The Influence of Process Parameters on the Microstructure and Microhardness of 304 Stainless Steel in Joule Heating Fused Filament Fabrication

: Using finite element simulation and single-variable experimental methods, this study analyzes the variations in the microstructure and hardness of a 304 stainless steel wire during Joule heating fused filament fabrication. The effects of current intensity, printing speed, and roller pressure on the macroscopic morphology, microstructure, and microhardness of a single-layer single-channel formation were investigated. The results indicate that when the current intensity is 400 A, the printing speed is 1000 mm/min, and the roller pressure is 0.3 N, the surface of the single-layer single-channel formation is smooth and exhibits optimal forming characteristics with a width-to-height ratio of 3.23, a dilution rate of 51.61%, and an average microhardness of 238.17 HV. As the current intensity increases, the microstructure in the fusion zone initially decreases in size and then increases; similarly, with the increase in printing speed, the microstructure in the fusion zone first decreases and then increases; as the roller pressure increases, the microstructure in the fusion zone initially increases in size and then decreases. The microhardness initially increases and then decreases with the increase in process parameters, resulting in uneven hardness distribution due to the variations in microstructure size. The optimal combination of process parameters achieves a balance between heat input, cooling rate, and growth rate, thereby achieving grain refinement and hardness improvement, ultimately enhancing the mechanical properties of the material.


Introduction
In recent years, the global space manufacturing sector has become fiercely competitive, with major countries adjusting their strategies to vie for dominance in space [1].However, the space sector faces challenges such as high costs, high-power heat sources, and equipment reliability, while transporting resources from Earth is time-consuming and costly [2].Therefore, the development of low-energy, small-sized, and low-cost equipment to improve space printing efficiency has become an urgent need to ensure space safety and strategic dominance.Additive manufacturing, as a new era manufacturing technology, can fulfill the manufacturing tasks required in space.Additive manufacturing (AM), also known as 3D printing technology, forms three-dimensional parts by stacking materials layer by layer [3].It can utilize the microgravity environment of space to manufacture special materials and structures, with metal additive manufacturing being particularly crucial in the aerospace field [4,5].Metal additive manufacturing technology slices the 3D model, plans the path, and controls the heat source for layer-by-layer processing, ultimately producing solid parts [6,7].Currently, this technology is mainly focused on metal powder bed and metal wire processes.The former, such as Selective Laser Melting (SLM), offers high precision and density but is costly and limited in microgravity environments [8].Therefore, in recent years, some organizations have turned to metal wire additive manufacturing technology.Professor Shujun Chen from the Welding Research Institute of Beijing University of Technology [20] have proposed a metal additive manufacturing method suitable for space environments, utilizing resistance heating to melt the wire, followed by microscopic analysis of the melted and cooled wire.The study found that in the microstructure of hypoeutectoid steel with higher carbon content, the area occupied by pearlite was larger.Li et al. [21] used the Joule heating additive manufacturing process to print single-layer single-channel samples and analyzed their microstructure, revealing significant temperature gradients in the fusion zone of the single-layer single-channel, which promotes grain growth.Currently, Joule heating additive manufacturing is not yet fully mature, suffering from issues such as process instability and poor surface quality.Ren Yake et al. [22] optimized the laser power, scanning speed, and scanning spacing using the selective laser melting (SLM) process.They discovered that these parameters have a nonlinear impact on the relative density, metallurgical defects, and microstructure of the Cu-1.93Cr-0.74Nballoy.Wang Enmao et al. [23] determined the range of the spot-welding process parameters for Q&P980 galvanized high-strength steel through orthogonal experiments.They found that the microstructure and hardness distribution of the weld joints followed specific patterns, although issues such as liquid metal embrittlement cracks were also observed.Das J et al. [24] studied the effects of process parameters on the welding joints of aerospace-grade 5052 aluminum alloy using friction stir welding.They identified the optimal combination of process parameters to enhance the mechanical properties and grain refinement of the joints.Wei Y et al. [25] successfully fabricated crack-free M2 high-speed steel using selective electron beam melting technology at a specific volumetric energy density and high powder bed preheating temperature.The resulting material exhibited a fully dense structure, fine grains, and carbides, demonstrating ultra-high hardness and excellent tribological properties.Despite some existing issues, Joule heat additive manufacturing holds great potential for use in space environments.
In this study, the microstructure and hardness of 304 stainless steel were analyzed through finite element simulation and process experiments using Joule heat fused filament fabrication.The effects of current intensity, printing speed, and roller pressure on the macroscopic morphology, microstructure, and microhardness of a single-layer singlechannel formation were investigated.Optimal process parameters within this window were identified to achieve a better single-layer single-channel formation.The novelty of this study lies in its pioneering exploration of the influence of process parameters on the microstructure in Joule heating fused filament fabrication.It summarizes the microstructural changes in a single-layer single-channel formation, providing a reference for multi-layer single-channel fabrication.Additionally, the study identifies the optimal process parameters within this window, resulting in single-layer single-channel components with superior performance and smooth surfaces.This research offers theoretical and data support for ground-based verification of Joule heat metal fused filament additive manufacturing, with implications for space metal additive manufacturing.Based on the entire content, a comprehensive flowchart has been created to facilitate readers in understanding the research content and objectives of the article more easily, as shown in Figure 1.
Crystals 2024, 14, x FOR PEER REVIEW 3 of 18 steel using selective electron beam melting technology at a specific volumetric energy density and high powder bed preheating temperature.The resulting material exhibited a fully dense structure, fine grains, and carbides, demonstrating ultra-high hardness and excellent tribological properties.Despite some existing issues, Joule heat additive manufacturing holds great potential for use in space environments.In this study, the microstructure and hardness of 304 stainless steel were analyzed through finite element simulation and process experiments using Joule heat fused filament fabrication.The effects of current intensity, printing speed, and roller pressure on the macroscopic morphology, microstructure, and microhardness of a single-layer singlechannel formation were investigated.Optimal process parameters within this window were identified to achieve a better single-layer single-channel formation.The novelty of this study lies in its pioneering exploration of the influence of process parameters on the microstructure in Joule heating fused filament fabrication.It summarizes the microstructural changes in a single-layer single-channel formation, providing a reference for multilayer single-channel fabrication.Additionally, the study identifies the optimal process parameters within this window, resulting in single-layer single-channel components with superior performance and smooth surfaces.This research offers theoretical and data support for ground-based verification of Joule heat metal fused filament additive manufacturing, with implications for space metal additive manufacturing.Based on the entire content, a comprehensive flowchart has been created to facilitate readers in understanding the research content and objectives of the article more easily, as shown in Figure 1.

Experimental Principle and Equipment
The principle of Joule heating metal wire additive manufacturing is illustrated in Figure 2.This paper aims to develop a new additive manufacturing method with low energy consumption and high quality.The positive terminal of the power supply is connected to the roller, while the negative terminal is connected to a copper ring beneath the substrate.The CNC machine's Z-axis is controlled by a computer to regulate the roller's pre-pressure on the metal wire.Feedback from the pressure sensor adjusts the pressure value accordingly.A programmable power supply applies current to the system.The purpose is to form a closed circuit consisting of the positive terminal of the power supply, roller, metal wire, substrate, copper ring, and negative terminal of the power supply, thereby generating Joule heat to metallurgically bond the wire to the substrate.When the current reaches

Experimental Principle and Equipment
The principle of Joule heating metal wire additive manufacturing is illustrated in Figure 2.This paper aims to develop a new additive manufacturing method with low energy consumption and high quality.The positive terminal of the power supply is connected to the roller, while the negative terminal is connected to a copper ring beneath the substrate.The CNC machine's Z-axis is controlled by a computer to regulate the roller's pre-pressure on the metal wire.Feedback from the pressure sensor adjusts the pressure value accordingly.A programmable power supply applies current to the system.The purpose is to form a closed circuit consisting of the positive terminal of the power supply, roller, metal wire, substrate, copper ring, and negative terminal of the power supply, thereby generating Joule heat to metallurgically bond the wire to the substrate.When the current reaches its maximum value, the platform moves along a predetermined trajectory to achieve sliding pressure printing.After the motion is complete, the current input is disconnected, and the roller is lifted, completing one layer of single-channel printing in Joule heating metal wire additive manufacturing.its maximum value, the platform moves along a predetermined trajectory to achieve sliding pressure printing.After the motion is complete, the current input is disconnected, and the roller is lifted, completing one layer of single-channel printing in Joule heating metal wire additive manufacturing.The experimental platform is depicted in Figure 3, with the experimental equipment shown in Figure 3a, the printing area illustrated in Figure 3b, and the chromium-zirconium-copper roller model displayed in Figure 3c.Based on the improvement of the CNC/M-1 (Xianhe Yuexin, Zibo, China) micro CNC milling machine, the parameters of the micro milling machine are listed in Table 2. Windows-based open computer control is used to achieve three-axis linkage control of the CNC machine tool.The programmable DC power supply used during the experiment is the IT-M3910D-10-1020 (ITECH, Nanjing, China), with parameters detailed in Table 3, providing a maximum voltage of 10 V and a maximum current of 1020 A. The vacuum environment is maintained by the SCROLLVAC 15 PLUS 1-PH (LEYBOLD, Cologne, Germany) mechanical pump, which can reduce the atmospheric pressure inside the tank to below 50 Pa within 10 min.Subsequently, the TURBOVAC 450 i (LEYBOLD, Cologne, Germany) semi-magnetic levitation molecular pump is activated to reduce the pressure inside the tank to the order of 10-2 Pa within 5 min.The main performance parameters of the mechanical pump and molecular pump are listed in Table 4.The entire vacuum system is equipped with the ZDF-III-PRO (Chengzhen, Chengdu, China) high-precision vacuum gauge for accurately detecting the actual vacuum inside the tank.The experimental platform is depicted in Figure 3, with the experimental equipment shown in Figure 3a, the printing area illustrated in Figure 3b, and the chromium-zirconiumcopper roller model displayed in Figure 3c.Based on the improvement of the CNC/M-1 (Xianhe Yuexin, Zibo, China) micro CNC milling machine, the parameters of the micro milling machine are listed in Table 2. Windows-based open computer control is used to achieve three-axis linkage control of the CNC machine tool.The programmable DC power supply used during the experiment is the IT-M3910D-10-1020 (ITECH, Nanjing, China), with parameters detailed in Table 3, providing a maximum voltage of 10 V and a maximum current of 1020 A. The vacuum environment is maintained by the SCROLLVAC 15 PLUS 1-PH (LEYBOLD, Cologne, Germany) mechanical pump, which can reduce the atmospheric pressure inside the tank to below 50 Pa within 10 min.Subsequently, the TURBOVAC 450 i (LEYBOLD, Cologne, Germany) semi-magnetic levitation molecular pump is activated to reduce the pressure inside the tank to the order of 10-2 Pa within 5 min.The main performance parameters of the mechanical pump and molecular pump are listed in Table 4.The entire vacuum system is equipped with the ZDF-III-PRO (Chengzhen, Chengdu, China) high-precision vacuum gauge for accurately detecting the actual vacuum inside the tank.
its maximum value, the platform moves along a predetermined trajectory to achieve sliding pressure printing.After the motion is complete, the current input is disconnected, and the roller is lifted, completing one layer of single-channel printing in Joule heating metal wire additive manufacturing.The experimental platform is depicted in Figure 3, with the experimental equipment shown in Figure 3a, the printing area illustrated in Figure 3b, and the chromium-zirconium-copper roller model displayed in Figure 3c.Based on the improvement of the CNC/M-1 (Xianhe Yuexin, Zibo, China) micro CNC milling machine, the parameters of the micro milling machine are listed in Table 2. Windows-based open computer control is used to achieve three-axis linkage control of the CNC machine tool.The programmable DC power supply used during the experiment is the IT-M3910D-10-1020 (ITECH, Nanjing, China), with parameters detailed in Table 3, providing a maximum voltage of 10 V and a maximum current of 1020 A. The vacuum environment is maintained by the SCROLLVAC 15 PLUS 1-PH (LEYBOLD, Cologne, Germany) mechanical pump, which can reduce the atmospheric pressure inside the tank to below 50 Pa within 10 min.Subsequently, the TURBOVAC 450 i (LEYBOLD, Cologne, Germany) semi-magnetic levitation molecular pump is activated to reduce the pressure inside the tank to the order of 10-2 Pa within 5 min.The main performance parameters of the mechanical pump and molecular pump are listed in Table 4.The entire vacuum system is equipped with the ZDF-III-PRO (Chengzhen, Chengdu, China) high-precision vacuum gauge for accurately detecting the actual vacuum inside the tank.

Joule Heating Metal Wire Theory Analysis
Throughout the entire experimental process, the maximum power does not exceed 0.5 kW, and the total weight of the experimental platform is 50 kg, occupying minimal space and facilitating operation.Compared to additive manufacturing methods using laser or arc as heat sources, the heat originates from the wire itself.In comparison to external heat sources, this not only saves energy but also reduces energy consumption.According to Joule's law, the total Joule heat Q produced by the system is calculated as follows: In the equation, i represents the current flowing through the wire, t 1 denotes the duration of current flow, R signifies the total resistance during the current flow process.R includes both the resistance of the wire itself and the contact resistance.
In this equation, R 1 represents the resistance of the metal wire, R 2 denotes the contact resistance between the metal wire and the roller, R 3 signifies the contact resistance between the metal wire and the substrate.Feulvarch [26] discovered that there is a close relationship between contact resistance and pressure, which is expressed by the following formula: In the equation, R c represents the contact resistance, R 0 denotes the contact resistance at zero pressure, P represents the contact pressure, α is a material-related coefficient ranging between 0.1 and 1.The specific value is influenced by factors such as the material surface treatment, pressure, temperature difference, etc.
In the equation, F represents the pressure applied to the wire by the chromiumzirconium-copper roller, and S denotes the contact area between the wire and the chromiumzirconium-copper roller.
During solidification, the crystalline morphology is influenced by the temperature gradient G at the solid-liquid interface and the growth rate R. The interface stability coefficient, represented by G/R.G/R is typically used to predict the crystalline morphology of the solidified structure.The cooling rate often used to predict the grain size of the solidified structure, is denoted by GR.A larger value of this coefficient indicates a smaller grain size, whereas a smaller value suggests larger grain sizes.The temperature gradient G and growth rate R can be expressed by the following equations: In the equation, G x , G y , G z represents the temperature gradient in the direction (x, y, z), v denotes the velocity, R indicates the direction of grain growth.The calculation formula for cosθ is as follows: In the equation, θ represents the angle between the grain growth direction and a specific crystal face normal, ranging from 0 to 90 degrees.It represents the deviation of the grain growth direction from a particular reference direction.

Establishment of Joule Heating Fused Filament Finite Element Model
To facilitate the understanding of the heat generation mechanism in Joule heating metal fused filament additive manufacturing [27] and to comprehend its impact on microstructure, we use COMSOL6.2 software for simulation.In the COMSOL, the dimensions of the substrate were reduced to 10 mm × 10 mm × 3 mm, and the chromium-zirconium-copper roller was simplified to a half-wheel with a radius of 6 mm and a thickness of 8 mm.This simplification aimed to facilitate the application of loads to the model.The dimensions of the metal wire were set to a diameter of 0.4 mm and a length of 10 mm.The model was assembled to ensure that the bottom surface of the roller was tangent to the surface of the wire.Meshing was performed with an overall mesh size of 0.2 mm, with the contact area appropriately refined to a minimum mesh size of 0.02 mm.The mesh is coarser in regions away from the wire, while the mesh is finer in the wire, at the contact surface between the wire and the substrate, and at the contact surface between the wire and the roller.This is mainly because the temperature gradient in these areas is significant, and a finer mesh is needed to accurately capture the temperature field changes.The chosen mesh type is tetrahedral because of its flexibility in handling irregular geometries.Tetrahedral meshes can adapt to the curves of the circular wire more easily and intuitively, ensuring the accuracy and convergence of the simulation results.The finite element model is depicted in Figure 4.
The total resistance R(t) in the system consists of four components: the bulk resistance R e of the roller, the bulk resistance R w of the wire/substrate, the contact resistance R ew between the roller and the wire, and the contact resistance R c between the wire and the substrate, represented as R(t) = R e + R ew + 2R w + R c .The specifics are illustrated in Figure 5.The total resistance () in the system consists of four components: the bulk resistance  of the roller, the bulk resistance  of the wire/substrate, the contact resistance  between the roller and the wire, and the contact resistance  between the wire and the substrate, represented as () =  +  + 2 +  .The specifics are illustrated in Figure 5.

Model Assumptions and Analysis Steps Configuration
Based on the constructed Joule heat metal wire additive manufacturing equipment, the model assumptions are as follows: (1) In the geometric model, material distribution is uniform, and the material is isotropic, following the principle of material yield.
(2) The contact types between the roller and the wire, and between the wire and the substrate are assumed to be smooth and continuous.(3) The vacuum environment is assumed to be an absolute vacuum, with no gas convection, hence the heat transfer occurs through both thermal conduction and thermal radiation.

Experimental Materials
The experiment employs 304 stainless steel wire with a diameter of 0.4 mm and 316 L stainless steel square substrate with a thickness of 3 mm and a side length of 100 mm.Given the high surface quality requirement for both wire and substrate, surface treatment is conducted prior to printing to ensure the repeatability of process parameters and minimize errors.The method is as follows: the substrate, which has been semi-finished through CNC milling, is polished using 1500-and 2000-grit sandpaper, then rinsed with water and sprayed with industrial alcohol.Afterward, the substrate and the drawn wire are wiped with oil-free paper and dried with a blow dryer to remove machining marks and dirt, ensuring that both surfaces are clean and free of impurities.Polishing and wiping  The total resistance () in the system consists of four components: the bulk resistance  of the roller, the bulk resistance  of the wire/substrate, the contact resistance  between the roller and the wire, and the contact resistance  between the wire and the substrate, represented as () =  +  + 2 +  .The specifics are illustrated in Figure 5.

Model Assumptions and Analysis Steps Configuration
Based on the constructed Joule heat metal wire additive manufacturing equipment, the model assumptions are as follows: (1) In the geometric model, material distribution is uniform, and the material is isotropic, following the principle of material yield.
(2) The contact types between the roller and the wire, and between the wire and the substrate are assumed to be smooth and continuous.(3) The vacuum environment is assumed to be an absolute vacuum, with no gas convection, hence the heat transfer occurs through both thermal conduction and thermal radiation.

Experimental Materials
The experiment employs 304 stainless steel wire with a diameter of 0.4 mm and 316 L stainless steel square substrate with a thickness of 3 mm and a side length of 100 mm.Given the high surface quality requirement for both wire and substrate, surface treatment is conducted prior to printing to ensure the repeatability of process parameters and minimize errors.The method is as follows: the substrate, which has been semi-finished through CNC milling, is polished using 1500-and 2000-grit sandpaper, then rinsed with water and sprayed with industrial alcohol.Afterward, the substrate and the drawn wire are wiped with oil-free paper and dried with a blow dryer to remove machining marks and dirt, ensuring that both surfaces are clean and free of impurities.Polishing and wiping

Model Assumptions and Analysis Steps Configuration
Based on the constructed Joule heat metal wire additive manufacturing equipment, the model assumptions are as follows: (1) In the geometric model, material distribution is uniform, and the material is isotropic, following the principle of material yield.
(2) The contact types between the roller and the wire, and between the wire and the substrate are assumed to be smooth and continuous.(3) The vacuum environment is assumed to be an absolute vacuum, with no gas convection, hence the heat transfer occurs through both thermal conduction and thermal radiation.

Experimental Materials
The experiment employs 304 stainless steel wire with a diameter of 0.4 mm and 316 L stainless steel square substrate with a thickness of 3 mm and a side length of 100 mm.Given the high surface quality requirement for both wire and substrate, surface treatment is conducted prior to printing to ensure the repeatability of process parameters and minimize errors.The method is as follows: the substrate, which has been semi-finished through CNC milling, is polished using 1500-and 2000-grit sandpaper, then rinsed with water and sprayed with industrial alcohol.Afterward, the substrate and the drawn wire are wiped with oil-free paper and dried with a blow dryer to remove machining marks and dirt, ensuring that both surfaces are clean and free of impurities.Polishing and wiping steps involve physical contact and friction to clean and smooth the surface, hence they are of mechanical nature.The chemical composition and physical properties of 304 stainless steel and 316 L are provided in Tables 2 and 3 [28].

Detection Method
The formed single-layer single-channel specimens are cut into blocks using wirecutting technology.After sectioning, the samples undergo rough grinding, fine grinding, and polishing.They are then subjected to surface corrosion using a prepared aqua regia solution (hydrochloric acid to nitric acid volume ratio of 3:1).After corrosion for 20 s, the samples are immediately rinsed with anhydrous ethanol and dried.Subsequently, the microstructure and cross-sectional morphology of the single-layer single-channel are observed using an optical microscope.The microstructure at five different positions, including the top, middle, and bottom of the single-layer single-channel, as well as the side and substrate, is observed using an AX10 optical microscope (ZEISS, Oberkochen, Germany).
Microhardness testing was conducted on the single-layer single-channel cross sections using the HM200 Vickers microhardness tester (Mitutoyo, Kanagawa, Japan).The samples were loaded with a force of 0.25 N for 15 s.Measurements were taken along the normal to the fusion line, with points sampled every 10 µm toward and away from the center of the wire.

Experiment Plan
During the experimental process, to minimize oxidation and ensure printing quality to the maximum extent, the experiment was conducted at an ambient temperature of 25 • C, humidity of 55%, and a vacuum pressure of 5 × 10 −2 Pa.The dry elongation of the wire was set at 15 mm, with the angle between the wire nozzle and the substrate set to 25 • .The roller cross-section width was 1.0 mm.Three process parameters, namely current intensity, printing speed, and roller pressure, were selected as variable factors.After preliminary research and continuous optimization of equipment [29], the final selected process parameters were as follows: current intensity ranged from 380 A to 420 A, printing speed ranged from 800 mm/min to 1200 mm/min, and roller pressure ranged from 0.2 N to 0.4 N. The specific process parameters are listed in Table 4.By analyzing macroscopic forming and microstructural changes, the optimal process parameters for single-layer single-channel within the selected window were determined.

Characteristics of Temperature Distribution Field
The simulation for the Joule heating fused filament additive manufacturing process used a 0.4 mm diameter 304 stainless steel wire.The chosen process parameters were: a current intensity of 400 A, a printing speed of 1000 mm/min, and a roller pressure of 0.3 N.
Figure 6 shows that the simulation process consists of four main stages: the current application stage (0-0.1 s), roller movement stage (0.1-0.9 s), current unloading stage (0.9-1.0 s), and cooling stage (1.0-2.0 s).At approximately 0.15 s, the temperature of the system tends to reach equilibrium, with the highest temperature exceeding the melting point (1450 • C), providing favorable temperature conditions for the metallurgical bonding between the metal wire and the substrate.At 1 s, the highest temperature decreases to 600 • C and gradually cools to room temperature (20 • C). Figure 7 shows that the calculated melting zone corresponds to the experimental results under the same process parameters.The simulation results exhibit slight differences compared to the experimental results, primarily in terms of the height and width of the melted area.This discrepancy may be due to inaccurate material property data and simplified experimental equipment in the simulation.Additionally, some assumptions were made during the simulation, such as neglecting the fluid behavior of the metal after melting.However, in general comparison, the simulation results are in good agreement with the experimental results, validating the reliability of the model and providing a theoretical basis for subsequent analysis of the heat generation mechanism's impact on microstructure.The calculated deviation is approximately 15.7%.
In Joule heating fused filament additive manufacturing, temperature distribution directly influences melting and solidification, determining the final microstructure.The temperature field results reveal that metallurgical bonding occurs between the wire and the substrate once the melting point is reached, leading to changes in the microstructure.Due to the presence of temperature gradients, variations in the size of the material's microstructure are observed.The heat generated by the current passing through resistance when the roller contacts the wire creates diverse temperature gradients, affecting grain size and distribution.Non-uniform temperature fields may result in localized overheating, uneven grain growth, and consequently, a reduction in material performance.Optimizing the temperature field can enhance the microstructure, reduce internal defects, and improve material properties.generation mechanism's impact on microstructure.The calculated deviation is approximately 15.7%.
In Joule heating fused filament additive manufacturing, temperature distribution directly influences melting and solidification, determining the final microstructure.The temperature field results reveal that metallurgical bonding occurs between the wire and the substrate once the melting point is reached, leading to changes in the microstructure.Due to the presence of temperature gradients, variations in the size of the material's microstructure are observed.The heat generated by the current passing through resistance when the roller contacts the wire creates diverse temperature gradients, affecting grain size and distribution.Non-uniform temperature fields may result in localized overheating, uneven grain growth, and consequently, a reduction in material performance.Optimizing the temperature field can enhance the microstructure, reduce internal defects, and improve material properties.rectly influences melting and solidification, determining the final microstructure.The temperature field results reveal that metallurgical bonding occurs between the wire and the substrate once the melting point is reached, leading to changes in the microstructure.Due to the presence of temperature gradients, variations in the size of the material's microstructure are observed.The heat generated by the current passing through resistance when the roller contacts the wire creates diverse temperature gradients, affecting grain size and distribution.Non-uniform temperature fields may result in localized overheating, uneven grain growth, and consequently, a reduction in material performance.Optimizing the temperature field can enhance the microstructure, reduce internal defects, and improve material properties.

Metallographic Analysis
Figure 8 illustrates the cross-sectional view of a single-layer single-channel.From Figure 8a, it is evident that the single-layer single-channel cross section mainly comprises melt width (W), melt height (H), melt depth (D), and total melt height (Hw).The singlelayer single-channel cross section consists of a forming layer, fusion zone, heat-affected zone, and substrate.Tables 5-7 show the macroscopic surface morphology under different process parameters.As indicated in Table 5, for wire No. 1, when the current intensity reaches 380 A, the lower current intensity results in more Joule heat being utilized for wire melting, resulting in a raised morphology.Conversely, for wire No. 3, with a current intensity of 420 A, the melt height decreases while the melt width increases, presenting a flattened morphology.Table 6 shows that with the increase in printing speed, the melt width of the wire decreases.This is because higher printing speeds reduce heat input per unit Crystals 2024, 14, 643 10 of 17 length, leading to insufficient wire spreading and decreased melt width.From Table 7, it can be observed that when the roller pressure is 0.2 N, the melt width of the wire decreases, and the contact with the substrate also decreases, resulting in lesser fusion with the substrate.When the roller pressure is 0.4 N, the effective contact area between the roller and the wire increases, causing a decrease in current density and a more dispersed distribution of current lines, resulting in reduced Joule heating and insufficient heat input.Additionally, the surface morphology becomes rougher with increasing pressure.By comparing the experimental results under different process parameters, it was found that the optimal morphology for single-layer single-channel formation, with a smooth surface, was achieved at a current intensity of 400 A, a printing speed of 1000 mm/min, and a roller pressure of 0.3 N. At these parameters, the characteristic parameters of the cross section, namely the width-to-height ratio (W/H) and dilution rate (η), were 3.23 and 51.61%, respectively.Continuous compression contact between the roller and wire made the wire surface rougher at lower printing speeds.Conversely, at higher printing speeds, the wire surface became smoother, albeit with reduced melt depth and width.Hence, printing speed is the second most important parameter primary in controlling wire formation morphology.Current intensity is the most critical factor influencing the morphology of the wire.The primary reason is that variations in current intensity affect the Joule heating effect, determining the degree of wire melting.Changes in current intensity significantly impact the wire's diameter and smoothness.process parameters.As indicated in Table 5, for wire No. 1, when the current intensity reaches 380 A, the lower current intensity results in more Joule heat being utilized for wire melting, resulting in a raised morphology.Conversely, for wire No. 3, with a current intensity of 420 A, the melt height decreases while the melt width increases, presenting a flattened morphology.Table 6 shows that with the increase in printing speed, the melt width of the wire decreases.This is because higher printing speeds reduce heat input per unit length, leading to insufficient wire spreading and decreased melt width.From Table 7, it can be observed that when the roller pressure is 0.2 N, the melt width of the wire decreases, and the contact with the substrate also decreases, resulting in lesser fusion with the substrate.When the roller pressure is 0.4 N, the effective contact area between the roller and the wire increases, causing a decrease in current density and a more dispersed distribution of current lines, resulting in reduced Joule heating and insufficient heat input.Additionally, the surface morphology becomes rougher with increasing pressure.By comparing the experimental results under different process parameters, it was found that the optimal morphology for single-layer single-channel formation, with a smooth surface, was achieved at a current intensity of 400 A, a printing speed of 1000 mm/min, and a roller pressure of 0.3 N. At these parameters, the characteristic parameters of the cross section, namely the width-to-height ratio (W/H) and dilution rate (η), were 3.23 and 51.61%, respectively.Continuous compression contact between the roller and wire made the wire surface rougher at lower printing speeds.Conversely, at higher printing speeds, the wire surface became smoother, albeit with reduced melt depth and width.Hence, printing speed is the second most important parameter primary in controlling wire formation morphology.Current intensity is the most critical factor influencing the morphology of the wire.The primary reason is that variations in current intensity affect the Joule heating effect, determining the degree of wire melting.Changes in current intensity significantly impact the wire's diameter and smoothness.

Microstructural Analysis
The original structure of the wire and substrate consists of equiaxed austenite grains, but there are differences in the shape and size of the wire and substrate grains.This variation arises from the different manufacturing processes employed for the wire (cold drawing) and substrate (forging), which can result in variations in grain size and shape.During the process of Joule heating metal wire additive manufacturing, as the heat input increases, some changes occur in the microstructure of the wire.The initial structure of the wire consists of austenite grains, which undergo processes such as heating, cooling, and solidification, transforming into a sequence of planar grains, columnar grains, equiaxed grains, and columnar grains from the fusion line to the top of the wire [20].As shown in Figures 9 and 10, the grains grow from the surface of the wire toward the center due to the large temperature gradient at the wire center and because the optimal crystallization direction aligns with the direction of the temperature gradient.Figure 10 depicts the actual EBSD image of a single-layer single-channel cross section.

Microstructural Analysis
The original structure of the wire and substrate consists of equiaxed austenite grains, but there are differences in the shape and size of the wire and substrate grains.This variation arises from the different manufacturing processes employed for the wire (cold drawing) and substrate (forging), which can result in variations in grain size and shape.During the process of Joule heating metal wire additive manufacturing, as the heat input increases, some changes occur in the microstructure of the wire.The initial structure of the wire consists of austenite grains, which undergo processes such as heating, cooling, and solidification, transforming into a sequence of planar grains, columnar grains, equiaxed grains, and columnar grains from the fusion line to the top of the wire [20].As shown in Figures 9 and 10, the grains grow from the surface of the wire toward the center due to the large temperature gradient at the wire center and because the optimal crystallization direction aligns with the direction of the temperature gradient.Figure 10 depicts the actual EBSD image of a single-layer single-channel cross section.

Microstructural Analysis
The original structure of the wire and substrate consists of equiaxed austenite grains, but there are differences in the shape and size of the wire and substrate grains.This variation arises from the different manufacturing processes employed for the wire (cold drawing) and substrate (forging), which can result in variations in grain size and shape.During the process of Joule heating metal wire additive manufacturing, as the heat input increases, some changes occur in the microstructure of the wire.The initial structure of the wire consists of austenite grains, which undergo processes such as heating, cooling, and solidification, transforming into a sequence of planar grains, columnar grains, equiaxed grains, and columnar grains from the fusion line to the top of the wire [20].As shown in Figures 9 and 10, the grains grow from the surface of the wire toward the center due to the large temperature gradient at the wire center and because the optimal crystallization direction aligns with the direction of the temperature gradient.Figure 10 depicts the actual EBSD image of a single-layer single-channel cross section.

Microstructural Analysis
The original structure of the wire and substrate consists of equiaxed austenite grains, but there are differences in the shape and size of the wire and substrate grains.This variation arises from the different manufacturing processes employed for the wire (cold drawing) and substrate (forging), which can result in variations in grain size and shape.During the process of Joule heating metal wire additive manufacturing, as the heat input increases, some changes occur in the microstructure of the wire.The initial structure of the wire consists of austenite grains, which undergo processes such as heating, cooling, and solidification, transforming into a sequence of planar grains, columnar grains, equiaxed grains, and columnar grains from the fusion line to the top of the wire [20].As shown in Figures 9 and 10, the grains grow from the surface of the wire toward the center due to the large temperature gradient at the wire center and because the optimal crystallization direction aligns with the direction of the temperature gradient.Figure 10 depicts the actual EBSD image of a single-layer single-channel cross section.

Microstructural Analysis
The original structure of the wire and substrate consists of equiaxed austenite grains, but there are differences in the shape and size of the wire and substrate grains.This variation arises from the different manufacturing processes employed for the wire (cold drawing) and substrate (forging), which can result in variations in grain size and shape.During the process of Joule heating metal wire additive manufacturing, as the heat input increases, some changes occur in the microstructure of the wire.The initial structure of the wire consists of austenite grains, which undergo processes such as heating, cooling, and solidification, transforming into a sequence of planar grains, columnar grains, equiaxed grains, and columnar grains from the fusion line to the top of the wire [20].As shown in Figures 9 and 10, the grains grow from the surface of the wire toward the center due to the large temperature gradient at the wire center and because the optimal crystallization direction aligns with the direction of the temperature gradient.Figure 10 depicts the actual EBSD image of a single-layer single-channel cross section.
solidification, transforming into a sequence of planar grains, columnar grains, equiaxed grains, and columnar grains from the fusion line to the top of the wire [20].As shown in Figures 9 and 10, the grains grow from the surface of the wire toward the center due to the large temperature gradient at the wire center and because the optimal crystallization direction aligns with the direction of the temperature gradient.Figure 10 depicts the actual EBSD image of a single-layer single-channel cross section.solidification, transforming into a sequence of planar grains, columnar grains, equiaxed grains, and columnar grains from the fusion line to the top of the wire [20].As shown in Figures 9 and 10, the grains grow from the surface of the wire toward the center due to the large temperature gradient at the wire center and because the optimal crystallization direction aligns with the direction of the temperature gradient.Figure 10 depicts the actual EBSD image of a single-layer single-channel cross section.Following the metallographic sample preparation procedure, metallographic samples were prepared for the selected process parameters.Optical microscopy was used to observe the microstructure of different regions of the specimens.Figure 11 shows the microstructure of sample 2, including the fusion zone, middle, top, sidewall of the wire, and substrate under the process parameters 400 A current intensity, 1000 mm/min printing speed, and 0.3 N roller pressure.Since austenite is more corrosion-resistant than ferrite, ferrite is preferentially corroded.As shown in Figure 11, the grains grow along the normal direction of the fusion line toward the center of the wire, as the optimal crystallization orientation aligns with the direction of the temperature gradient.In the melted area of the wire, clear temperature gradient lines can be observed.This is due to the difference in grain nucleation and growth rates.The grains gradually become denser from the fusion zone to the center because the grains in the fusion zone have sufficient time to grow, while the central area experiences a large temperature gradient and short cooling time, resulting in rapid solidification of grains before they can grow significantly.
Figures 12-14 depict the microstructures of the fusion zone under different process parameters.It can be observed that there are partial differences in the size and distribution of the wire's microstructure, indicating that the process parameters of Joule heating metal wire additive manufacturing have a significant influence on the microstructure.In the fusion zone, ferrite is mainly densely distributed, while above the fusion zone, there is a distribution of bright austenite.The grain size in the fusion zone is smaller compared to other areas.This is because, in the JHAM process, the wire undergoes heating and cooling, and the heat generated on the wire surface in contact with the substrate is quickly conducted from the substrate.The grains solidify before they have a chance to grow significantly.However, larger temperature gradients and longer cooling times allow grains to grow more in other areas.Due to the large temperature gradient at the top of the wire, grain growth is most favorable, resulting in an average grain size of 50 µm at the wire top, 10 µm in the middle, and 20 µm in the fusion zone.
wire additive manufacturing have a significant influence on the microstructure.In the fusion zone, ferrite is mainly densely distributed, while above the fusion zone, there is a distribution of bright austenite.The grain size in the fusion zone is smaller compared to other areas.This is because, in the JHAM process, the wire undergoes heating and cooling, and the heat generated on the wire surface in contact with the substrate is quickly conducted from the substrate.The grains solidify before they have a chance to grow significantly.However, larger temperature gradients and longer cooling times allow grains to grow more in other areas.Due to the large temperature gradient at the top of the wire, grain growth is most favorable, resulting in an average grain size of 50 µm at the wire top, 10 µm in the middle, and 20 µm in the fusion zone.

Model Assumptions and Analysis Steps Configuration
Figure 12 illustrates the microstructure of the fusion zone under different current intensities.It can be observed that when the current intensity is low, specifically at 380 A, the fusion zone exhibits finer and less uniform microstructure.As the current intensity increases, reaching 420 A, the microstructure of the fusion zone gradually becomes coarser and more concentrated.This phenomenon primarily arises due to the consistent parameters except for current intensity.At lower current intensities, the reduced Joule heating generated by the contact between the roller and the wire leads to decreased thermal input during the additive manufacturing process.This results in a smaller temperature gradient, longer cooling time, reduced solidification rate, and subsequently slower grain growth, leading to finer grains.Conversely, at higher current intensities, the increased Joule heating from the contact between the roller and the wire amplifies the thermal input during the additive manufacturing process.Consequently, a larger temperature gradient, shorter cooling time, increased solidification rate, and coarser grain formation occur.

The Influence of Printing Speed on Microstructure
Figure 13 presents the microstructure of the fusion zone under different printing speeds.It is observed that at lower printing speeds, specifically at 800 mm/min, the fusion zone exhibits coarser microstructure.As the printing speed increases to a certain value, namely 1000 mm/min, the microstructure of the fusion zone becomes finer.This phenomenon arises due to consistent parameters except for printing speed.At lower printing speeds, the thermal input per unit length of the wire increases, resulting in a larger temperature gradient, shorter cooling time, and increased solidification rate, leading to coarser grains.When the printing speed reaches a certain threshold, the thermal input per unit length of the wire decreases, causing a smaller temperature gradient, longer cooling time, and reduced solidification rate, resulting in finer grains.With further increases in printing speed beyond a certain threshold, the thermal input per unit length of the wire decreases even more, leading to a smaller temperature gradient and longer cooling time, thus slowing down the grain growth rate and resulting in finer grains.

The Influence of Roller Pressure on Microstructure
Figure 14 illustrates the microstructure of the fusion zone under different roller pressures.It is observed that at lower roller pressures, specifically at 0.2 N, the fusion zone exhibits a coarser microstructure.As the roller pressure increases to a certain value, namely 0.3 N, the microstructure of the fusion zone becomes finer.This phenomenon arises due to consistent parameters except for the roller pressure.At lower roller pressures, the effective contact area between the roller and the wire is smaller, resulting in increased current density, higher thermal input, larger temperature gradient, shorter cooling time, and increased solidification rate, leading to coarser grain growth.When the roller pressure increases to a certain value, the effective contact area between the roller and the wire gradually increases, causing a decrease in current density and thermal input.This results in a smaller temperature gradient, longer cooling time, and increased solidification rate, leading to the growth of finer grains.With further increases in roller pressure, the effective contact area between the roller and the wire continues to increase, causing a decrease in current density, a more dispersed distribution of current lines, and a decrease in total resistance.This leads to reduced Joule heating, insufficient thermal input, longer cooling time, and slower grain growth, resulting in finer grains.

Microhardness Analysis
According to Figure 15, it can be observed that the microhardness values in the fusion zone are higher than those in the base metal and other areas of the wire cross section for all process parameters.With variations in different process parameters, the microhardness of the single-channel forming layer exhibits a sawtooth-like fluctuation.This phenomenon mainly arises due to grain refinement and uneven, irregular distribution.The highest hardness near the fusion line is primarily because of the denser distribution of ferrite, which also contributes to the increased hardness.

Microhardness Analysis
According to Figure 15, it can be observed that the microhardness values in the fusion zone are higher than those in the base metal and other areas of the wire cross section for all process parameters.With variations in different process parameters, the microhardness of the single-channel forming layer exhibits a sawtooth-like fluctuation.This phenomenon mainly arises due to grain refinement and uneven, irregular distribution.The highest hardness near the fusion line is primarily because of the denser distribution of ferrite, which also contributes to the increased hardness.Figure 15a shows the microhardness curves at different current intensities.With increasing current intensity, the microhardness first increases and then decreases.This trend occurs mainly within a certain range of current intensity (380-400 A).With the increase in heat input, the cooling time decreases, and the system temperature gradually reaches thermal equilibrium, resulting in finer and more uniform grains, thus gradually increasing the hardness.However, when the current intensity increases to 420 A, the heat input increases, the cooling time decreases, and the system's heat input gradually increases, causing the system temperature to exceed thermal equilibrium, leading to coarser grains and hence decreased hardness.
Figure 15b depicts the microhardness curves at different printing speeds.With increasing printing speed, the microhardness first increases and then decreases.This behavior mainly occurs within a certain range of printing speeds (800-1000 mm/min).As the heat input per unit length of wire gradually decreases, the temperature gradient decreases, the cooling time increases, and the solidification rate decreases.Consequently, the system gradually reaches thermal equilibrium, and the organization of the single-channel forming layer gradually becomes finer and more uniform, resulting in higher hardness.However, when the printing speed increases to 1200 mm/min, the heat input per unit length of wire decreases, the cooling time increases, but the faster printing speed causes uneven temperature gradient changes, varying growth rates of grains in different regions, leading to unevenness in grain size and thereby reducing the hardness.
Figure 15c shows the microhardness curves at different roller pressures.With increasing roller pressure, the microhardness first increases and then decreases.This phenomenon mainly occurs within a certain range of roller pressures (0.2-0.3 N).A higher pressure leads to a larger effective contact area between the roller and the wire, gradually reducing the heat input, lengthening the cooling time, and gradually reaching thermal equilibrium.Consequently, the grains are relatively finer and more uniform, resulting in higher hardness.However, when the roller pressure increases to 0.4 N, insufficient heat input leads to excessive cooling of the wire, uneven grain growth, and lower hardness.

Conclusions
(1) Effect of Current Intensity: As the current intensity increases, the microstructure in the fusion zone of the single−channel forming layer transitions from fine to coarse, and the microhardness first increases and then decreases.Specifically, within the current intensity range of 380 A to 420 A, the changes in microstructure and microhardness are significant, with the optimal current intensity being 400 A. (2) Effect of Printing Speed: As the printing speed increases, the microstructure in the fusion zone of the single−channel forming layer initially becomes finer, then coarser, and the microhardness exhibits a similar trend of first increasing and then decreasing.

Figure 3 .
Figure 3. Joule Heating Metal Wire Additive Manufacturing Experimental Platform: (a) Experimental Equipment; (b) Experimental Printing Area and Chromium-Zirconium-Copper Wheel Model; (c) The chromium-zirconium-copper roller model.

Figure 4 .
Figure 4. Finite Element Model of Joule Heating Metal Wire Additive Manufacturing.

Figure 5 .
Figure 5. Composition of total resistance in single-layer single-channel system.

Figure 4 .
Figure 4. Finite Element Model of Joule Heating Metal Wire Additive Manufacturing.

Figure 4 .
Figure 4. Finite Element Model of Joule Heating Metal Wire Additive Manufacturing.

Figure 5 .
Figure 5. Composition of total resistance in single-layer single-channel system.

Figure 5 .
Figure 5. Composition of total resistance in single-layer single-channel system.

Figure 6 .
Figure 6.Temperature variation of the entire process system for single-layer single-channel wire material: (a) System temperature change data diagram (b) System flow chart of the temperature change different time.

Figure 6 .
Figure 6.Temperature variation of the entire process system for single-layer single-channel wire material: (a) System temperature change data diagram (b) System flow chart of the temperature change different time.

Figure 6 .
Figure 6.Temperature variation of the entire process system for single-layer single-channel wire material: (a) System temperature change data diagram (b) System flow chart of the temperature change different time.

Figure 9 .
Figure 9.The formation process of microstructure.Figure 9.The formation process of microstructure.

Figure 9 .
Figure 9.The formation process of microstructure.Figure 9.The formation process of microstructure.

Figure 9 .
Figure 9.The formation process of microstructure.

Figure 11 .
Figure 11.Single-channel overall and local microstructure: (a) Top of filament; (b) Side of filament; (c) Middle of filament; (d) Substrate; (e) Melting zone of filament.

Figure 14 .Figure 12 . 18 Figure 12 .
Figure 14.The microstructure of the fusion zone under different roller pressure: (a) 0.2 N; (b) 0.3 N; (c) 0.4 N. 3.3.1.Model Assumptions and Analysis Steps Configuration Figure 12 illustrates the microstructure of the fusion zone under different current in-

Figure 14 .
Figure 14.The microstructure of the fusion zone under different roller pressure: (a) 0.2 N; (b) 0.3 N; (c) 0.4 N. 3.3.1.Model Assumptions and Analysis Steps Configuration Figure 12 illustrates the microstructure of the fusion zone under different current in-

Figure 14 .
Figure 14.The microstructure of the fusion zone under different roller pressure: (a) 0.2 N; (b) 0.3 N; (c) 0.4 N. 3.3.1.Model Assumptions and Analysis Steps Configuration Figure 12 illustrates the microstructure of the fusion zone under different current intensities.It can be observed that when the current intensity is low, specifically at 380 A,

Figure 15 .
Figure 15.The hardness distribution of single−layer single−channel cross section under different process parameters: (a) Microhardness curves under different current intensities; (b) Microhardness curves under different printing speeds; (c) Microhardness curves under different roller pressures.

Figure 15 .
Figure 15.The hardness distribution of single−layer single−channel cross section under different process parameters: (a) Microhardness curves under different current intensities; (b) Microhardness curves under different printing speeds; (c) Microhardness curves under different roller pressures.

Table 2 .
The chemical composition of 304 stainless steel and 316 L (Mass fraction/%).

Table 5 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 5 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.Number Speed/mm/mi n Macroscopic Morphology Cross-Sectional Morphology Crystals 2024, 14, x FOR PEER REVIEW 11 of 18

Table 5 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.Number Speed/mm/mi n Macroscopic Morphology Cross-Sectional Morphology Crystals 2024, 14, x FOR PEER REVIEW 11 of 18

Table 5 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.Number Speed/mm/mi n Macroscopic Morphology Cross-Sectional Morphology Crystals 2024, 14, x FOR PEER REVIEW 11 of 18

Table 5 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.Number Speed/mm/mi n Macroscopic Morphology Cross-Sectional Morphology Crystals 2024, 14, x FOR PEER REVIEW 11 of 18

Table 5 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.Number Speed/mm/mi n Macroscopic Morphology Cross-Sectional Morphology Crystals 2024, 14, x FOR PEER REVIEW 11 of 18

Table 5 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.NumberSpeed/mm/mi n Macroscopic Morphology Cross-Sectional Morphology

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 7 .
Macroscopic morphology and cross-sectional morphology under different roller pressures.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 7 .
Macroscopic morphology and cross-sectional morphology under different roller pressures.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 7 .
Macroscopic morphology and cross-sectional morphology under different roller pressures.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 7 .
Macroscopic morphology and cross-sectional morphology under different roller pressures.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 7 .
Macroscopic morphology and cross-sectional morphology under different roller pressures.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 7 .
Macroscopic morphology and cross-sectional morphology under different roller pressures.

Table 7 .
Macroscopic morphology and cross-sectional morphology under different roller pressures.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 7 .
Macroscopic morphology and cross-sectional morphology under different roller pressures.

Table 6 .
Macroscopic morphology and cross-sectional morphology under different current intensities.

Table 7 .
Macroscopic morphology and cross-sectional morphology under different roller pressures.