sintering of conductive adhesive f) Finished esla-module. Considering described process chain, integration of conductive adhesive into parts requires structures with low values in height. In addition, wettability of conductive adhesive on substrate, e contact angle, as well as accessible electrical conductivity of conductive circuits after laser sintering process p are main factors which must be taken into account. Particularly geometry of conductive structures depends on dispensing d parameters like velocity, pressure, nozzle diameter and viscosity of conductive adhesive. Due to SLA-process it is important that dispensed structures both on contact points of functionall components and on top layer of part do not t hinder layer wise building process. For that reason it is necessary to investigate minimum height of e conductive structures. In this context, geometry of dispensed structure has to be determinedd two times, after dispensing process and after sintering process. However, measured heights of sintered conductivee adhesive are finally determinant d factor for creating dimension of cavities getting a failure-free building process. 3. Materials and Methods 3.1. Resin The utilized material is an epoxy-based SLA-resin which approximately achieves mechanical and rmal properties of technical plastics in electronicc production. The used SLA-resin in this work is Somos S NanoTool (DSM) which enables production of parts with a highh stiffness and a high temperature resistance (up to 263 C). In this s study, Somos NanoTool is additionally doped with 1 wt.-% of aluminum powder (particle diameter d 1.5 μm). According to Amend et al. 2013, investigations with regard to selective activation of SLA-resin Somos NanoTooll were carried d out in previous studies. In this context, polymerized, doped resin is locally illuminated by laser radiation, whereby aluminum particles are exposed on surface. By doing this, copper particles can c deposit onn activated structures during a subsequent chemical metallizing bath (Amend, et al., 2013). This can be an alternative to create conductive circuits byy means of laser structuring and subsequent chemical metallization. It can also be used for contacting conductive circuits inside and outside of parts Conductive adhesive A single component (1 K) silver-filled, solvent-free conductive adhesive is used in experiments. The amount of silver particles in conductive adhesive is approximately 80 wt.-% and viscosity iss 4000 to 5000 mpas. Thee small size of silver particles with < 10 μm and low viscosity allows dosing by means of a dispensing system. The usuall sintering process of conductive adhesive is realized by oven sintering att 110 to 150 C for 10 to 300 minutes. Besides, absorption coefficient of se conductive adhesives was investigated in variouss researches testing conductive adhesive filled with silver nanoparticles in combination with laser sintering in electronic use (Maekawa, et al., 2009; Maekawa, M et al. 2010). According to se investigations, a high absorption in wavelength range of 400 nm to 500 nm was determined. d

4 Bernd Niese et al. / Physics Procedia 56 ( 2014 ) Dispensing system The dispensing process described in Fig. 2 should enable creation of conductive circuits and contacting of functional components. The used dispenser in this work is MicroDispensing System MDS 3200A (VermesTechnik). This system is suitable for dispensing high viscous media and provides very high frequencies of up to 150 Hz. The dispenser is mounted on a Pick and Place system called Speed Mounter², integrated in SLA-machine. Therefore it is possible to generate 2 ½-D structures on SLA-parts with a speed up to 5000 mm/s. The used conductive adhesive is dispensed by a nozzle having a diameter of 200 μm. Testing nozzle with a diameter of 100 μm leads to agglutination of conductive adhesive Laser system The laser beam source used for laser sintering is a pulsed solid state laser (Nd:YVO 4 ) with wavelength of 355 nm. The maximum nominal power is 3 W at a frequency of 40 khz. The laser beam is focused on working space after redirecting by a scanner system. The sintering process of conductive adhesive is carried out with a focus diameter of approximately d = 50 μm. 4. Experimental investigations of laser sintering of conductive adhesive For experimental tests concerning dispensing of conductive adhesive on substrates, dispensed structures are lines with a length of 20 mm. By variation of parameter dispensing velocity v d, experimental test shall result in creating conductive circuits with preferable low heights and low contact angles. The dispensing parameter pressure (p = 6 bar), nozzle diameter (d n = 200 μm) and nozzle temperature (t n = 40 C) are used and remain constant according to manufacturer s recommendation (Panacol, 2009). Additionally dispensing of conductive adhesive with a nozzle diameter of 100 μm was investigated as well but results are not satisfactory. Even raising pressure, effect is agglutination within nozzle due to large particle size of < 10 μm on average. For experimental tests concerning laser sintering of dispensed conductive adhesive, test structure is shown in Fig. 3. The first laser sintering tests are carried out with a focus diameter of d = 50 μm to get an impression of interaction between laser and material. After dispensing a line, it is necessary to determine a laser parameter for a fast sintering process without any rmal damage neir substrate nor conductive adhesive. Therefore, laser sintering strategy of dispensed conductive adhesive consists of exposing in center of dispensed line and pursuing structure with a defined velocity v s. The varied laser parameters in experimental investigations are laser power P L, frequency f, velocity v s and number of loops n, see Fig. 3. In studies laser power P L for sintering process is varied from 0.5 to 2.5 W. The investigations can be described by showing three characteristic results. Starting with too low laser power (P L = 0.7 W), rar with sufficient laser power (P L = 1.4 W) and ending with too high laser power (P L = 2.2 W). The variation of focus diameter is even a proper parameter have to investigated and will be considered in next studies, especially for contacting electrical components by this way. The electrical resistance R of sintered conductive adhesive structure is measured by four point measurement method. In this case, two pairs of measurement points are connected to ends of conductive circuits. The measured conductive circuits are 20 mm in length. The volume of conductive circuits is primarily depended on dispensing parameter. However, after sintering process a volume shrinkage is to be expected resulting in removal of isolated coating enclosing silver particles (Franke, 2013; Hörber, et al., 2013). The values for height, width and contact angle of structures are measured before and after laser sintering process.

5 340 Bernd Niese et al. / Physics Procedia 56 ( 2014 ) Fig. 3. Test structure and laser sintering strategy. 5. Results and discussion 5.1. Laser sintering of conductive adhesive structures Directly after dispensing, color of conductive adhesive (Elecolit 3043, Co.Panacol) is matte grey. This is an indication that conductive adhesive contains organic solvent respectively silver particles are enclosed by organic solvent and refore dispensed lines are not electrically conductive and pasty. Thus, a subsequent sintering processs is necessary. By using rmal or photonic energy, organic solvent within conductive adhesive can be evaporated and t silver particles sinter toger. After this sintering process, color off conductive adhesive changes fromm grey to silver glossy and conductive adhesive is electrically conductive (e. g. see Fig. 4 for laser sintering technique). To characterize t structures it is necessary to detect electrical resistance of m by four point measurement. For first investigation of applied structures, dispensing velocity is constant at 800 mm/min, thus volume flow is same at each structure. Concerning laser sintering,, at low laser power (P L = 0.70 W) surface of dispensed structure is sintered evidently at silver glossy top, but small amounts of not sintered conductive adhesive still exists in deeper areas. The lines sintered with low laser power (PL = 0.7 W) have a high resistance ( = 0.43 ) and a low adhesive strength on substrate can be detected by scratch tests. By increasing laser power P L, surface of conductive adhesive starts to get damaged resulting in traces of powder, even though resistance of structures decreases (P L = 1.4 / = 0.25, P L = 2.22 W / = 0.20 ). Furrmore, conductive adhesive withinn and alongside laser impact zone gets more and more ablated. With highest laser power in Fig. 4, not only conductive adhesive but also substrate getss damaged resulting in a deep ablation crater. Even volume of dispensed structures is known, a measurement of specific resistance is not expedient because of different numbers of pores inside structures and different amount of ablated section. Despite a large number of holes and pores within laser sintered structures, electrical resistance of se is fairly good compared with rmal sintered conductive circuits. The electrical resistance of four oven sintered structures of same geometry and of same measured length,, as describedd in Fig. 4, is on average = 0.27 (Sintering T = 150 C / t = 10 min. according to manufacturer s data). Regarding more specifically cross-cuts of conductive circuits in Fig. 4, it is obvious that many holes and pores have formed along cross-section independently of amount of used laser l energy. One possible explanation for this could be e high rmal energy caused by ablation of laser radiation. This effects an immediately sintering in upper areas of conductive adhesive. The consequence is that re is not enough time for outgassing process of organic solvent underneath this sintered top layer to evaporate completely and so it forms gas inclusions or pores and cavities or hole, see Fig. 4 (Hörber, et al., 2013). This effectt is much lower for oven sintered conductive circuits. Here, re are only a few small pores within conductive circuit andd hardly any holes or cavities linked to surface. This is due to much longer sintering time and an almost simultaneous sintering of complete conductive circuits.

6 Bernd Niese et al. / Physics Procedia 56 ( 2014 ) Fig. 4. Influence of laser power P L L= 0.7, 1.4, 2.2 W on optical appearance of laser l sintered conductive lines (Constant parameters: v s = 1 mm/s, f = 20 khz,, d = 50 μm, n = 2). At moment best results are achieved for laser parameters P L = 1.4 W, f = 20 khz, v s = 1 mm/s and n = 2, see Fig. 4. The conductive circuits exposed with se laser parameters show only minimal traces of powder r in laser impact zone but surface sare completely sintered and have a good adhesive strength on substrates. For that reason, all following experiments concerning exposure of conductivee adhesive aree carried out with se laser parameters in following. Due to relative fast laser sintering process in comparison with oven sintering process, formation of pores iss hard to avoid by variation of laser parameters only. However, re is a need to investigate alternative exposure strategies. This will be done in furr test studies Dispensing of conductive adhesive structures After investigations on suitable parameters for laser sintering, re is a need to findd suitable dispensing parameter embedding conductive adhesive in cavities. The manufacturing of SLA-parts using u a layer thickness of 0.1 mm results in creating of conductive structures which enable a failure-free have to be investigated regarding minimum m possible height before and after building process. Forr that reason, dispensing structures of used conductive adhesive on SLA-material laser sintering process.the dimensions (width, height, contact angle) of dispensed structures are measured by a laser scanning microscope (LSM) before and after laser l sinteringg process to get information about shrinkage. According to manufacturer s recommendation, dispensing parameter pressure (p = 6 bar), nozzle diameter (d n = 200 μm) and nozzle temperature (t n = 40 C) remain constant during tests. For each velocity step from 800 to 1800 mm/min,, three lines with a length of 20 mm are dispensedd to get a statistical value. Depending on variation of dispensingg velocity v d Fig. 5 shows values for width, height and contact angle of measured structures. The continuous lines represent measure values before laser sintering and dotted lines afterwards. According to Fig. 4, parameters for laser sintering process are P L = 1.4 W, v s = 1mm/s, f = 20 khz and n = 2. By using se parameter values best overalll results are achieved regarding sintered material, adhesive strength on substrates and low rmal damage by laser radiation. To get a process window for dispensing process of chosen conductive adhesive re is a need carry out investigations concerning dimension of structures. As shown in Fig. 5, with an increase in dispensing velocity dimensions of dispensed structures (width, height, and contact angle) adopt lower values until velocity of 1600 mm/min.

7 342 Bernd Niese et al. / Physics Procedia 56 ( 2014 ) Heree values have been reduced from about 1200 μm to 800 μm for width, from about 533 to 22 forr contact angle and from about 280 μm to 180 μm for height on average. Above velocity of mm/min, structures on substrate curl and tend to interruptions during dispensing process. Fig. 5. Geometry of dispensed conductive adhesive before b and after laser sintering (Parameters: v d = mm/min, P L = 1.4 W, vs = 1 mm/s, f = 20 khz, n = 2). The heights of structures dispensed with a velocity of 1600 mm/min vary between 193 μm and 161 μm after laser sintering. However, minimum heights are still tooo high for a failure-free dispensing process on topp layers of e SLA-part which needs heights below 100 μm. In this case, itt is necessary y to find or solutions to embed structures within matrix. One possibility is to design cavities, in which e conductive adhesive can be dispensed. The sufficient depth of such cavities would be approximately about 200 μm. In terms of creating se cavities, re are several unresolved questionss for examplee cleaning of cavities during SLA-building process and exact placing off conductive adhesive into m have to be investigated. Therefore, furrr experiments facing se challenges are needed Comparison between oven and laser sintering Regarding sintering of conductive adhesive Elecolit 3043, followingg section shows difference in resistance measurement concerning oven sintered process, laser sintered process and combination of both, see Fig. 6. Considering previous results in tests, best parameters for dispensing (v d = 1600 mm/min) and laser sintering (P L = 1.4 W) are used now. After SLA-process, part respectivelyy resin in cavities iss not completely hardened by laser exposure. Thus, laser sintering with a subsequent post-curing by oven sintering is also important to investigate. So it is necessary to polymerize residual resin within parts by a rmal downstream process like oven sintering. By doing that, re is possibility to get a dual benefit with an additional processs step. On one hand, e residual resin gets completely polymerized by rmal energy and on or hand conductivee adhesive is sintered by laser exposure and rmal energy. The oven sintered process of conductive adhesivee is carried out at 150 C for 10 minutes. For laser sintering process, most appropriate parameters are determined in chapter 5.1 and 5.2. With se parameterss conductive adhesive is i sintered and electrical resistance is measured by four point probe measurement.

8 Bernd Niese et al. / Physics Procedia 56 ( 2014 ) Fig. 6. Measured resistance values of conductive adhesive sintered by oven by laser and in combination of both. The results in Fig. 6 show that sintering process has an influence on electrical resistance of conductive circuits. In this diagram, oven sintered structures have an electrical resistance of = In I comparisonn to oven sintered structures, laser sintered, respectively laser and oven sintered structures, have an a electrical resistance of = 0.39 or = 0.33 on average. So, electrical resistance of structures is reduced to about 30 percentt by laser exposure and to about 40 percent by laser exposure and oven sintering. This could be explained by fact that conductive adhesive is sintered in oven at lower temperaturee and for a longer time. For this case, conductive adhesive iss completely sintered along cross-section of dispensedd structures but low temperature of 150 C effects only an electrical connection between gains due to building of sintering necks. Contraryy to this, laser sintering process causes immediate sintering and evaporation of organic solvent within exposure duration time, which leads to e formation off pores. This indicates that a real melting process is mediated due to absorption of laser radiation by silver particles which enhance conductivity c in laser impact zone and surrounded areas. In this context, electrical resistance would be reduced locally because of melting spots of silver particles. In some places, fast sintering of conductive adhesive in upper areas of structures causess a lower sintering of particles in deeperr sections. The result is, for example ann electric conductive area in upper section of structure and a non-electric conductive area in deeper sections combinedd with a lowerr adhesive strength to substrate. An additional rmal post-curing by oven sintering affects a completely sintering of conductive adhesive, especially in sections near substrate. 6. Conclusions A new approach for flexible manufacturing of mechatronic modules by stereolithographyy and selective laser based creation of conductive circuits is presented. For integration of conductive circuits and d electronic components c in SLA-parts, creation of se conductivee structures should be possible during SLA-process and structures mselves should not interrupt building process of part. Inn this context, suitability of laser sintering of conductive adhesive, dispensed on top layers of part, is investigated in this work. The evaluation of experiments shows good results for sintering of conductive adhesive by laser radiation. The laser sintering processs is fast and flexible and does not induce a rmal damage to e substrate when using adequate parameters. Only fast sintering of conductive adhesive prevents a slowly evaporation of organic solvent. The result is a formation of pores and cavities within conductive circuits, c whichh seems to have no significant influence on electrical conductivity. For a failure-free SLA-building process, conductive circuits should not exceed a definedd height dimension.taking into account shrinkage of conductive adhesive a afterr laser sintering process, investigations were done determining a suitable dispensing parameter to minimize height. By increasing dispensingg velocity to v d = 1600 mm/min, conductive circuits with an average height of 180 μm can be achieved. This value is 80 μm beyond target value of 100 μm. At higher velocities, dispensed lines are curled and tend to disconnections. Creating embedded structures inside SLA-parts, one possibility is to design cavities embedding dispensed structures within part. Furrmore, electrical resistance measurement of structures whichh are sintered by b oven, by laser and by laser and as well by oven is carried out. The experimental results indicate thatt a better conductivity of structures can be reached by laser sintering compared to oven sintering. One possible reason r could be that high rmal energy in upper sectionss of conductive

9 344 Bernd Niese et al. / Physics Procedia 56 ( 2014 ) adhesive effects a local melting of silver particles. The result for laser sintering is a lower resistance of conductive circuits in spite of forming pores. The conductive circuits sintered by oven have a significant lower amount of pores within structures but a higher electrical resistance on average as well. The proposed approach can be a suitable solution for manufacturing embedded conductive structures in parts by means of laser sintering and stereolithography. Furr work will consider embedding of conductive adhesive, ir characterization and contacting of functional components. Acknowledgements This research is supported by Deutsche Forschungsgemeinschaft (DFG) within project embedded stereolithography. Furr authors want to thank company alphaform, Feldkirchen, Germany for supporting this work. References Amend, P. et al., Fast and flexible generation of cunductive circuits. In: Proceedings of LAMP th International Congress on Laser Advanced Materials Processing. Ayers, K. & Hilbrandt, F. J., The usage of Stereolithographic Parts as Final Product. San Antonio, Texas: 3D Systems North America Stereolithography User Group Meeting. N.N.: Data sheet Somos NanooTool, DSM, [Online] Available at: [Accessed 18 February 2014]. Franke, J., Räumliche elektronische Baugruppen (3-D MID). München: Carl Hanser. Hörber, J. et al., Selektives Laser- und Lichtsintern von Aerosol-Jet gedruckten Nano-Silbertinten für rmoplastische Schaltungsträger.Bamberg: Meisenbach. Maekawa, K. et. al., Influence of Wavelength on Laser Sintering Ink-jet Printed Conductive Microstructures Containing Nano-sized Silver Particles. San Diego, USA: 59th Electronic Components and Technology Conference - ECTC. Maekawa, K. et al., Laser Sintering of Silver Nanoparticles for electronic Use. In: Materials Science Forum Vols Trans Tech Publications, Switzerland, pp N.N.: Data sheet Elecolit 3043, Panacol, [Online] Available at: Runge, W., Electronics inside Transmission - Chances and signes of Mechatronic Control moduls, Göppingen: Deutsches IMAPS Seminar.