A. Microstructural characterization

The investigation of welds was performed from BM across the HAZ to WM. Figures 1(a) and 1(b) illustrate the low magnification SEM macrographs of both the welds, sound welded joints without any defects were observed. Figures 1(c) and 1(d) show the optical micrographs of DSS 2205 and AISI 430 FSS BM, respectively. Both BMs had a different microstructure, DSS 2205 BM showed combined ferrite and austenite dual hybrid phases, ^{Reference Verma, Taiwade, Khatirkar, Sapate and Gaikwad13} while AISI 430 FSS BM illustrated ferrite with an island of grain boundary martensite. ^{Reference Ramkumar, Chandrasekhar, Singh, Ahuja and Arivazhagan8}

FIG. 1.

(a) E2209 weld macrograph, (b) E309LMo weld macrograph, (c) DSS 2205 BM microstructure, (d) AISI 430 FSS BM microstructure, (e) X-ray diffraction pattern of DSS 2205 BM, (f) X-ray diffraction pattern of AISI 430 FSS BM.

X-ray diffraction technique [Figs. 1(e)–1(f)] revealed the phases present in the BMs and confirmed the various peaks of austenite [(111), (200), (220), and (222)] and ferrite [(110), (220), and (211)] in DSS 2205 BM [Fig. 1(e)]. Whereas in AISI 430 FSS BM [Fig. 1(f)], (110), (200), (211), and (220) peaks of ferrite and (110) peak of martensite were evident. However, variation in the peak intensity was noticed and no other intermetallics were observed in the BMs.

The HAZ of both the BMs side is shown in Figs. 2(a) and 2(b). DSS 2205 side HAZ illustrated ferrite grains with grain boundary austenite (GBA), Widmanstatten austenite (WA) and intragranular austenite (IGA), ^{Reference Verma and Taiwade3,Reference Aguilar, Tabares and Serna15,Reference Lippold and Kotecki22} while large grains of ferrite were observed along with martensite in AISI 430 FSS side HAZ. It was reported ^{Reference Mas, Martin, Lhuissier, Bréchet, Tassin, Roch, Todeschini and Simar23} that the microstructure of HAZ was mainly influenced by welding heat source and physical properties of the materials, whereas WM microstructure can be controlled by both chemical composition of electrodes and welding parameters.

FIG. 2.

Optical micrographs of dissimilar weldments (a) DSS 2205 BM side HAZ, (b) AISI 430 FSS BM side HAZ, (c) Espy a modified Scheaffler diagram, (d) Fe–Cr–Ni pseudo binary diagram, (e) E2209 weld, (f) E309LMo weld.

The weld microstructure was predicted by Espy diagram [Fig. 2(c)] by considering the chemical composition of BMs, electrodes and weld dilution.

The level of mixing between BMs and electrodes majorly affects solidification mode and ferrite content, as the dilution increases solidification mode changes from fully ferritic (F) to ferritic-austenitic (FA) and fully austenitic (A) mode. ^{Reference Eghlimi, Shamanian, Eskandarian, Zabolian, Nezakat and Szpunar24} The measurement of the WM dilution level was quantified through the area ratio of BM fused [A _BMF = (A _{430 side BMF} + A _{2205 side BMF})] of both sides to the total weld deposit (A _TWD) and it is expressed as a percentage dilution followed by equation. ^{Reference Verma, Taiwade, Khatirkar, Sapate and Gaikwad13,Reference Verma and Taiwade14,Reference Mukherjeey and Pal25}

Metallographic method (by SEM) and ImageJ software were used to calculate the individual geometrical cross-sectional area of the melted BM and deposited electrode. Since in dissimilar joint both DSS 2205 and AISI 430 FSS showed different dilution behavior due to different thermal properties. The individual dilution of DSS 2205 and AISI 430 FSS in the welds were calculated as mentioned in Table I.

Cr_eq/Ni_eq ratio of BMs, electrodes and welds (as per ≥46.8 and ≥48.7) were calculated and indicated in Espy diagram [Fig. 2(c)]. The calculation was done as per given equation. ^{Reference Verma, Taiwade, Khatirkar, Sapate and Gaikwad13,Reference Verma and Taiwade14}

Both the welds showed distinct ferrite and austenite morphology as per Espy diagram.

The Cr_eq/Ni_eq ratio of the welds and BMs were also plotted in Fe–Cr–Ni pseudo-binary diagram which is the most important tool to predict the final microstructure corresponding to solidification mode ^{Reference Verma, Taiwade, Khatirkar, Sapate and Gaikwad13,Reference Verma and Taiwade14} as it is shown in Fig. 2(d).

Figures 2(e) and 2(f) illustrates the peculiar microstructure of both the welds. In E2209 weld, the morphological features [Fig. 2(e)] were quite different than that of E309LMo weld, it appeared as a hybrid phase of austenite and ferrite due to F mode of solidification, L → L + δ → δ → δ + γ (where “L” is liquid, “δ” is delta ferrite and “γ” is austenite). ^{Reference Verma and Taiwade3} It was reported that if Cr_eq/Ni_eq > 1.95 considered to be an F mode. ^{Reference Verma and Taiwade3} E2209 weld showed higher Cr_eq/Ni_eq ratio than that of E309LMo weld, therefore, different microstructure evolved as indicated in Fig. 2(d).

The primary austenite which directly nucleated from columnar ferrite phase during solid state transformation, then WA a needle-like shape, branched from the GBA (primary austenite) showing to displacement and diffusion mechanism of alloying elements and IGA within the ferrite grains precipitated only at a lower temperature with higher undercooling. ^{Reference Verma and Taiwade3,Reference Eghlimi, Shamanian and Raeissi26,Reference Yang, Yan, Li and Wang27}

On the other hand, 309LMo weld [Fig. 2(f)] solidified in an F mode with lower Cr_eq/Ni_eq ratio [Fig. 2(d)] compared to E2209 weld, the reason of shifting toward F mode was due to high dilution of DSS and FSS BMs in the weld. The WM primarily consisted of a solidified dendritic structure of austenite with acicular ferrite in the cores of subgrain of austenite and remain stable as δ-ferrite at room temperature. ^{Reference Eghlimi, Shamanian, Eskandarian, Zabolian and Szpunar28} However, multi-directional grain growth comprising of cellular, dendritic and equiaxed grains were observed due to the chemical composition of the austenitic electrode having more austenite former elements (C, Ni and Mn). ^{Reference Eghlimi, Shamanian, Eskandarian, Zabolian and Szpunar28} Higher dilution of AISI 430 FSS and DSS 2205 was also expected in the weld during welding which increased ferrite content in the weld significantly resulted in acicular ferrite formation in the E309LMo weld.

Ferrite content of the welds was measured by ferritoscope based on the magnetic method. The measured ferrite content [in ferrite number (FN)] of the welds was 47 FN (34.3 vol%) in E2209 and 24 FN (22.9 vol%) in E309LMo. Whereas the predicted ferrite content according to Espy diagram [Fig. 2(c)] in volume percentage (vol%) was measured ∼42 vol% in E2209 and ∼20 vol% in E309LMo weld which was also the good agreement of ferritoscope measurement.

B. EBSD analysis

The EBSD was used to study the evolution of the microstructure (structure of a material revealed by microscopy) and micro-texture (set of crystallographic orientations) with major emphasis on phase analysis of WM and HAZ. EBSD maps were acquired on the transverse direction plane of the rolled sheets that were welded with the welding direction being normal to the plane of the paper.

Inverse pole figure (IPF) map in the Z direction (perpendicular to welding direction, i.e., the transverse plane) of different regions (schematic of the welded joint subjected to EBSD) is shown in Fig. 3(a). The various colors of the IPF map indicated crystal orientations. Figures 3(b) and 3(c) show the IPF of DSS 2205 and AISI 430 FSS BM respectively, which indicated heterogeneous structure in the BMs. Duplex 2205 BM [Fig. 3(b)] shows a dual-phase microstructure comprising of body center cubic (BCC) ferrite and face center cubic (FCC) austenite phase. ^{Reference Eghlimi, Shamanian and Raeissi26,Reference Eghlimi, Shamanian, Eskandarian, Zabolian and Szpunar28} Fig. 3(d) shows the phase map of DSS 2205 BM, both the phases showed a typical banded structure and the ferrite phase showed a high volume fraction (0.6723) of high angle grain boundary (HAGB) as it is mentioned in Table II.

FIG. 3.

Schematic, EBSD IPF map and phase map of both the weldments.

TABLE II.

Volume fraction of misorientation angle.

On the other hand, AISI 430 FSS BM illustrates [Fig. 3(c)] ferrite (BCC) matrix with intergranular martensite, ^{Reference Lippold and Kotecki22} XRD patterns [Fig. 1(f)] revealed the presence of martensite phase in AISI 430 FSS BM. The AISI 430 FSS BM was deformed, which reflected very low angle grain boundary (VLAGB) with misorientation angles of smaller than 5°, the measured fraction of VLAGB was 0.6272 (Table II).

Figure 3(f) shows the IPF map of AISI 430 FSS side HAZ, the large size of ferrite grains were observed with large variation of martensite (width and length varying from less than ∼4 µm, similar for both electrode welds). In the HAZ of ferritic BM side, significant grain growth was observed, the higher local average misorientation (LAM) in this region was also the indication of such phenomenon. The fraction of VLAGB was high, which indicated higher dislocation density within the grains. It was reported by Lippold and Kotecki ^{Reference Lippold and Kotecki22} that any austenite which forms at elevated temperature will transform to martensite upon cooling to room temperature leaving no carbon available for carbide formation. There was grain growth in the FSS HAZ due to lack of a significant second pinning phase. ^{Reference Ramkumar, Kumar, Radhakrishna, Kothari, Sridhar, Arivazhagan and Kuppan16}

Phase map of ferritic HAZ [Fig. 3(h)] indicated a distinct color to differentiate the phases present in the HAZ, ferrite and martensite phase with the help of color variation (red color = ferrite, blue color = martensite). The large ferrite grain was measured ∼0.31 mm, whereas the average HAZ width was ∼1.2 mm for both the electrode welds, which may be attributed to higher thermal conductivity of FSS. ^{Reference Lippold and Kotecki22}

On the other hand, in DSS 2205 side HAZ [Fig. 3(g)] more ferritization occurred due to the higher thermal conductivity of ferrite. Moreover, during heating, some austenite grains were formed and the presence of austenite in the ferritic matrix of DSS 2205 side HAZ controlled the grain growth and hence the width of HAZ (∼0.83 mm) was considerably smaller than AISI 430 FSS side HAZ. ^{Reference Singh12,Reference Lippold and Kotecki22} More grains with 〈111〉 orientation were apparently dominant in the HAZ. However, the presence of 〈111〉 texture along ND can be attributed to the tensile forces along the direction of heat flow that results in compressive stresses along TD.

The large volume fraction of HAGB and LAM (Table II) indicated grain growth. Figure 3(i) illustrates the phase map of DSS 2205 side HAZ.

E2209 weld [Fig. 3(j)] comprised of austenite and ferrite hybrid micro-constituents, but the morphological features and orientation were typically different than that of DSS 2205 BM. In WM region, very large fraction of HAGB (Table II) indicated pronounced nucleation and resulted in relatively small recrystallized grains because of high cooling rate. ^{Reference Verma and Taiwade3,Reference Verma, Taiwade, Khatirkar and Kumar17} It was reported ^{Reference Eghlimi, Shamanian, Eskandarian, Zabolian and Szpunar28,Reference Shamanian, Mohammadnezhad, Amini, Zabolian and Szpunar29} that the dendritic growth with 〈100〉 grains during welding have a detrimental effect on the mechanical properties and leads to brittle fracture.

Phase map [Fig. 3(l)] of E2209 weld distinguished the ferrite and austenite phase with the help of color variation (green color = austenite, red color = ferrite). However, chemistry variation synergistically affects the phase balance. ^{Reference Eghlimi, Shamanian, Eskandarian, Zabolian, Nezakat and Szpunar24,Reference Shamanian, Mohammadnezhad, Amini, Zabolian and Szpunar29}

Whereas 309LMo weld [Fig. 3(k)] showed dendritic growth of acicular ferrite (with ∼86 µm length) with 〈111〉 dominated austenite orientation, the formation of acicular ferrite was due to higher ferrite former Mo content, which increased Cr_eq/Ni_eq ratio (Table I) as mentioned above and depicted in Fig. 2(d). Phase map illustrates [Fig. 5(h)] the ferrite and austenite phases with the help of color variation (red color = austenite, black color = ferrite). Variation in the chemistry changed the solidification mode and phase balance. ^{Reference Eghlimi, Shamanian, Eskandarian, Zabolian, Nezakat and Szpunar24} In both the cases, welded region did not illustrate the presence of substructure or any detrimental precipitates. Substructure is an arrangement of dislocations to form low energy configurations, that is boundaries with <15° misorientation. ^{Reference Eghlimi, Shamanian, Eskandarian, Zabolian, Nezakat and Szpunar24,Reference Shamanian, Eghlimi, Eskandarian and Szpunar30}

C. Hardness and tensile test

Vicker’s micro-hardness test was carried out on the transverse section of the weldment (BM, HAZ and weld). Figure 4(a) shows the hardness variations of both the weldments. Higher hardness was observed in E2209 weld (258.8 HV_0.5) compared to E309LMo weld (236.1 HV_0.5) attributed to solid solution strengthening (interstitial solid solution by N and C and substitutional solid solution by Mo and Cr) and the difference in developed microstructure by different Cr_eq/Ni_eq ratio and level of dilution. ^{Reference Verma and Taiwade3,Reference Aguilar, Tabares and Serna15} The hardness of DSS 2205 BM was measured 264 HV_0.5 due to well-balanced ferrite and austenite phases, whereas 176.9 HV_0.5 was measured in AISI 430 FSS BM due to lower amount of alloying elements and morphological features. ^{Reference Verma, Taiwade, Khatirkar, Sapate and Gaikwad13,Reference Lippold and Kotecki22}

FIG. 4.

(a) Microhardness profile of the weldments, (b) schematic diagram of tensile specimen, (c) and (d) tensile fractograph of E2209 and E309LMo welds respectively, attached with tested fractured tensile specimen and (e, f) Strauss fractograph of both E2209 and E309LMo welds, attached with tested fractured tensile specimen.

The tensile test was conducted for both the welds and the dimensions are represented in a schematic diagram as it is illustrated in Fig. 4(b). The fractured micrographs of tensile tested samples are illustrated in Figs. 4(c) and 4(d), which belongs to a predominantly brittle fracture. Since both electrodes welded samples were broken from AISI 430 FSS BM, only far distance from the weld, hence it was inconclusive to discuss in detail, only the weld restraint helps to increase the tensile strength of the weld. ^{Reference Sathiya, Aravindan and Haq6}

E2209 weld showed higher tensile strength (575 MPa) than that of E309LMo weld (561 MPa). The higher restraint of E2209 weld could be attributed to higher solid solution strengthening due to higher dilution and Cr_eq/Ni_eq ratio. ^{Reference Verma and Taiwade3,Reference Sathiya, Aravindan and Haq6,Reference Verma, Taiwade, Khatirkar and Kumar17}

On the other hand, DSS 2205 BM showed 626 MPa tensile strength, while 430 FSS BM showed 476 MPa, the difference in both the BMs was due to different morphological features and alloying elements. ^{Reference Verma and Taiwade3,Reference Aguilar, Tabares and Serna15}

In addition modified Strauss test also carried out to check the susceptibility of weldments to IGC attack. Modified tensile test gives the considerable changes in mechanical properties, like strength and ductility by the intergranular attack in the Strauss test medium.

Tensile specimens were experienced in the Strauss medium (Cu–CuSO₄–16% H₂SO₄), boiling at 100 °C for 36 h and then followed by the tensile test, the fracture occurred from AISI 430 FSS side HAZ in both the welds, which indicated the intergranular attack in the HAZ. Moreover, the significant reduction in strength was evident (295 MPa) in the Strauss test. Fractograph [Figs. 4(e) and 4(f)] revealed the intergranular fracture in both the cases. From the literature, ^{Reference Joshi and Stein31} it was believed that the Cr depletion or carbide formation may cause the IGC attack in this region and more dissolution of the ferrite phase in this environment resulted in the reduction in strength.

D. Impact test

The Charpy V-notch test was conducted at room temperature to know the response of the weldments to sudden impact loads. ^{Reference Verma and Taiwade3} The dimensions were mentioned in the schematic diagram as it is shown in Fig. 5(a).

FIG. 5.

Fractograph of impact specimen of weldment (a) schematic diagram of impact specimen, (b) AISI 430 FSS BM, (c) DSS 2205 BM, (d) E2209 weld, (e) E309LMo weld, attached with fractured samples.

Experimental results indicated lower impact energy of AISI 430 FSS BM (∼6 J) due to lower alloying elements and formation of hard martensite. Fractograph [Fig. 5(b)] revealed flat facets and a very shallow tearing ridge. Whereas DSS 2205 BM showed higher impact energy (∼69 J) owing to large fibrous network attributed to proper phase balance as it is shown in Fig. 5(c). The next higher impact energy was measured in case of E2209 weld (∼48 J) as compared to E309L weld (∼24 J). The higher impact energy of E2209 weld compared to its counterpart could be attributed to a higher dilution resulted in higher Cr_eq/Ni_eq ratio and different morphological features. ^{Reference Verma, Taiwade, Khatirkar, Sapate and Gaikwad13,Reference Verma and Taiwade14} It was also reported that the presence of higher N in E2209 weld helps to increase the toughness of E2209 weld. ^{Reference Verma and Taiwade3,Reference Aguilar, Tabares and Serna15}

From the fractograph [Fig. 5(d)], it was noticed that E2209 weld showed fine and large amounts of dense, well-developed dimples, indicated adequate ductility. While E309LMo illustrated [Fig. 5(e)] mix dimples with coarse and fine, quasicleavage was also noticed in some region, the fractured sample (in E309LMo weld) showed that the fracture occurred first from the ferritic HAZ side then propagated toward the weld center.

E. Pitting corrosion

Pitting corrosion represents the localized small hole, which causes of catastrophic failure of the materials. Pitting resistance of the materials is indexed by PREN, the sudden change in chemical composition, changes the pitting behavior. ^{Reference Verma and Taiwade3,Reference Yang, Yan, Li and Wang27} This order is consistent with given formula ^{Reference Verma and Taiwade3}

The potentiodynamic polarization curve is shown in Fig. 6(a). From the curve, it was observed that the DSS 2205 BM had highest E _pit value, thereby being the most resistant to pitting corrosion among welds and ferritic BM. The balanced phase ratio of austenite and ferrite which can inhibit the preferential dissolution of the ferrite phase due to the galvanic coupling between these two phases. ^{Reference Verma and Taiwade3,Reference Eghlimi, Shamanian and Raeissi26,Reference Yang, Yan, Li and Wang27} From the micrograph [Fig. 6(b)] of DSS 2205 BM, no pit was observed.

FIG. 6.

(a) Potentiodynamic anodic polarization of weldments, (b) micrograph of 2209 DSS BM, (c) micrograph of AISI 430 FSS BM showing pits, (d) E2209 weld showing pits, (e) E309LMo weld showing pits, (f) schematic diagram of galvanic set-up, (g) galvanic current density/time graph of weld couples.

The next most resistant was E2209 weld (E _pit = 0.79 V) followed by E309LMo weld (E _pit = 0.32 V. It is well known that, Mo and Cr-enriched the ferrite phase, whereas N promoted the austenite phase. ^{Reference Verma and Taiwade3,Reference Verma, Taiwade, Khatirkar and Kumar17} However, E2209 weld had higher Cr_eq/Ni_eq ratio and higher dilution than that of E309LMo weld. Nitrogen associates in repassivation, whereas Mo suppresses initiation of pits. ^{Reference Eghlimi, Shamanian and Raeissi26,Reference Yang, Yan, Li and Wang27}

Pits were observed mostly in austenite phase in both the welds, E2209 [Fig. 6(d)] and E309LMo [Fig. 6(e)]. However, in E2209 weld more pits were evident in IGA compared to GBA and WA.

DSS 2205 BM showed highest PREN (35.27) followed by E2209 (34.54) then E309LMo (31.2), while ferritic BM showed lowest PREN (16.24). The higher PREN of DSS 2205 BM was attributed to the synergy of higher concentration of Mo, Cr, and N content and hence more immune for pitting corrosion. While AISI 430 FSS BM had the highest content of the C and the lowest content of Cr and Mo, which showed lowest E _pit (−0.29 V), therefore, chloride ions easily accumulate in the pits, changing the micro-environment. Heavy pits were observed in martensite phase than ferrite phase as it is shown in Fig. 6(c). From the observed trend, it was suggested that C was detrimental to the pitting resistance.

F. Galvanic corrosion

Figure 6(f) illustrates a schematic diagram for galvanic corrosion test. The couple was formed between WM part (excluding HAZs and BM) and BM. Since HAZ was very small and difficult to test separately, hence only WM was tested.

Figure 6(g) shows the galvanic current densities (I _G) of BM and weld couples as a function of time. It was observed that the I _G of all the couples were varied with time (12 h test) and showed very low galvanic corrosion rates only in the range of nA/cm². However, the galvanic corrosion rate of the galvanic couple depends on the potential difference and electrochemical behavior of different materials based on composition and developed microstructure. ^{Reference Tsai and Chen32} From the experimental results, it was noticed that the I _G for DSS 2205 BM-E2209 weld couple was lowest (−42.78 nA/cm²) due to the small difference in δ-ferrite between the couple. ^{Reference Tsai and Chen32,Reference Alinejad, Korojy and Ebrahimi33} While highest I _G was measured in AISI 430 FSS BM-309LMo weld couple (13.77 nA/cm²) attributed to the large δ-ferrite difference between couple. However, 5 nA/cm² I _G was measured in AISI 430 FSS BM-E2209 weld couple, whereas −7.65 nA/cm² was in case of 2205 BM-309LMo weld couple. From the results, it was suggested that the large potential difference is enough for an acceleration of the galvanic phenomenon. No pit or discoloration was noticed on the surface (Note: All the data related to pitting and galvanic corrosion can be referred in Supplementary Material Table SI).