Latest progress in performance and understanding of laser welding

Available online at www.sciencedirect.com Physics Procedia 39 (2012 ) 8 16 LANE 2012 Latest progress in performance and understanding of laser welding Seiji Katayama, Yousuke Kawahito, Masami Mizutani JWRI - Joining and Welding Research Institut - Osaka University, 11-1 Mihogaoka, Ibaraki, Osaka 567-0047, Japan - Keynote Paper - Abstract This paper describes a variety of fundamental research results of laser welding which the authors have recently performed. The behavior and characteristics of a laser-induced plume were elucidated. Especially, in remote welding with a fiber laser, the effect of a tall plume leading to shallow weld was interpreted by considering the interaction of an incident laser beam against the zone of a low refractive index from the Mickelson interferometer results. The laser absorption in the plate was higher in the case of a smaller focused beam of fiber laser, lower welding speed and higher power, and the reason was interpreted by considering the size and location of a keyhole inlet and a beam spot. High power tandem laser beams could produce deep penetration, and laser welding in vacuum was developed for production of deeply penetrated welds. Laser direct joining was also developed for joining of metal to plastic or CFRP. 2012 2011 Published by by Elsevier B.V. Ltd. Selection and/or review under responsibility of of Bayerisches Laserzentrum GmbH Open access under CC BY-NC-ND license. Keywords: fiber laser; disk laser; welding phenomena; deep penetration ; metal-plastic joining 1. Introduction Laser welding has received great attention as promising joining technology with high quality, high precision, high performance, high speed, good flexibility and low deformation or distortion, in addition to the recognition of easy and wide applications due to congeniality with a robot, reduced man-power, full automation, systematization, production lines, etc. Thus, applications of laser welding are increasing. * Corresponding author. Tel.: +81-6-6879-8662. E-mail address: katayama@jwri.osaka-u.ac.jp. 1875-3892 2012 Published by Elsevier B.V. Selection and/or review under responsibility of Bayerisches Laserzentrum GmbH Open access under CC BY-NC-ND license. doi:10.1016/j.phpro.2012.10.008

Seiji Katayama et al. / Physics Procedia 39 ( 2012 ) 8 16 9 Therefore, to understand the mechanism of weld penetration and the phenomena in laser welding, a variety of researches have been performed concerning laser-induced plume behavior, the interaction between a laser beam and its induced plume/plasma, melt flows, keyhole behavior, and bubbles generation in the molten pool leading to the porosity formation in the weld fusion zones. Especially, remote welding phenomenon was elucidated by Mickelson interferometer. Welding in low vacuum for sound deep penetration or joining of metal to plastic or CFRP has been newly developed in terms of effective penetration and lighter-weight applications. 2. Behavior and Characteristics of Laser-Induced Plume, and Its Interaction to Laser Beam Spectroscopic measurement and analysis of a plume were performed during welding of Type 304 stainless steel with a 10 kw fiber laser beam at the ultra-high power density of about 1 MW/mm2 in Ar shielding gas. It was revealed that almost all peaks came from the emission of neutral atoms, and the emission from Ar gas was not detected. The inverse Bremsstrahlung effect seems to be negligibly small in welding with a fiber, disk or YAG laser beam of about 1 m wavelength. According to Saha equation, it is estimated that a plume is in the state of a weakly-ionized plasma (about 6000 K and 2% atomic ions)[1]. It was confirmed that the refraction and attenuation of a probe laser beam existed due to an inclined refraction index profile caused by slanted distribution of vapors density at high temperatures and Rayleigh scattering, respectively. The interaction of such a small-sized plume on a laser beam of about 1.06 m wavelength was also judged to be small. Fig. 1. Schematic arrangement of Michelson interferometer for measuring

10 Seiji Katayama et al. / Physics Procedia 39 ( 2012 ) 8 16 Fiber laser remote welding was performed on 270 MPa steel sheet with and without fan blowing under the conditions of 4 kw and 5 m/min. Without fan-blowing, a full penetration weld was changed into a partial penetration during welding, while it was formed all over the weld length with fan-blowing. To understand the effect of tall plume on the weld penetration, a laser induced plume and refractive index distribution were observed by Mickelson interferometer, as the system is schematically shown in Fig. 1 [2]. The observation results are shown in Fig. 2 [2]. It is considered that a tall plume over the specimen is heated, and thus the distribution of the refractive-index is so different that the incident laser beam is defocused and/or inclined. As a result, the transition in weld penetration depth occurred in air without fan-blowing. It is concluded that the influence of a short plume is small but the effect of a tall plume is great during remote fiber laser welding. Fan OFF Fan ON Fig. 2. Plume behaviour and refraction index distribution during remote laser welding of steel without and with fan-blowing 3. Laser Absorption Laser absorption was measured by calorimetric method. Fig. 3 shows a comparison of fiber and YAG laser absorption. Laser power: 2.5 kw, Welding speed: 1 m/min Fiber laser Spot diameter: 200 m YAG laser Spot diameter: 580 m Defocused distance [mm] Defocused distance [mm] Beam diameter [mm] Absorptivity [%] 3 mm 3 mm Beam diameter [mm] Defocused distance [mm] Fig. 3. Absorption of fiber and YAG laser in Type 304 stainless steel as function of defocused distance

Seiji Katayama et al. / Physics Procedia 39 ( 2012 ) 8 16 11 The absorption of tightly focused fiber laser of about 0.2 mm diameter was higher than that of YAG laser of about 0.6 mm diameter. The laser absorption was higher in the case of higher power and/or lower welding speed. The keyhole behavior during welding was observed. The results are shown in Fig. 4. At low welding speeds, the incident beam was shot into the keyhole, but the molten pool in front of a keyhole became smaller with an increase in the welding speed, and the laser beam was shot on the front wall of the keyhole and partly on a the molten pool and solid plate in front of the keyhole, which results in the decrease in the absorption. Consequently, the reason for the difference in laser absorption is interpreted by considering the relationship between the size and location of a keyhole inlet and beam spot size. Welding speed High speed images 2 m/min 3 m/min 6 m/min 10 m/min 15 m/min 1 mm Schematic illustration Beam ratio to keyhole Absorption Molten pool Keyhole Laser spot 100 % 84 % 100 % 85 % 69 % 66 % 84 % 79 % 74 % 68 % Fig. 4. Fiber laser absorption in Type 304 stainless steel during welding at 10 kw and various welding speeds 4. High Power Laser Welding in Vacuum Laser welding in low vacuum was established by using a new chamber, whose vacuum conditions were achieved by rotary pumps [3]. Two continuous wave (CW) disk lasers (TRUMPF 10003 and 16002 with the wavelength of 1030 nm) were used, as shown in Fig. 5. Feeding fiber: 300 m Peak power: 10 kw Wavelength: 1,030 nm BPP : 12 mm*mrad TRUMPF L A S E R TruDisk 10003 Feeding fiber: 200 m Peak power: 16 kw Wavelength: 1,030 nm BPP : 8 mm*mrad TRUMPF L A S E R TruDisk 16002 Pumping speed: 162 /min Rotary pump Guard glass Disk laser Pumping speed: 500 /min Pressure gauge High speed camera Flowmeter Frame rate: 5,000 f/s Frame rate: 100 f/s N 2 Specimen Laser diode (Illumination) : 980 nm P: 30 W Fig. 5. Schematic experimental setup for laser welding under low vacuum

12 Seiji Katayama et al. / Physics Procedia 39 ( 2012 ) 8 16 The maximum laser powers are 10 and 16 kw, and the beam parameter products (BPP) were 12 and 8 mm*mrad, respectively. The maximum power of combined laser beams was 26 kw. A focusing optic of the focal distance of about 1 m was employed. A spot diameter at the focus position was about 0.4 to 0.5 mm. High power fiber or disk laser bead-on-plate welding was performed on Type 304 stainless steel or A5052 aluminum alloy plate at the powers of 10, 16 and 26 kw at various welding speeds or under various defocused conditions in low vacuum. Examples of cross sections of weld beads in A5052 aluminum alloy under the different vacuum conditions are shown in Fig. 6. Pressure 0.1 kpa 10 kpa 101.3 kpa Bead surface Cross section Fig. 6. Surface appearances and cross sections of laser weld beads in A5052 obtained at 0.3 m/min Examples of Type 304 stainless steels as a function of defocused distance are exhibited in Fig. 7. The penetration depth increased with the decrease in pressure (vacuum) in both aluminum alloy and stainless steel. The increase was remarkable at lower welding speed. However, humping beads were produced near the focal point under lower vacuum. Such a humping bead was suppressed under the proper defocused conditions as shown in Fig. 7 [4].

Seiji Katayama et al. / Physics Procedia 39 ( 2012 ) 8 16 13-40 mm Defocused distance -20 mm 0mm +20 mm Bead surface Cross section (b) Welding speed of 0.3 m/min Fig. 7. Bead surfaces and cross sections of 16 kw disk laser welds made in SUS 304 at various defocused distances and welding speed of 0.3 m/min (a) f d = -40 mm (b) f d = + 2 0 mm Fig. 8. Cross sections of Type 304 wed beadsmade with disk laser at 26 kw, 0.3 m/minand 0.1 kpa, showing formation of sounddeep- penetration weld The penetration depth was further increased under the defocused conditions. The wide bead zone was formed about below the focal point in Type 304 welds. Examples of weld cross sections produced with 26 kw disk lasers (combined by 10 kw and 16 kw beams) at the defocused distances of +20 mm and - 40 mm is shown in Fig. 8 [4]. At +20 mm and 40 mm, ideal and deep weld beads of about 46 mm and 73 mm depth were formed in Type 304 steel, respectively. Sound welds of more than 50 and 70 mm in penetration depth could be produced in Type 304 at the pressure of 0.1 kpa, the speed of 0.3 m/min and the power of 16 kw and 26 kw. Similar welding results of deep penetration were also obtained in A5052 alloy [4]. A swelled part was not present in A5052 alloy although it was located at about under the focal point in Type 304 steel. According to welding phenomena, weak plume or evaporation, a stable molten pool on the surface and reduced spattering were observed although the keyhole in the molten pool was actively moved. In low vacuum with a focusing optics of long focal length, deep penetration welds could be produced under the conditions of high powers, low welding speeds and (minus) defocused distance. 5. Laser Direct Joining between Metal and Plastic or CFRP Recently, the use of light plastics (resin) has been increasing. The joining of metals and plastics is important in the fields of manufacturing. Therefore, we have developed a new Laser-Assisted Metal and Plastic (LAMP) direct joining process with a YAG, diode, disk or fiber laser, which has many advantages over the conventional bonding methods using adhesives or mechanical fastening with bolts and nuts. The specimen after the tensile shear test of LAMP joint of Type 304 and polyethylene terephthalate (PET) exhibited a long elongation of the base PET plastic, as shown in Fig. 9 [5,6]. TEM observation was also performed. The photo is shown in Fig. 10.

14 Seiji Katayama et al. / Physics Procedia 39 ( 2012 ) 8 16 Plastic Cro ss s e ct i o n We ld fus io n z one Type 304 Plastic :PE T 104 000 10 4 3 m m Bubbles (a) (a) C r 2 O 3 Beam // [ 010] Ty T y pe 304 PET 002 11 1 11 1 00 0 11 1 (a) LA M P jo i n t b etween TYPE304 stainl ess s t eel a n d PET 5 nm TEM photo (b) 11 1 002 ( b ) T YPE304 B ea m // [ 110] Elect ro n dif f racti o n p a t te rn s (b) L A MP j o i n t af ter ten si l e share test Fig. 8. Tensile shear test specimens of Type 304 and PET joined with laser (a) before and (b) after test, showing elongation of PET base plastic Fig. 9. TEM photo near joint interface of Type 304 steel and PET plastic, and electron diffraction patterns obtained from oxide film and metal It suggests that the joint was produced by bonding melted plastic onto the oxide film covering all over the metal. The LAMP joining could produce strong joints between almost all metals and some plastics such as PET, polyamide (PA), and polycarbonate (PC) [5,6]. The joint specimens of Type 304 stainless steel plates and PA or PET plastic sheets of 2 mm thickness and 30 mm width possessed tensile shear loads of more than 3000 N, where the fracture occurred in the base plastic sheets. The LAMP joining was characterized by the formation of small bubbles in the plastic on the metal plate surface. SEM and TEM observation of the bonding parts, XPS analyses of the bonding interface and the special bubble-inside-gas analyzer suggested that such high strengths were attributed to three joining mechanisms of mechanical, physical and chemical bonding, as indicated in Fig. 11 [5,6]. Dissimilar lap joints between Type 304 stainless steel plate, Zn-coated steel sheet or A5052 aluminum alloy sheet and CFRP sheet could be produced by irradiating a disk laser on Type 304 steel plate [7]. An example is shown in Fig. 12. The mechanical properties of the joints were evaluated by the tensile shear test. The tensile strength or loads of Type 304-CFRP joints of 3 mm thickness and 20 mm width were high. Fracture occurred in the CFRP base plastic to connect some bubbles. Some joints were very strong depending upon the joining areas according to the conditions. Especially, the load or strength of the joint reached 4770 N tensile shear load [8].

Seiji Katayama et al. / Physics Procedia 39 ( 2012 ) 8 16 15 Laser beam Plastic Metal Heat transfer High vapor pressure due to rapid expansion Plastic Heat transfer Molten area Flow Metal Bubble Laser heated zone Bubble Oxide film (Higher magnification) Laser beam Fig. 10. Mechanisms of laser direct joining between metal and plastic, showing absorption of laser in metal, melting of plastic due to absorbed heat, and formation and expansion of bubbles inducing plastic flows onto oxide covering metal. The joining is attributed to anchor effect, Van der Waals force and chemical bonding between melted plastic and oxide film 16 kw Disk laser, fd = +20mm, spot diameter = 0.3 mm P = 2 kw, v = 5 m/s Appearances Crosssection Magnified photo Fig. 11. Tensile shear test results of specimens made between Type 304 and CFRP sheets (left), and example of specimen and cross section and magnified photo near joint interface showing bubbles in CFRP 6. Conclusions Laser welding phenomena are satisfactorily understood. The behavior and characteristics of a laser induced plume were elucidated. In remote welding with a fiber laser, the effect of a tall plume leading to shallow weld was interpreted by considering the interaction of an incident laser beam against the zone of a low refractive index from the Michelson interferometer results. The laser absorption in the plate was higher in the case of a smaller focused beam of fiber laser, lower welding speed and higher power, and the reason for high or low absorption was interpreted by considering the size and location of a keyhole

16 Seiji Katayama et al. / Physics Procedia 39 ( 2012 ) 8 16 inlet and a beam spot. High power tandem laser beams could produce deep penetration, and laser welding in vacuum was developed for production of deeply penetrated welds in stainless steel and aluminum alloy. Laser direct joining between metal and plastic or CFRP sheet was also developed. Strong joints were produced easily by irradiating a laser beam on the metal plate. Acknowledgements We would like to acknowledge Mr. Naoki Matsumoto, Mr. Shimpei Oiwa, Mr. Youhei Abe, Kwangwoon Jung, and so on of former Graduate Students of Osaka University. References [1] Kawahito Y., Kinoshita K., Matsumoto N., Mizutani M. and Katayama S., Q. J. Jpn. Weld. Soc., 2007 25-3 461-467. (in Japanese). [2] Katayama S., Oiwa S., Ishida H., Ozawa N., Mizutani M. and Kawahito Y., 61st Annual Assembly of Int. Inst. Welding (IIW), Graz, IIW Doc. IV-962-08. [3] Abe Y., Mizutani M., Kawahito Y., and Katayama S., Proc. of ICALEO 2010, LIA, 2010, 103 648-653. [4] Katayama S., Abe Y., Mizutani M., and Kawahito Y., Physics Procedia, 2011, 12-Part 1, 75-80. [5] Katayama S. and Kawahito Y., Scripta Materialia, 2008, 59-12 1247-1250. [6] Katayama, S., Kawahito Y., Niwa Y. and Kubota S., Proc. of LANE 2007, Germany, 2007, 41-51. [7] Niwa Y., Kawahito Y., Kubota S. and Katayama S., Proc. ICALEO 2008, LIA, CA, USA, 2008, 101 311-317. [8] Jung K.-W., Kawahito Y. and Katayama S., Science and Technology of Welding and Joining, 2011, 16-8 676-680.