Weldability of Iron Based Powder Metal Alloys Using Pulsed GTAW Process

effective techniques joining. introduce relatively established methodologies and techniques which have been studied, developed and applied in industries researches. State-of-the-art development aimed at improving technologies will be presented covering topics such as weldability, technology, automation, modelling, and measurement. This book also seeks to provide effective solutions to various applications for engineers and researchers who are interested in arc material processing.



Maintain close dimensional tolerances  Permits a wide variety of alloy systems  Produces good surface finishes  Provides materials which may be heat-treated for increased strength or increased wear resistance  Facilitates manufacture of complex or unique shapes which would be impractical or impossible with other metalworking processes.  Suited to moderate-to high volume component production requirements  Offers long-term performance reliability in critical applications  Cost effective  Provides controlled porosity for self-lubrification or filtration

Influence of the porosity
The most prominent microstructural feature of a PM component is its porosity, which affects virtually all its physical properties and, consequently, its weldability. The nature of the porosity is controlled by several processing variables such as green density, sintering temperature and time, alloying additions, and particle size and type of the initial powders.
In particular, the fraction, size, distribution and morphology of the porosity have a profound impact on mechanical behaviour, especially in components under welding conditions (Chawla & Deng, 2005;Sudhakar et al., 2000). Firstly, the pores act as thermal insulators which slow the transfer of heat, affecting considerably the thermal conductivity of the PM material to be joined. As the change in heat transfer naturally affects the welding parameters, the welder needs constantly to adjust them to assure the good quality of the weldment. Also, since the amount of porosity reduces the thermal conductivity, the cooling rate of the material also slows, reducing the hardening tendency (Hamill, 1993;Kurt et al., 2004;Kumar et al., 2007). The thermal expansion is another important physical characteristic which is influenced by the porosity. Potential changes in the porosity volume fraction during welding, due to smaller particle melting or filler metal infiltration, can result in excessive shrinkage or growth. As a consequence, subsequent cracking can occur in the heat affected zone (HAZ) or in the fusion zone (ZF) of the PM base metal. Porosity can also cause erratic fluctuations in welding performance as well as other welding defects because of entrapped oxides or impurities within the structure. These oxides and impurities may be originated from lubricant residues and quench oils.

PM welding process
The selection of the welding process more suitable to join PM parts should be made taking an account the requirements desired such as strength, environmental factors, appearance and the porosity volume. According to the literature (Hamill, 1993;Jayabharat et al., 2007), fusion welding processes are used successfully to join ferrous powder metal parts with high density (> 7.0 g/cm 3 ) once these high density PM parts typically have the same weldability as forged, rolled or cast materials. Indeed, research of Hinrichs et al showed that it is possible to obtain good quality dissimilar weldments of low and medium carbon PM steels joined with forged steels using the most common fusion welding processes. The researchers also showed that in the case of the welding of PM high-strength low-alloy (HSLA) steels, procedures such as pre-heating and hydrogen control should be adopted to guarantee the success of the joining. Low and intermediate density PM parts (< 7.0 g/cm 3 ) should be joined using welding processes which minimize the volume of molten weld metal such as resistance projection welding, friction welding and brazing. The reason is that the low fracture resistance of these PM materials, caused by the small number of bonding between the particles, does not allow that these absorb the residual stress produced by the high densification that occurs in the heat affected zone and shrinkage of the weld metal, resulting in subsequent cracking in or near the weld interface. Additionally, when choosing the brazing process to join the low or intermediate density PM parts, a special attention should be paid to the capillary force o f the pores once the porosity near the joint wicks the copper brazing filler metal into the pores, leaving an insufficient amount of filler metal to establish the satisfactory weld strength. To overcome this problem, the PM parts must be copper infiltrated before brazing (Hamill, 1993;Jayabharat et al., 2007).

Pulsed GTAW process
The Gas Tungsten Arc Welding (GTAW) is a fusion welding process, where arc is produced between non-consumable tungsten electrode and base metal. This process provides suitable results in many situations because of its ability to control the welding parameters (heat input, travel speed, feed rate and type of filler metal) during the welding and subsequent weld metal and HAZ hardness. The higher control of the heat input in this process compared with other fusion welding processes may be mainly attributed to the fact of the welding arc does not suffer direct interference of the metal transference during welding. Pulsed GTAW involves cycling of the welding current from a high level to a low level at a selected regular frequency. Thus, pulsing the current introduces additional operational parameters, which include peak current, base current, peak pulse time and base pulse time (Pawan et al., 2011). Figure 1 shows the representation for the pulsed current.

www.intechopen.com
Pulsed GTAW is frequently used for welding of several materials as heat input can be precisely controlled. Also, this process is strongly characterized by the bead geometry control, which plays an important role in determining the mechanical properties of the weldment (Juang & Tarng, 2002).
In contrast to continuous current welding, in the pulse mode the heat energy required to melt the base material is supplied only during peak current pulses for brief intervals of time allowing the heat to dissipate into the base material. With this, it is possible to achieve the maximum penetration without excessive heat build-up (Juang & Tarng, 2002).
As a result of all mentioned above, it is possible to obtain weldments with a narrower HAZ as well as reduction of segregation of alloying elements, of residual stress and of the hot cracking sensitivity (D' Oliveira et al., 2006;Wang et al., 2006). Current pulsing also results in periodic variations of the arc forces and in an increase of the melt pool agitation leading to additional fluid flow, which lowers the temperature in front of the solidifying interface. This temperature fluctuations leads to the continual changes in the weld pool size and shape favouring the growth of new grains (Pawan et al., 2011). As a consequence of this grain refinement in the fusion zone, an improvement of the mechanical properties, such as tensile and fatigue resistance is achieved. Therefore, through the use of pulsed parameters, it is possible working with high current peaks without increase the average heat input to the base material, which enables itself as a good choice for welding powder metals alloys.

Weldability study of iron based powder metal alloys
According to the literature (Hamill, 1993), a unique conventional GTAW application involving welding of PM parts was the replacement of a casting for two welded PM components together for use in a commercial truck differential. It was found that the welded PM components exhibited higher and more consistent strength values than a bolted gray iron casting along with providing a 35% cost savings compared with the previous method of manufacture. However, there was not enough information about the welding procedures (parameters, heat input, gas shield flow, etc) used to carry out the welding. This topic therefore intends to provide to the readers experimental information about welding of iron based PM alloys using the pulsed GTAW.
The materials involved in this study were three different iron based powder metal alloys whose compositions and features are given in table 1. As mentioned before, these alloys are the largest and the most effective alternative PM parts group for replacing castings, forged and machined parts, mainly in the automobile industry. In order to obtain the PM alloys, the powder metal were first mixed with lubrificant (zinc stearate) in a ball mill according to the chemical compositions of the alloys given in table 1 to produce a homogeneous mixture of ingredients. After that, the mixed powder of each alloy was compacted to about 90% relative density in a press (green compact) and then sintered in a pure hydrogen atmosphere according to the thermal cycle shown in Figure 2 to complete the metallurgical bonds between powder particles. The powder metal samples (dimensions: 100 mm x 20mm x 7 mm) produced were welded in the butt joint, flat position with three different filler metals (AWS R 70s-6, AWS R 309L, AWS R Fe-Ni) using four passes weld by a manual pulsed GTAW process. The pulse welding parameters used were the same to the three different alloys and were chosen after preliminary tests that guaranteed an arc stability and lower heat input. A flow rate of argon (99.99% purity) of 7 l/min was used as a shielding gas. The travel speed was adjusted to give an adequate penetration and weld bead contour. These welding parameters are given in table 2. After welding, the test samples were transverse sectioned, polished with Al 2 O 3 and etched with 2% nital (HNO 3 + alcohol). Microstructural examination of the specimens was carried out using standard optical microscopy and scanning electronic microscopy (SEM). Vickers hardness values were taken across the transverse section using a 10 Kg load. The tensile test samples geometry is shown in Figure 3 and these were in accordance with ISO 2740 standard.     In general, phosphorus is intentionally added in powder metal iron based alloys to increase the densification of the iron powder once this element allows the formation of a transient liquid phase during sintering. Furthermore, phosphorus is known to improve the corrosion resistance and magnetic properties of the powder metal iron-based parts (ASM Handbook, www.intechopen.com 1999). However, phosphorus additions are not particularly attractive for fusion welding applications because its presence in the metal base composition is associated with the formation of the eutectic M 3 P, which may promote solidification cracking in the fusion zone. Therefore, the amount of phosphorus added in the alloy must be rigorously controlled (Correa et al., 2008). b. Powder metal Fe-Ni-P According to author (Beiss, 1989), PM carbon steels with additions of 0,35% P may be fusion successfully welded without the occurrence of solidification cracks since the carbon content is lower than 0,2%. PM carbon steels with higher carbon contents tend to facilitate the segregation of the phosphorus and the formation of the M 3 P eutectic. Figure 6 shows the microstructure of the transverse section of weld joint of the alloy Fe-Ni-P using Fe-Ni (60% Fe-40% Ni) filler metal. Despite the good toughness of the Fe-Ni filler metal to absorb the shrink stresses during the weld metal cooling, the weld metal presented solidification cracks and pores after the pulsed GTA welding. The weld solidification cracking can be mainly attributed to the presence of the low-melting eutectic Fe 3 P and/or Ni 3 P in the weld pool. According to the literature (Lancaster, 1987;Lippold & Kotecki, 2005), the excessive amount of phosphorus combines with nickel or iron forming the low-melting eutectic Ni 3 P or Fe 3 P. The continuous presence of the segregated Ni 3 P or Fe 3 P liquid film in the last stages of solidification o fthe weld pool combined with the higher shrinking stress due to faster cooling rates during fusion welding, may have contributed to the appearance of solidification cracking (Briskman, 1979). In addition, considering the potential of the nickel as an austenite stabilizer, the Fe-Ni filler metal solidifies in the austenitic mode, which increases the segregation of phosphorus in the weld pool and, consequently, the susceptibility to cracking (Lippold & Kotecki, 2005). Figures 11 and 12 show the presence of phosphorus eutectic in the HAZ and fusion zone of the powder metal Fe-Ni-P alloy using Fe-Ni filler metal, which is characterized by the presence of small "islands" in the ferritic grains. As can be seen in Figure 13, a complete elimination of the weld solidification cracking and porosity in the fusion zone of the Fe-Ni-P alloy was possible using the filler metal 309L stainless steel and adjusting the welding parameters to those values shown in table 2. According the literature (Lippold & Kotecki, 2005), the principal reason for the absence of solidification cracking in the weld metal of the 309L filler metal is the low carbon content of the 309L filler metal and, mainly, the presence of a two-phase austenite/ferrite mixture in the microstructure of the weld metal. (See Fig 14).    Image analysis results (Table 3) from optical micrographs, similar to that showed in Fig 14, together with observations in the WRC-1992 diagram, indicated that the ferrite number (FN) in the 309L stainless steel weld metal was approximately 7 FN. In general, above 3 FN, but less than 20 FN, solidification of austenitc stainless steels is most likely in the FA mode (Suutala, 1983 Table 3. Image analysis results of the delta ferrite volume fraction of the 309L weld metal. In the FA mode, the duplex microstructure (delta ferrite + austenite) presents at the end of solidification, increase the amount of tortuous phase boundaries that resist wetting by liquid films and along which cracks might propagate. Thus, once the crack is nucleated, it becomes very difficult for it to propagate along to the nonplanar crack path generated to these tortuous bondaries (Briskman, 1979). Additionaly, as the solubility of the phosphorus in the ferrite is higher than that observed in the austenite, the delta ferrite in the weld metal is prone to absorb a significant amount of this element, which reduce the concentration of the phosphorus in liquid film, avoiding the permanence of the segragated low-melting eutectic until the last stages of solidification and, consequently, the solidification cracking. It can be also seen in Fig. 14 the vermicular morphology of the delta ferrite. In general, this ferrite morphology is present when welding cooling is moderate and/or when the Creq/Nieq is low but still within the FA mode (Lippold & Kotecki, 2005).
It is worthwhile mentioning that other significant factors involving the pulsed GTA welding of teh powder metal Fe-Ni-P alloy may have also contributed to minimize the occurence of solidification cracking. These may be, for instance the utilization of pulsed current and multipass weld, which have the effect of refinement of the as-cast microstructure in the fusion zone (Balasubramanian et al., 2008). Also, it can be noted that there was no presence of pores in the weld metal. However, th e pores in the base metal of this alloy (see Fig. 15) were large and rounded with higher densification of ferrite in their surroundings. The size of the Fe-P pre-alloyed particles added in this alloy probably is the cause of the pores characteristics (large and rounded) in the base metal. Figure 16 shows that Fe-Ni-P alloy welded with 309L austenitic stainless steel filler metal did not evidence significant changes of hardness profile in the HAZ in comparison with the base metal, even though the phosphorus is prone to harden the ferrite by solid solution.

Tensile tests of the weldments
Tensile tests carried out in the welded samples of the pure Fe and Fe-Ni using AWS R 70S-6 and AWS R Fe-Ni filler metals (filler metal of mild steel and Fe-Ni alloy) and Fe-Ni-P powder metal alloy using AWS R 309L filler metal showed that the failures of samples occurred always in the base metal. Furthermore, the welded samples of these alloys presented ultimate tensile strength slightly higher than unwelded samples (see Table 4). The higher tensile strength of the welded samples may be attributed to the residual stress in the samples due to their small dimensions (width and length) combined with the high heat input and the relatively rapid cooling of the weld metal during welding. It is worthwhile www.intechopen.com  mentioning that the welded samples were not annealed after welding. However, the tensile tests results together with the hardness results indicated that weldments of these materials presented a good continuity of the mechanical properties with relation to the base metal, even when these materials are welded by a fusion welding process.

Conclusion
In this chapter was shown the benefits of the PM components and the advantages of their utilization in replacing casting, machined and forged materials. However, experimental information about joining these PM materials using fusion welding processes is still scarce.
Through of the study presented in this chapter, it was possible to verify that PM iron based alloys pure Fe and Fe-Ni may be successfully welded by pulsed GTAW process without additional techniques such as preheating, post-heating or special joint configuration.
Concerning to the Fe-Ni-P alloy, this alloy also may be successfully welded by pulsed GTAW process but a discerning selection of the filler metal and careful control over the welding parameters (heat input, peak current, base current, peak time, base time and travel speed) must be done. Due to the presence of phosphorus, the correct selection of the filler metal may avoid the presence of low-melting eutectic films at the end of the solidification and, consequently, the solidification cracking. A rigid control of the heat input, in turn, prevents the higher dilution of the base metal, which decreases the shrink stress in the fusion zone during the solidification of the weld pool leading to a lower susceptibility to cracking of the weldment.
Hardness results showed that no excessive hardening was observed in the weld metal and HAZ of the iron powder metal alloys studied. This is in agreement with the base metal fracture location in the tensile tests and with the slight increase of the tensile properties of welded samples in comparison with unwelded samples. In general, excessive hardening (higher Vickers hardness values) is prone to increase significantly the tensile strength and decrease the toughness.

Acknowledgment
The author acknowledges the Brazilian Government Agencies CNPq, CAPES and FAPEMIG for the financial support to carry out this study. ASM Handbook, (1999). Metallography and Microstructure,, vol. 9, ASM, USA Briskman, A.N. (1979). The effect of welding currente pulses on the susceptibility of weld metal to hot cracking during argon TIG welding, Aut. Weld. (7), pp. 40-43.