Process studies on copper laser beam welding over gap by using disc laser at green wavelength

Process studies on copper laser beam welding over gap by using disc laser at green wavelength

Journal Pre-proof Process studies on copper laser beam welding over gap by using disc laser at green wavelength W.-S. Chung , A. Olowinsky , A. Gilln...

1MB Sizes 0 Downloads 11 Views

Journal Pre-proof

Process studies on copper laser beam welding over gap by using disc laser at green wavelength W.-S. Chung , A. Olowinsky , A. Gillner PII: DOI: Reference:

S2666-3309(20)30007-8 https://doi.org/10.1016/j.jajp.2020.100009 JAJP 100009

To appear in:

Journal of Advanced Joining Processes

Received date: Revised date: Accepted date:

18 November 2019 28 January 2020 28 January 2020

Please cite this article as: W.-S. Chung , A. Olowinsky , A. Gillner , Process studies on copper laser beam welding over gap by using disc laser at green wavelength, Journal of Advanced Joining Processes (2020), doi: https://doi.org/10.1016/j.jajp.2020.100009

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 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/)

Process studies on copper laser beam welding over gap by using disc laser at green wavelength W.-S. Chung1, A. Olowinsky1, A. Gillner1, 2 1

Fraunhofer Institute for Laser Technology ILT, 52074 Aachen, Germany 2

Chair of Laser Technology LLT, 52074 Aachen, Germany

The increasing demand for the substitution of the internal combustion engine vehicles to the battery electric vehicles requires beside battery cells high performance power electronic devices such as power control units (PCU). However, a combined requirement of high junction temperature stability and a large joint area of the interconnection on the PCU is a challenge for the conventional joining method such as soldering and wire bonding process. The Laser Impulse Metal Bonding (LIMBO) process enables a high temperature stable weld joint and large joint area. During the LIMBO process only minimized thermal stress is induced into the underlying substrate by a spatial separation between both joining partners in an overlap configuration with a gap. Hence, an energetic separation between the melting and joining phase is given.

In this paper, the LIMBO process is firstly investigated with the disc laser at wavelength λ = 515 nm. Due to the enhanced absorptivity of the laser beam at this wavelength on copper material, the process duration of the LIMBO process is about the half compared to the LIMBO process with wavelength λ = 1064 nm.

Keywords: LIMBO; Green laser beam; Laser beam micro joining; Heat conduction welding mode; Shadow projection

1. Introduction The laser beam welding process of copper material with a wavelength at λ = 1064 nm is a well-known challenge due to the high reflectivity R = 97 % and high thermal conductivity 𝑘 = 400 W/mK in ambient atmosphere [3]. The high reflectivity and therefore a low absorptivity of the laser beam leads to instable energy coupling into the material and the high thermal conductivity impede the local heat accumulation. Hence, the low absorptivity of copper material is most crucial for the spot welding process, especially for the heat conduction welding mode. In order to enhance the absorptivity of the laser beam at the surface, several methods for the modification of the copper surface such as pre-oxidization [8, 9] and roughened surface [4, 7] has been investigated. However, as soon as the modified surface melts and vaporizes, the beneficial of enhanced absorptivity due to the surface modification is not given anymore. As a consequence, a stable melt pool formation is difficult to the point of keyhole welding mode.

The main approach of the LIMBO process is to induce minimal thermal load on the underlying joining partner due to the energetic separation of melting and joining stage. Both joining partners are therefore spatially separated with a defined gap in between. The precise differentiation of melting and joining stage with the temporal laser beam modulation is the key for successful welding performance. However, the LIMBO process also confronts mentioned afore difficulties with the laser beam in wavelength with the infrared regime. In this paper the LIMBO process is investigated with the enhanced laser

beam absorptivity on the copper substrate by utilizing the laser beam source with λ = 515 nm to demonstrate a more efficient LIMBO process.

2. State of the art

2.1 Laser Impulse Metal Bonding (LIMBO) process

The LIMBO process is a spot welding process which is executed by three different stages in order to minimize the thermal load on the underlying substrate of the joining configuration. In addition, a defined gap between both joining partners in overlap configuration is required as shown in the figure 1 which shows a schematic representation of the LIMBO process.

1. Pre-heating stage Laser beam

Upper and lower joining partner

2. Deflection stage ‘Lens-like’ melt

3. Joining stage

Metal vapour plume

Melt pool

Weld joint

Gap

Laser beam power P1

Laser beam power P1 < P2

Laser beam power P2

Figure 1: Schematic representation of the LIMBO process [1]

The upper joining partner is melted in the first stage of the LIMBO process, the so called ‘pre-heating stage’, with laser power P1 by heat conduction welding mode to obtain a ‘melt pool at the lower side of the upper joining partner. In the following ‘deflection stage’, the melt is accelerated towards underlying substrate over a gap due to the sudden vaporization of the melt surface from the upper joining partner. The sudden vaporization

is attributed to the metal plume recoil pressure which is induced by the temporal laser beam power modulation to P2 for a short duration of time. As soon as the accelerated melt due to the recoil pressure bridges the foreseen gap, the melt contacts the underlying substrate and consequently wets on the surface of underlying joining partner. After the wetting, the storage internal heat in the melt and the further energy input from the laser beam irradiation melts the lower joining partner. During the joining stage, the underlying substrate is not directly irradiated by the laser beam nor penetrated through the keyhole. Consequently, a weld joint is formed between both joining partners.

2.2 Absorptivity of the copper material At a wavelength of λ = 1064 nm, copper material represents a major challenge due to its low absorptivity of approximately 3 % at ambient conditions and its high thermal conductivity of approx. 𝑘 = 400 W/mK [3]. Due to the limited absorptivity of the copper material, a small absolute change in the absorptivity value leads to relatively large influence on the welding process compared to materials with a high absorptivity such as steel. On technical surfaces of the metallic material, the absorptivity 𝛼 varies by ± 1 %, which for copper materials with an absorptivity of 3 % at the wavelength λ = 1070 nm corresponds to a large fluctuation of the absorptivity by 66.7 %. [4] On the other hand, the laser beam with the wavelength at λ = 532 nm has a high absorptivity with over 50 %. Hence, the process efficiency is expected to be increased when the laser radiation of wavelength λ = 532 nm is utilized instead of infrared laser. As the absorptivity is dependent to the welding mode, a direct comparison of 532 nm wavelength and 1064 nm wavelength is investigated by [6] in terms of spot welding process (figure 2).

70 60

λ = 532 nm, ds=200 µm; λ = 1064 nm, ds=25 µm

ds = 1000 µm

58.1

0.6

Absorptivity 𝛼

55.9 50

49.9

2.0

0.7

40

32.2

8.3

30

15.7

20 9.3

10 0

No-melting

4.7

0.4

Heat conduction

Pulse duration : t = 1.2 ms Specimen : C1020 (t = 1.00 mm) Laser beam spot diameter: ds

Keyhole 532 nm 1064 nm

Figure 2: Absorptivity for both 532 nm and 1064 nm in no-melting, heat conduction and key-hole welding mode according to [6]

As a result, the absorptivity at 532 nm shows for all welding modes higher values, especially at solid state of the copper material at the ‘no-melting’ mode. The absorptivity of the laser beam with λ = 532 nm at heat conduction welding mode shows lowest value with 49.9 ± 0.7% compared to other welding modes. However, the absorptivity value of the laser beam with λ = 532 nm for all welding mode value is sufficient to neglect the variation of the absorptivity of ± 1% due to the surface state. Hence, more stable energy coupling is expected with the green laser beam for copper material.

2.3 Heat conduction welding mode of the LIMBO process

The heat conduction welding mode of the LIMBO process is discussed based on the geometrical value such as the melt pool diameter ratio ζ. The geometrical value ζ provides informations about the welding mode during the pre-heating stage where only

heat conduction welding is applied. When a keyhole welding mode occurs during the preheating stage, the laser beam directly irradiates through the keyhole. As a consequence, the melt volume will be ejected in form of spatter and the underlying substrate will be damaged due to the direct laser beam irradiation. The conservation of melt pool is given for the heat conduction welding mode and therefore the melt pool diameter at the lower side of substrate db cannot be larger than the melt pool diameter at the upper side of the substrate dt. The ratio between the melt pool diameters, the melt pool diameter ratio ζ is investigated for the reproducibility of heat conduction welding mode (figure 3). [2]

The bead-on-plate welding is performed on the upper joining partner and the determination of mentioned melt pool diameters is based on the post measurement method with a microscope. The average size of the melt pool diameter at the upper side of substrate dt is dt_avg = 333 µm for all values in figure 6. Also, the average size of the melt pool diameter at the lower side of substrate db is db_avg = 199 µm for all corresponding values in figure 6.

Laser beam Melt pool

dt

db

Melt pool diameter ratio: ζ = with 0 < ζ < 1 for heat conduction welding

Figure 3: Schematic representation of the melt pool diameter ratio according to [2]

3 Methods and experimental set up

3.1 Methods

The process study for the laser beam welding over a gap with the laser beam source at wavelength λ = 515 nm is divided into establishing parameter combinations for the preheating stage and the deflection stage. The pre-heating stage is firstly investigated with a wide range of parameter combinations in order to find parameter combinations for reproducible heat conduction welding mode for melting the upper joining partner over the material thickness. When the material is molten dominantly by heat conduction welding mode, a ‘lens-like’ melt formation at the lower side of the upper joining partner is given.

After determining the process parameter for the pre-heating stage, the temporal laser beam power modulation is applied to increase the laser beam power during the deflection stage. The deflection of the melt pool should be sufficient to bridge the gap between both joining partners and guarantee a certain residual time for the melt to penetrate into the underlying metallization. However, the residual time of the melt should not exceed the thermal threshold for the underlying thermally sensitive substrate.

3.2 Experimental set up The laser beam source TRUMPF TruDisk Pulse 421 with a wavelength of λ = 515 nm is used for this experimental work. The maximum average power is Pavg = 400 W and the peak power is Ppeak = 4 kW. The limited maximum pulse length is set to t = 50 ms, as the maximum pulse energy of Epulse= 40 J cannot be exceeded. The experimental setup is shown in figure 4.

Beam deflection system Beam collimation

High speed camera Shadow projection F-theta lens

Lower side of the interconnector Gap Upper side of the copper layer

Laser beam with λ = 515 nm

z y

Weld joint

x

Gap

z

y

x

Interconnector and FR4 Substrate with Cu layer Diode laser with λ = 940 nm

Figure 4: Experimental set up and exampling image of the weld joint shadow projection

The laser beam source is coupled to an optical fiber which has a diameter of 100 µm. At the fiber exit, the laser radiation has a BPP of 4 mm*mrad which is connected to a collimator before the scanner. The displacement of the beam can be achieved by using the galvanometer scanner (beam deflection system in figure 4). The spot diameter of the laser beam after a f-theta lens with a focal length of 163 mm is ds = 238 µm at the processing plane respectively.

The upper joining partner used for this experiment is CuFe2P with a = 200 µm thickness without any surface treatment. For the underlying joining partner, a glass fiber reinforced epoxy resin with a thin copper layer is selected. A thin copper layer above the glass fiber reinforced epoxy resin has the thickness of a = 100 µm. A gap between both the joining partners is realized by using a spacer with a thickness of a = 120 µm.

To observe the melt dynamics between the gap, a high speed camera is used to capture the shadow projection as shown in figure 4. A strong light source in form of a laser beam source is required to generate a shadow projection of the deflected melt which is aligned

coaxially to the high speed camera. The diode laser beam source with the wavelength of λ = 940 nm with Pmax = 50 W is used for this set up and the frame rate of the high speed camera is set to f = 100,000 fps.

4 Results

4.1 Process study for the pre-heating stage

The pre-heating stage is executed dominantly by heat conduction welding mode to obtain a ‘lens-like’ melt at the lower side of the thin copper plate. At focal position of the laser beam, 22 parameter combinations are investigated in order to generate a ‘lens-like’ melt and the results are shown in figure 5 below. Furthermore, the applied parameters are listed in table 1 below.

Laser power (P)

Pulse duaration (t)

Energy E

Spot size (d)

Intensity (I)

Evaluation

[W]

[ms]

[J]

[µm]

[MW/cm2]

[-]

1

250

10

2.50

238

0.56

No `lens-like´ melt

2

250

13

3.25

238

0.56

No `lens-like´ melt

3

250

15

3.75

238

0.56

No `lens-like´ melt

4

300

8

2.40

238

0.67

No `lens-like´ melt

5

300

10

3.00

238

0.67

No `lens-like´ melt

6

300

12

3.60

238

0.67

No `lens-like´ melt

7

350

8

2.80

238

0.79

No `lens-like´ melt

8

350

10

3.50

238

0.79

No `lens-like´ melt

9

350

12

4.20

238

0.79

No `lens-like´ melt

10

400

8

3.20

238

0.90

No `lens-like´ melt

11

400

10

4.00

238

0.90

Reproducible

12

400

12

4.80

238

0.90

Reproducible

13

400

9

3.60

238

0.90

Reproducible

14

450

8

3.60

238

1.01

Unreproducible

15

450

10

4.50

238

1.01

Reproducible

16

450

12

5.40

238

1.01

Reproducible

17

500

5

2.50

238

1.12

Unreproducible

18

500

6

3.00

238

1.12

Reproducible

19

500

7

3.50

238

1.12

Melt ejection

Parameter combination

1,4

No 'lens-like' melt Unreproducible Reproducible ejectionMelt ejection

500

8

4.00

238

1.12

Melt

21

500

10

5.00

238 1,2

1.12

Melt ejection

22

550

8

4.40

1.24

Melt ejection

Intensity [MW/cm²]

20

238 1,0

Table 1: Investigated parameters for the pre-heating stage

1,2

Intensity [MW/cm²]

Intensity [MW/cm²]

Power density at focal position I [MW/cm²]

0,8

0,6

0,4 2,0

0,6

0,6

0,4 3,0

0,4

4,0

4,5

5,0 2,0

Energy [J] 2,0

0,6

3,5

2,5

2,5

3,0

3,5

4,0

4,5

5,0

5,5

6,0

Energy [J]

0,8

2,5

2,0

0,8

1,0

1,0

0,8

0,4 1,0

1,2

No 'lens-like' melt Unreproducible Reproducible Melt ejection

No 'lens-like'melt melt No 'lens-like' Unreproducible Unreproducible Reproducible Reproducible Melt ejection Melt ejection

1,2

1,4

1,0

1,2

1,4

0,6

Intensity [MW/cm²]

1,4

No 'lens-like' melt Unreproducible Reproducible Melt ejection

0,8

1,4

3,0

5,5 2,5

6,0 3,0

3,5

4,0

4,5

5,0

5,5

6,0

Energy [J] 3,5

4,0

4,5

5,0

5,5

6,0

Energy [J]

0,4 2,0

2,5

3,0

3,5

4,0

4,5

5,0

5,5

6,0

Energy E [J]

Figure 5: Investigated parameter combination for the pre-heating stage

In total, 22 parameter combinations have been investigated with five repetitions each. The definition of ‘no lens-like melt’ indicates that the penetration depth of the heat conduction welding during the pre-heating stage is lower than the material thickness. So, no ‘lens-like’ melt is observed. The definition of ‘unreproducible’ is when less than five ‘lens-like’ melts are generated with given parameter for five repetitions. The definition of the ‘reproducible’ is when a ‘lens-like’ melt is given for all five repetition. Furthermore, the definition of ‘melt ejection’ means that at least one of the conducted pre-heating stage has led to keyhole welding mode. Consequently, the melt pool diameter ratio is above the value 1 (refer to sub-section 2.3).

The laser beam intensity value between 0.9 MW/cm² and 1.1 MW/cm² poses to be suitable to penetrate a melt over 200 µm dominantly with the heat conduction welding mode. The laser beam energy value between 3 J and 5.5 J is tolerated for the pre-heating stage combined with the mentioned intensities. As for the investigated parameter combinations, a wide range of the laser beam energy value leads to ‘lens-like’ melt generation for the lower laser beam intensity I = 0.9 MW/cm². On the other hand, only one laser beam energy value is valid for the pre-heating stage at higher laser beam intensity I = 1.12 MW/cm². A further reduction of the laser beam energy for the heat conduction welding mode can be expected when a sufficient laser beam intensity is applied. However, the enhanced laser beam intensity is not only advantageous for the LIMBO process due to the separated stages of LIMBO process. After the pre-heating stage, the melt pool should bridge the gap in the following deflection stage with higher laser beam power. Hence, the high laser beam power required in the deflection stage may damage the underlying substrate due to the higher internal energy applied in the process stage beforehand.

The process parameter study shows, that six parameter combinations are qualified for the pre-heating stage. The ‘reproducible’ parameter combinations are compared with the melt pool diameter ratio in figure 6. Equally to the previous investigation, five repetitions are conducted for each melt pool diameter ratio value. For every measured melt pool diameter value, the laser beam intensity and energy value are assigned.

Melt pool diameter ratio z [-]

1,0 0,8 I = 1.12 MW/cm²

0,6

I = 1.01 MW/cm² I = 1.01 MW/cm²

0,4

I = 0.9 MW/cm² I = 0.9 MW/cm²

I = 0.9 MW/cm²

0,2 0,0 3,0

3,5

4,0

4,5

5,0

5,5

Energy E [J]

Figure 6: Melt pool diameter ratio for reproducible parameter combinations

Between the investigated parameter combinations, three measurements show low standard deviation value where another three measurements show relatively high standard deviation. The laser beam intensity value correlates with the melt pool diameter ratio. As the melt pool diameter ratio increases, the expected weld joint diameter will be also larger. When higher laser beam intensity is applied, the melt pool diameter ratio value over 0.5 is achieved. On the one hand, the deviation between each melt pool ratio increases as the applied laser beam intensity enhances. On the other hand, the laser beam intensity lower than I = 1 MW/cm² leads to independent melt pool ratio value even for different amount of the applied laser beam energy value. Furthermore, the deviation between all investigated melt pool diameter ratios is 0.038 respectively. The lowest standard deviation value with 0.005 is achieved with lowest laser beam energy and intensity value respectively. Since the main idea of the LIMBO process is to minimize the thermal load to the underlying substrate, lower energy amount during the pre-heating stage storage in the upper joining partner is advantageous. Therefore, the parameter

combination with lowest energy value E = 3 J (with I = 1.12 MW/cm²) is chosen for the further LIMBO process.

4.2 Influence of the laser beam power for the deflection and joining stage

In the deflection stage, the temporal laser beam power modulation leads to sudden vaporization of the molten copper surface. Consequently, the metal plume recoil pressure induced by the sudden vaporization deflects the melt pool towards underlying substrate. In a previous study of the LIMBO process with a wavelength of λ = 1064 nm, the laser beam power over P = 2000 W is needed to deflect the melt with a duration of t = 0.3 ms [5].

However, due to high absorptivity of copper material with the laser beam with λ = 515 nm, the laser beam power for the deflection stage is applied with P = 600 W constantly instead of P = 2000 W at laser beam with λ = 1064 nm. The influence of the laser beam duration in the deflection stage is investigated on the weldability in terms of LIMBO process. The deflection stage is conducted with five repetitions. The corresponding interface joint diameter d I is shown in figure 7. The interface joint diameter is a geometrical value which describes the joint diameter between the solidified melt pool and the underlying substrate surface over a gap (refer to figure 8 and figure 9).

600

Interface joint diameter dI [µm]

Invalid value due to melt pool destruction

0,48 J

400

0,18 J 0,3 J

200

0,165 J

0

For all result with I = 1.35 MW/cm² 0,0

0,1

0,2

0,3

0,4

0,5

Energy E [J]

Figure 7: Interface joint diameter value after the deflection and joining stage

The interface joint diameter dI is also an indicator to distinguish whether the deflected melt has formed a weld joint between both joining partners. The interface joint diameter value 0 indicates that the applied parameter combination for the deflection stage is not sufficient to bridge the gap between both joining partners. However, when the applied laser beam energy exceeds the threshold for the deflection stage, the melt pool between the gap destructs and leads also to weld joint failure. The destruction of the melt pool during the deflection stage occurs when the deflected melt is overheated due to high laser beam energy and the melt pool cannot maintain its given geometry by the surface tension anymore.

The laser beam energy value below E = 0.18 J (t = 0.3 ms) shows an interface joint diameter of 0 µm. At E = 0.18 J the melt pool is successfully deflected towards the underlying substrate and forms a weld joint with an interface joint diameter of dI = 293.9 ± 62.9 µm respectively. The further laser beam irradiation leads to comparable

interface joint diameter values. When a higher laser beam energy values with E = 0.48 J (t = 0.8 ms) is applied for the deflection stage, the interface joint diameter value increases but a destruction of the melt pool has been observed twice within five repetitions. The measured interface diameter value in figure 7 is representative for three interface diameter values which have been successfully performed. Thus, the weldability of the deflection stage with the laser beam energy value E = 0.48 J is not guaranteed. To sum up, the laser beam energy value between 0.18 J < E < 0.48 J (0.3 < t < 0.8 ms) with the laser beam intensity value I = 1.35 MW/cm² is suitable for the deflection stage.

A deflection stage and the joining stage are visualized with the shadow projection in figure 8 by using a laser beam energy of E = 0.18 J and a laser beam intensity of I = 1.35 MW/cm².

dI

Figure 8: Image sequence of the shadow projection during the deflection and joining stage

The timeline for the shadow projection is set to t = 0 ms for the moment when the ‘lenslike’ melt is deflected towards the underlying substrate. At t = 0.04 ms the deflected melt wets the underlying substrate surface and wets directly on the surface. With further

irradiation of the laser beam, a symmetrical melt geometry is given for t = 0.09 ms which continuously increases its volume until t = 0.27 ms. When the laser beam irradiation is shut off after t = 0.3 ms, the melt volume solidifies in a symmetrical weld joint geometry with an interface joint diameter dI = 373.94 µm respectively. The cross section of the weld joint from the shadow projection in figure 8 is shown in figure 9 left. In addition, the applied pulse duration and corresponding laser beam pulse form for this paper is shown in figure 9 right below. 800

Laser beam power P [W]

700

100 µm

λ =515 nm ds = 238 µm

Pulse for the deflection stage Pulse for the pre-heating stage

600

Pdeflection stage = 600 W tdeflection stage = 0.3 ms

500

400 300

Ppre-heating stage = 500 W tpre-heating stage = 6 ms

200 100 0 0

1

2

3

4

5

6

7

8

9

10

Pulse duration t [ms]

Figure 9: Left: Weld joint cross section of the LIMBO process; Right: Pulse duration and pulse form for successful LIMBO process with a wavelength λ = 515 nm

5 Conclusion and outlook

The Laser Impulse Metal Bonding process has been investigated with the laser beam source at the wavelength of λ = 515 nm. The copper material shows higher absorptivity of the laser beam at wavelength λ = 515 nm compare to the laser beam at a wavelength λ = 1064 nm. The high absorptivity enables to process the copper material with the spot welding process even for the heat conduction welding mode without any additional surface treatment.

From 22 parameter combinations, the reproducible ‘lens-like’ melt is achieved for six parameter combinations for the pre-heating stage. The chosen six parameter combinations are compared with the melt pool diameter ratio and required laser beam energy. As a result, the laser beam energy of E = 3 J with the laser beam intensity value I = 1.12 MW/cm² is chosen.

After the pre-heating stage, the melt pool is deflected towards underlying substrate due to temporal laser beam power modulation. The laser beam power is enhanced in the deflection stage to P = 600 W and the only different duration of the deflection stage is investigated in terms of weldability. The laser beam energy value between 0.18 < E <0.48 J with laser beam intensity laser beam intensity value I = 1.35 MW/cm² leads to successful weld joint over a gap of 120 µm.

The total duration of the LIMBO process with the laser beam source at wavelength λ = 515 nm is t = 6.3 ms respectively. The total duration of the LIMBO process with a laser beam source at wavelength λ = 1064 nm is t = 17.8 ms [1]. A reduction of the process duration compare to the LIMBO process with the laser beam source at wavelength λ = 1064 nm is achieved.

As for the further investigation, the expansion of the weld joint with overlap welding process should be investigated. For this purpose, single LIMBO weld joints are overlapped to each other according to [2]. Also, the deflection stage should be investigated with temporal laser beam spot size modulation which allows to enhance only the laser beam intensity value according to [1]. Furthermore, the LIMBO process should be investigated with the laser beam source at wavelength λ = 450 nm, to see whether the

process duration can be further reduced due to an absorptivity on copper material of more than 60 %.

6 Reference [1] S. W. Britten: “Bauteilschonende Verbindungstechnik auf Metallisierungen durch moduliertes Laserstrahl-schweißen“ ed. by Apprimus Verlag (Publisher, Aachen, 2017) [2] W. Chung, A. Haeusler, A. Olowinsky, A. Gillner, R. Poprawe: JLMN, 13, 2, (2018) [3] Ramsayer, R.M.: Prozessstabilisierung beim gepulsten Laserstrahl-Mikroschweißen von Kupferwerkstoffen. Dissertation, Rheinisch-Westfälischen Technischen Hochschule (2004) [4] Heß, A.: Vorteile und Herausforderungen beim Laserstrahlschweißen mit Strahlquellen höchster Fokussierbarkeit. Dissertation, Universität Stuttgart (2012) [5] Britten, S. W.; Wein, S.; Olowinsky, A. u. Gillner. A.: Laser Impulse Metal Bonding with Temporal Power Modulation. PCIM Europe 2015, 19- 21 Mai 2015 Nürnberg [6] Y. Okamoto, N. Nishi, S. Nakashiba, T. Sakagawa, A. Okada, "Smart laser microwelding of difficult-to-weld materials for electronic industry," Proc. SPIE 9351, Laser-based Micro- and Nanoprocessing IX, 935102 (12 March 2015) [7] Chen, G.: Verhalten verschiedener Werkstoffe beim Schweißen mit dem Nd:YAG – Laser. Dissertation, Technische Hochschule Braunschweig (1984) [8] Moalem, A., Witzendorff, P. von, Stute, U., Overmeyer, L.: Reliable Copper Spot Welding with IR Laser Radiation through Short Prepulsing. Procedia CIRP (2012). [9] Ramsayer, R., Engler, S., Schmitz, G. (eds.): New Approaches for Highly Productive Laser Welding of Copper Materials. 1st International Electric Drives Production Conference (EDPC), Nürenberg, Germany, 28 - 29 Sept. 2011. IEEE, Piscataway, NJ (2011)