A parametric study and experimental investigations of microstructure and mechanical properties of multi-layered structure of metal core wire using wire arc additive manufacturing

In the present study, the Gas metal arc welding (GMAW) based Wire-arc additive manufacturing (WAAM) process was preferred for the fabrication of multi-layered structures and their investigations of mechanical properties on metal core wire. Based on literature work, preliminary trials, machine limits, travel speed (TS), voltage (V), and gas mixture ratio (GMR) were identified as machining parameters along with output factors of bead width (BW), bead height (BH), and depth-of-penetration (DOP). Experiments were conducted by following the Box-Behnken design. The feasibility of the generated non-linear regression models has been validated through the statistical analysis of variance and residual plots. The multi-layered structure has been successfully fabricated at the optimized parametric settings of TS at 24 mm/s; the voltage at 24 V, and GMR at 1 which was obtained through the heat transfer search (HTS) algorithm. The fabricated structure was observed to be uniform. The structure exhibited uniform bead-on-bead deposition for the deposited layers. The fabricated multi-layered structure underwent a detailed microstructural and mechanical examinations. Microstructural examination revealed dense needles at the bottom section of the structure as compared to the top section, as the bottom section undergoes multiple heating and cooling cycles. When comparing the multilayer structure to the metal core wire, all the properties exhibited favorable tensile characteristics. The obtained strength from the impact test results highlights the impressive ductility of the multi-layer deposition. Fractography of tensile and impact test specimens has shown the occurrences of larger dimples and suggested a ductile fracture. Lastly, the hardness value in all the sections of the built structure was observed to be uniform, suggesting uniform deposition across the built multi-layer structure. The authors consider the current work will be highly beneficial for users in fabricating multi-layer structures at optimized parametric settings and their investigations for mechanical properties for metal core wire.


Introduction
In recent years, the additive manufacturing technique has been widely accepted to build intricate 3D designs in layers by depositing material, resulting in about 50 % savings on fabrication compared to traditional manufacturing (Chen et al., 2022, Nguyen et al., 2022, Zhou et al., 2022).Additive manufacturing enables the creation of highly complex items such as aerospace components, valves, gears, and biomedical prosthetics while conserving materials and production time (Al Rashid et al., 2021, Frascio et al., 2021).Among the several available techniques of the additive process, wire-arc additive manufacturing (WAAM) from direct energy deposition has proved to be an advantageous method for metallic components which utilizes a wire feeding mechanism, welding torch/power source, and computer control system (Zahidin et al., 2023, Rauch et al., 2021, Taheri et al., 2022) WAAM process produces the nearly net-shaped components.Thus it reduces the need for additional finishing.The electric arc is utilized as the heat source and the wire as feed (Elsokaty et al., 2023).WAAM process depends on the fundamental concepts of Gas Tungsten Arc Welding, Plasma Arc Welding, and Gas Metal Arc Welding (GMAW).Owing to the advantageous features like effortlessness in the deposition of material with a higher deposition rate and fabrication of thin multi-walled structures with minimum capital cost, GMAW-based WAAM has been broadly adopted (Sinha et al., 2022, Aldalur et al., 2022, Vora et al., 2023).Large components can also be produced by adjusting the process parameters.WAAM is based on welding principles and has superior metallurgical and mechanical properties compared to subtractive manufacturing for most of the alloys (Giarollo et al., 2022, Wankhede andVinodh, 2022).
During the production of Specimens through WAAM, a significant obstacle does occurs in residual stresses and distortion.It is vital to confirm that the quality of the final specimen meets industry standards.Thus, it is necessary to optimize the process parameters that govern the specimens' quality (Kumar andMaji, 2020, Baghel andGupta, 2022).Process variables such as wire feed speed, torch angle, current intensity, voltage (V), stand-off distance, shielding gas type, gas mixture ratios (GMR), gas flow rate, and travel speed (TS) play a crucial role in the manufacturing of components through WAAM.Controlling these variables for a suitable multi-layered structure is essential.Among the output responses, depth of penetration (DOP), bead width (BW), and bead height (BH) played a crucial role during the fabrication of the components through the GMAW-based WAAM process.
Low Alloy Steels are largely suitable for fabricating parts that should withstand stress and temperature (Tomita, 2000).By examining steel's mechanical, chemical, and physical attributes, it becomes apparent that it is ideal for use in fluctuating temperatures due to its safe and reliable properties (Das et al., 2021).As the operating temperature of parts increases, signs of degradation become more noticeable, making post-processing (welding and heat treatment) important for the best performance.Two types of wires are generally used in welding: metal core and flux core.While metal core wires have a higher deposition rate, flux core wires are equipped with a shielding flux that improves overall welding quality.Metal core wires are the more efficient option owing to their capacity to enhance current density and deposition (Das et al., 2021).
Many attempts have been made to optimize parameters for welding and investigations of the mechanical properties of multi-layered structures.Dinovitzer et al. (2019) present an experimental investigation of the impact of process variables on the microstructure and bead geometry of WAAM-deposited AISI 316L SS.They systematically varied the process variables, such as voltage, wire-feed ratio, and travel speed to optimize the deposition process and study its consequence on the bead geometry and microstructure.The results show that the optimized process parameters produced beads with a more uniform cross-sectional area and reduced height variation.They have used statistical analysis to identify the significant process variables and their interactions with the bead geometry and microstructure.Sarathchandra et al. (2020) analyzed the impact of WAAM variables on the microstructure and mechanical characteristics of SS304 beads.They employed optical microscopy and microhardness testing to assess the qualities of the deposition beads.The study discovered that the microstructure and mechanical characteristics of the deposited beads were considerably impacted by the process variables.To make beads with the required properties, the ideal variables were found.The study emphasizes how crucial it is to control the heat input throughout the WAAM to obtain the deposited beads' appropriate properties.Mai et al. (2021) presented a comprehensive study on optimizing processing variables and material aspects of 308L stainless steel components produced by WAAM.The paper mainly focuses on optimizing processing variables for WAAM of 308L SS.They conducted a series of experiments to explore the impact of various variables, such as wire feed speed, voltage, and travel speed, on the quality of the deposited material.Kumar and Anandakrishnan (2020) present a study on the influence of process variables on the layer geometry of Inconel 825 in WAAM.They also developed mathematical models to predict the layer height and width based on the process parameters.The process variables were optimized using the generated mathematical models to attain the desired layer geometry.Vora et al. (2022) have used the Box-Behnken Design (BBD) to create the experimental matrix and explore the impact of various process variables on the bead morphology.The results showed that the optimized process variables resulted in a bead with a uniform cross-sectional area and minimal height variation.Overall, the paper provided valuable insights into optimizing GMAW-based WAAM processes for fabricating 2.25 Cr-1.0 Mo steel components with desired bead morphology and mechanical characteristics.Chaudhari et al. (2022) studied the impact of various process variables on the bead geometries of GMAW-based WAAM of 316L SS.They used a design of experiments approach to systematically vary the process variables and evaluate their impact on the bead geometries.The results show that the optimized process parameters resulted in beads of uniform width, height, and penetration depth.Palmeira et al. (Belotti et al., 2022) provided a comprehensive analysis of the microstructural characteristics of WAAM stainless steel.Their obtained results have shown a detailed analysis of the microstructural features of the WAAM stainless steel, such as the grain size, morphology, and phase distribution.The article highlighted the importance of microstructural analysis in understanding the properties and performance of WAAM materials.Different methods of characterization, including optical microscopy, scanning electron microscopy, and X-ray diffraction, were utilized to analyze the microstructure of WAAM stainless steel.The study shows that variations in process parameters can significantly affect the grain size, morphology, and phase distribution of the material, which in turn affects its mechanical properties.Vora et al. (2022) provided valuable insights into the mechanical properties of multi-layered structures created using GMAW-based WAAM of SS316L.They conducted a series of experiments to evaluate the tensile, compressive, and fatigue properties of the fabricated structures.The results show that the fabricated structures have high tensile and compressive strength and can withstand cyclic loading without significant damage.They also analyzed the microstructure of the samples and observed that the grain size and porosity were within acceptable limits.The study concluded that GMAW-based WAAM is a promising technology for fabricating multi-layered structures with excellent mechanical properties.With the assistance of Taguchi-based regression analysis, Hussein et al. (2022) presented an experimental study on optimizing process variables for the WAAM of aluminum alloy 4043.To optimize the procedure and explore the impact of these variables on the deposition quality, they adjusted process variables such as wire feed speed, travel speed, and voltage.The benefits of applying this strategy to process optimization are presented in the study, along with a discussion of the Taguchi-based regression analysis used to improve the process variables.In another study, Chaudhari et al. (2022) studied the impact of multi-walled structures on the microstructure and mechanical properties of 1.25Cr-1.0Mosteel fabricated by GMAW-based WAAM using metal-cored wire.They conducted a series of experiments to evaluate the microstructure, tensile strength, and impact toughness of the fabricated structures with multi-walled and single-walled structures.The results show that the multi-walled structure has a positive impact on the microstructure and mechanical properties of the fabricated samples.The multi-walled structure results in refined grains, reduced porosity, and increased strength and toughness.They also analyzed the fracture surfaces and observed that the multi-walled structure improves the ductility of the samples.
Heat transfer search (HTS) is a metaheuristic optimization method that can easily implement and solve complex engineering problems.In recent past studies, researchers have successfully applied the HTS method effectively in various manufacturing-related processes (Chaudhari et al., 2021, Vora et al., 2021a, Vora et al., 2021b, Chaudhari et al., 2019).
The WAAM technique, including the entire additive manufacturing domain, is primarily held back due to productivity issues (time/cost).As R. Chaudhari et al. per the studied literature, minimal work has been carried out to optimize the process variables for fabricating multi-layered structures of the metal core wire.In the present study, an attempt has been made to narrow down the aforementioned research gap by using the GMAWbased WAAM process for metal core wire.Based on literature work, preliminary trials, and machine limits, TS, voltage, and GMR were identified as machining parameters along with BW, BH, and DOP as output factors.Experiments were conducted by following the Box-Behnken design.Non-linear regression models were developed from the derived results of the conducted trials.The feasibility of these models was validated through the statistical approach of analysis of variance and residual plots.In the present work, the authors have adapted the meta-heuristic optimization approach of the HTS technique to optimize the weld bead's process parameters.The multi-layered structure was fabricated at optimized parametric settings.The fabricated multi-layered structure underwent a detailed microstructural and mechanical examination, including tensile testing, impact testing, hardness, and fractography.The authors consider the current work with optimized parametric settings will be highly beneficial for users in fabricating multi-layer structures.

Experimental setup and plan
The present study has utilized a 1.2 mm diameter of the metal core wire and bead-on-plate trials on Mild Steel 2062 grade substrate with a thickness of 20 mm using the GMAW process.The make of wire is Metalloy 80N1 from Hobart Brothers (TRI-MARK) company.The chemical composition of the metal core wire is shown in Table 1.The WAAM Machine system was utilized to perform the test.The substrate plate was carefully cleaned and dried before the depositions.The experimental setup involved a GMAW torch, Wire Feeder, Shielding Gas Cylinders (Argon and Carbon dioxide), controller, and a specialized machine.The controller controls the movement of the nozzle.The torch was movable in any direction and was used to deposit the material on the clamped base metal.The torch was kept exactly vertical and perpendicular to the deposition path.Before starting the program, the shielding gas was supplied to keep deposited material from contacting any ambient gases.The experimental setup employed in the present work is depicted in Fig. 1.The machining setup has a built volume of 220 × 220 × 500 mm.
The depositing of a single bead was performed on a substrate plate using an experimental matrix generated through the BBD using the response surface methodology (RSM).Input variables GMR, TS, and voltage were utilized in the study.GMR represents the ratio of % CO 2 gas, and the remainder is argon.The flow rate was maintained above 15 L/min due to recent literature research and machine capabilities (Chaudhari et al., 2022).The bead length was set at 150 mm.During the experimentations, the GMAW torch was used with a metal core wire having a diameter of 1.2 mm.The process parameters of experimental conditions used in the study are presented in Table 2.The BBD, which employs RSM, was used to properly organize the experimental matrix to attain the best possible response (Chaudhari et al., 2022).This method minimizes experimentation, saving labor costs and time (Javidikia et al., 2023).It also establishes a correlation between the answers and machining variables.Fifteen trials were conducted by adjusting the input variables.All the experimentations were repeated three times to check the repeatability and accuracy of the trials, and their average values were selected for analysis.

Measurement of output responses
The output parameters considered in the present work included DOP, BH, and BW.Each bead deposition was cut into a small cross-section to measure the bead geometries.The WEDM process was utilized to create specimens for the examination of output variables.All the samples were initially cleaned and polished with different abrasive papers.Optical Macroscopy was used for analyzing DOP, BH, and BW.In each bead deposition, three readings across the cross-section were taken for better accuracy.Their average value has been considered during the analysis.Fig. 2 shows the measurement process of output measures.Fig. 3 displayed the cross-sections of 15 trials in correspondence with the BBD design matrix, as shown in Table 3.

Optimization using HTS algorithm
The HTS algorithm was utilized to optimize WAAM variables in the present work.Single-response and multi-response optimization was carried out by considering maximization criteria for BH, and DOP along with minimizing BW response.The HTS algorithm is a parameterless technique that is very easy to apply, fast and establishes improved convergence for the desired outcome.HTS is a metaheuristic optimization method that can easily implement and solve complex engineering problems (Chaudhari et al., 2020).The HTS algorithm, developed by Patel and Savsani (2015), aims to achieve thermal equilibrium in an unbalanced system by incorporating the three primary modes of heat transfer: conduction, convection, and radiation.Each mode is given an equal opportunity to transfer heat, and the algorithm randomly selects one of these modes for each generation.The algorithm initiates with a population of molecules having random temperature levels, and in each generation, it is updated using one of the heat transfer modes.Better solutions are accepted, while elite ones replace poorer ones.The process is represented in Fig. 4. Overall, the objective of the HTS algorithm is to optimize the system's thermal equilibrium by maintaining balance among the three modes of heat transfer.
Heat transfer by conduction mode.Eqs. ( 1) and (2) were used to improve the solutions by conduction mode.
Heat transfer by convection mode.Eqs. ( 3) and ( 4) were used to improve the solutions by conduction mode.
Heat transfer by radiation mode.Eqs. ( 5) and ( 6) were used to improve the solutions by conduction mode.

Examination of microstructure and mechanical properties
The fabricated multi-layered structure underwent a detailed microstructural and mechanical examination.The multilayer structure of 45 mm in height and 150 mm in length was fabricated.Fig. 5 displays the various testing locations of the built multi-layered structure.A cuboid of 105 × 35 × 35 mm was created from the structure utilizing the wire electrical discharge machining process.The WEDM process was utilized to create specimens for metallographic analyses and mechanical characteristics.
All the samples were initially cleaned and polished.The chemical composition of the multi-layered structure has also been tested in multiple locations.The obtained results of chemical compositions were observed to be within the prescribed limit of metal core filler wire.The fabricated multi-layered structure underwent a detailed microstructural and mechanical examination, including tensile testing, impact testing, hardness, and fractography.
The examination of microstructure has been carried out by using an   optical microscope of RAD ICAL.An ASTM E8M standard-compliant tensile specimen sample was cut through WEDM.The impact test was performed using the Charpy impact testing machine at three different locations.Fig. 5 depicts a schematic representation of the impact testing locations and samples.Measurements of Hardness were made in accordance with ASTM E-10.A 187.5 kg load with a 2.5 mm ball diameter was applied for the indentations during the Brinell hardness measurement.Multiple readings were collected for each specimen, and their average value was used for better accuracy.Fractography of tensile and impact specimens was investigated by using a scanning electron microscopy (Zeiss Ultra-55) setup.

Results and discussions
Table 3 depicts the results of output measures by following the BBD matrix of the RSM approach.Maximum bead height, depth of penetration, and minimum bead width are desirable for multi-layer fabrication.Multivariable non-linear regressions among the process parameters and responses were generated.Analysis of variance, residual plots, the effect of input variables on responses, and optimization of variables were discussed in detail in this section.
Multivariable non-linear regressions among the process parameters and responses were generated.BBD design of RSM was integrated into Minitab v17 to generate these equations.They established the relationships between the WAAM variables and bead height, depth-ofpenetration, and bead width responses.These equations will serve as a starting point to evaluate the responses' values beyond the BBD experimental matrix.Eqs.(1-3) represent multivariable non-linear regressions for bead height, bead width, and depth-of-penetration respectively.Eqs.1-3 are regression equations that primarily explain the effect of independent parameters (TS, V, and GMR) and their interactions on the dependent quality parameter (BH, BW, and DOP).
ANOVA for output responses ANOVA analysis was employed to validate the appropriateness and reliability of the obtained regression equations.Minitabv17 was employed to evaluate relevant and non-relevant model terms, with a 95 % confidence level.Probability values less than 0.05 indicate the respective term's significant impact on the response variable (Rathi et al., 2020, Tiwary et al., 2023).Table 4 depicts ANOVA results for  output responses.The obtained statistical result for output factors has shown a noteworthy contribution as their regression, linear, square, and 2-way interaction models have a significant effect.In the case of BW, and DOP response, all three WAAM variables of TS, V, and GMR were considered essential factors.V and GMR had significant parameters for BH response.Among the WAAM variables, V was found to have the highest contributing factor for BH, while TS was a highly contributing effect for both BW and DOP.Additionally, the error term has a minimal impact on all responses, indicating that the developed model accurately predicted responses with minimal errors.For BH response, percentage contributions of TS, V, and GMR were observed to be 1.21 %, 94.17 %, and 4.62 %, respectively.In the case of BW, the percentage contributions of 89.31 %, 8.54 %, and 2.15 % were found for TS, V, and GMR, respectively.Lastly, DOP response has shown 80.83 %, 7.49 %, and 11.68 % contributions for TS, V, and GMR, respectively.Furthermore, lack-of-fit was also noticed to be statistically non-significant.This, in  turn, is again a crucial factor that confirms the accuracy of the ANOVA results.Therefore, the lack-of-fit's insignificance for the response implies that the model is appropriate for predicting the output value (Fuse et al., 2021).This significance of the created model terms shows that the generated regression equations are acceptable and dependable for forecasting the values of all the output measures.R 2 value close to one indicates that the regression effectively predicts the value (Vora et al., 2022).The proposed model's adequacy was evaluated using R 2 values.The R 2 values for BH, BW, and DOP were recorded at 0.9579, 0.9824, and 0.9920, respectively.The results indicated that the R 2 values for all responses were near one, suggesting that the model fits the data well and can effectively anticipate new observations.In order to ensure the ANOVA model's appropriateness, certain assumptions need to be met, and residual plots are used to verify the analysis results.The residual plots for BH, which consist of four plots, are presented in Fig. 6 a.The normality plot shows a linear trend that supports the model's suitability.The verses fit plot confirms that the fits were randomly distributed around the source, and the histogram plot exhibits a bell-shaped curve indicating the adequacy of ANOVA results.The versus order plot's absence of any particular pattern further confirms the ANOVA statistics.These four plots demonstrate the ANOVA statistics, leading to better future outcome predictions.Similar outcomes were observed for BW and DOP, as shown in Fig. 6 b and c respectively.As a result, all four plots successfully validate the ANOVA statistics, improving the accuracy of future outcome predictions for BH, DOP, and BW.

Main effect plots for BH, BW, and DOP
The impact of WAAM variables (TS, V, and GMR) was investigated on BH, BW, and DOP responses.Fig. 7 depicts the impact of WAAM variables on BH.The BH response has shown a marginal decline trend with the increased value of TS.The reason behind this trend was the larger speed of the torch.As the torch speed increases, a lower amount of material is deposited; therefore, BH decreases (Thakur andChapgaon, 2016, Kumar et al., 2021).The voltage had a significant impact on the BH response.BH was reduced with an increased voltage from 24 V to 28 V.This decrease is caused by increased arc length and more molten material being deposited (Vora et al., 2022, Mistry, 2016).As a result of spreading these melted droplets, a reduction in BH was noticed.GMR hasn't shown any significant impact on the BH response.It primarily protects the weld pool from atmospheric contamination and does not significantly affect the weld BH (Jurić et al., 2019).Maximum BH, DOP, and minimum BW are desirable for multi-layer fabrication.For attaining maximum BH response, it was desirable to have WAAM variables of TS at 20 mm/s, voltage at 24 V, and GMR at 1.
The main effect plot presented in Fig. 8 illustrates the effect of WAAM variables on BW.The BW response demonstrated a decreasing trend as TS increased from 20 mm/s to 26 mm/s.The increased value of TS has reduced the material deposition owing to the increased torch speed (Xu et al., 2015).Thus, the BW value also decreases.BW was observed to increase with an increased voltage value.This increase caused increased arc length and more molten material deposited at a higher voltage.This leads to the spreading of molten droplets, resulting in a noticeable increase in BW (Chaudhari et al., 2022, Wu et al., 2018).The gas mixture ratio, on the other hand, has minimal impact on the BW response.Its primary function is to protect the weld pool from atmospheric contamination and does not significantly affect the weld BW.For attaining minimum BW response, it was desirable to have WAAM variables of TS at 26 mm/s, voltage at 24 V, and GMR at 9.
The effect of WAAM variables on DOP response is represented in Fig. 9.It was observed that the DOP value was improved with the increased value of TS.This was due to the larger speed of the torch, resulting in a lower amount of material being deposited (Nagasai et al., 2022).Conversely, the voltage minimally impacts the DOP response as the voltage increases from 24 V to 28 V.With the increased voltage, larger material gets deposited as it enhances the arc length (Huang and Shi, 2022).This in turn increased the BW value.The gas mixture ratio has a positive impact on the DOP response.As the carbon percentage increases, the weld penetration increases, and therefore DOP increases.For attaining maximum DOP, it was desirable to have WAAM variables of TS at 26 mm/s, voltage at 28 V, and GMR at 9.

Optimization using the HTS algorithm
The contradictory conditions of WAAM variables were evident from the ANOVA and main effect plots for attaining minimum BW, maximum BH, and DOP.This, in turn, raises the need to optimize WAAM variables.The HTS algorithm was utilized to optimize WAAM variables in the present work.According to this, during the execution of HTS, the range of WAAM variables used includes 20 ≤ WFS ≥ 26; 24 ≤ V ≥ 28, and 1 ≤ GMR ≥ 9.
A single-response optimization has been implemented for BH, BW, and DOP.Table 5 depicts the results of single-response optimization for BH, BW, and DOP showing the predicted values from the HTS algorithm and actual experimental values.Similar to the conclusion from the main effect plot, single-response optimization results yielded contradictory conditions of WAAM variables for BW, BH, and DOP.Verification experiments were carried out to verify the adequacy of the results produced by the HTS algorithm.An acceptable deviation of less than 5 % was witnessed among the actual and predicted value results.To tackle the contradictory situation, it is necessary to optimize BH, BW, and DOP responses simultaneously.Pursuant to this, multi-objective HTS (MOHTS) was employed, producing non-dominated solutions.Thus, simultaneous optimization was carried out using the MOHTS algorithm, which is easy to implement and can solve complex  of generated regression models and the HTS algorithm.
The objective function was selected by assigning equal importance to BH, BW, and DOP responses.MOHTS was used for obtaining the levels of WAAM variables, and it has shown at TS at 24 mm/s; the voltage at 24 V, and GMR at 1 with response values of BH at 5.75 mm, BW at 5.65 mm, and DOP at 0.93 mm.A multi-layered structure will be fabricated based on these parametric settings.

Fabrication of multi-layered structure
The multi-layered structure has been successfully fabricated at the optimized parametric settings of TS at 24 mm/s; the voltage at 24 V, and GMR at 1.The fabricated structure was observed to be uniform.Small metal lumps were observed on one side of the layer's deposition.However, this can be easily removed during post-processing.Thus, the side surfaces of the structure were removed by using the WEDM process.Fig. 10 shows the fabricated multi-layered structure after removing from the base plate by using the WEDM process.The structure exhibited uniform bead-on-bead deposition, with each layer being identical.To mitigate the occurrence of excessive residual strains and deformation, a cooling interval of 60 seconds was implemented between successive layers.This cooling time also aids in the solidification process between continuous layers.Fig. 10 demonstrated proper bonding and displayed smooth fusion between the manufactured layers.Thus, the optimized process parameters, implemented through GMAW-based WAAM technology, effectively yielded a successful multi-layered structure of high quality on an MS substrate.The mechanical properties of the multilayered structure were thoroughly examined in the study.

Microstructure of multi-layered structure
The microstructural images of the bottom, middle, and top sections have been reported as shown in Fig. 11 a-c respectively.It can be observed that a series of acicular ferrites having the traditional shape of needles have been observed.These constituents are a result of the displacive transformation, where austenite transforms to carbon super-saturated ferrite and is immediately followed by carbon partitioning from the ferrite to the surrounding austenite (Cho et al., 2022, Jorge et al., 2019).This acicular ferrite is also characterized as an intergranular upper bainite which possesses fairly greater hardness.These can also be confirmed by the hardness values of all three sections wherein a hardness of less than 220 has been achieved.It can also be noticed from the microstructures that the needles available at the bottom section are dense as compared to the top section as the bottom section undergoes multiple heating and cooling cycles (Kumar et al., 2021, Wang et al., 2019).The cooling rates experienced by the bottom structure are harsher due to the colder surroundings.

Mechanical properties of multi-layered structure
The fabricated multi-layered structure underwent a detailed mechanical examination, including tensile testing, impact testing, hardness, and fractography.For better accuracy and reproducibility of the results, multiple specimens were tested for each condition, and their average value was considered for analysis.

Hardness testing
The Hardness testing (Brinell Hardness Test) of the multilayer structure was performed as per ASTM E-10.Testing has been carried out in different locations (top, middle, and bottom sections) of the built structure.In each section, five readings were taken for evaluation.The average values obtained in the top section, middle section, and bottom sections were 201 ± 3.88 HBW, 198 ± 4.11 HBW, and 203 ± 3.96 HBW respectively.The average Hardness value was found to be 200.7 HBW.Hardness value in all the sections of the built structure was observed to be uniform, suggesting uniform deposition across the built multi-layer structure.

Tensile testing
Tensile testing was conducted on the bead multilayer structure to assess its mechanical properties.Tensile test specimens were prepared by following ASTM E8M standards using a WEDM setup.The specimens exhibited both elastic and plastic deformation before fracturing.Table 7 displays the tensile test results along with the mechanical properties of used filler wire material.The obtained results revealed that the multilayer structure had an ultimate tensile strength (UTS) of 659.6 ± 7.23 MPa, yield strength (YS) of 595.6 ± 8.94 MPa, and a percentage elongation of 26 ± 0.78 %.These tensile properties were then compared to those of the metal core wire.Table 7 provides a summary of the tensile test results of metal core wire indicating UTS, YS, and elongation values of 650 MPa, 470 MPa, and 24 %, respectively.The WAAM components displayed UTS, YS, and elongation values that fell within the range of the metal core wire (E80C Ni1) values.When comparing the multilayer structure to the metal core wire, all the properties exhibited favourable tensile characteristics.The obtained results were also in line with the sensible agreement given by Das et al. (2021).Consequently, the GMAW-based WAAM component adheres to industry standards for commercial use, as its tensile properties satisfy the requirements for metal core wire.
Fractography of the tensile specimen has been carried out by using scanning electron microscopy as shown in Fig. 12.A consistent occurrence of a higher amount of dimples was found on the specimen of the fractured surface.This suggests a greater toughness of the structure obtained through the GMAW-based WAAM process.The higher amount of dimples next to others also established the superior ductility of the parts from multi-layered structure parts.

Impact testing
The Charpy impact test was performed on specimens prepared as per the ASTM E-23 standards.The specimens with dimensions of 10 mm x 2.5 mm x 55 mm were prepared through WEDM and featured a 45 • V-notch with a root radius of 0.25 mm at their centre.The impact test was conducted at three different locations (top, middle, and bottom sections) of the built structure.Table 8 displays the results of the impact testing for the multilayer structure, indicating its impact strength.The impact test values obtained in the top section, middle section, and bottom sections were 22 J, 22 J, and 24 J respectively.The average impact energy value was calculated as 22.67.The impact test results for all sections exhibited good impact strength as per the requirement as shown in Table 8.Furthermore, the obtained strength from the impact test results highlights the impressive ductility of the multi-layer deposition.Therefore, the GMAW-based WAAM component meets the requirements for commercial applications.
Fractography of the impact specimen has been carried out by using SEM as shown in Fig. 13.The fracture surface of the specimen obtained through SEM depicted a large amount of dimples along with uniform circulations.This has again shown the ductile nature of the specimens.The fabricated multi-layered also depicted superior impact properties as the surface was observed to be free from micro-pores and micro-cracks.Thus, the obtained results show the adequacy of the GMAW-based WAAM process for manufacturing the multi-layered structure of metal-cored wire.

Conclusions
The present study used the GMAW-based WAAM process for the metal core wire to fabricate multi-layer structures and investigate their microstructure and mechanical properties.Important conclusions derived from the present work were summarized as follows: • Statistical analysis through ANOVA has successfully validated the generated multivariable non-linear regressions.Among the input variables, voltage was found to have the highest contributing factor for BH, while TS was a highly contributing effect for both BW and DOP.• Single-response optimization using the HTS technique has shown a maximum BH of 6.13 mm, minimum BW of 4.91 mm, and maximum DOP of 1.08 mm.The simultaneous optimization through heat transfer search (HTS) algorithm yielded optimal parametric settings of TS at 24 mm/s; voltage at 24 V; and GMR at 1 for the fabrication of a multi-layered structure.• The multi-layered structure has been successfully fabricated at the optimized parametric settings.The fabricated structure was observed to be uniform.The structure exhibited uniform bead-on-bead deposition, with each layer being identical.• Microstructural examination revealed dense needles at the bottom section of the structure as compared to the top section, as the bottom section undergoes multiple heating and cooling cycles.• The average Hardness value was found to be 200.7 HBW.The obtained results of the tensile test revealed that the multilayer structure had a UTS of 659.6 ± 7.23 MPa, YS of 595.6 ± 8.94 MPa, and a percentage elongation of 26 ± 0.78 %.The WAAM components displayed UTS, YS, and elongation values that fell within the range of the metal core wire (E80C Ni1) values.The impact test values obtained in the top section, middle section, and bottom sections were 22 J, 22 J, and 24 J respectively.The average impact energy value was calculated as 22.67 J.The impact test results for all sections exhibited good impact strength as per the requirement.Fractography of tensile and impact test specimens has shown the occurrences of larger dimples and suggested a ductile fracture.• The authors consider the current work will be highly beneficial for users in fabricating multi-layer structures at optimized parametric settings and their investigations for mechanical properties for metal core wire.

Table 1
Chemical composition of the Metal core wire and substrate plate.

Table 2
Process variables of WAAM.

Table 3
Obtained results of BH, BW, and DOP.

Table 4
ANOVA for BH, BW, and DOP.

Table 6
Results of multi-objective optimizations for BH, BW, and DOP.

Table 7
Tensile properties of multilayer structure and metal core wire.
R.Chaudhari et al.

Table 8
Impact testing results.