Laser Processing of Copper and Aluminium Thin Sheets with Green (532nm) and Infrared (1064nm) Pulsed Laser Beam Sources Paper 1804

Laser Processing of Copper and Aluminium Thin Sheets with Green (532nm) and Infrared (1064nm) Pulsed Laser Beam Sources Paper 1804 Paola De Bono, Ioan...
Author: Mercy Russell
24 downloads 0 Views 639KB Size
Laser Processing of Copper and Aluminium Thin Sheets with Green (532nm) and Infrared (1064nm) Pulsed Laser Beam Sources Paper 1804 Paola De Bono, Ioannis Metsios, Jon Blackburn, Paul Hilton TWI Ltd, Granta Park, Abington, Cambridge, CB21 6AL, UK Abstract

to lead to a sensitive process with low reproducibility. While laser welding of pure Al presents similar issues to those associated to pure Cu, certain Al alloys can also present problems of crack formation, porosity and degradation of mechanical properties in the heat affected zone [2]. Nevertheless, laser welding of Cu and Al is very attractive to industry, due to the potential for high productivity joining, ease of automation, and minimised and localised heat-input.

In this study processing of thin copper C101 (Cu C101) and aluminium 3003 (Al 3003) were investigated, using both infrared (IR), at 1064nm wavelength, and green, at 532nm wavelength, pulsed laser beam sources. The power density (peak power divided by the area of the minimum focused laser beam width) thresholds for keyhole initiation, when processing with the 1064nm laser source, were identified as 59kW/mm2 and 21kW/mm2 for 1mm thickness Cu C101 and Al 3003 coupons respectively. Melt depths up to 400μm were achieved in Cu C101 by tailoring the pulse energy. Melt depths up to 650µm were achieved in the case of Al 3003, with the power density of the laser pulse having a significant effect on the resultant melt depth. Conduction limited melting and welding was carried out in the case of the green wavelength, due to the limited output power of the 532nm pulsed laser sources available. The increased absorptivity of Cu C101 at 532nm introduced improved stability in the process, compared to the results achieved with the 1064nm wavelength laser beam. However, the same behaviour was not observed in the case of Al 3003.

Chen [3] studied the interaction of a continuous wave (CW) IR laser source with pure Cu prepared by different methods, including as-received, sandblasted, painted black and with a nano-composite material applied to the surface. Other studies include the use of oscillating IR continuous wave (CW) laser beams [4] and of green wavelength laser sources [5]. An improvement of process stability was shown in both approaches. Laser power modulation, when using a CW laser source at 1μm wavelength, has also been reported as a successful approach to control process stability and reduce weld imperfections when laser welding pure Cu. Heider et al [6], Herrmann [7] and Schmitt et al [8] observed that a sinusoidal power modulation led to a strong reduction of melt ejections and also to an increase in melt depth.

Introduction Cu and Al exhibit high electrical and thermal conductivity, therefore, they are amongst the most important materials for conducting electric power and for building up electrical systems. Several market opportunities for joining these materials have been identified across a wide range of industry sectors, including automotive and renewable energy. Unfortunately, the properties that make these materials outstanding conductors are also the main reason why high purity Cu and Al alloys are so difficult to laser weld.

A superposition of 515 and 1030nm wavelengths showed an improvement in reproducibility and greater melt depths for welding of Cu [8]. The study observed that weld imperfections significantly decreased by combining a modulated 1030nm wavelength laser beam and a CW 515nm wavelength laser beam. Schmidt et al [8] investigated laser-beam welding of 5000 series Al-Al joints, using 1mm thickness sheets for electrical connections in battery modules. The resulting joints almost reached the conductivity of the base material. A 3kW modulated fibre laser was used and an electrical resistance of approximately 46μΩ was achieved in the case of a joining area of 40mm2.

The high reflectivity of Cu at room temperature (approximately 97%) in the IR, has always been a prevailing barrier in implementing laser welding of copper [1]. The low absorptivity causes high sensitivity to variation in surface conditions, such as roughness and oxidation. In addition, absorptivity rises as a function of temperature. These characteristics tend

Similarly, pulsed laser approaches showed an improvement of the process stability on Cu when using a 532nm wavelength laser source [9]. Other studies investigated real-time adaptive feedback control [10, 11] to improve process stability. Specifically, Piqué et al [11] developed a control algorithm based on several parameters that were monitored simultaneously during 1

welding of Cu. A number of sensors were needed to control the variation of reflectivity and thermal conductivity during the process. However, further optimisation was suggested with respect to the accuracy in the measurement of the temperature profile (which was often affected by the gap between the parts to join), in order to obtain the full performance of this technique.



A frequency doubled Miyachi ML 8150 Nd:YAG laser source, operating at 532nm wavelength. The laser beam had a maximum peak power of 1.5kW, maximum pulse width of 5ms, maximum pulse energy of 4J, an average power of 5W, and a maximum repetition rate was 30Hz. The optical system produced a nominal calculated beam width of 200μm.

The work presented in this paper focused on laser processing of Cu C101 and Al 3003 foils and sheets (thickness range from 0.1 to 1mm) with pulsed laser sources of 1064 and 532nm wavelengths. Both these two metals dominate conductive element manufacturing. In the trials at 1064nm wavelength, the power density thresholds for molten pool and keyhole initiation were investigated on both 1mm thickness Cu C101 and Al 3003 sheets. Tailored pulse energies, with the peak power of the initial sectors sufficient to overcome the threshold for keyhole initiation, were investigated to establish their influence on melt depth and melt width. In the trials at 532nm wavelength, conduction limited processing was carried out on both Cu C101 and Al 3003 sheets and foils (the thickness range was from 0.1 to 1mm). The effect on melt depth was investigated by performing bead-on-plate (BOP) single and overlapping spot melts and welds. For Cu C101 both single sheets (0.5mm thickness) and up to five overlap-clamped together 0.1mm thickness foils (of equivalent total thickness of 0.5mm) were processed to investigate thermal mass effects.



A Miyachi ML 2050 laser source, operating at 1064nm wavelength. The laser beam had a maximum pulse energy of 70J, maximum peak power of 5kW, maximum pulse width of 100ms and maximum repetition rate of 30Hz. The optical system produced a nominal calculated beam width of 200μm.

For all trials, the samples were clamped in a standard welding fixture. The fixture was mounted on an XY stage, which provided motion of the workpiece relative to a stationary laser beam. Argon gas, of 99.98% purity, was used as the shielding gas during all the experiments. The gas was supplied at a flow rate of 10ℓ/min through a Cu pipe of 10mm internal diameter. During all experimental trials, the laser beam was inclined at 5º from the normal to the sample surface, to minimise back reflections. Table 1 gives a general overview of the thickness of the Cu C101 and Al 3003 materials processed with the laser sources used in this study. Table 1 General overview of thickness of Cu C101 and Al 3003 processed with the equipment used.

Experimental Approach Materials and materials preparation

Material Welding trials were conducted on Al 3003 (98.4wt% Al content) and Cu C101 (99.95wt% Cu) foils (0.1mm in thickness) and thin sheets (0.5 and 1.0mm in thickness). Coupons measuring 10x30mm were used. The specimens were cleaned with acetone immediately prior to being processed in order to remove any grease or debris.

Cu C101 1mm thickness Al 3003 1mm thickness Cu C101 0.1 and 0.5mm thickness Al 3003 1mm thickness

Equipment and Experimental procedures The following laser systems were used during the experiments: 

A LASAG FL 542 laser source, operating at 1064nm wavelength. The laser source had a peak power and maximum pulse energy of 15kW and 70J respectively. The optical system produced a nominal calculated beam width of 500µm. The average power was limited to 400W.

LASAG FL 542 Keyhole initiation trials Keyhole initiation trials

Miyachi ML 8150

Miyachi ML 2050

-

-

-

-

-

Conduction limited trials

-

-

Conduction limited trials

Weld Quality Assessment The face and root of all BOP single and overlapping spot melts and lap welds were inspected visually by a 10x loupe and, where necessary, by higher 2

magnification optical microscopy, to determine the presence of imperfections/defects, such as spatter and surface breaking porosity. Selected melts and welds were cross-sectioned to assess the weld profile and also to establish the threshold conditions for melting and keyhole initiation. Cross-sections, longitudinal to the welding direction, were also taken for selected welds.

series of experiments, producing single and overlapping spot melts and welds were carried out to establish the reproducibility of results. During experiments, the pulse energy was kept constant at approximately 4J as this was the maximum amount of energy that could be associated with a pulse. Peak power was varied from 1 to 1.5kW (corresponding pulse width varied from 4 to 2.8ms respectively) in order to identify whether a transition from conduction limited to keyhole processing could be observed. Due to the limited repetition rate, processing speed was limited to 25mm/min.

Scope of Work The work reported here focused on understanding the effect of pulsed laser beam processing, with industrially available sources, on the welding process stability of thin Cu C101 and Al 3003 sheets and foils. All experiments were performed with the beam width positioned at the top-surface of the workpiece.

Welding of Al 3003 with a 532nm wavelength laser beam and direct comparison with 1064nm results Experiments with the Miyachi ML 8150 laser source were performed on 1mm thickness Al 3003 sheets. Conduction limited single spot melts were carried out to establish the reproducibility of results at 532nm wavelength. Similarly to the experiments on Cu C101 at 532nm wavelength, the pulse energy was kept constant at approximately 4J, peak power was varied from 1 to 1.5kW (corresponding pulse duration varied from 4 to 2.8ms respectively) in order to identify whether a transition from conduction limited to keyhole welding could be observed. The processing speed was kept constant at 25 mm/min.

Welding of Cu C101 with a 1064nm wavelength laser beam Experiments with the LASAG FL 542 laser source were performed on 1mm thickness Cu C101 sheets in two distinct phases: 1. Initially, BOP spot melts were produced in order to determine the power density thresholds for melting and keyhole initiation (only single sector square pulses were used). 2. Based on the results achieved on the power density threshold experiments above, tailored pulse energies were applied to determine their effect on melt depth and melt width. Specifically pulses with a leading high peak power sector were applied to meet the melting threshold, while different trailing sectors were tested in order to determine their effect on the resulting depth and width of the BOP spot melts.

To compare the welding behaviour of Al 3003 at 532 and 1064nm, additional melting trials were performed using the ML2050 Nd:YAG laser source (1064nm wavelength), with the same minimum focused laser beam width as the frequency doubled Miyachi ML 8150 laser source (532nm wavelength). Argon was used as shielding gas in all experiments.

Experimental Results and Discussion

Welding of Al 3003 with a 1064nm wavelength laser beam

Welding of Cu C101 with a 1064nm wavelength laser beam

Experiments with the LASAG FL 542 laser source were performed on 1mm thickness Al 3003 sheets. The trials had a nominally identical scope to those performed with the LASAG FL 542 on Cu C101 – ie determining the power density thresholds for melting and keyhole initiation, and then identifying the influence of tailored pulse energies on the resulting melt depth and width.

Figure 1 shows the transition from melting to keyhole initiation in 1mm thickness Cu C101. For welds performed in argon shielding gas, it was observed that a power density threshold of 30kW/mm2 (500μm minimum focused laser beam width at the top-surface of the workpiece) and a pulse energy of 40J, was sufficient to start melting Cu C101 (Figure 1b). Keyhole initiation in Cu C101 was observed at a power density threshold of 59kW/mm2 (500μm minimum focused laser beam width at the top-surface of the workpiece) and pulse energy of 10.5J (Figure 1c).

Welding of Cu C101 with a 532nm wavelength laser beam Experiments with the Miyachi ML 8150 laser source were performed on Cu C101 thin sheets and foils (the thickness range investigated was from 0.1 to 1mm). A 3

a) Figure 2 The effect of changing the power density on melt depth when irradiating 1mm thickness Cu C101. Following the determination of the power density thresholds required to for melting and keyhole initiation, single spot melts were produced using an initial pulse sector of either 0.5ms or 1ms pulse duration and sufficient power density to initiate a keyhole. The trailing sector of the pulse was varied in order to determine its effect on the resultant spot melt.

b)

In all the trials performed, the melt width and depth appeared to be governed by the interaction of the initial sector of the pulse with the material. Specifically, the melt width (measured at the melt face) did not change as a result of varying the width of the trailing sector and corresponding increase in pulse energy. Similarly, the melt depth varied randomly without showing an ascending or descending pattern. The variations in melt depth and melt width are likely related to surface reflectance of the material during the interaction with the initial sector of the pulse. The results achieved for three different initial sector conditions are summarised in Tables 2, 3 and 4.

c) Figure 1 Transition from melting (from a to b) to keyhole initiation (from b to c) in Cu C101 single spot melts.

Table 2 Single spot melt width and depth values achieved with an initial sector of 1ms at 12.5kW and tail at 20% of the initiator’s peak power.

Figure 2 shows the effect of changing the power density on melt depth when irradiating 1mm thickness Cu C101 and demonstrates the onset of the keyhole initiation at power density of approximately 59kW/mm². At the onset of keyhole initiation, it can be seen that the rate at which melt depth increases as a function of the power density changes significantly.

Sample No 20a 20b 20c

4

Energy (J) 18.8 25.05 37.5

Pulse tail width (ms) 1.5 4 9

Melt width (μm) 727 814 775

Melt depth (μm) 396 413 341

Table 3 Single spot melt width and depth values achieved with an initial sector of 0.5ms at 15kW and tail at 20% of the initiator’s peak power.

Sample No 21d 21b 21c

Energy (J) 14.27 21.2 37.1

Pulse tail width (ms) 1.5 4 9

Melt width (μm) 710 633 626

Figure 4 depicts the effect of changing the power density on melt depth when irradiating Al 3003 with a 5J pulse energy. The curve demonstrates the onset of the keyhole initiation at power densities of approximately 21kW/mm². At the onset of keyhole initiation, it can be seen that the rate at which melt depth increases as a function of the power density changes significantly. The melt depth increase would normally be expected to continue with the same rate for bulk material. However, in this case, it is thought a lack of effective heat dissipation eventually dominated and energy deposited to the material achieved ablation and limited further conduction and growth. This effect is more evident in thin sheet material as heat sinking is further restricted in comparison to thicker workpieces. A line has been drawn on the graph to demonstrate the anticipated behaviour if this work was performed in material thicknesses above 1mm, or if ablation effects were suppressed.

Melt depth (μm) 196 190 105

Table 4 Single spot melt width and depth values achieved with an initial sector of 1ms at 12.5kW and tail at 50% of the initiator’s peak power.

Sample No 22a 22b 22c

Energy (J) 24.2 39.8 42

Pulse tail width (ms) 1.5 4 9

Melt width (μm) 699 735 750

Melt depth (μm) 310 241 353

Welding of Al 3003 with a 1064nm wavelength laser beam In the top micrographs of Figure 3 it is shown that only surface effects were achieved on the 1mm thickness Al 3003 workpiece at a power density of 6kW/mm2 and a pulse energy of 5J. The middle micrographs of Figure 3 show that the material melting threshold was surpassed when reaching a power density of approximately 8kW/mm2 at a pulse energy of 5J. With a further power density increase to 21kW/mm2 and a pulse energy of 5J, a keyhole was initiated in the workpiece.

Figure 4 The effect of changing the power density on melt depth when processing Al 3003 with a laser pulse of 5J pulse energy. Following the determination of power density thresholds for melting and keyhole initiation, single spot melts were produced using an initial pulse sector of 1ms and sufficient power density to initiate a keyhole. The trailing sector of the pulse was varied in order to determine its effect on the resultant spot weld. Similarly to the results in the Cu C101 material, in all performed trials, the melt width and depth appeared to be governed by the interaction of the initial sector of the pulse, at specific peak power, with the material. Specifically, the diameters of the melt bead (measured at the melt face) did not change as a result of the varying width of the trailing sector and corresponding increase in pulse energy. The melt depth varied randomly without showing an ascending or descending pattern, again in accordance with the results observed on Cu C101. Specifically this behaviour was observed in each of the peak power conditions of the initial sector, which are summarised in Tables 5, 6 and 7.

Figure 3 Transition from surface effects to molten pool to keyhole initiation in Al 3003.

5

Table 5 Single spot melt width and melt depth values achieved with an initial sector of 1ms at 2.5kW and tail at 20% of the initiator’s peak power.

Sample No 15c 15b 15a 16a 16b 16c

Energy (J) 4.08 5.5 8.7 14.3 19.7 25.1

Pulse tail width (ms) 1.5 4 9 19 29 39

Melt width (μm) 500±25 500±25 500±25 500±20 490±10 490±10

parameters used for the results detailed in Figure 5 were a pulse width of 2.8ms, a pulse energy of 4J and a peak power of 1.5kW.

Melt depth (μm) 117 103 122 127 117 121

Figure 5 Image of single spot melts on 0.5mm thickness Cu C101; produced with a pulse width of 2.8ms, a pulse energy of 4.2J, and a peak power: 1.5kW.

Table 6 Single spot melt width and melt depth values achieved with an initial sector of 1ms at 5kW and tail at 20% of the initiator’s peak power.

Sample No 17a 17b 17c 18a 18b 18c

Energy (J) 7.48 10.3 15.5 26 37 47.8

Pulse tail width (ms) 1.5 4 9 19 29 39

Melt width (μm) 1055 927 1054 1108 948 1020

Bead-on-plate spot melts, using single pulses of variable duration, on 0.5mm thickness Cu C101 sheets were performed in order to observe how the melt depth varied as function of the applied power density. The Cu C101 sheets were sectioned and polished to determine the melt depth. The results of plotting melt depth as a function of the power density in the spot, for fixed pulse energy, are presented in Figure 6 and also show the rate at which the melt depth increases as function of the power density is gradual. This indicates that the process was characterised by a conduction limited melting behaviour. The pulse energy was kept constant at 4J for all trials. A plateau of melt depth values was observed from a power density of approximately 4x106W/cm2 to the maximum power density of approximately 4.7x106W/cm2, which was achievable with this laser system. The plateau showed that higher power density values than 4.7x106W/cm2 were needed to observe a transition from conduction limited behaviour to keyhole initiation.

Melt depth (μm) 511 458 473 492 405 468

Table 7 Single spot melt width and melt depth values achieved with an initial sector of 1ms at 5kW and tail at 50% of the initiator’s peak power.

Sample No 19a 19b 19c

Energy (J) 9.75 16 28.5

Pulse tail width (ms) 1.5 4 9

Melt width (μm) 1069 1058 1068

Melt depth (μm) 657 663 605

The small variations in melt depth and melt width observed are likely related to surface reflectance of the material during the interaction with the initial sector of the pulse. Melt depths up to 650µm were achieved in the case of Al 3003, with the initial peak power of the laser pulse having a significant effect on the resultant melt depth. Figure 6 Melt depth versus power density for BOP spot melts performed on 0.5mm thickness Cu C101 sheets. All trials performed at a constant pulse energy of 4J.

Welding of Cu C101 with a 532nm wavelength laser beam When using the 532nm wavelength, when compared to the 1064nm wavelength, improved reproducibility can be seen. Figure 5 shows single spot melts performed on 0.5mm thickness Cu C101 sheets. The laser

It was observed that melt depth varied from a minimum of 11µm (power density 3.2x106 W/cm2) to a 6

maximum of 52µm (power density 4.7x106 W/cm2). The maximum melt depth was found at the maximum performance of the laser, in terms of peak power and pulse energy. It was observed that at least 1kW peak power was needed to observe the start of melting on the surface of the workpiece.

Welding of Al 3003 with a 532nm wavelength laser beam and direct comparison with 1064nm results All the BOP single spot melts produced on Al 3003 sheet showed good reproducibility, at least from a visual observation of the surface of the melts. Figure 8 shows an example of a series of single spot melts performed on 1mm thickness Al 3003 sheet using a pulse duration of 3.1ms, a pulse energy of 4J and a peak power of 1.3kW.

Similar results, in terms of melt depth and reproducibility, were observed in the case of BOP overlapping spot melts, on 0.5mm thickness Cu C101 thin sheets. A weld melt depth of 61μm was measured in this case. When the maximum achievable peak power (using the same laser system) was applied on five 0.1mm thickness Cu C101 foils, overlap-clamped together to achieve an equivalent thickness of 0.5mm, it was observed that a weld penetration depth of the order of 250μm was achieved, fully penetrating two foils and partially penetrating into the third. This value of weld penetration depth is much greater than the 52µm melt depth achieved in the case of using a 0.5mm Cu C101 sheet, and the same laser processing conditions. This might be related to the higher heat sinking capability present in the bulk material, while the presence of air/discontinuities at the interface of the thin foils clamped together, might contribute to better control of heat dispersion during the welding process. This again highlights the importance of thermal mass and clamping in this type of experiment. Figure 7 shows a macro-section of the BOP single spot melt on 0.5mm Cu C101 sheet a) and a lap weld achieved between the three foils b).

Figure 8 Image of single spot melts on 1mm thickness Al 3003, showing reproducibility of the process. The variation of melt depth as a function of the power density was observed by changing pulse duration from 2.8ms to 4ms and peak power from 1.5 to 1.0kW (for a fixed pulse energy of 4J). The results of analysing melt depth as a function of power density in the spot (Figure 9) showed that melt depth varied from a minimum of 115μm (power density 3.2x106 W/cm2) to a maximum of 162μm (power density 4.8x106 W/cm2).

b) a) Figure 7 a) Single spot BOP melt on 0.5mm thickness Cu C101 sheet and b) single spot lap weld of three Cu C101 foils of 0.1mm thickness, produced with identical parameters using a 532nm wavelength laser beam source.

Figure 9 Melt depth versus power density for BOP spot melts performed on 1mm thickness Al 3003 sheets. All trials performed at a constant pulse energy of 4J.

All results achieved at 532nm wavelength showed a reproducible performance with good weld quality. The use of a frequency doubled Nd:YAG laser enabled some of the difficulties of Cu laser welding to be overcome. The absorptivity of Cu at 532nm and at room temperature is approximately 30%, which is ten times higher than at 1064nm at the same temperature (approximately 3%).

To compare the welding behaviour of Al 3003 at 532nm and 1064nm, additional melting trials were performed using the ML2050 Nd:YAG laser source (1064nm wavelength), with the same minimum focused laser beam width as the frequency doubled Miyachi ML 8150 laser source (532nm wavelength). BOP single spot melts were performed on Al 3003 1mm thickness sheets. Parameters varied during experiments were peak power (from 1 to 1.5kW) and pulse duration (from 2.8ms to 4ms), while pulse 7



energy was kept constant at 4J. In all cases, results of cross sections showed that melt depth was less when using the 1064nm beam, compared with the 532nm wavelength, under the laser processing conditions. Figure 10 presents melt depth plotted as function of power density, where a progressive decrease of melt depth is observed when decreasing the power density, while keeping pulse energy fixed to 4J.

Melt depths up to ~650µm were achieved, and this was found to be dependent upon the initial peak power of the laser pulse.

Experiments carried out on Cu C101 at 532nm wavelength:  The use of 532nm light gives enhanced reproducibility of spot and overlapping spot welds on thin Cu C101 sheets, when compared to results using pulsed 1064nm light.  Up to three Cu C101 foils, 0.1mm thickness, were successfully spot lap joined achieving a maximum melt depth of 250μm.  The use of 532nm light on Al 3003 did not show major differences in terms of reproducibility and stability of the process when comparing results achieved using 1064nm light. The main differences when using the two wavelengths, under the same processing conditions, were in terms of melt depth.

The absorptivity of Al 3003 at 1064nm is approximately 4% less than that at 532nm. Crosssections of BOP single spot melts on 1mm thickness Al 3003 plates, using both wavelengths, showed that there are no major differences, in terms of reproducibility and stability, of the process. In terms of penetration, the green laser outperformed the 1064nm laser and the better performance was stronger at lower applied power density. A green wavelength could therefore be considered for the energy effective welding of thin Al 3003, however, the higher cost of the green laser, would have to be taken into account in any industrial application.

Acknowledgements The research leading to these results has received funding from:  The European Community's Seventh Framework Programme (FP7/2007-2014) under grant agreement n° 260153.  The Industrial Members of TWI, as part of the Core Research Programme. References [1] Steen W M, 2003: ‘Laser Materials Processing’, Published by Springer-Verlag London Ltd. ISBN 185233-698-6.

Figure 10 Penetration versus power density of green (blue line) and IR-laser (red line) beams on Al 3003, 1mm thickness. All trials performed at a constant pulse energy of 4J.

[2] Thorstensen B, 1989: Laser welding of aluminium, The industrial laser annual handbook, Edited by Belforte David and Levitt Morris, PennWell Publishing, Tulsa, Oklahoma.

Conclusions Work using the three laser sources outlined in Table 1 has allowed the following conclusions to be drawn within the scope of work performed:

[3] Chen H C, Bi G, Nai S M L, Wei J, 2012: Influence of Surface Condition in Fiber Laser Welding of Pure Copper; 31st International Congress on Applications of Lasers and Electro-Optics, September.

Experiments carried out on Cu C101 at 1064nm wavelength:  Power densities of 30 and 59kW/mm2 were required to achieve melting and keyhole initiation in 1mm thickness Cu C101 coupons respectively.

[4] Olowinsky A, A Boglea, J Gedicke, 2008: Innovative Laser Welding Processes: TWIST and SHADOW in Microtechnology LTJ, May, No. 3 Micro Material Processing.

Experiments carried out on Al 3003 at 1064nm wavelength:  Power densities of 8 and 21kW/mm2 were required to achieve melting and keyhole initiation in 1mm thickness Al 3003 coupons respectively.

[5] Ramsayer R M, Engler S, Schmitz G, Bosch GmbH R, 2011: New approaches for highly productive laser welding of copper materials, Electric Drives Production Conference (EDPC), [1-4577-1371-3], pp69-73. 8

[6] Heider A, Stritt P, Hess A, Weber R, Graf T, 2011: Process Stabilization at welding Copper by Laser Power Modulation Physics Procedia 12, 81–87. [7] Herrmann D, Progress in laser welding of copper, AKL 2012 International Laser Technology Congress, Fraunhofer ILT, Aachen, 9-11 May. [8] Schmitt F, Mehlmann B, Olowinsky A, Roesner A, Laser micro-joining – Processes and applications in research and development, AKL 2012 International Laser Technology Congress, 9-11 May. [9] Shannon G and Severloh P, 2009: Laser Microwelding of copper, Industrial Laser Solutions for Manufacturing, January. [10] Amorosi S, Sidler Th, Salathé R P, Schwob H P and Hertzberg J, 2004: ‘Laser microspot welding of copper’, Journal of Laser applications, 16, 134. [11] Piqué A, Sugioka K, Herman P R, Fieret J, Bachmann F G, Dubowski J J, Hoving W, Washio K, Geohegan D B, Träger F, Murakami K, 2003: Process spread reduction of laser micro-spot welding of thin copper parts using real time control, Proceeding of SPIE, Vol. 4977, pp493-507. Meet the Authors Ms Paola De Bono is a Senior project Leader at TWI Ltd. She has a Master Degree in Chemical Engineering and over 5 years of specific experience in laser processing for micro-scale applications. In October 2010 she started an Engineering Doctorate in Birmingham University to investigate the interaction laser-matter for industrial applications, specifically focusing on laser welding of high reflective metals. Dr Ioannis Metsios is a Senior project leader at TWI Ltd. He specialises in laser surface and microprocesses. Ioannis has 10 years experience in developing laser manufacturing processes and photonics engineering. Dr Jon Blackburn is Manager of the Laser and Sheet Processes Section at TWI Ltd, covering TWI’s laser cutting, welding and surfacing activities. Dr Paul Hilton is the Technology Fellow for Laser Materials Processing at TWI Ltd and has over twenty years of laser processing experience. He has previously been conference chair for LMP at ICALEO and is the current Chairman of the governing body of ELI, the European Laser Institute.

9

Suggest Documents