Elsevier

Additive Manufacturing

Volume 36, December 2020, 101451
Additive Manufacturing

Research Paper
Weld seam morphology and bond strength of infrared and vibration welded SLS parts of polyamide 12 as a function of the layer build-up direction and the welding process

https://doi.org/10.1016/j.addma.2020.101451Get rights and content

Abstract

Selective laser sintering (SLS) offers the possibility of manufacturing complex parts without tools and thus enables the cost-effective production of individualized products. In contrast to this advantage, the producible component dimension is restricted due to the dimensions of the building chamber of commercial SLS systems. Furthermore, a great amount of cost-intensive powder is necessary for the production of large SLS parts. In order to overcome these limitations, reliable joining processes, which enable high mechanical bond properties in SLS assemblies, are indispensable. Especially welding processes are suitable in this case. However, there is only limited knowledge available about the welding of SLS parts. In particular, the influence of the characteristics of the SLS process on the resulting bond quality has not been analyzed so far. Within this paper, the influence of the SLS layer build-up direction on the resulting weld seam morphology and bond strength as a function of the welding process is investigated. Infrared and vibration welding, which differ in the method of energy input into the joining zone, are used.

The investigations carried out show that high weld seam strengths can be achieved in SLS assemblies for both welding processes. The resulting bond strength is thereby largely independent of the SLS build-up direction. Vibration welded SLS parts exhibit high bond strengths in the range of the base material strength. SLS joints produced via infrared welding have lower bond strengths, which is expected to be caused by an already beginning recrystallization in the generated melt layer during the changeover phase. Furthermore, the investigations show that the weld seam morphology for SLS parts deviates from the characteristic structure of polymeric weld seams, as welded SLS samples do not show the typical separated multi-layer structure. The weld seam shows a very fine spherulitic, partly optically amorphous structure. This is probably the case because of additives contained in the SLS powder (e.g. pyrogenic silicon dioxide), which have an accelerating effect on crystallization. Deformed spherulites in the transition area to the base material are only visible for vibration welded SLS specimens. Due to larger occurring spherulites within the SLS samples compared to injection molded parts, their deformation takes place over a much wider area. The analysis of the weld seam morphology further substantiates the assumption that recrystallization already begins during the changeover phase in the infrared welding process. A highly oriented area with growing spherulites can be detected in the area of the joining plane. In summary, however, it can be noted that both welding processes are well suited for joining SLS parts.

Introduction

The additive manufacturing process of selective laser sintering (SLS) offers the possibility of manufacturing complex parts without tools and thus enables the cost-effective production of individualized products even with a batch size of one. In contrast to this advantage, there is a current limitation of the producible component dimension. This limitation results from the dimensions of the building chamber of commercial SLS systems. Furthermore, a great amount of the cost-intensive polyamide 12 powder is necessary for the manufacturing of large SLS parts. In order to overcome these limitations, reliable joining processes, which enable high mechanical bond properties in assemblies are also indispensable for SLS components. Generally, joining processes can be separated into adhesive joints, mechanical fasteners, and welding processes. Welding is characterized in particular by the ability to produce durable media-tight joints that allow high bond strengths even without the introduction of additives.

In order to be able to exploit the high mechanical properties of SLS parts in SLS assemblies, the production of SLS-SLS connections by means of welding processes shall be analyzed within the scope of this paper. In particular, the effects of the SLS layer build-up direction and the influence of the welding technology on the resulting weld morphology and the bond strength are to be investigated. Infrared and vibration welding are applied as welding methods, which differ in the way of energy input into the joining zone.

The vibration welding process belongs to the friction welding processes. Heating of the joining partners is thereby accomplished by outer friction. The process is characterized by short cycle times combined with a high reproducibility, high energy efficiency, and the suitability to join large components [1].

Linear vibration welding is based on a frictional relative movement of the joining partners [2]. Due to the applied joining pressure during the oscillating movement, a melting (semi-crystalline thermoplastic) or softening (amorphous thermoplastic) of the parts take place [2]. After the end of the welding process, the joining partners cool down under pressure [2].

Fundamental investigations on the vibration welding process [[3], [4], [5]] have shown a characteristic course of the weld path (time-dependent course of the displaced melt layer during the welding process), which is independent of the part geometry, material and process parameters. This resulting course can be divided into four phases. At the beginning of the oscillating vibration movement, the temperature of the joining surfaces continuously rises due to the solid friction. In this phase, surface roughnesses are smoothed by plastic- as well as elastic deformation and a thin melt film is formed locally as soon as the crystalline melting temperature (semi-crystalline thermoplastic) or the glass transition temperature (amorphous thermoplastic) is exceeded. In phase two the melt layer thickness increases steadily until an equilibrium is reached between the energy supplied by vibration and the energy flowing off into the weld bead. The result of this equilibrium is a stable melt layer thickness in phase three. In phase four, the friction-relative movement ends, and the joining partners cool down (passively) under pressure.

In order to achieve high bond strengths, it is necessary to reach the phase of the steady-state melt flow [6,7]. Therefore, the achievement of this third phase is generally regarded as a quality criterion in vibration welding [8]. The occurring energy equilibrium leads to a constant joining zone temperature during the third process phase [4,9,10]. Vibration welding is therefore classified as a self-regulating process.

In infrared welding, the heating of the parts takes place separately before the joining process [11]. The energy required for the melting or the softening of the material is introduced into the joining zone using an infrared emitter. The absorption of the radiation is dependent of the wavelength of the radiator, as well as of the morphological and chemical structure of the polymers [12]. The advantage of infrared welding in comparison to friction welding processes, is the contactless heating that enables a high freedom of the part geometry [13,14].

The infrared welding process can be divided into three process phases, which were described in detail by Heil [12]. In the beginning, the joining partners are heated up using an infrared radiator placed between the joining zones of the parts. The heating takes place contactless and no aligning of the joining partners occurs. For this reason, production-related tolerances can lead to fluctuations in the radiator spacing and thus, to uneven energy transfer. According to Ehrenstein [11], an unevenness greater than 0.2 mm should be avoided in order to achieve a homogeneous plasticization of the part surface. In contrast to other welding processes, there is no squeeze flow during the heating phase due to the pressureless heating. After the formation of a sufficient melt layer, the radiator is removed and the joining partners are contacted during the changeover phase. A minimum changeover time must be ensured in order to limit the cooling of the materials in this phase. The joining and cooling phase begins with the contact of the plasticized joining partners. The joining pressure builds up continuously from the beginning of the contact until the selected joining pressure is reached. The melt is thereby displaced into the welding bead. After a defined cooling time, which is necessary for sufficient stability of the parts, the joining process is completed and the component can be removed from the fixture [12,13].

The morphology of weld joints differs in its structure depending on the welding process. For vibration-welded polyamide, investigations by Chung et al. [15] show a heat-affected zone (HAZ) consisting of a core layer with an amorphous zone and a recrystallized zone and a layer of deformed spherulites, Fig. 1 (left). The recrystallized area is most prominent for low welding pressures [15]. An increase in the weld pressure for a constant amplitude leads to a decrease in the size of the spherulites. The optically amorphous region exhibits no recognizable superstructures, but is still crystalline, as investigations by Chung et al. (PA6) [15] or Stevens (PA66) [16] have shown. The deformed layer originates from material deformation in the rubbery state above the glass transition temperature Tg [15]. Spherulites are only partially melted and are orientated in the flow direction of the squeeze flow. The thickness of the zone of deformed spherulites increases with an increase in weld pressure, whereas the thickness of the core layer decreases [17].

Due to the separation of heating and joining and thus, the reduced molecular movements during the heating phase, weld seams produced by convection or radiation exhibit a different seam formation, Fig. 1 (right), in comparison to weld seams generated by friction based welding. A recrystallized zone is formed in the seam area, which has a mixture of a fine and a large spherulitic structure. Due to the increased cooling rate, the spherulite size decreases with increasing distance to the seam center. The flow zone then constitutes of an area with high orientations and has no recognizable spherulitic structures. Deformed spherulites represent the transition to the base material, analogous to vibration welding [18,19].

Selective laser sintering (SLS) is a powder and beam-based additive manufacturing process that can be divided into three process steps [20]. First, a new layer of powder with a defined thickness is applied to the building platform by a blade or roller. The powder is then heated up to the specific building temperature, which is between the crystallization and melting temperature of the semi-crystalline polymer. In the last step, the cross-section area of the resulting part is melted by a laser, with the surrounding powder serving as a supporting structure. Then the building platform is lowered and the three phases are repeated until the component has been completely generated [21,22].

Compared to other additive manufacturing processes, the SLS process offers high mechanical properties (e.g. tensile strength), which are comparable to injection-molded components [23]. However, the mechanical properties show an anisotropic behavior depending on the part orientation caused by the layerwise build-up. The highest mechanical properties are reached in layer plane direction, which is indicated by an increased tensile strength and elongation at break compared to loads acting perpendicularly on the layers [23]. Due to the pressure-free process and the surrounding powder bed, SLS parts have a complex surface roughness [24] as well as a porous inner structure [24,25] in the range of 3–5 %. In contrast to injection molding, in which the melt cools down below the crystallization temperature within seconds due to the contact with the cold mold, the SLS process has cooling times in the range of hours and days. These are caused by the high heat capacity of the powder and lead to a high crystallinity of SLS parts.

The additive manufacturing of polymer components using the SLS process is increasingly moving into the focus of user applications. Although the SLS process offers high integration possibilities [26], the necessity for joining processes cannot be completely ignored. Joining individual SLS components to assembly groups offers several advantages. On the one hand, part dimensions that are restricted by the given machine space can be extended. On the other hand, dimensional deviations due to thermal deformation in the SLS process can also be avoided by dividing large-volume SLS parts into several joining components [27].

Possible processes for producing joints are mechanical fasteners, gluing, and welding. In the field of additive manufacturing, however, there is hardly any available knowledge about joining. Adhesive bonds between at least one additively manufactured component and a counterpart can be found e.g. in the interior of vehicles [28]. In addition to welding processes, adhesive bonding offers the possibility to realize material- and media-tight joints [11]. However, in order to achieve sufficient bond strengths in adhesive joints, time-consuming as well as expensive pretreatment methods of the individual parts are often necessary [28]. Moreover, adhesives usually contain harmful ingredients and have to be tailored and approved for customer-specific use. Investigations by Fieger et al. [28] on adhesively joined SLS components showed achievable bond strengths of approx. 6 N/mm² for SLS-SLS joints made of PA12. The use of design elements (catching structures, grooves) or surface pretreatments (etching, plasma) increased the achievable strength slightly. However, the maximum achievable bond strength of 11.4 N/mm² was still well below the base material strength, which is in the range of 45−50 N/mm² [23] for SLS components made of PA12.

The common option of connecting SLS parts by using mechanical connectors (e.g. screws, snap fits) is a cost-effective and efficient way of realizing detachable or permanent bonds. However, the anisotropy of the properties of SLS parts can have a negative influence on the durability of load-bearing features [29]. Furthermore, the process-inherent staircase effect can also lead to a restriction of movement in sliders and other kinematic elements [29]. Media tightness can generally not be achieved with mechanical connectors.

Joints between laser-sintered PA12 parts created by hot plate welding have proven to remain mostly unaffected by typical process-related interferences of the SLS process, such as the anisotropy of the part properties [30]. Depending on the selected welding parameters, this joining process can achieve high weld strengths in the range of the base material strength. Heilig et al. [31] were able to show that welding of SLS components is also feasible by the use of ultrasonic welding. The achieved joints showed a high tensile strength. In addition, the investigation revealed that the anisotropy of the part properties influences the resulting weld quality. This influence can be explained by the geometric form of the energy director (ED) that exhibits rounded edges or staircase effects depending on the SLS build-up direction. A variation of the geometry of the ED significantly affects the energy input during the ultrasonic welding process. The energy density used in the SLS manufacturing process, as well as the selected ratio between recycled and new powder material, showed no influence on the resulting bond strength for both welding processes.

In summary, it can be stated that welding offers the possibility of exploiting the high mechanical properties of SLS components to a high degree. Furthermore, media-tight connections can be achieved by welding without the need for additives and health-endangering ingredients. However, only minor knowledge is available about the influence of the SLS process (e.g. complex surface roughness, layer build-up, porosity) on the welding process and the resulting bond qualities (e.g. strength, weld seam morphology). For this reason, first interactions between the layer build-up direction in the SLS manufacturing process and the thermal joining processes infrared welding and vibration welding are investigated within this study. Among other aspects, the resulting joining zone temperature, the weld seam morphology, and the bond strength are analyzed depending on the SLS build-up direction and the welding process. In order to evaluate and classify the SLS joints, comparative welding trials with injection-molded parts are carried out as a reference.

Section snippets

Materials and test specimens

For selective laser sintering, the semi-crystalline thermoplastic polyamide 12 (PA12) has established itself as a standard. In this work, PA12 of the type PA2200 (EOS GmbH, Germany, Krailling) was used. The commercially available polyamide 12 of the type Vestamid L1901 (Evonik Industries AG, Essen, Germany) was used as a reference material in the injection molding (IM) process. Before the IM-process, the material was dried at 80 °C for 12 h according to the manufacturer’s specifications.

The

Melting behavior

The results of the DSC analysis show differences in the melting behavior depending on the manufacturing process, Fig. 4. Injection-molded test specimens made of Vestamid L1901 have a melt peak temperature of 179 °C, whereas SLS specimens made of PA2200 have a slightly increased melt peak temperature of 182 °C. Previous investigations of injection-molded PA2200 showed a melt peak temperature of 178 °C [32], which corresponds to that of injection-molded Vestamid L1901. Consequently, it can be

Conclusion

The conducted investigations show that high weld strengths can be achieved by joining SLS parts by infrared and vibration welding. The resulting bond strength is thereby largely independent of the SLS build-up direction. Vibration welded SLS parts show a high bond strength in the range of the base material strength. SLS joints produced by infrared welding have reduced weld factors in the range of 0.7. This reduced bond quality is expected to be caused by a premature start of the

CRediT authorship contribution statement

Michael Wolf: Conceptualization, Methodology, Investigation, Writing - original draft, Writing - review & editing, Visualization, Funding acquisition. Jannik Werner: Investigation, Writing - review & editing. Dietmar Drummer: Supervision, Funding acquisition.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors thank the Bayerische Forschungsstiftung (BFS) for funding this study within the project FAB-Weld (AZ-1352-18). The authors are also grateful to the companies’ bielomatik GmbH (Neuffen, Germany), BMW AG (Munich, Germany), and Sintermask GmbH (Lupburg, Germany). Further, the authors thank the Evonik Industries AG (Essen, Germany) for providing the injection molding material.

References (38)

  • R.D. Goodridge et al.

    Laser sintering of polyamides and other polymers

    Progess Mater. Sci.

    (2012)
  • C. Klahn et al.

    Design guidelines for additive manufactured snap-fit joints

    Procedia CIRP

    (2016)
  • B. Caulfield et al.

    Dependence of mechanical properties of polyamide components on build parameters in the SLS process

    J. Mater. Process. Technol.

    (2007)
  • S. Friedrich

    Lineares Vibrationsschweißen von Kunststoffen im industriellen Umfeld. Dissertation

    (2014)
  • V.K. Stokes

    Vibration welding of thermoplastics. Part I: phenomenology of the welding process

    Polym. Eng. Sci.

    (1988)
  • V.K. Stokes

    Vibration welding of thermoplastics. Part II: Analysis of the welding process

    Polym. Eng. Sci.

    (1988)
  • H. Potente et al.

    Analysis of vibration welding

    Kunststoffe - German Plastics

    (1987)
  • A.K.H. Schlarb

    Zum Vibrationsschweißen von Polymerwerkstoffen

    (1989)
  • V.K. Stokes

    Vibration welding of thermoplastics. Part III: strength of polycarbonate butt welds

    Polym. Eng. Sci.

    (1988)
  • A.K.H. Schlarb et al.

    The impact strength of butt welded vibration welds related to microstructure and welding history

    Polym. Eng. Sci.

    (1989)
  • J. Vetter et al.

    On-line quality determination in vibration welding

    Kunststoffe

    (1998)
  • J. Vetter et al.

    Biaxial vibration welding of polypropylene

    Annual Technical Conference - ANTEC, Conference Proceedings

    (1999)
  • P.J. Bates et al.

    Real time temperature measurement of nylon 66 butt-joints during vibration welding

    Annual Technical Conference - ANTEC, Conference Proceedings

    (2003)
  • G.W. Ehrenstein

    Handbuch Kunststoff-Verbindungstechnik

    (2004)
  • M. Heil

    Heizstrahlerschweißen thermoplastischer Kunststoffe

    (1994)
  • D.A. Grewell et al.

    Plastics and Composites Welding Handbook

    (2003)
  • R. Fuhrich

    Infrarotschweißen von Kunststoffen mit thermischen Strahlungsemittern

    (2013)
  • Y.-M. Chung et al.

    Morphology of PA‐6 vibration welded joints and its effect on weld strength

    Polym. Eng. Sci.

    (2008)
  • S.M. Stevens

    Characterisation of nylon 66 vibration welds

    Annual Technical Conference - ANTEC, Conference Proceedings

    (1997)
  • Cited by (4)

    View full text