Journal of Materials Processing Tech.

309 (2022) 117738

Contents lists available at ScienceDirect

Journal of Materials Processing Tech.

journal homepage: www.elsevier.com/locate/jmatprotec

Pore formation and melt pool analysis of laser welded Al-Cu joints using
synchrotron radiation
S. Hollatz a, *, M. Hummel b, A. Olowinsky a, A. Gillner a, b, F. Beckmann c, J. Moosmann c
a
Fraunhofer Institute for Laser Technology ILT, Steinbachstr. 15, 52074 Aachen, Germany
b
Chair for Laser Technology LLT, RWTH Aachen University, Steinbachstr. 15, 52074 Aachen, Germany
c
Institute of Materials Physics, Helmholtz-Zentrum HEREON, Max-Planck-Str. 1, 21502 Geesthacht, Germany

A R T I C L E I N F O A B S T R A C T

Associate Editor: Hui-Ping Wang Known as challenging material combination, the welding of aluminium and copper, both with strongly different
thermophysical properties, causes joining failures such as pores, cracks or intermetallic phases in the solidified
Keywords: welding area. To investigate the mixing of the materials and the occurrence of pores, the laser welding process is
Laser welding observed with synchrotron radiation which visualizes the phase boundaries between solid, liquid and gaseous
Synchrotron
material phases. This allows the visualization of pore formation and density differences of the materials inside
Aluminium
the melt pool. In this investigation, pore formation in front and bottom of the keyhole is observed. The movement
Copper
Melt pool dynamics of the bubbles in the melt pool can be tracked until solidification at the material transition. Regarding the
Pore formation intermixing of the materials, the high-speed images show a fluctuating copper flow towards the keyhole and a
material mixing over the entire aluminium melt pool depth. By understanding the mechanisms, compensatory
measures for an improved process can be developed to enable the usability of aluminium and copper connec­
tions, for example in electromobility applications.

1. Introduction In laser beam welding, a differentiation is made between heat con­

duction welding and deep penetration welding. Deep penetration
To reduce the CO2-emission, the international automotive industry is welding is based on evaporation of material due to absorption of laser
developing an increasing number of battery electric vehicles. Increasing radiation. When a material-dependent intensity threshold is exceeded, a
electrification and the development of battery storage systems require a vapor capillary (keyhole) is formed. The resulting keyhole is moved
variety of high-quality electrical connections. Due to the aluminium and through the workpiece by laser movement and leads, compared to heat
copper electrodes of the Li-ion cells, dissimilar material joints must be conduction welding, to narrow and deep weld seams. During a deep
created for the connection of battery cells to a battery storage system penetration welding process, different melt pool flows occur. The
(Lee et al., 2010). The challenge in welding aluminium and copper is the displacement of material due to the relative movement of the vapor
formation of pores, intermetallic phases and cracks which increase the capillary and the workpiece, results in a backward flow of material
electrical resistance of the joint (Schmidt et al., 2012). around the capillary. At the same time, the metal vapor escaping from
Commonly used manufacturing techniques in battery applications the capillary generates an acceleration of the adjacent melt in the di­
are resistance spot welding, ultrasonic welding, and laser beam welding rection of the capillary opening. A third effect is based on the temper­
(Das et al., 2018). Talking about aluminium and copper, the high ther­ ature dependence of the surface tension of the melt. This moves the near-
mal and electrical conductivity is challenging for the cheap and surface melt from hot areas close to the capillary towards the cooler edge
well-known resistance welding process. Ultrasonic welding is limited of the melt pool (Hügel and Graf, 2009).
due to material thickness, high demand on the surface condition and To increase the process stability and to adjust the weld seam geom­
induced vibration stress (Brand et al., 2015). The high level of auto­ etry, a spatial power modulation (a superposition of the feed rate with
mation, process speed and mechanical contactlessness make the laser an oscillation movement) is used (Schmitt, 2012).
welding process a promising process for this application (Das et al., There are several investigations on the mixing of aluminium and
2018). copper during laser welding. Balu et al. (2011) have investigated

* Corresponding author.
E-mail address: soeren.hollatz@ilt.fraunhofer.de (S. Hollatz).

https://doi.org/10.1016/j.jmatprotec.2022.117738
Received 7 April 2022; Received in revised form 20 July 2022; Accepted 27 July 2022
Available online 29 July 2022
0924-0136/© 2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
S. Hollatz et al. Journal of Materials Processing Tech. 309 (2022) 117738

different welding arrangements, for example Cu on top of Al or vice the high energy beamline P07 (EH4) of Petra 3 at Deutsches Elektronen
versa. The weld seam geometry in cross section has been analysed, Synchrotron DESY in Hamburg, Germany (Schell et al., 2013). To
resulting in a wider weld seam on aluminium side due to the lower observe the welding process, the x-ray radiation is transmitting the
thermal conductivity and melting point. Hardness measurements sug­ metal probe and is detected by a CdWO4 (Cadmiumwolframat) scintil­
gest the occurrence of intermetallic phases (Balu et al., 2011). Further lator. Using a Photron Fastcam SA5, the intrinsic glow of the scintillator
research is dealing with the influence of the laser parameters and the screen stimulated by the X-rays is recorded. The pictures were taken
usage of beam oscillation on mechanical and electrical properties of the with 1024 × 768 pixels with a frame rate of 1 kHz and 1 ms exposure
dissimilar joint. Kraetzsch et al. (2011) could achieve crack free welds time. The synchrotron setup and parameters are shown in Table 1.
by using a high frequency beam oscillation. Dimatteo et al. (2019) have The setup for the laser welding includes a high-speed axis to apply
discovered increase mechanical properties by decreasing the penetra­ the feed rate, a scanning system with optics and the laser beam sources.
tion depth in lap joints. An investigation of the intermetallic phases For the investigations two laser beam sources with different spot di­
relative to the percentage of aluminium in the melt pool has been con­ ameters are used. A schematic of the setup is shown in Fig. 1.
ducted by Bantel (2018). The EDX analysis leads to the conclusion, that a More information about the set up can be found in Wagner et al.
higher percentage of aluminium in the melt pool reduces the occurrence (2021).
of critical intermetallic phases and improves the mechanical properties
of the lap joint. Measurements of the electrical resistance of each 2.1. Laser beam characteristics for experimental investigation
intermetallic phase were performed by Rabkin et al. (1970). Compared
to the base material, the resistance increases for all intermetallic phases. To compare the influence of different spot diameters on the shape of
For the theta phase, there is an increase by a factor of almost six. the keyhole, a single-mode fibre laser and a multi-mode disk laser are
A major influence on intermetallic phase and crack formation is the used. The characteristics of the laser beam sources are shown in Table 2.
mixing of the materials. This in turn is dependent on the melt pool dy­ The used laser beams sources differ by a factor of about three in the
namics, caused by the vapor capillary and the weld depth as a measure resulting focal diameter. The laser beams are guided through a scanner
for the material ratio in the melt (considering a lap joint). Using an x-ray system and f-theta optics to the workpiece surface. The beam position is
phase contrast method, the keyhole and solid-liquid interface can be fixed, while the workpiece is moving with the feed rate. The stationary
observed during laser welding. Miyagi et al. (2017) have identified in­ laser beam is necessary to have a fixed position for the x-ray beam. The
stabilities at the keyhole bottom as a reason for porosity formation in scanner system is used to initially arrange the laser beam and to apply a
aluminium using x-ray analysis. They have observed pores, formed at circular spatial power modulation. Thus, the following results differ in
the keyhole bottom, moving towards the surface, and being trapped at focal diameter, laser power, feed rate as well as amplitude (radius) and
the solid-liquid interface in the back of the melt pool. Keyhole in­ frequency of the spatial power modulation.
stabilities are also identified by Fetzer et al. (2018) as a main driving
force for the formation of pores in aluminium welding with x-ray anal­ 2.2. Material preparation for X-ray analysis
ysis. As a different method to identify pore formation during the welding
process, a butt joint configuration with glass and stainless steel can be The workpieces for the investigation are based on 3 mm thick metal
used. The laser beam is focused on the steel close to the edge. With this sheets which are irradiated by the x-rays. The material samples are
setup, Xu et al. (2018) have observed a keyhole fluctuation and collapse, 30 mm in height and 100 mm in length. For the investigation of the
resulting in bubble formation. dissimilar lap joints, sheets with 0.5 mm in height are welded to the top
Cunningham et al. (2019) suggest a relationship between the of the base metal sheets. As top layers Al99.5 and Cu-ETP are used with
appearance of pores and the aspect ratio of the keyhole depending on the CuSn6 and Al99.5 as base materials. The properties of the used materials
ratio of laser power and feed rate. By using x-ray synchrotron radiation, are shown in Table 3.
they could observe keyhole instabilities and bubble formation while
welding titanium alloys for application in laser powder bed fusion. A 2.3. Process observation with synchrotron radiation
detailed visualisation of a keyhole collapse, bubble formation and an
analysis of the bubble size including a pore splitting are published by The captured grey scale images are post processed using shading
Zhao et al. (2020). They assume that acoustic waves as well as a fluc­ technology and Kalman filtering (Kalman, 1960) to improve the visi­
tuating keyhole can drag away the bubbles from the keyhole to the so­ bility of the phase boundaries and contrasts. Images are recorded with a
lidification area resulting in pore trapping. The acoustic waves occur frame rate of 1000 images per second. In Fig. 2 individual frames of
after the bubble has detached from the keyhole, as it contracts into a three different welding processes are compared as an example. On the
sphere. left, a bead on plate welding of aluminium is shown as a reference. The
Welding aluminium and copper, x-ray analysis additionally enables phase boundary of the melt pool and the keyhole are visible, while the
the visibility of the material mixing due to different density. Leitz (2016) position of the laser beam is indicated. Using aluminium and copper
derives melt pool flows from x-ray videography and metallographic connections, the higher optical density of the copper material prevents
analysis for an aluminium and copper welding process. The results of an visibility inside the copper.
x-ray analysis for aluminium and copper welds are also addressed in this However, in the aluminium material of the dissimilar joints the
paper. In contrast to the existing investigations, synchrotron radiation is keyhole, partially the phase boundary of the melt pool, pores and darker
used in this work with the goal of improving the quality and contrast of veils in the melt pool can be identified. These darker areas aren’t visible
the images. The pore formation and movement as well as the melt pool
dynamics are in the focus of the discussions. To analyse in detail the Table 1
formation mechanisms in dissimilar material welding configurations, Characteristics and parameters of the synchrotron (Hummel et al., 2021).
aluminium and copper are welded in overlap configuration with two
Parameter Unit P07 Petra 3
different infrared laser beam sources of different laser focus diameters
(see Table 2). Operation Mode [-] Low Beta
Photon energy [keV] 37,7
Beam area [mm2] 2×2
2. Experimental setup for laser welding of aluminium and Scintillator material [-] CdWO4
copper with synchrotron radiation Scintillator size [mm2] 11 × 11
Scintillator thickness [µm] 300
Distance scintillator – material sample [mm] 800
The experimental setup for this investigation is based on the usage of

2
S. Hollatz et al. Journal of Materials Processing Tech. 309 (2022) 117738

as well as on the intermixing of the materials.

3.1. Analysis of pore formation and movement

Laser beam

3.1.1. Aluminium on copper welds

The formation of pores is a usual defect for aluminium and copper
X-ray beam
welds. Using synchrotron radiation, the point of origin as well as the
Highspeed-Camera
movement of the pores within the melt pool can be analysed.
Sample Scintillator
From the investigations three main locations for pore formation
Linear motion axis
depending on the welding configuration are identified:

Fig. 1. Schematic of the used synchrotron setup (Wagner et al., 2021).

1. Formation of pores in front of the keyhole at the material transition
layer of aluminium-to-copper-welds.
2. The solidification area in aluminium while welding aluminium on
Table 2 copper connections.
Characteristics of the used laser beam sources. 3. The keyhole ground when welding copper on aluminium.
Parameter Unit YLR-2000WC TruDisk 5000
The first phenomenon is presented in Fig. 3. In front of the keyhole, a
Wavelength [nm] 1070 1030
Laser power [W] 2000 5000
semi-circular bubble is formed, see marked circle. In the image sequence
Fibre diameter [µm] 14 50 the bubble is growing. The movement and solidification of the bubble
Focal length collimator [mm] 70 80 cannot be seen due to the lack of visibility in the copper material. A
Focal length optics [mm] 163 163 possible reason is an uprising movement along the keyhole due to the
Focal diameter (measured) [µm] 34 117
rising metal vapour. It can be assumed that the material transition is an
obstacle for vapour and melt pool flows due to different physical prop­
erties of the materials.
Table 3 Another possible reason for the formation of the pores is the different
Material properties of Al99.5, Cu-ETP and CuSn6 (Hummel et al., 2020; DKI melting temperature of the two materials. Due to the higher melting
Deutsches Kupferinstitut, 2005a, 2005b; Wieland SMH, 2018; Sicius, 2021). point of copper and the higher energy content in the melt, the
Material Property Unit Al99.5 Cu-ETP CuSn6 aluminium melt can overheat locally on contact layer and thus promote
(T = 20 ◦ C) the formation of pores.
Density [g/cm3 ] 2.7 8.9 8.8 Increasing the focal diameter in the same configuration leads to an
Absorption (λ ≈ 1 µm) [%] ≈9 ≈5 ≈ 10 elongation of the melt pool in aluminium. More significant melt pool
Melting point [◦ C] 660 1083 1050
waves can be recognized on the surface in Fig. 4. The single images are
Evaporation Temperature [◦ C] 2470 (Al) 2595 2595
(Cu) (Cu)
showing different welding parameters with the same focal diameter. In
Electrical conductivity [MS/m] 34–36 57 9 the left image the significant number of bubbles in the solidification area
Thermal conductivity [W/m•K- 210–220 394 75 is noticeable. Reviewing the full video sequence, the bubbles float back
1] and forth, while bubbles are rising from the copper. The melt pool in
Coefficient of thermal [10–6•K- 24 17 19
copper is presumably shorter than in the aluminium. Therefore, the
expansion 1]
vertical flows disappear in the solidification area of the aluminium. Most
of the bubbles can’t reach the material surface and resume in the so­
in the aluminium reference, showing material intermixing with the lidified weld seam as pores.
denser copper material. These visible process characteristics are then By increasing laser power and feed rate, this phenomenon cannot be
used to derive explanatory approaches and solution strategies. detected, see Fig. 4 right. For a further investigation in this configura­
tion, a view into the copper is necessary for a detailed discussion of the
3. Results and discussion influence of the spot diameter. At present, differences in the width of the
vapor capillary and the melt pool length can be identified. Pore for­
In the following the results of the experimental study are shown and mation can occur with both beam diameters.
analysed. The investigation is focused on pore formation and movement

Fig. 2. Synchrotron images of the configurations (left: Al99.5 bead on plate, centre: Al99.5 to CuSn6, right: Cu-ETP to Al99.5).

3
S. Hollatz et al. Journal of Materials Processing Tech. 309 (2022) 117738

Fig. 3. Bubble formation in front of the keyhole - aluminium on copper (PL = 750 W, vf = 50 mm/s, df = 34 µm, Δt = 1 ms, related video sequence online available:
https://doi.org/10.5281/zenodo.6411749).

Fig. 4. Floating bubbles in the aluminium melt pool – aluminium on copper (df = 117 µm; left: PL =1000 W, vf = 50 mm/s; right: PL = 1500 W, vf = 100 mm/s).

3.1.2. Copper on aluminium welds movement. The movement direction is changed, and the upward
When changing the top material from aluminium to copper, the pore movement prevented.
formation and movement in the melt pool ground becomes visible, see Additionally, it can be assumed that there is a movement in the
Fig. 5. The bubbles occur in the keyhole ground and are moving along image plane, vertically to the feed direction in frame 4–6. Due to the
the melt pool ground (B). Due to the lower thermal conductivity, the narrower copper melt pool, the copper shades are penetrating centrally
copper melt pool on top is smaller than the aluminium melt pool below. into the aluminium. The forward copper flow pushes the bubble to the
The bubbles, moving along the solidification line, are trapped in the melt pool sidewall where it solidifies as a pore. This hypothesis is sup­
material transition and solidify as pores (A). Some of them detach from ported by the metallographic analysis in Fig. 8 which shows pore for­
the keyhole ground and directly rise to the material transition (C). Both mation at the upper part of the lower joining partner at the outside of the
motion paths show two distinct flow fields within the melt pool (metal weld seam.
vapour induced flow and capillary flow) (Beck, 1996). The presented experimental results require a further investigation to
The cross-section of the weld seam confirms the effect, see Fig. 6. The outline the influence of the laser spot diameter and process parameters
number of pores in the lower aluminium material is significantly higher on the pore formation as well as a visibility with the given observation
than in copper. In the transverse cross-section one of the pores is trapped method inside the copper material.
directly under the unmolten copper material. The longitudinal cross-
section indicates pores along the whole depth of the weld seam.
The intermixing of the materials can also have an influence of the 3.2. Analysis of melt pool dynamics
bubble movement and prevents their rising to the surface. In Fig. 7, the
movement of a pore is marked in an image sequence. The last image The intermixing of the materials is a major concern during the
summarizes the movement path on one specific bubble indicated with welding of aluminium and copper due to the formation of intermetallic
the number of the images and the resulting path geometry. In the images phases. In the following, the movement and mixing of the copper in the
4–6 dark shades (e.g. copper melt) are influencing the bubble aluminium melt pool is analysed. Using synchrotron radiation, new
opportunities in the analysis are available compared to conventional

Fig. 5. Pore formation and movement in the melt pool ground – copper on aluminium (PL = 1000 W, vf = 50 mm/s, df = 117 µm, Δt = 2 ms, related video sequence
online available: https://doi.org/10.5281/zenodo.6411749).

4
S. Hollatz et al. Journal of Materials Processing Tech. 309 (2022) 117738

Fig. 6. Transverse and longitudinal cross-section of the weld seam – copper on aluminium.

Fig. 7. Bubble movement influenced by material intermixing – copper on aluminium (PL = 1500 W, vf = 100 mm/s, df = 117 µm, Δt = 2 ms, related video sequence
online available: https://doi.org/10.5281/zenodo.6411749).

1 is mixed along the whole depth, the copper flow with increased feed
rate is much smaller and closer to the material transition. It is to be
mentioned, that the increase of the feed rate leads to a reduction of the
weld depth. The marked weld depth is determined by analysing the full
video due to the challenging visibility in a single frame.
Images 3 and 4 are welded using a circular spatial power modulation.
Again, the mixing appearance is different compared to frames 1 and 2.
The higher frequency of the oscillation leads to a more homogeneous
mixing of the materials in frame 4. Both comparisons allow the
assumption that due to the higher movement speed of the laser beam on
the material surface and thus of the keyhole in the melt, the flow vectors
Fig. 8. Transverse cross-section to confirm the appearance of pores at the are more strongly aligned in the horizontal direction and thus less
outside of the weld seam – copper on aluminium. mixing of the materials can take place in the vertical direction. On the
other hand, more bubbles along the material transition are visible as
metallographic analysis. The image sequences analysed in this investi­ well. Especially for the high dynamic spatial power modulation a higher
gation are showing a batchwise flow of copper from the backside of the frame rate needs to be achieved for the synchrotron analysis to capture
melt pool, see Fig. 9. The material is flowing towards the keyhole, where every position of the keyhole inside the melt pool. Typical oscillation
it is entrained by the capillary flow and swirled into the aluminium. The frequencies up to 1000 Hz are not covered with the actual set up but are
copper is mixed along the whole depth of the weld seam. subject to further investigations.
The same phenomenon is visible in the experiments with the smaller
spot diameter. Using the parameters mentioned in Fig. 10, the copper
shades are moving towards the keyhole and are pushed to the melt pool 3.3. Discussion of the melt pool movement in aluminium and copper
ground while intermixing with the aluminium. welding
Depending on the laser beam diameter, the images are showing a
significant difference in the appearance of bubbles. While the welding Based on the results, a schematic illustration of the melt flow for
process with df = 117 µm, shown in Fig. 5, Fig. 7 and Fig. 9, lead to aluminium and copper welds can be created, see Fig. 12.
bubbles, the welding process with df = 34 µm has no visible bubble in During evaporation in the keyhole, the metal vapor rises and causes
the presented sequence in Fig. 10. an upward melt flow along the capillary wall. A backward flow of ma­
The welding parameters have a strong influence on the intermixing terial occurs due to the movement of the keyhole through the metal. It
of the materials. In Fig. 11, single frames of four different parameter sets can be assumed that the combination of flow mechanisms creates a
are presented. Frames 1 and 2 differ in the feed rate. While the copper in backward vortex in the copper. The copper is entering the aluminium at
the end of the copper melt pool. Due to the wavy melt pool surface

5
S. Hollatz et al. Journal of Materials Processing Tech. 309 (2022) 117738

Fig. 9. Copper flow towards the keyhole – copper on aluminium (PL = 1000 W, vf = 50 mm/s, df = 117 µm, Δt = 1 ms, related video sequence online available:
https://doi.org/10.5281/zenodo.6411749).

Fig. 10. Copper flow with smaller laser spot diameter – copper on aluminium (PL = 750 W, vf = 50 mm/s, df = 34 µm, Δt = 1 ms, related video sequence online
available: https://doi.org/10.5281/zenodo.6411749).

Fig. 11. Parameter-dependent copper flow (1 (Basic parameters): PL = 750 W, vf = 50 mm/s, df = 34 µm, 2: vf = 100 mm/s, 3: As = 0,2 mm, fs = 250 Hz, 3: As
= 0,2 mm, fs = 500 Hz).

characteristic the batchwise copper flow is induced. The higher density Especially the influence of different parameters, as indicated by
of copper supports the penetration of the copper melt into the doubling the feed rate in Fig. 11, has a major influence on the melt flow.
aluminium. (Leitz, 2016) explained the vertical backward flow with the With regard to a comparison of the two beam diameters, it can be stated
evaporation pressure that accelerates the melt in front of the keyhole. By that the batchwise penetration of the copper melt in the direction of the
reaching the keyhole, the copper shade is accelerated to the melt pool keyhole occurs in both cases but differs in terms of the intensity and size
ground and moves along the solidification border to the back of the melt of the copper stream. To gain further information about the melt flow,
pool. Comparable to the bubble movement the copper flow is deflected the visibility of the copper melt pool in the x-ray analysis should also be
at the material transition due to the smaller copper melt pool. A mixing achieved in further investigations.
along the whole depth is visible.
The illustrated melt flows are approaches for an explanation of the 4. Conclusion
material mixing. The high dynamics of the flows in the melt pool lead to
complex flow situations, which are shown in simplified form in Fig. 12. In conclusion, it was shown that high energy synchrotron radiation is
For the confirmation of the hypotheses, further investigations with able to give an insight into the welding process of aluminium and cop­
higher temporal resolution of the imaging method are necessary. per. The approach can help to understand the melting and mixing

6
S. Hollatz et al. Journal of Materials Processing Tech. 309 (2022) 117738

Declaration of Competing Interest

vf
The authors declare the following financial interests/personal re­
Laser beam lationships which may be considered as potential competing interests:
Marc Hummel reports equipment, drugs, or supplies was provided by
IPG Photonics Corp. Marc Hummel reports equipment, drugs, or sup­
Keyhole plies was provided by TRUMPF Laser GmbH.

Data Availability

Melt pool boundary Links to high speed videos are included in the paper.

Copper melt flow Acknowledgements

The presented investigations were carried out at RWTH Aachen

University within the framework of the Collaborative Research Centre
SFB1120–236616214 “Bauteilpräzision durch Beherrschung von
Fig. 12. Schematic illustration of the melt pool flows. Schmelze und Erstarrung in Produktionsprozessen” and funded by the
Deutsche Forschungsgemeinschaft e.V. (DFG, German Research Foun­
behaviour of the materials as well as the pore formation considering the dation). The sponsorship and support are gratefully acknowledged.
different parameters applied. The following observations can be derived This research was supported by TRUMPF GmbH & Co. KG and IPG
from this work: Laser GmbH. We would like to thank all people involved for their sup­
port. The presented investigations were carried out in cooperation with
• The different material properties and thus different melt pool ex­ DESY in Hamburg at PETRA III and we would like to thank F. Beckmann
tensions in dissimilar material joints cause a difference in the melt and J. Moosmann for assistance in using P07 EH4.
pool length of copper and aluminium. The uprising bubbles are
trapped at the material transition area and remain as pores in the References
joint.
• Regarding the intermixing of the materials a batchwise penetration Balu, Prabu; Carlson, Blair; Kovacevic, Radovan (2011): An Investigation into the Laser
of the copper into the aluminium could be observed in the copper Micro-Welding of Aluminum and Copper in Lap Joint Configuration. In TMS annual
Meeting (Vol. 3), pp. 295–307. doi: 10.1002/9781118062173.ch37.
and aluminium joint. Bantel, C., 2018. Laserstrahlschweißen von Aluminium-Kupfer-Verbindungen.
• Changing the welding parameters, such as laser power, feed rate, Dissertation: Technische Universität Dresden.
spot diameter or spatial power modulation, can lead to a massive Beck, Markus, 1996. Modellierung des Lasertiefschweißens. Teubner (Laser in der
Materialbearbeitung Forschungsberichte des IFSW), Zugl.: Stuttgart, Univ., Diss.
change of the melt flows and intermixing. In the parameter range Stuttgart.
used, an increase in the feed rate leads to a reduced penetration Brand, Martin J., Schmidt, Philipp, A., Zaeh, Michael, F., Jossen, Andreas, 2015. Welding
depth of the copper into the aluminium below. techniques for battery cells and resulting electrical contact resistances. J. Energy
Storage 1, 7–14. https://doi.org/10.1016/j.est.2015.04.001.
• The observed penetration of the copper into the aluminium occurs Cunningham, R., Zhao, C., Parab, N., Kantzos, C., Pauza, J., Fezzaa, K., Rollett, A.D.,
with both beam diameters used. The number and size of pores might 2019. Keyhole threshold and morphology in laser melting revealed by ultrahigh-
be lower with the smaller spot diameter when aluminium is the speed x-ray imaging. Science 363 (6429), 849–852. https://doi.org/10.1126/
science.aav4687.
bottom material. Das, Abhishek, Li, Dezhi, Williams, David, Greenwood, David, 2018. Joining
technologies for automotive battery systems manufacturing. WEVJ 9 (2), 22. https://
To reduce the appearance of intermetallic phases in these dissimilar doi.org/10.3390/wevj9020022.
Dimatteo, Vincenzo, Ascari, Alessandro, Fortunato, Alessandro, 2019. Continuous laser
joints, the influence of the parameters can be investigated in future
welding with spatial beam oscillation of dissimilar thin sheet materials (Al-Cu and
experiments. Upcoming studies will investigate the influence on Cu-Al): process optimization and characterization. J. Manuf. Process. 44, 158–165.
reducing the intermetallic phases by adapting process parameters ac­ https://doi.org/10.1016/j.jmapro.2019.06.002.
cording to the results of the x-ray observation. To increase the precision DKI Deutsches Kupferinstitut (2005a): Werkstoff-Datenblatt: Cu-ETP. Available online at
〈https://www.kupferinstitut.de/wp-content/uploads/2019/11/Cu-ETP.pdf〉,
of the presented analyzation method, future improvements must deal checked on 7/23/2021.
with the frame rate and the visibility of melt pool in copper. This allows DKI Deutsches Kupferinstitut (2005b): Werkstoff-Datenblatt: CuSn6. Available online at
differences between the two material arrangements to be analysed. The 〈https://www.kupferinstitut.de/wp-content/uploads/2019/11/CuSn6–1.pdf〉,
checked on 2/14/2022.
increasing understanding of the formation of pores and intermetallic Fetzer, F.; Berger, P.; Hu, H.; Weber, R.; Graf, T. (2018): Pores in laser beam welding:
phases enables the development of specific compensation methods such generation mechanism and impact on the melt flow. In Proc. SPIE 10525, High-
as beam shaping, temporal or spatial power modulation. A deep process Power Laser Materials Processing: Applications, Diagnostics, and Systems VII
105250D. doi: 10.1117/12.2295798.
knowledge is the basis for the industrial use of aluminium and copper Hügel, Helmut, Graf, Thomas, 2009. Laser in der Fertigung, Vol. 2. Vieweg+ Teubner,
connections in electromobility. Wiesbaden.
Hummel, Marc, Schöler, Christoph, Häusler, André, Gillner, Arnold, Poprawe, Reinhart,
2020. New approaches on laser micro welding of copper by using a laser beam
CRediT authorship contribution statement source with a wavelength of 450 nm. J. Adv. Join. Process. 1 (100012) https://doi.
org/10.1016/j.jajp.2020.100012.
S. Hollatz: Conceptualization, Methodology, Validation, Investiga­ Hummel, Marc Daniel; Halm, Ulrich; Hagenlocher, Christian; Lind, Jannik; Hollatz, S.ö
ren; Häusler, André et al. (2021): Temporally and spatially highly resolved
tion, Writing – original draft. M. Hummel: Conceptualization, Meth­
reconstruction of vapor capillary geometry during laser beam welding using
odology, Investigation, Writing – review & editing. A. Olowinsky: synchrotron radiation. Edited by Lasers in Manufacturing Conference 2021.
Conceptualization, Methodology, Investigation, Writing – review & München (RWTH-2021–05994). Available online at 〈https://publications.rwt
editing. A. Gillner: Conceptualization, Supervision, Project adminis­ h-aachen.de/record/821069〉.
Kalman, R.E., 1960. A new approach to linear filtering and prediction problems. J. Basic
tration, Funding acquisition. F. Beckmann: Methodology, Investiga­ Eng. 82 (1), 35–45. https://doi.org/10.1115/1.3662552.
tion. J. Moosmann: Methodology, Investigation. Kraetzsch, M., Standfuss, J., Klotzbach, A., Kaspar, J., Brenner, B., Beyer, E., 2011. Laser
beam welding with high-frequency beam oscillation: welding of dissimilar materials

7
S. Hollatz et al. Journal of Materials Processing Tech. 309 (2022) 117738

with brilliant fiber lasers. Phys. Procedia 12, 142–149. https://doi.org/10.1016/j. Schmidt, Philipp A.; Schweier, Markus; Zaeh, Micheal F. (2012): Joining of lithium-ion
phpro.2011.03.018. batteries using laser beam welding: Electrical losses of welded aluminum and copper
Lee, S.Shawn; Kim, Tae H.; Hu, S.Jack; Cai, Wayne W.; Abell, Jeffrey A. (2010): Joining joints. In ICALEO 2012, pp. 915–923. doi: 10.2351/1.5062563.
Technologies for Automotive Lithium-Ion Battery Manufacturing: A Review. In Schmitt, F., 2012. Laserstrahl-Mikroschweißen mit Strahlquellen hoher Brillanz und
Proceedings of the ASME 2010 International Manufacturing Science and Engineering örtlicher Leistungsmodulation. Dissertation. RWTH Aachen University. Aachen:
Conference, pp. 541–549. doi: 10.1115/MSEC2010–34168. Shaker.
Leitz, Andreas (2016): Laserstrahlschweißen von Kupfer- und Aluminiumwerkstoffen in Sicius, Hermann, 2021. Handbuch der chemischen Elemente. Springer Berlin Heidelberg,
Mischverbindung. Dissertation, Universität Stuttgart. München: Herbert Utz Verlag Berlin, Heidelberg.
(Laser in der Materialbearbeitung - Forschungsberichte des IFSW). Wagner, Jonas, Hagenlocher, Christian, Hummel, Marc, Olowinsky, Alexander,
Miyagi, Masanori, Kawahito, Yousuke, Kawakami, Hiroshi, Shoubu, Takahisa, 2017. Weber, Rudolf, Graf, Thomas, 2021. Synchrotron X-ray analysis of the influence of
Dynamics of solid-liquid interface and porosity formation determined through x-ray the magnesium content on the absorptance during full-penetration laser welding of
phase-contrast in laser welding of pure Al. J. Mater. Process. Technol. 250, 9–15. aluminum. Metals 11 (5), 797. https://doi.org/10.3390/met11050797.
https://doi.org/10.1016/j.jmatprotec.2017.06.033. Wieland SMH (2018): Werkstoff-Datenblatt: EN AW-1050A. Available online at 〈htt
Rabkin, D.M., Ryabov, V.R., Lozovskaya, A.V., Dovzhenko, V.A., 1970. Preparation and ps://www.wieland-smh.de/files/smh/wieland_smh/dat%20sheets/EN-AW-1050A
properties of copper-aluminum intermetallic compounds. Powder Met. Met. Ceram. _DE.pdf〉, checked on 7/23/2021.
9, 695–700. https://doi.org/10.1007/BF00803820. Xu, J., Rong, Y., Huang, Y., Wang, P., Wang, C., 2018. Keyhole-induced porosity
Schell, Norbert; King, Andrew; Beckmann, Felix; Fischer, Torben; Müller, Martin; formation during laser welding. J. Mater. Process. Technol. 252, 720–727.
Schreyer, Andreas (2013): The High Energy Materials Science Beamline (HEMS) at Zhao, C., Parab, N.D., Li, X., Fezzaa, K., Tan, W., Rollett, A.D., Sun, T., 2020. Critical
PETRA III. In Mechanical Stress Evaluation by Neutrons and Synchrotron Radiation instability at moving keyhole tip generates porosity in laser melting. Science 370
VI 772, pp. 57–61. doi: 10.4028/www.scientific.net/MSF.772.57. (6520), 1080–1086. https://doi.org/10.1126/science.abd1587.