Study on the Mechanical Performance of Dissimilar Butt Joints between Low Ni Medium-Mn and Ni-Cr Austenitic Stainless Steels Processed by Gas Tungsten Arc Welding

In the present work, dissimilar butt joints between a low-Ni, medium-Mn austenitic stainless steel, M-Mn SS, and a Ni-Cr austenitic stainless steel, Ni-Cr SS, were processed by utilizing the gas tungsten arc welding (GTAW) technique at different heat inputs. A filler metal of ER308 was employed in the welding process. The filler yields 480 MPa, which is equivalent to the yield strength of M-Mn SS. The microstructural analysis and mechanical performance (i.e., tensile strength and hardness properties) of the concerned joints were studied by using an optical microscope and uniaxial tensile tests, respectively. The results revealed that a duplex structure from austenite matrix and delta ferrite is promoted in the fusion zone (FZ) of the dissimilar joints processed with low and high energy inputs (0.486 kJ/mm and 0.558 kJ/mm). The FZ of the specimens welded at high heat input exhibited the lowest hardness value (151.2 HV) in comparison to heat affected zone (HAZ) (166.3 HV). Moreover, the joints exhibited a low tensile strength of 610 MPa. The achieved strength is significantly lower than the strengths of the base metals (BMs) M-Mn SS and Ni-Cr SS. This is mainly attributed to the inhomogeneous dendritic structure of the FZ with Cr-carbides precipitation.


Introduction
Austenitic stainless steels (ASSs) are widely used in many industrial applications such as automotive, chemical, petrochemical, marine structure, steel structure, and furniture industries [1]. ASS has excellent formability, superior mechanical properties such as strength and ductility, oxidation resistance at elevated temperatures, and high corrosion resistance [2]. The typical chemical composition of ASSs 300 series contains 18-26 wt.% Cr and 8-37 wt.% Ni, in addition to low C content ranges (0.03-0.25 wt.%) [3]. Ni is responsible for stabilizing the austenitic microstructure at room temperature; however, it increases the cost of alloy. Mn is combined with N to produce the same effect of Ni as an austenitic stabilizer with low cost [4]. Moreover, the N addition enhances the austenite phase range, it improves mechanical properties, and promotes the passivation process, resulting in improvements of corrosion resistance properties [5]. Cr-Mn stainless steels are developed under a new grade of austenitic stainless steels called the 200 series, reducing Ni content to less than 1 wt.% to meet market needs with higher quality and lower cost compared to the ASSs 300 series [4,6]. Cr alloying element is the responsible element for developing corrosion resistance of ASS [7]. Therefore, Cr-Mn SS (200 series) naturally possesses lower corrosion resistance compared to Cr-Ni SS (300 series) due to the low content of Cr and Ni (low heat input) since the cooling rate was relatively high, which provided an amount of short time for dendrites' formation and growth in the FZ. Taiwade et al. [29] reported the existence of carbide precipitation, and grain coarsening increased with an increase in passes relative to the number of similar welding of 304 SS and Cr-Mn SS. It was found that the tensile strength decreased with an increase in passes number.
Dissimilar welding of medium-Mn stainless steel with another steel is scarcely investigated. However, Chuaiphan and Srijaroenpramong [20] studied the mechanical properties and corrosion behavior of dissimilar joints 204Cu and 304 by GTA with one heat input. The purpose of the present paper is to comprehensively investigate the dissimilar weldability of M-Mn SS with Ni-Cr SS by applying the welding technique, GTAW, at different heat inputs. The microstructure evolution, mechanical tensile strength, and hardness properties of the weldments were depicted in order to determine the welding efficacy.

Materials and Methods
Low-Ni, medium-Mn austenitic stainless steel, M-Mn SS, and Ni-Cr austenitic stainless steel, Ni-Cr SS, were used for dissimilar butt joints. The base metals were received in the form of 2 mm thick sheets. Filler metal of ER308L was used for the welding experiments. Table 1 shows the chemical compositions of the studied steels and filler metals. The chemical analysis was carried out by utilizing optical emission spectroscopy (FOUNDRY-MASTER Pro, OXFORD INSTRUMENTS, High Wycombe, UK). The edges of the cut specimens were cleaned by a shaper machine via mechanical removal of 0.5 mm from the surfaces of specimens that will be welded. In addition, the surfaces of the specimens were chemically cleaned with ethanol to avoid any greases, oil, or dust on the edges.  Figure 1 shows a schematic illustration of the butt joints with a root gap of 1 mm. The welding process was carried out using the GTAW technique (ESAB Tig 4300i AC/DC) for the root pass and the cup pass. A pure argon shielding gas of purity 99.8% was used for all the welding experiments with 15 L/min. Argon gas was used to protect the root side from oxidation and make welds with high quality at the same flow rate of 15 L/min. No preheat or post-weld heat treatment was applied to the specimens The welding current is considered as the most influential parameter, which affects the melting rate of the filler metals as well as the mechanical properties of base metals at the HAZ. In addition, it affects the speed and heat input of the welding process. Hence, welding was performed at the current range varying between 50 to 110 A. The heat input was calculated according to Equation (1) as follows: Heat input (kJ/mm) = η × I × V S × 1000 (1) where η is the arc efficiency for the GTAW process and η equals 0.6 according to BS EN 1011-1 [30]. I is the welding current in ampere (A), V is the arc voltage in volt (V), and S is the welding speed in mm/s. The welding current is considered as the most influential parameter, which affects the melting rate of the filler metals as well as the mechanical properties of base metals at the HAZ. In addition, it affects the speed and heat input of the welding process. Hence, welding was performed at the current range varying between 50 to 110 A. The heat input was calculated according to Equation (1) as follows: Heat input (kJ/ mm) = η × I ×V S ×1000 (1) where η is the arc efficiency for the GTAW process and η equals 0.6 according to BS EN 1011-1 [30]. I is the welding current in ampere (A), V is the arc voltage in volt (V), and S is the welding speed in mm/s. Table 2 represents the welding parameter of the GTAW process at two levels of different heat inputs, namely, low and high heat inputs. After the completion of the joining processes, the welded metals were examined by visual methods and radiographic test (RT) to inspect the weld quality of the welded specimens. Figure 1 shows preparations of specimens in perpendicular sections relative to the weld direction to observe microstructure changes after welding processes under low and high heat inputs. The specimens were cut from the welding samples using an electric discharge machine (EDM) (Chmer, Taichung, Taiwan).
The dissimilar butt joints of M-Mn/Ni-Cr SS were transversely cross sectioned and mechanically grounded and polished to prepare metallographic specimens. The specimens were etched via an electrolytic oxalic acid etchant according to the ASTM E407-07 (10 g of oxalic acid and 100 mL distilled water at 6 V for 1 min) [31]. The microstructures of the weldments were investigated by using an optical microscope (Olympus PMG 3, Waltham, MA, USA), which coupled with an image analyzing software. Moreover, scanning electron microscopy (SEM) (QUANTA FEG 250) equipped with energy dispersive spectroscopy (EDS) (FEI, Hillsboro, OR, USA) was used to determine the chemical composition of the weld metals and provide additional high-resolution images of the weld  Table 2 represents the welding parameter of the GTAW process at two levels of different heat inputs, namely, low and high heat inputs. After the completion of the joining processes, the welded metals were examined by visual methods and radiographic test (RT) to inspect the weld quality of the welded specimens. Figure 1 shows preparations of specimens in perpendicular sections relative to the weld direction to observe microstructure changes after welding processes under low and high heat inputs. The specimens were cut from the welding samples using an electric discharge machine (EDM) (Chmer, Taichung, Taiwan).
The dissimilar butt joints of M-Mn/Ni-Cr SS were transversely cross sectioned and mechanically grounded and polished to prepare metallographic specimens. The specimens were etched via an electrolytic oxalic acid etchant according to the ASTM E407-07 (10 g of oxalic acid and 100 mL distilled water at 6 V for 1 min) [31]. The microstructures of the weldments were investigated by using an optical microscope (Olympus PMG 3, Waltham, MA, USA), which coupled with an image analyzing software. Moreover, scanning electron microscopy (SEM) (QUANTA FEG 250) equipped with energy dispersive spectroscopy (EDS) (FEI, Hillsboro, OR, USA) was used to determine the chemical composition of the weld metals and provide additional high-resolution images of the weld metal. In addition, SEM was used to provide the details of the fractured surfaces after tensile testing in order to examine the mode of fracture. The hardness test was performed according to the Vickers hardness (HV) scale (Matsuzawa) at room temperature. The hardness tester is equipped with a diamond indenter under a load of 10 kg for a penetration holding time of 15 s.
Tensile specimens were cut parallel to the rolling direction of the sheets and perpendicular to the welding pool direction. Three specimens per heat input were machined out from the weld pads by EDM. Dog-bone-shaped specimens were prepared according to the ASTM E8/E8M standard with a gauge length of 50 mm [32]. Uniaxial tensile testing was performed by using a hydraulic universal testing machine (UH-F1000kNI SHIMADZU, Tokyo, Japan) at a quasi-static strain rate of 10 −3 s −1 at room temperature.

Results and Discussion
3.1. Metallographic Characterization 3.1.1. Macrostructure Investigation Figure 2 shows the radiographic films after dissimilar welding of M-Mn SS and Ni-Cr SS at low and high heat inputs. Clearly, the weldments are sound without weld defects, such as porosity, undercuts, or blowholes, with full penetration. This shows that the welding process parameters applied during the joining of the paired steels are appropriate.
tion holding time of 15 s.
Tensile specimens were cut parallel to the rolling direction of the sheets and perpendicular to the welding pool direction. Three specimens per heat input were machined out from the weld pads by EDM. Dog-bone-shaped specimens were prepared according to the ASTM E8/E8M standard with a gauge length of 50 mm [32]. Uniaxial tensile testing was performed by using a hydraulic universal testing machine (UH-F1000kNI SHI-MADZU, Tokyo, Japan) at a quasi-static strain rate of 10 −3 s −1 at room temperature. Figure 2 shows the radiographic films after dissimilar welding of M-Mn SS and Ni-Cr SS at low and high heat inputs. Clearly, the weldments are sound without weld defects, such as porosity, undercuts, or blowholes, with full penetration. This shows that the welding process parameters applied during the joining of the paired steels are appropriate.  Table 3 illustrates the width of the face and root weld metals of the joints. It is observed that the widths of the face and root weld metals increase with the heat input. Since by increasing the heat input, the produced molten pool size is higher. Consequently, the FZ dimensions increase.  Figure 3 shows the microstructures of the base metals, M-Mn SS, and Ni-Cr SS, which were observed via optical microscopy. It is observed that the microstructures of the two alloys consist of a fully austenitic structure with annealing twins.  Table 3 illustrates the width of the face and root weld metals of the joints. It is observed that the widths of the face and root weld metals increase with the heat input. Since by increasing the heat input, the produced molten pool size is higher. Consequently, the FZ dimensions increase.   The microstructural characteristics of the FZ welded at low and high heat inputs are shown in Figure 4. It is observed that the FZs of both heat inputs consist of austenitic matrix and delta ferrite phase. Moreover, the volume fraction of delta ferrite was measured by a ferritescope. The recorded ferrite contents of the weldments are 7.8% and 6.2% at low and high heat inputs, respectively. Therefore, we can conclude that the amount of the delta ferrite increases with the decrease in heat input. The increase in the delta ferrite content at low heat input is attributed to the higher cooling rate, which restricted ferrite transformation to austenite. This is in agreement with the work of Kumar and Shahi [27].  Table 3 shows the values of the dendrite length and interdendritic spacing at both heat inputs. Clearly, the dendrite length and interdendritic spacing increased from 109.2 ± 5 µm and 17.1 ± 3 µm at the low heat input to 171.9 ± 6 µm and 28.6 ± 4 µm at the high heat input, respectively. It seems that the high heat input does allow the dendritic structure to enlarge in the weld zone. Figures 5 and 6 show the optical micrographs of dissimilar welding of M-Mn SS and Ni-Cr SS at low and high heat inputs, respectively. Clearly the optical micrographs of the two heat inputs consist of four regions: FZ, partially melted zone (PMZ), HAZ, and BM, The microstructural characteristics of the FZ welded at low and high heat inputs are shown in Figure 4. It is observed that the FZs of both heat inputs consist of austenitic matrix and delta ferrite phase. Moreover, the volume fraction of delta ferrite was measured by a ferritescope. The recorded ferrite contents of the weldments are 7.8% and 6.2% at low and high heat inputs, respectively. Therefore, we can conclude that the amount of the delta ferrite increases with the decrease in heat input. The increase in the delta ferrite content at low heat input is attributed to the higher cooling rate, which restricted ferrite transformation to austenite. This is in agreement with the work of Kumar and Shahi [27].  The microstructural characteristics of the FZ welded at low and high heat inputs are shown in Figure 4. It is observed that the FZs of both heat inputs consist of austenitic matrix and delta ferrite phase. Moreover, the volume fraction of delta ferrite was measured by a ferritescope. The recorded ferrite contents of the weldments are 7.8% and 6.2% at low and high heat inputs, respectively. Therefore, we can conclude that the amount of the delta ferrite increases with the decrease in heat input. The increase in the delta ferrite content at low heat input is attributed to the higher cooling rate, which restricted ferrite transformation to austenite. This is in agreement with the work of Kumar and Shahi [27].  Table 3 shows the values of the dendrite length and interdendritic spacing at both heat inputs. Clearly, the dendrite length and interdendritic spacing increased from 109.2 ± 5 µm and 17.1 ± 3 µm at the low heat input to 171.9 ± 6 µm and 28.6 ± 4 µm at the high heat input, respectively. It seems that the high heat input does allow the dendritic structure to enlarge in the weld zone. Figures 5 and 6 show the optical micrographs of dissimilar welding of M-Mn SS and Ni-Cr SS at low and high heat inputs, respectively. Clearly the optical micrographs of the two heat inputs consist of four regions: FZ, partially melted zone (PMZ), HAZ, and BM,  Table 3 shows the values of the dendrite length and interdendritic spacing at both heat inputs. Clearly, the dendrite length and interdendritic spacing increased from 109.2 ± 5 µm and 17.1 ± 3 µm at the low heat input to 171.9 ± 6 µm and 28.6 ± 4 µm at the high heat input, respectively. It seems that the high heat input does allow the dendritic structure to enlarge in the weld zone. Figures 5 and 6 show the optical micrographs of dissimilar welding of M-Mn SS and Ni-Cr SS at low and high heat inputs, respectively. Clearly the optical micrographs of the two heat inputs consist of four regions: FZ, partially melted zone (PMZ), HAZ, and BM, as depicted in Figures 5 and 6. The PMZ appears similar to a layer after the FZ, where the BM has been melted and solidified without mixing with filler metal. The HAZ was significantly influenced by the value of the heat input during the welding process. On the one hand, under low heat input conditions, the width of PMZ and HAZ approximately equals 388 µm and 313 µm for the M-Mn SS side and the Ni-Cr SS side, respectively. On the other hand, under high heat input, the width of PMZ and HAZ approximately equals 432 µm and 373 µm for the M-Mn SS side and the Ni-Cr SS side, respectively. as depicted in Figures 5 and 6. The PMZ appears similar to a layer after the FZ, where the BM has been melted and solidified without mixing with filler metal. The HAZ was significantly influenced by the value of the heat input during the welding process. On the one hand, under low heat input conditions, the width of PMZ and HAZ approximately equals 388 µm and 313 µm for the M-Mn SS side and the Ni-Cr SS side, respectively. On the other hand, under high heat input, the width of PMZ and HAZ approximately equals 432 µm and 373 µm for the M-Mn SS side and the Ni-Cr SS side, respectively.  It is observed that as the heat input increases, the width of PMZ and HAZ increases at the two sides of the weld metal. This is attributed to the low cooling rate associated with the high heat input, allowing sufficient time for grain growth, and hence the size of the PMZ and HAZ was increased. Moreover, the widths of PMZ and HAZ of M-Mn SS side are higher than those of Ni-Cr SS side. This is attributed to the effect of volumetric heat capacities of M-Mn SS lower than Ni-Cr SS and the higher susceptibility of M-Mn SS towards inter granular precipitations, which agrees with the literature [20,24]. Figure 7 shows Cr carbides formation at the grain boundaries of the microstructure of the HAZ at both sides of the weld metal and the FZ adjacent to the weld interface of M-Mn SS and Ni-Cr SS at the high heat input. It is observed that the Cr carbides formation as depicted in Figures 5 and 6. The PMZ appears similar to a layer after the FZ, where the BM has been melted and solidified without mixing with filler metal. The HAZ was significantly influenced by the value of the heat input during the welding process. On the one hand, under low heat input conditions, the width of PMZ and HAZ approximately equals 388 µm and 313 µm for the M-Mn SS side and the Ni-Cr SS side, respectively. On the other hand, under high heat input, the width of PMZ and HAZ approximately equals 432 µm and 373 µm for the M-Mn SS side and the Ni-Cr SS side, respectively.  It is observed that as the heat input increases, the width of PMZ and HAZ increases at the two sides of the weld metal. This is attributed to the low cooling rate associated with the high heat input, allowing sufficient time for grain growth, and hence the size of the PMZ and HAZ was increased. Moreover, the widths of PMZ and HAZ of M-Mn SS side are higher than those of Ni-Cr SS side. This is attributed to the effect of volumetric heat capacities of M-Mn SS lower than Ni-Cr SS and the higher susceptibility of M-Mn SS towards inter granular precipitations, which agrees with the literature [20,24]. Figure 7 shows Cr carbides formation at the grain boundaries of the microstructure of the HAZ at both sides of the weld metal and the FZ adjacent to the weld interface of M-Mn SS and Ni-Cr SS at the high heat input. It is observed that the Cr carbides formation It is observed that as the heat input increases, the width of PMZ and HAZ increases at the two sides of the weld metal. This is attributed to the low cooling rate associated with the high heat input, allowing sufficient time for grain growth, and hence the size of the PMZ and HAZ was increased. Moreover, the widths of PMZ and HAZ of M-Mn SS side are higher than those of Ni-Cr SS side. This is attributed to the effect of volumetric heat capacities of M-Mn SS lower than Ni-Cr SS and the higher susceptibility of M-Mn SS towards inter granular precipitations, which agrees with the literature [20,24]. Figure 7 shows Cr carbides formation at the grain boundaries of the microstructure of the HAZ at both sides of the weld metal and the FZ adjacent to the weld interface of M-Mn SS and Ni-Cr SS at the high heat input. It is observed that the Cr carbides formation increased at the HAZ of M-Mn side and the FZ adjacent to the weld interface of M-Mn compared to HAZ of Ni-Cr side and the FZ adjacent to the weld interface of Ni-Cr.

Macrostructure Investigation
Moreover, the dark particles have a high content of Cr, while the white regions have a low content of Cr as examined by the EDS affiliated with the SEM. This is attributed to the diffusion of C atoms from the base metals to the weld zone forming Cr carbides adjacent to the weld interface of both sides, which was concluded by the research conducted by Vashishtha et al. [33], Chuaiphan et al. [28,34], and Kumar et al. [27]. increased at the HAZ of M-Mn side and the FZ adjacent to the weld interface of M-Mn compared to HAZ of Ni-Cr side and the FZ adjacent to the weld interface of Ni-Cr. Moreover, the dark particles have a high content of Cr, while the white regions have a low content of Cr as examined by the EDS affiliated with the SEM. This is attributed to the diffusion of C atoms from the base metals to the weld zone forming Cr carbides adjacent to the weld interface of both sides, which was concluded by the research conducted by Vashishtha et al. [33], Chuaiphan et al. [28,34], and Kumar et al. [27].

Solidification Mode and Delta-Ferrite Content
In the present work, we employed the WRC-1992 Constitution Diagram for Stainless Steel Weld Metals to estimate the predicted ferrite content related to the values of the chromium equivalent (Cr eq ) and nickel equivalent (Ni eq ) of the weld zone [35,36]. The chemical compositions of the FZs were determined by EDS to calculate Cr eq and Ni eq according to the Equations (2) and (3), as follows.
Cr eq = Cr + Mo + 0.5 Nb + 1.5 Si (2) Ni eq = Ni + 30 (C + N) + 0.5 (Mn + Cu) However, EDS is not accurate for low atomic number elements, such as C and N, since the sensitivity of EDS analysis is approximately 0.1 wt.% for all elements, i.e., higher C content of the paired steels and the filler metal [37]. The C contents of the FZs were calculated by stoichiometry method, which estimates the quantitative relationship between the C contents of the paired steel sheets (M-Mn SS and Ni-Cr SS) and the filler metal (ER308L) and the dilution percent of each steel in the joint according to the Equation (4): where C FZ is the C content of the FZ, (C.D) Mn SS denotes the C content and dilution percent of M-Mn SS, (C.D) NiCr SS denotes C content and dilution percent of Ni-Cr SS, and (C.D) FM denotes C content and dilution percent of the filler metal. Based on the measured delta ferrite content, the dilution precents of the paired steels were taken as D = 15% and D = 70% for the filler metal. Consequently, the C content of the FZ is 0.032 wt.% at the low heat input.
Similarly, the C content of the FZ was calculated to be 0.037 wt.% at the high heat input. Table 4 shows the EDS elemental analysis of the FZs at low and high heat input. The corresponding Cr eq , Ni eq , and the predicted content of the delta ferrite are shown in Table 4. For comparison, the actual delta ferrite was measured by a ferrite scope. It is now well established that there are different solidification modes of ASSs during welding based on the ratio of Cr eq / Ni eq , as shown below [38]

Solidification Mode and Delta-Ferrite Content
In the present work, we employed the WRC-1992 Constitution Diagram for Stainless Steel Weld Metals to estimate the predicted ferrite content related to the values of the chromium equivalent (Creq) and nickel equivalent (Nieq) of the weld zone [35,36]. The chemical compositions of the FZs were determined by EDS to calculate Creq and Nieq according to the Equations (2) and (3), as follows.
However, EDS is not accurate for low atomic number elements, such as C and N, since the sensitivity of EDS analysis is approximately 0.1 wt.% for all elements, i.e., higher C content of the paired steels and the filler metal [37]. The C contents of the FZs were calculated by stoichiometry method, which estimates the quantitative relationship between the C contents of the paired steel sheets (M-Mn SS and Ni-Cr SS) and the filler metal (ER308L) and the dilution percent of each steel in the joint according to the Equation (4): where C is the C content of the FZ, C. D C. D denotes C content and dilution percent of the filler metal. Based on the measured delta ferrite content, the dilution precents of the paired steels were taken as D = 15% and D = 70 % for the filler metal. Consequently, the C content of the FZ is 0.032 wt.% at the low heat input. Similarly, the C content of the FZ was calculated to be 0.037 wt.% at the high heat input. Table 4 shows the EDS elemental analysis of the FZs at low and high heat input. The corresponding Creq, Nieq, and the predicted content of the delta ferrite are shown in Table  4. For comparison, the actual delta ferrite was measured by a ferrite scope. It is now well established that there are different solidification modes of ASSs during welding based on the ratio of Creq/ Nieq, as shown below [38]  In the present instance, the Creq/Nieq values seem to result in inducing delta ferrite within the austenitic matrix, as indicated by locating the Creq/Nieq values of the two heat inputs on the WRC diagram in Figure 8. It is observed that the predicted and the measured contents of the delta ferrite content are comparative since that the predicted delta ferrite

Solidification Mode and Delta-Ferrite Content
In the present work, we employed the WRC-1992 Constitution Diagram for Stain Steel Weld Metals to estimate the predicted ferrite content related to the values of chromium equivalent (Creq) and nickel equivalent (Nieq) of the weld zone [35,36]. T chemical compositions of the FZs were determined by EDS to calculate Creq and Nieq cording to the Equations (2) and (3), as follows.
However, EDS is not accurate for low atomic number elements, such as C and since the sensitivity of EDS analysis is approximately 0.1 wt.% for all elements, i.e., hig C content of the paired steels and the filler metal [37]. The C contents of the FZs w calculated by stoichiometry method, which estimates the quantitative relationship tween the C contents of the paired steel sheets denotes C content and dilution percent of the filler metal. Based on the me ured delta ferrite content, the dilution precents of the paired steels were taken as D = 1 and D = 70 % for the filler metal. Consequently, the C content of the FZ is 0.032 wt.% the low heat input. Similarly, the C content of the FZ was calculated to be 0.037 wt.% the high heat input. Table 4 shows the EDS elemental analysis of the FZs at low and high heat input. T corresponding Creq, Nieq, and the predicted content of the delta ferrite are shown in Ta 4. For comparison, the actual delta ferrite was measured by a ferrite scope. It is now w established that there are different solidification modes of ASSs during welding based the ratio of Creq/ Nieq, as shown below [38]  In the present instance, the Creq/Nieq values seem to result in inducing delta fer within the austenitic matrix, as indicated by locating the Creq/Nieq values of the two h inputs on the WRC diagram in Figure 8. It is observed that the predicted and the measu contents of the delta ferrite content are comparative since that the predicted delta fer γ, (Cr eq /Ni eq ) < 1.25; 2.

Solidification Mode and Delta-Ferrite Content
In the present work, we employed the WRC-1992 Constitution Diagram for Stainles Steel Weld Metals to estimate the predicted ferrite content related to the values of th chromium equivalent (Creq) and nickel equivalent (Nieq) of the weld zone [35,36]. Th chemical compositions of the FZs were determined by EDS to calculate Creq and Nieq ac cording to the Equations (2) and (3), as follows.

Creq
= Cr + Mo + 0.5 Nb + 1.5 Si Nieq = Ni + 30 (C + N) + 0.5 (Mn + Cu) However, EDS is not accurate for low atomic number elements, such as C and N since the sensitivity of EDS analysis is approximately 0.1 wt.% for all elements, i.e., highe C content of the paired steels and the filler metal [37]. The C contents of the FZs wer calculated by stoichiometry method, which estimates the quantitative relationship be tween the C contents of the paired steel sheets (M-Mn SS and Ni-Cr SS) and the filler meta (ER308L) and the dilution percent of each steel in the joint according to the Equation (4): C. D denotes C content and dilution percent of the filler metal. Based on the meas ured delta ferrite content, the dilution precents of the paired steels were taken as D = 15% and D = 70 % for the filler metal. Consequently, the C content of the FZ is 0.032 wt.% a the low heat input. Similarly, the C content of the FZ was calculated to be 0.037 wt.% a the high heat input.  In the present instance, the Creq/Nieq values seem to result in inducing delta ferrit within the austenitic matrix, as indicated by locating the Creq/Nieq values of the two hea inputs on the WRC diagram in Figure 8. It is observed that the predicted and the measured contents of the delta ferrite content are comparative since that the predicted delta ferrit

Solidification Mode and Delta-Ferrite Content
In the present work, we employed the WRC-1992 Constitution Diagram Steel Weld Metals to estimate the predicted ferrite content related to the v chromium equivalent (Creq) and nickel equivalent (Nieq) of the weld zone chemical compositions of the FZs were determined by EDS to calculate Creq cording to the Equations (2) and (3) denotes C content and dilution percent of the filler metal. Based o ured delta ferrite content, the dilution precents of the paired steels were taken and D = 70 % for the filler metal. Consequently, the C content of the FZ is 0 the low heat input. Similarly, the C content of the FZ was calculated to be 0 the high heat input.  In the present instance, the Creq/Nieq values seem to result in inducing within the austenitic matrix, as indicated by locating the Creq/Nieq values of inputs on the WRC diagram in Figure 8. It is observed that the predicted and th contents of the delta ferrite content are comparative since that the predicted (L + γ + δ)

Solidification Mode and Delta-Ferrite Content
In the present work, we employed the WRC-1992 Constit Steel Weld Metals to estimate the predicted ferrite content r chromium equivalent (Creq) and nickel equivalent (Nieq) of chemical compositions of the FZs were determined by EDS to cording to the Equations (2) and (3)  In the present instance, the Creq/Nieq values seem to resu within the austenitic matrix, as indicated by locating the Creq inputs on the WRC diagram in Figure 8. It is observed that the p contents of the delta ferrite content are comparative since tha (γ + δ), 1.25 < (Cr eq /Ni eq ) < 1.48; 3.

Solidification Mode and Delta-Ferrite Content
In the present work, we employed the WRC-1992 Constitution Diagram for Stainless Steel Weld Metals to estimate the predicted ferrite content related to the values of the chromium equivalent (Creq) and nickel equivalent (Nieq) of the weld zone [35,36]. The chemical compositions of the FZs were determined by EDS to calculate Creq and Nieq according to the Equations (2) and (3), as follows.

Creq
= Cr + Mo + 0.5 Nb + 1.5 Si (2) Nieq = Ni + 30 (C + N) + 0.5 (Mn + Cu) However, EDS is not accurate for low atomic number elements, such as C and N, since the sensitivity of EDS analysis is approximately 0.1 wt.% for all elements, i.e., higher C content of the paired steels and the filler metal [37]. The C contents of the FZs were calculated by stoichiometry method, which estimates the quantitative relationship between the C contents of the paired steel sheets (M-Mn SS and Ni-Cr SS) and the filler metal (ER308L) and the dilution percent of each steel in the joint according to the Equation (4): where C is the C content of the FZ, C. D denotes C content and dilution percent of the filler metal. Based on the measured delta ferrite content, the dilution precents of the paired steels were taken as D = 15% and D = 70 % for the filler metal. Consequently, the C content of the FZ is 0.032 wt.% at the low heat input. Similarly, the C content of the FZ was calculated to be 0.037 wt.% at the high heat input. Table 4 shows the EDS elemental analysis of the FZs at low and high heat input. The corresponding Creq, Nieq, and the predicted content of the delta ferrite are shown in Table  4. For comparison, the actual delta ferrite was measured by a ferrite scope. It is now well established that there are different solidification modes of ASSs during welding based on the ratio of Creq/ Nieq, as shown below [38]  In the present instance, the Creq/Nieq values seem to result in inducing delta ferrite within the austenitic matrix, as indicated by locating the Creq/Nieq values of the two heat inputs on the WRC diagram in Figure 8. It is observed that the predicted and the measured contents of the delta ferrite content are comparative since that the predicted delta ferrite However, EDS is not accurate for low atomic number elements, such a since the sensitivity of EDS analysis is approximately 0.1 wt.% for all elements C content of the paired steels and the filler metal [37]. The C contents of th calculated by stoichiometry method, which estimates the quantitative relat tween the C contents of the paired steel sheets (M-Mn SS and Ni-Cr SS) and the (ER308L) and the dilution percent of each steel in the joint according to the Eq  In the present instance, the Creq/Nieq values seem to result in inducing d within the austenitic matrix, as indicated by locating the Creq/Nieq values of th inputs on the WRC diagram in Figure 8. It is observed that the predicted and th contents of the delta ferrite content are comparative since that the predicted d

Solidification Mode and Delta-Ferrite Content
In the present work, we employed the WRC-1992 Constitu Steel Weld Metals to estimate the predicted ferrite content re chromium equivalent (Creq) and nickel equivalent (Nieq) of th chemical compositions of the FZs were determined by EDS to cording to the Equations (2) and (3) denotes C content and dilution percent of the filler ured delta ferrite content, the dilution precents of the paired ste and D = 70 % for the filler metal. Consequently, the C content the low heat input. Similarly, the C content of the FZ was calc the high heat input. Table 4 shows the EDS elemental analysis of the FZs at low corresponding Creq, Nieq, and the predicted content of the delta 4. For comparison, the actual delta ferrite was measured by a f established that there are different solidification modes of ASSs the ratio of Creq/ Nieq, as shown below [38]  In the present instance, the Creq/Nieq values seem to resu within the austenitic matrix, as indicated by locating the Creq/N inputs on the WRC diagram in Figure 8. It is observed that the p contents of the delta ferrite content are comparative since that

Solidification Mode and Delta-Ferrite Content
In the present work, we employed the WRC-1992 Constitution Diagram for Stainless Steel Weld Metals to estimate the predicted ferrite content related to the values of the chromium equivalent (Creq) and nickel equivalent (Nieq) of the weld zone [35,36]. The chemical compositions of the FZs were determined by EDS to calculate Creq and Nieq according to the Equations (2) and (3), as follows.

Creq
= Cr + Mo + 0.5 Nb + 1.5 Si (2) Nieq = Ni + 30 (C + N) + 0.5 (Mn + Cu) However, EDS is not accurate for low atomic number elements, such as C and N, since the sensitivity of EDS analysis is approximately 0.1 wt.% for all elements, i.e., higher C content of the paired steels and the filler metal [37]. The C contents of the FZs were calculated by stoichiometry method, which estimates the quantitative relationship between the C contents of the paired steel sheets (M-Mn SS and Ni-Cr SS) and the filler metal (ER308L) and the dilution percent of each steel in the joint according to the Equation (4): where C is the C content of the FZ, C. D denotes the C content and dilution percent of M-Mn SS, C. D denotes C content and dilution percent of Ni-Cr SS, and C. D denotes C content and dilution percent of the filler metal. Based on the measured delta ferrite content, the dilution precents of the paired steels were taken as D = 15% and D = 70 % for the filler metal. Consequently, the C content of the FZ is 0.032 wt.% at the low heat input. Similarly, the C content of the FZ was calculated to be 0.037 wt.% at the high heat input. Table 4 shows the EDS elemental analysis of the FZs at low and high heat input. The corresponding Creq, Nieq, and the predicted content of the delta ferrite are shown in Table  4. For comparison, the actual delta ferrite was measured by a ferrite scope. It is now well established that there are different solidification modes of ASSs during welding based on the ratio of Creq/ Nieq, as shown below [38]  In the present instance, the Creq/Nieq values seem to result in inducing delta ferrite within the austenitic matrix, as indicated by locating the Creq/Nieq values of the two heat inputs on the WRC diagram in Figure 8. It is observed that the predicted and the measured contents of the delta ferrite content are comparative since that the predicted delta ferrite

Solidification Mode and Delta-Ferrite Content
In the present work, we employed the WRC-1992 Constitution Diagram for Stainless Steel Weld Metals to estimate the predicted ferrite content related to the values of the chromium equivalent (Creq) and nickel equivalent (Nieq) of the weld zone [35,36]. The chemical compositions of the FZs were determined by EDS to calculate Creq and Nieq according to the Equations (2) and (3), as follows.

Creq
= Cr + Mo + 0.5 Nb + 1.5 Si (2) Nieq = Ni + 30 (C + N) + 0.5 (Mn + Cu) However, EDS is not accurate for low atomic number elements, such as C and N, since the sensitivity of EDS analysis is approximately 0.1 wt.% for all elements, i.e., higher C content of the paired steels and the filler metal [37]. The C contents of the FZs were calculated by stoichiometry method, which estimates the quantitative relationship between the C contents of the paired steel sheets (M-Mn SS and Ni-Cr SS) and the filler metal (ER308L) and the dilution percent of each steel in the joint according to the Equation (4): where C is the C content of the FZ, C. D denotes C content and dilution percent of the filler metal. Based on the measured delta ferrite content, the dilution precents of the paired steels were taken as D = 15% and D = 70 % for the filler metal. Consequently, the C content of the FZ is 0.032 wt.% at the low heat input. Similarly, the C content of the FZ was calculated to be 0.037 wt.% at the high heat input. Table 4 shows the EDS elemental analysis of the FZs at low and high heat input. The corresponding Creq, Nieq, and the predicted content of the delta ferrite are shown in Table  4. For comparison, the actual delta ferrite was measured by a ferrite scope. It is now well established that there are different solidification modes of ASSs during welding based on the ratio of Creq/ Nieq, as shown below [38]  In the present instance, the Creq/Nieq values seem to result in inducing delta ferrite within the austenitic matrix, as indicated by locating the Creq/Nieq values of the two heat inputs on the WRC diagram in Figure 8. It is observed that the predicted and the measured contents of the delta ferrite content are comparative since that the predicted delta ferrite However, EDS is not accurate for low atomic number elements, such as C an since the sensitivity of EDS analysis is approximately 0.1 wt.% for all elements, i.e., h C content of the paired steels and the filler metal [37]. The C contents of the FZs calculated by stoichiometry method, which estimates the quantitative relationshi tween the C contents of the paired steel sheets (M-Mn SS and Ni-Cr SS) and the filler m (ER308L) and the dilution percent of each steel in the joint according to the Equation denotes C content and dilution percent of the filler metal. Based on the m ured delta ferrite content, the dilution precents of the paired steels were taken as D = and D = 70 % for the filler metal. Consequently, the C content of the FZ is 0.032 wt the low heat input. Similarly, the C content of the FZ was calculated to be 0.037 wt the high heat input. Table 4 shows the EDS elemental analysis of the FZs at low and high heat input corresponding Creq, Nieq, and the predicted content of the delta ferrite are shown in 4. For comparison, the actual delta ferrite was measured by a ferrite scope. It is now established that there are different solidification modes of ASSs during welding base the ratio of Creq/ Nieq, as shown below [38]  In the present instance, the Creq/Nieq values seem to result in inducing delta f within the austenitic matrix, as indicated by locating the Creq/Nieq values of the two inputs on the WRC diagram in Figure 8. It is observed that the predicted and the meas contents of the delta ferrite content are comparative since that the predicted delta f (γ + δ), (Cr eq /Ni eq ) >1.95. In the present instance, the Cr eq /Ni eq values seem to result in inducing delta ferrite within the austenitic matrix, as indicated by locating the Cr eq /Ni eq values of the two heat inputs on the WRC diagram in Figure 8. It is observed that the predicted and the measured contents of the delta ferrite content are comparative since that the predicted delta ferrite content by the WRC diagram are approximately equal 8.5% and 7% at low and high heat inputs, respectively. As mentioned previously, the ferritescope measurements showed that the contents of delta ferrite phase in the FZ are 7.8% and 6.2% at low and high heat input, respectively.      Clearly, the welded metals have lower hardness values than the base metals. Moreover, the welded metal under high heat input has lower hardness values than welded metal under low heat input. These decreases in the hardness values are attributed to the preformed delta ferrite present in the weld metal and/or the effect of chromium precipitation formed during the solidification stage in this zone, which agrees with the literature [39].

Hardness Results
The PMZ exhibited higher hardness records than the weld metal and HAZ at low and high heat inputs due to the fact of partially unmelted grains at the fusion boundary zone as well as the effect of chromium precipitation formed during the solidification stage in the FZ. The hardness values of HAZs for both sides at low and high heat inputs near the PMZ are lower than the HAZs near the base metals due to the slower cooling rate and grain coarsening presence in the adjacent HAZ to the PMZ. Kumar et al. [27] and Bansod et al. [39] reported similar trends for microhardness values during the investigation of GTAW welded 304 SS and SMAW welded low nickel SS to 304 SS, respectively, at different heat inputs. Figure 10 displays the tensile behavior of the base metals and dissimilar welding of M-Mn SS and Ni-Cr SS at low and high heat inputs. It is observed that the ultimate tensile strength increases with the decrease in heat input for dissimilar welding. Moreover, the elongation percentage increases with the decrease in heat input. These results can be attributed to the high content of delta ferrite in the weld metal at the low heat input. In addition, the small dendrite sizes and interdendritic spacing in the weld metal enhance the tensile properties [13,27]. The joint efficiency was calculated according to Equation (5) [40,41] and presented in Table 5. Furthermore, the fracture of welded tensile specimens happened at the weld metal because the filler metal has an ultimate tensile strength lower than the base metals, as shown in Figure 11. Joint efficiency = Ultimate tensile strength of the weld metal Ultimate tensile strength of the soft base metal × 100 (5) Table 5 listed the tensile mechanical properties of the dissimilar joints comparing with those of the BMs. It is observed that the dissimilar joints exhibit lower strengths with a joint efficiency of~80% since the quasi-static mechanical properties of the joints are characterized by a low ductility of~20% and low strengths of~(580-610 MPa). The dissimilar joint welded with the low heat input exhibited higher tensile strength. This could be explained by the fact that by increasing the heat input, the cooling rate decreases. As a result, a softer grain structure is formed. Consequently, the mechanical strength decreases. Furthermore, the fast cooling rate associated with the low heat input results in higher delta ferrite content promoted in the FZ. This also enhances the strength of the joint welded at low heat input, which is in agreement with the literature [13,33,38]. [40,41] and presented in Table 5. Furthermore, the fracture of welded tensile specimens happened at the weld metal because the filler metal has an ultimate tensile strength lower than the base metals, as shown in Figure 11. Joint efficiency = Ultimate tensile strength of the weld metal Ultimate tensile strength of the soft base metal × 100 (5) Figure 10. Stress-strain curves of base metals and weld metals at low and high heat inputs.  Table 5 listed the tensile mechanical properties of the dissimilar joints comparing with those of the BMs. It is observed that the dissimilar joints exhibit lower strengths with Figure 10. Stress-strain curves of base metals and weld metals at low and high heat inputs. [40,41] and presented in Table 5. Furthermore, the fracture of welded tensile specimens happened at the weld metal because the filler metal has an ultimate tensile strength lower than the base metals, as shown in Figure 11. Joint efficiency = Ultimate tensile strength of the weld metal Ultimate tensile strength of the soft base metal × 100 (5) Figure 10. Stress-strain curves of base metals and weld metals at low and high heat inputs.  Table 5 listed the tensile mechanical properties of the dissimilar joints comparing with those of the BMs. It is observed that the dissimilar joints exhibit lower strengths with

Fractography
The fracture surface morphology of the dissimilar joints that underwent tensile testing was examined by using SEM to reveal the fracture mechanism. Figure 12 shows the fractography of tensile specimens of dissimilar M-Mn SS and Ni-Cr SS welded joints at low and high heat inputs. It is observed that the fracture surfaces of both joints welded at low and high heat inputs exhibit a dimpled feature, which is a typical characteristic feature of the ductile fracture. Interestingly, the fracture surface of the joints welded with low heat input contains small dimples of various sizes. This indicates that a higher ductility of the joint could be achieved. In contrast, coarse and elongated dimples are promoted in the fracture surface of the joints welded with high heat input. This reveals grain coarsening in the structure of the FZ. Hence, the mechanical strength of the joint decreases. This matches with the tensile curves of the weldments with different heat inputs, as shown in Figure 10.

Materials
Heat Input

Fractography
The fracture surface morphology of the dissimilar joints that underwent tensile testing was examined by using SEM to reveal the fracture mechanism. Figure 12 shows the fractography of tensile specimens of dissimilar M-Mn SS and Ni-Cr SS welded joints at low and high heat inputs. It is observed that the fracture surfaces of both joints welded at low and high heat inputs exhibit a dimpled feature, which is a typical characteristic feature of the ductile fracture. Interestingly, the fracture surface of the joints welded with low heat input contains small dimples of various sizes. This indicates that a higher ductility of the joint could be achieved. In contrast, coarse and elongated dimples are promoted in the fracture surface of the joints welded with high heat input. This reveals grain coarsening in the structure of the FZ. Hence, the mechanical strength of the joint decreases. This matches with the tensile curves of the weldments with different heat inputs, as shown in Figure 10.

Conclusions
Dissimilar welding of M-Mn SS to Ni-Cr SS was performed by the GTAW technique at low and high heat inputs (0.486 kJ/mm and 0.558 kJ/mm). There were no defects such

Conclusions
Dissimilar welding of M-Mn SS to Ni-Cr SS was performed by the GTAW technique at low and high heat inputs (0.486 kJ/mm and 0.558 kJ/mm). There were no defects such as porosity, undercuts, or cracks at the weld metals, and full penetration of the welding path was obtained for dissimilar welding at low and high heat inputs. The following conclusions were drawn from the current investigation: 1.
The weld metals were solidified in FA mode with conformable delta ferrite content (3-10) vol.% to avoid solidification cracks. Moreover, the delta ferrite content increased with the decrease in the heat input.

2.
The microstructure of weld metals consists of a duplex structure containing austenite matrix and delta ferrite for both heat inputs. Moreover, the dendrite length and interdendritic spacing increased with the increase in heat input.

3.
Cr carbides were precipitated at the grain boundaries in the HAZ and FZ at the high heat input due to the diffusion of the C atoms from the lower Cr base metals to the higher Cr weld metal. 4.
The weld metals have hardness values that are lower than the base metals under the two different heat inputs. Moreover, the hardness value of weld metals decreased with the increase in heat input.

5.
At low heat inputs, the ultimate tensile strength and elongation percentage were higher than those of the high heat input conditions. However, fractures occurred in the weld metal under the two heat inputs in a ductile mode.
Based on the present investigation, it is recommended to perform dissimilar welding of M-Mn SS and Ni-Cr SS using the GTAW technique at low heat inputs (0.486 kJ/mm) because the tensile strength and ductility are higher than the high heat input conditions.