Weldability Characteristics of Sintered Hot-Forged AISI 4135 Steel Produced through P/M Route by Using Pulsed Current Gas Tungsten Arc Welding

Materials required

The chemical compositions in weight percentage for the base metal (AISI 4135) and the filler wire (ER 80S B2) are listed in Table 1. The scanning electron microscope (SEM) photograph of the morphology of powder (Figure 1) clearly shows the difference in shape and size of the particles. It also exhibits the weak aggregation of particles. Atomized iron powder of –150 μm was obtained from M/s Sundaram Fasteners Limited, Hyderabad, Andhra Pradesh, India. The chemical analysis of this powder revealed 99.6% purity with 0.4% insoluble impurities. Sieve shaker with sieves was required for sieve size analysis of iron powder and a pot mill with stainless steel pots for blending the powder mixes was also required. Graphite powder of 3–5 µm was procured from the M/s Ashbury Inc., USA, for using in the steels as alloying element to be experimented. High-carbon high-chromium steel bars of 30 mm diameter and 200 mm length were required for designing, fabricating and suitably heat treating the compaction die. The inner diameter of this die was 30 ± 0.10 mm and length was 145 mm. Punch height was 160 mm with a diameter of 30 ± 0.10 mm and the bottom insert was of 15 mm height with a diameter of 30 ± 0.10 mm. However, flat-bottomed platen of length 240 mm width 140 mm and thickness of 75 mm of molybdenum die steel was manufactured. Similarly the top punch was of similar design with the provision to be firmly attached with the top punch of the friction screw press.

Figure 1:

Morphology of iron powder.

Powder blend preparation

Powder blends of required compositions were prepared by mixing various elemental powder constituents by weight and adequately blending them on a pot mill so as to yield alloy compositions after sintering as corresponding to AISI4135 steels, respectively. Zinc stearate was required as lubricant. These mixes were blended for a period of 30 h so as to obtain homogeneous blends. The powder mix to porcelain balls of 15–20 mm diameter ratio by weight was maintained as 1:1. At the end of every 1 h of lapse, the pot mill was switched off and 100 g powder was taken out to measure the flow rates and apparent densities. Immediately after the completion of measurements, the powder mixes were returned back to their respective pots and the pot mill was operated again. Once the last three readings of flow rates and densities were consistently almost same, the blending operation was discontinued and the powder blends were ready for compaction. Iron powder and AISI4135 steel blends were characterized as shown in Table 2. The morphology of the particles after blending was analysed by using the SEM shown in Figure 2. Element mapping by energy-dispersive X-ray spectrometry (EDS) of blended powder is shown in the figure. It is shown that all the elemental powders were uniformly distributed throughout the powder blended for making samples.

Figure 2:

EDS mapping of elements in blended powder of AISI 4135 steel.

Powder blend compaction

AISI 4135 steel blends were compacted in the die designed, fabricated and suitably heat treated. In order to compact powder blends to initial compact aspect ratio of 1:1 with the diameter being 30 mm, a preweighed powder blend was taken which was calculated for the initial compact density of 85 ± 1% of theoretical. This was obtained at a pressure of 480 ± 10 MPa. All compacts of 1:1 aspect ratios for all P/M steels tested were made by taking accurately weighed powder blends with well-monitored applied pressures. Compact ejections were made by using a hollow cylinder of 75 mm inner diameter and 100 mm outer diameters of length 180 mm compact with the bottom insert along with punch were placed on this cylinder and assembly was placed on the UTM platform and the punch pressured the bottom insert. The compacts were cushioned and similarly the punch was separately removed and after cleaning the assembly was ready for compaction again.

Protection of compacts during sintering

During sintering compacts get oxidized if they were not sintered under protective environments. In the present investigation, indigenously [31] developed ceramic coating was applied to all the compacts on their entire surfaces. This first coating was allowed to dry under ambient condition for a period of 12 h. These ceramic-coated compacts were recoated 90° to the direction of the previous coating and were once again dried under the ambient conditions for another period of 12 h. Now these compacts were ready for sintering.

Sintering of ceramic-coated compacts

Ceramic-coated compacts were arranged in four ceramic trays. A tray contained only one composition. At a given time only one tray containing a particular composition was charged in an electric muffle furnace and the furnace temperature was raised to 750°C and retained the ceramic-coated compacts in the furnace for a period of 30 min and then the temperature of the furnace was raised to 1,150°C and the compacts were retained for a further period of 60 min and now the sintered compacts were ready for forging.

Hot upset forging to flat plates

Cylindrical compacts of initial aspect ratio of 1:1 for all samples were forged to flat plates of cross section of 50 mm × 5 mm and of length 120–140 mm. The forging schedule in these cases followed was first upsetting then rotating 90° angle and forging, again rotating 90° angle and forging. Nearly nine such steps were carried out for each compact to transfer to the required dimensions. Further these samples were cooled inside the furnace to maintain the homogeneity of microstructure.

Welding

The sintered hot-forged P/M plates of steel with a density of 99 ± 0.05 of theoretical density were machined into 120 mm × 50 mm × 3 mm by surface grinding. The initial joint configuration was obtained by securing the plates in position using tack welding. Normal butt joints were fabricated using PCGTAW. All necessary care was taken to avoid joint distortion and the joints were made by applying clamping devices. The plates were preheated to a temperature of 150 ± 10°C for avoiding distortion during welding. The welding conditions and process parameters used to fabricate the joints are listed in Table 3. High-purity (99.99%) argon gas was used as shielding gas in PCGTAW process. The reliability of all the welded plates was checked using radiographic testing. The welded joints were sliced using a wire-cut electron discharge machine and then machined to the required dimensions for preparing tensile test specimens. The unnotched smooth tensile specimens were prepared to evaluate the transverse tensile properties of joints such as tensile strength and percentage of elongation. The tensile test was conducted with the help of a 2 kN, electromechanically controlled Hounsfield Tensometer. ASTM-E8/E8M-11 standard guidelines were followed for preparing and testing the tensile specimens. Vickers’s microhardness testing machine was utilized for measuring the hardness with a 500 g load and dwell time of 12 s. Microstructural examination was carried out using a light optical microscope incorporated with an image analysing software. The specimens for metallographic examinations, both macrostructure and microstructure, were sectioned to the required size from the joint comprising weld metal, HAZ and base metal regions, and were polished using different grades of emery papers. The final polishing was done using the diamond paste in the disc polishing machine. The specimens were etched with 3% Nital solution. The fractured surfaces of the tensile tested specimens were analysed by taking SEM fractographs, revealing the morphology of the fractured surfaces.

Microstructural analysis of weld joint

Figure 3 shows the macrostructure of the pulsed current GTAW AISI 4135 joint. The welding cross section exhibited no volumetric defect. The macrostructure of the joint could be split into three distinct regions, namely fusion zone, heat-affected zones and the base metal. Figure 4(a) and 4(d) depicts optical micrographs of the parent metal etched for metallographic analysis. The parent metal microstructure was found to be ferritic–pearlitic, with well-defined equiaxed ferrite grains present all over. The fine-grained ferrites were observed to have well-defined grain boundaries. Fineness of the ferrite structure was attributed to the forging treatment carried out after sintering. Unresolved pearlites along with numerous, round carbide particulates were observed along grain boundaries. These microphotographs clearly showed that they mainly contained a mixture of ferrites and pearlites, which were in elongated forms like fibres in the horizontal direction. Further, the presence of porosities was negligibly small as the achieved density was beyond 99% of the theoretical of the parent steel. Figure 4(c) and 4(f) exhibits microstructures of the weld zone. The entire weld zone consisted of acicular ferritic microstructure. However, the heat input of welding, cooling rate and concentrations of constituents alloying elements in the steel were the major governing factors for formation of acicular ferrite structure in the weld zone. Since the preheating operation of the plates prior to welding was performed, the faster cooling rate was restricted and this resulted in the formation of non-martensite structure in the weld zone. Apart from these, the low carbon content of the filler metal also restricted the formation of martensite structure in the weld zone. As a ferrite stabilizer, chromium, which is one of the major constituent in the filler wire, also played a role in the formation of ferritic structure in the weld metal zone. The excess carbon content in base metal might promote brittle martensite phase formation in the weld metal. But the low carbon content in filler metal as well as base metal has played a role in promoting acicular structure of the weld zone. Thus, the homogenized acicular ferrite formed inheriting fine grains throughout the weld zone. Figure 4(b) and 4(e) shows the microstructure of heat-affected zone. Even though the heat-affected zone did not exhibit any such common weld defects, the inherent porosity of P/M alloys was still observed in heat-affected zone. Varying concentration of fine bainites were observed in the heat affected zone which resulted in an increase in the hardness values.

Figure 3:

Macrograph of welded sample with 20×.

Figure 4:

Microstructure of (a) base metal, (b) heat-affected zone, (c) weld metal on 200× and microstructure of (d) base metal, (e) heat-affected zone and (f) weld metal on 400×.

Weld joint hardness analysis

The hardness across the weld cross section was measured using a Vickers microhardness testing machine, and the results are presented in Figure 5. The hardness values of the heat affected zone and weld metal region were found to be greater than that of the base metal region. The fusion zone region consisted of acicular ferrite, and the heat-affected zone region primarily consists of coarse and fine bainitic structure which was non-uniformly distributed. Further the micrographs clearly showed that there was an appreciable difference in the grain size in the fusion zone regions compared to the base metal regions. The microhardness across the mid-thickness of the welded joint was measured and the hardness of the as-received base metal was approximately HV 270. The hardness of the fusion zone varied from HV 290 to HV 293, and in heat-affected zone it gave HV 336. The higher hardness was due to the formation of bainitic structure at the heat-affected zone. But weld zone was having lower hardness than the heat-affected zone, due to the lower concentration of bainite.

Figure 5:

Microhardness values (VHN) of welded joints.

Tensile test results and fractography analysis

The tensile strength of carbon steel weld joint depends on the degree of grain refinement, the amount of pro-eutectoid ferrite and the porosity content. The maximum tensile strength and joint efficiency were achieved in PCGTAW due to reduced heat input. The transverse tensile properties such as tensile strength and percentage of elongation of the AISI4135 joints were evaluated. P/M alloy samples produced had attained 99 percentage of the theoretical density. The specimens of base metal and welded metals were tested, and the results are presented in Figure 6. The average tensile strength of the base metal was 685 MPa and the tensile strength of the PCGTAW joints were 703 MPa. These indicate that there was moderate increase in strength values in the PCGTAW. The welded alloy exhibited almost equivalent in ductile properties such as percentage elongation in length compared to parent alloy metal. But the P/M route base metal showed low ductile properties due to strain-hardening effect produced during hot upsetting process that could have considerably reduced ductility. The elongation of the base metal was 8.60%, and the elongation of PCGTAW joints was 8.50%. This suggests that there was a negligible difference in ductility after PCGTAW. The elongation difference is due to the difference between yield stress and stress at fracture. The failure occurred at base metal region also on the welded sample after tensile testing. This shows that welded alloy was more stronger than the base metal. Uniform grain refinement in weld zone, i.e. formation of acicular ferrite at the weldment and residual stresses formed at base metal after welding were the reasons for increase in strength of weldment. Further the absence of porosity, blowholes and other defects in the weld metal had promoted greater strength of the welded joint.

Figure 6:

Tensile properties of AISI 4135 base metal and welded joint.

The X-ray diffraction (XRD) patterns of AISI 4135 base metal and as-welded sample are shown in Figure 7. In Figure 7(a), there are sharp peaks that correspond to the XRD pattern of iron. Also, there are other weak peaks that indicate the presence of alloying element like chromium phases which is the major alloying element in the Base metal. In Figure 7(b), the sharp peaks of iron and weak peaks of Cr₃C, Cr₃Ni₂Si and Mo₃Si found correspond to the XRD pattern. Chromium in the form of chromium carbide precipitates increased the strength by means of precipitation strengthening. Secondary chromium carbides pinned the grain boundaries and inhibited the grain growth. This resulted in grain refinement and the presence of second-phase particles made dislocation movement more difficult. Second-phase particles like chromium carbide in the matrix increased the energy required for elastic/plastic deformation, and hence created higher strength in the alloy. Nickel did not produce any second-phase particle. Ni was found mostly in the form of solid solution in the ferrite, so Ni increased the strength of the steel by solid solution strengthening. Ni also lowered the transformation temperature, so the lower transformation temperature produced smaller ferrite grains. Besides that, from the microstructural observation it was seen that the presence of Ni promoted acicular ferrite formation. Change in morphology of the ferrite to acicular ones also produced obstacle in dislocation glide. Thus nickel increased the strength by refining the grains by lowering the transformation temperature and also changing the morphology of the ferrite grains. Cr3Ni2Si (siliconitride) was found in iron-based high-temperature alloys. The lattice was FCC (face-centred cubic) and the space group was Fd 3m. The space group was identical with M6C and the lattice constant was very close to M23C6, which might be present together with Cr3Ni2Si(C, N). Mo₃Si formed through a peritectic reaction, L + Mo = Mo₃Si at 2,025°C. Excellent oxidation resistance was observed in ternary (Cr, Mo) 3Si alloys, suggesting that Mo₃Si might have good oxidation resistance. Mo₃Si exhibited very high hardness and excellent hardness retention even at temperature of 1,300°C.

Figure 7:

XRD patterns of (a) AISI 4135 base metal, (b) weld portion.

Figure 8 displays the SEM fractographs of tensile-tested specimens of base metal and welded AISI 4135 joints. The SEM was preferred for fractographic examinations due to its better resolution and depth of field than optical microscope that revealed the topographical features of fractured surfaces. The displayed fractographs of numerous spherical dimples indicated that most of the tensile-tested specimens in Figure 8(a) and 8(c) failed in a ductile manner under the action of tensile loading. The SEM fractographs taken on the fractured surface of the tensile-tested specimen was prepared from the sintered, hot-forged, homogenized and furnace-cooled specimen. Presence of large number of dimples was the manifestation of ductile failure, but there is also the presence of residual porosities that had ruptured during tensile testing which added to brittleness and thus the overall fracture mode could be termed as monthly ductile but partially brittle.

Figure 8:

Fractographs of tensile specimens: (a) and (c) base metal, and (b) and (d) welded metal.

However, parabolic-shaped dimples characteristic of ductile fracture can be observed in Figure 8(b) and 8(d) of welded joints compared with the base metal. Number of flat cleavage-like facets are visible, but, at the same time, the presence of good number of dimples can also be seen in the fractograph. Presence of dimples and also few particles of undissolved are seen in the fractograph. At places, large-sized pores are visible, indicating the situation where the rupture had initiated along their surfaces. However, the enhanced strength and hardness indicates the mixed mode of failure, i.e. partly brittle and partly ductile. Few dimples contained the fractured fine fragments of harder particles, namely molybdenum. But in both cases the elemental powders agglomeration had taken place in some areas, which could be a reason for the failure. And in weld area, the grain refinement had taken place during welding process and led to weld area being stronger than base metal. The EDS analysis of fractured surfaces of base metal and welded samples is given in Figure 9. In both fractured surfaces, elemental powders agglomeration had taken place. Figure 9(a) shows that fractured surface of base metal, silicon and manganese powders have got agglomerated. This shows that fractional percentage of elemental powders in this composition in the marked area was higher than that of overall composition. But in Figure 9(b) the percentage of chromium and molybdenum is nil. So, overall composition was matching with agglomerated combination of silicon and manganese composition.

Figure 9:

EDS analysis of fractured surface of (a) base metal and (b) welded joint.