L

Ma

b

a

ARRAA

KLWADSNA

1

rwnntantt

ntteC

ss

w

h0

Journal of Materials Processing Technology 216 (2015) 381–384

Contents lists available at ScienceDirect

Journal of Materials Processing Technology

jo ur nal ho me page: www.elsev ier .com/ locate / jmatprotec

aser welding of duplex stainless steel with nitrogen as shielding gas

. Keskitaloa,∗, K. Mäntyjärvia, J. Sundqvistb, J. Powellb, A.F.H. Kaplanb

University of Oulu, Oulu Southern Institute, Pajatie 5, FI-85500 Nivala, FinlandLuleå University of Technology, Department of Engineering Sciences and Mathematics, SE-97 451 Luleå, Sweden

r t i c l e i n f o

rticle history:eceived 13 May 2014eceived in revised form 2 October 2014ccepted 6 October 2014vailable online 16 October 2014

eywords:

a b s t r a c t

Nitrogen loss from laser welding melts pools and can have a deleterious effect on weld toughnessfor duplex stainless steels. This effect can be alleviated by using nitrogen as the shielding gas duringlaser welding. The use of nitrogen results in increased austenite levels in the weld metal and improvedtoughness levels.

© 2014 Elsevier B.V. All rights reserved.

aser weldingeld toughness

ustenite contentuplex stainless steelhielding gasitrogen

rgon

. Introduction

Duplex stainless steels are a combination of austenite and fer-ite, and some grades employ nitrogen as an alloying ingredienthich promotes austenite formation in welds and improves tough-ess. In standard laser welding, using argon as the shield gas,itrogen can be lost from the weld pool and the austenite con-ent of the weld will be reduced. Kyröläinen and Lukkari (1999)nd the Outokumpu welding handbook (Outokumpu, 2010) haveoted that a low austenite content can lead to nitride precipita-ion, which has a negative effect on weld corrosion properties andoughness.

It has been noticed by Keskitalo and Mäntyjärvi (2013) thatitrogen shielded austenitic laser welds have a higher hardnesshan argon shielded welds. Westin and Serrander (2012) have notedhat using nitrogen as the backing gas had a measurable positiveffect on the weld metal austenite formation when welding withO2 lasers.

This paper investigates the idea that if nitrogen is used as thehield gas some of it would dissolve into the weld pool, compen-

ating for the amount lost by evaporation.

Kyröläinen et al. recommend the WRC-92 diagram for duplexeld material. The presence of nitrogen promotes the formation

∗ Corresponding author. Tel.: +358 407750337.E-mail address: markku.keskitalo@oulu.fi (M. Keskitalo).

ttp://dx.doi.org/10.1016/j.jmatprotec.2014.10.004924-0136/© 2014 Elsevier B.V. All rights reserved.

of austenite in duplex stainless steels according to the followingequations (Kyröläinen and Lukkari, 1999):

Ni-equivalent (Ni eq) = %Ni + (%C × 35) + (%N × 20)

+ (%Cu × 0.25) (1)

Cr-equivalent (Cr eq) = %Cr + %Mo + (%Nb × 0.7) (2)

According to Sakai et al. (1989) the austenite content of the steelshould be higher than 50% in order to achieve high impact tough-ness, and according to Miura and Ogawa (2000), the lowest pittingcorrosion rate is also associated with an austenite content of 50%or more.

In addition to the problem of nitrogen loss from the laser weld,the high solidification rates associated with laser welding tend tosuppress austenite formation. Welds were therefore produced atprocess parameters designed to show the effects of two differentsolidification rates as well as the effect of nitrogen as a shield gas.

2. Experimental procedure

The test welds were made by using a 4 kW disc laser with

300 mm focusing optics. Shielding gas was blown over the weldusing a pipe in front of the keyhole, with a 60 mm shielding gasnozzle behind the keyhole. The root sides of the welds were alsogas shielded, see Fig. 1.

382 M. Keskitalo et al. / Journal of Materials Processing Technology 216 (2015) 381–384

Table 1The weld parameters used.

Material Laser power (kW) Speed (m/min) Focal point position (mm) Shielding gas and flow rate Energy input (J/mm)

LDX2101®

1.5 mm

3.0 9.0 −1.0 Argon 30 l/min 203.0 9.0 −1.02.0 1.5 −6.0

2.0 1.5 −6.0

sm1

Fig. 1. The shielding gas arrangements.

The welded material was 1.5 mm LDX 2101® duplex stainlessteel. The Ni-equivalent was 7.3 and Cr-equivalent was 21.7, whicheans that the austenite content was 45% (Kyröläinen and Lukkari,

999).

Fig. 2. Cross sections of the 3 kW, 9 m/min welds:

Fig. 3. Cross sections of the 2 kW, 1.5 m/min w

Nitrogen 30 l/min 20Argon 30 l/min 80Nitrogen 30 l/min 80

The laser weld parameters, shown in Table 1, were chosen inorder to show the influence of heat input and nitrogen shieldinggas on the austenite content of the weld.

Five samples of both argon and nitrogen shielded welds under-went bend testing using a manually operated bending machine.

3. Results

Cross-section macrographs of the four welds are presented inFigs. 2 and 3. There are minor differences in weld shape as a resultof changing the shield gas, but the main difference is the increasein weld width between the fast (9 m/min) and slow (1.5 m/min)welds.

This increase in weld width at reduced welding speed is a resultof increased lateral thermal conduction associated with longerlaser-material interaction times.

The cross sections presented in Fig. 4 were etched in NaOH liq-uid using a voltage of 2.5 V for 10 s. This etching technique makesthe ferrite areas corrode more and become darker. The austen-ite and ferrite phase contents of the welds were then measured

(a) argon shield gas, (b) nitrogen shield gas.

elds: (a) argon shield, (b) nitrogen shield.

M. Keskitalo et al. / Journal of Materials Processing Technology 216 (2015) 381–384 383

Fig. 4. Micrographs of etched welds E = 20 J/mm. (a) Argon shield gas. (b) Nitrogen shield gas.

Table 2The austenite content of the Laser welds (the Austenite content of the unweldedmaterial is 45%).

Shielding gas Energy input 20 J/mm(9 m/min)

Energy input 80 J/mm(1.5 m/min)

fawgp

3

hdr(ac

3

a

Fig. 5. The hardness profiles of LDX 2101® steel laser welds with different heatinputs and shielding gases.

Table 3The maximum bending angles for different parameters.

Shieldinggas

Energy input20 J/mm (9 m/min)

Energy input80 J/mm(1.5 m/min)

Nitrogen Max bending angle (◦) 180 180

N2% austenite 8.1 28.8Ar % austenite 1.3 16.8

rom the cross sections by using the area measurement system of Keyence VHX 2000 microscope. The austenite contents of theelds are shown in Table 2. It is apparent that the use of nitro-

en as the weld shield gas increases the austenite level of the weldroduced.

.1. Hardness results

Typical hardness profiles of the four laser welds with differenteat input and shielding gases are shown in Fig. 5. It is evi-ent from these results that the slower, lower heat input weldsetained almost the same hardness as the as-received materialabout 260 Hv). The higher speed, lower energy input welds had

much higher hardness – reaching values of over 360 Hv near theentre of the welds.

.2. Bending tests

The results of the bending tests are shown in Table 3.Fig. 6 shows cross sectional micrographs of argon shielded

nd nitrogen shielded welds. It is evident that the argon shielded

5 samples in each case.Argon Max bending angle (◦)

5 samples in each case.30–40 180

Fig. 6. (a) 20 J/mm, argon as shielding gas. Brittle fracture at 30◦ bending angle, (b) 20 J/mm, nitrogen as shielding gas. 180◦ bending.

3 Proce

waob

4

rbLot

erthrmihri

iapTwbssiw1

84 M. Keskitalo et al. / Journal of Materials

eld is far less ductile than the nitrogen shielded sample. Thergon shielded weld experienced brittle fracture at a bend anglef approximately 30◦ whereas the nitrogen shielded weld could beent through 180◦ without failing.

. Discussion

When LDX 2101® Duplex stainless steel is welded the mate-ial solidifies as Ferrite and then some of it transforms to austenitey solid state diffusion during the cooling process (Kyröläinen andukkari, 1999). This transformation is supported by the presencef dissolved nitrogen. From the results presented here we can seehat:

Austenite formation is reduced at higher welding speeds. Thisffect is shown in Table 2 where the 9 m/min (20 J/mm) weldsesulted in structures containing considerably less austenite thanhe slower 1.5 m/min (80 J/mm) welds. The reason for this is thatigher welding speeds involve higher solidification and coolingates. If the material cools down too quickly there is not enoughobility in the lattice for the rearrangement from ferrite to austen-

te. The result is a stressed ferritic structure which has a higherardness than the original material’s 45:55 mix of austenite and fer-ite. This increase in hardness for the higher speed weld is evidentn Fig. 5.

The use of nitrogen as the weld assist gas increases the austen-te content of the weld (see Table 2). This is probably due to thebsorption of nitrogen by the melt which, to some extent, com-ensates for the loss of nitrogen experienced during laser welding.he increase in austenite levels in both the low and high speedelds did not have a strong effect on weld hardness (see Fig. 5)

ut produced a considerable improvement in weld toughness –ee Fig. 6 and Table 3. The higher speed welds created with argon

hield gas could only be bent between 30 and 40◦ before experienc-ng brittle fracture. The weld produced under the same conditions

ith a nitrogen gas shield could be successfully bent through80◦.

ssing Technology 216 (2015) 381–384

5. Conclusions

• Laser welding with argon shield gas results in a greatly reducedlevel of austenite in the weld metal.

• Higher welding speeds are linked to lower austenite levels andharder, more brittle welds.

• The use of nitrogen (instead of argon) as the weld shield gasincreases the austenite content of the weld. This is probablylinked to the uptake of nitrogen by the weld pool and this com-pensates to some extent for nitrogen lost by the weld pool as aresult of the laser welding process.

• Nitrogen as a shield gas has minimal effects on weld hardness butimproves weld toughness.

Acknowledgements

The research was carried out as part of the programme Inter-reg IV A Nord, project PROLAS, no. 304-58-11. The authors wouldlike to acknowledge financial support from the Interreg IV A Nordprogram, Lapin Liitto and Länsstyrelsen Norrbotten as the nationalfunding bodies. We would also like to thank the Private financiersOutokumpu Oyj, Randax Oy, Ocotec Oy and Miilux Oy of thePROLAS-project. The authors would especially like to express theirgratitude to Outokumpu Oyj for providing the test materials inaddition to their valuable support during the research.

References

Keskitalo, M., Mäntyjärvi, K., 2013. The influence of the shielding gas to the static anddynamic strength properties of laser welded work-hardened nitrogen alloyedaustenitic stainless steel. Key Eng. Mater. 549, 471–476.

Kyröläinen, A., Lukkari, J., 1999. Ruostumattomat teräkset ja niiden hitsaus. Metal-liteollisuuden keskusliitto, MET.

Miura, M., Ogawa, K., 2000. Hydrogen embrittlement cracking in duplex stainlesssteel weld metal. In: IIW Doc., IX-1461-87.

Outokumpu, O., 2010. Welding Handbook, first ed.Sakai, Y., Aida, G., Suga, T., Nakano, T., 1989. Development of various flux-cored wires

and their application in Japan. In: IIW/IISS Doc. XII-1131-89.Westin, E.M., Serrander, D., 2012. Experience in welding stainless steels for water

heater applications. Weld. World 56 (May (5–6)), 14–28.