Microstructure and Mechanical Properties of a Weld Seam from Magnetron High-Current CO2 Welding

Abstract

External magnetic field (EMF)-assisted high-current CO₂ welding is beneficial for improving the large spatter and poor performance of the welding heat-affected zone for mild steels under high-current welding specifications. In this paper, the droplet transfer behaviors were determined using a high-speed camera on a self-developed magnetically controlled CO₂ welding system. Based on these welding specifications, a three-dimensional, transient, multi-energy field coupling welding system model to investigate the mechanism of the droplet and molten pool in EMF-assisted welding was developed. The microstructure and mechanical properties of the welded joint were systematically studied. The results show that the Lorentz force applied by the EMF to twist the droplet decreases the accumulated energy in the short-circuited liquid bridge and changes the liquid metal flow condition, both of which reduce the spatter by 7% but increase the welded joint hardness by 10% and tensile strength by 8%.

1. Introduction

CO₂ gas shielded welding (CO₂ welding) has the advantages of a low welding cost, high productivity, and wide range of application. However, the drawbacks, such as high current spatter and poor weld formation, limit its wide application [1,2]. According to electro-explosive theory, spatter occurs mostly in the short-circuit stage. The droplet contacts the molten pool to form a short-circuit liquid bridge. Meanwhile, the short-circuit current rises rapidly, which leads to a sharp increase in the electromagnetic shrinkage force [3,4]. Therefore, the droplets that have not completely transferred are repelled out of the molten pool by the instantaneously increasing electromagnetic shrinkage force, which lead to the formation of the spatter [5,6,7,8].

To solve the aforementioned drawbacks, the method of EMF control has been proposed [9,10,11,12,13,14]. The mechanism of the EMF on the droplet transfer frequency in CO₂ welding was studied by Jiang et al. [15,16]. It was found that the EMF produces a radial inward electromagnetic force to accelerate the transition of droplets, which inhibits the generation of electric explosion spatter. Welding is more stable, and the quality is significantly enhanced. Hua et al. [17] studied magnetron metal active gas (MAG) welding using a high-speed camera. The results showed that after applying a longitudinal magnetic field (LMF), the relatively stable cone arc is transformed into a spiral cone under the EMF and rotates at a high speed, resulting in a droplet trajectory that departs from the center of the wire. This phenomenon is more obvious at a large magnetic field intensity, which can accelerate the droplet frequency and inhibit the weld finger depth phenomenon. An EMF was used by Wang et al. [18,19] to alter the anode spot size in gas metal arc welding (GMAW). The temperature gradient is effectively reduced by the change in the arc morphology. The surface tension and flow of molten metal were altered, which avoids welding defects and enhances the joint strength.

The fast development in the computer area has brought about great breakthroughs in numerical simulation technology, which makes up for the inadequacy of the ordinary external observation method that is unable to observe and explain the internal mechanism. A three-dimensional numerical model was developed by Cho et al. [20]. The model uses the volume of fluid (VOF) method to track the liquid metal surface in accordance with conservation equations of energy, momentum, and mass, which allows for the visualization of the changes in the morphology as well as in the flow and velocity fields. Ogino et al. [21,22,23] developed an “arc-droplet” coupling model to analyze the welding arc and droplet under pure Ar and Ar + 18% CO₂ shielding gas. It was found that the arc plasma morphology is more contracted under Ar + CO₂ shielding gas than under pure Ar shielding due to the higher specific heat. The increase in arc pressure with mixed shielding gas prevents the droplets from falling off, while pure Ar shielding results in droplet transfer. Xiao et al. [24,25] analyzed the influence of an EMF on the heat and mass transfer behaviors of GMAW using Fluent calculations. It was found that the Lorentz force generated by the LMF can divert arc plasma and metal vapor away from the wire axis, which leads to the contraction of the arc and an increase in both the peak temperature and current density. The coupled modeling of the welding arc and droplet under an EMF was established by Wang et al. [8]. The three-dimensional liquid metal flow and temperature of the GMAW butt weld under an EMF were simulated. The impact and heat input from the arc and the droplet were also considered, which can visualize and evaluate the free surface of the pool.

With the comprehensive research results of the above scholars, it is not difficult to notice that most of the experiments and simulations at the present stage are focused on GMAW, and there are fewer studies on the numerical simulation of the short-circuit phase of CO₂ welding. In fact, due to the specificity of CO₂, its short-circuit transfer is unique and completely different from general GMAW. It is difficult to explain the magnetically controlled droplet transfer mechanism. In this paper, the mechanism of the EMF acting on the droplet and molten pool in short-circuit transfer was studied using a combination of numerical simulation and experimentation. On this basis, the microstructure and mechanical properties of the welded joint were systematically investigated.

2. Materials and Methods

2.1. Experimental Setup

The experimental analysis of the short-circuit transfer process of EMF-assisted high-current CO₂ welding was carried out on the EMF-assisted welding system designed by the platform of the Liaoning Key Laboratory of Advanced Welding Technology, as shown in Figure 1. The experimental platform is composed of three parts: a welding system, magnetic field generator, and high-speed camera acquisition system. Among them, the welding system mostly consists of a Fronius TPS 500i welding power source and wire feeder. The magnetic field generation device is mainly for the magnetic head, MCWE-20/20 coupled excitation power supply, and cooling water tank. The magnetic field generator is made up of three components, which stabilizes the output magnetic field during the welding process. As it is difficult to clearly capture the complete droplet transfer process using conventional camera equipment, which is attributed to strong arc light, metal vapor, and spatter during the welding process, a Thousand Eyes Wolf X113 high-speed camera (Zhongke Junda Shijie Technology Co., Ltd., Hefei, China) and laser backlighter were used to form an acquisition system in this experiment to capture and analyze the droplet transfer frequency, transition morphology, and transition mode during the welding process. The welding current, welding voltage, welding speed, protective gas flow, wire feed speed and elongation of wire are 240 A, 26.4 V, 8 mm/s, 18 L/min, 7.4 m/min and 12 mm, respectively.

2.2. Sample Preparation

The size of Q235B mild steel is 200 mm × 100 mm × 6 mm, and an ER50-6 wire with a diameter of 1.2 mm is used as the filler material for the weld in accordance with the principle of equal-strength matching. The chemical compositions are shown in Table 1.

The weld was studied with different parameters. Firstly, the welds were cut into metallographic specimens, and the specimens were polished after sanding with different types of sandpaper. Secondly, the weld cross-section was corroded by using a nitric acid alcohol solution at a concentration of 4%. Finally, the weld cross-section was imaged using a body microscope. The weld depth, width, and residual height were measured using a scale for each parameter. The obtained data can be linked to the melt droplet and pool fluid in Chapter 3 to analyze the law of action of the LMF on the molten pool. The characteristic image of the weld and the measurement positions of the melt depth, width, and height are shown schematically in Figure 2.

2.3. Microstructure Characterization

To study the microstructure of the joints under different parameters, the weld microstructure was studied and analyzed using an MC-4XC trinocular inverted metallurgical microscope and Hitachi S-3400N scanning electron microscope (Mingbo Environmental Protection Technology Co., Ltd., Qingdao, China). The weld was wire cut to the designed specimen size, and the specimens were ground, polished, and corroded. Images of the weld microstructure were captured using a light microscope, and high-magnification microstructure observation was performed using a scanning electron microscope to study the impact of the EMF on weld organization and to establish a link with the weld properties. An EBSD test system was undertaken to collect data information such as the grain morphology, crystal orientation, grain boundary characteristics, and object identification of the samples, and the subsequent experimental data were processed and analyzed on OIM 7 and AZtecCrystal 2.1 software.

2.4. Mechanical Property

The hardness of the welds at different positions under different magnetic field parameters is tested with an MHVS-1A Vickers hardness tester (Mice Technology Co., Ltd., Shanghai, China), which can be used with numerical simulation and microstructure analysis to explain the effect of the EMF on welds more systematically. Tensile parts perpendicular to the weld are prepared, and tensile tests are carried out using a WDW-100 tensile testing machine (Limei Electromechanical Technology Co., Ltd., Jinan, China) to evaluate the mechanical properties of the weld.

2.5. Numerical Simulation

• (1). Governing equations

To improve the computational efficiency and save costs, it is necessary to simplify the model reasonably while highlighting the main research points. Therefore, the following simplifications and assumptions are made when establishing the droplet–molten pool coupling model [26,27,28,29,30,31,32]:

• (a). The model calculation area is divided into two parts: metal and gas. The liquid metal is regarded as an ideal Newtonian fluid, which is laminar and incompressible.

• (b). The composition of the welding wire used in the experiment is close to that of the base metal, which is considered to have the same thermophysical properties, and the thermophysical properties of the material are only related to the temperature.

• (c). The Boussineaq hypothesis is used to deal with the buoyancy and density changes of liquid metal in welding, and the effects of solid phase transformation and metal evaporation on the volume and temperature of the pool are ignored.

The VOF equation is employed to track the free surface of the molten pool. These equations for the conservation of mass, momentum, and energy are undertaken to describe the heat transfer and melt flow during the welding process [33,34,35,36].

• (2). Heat source modeling

In simulation, it is essential to consider the relationship between the arc, droplet, and molten pool. Considering the influence of the movement of the welding torch when the arc heats the base metal, the double-ellipsoidal heat source model is used to represent the heating effect of the arc on the base metal. At the same time, due to the existence of a certain enthalpy of the droplet, it also needs to be considered in the calculation process. Therefore, the double-ellipsoidal heat source + droplet transfer model was used to simulate the CO₂ welding.

In this paper, the numerical simulation of droplet transfer is realized by adding a mass source term, and the static force balance theory is adopted; that is, different forces act on the droplet transfer together. The focus is on the flow behavior of the droplet in contact with the molten pool and after its entry. To simplify the calculation model, special treatment is needed for the droplet process, which is set as the liquid metal flowing out from the circular region determined above the molten pool with a certain velocity, forming a molten droplet after the outflow and continuing to move downward to the molten pool to complete the transition. The boundary condition of this area is set as the velocity inlet. The droplet model is shown in Figure 3.

The FLUENT 21.0 software package, which is widely recognized for its ability to simulate complex fluid dynamics and heat transfer processes, was used to solve the conjugate thermodynamics problem in this study. It has a powerful solver and an extensive library of physical models that allow us to accurately simulate droplet transfer and liquid metal flow behavior under the influence of a longitudinal magnetic field (LMF).

Several requirements are imposed to ensure the convergence of the numerical results. First, a refined mesh is used to capture the detailed flow pattern and temperature distribution inside the molten pool. Second, an iterative solution method is employed using a convergence criterion that is set based on the residuals of the control equations. In addition, the effects of various numerical parameters (e.g., grid size and time step) on the solution are evaluated to further improve the accuracy of the results.

3. Results and Discussion

3.1. Droplet Transfer

High-speed photographs of the welding process without and with magnetism are illustrated in Figure 4. Under the same exposure conditions, the intervention of the EMF causes the arc waist position to shrink inward and the radius to decrease. The overall arc morphology shows a downward pressure phenomenon, and the high-temperature region of the arc is more concentrated.

The weld depth of fusion, width of fusion, and residual height under different magnetic field parameters were measured and summarized. The variation in the weld size under different parameters is shown in Figure 5. The width is enhanced with the introduction of the LMF, with an increase of 10.30% compared to the without-magnetic-field condition. A comparison with the baseline depth of the weld without the EMF shows that the weld depth is reduced by 10.65% after the introduction of the EMF. The residual height was reduced by 13.73%. The weld cross-section is characterized by a wide and shallow morphology.

3.2. The Temperature Distribution and Flow Behavior

Based on the above experimental results, the numerical simulation is performed to explain the differences in the droplet and molten pool with and without the EMF. To verify the accuracy of the simulation results, it is essential to compare the simulation with actual welding results. As shown in Figure 6, it can be seen that the measured value of the penetration ratio obtained using numerical simulation is slightly shallow, and the measured value of the width and height ratio is slightly smaller. The error between the melting width of the actual and simulated welds is about 4%, the error in the residual height is about 7%, and the melting depth error is about 9%. Therefore, it can be considered that the laws obtained using numerical simulation are basically consistent with the actual weld morphology, which confirms the accuracy of the numerical simulation.

During EMF-assisted CO₂ welding, the magnetic field affects the direction and magnitude of the forces acting in the droplet and molten pool. Therefore, the temperature and flow field changes before and after the application of the LMF were simulated by the established droplet–molten pool model, as shown in Figure 7. Assuming that the starting time of welding is t₀, when t = t₀ + 0.05 s, the liquid metal moves to the edge. The Lorentz force generated by the intervention of the EMF in conjunction with the current in the molten pool encourages the flow of liquid metal, which results in a reduction in the depression on the surface compared to the absence of the EMF. At t = t₀ + 0.30 s, the short circuit ends. The liquid metal flow velocity at the neck is faster, and the radial velocity component is stronger, which leads to the enhancement of liquid metal mobility. Before the droplet comes into contact with the pool, it is subjected to an LMF using a mechanism similar to that of the electric arc. The presence of a radial component of the current inside the droplet causes the liquid metal inside it to be rotated by the Lorentz force of the EMF. The presence of a radial fraction of the current inside the droplet causes the liquid metal inside it to be rotated by the EMF. Meanwhile, the arc plasma in the LMF will also be around the central axis of the high-speed spiral motion, which further promotes the droplet liquid metal flow. Under the EMF, the droplet morphology is spherical or ellipsoidal. The shortening of the ignition time means that the control of the EMF speeds up the transfer of the droplet. The accelerated transfer frequency of the droplets within the same time period means that the controllability of each overshoot is improved, and the possibility of spattering is reduced. According to the measurements, the spatter is reduced by about 7%.

Figure 7c,d,h,i are the temperature and flow field of the droplet in contact with the molten pool in different transition periods. It can be found that after applying the LMF, the liquid bridge will undergo a degree of torsion, especially when t = t₀ + 0.60 s, which is when this phenomenon is the most obvious. When the contact is short-circuited, the electromagnetic contraction force generated by the current in the short-circuited liquid bridge is an important reason for facilitating the droplet transfer by pulling off the liquid bridge. The droplet is short-circuited and transferred by the combined effect of gravity and surface tension, etc., and the liquid bridge appears to be necking. The force analysis of the liquid bridge during a short circuit is shown in Figure 8. Taking the minimum interface after the necking of the liquid bridge as the boundary, F₁ is the electromagnetic shrinkage force on the upper part of the liquid bridge, and F₂ is the lower part. The combined force of the two is F. The formulas for all three are shown below:

(1)$F={ F}_{1 }+{ F}_2$

(2)$F_1=K{I}^2\mathit{ln}\left({\displaystyle \frac{r_0}{r_1}}\right)$

(3)$F_2=K{I}^2\mathit{ln}\left({\displaystyle \frac{r_2}{r_1}}\right)$

(4)$K={\displaystyle \frac{{\mu }_0}{4\pi }}$

where μ₀ denotes vacuum permeability; I denotes current; K denotes the magnetic flux density per unit current element produced in vacuum; r₀ is the radius of the welding wire; r₁ is the minimum interface radius after necking of the liquid bridge; and r₂ is the radius of the contact surface between the liquid bridge and the molten pool.

After the LMF is applied, the radial inward electromagnetic force is generated in the liquid bridge, which reduces the radius r₁ at the neck and increases the electromagnetic forces F₁ and F₂ on the upper and lower parts of the liquid bridge. It produces electromagnetic contraction forces in opposite directions. When the LMF parameters are appropriate, the generated electromagnetic contraction force helps to break the liquid bridge, so the short-circuit time is shortened under the applied LMF. Reducing the transition time means that the energy accumulated at the neck is reduced, so the amount of spatter generated by the liquid-bridge burst is reduced.

Since the current inside the droplet and molten pool inevitably has a radial component, the Lorentz force generated by the current and the LMF will have an influence on the liquid metal flow. As shown in Figure 7e, compared with the welding without an EMF at the same time, the radial movement of the liquid metal is obviously accelerated after the LMF is applied, and the LMF has a stirring effect on the liquid metal. Since the CO₂ welding arc is always below the droplet, it will deviate from the wire axis direction without the EMF. By applying an LMF, the droplet can produce circumferential motion. The behavior of the deviation from the wire axis is improved, which makes the transfer more controllable.

In CO₂ welding, surface tension affects the flow of metal. High temperatures reduce the surface tension, and a temperature gradient is created in this regard. The temperature in the center is much higher than that in surrounding areas, and metal will flow from areas of low surface tension to high. As shown in Figure 8, the movement from the center to the surrounding area eventually cools to form the melt depth and width. The molten pool at this stage is subjected to an electromagnetic stirring effect by the EMF. Arc heating at the current density is greater, but the Lorentz force is greater, and the liquid metal flow in the EMF and arc changes are more obvious. A large number of rotating charged particles generate an induced current that produces an electromagnetic force. It interacts with the EMF to agitate the pool, which increases the flow rate and area of the molten metal. The metal in the molten pool circulates upward, outward, and downward.

The temperature and flow field distribution of the molten pool on the surface of the workpiece (Z = 0 mm) after applying the LMF are shown in Figure 9. The radial motion of liquid metal is obviously enhanced after the LMF is applied, which increases the weld width. The higher current density in the pool results in the liquid metal by the magnetic field effect being more obvious and an overall clockwise direction of the flow trend. Under the same magnetic field conditions, when the arc plasma around the direction of the wire axis undergoes a clockwise spiral movement, the LMF arc will further produce a stirring effect on the molten pool. The electromagnetic stirring effect of the magnetic field on the pool has a certain effect on the weld size and organizational properties.

3.3. Mechanism of Action

As shown in Figure 10, the flow of metal after the application of an LMF is mainly subject to the following forces: buoyancy, gravity, surface tension, gas shear stress, and electromagnetic forces exist inside the molten pool. On the outside, there is arc pressure, droplet impact force, and outer wall support force. The influences of gravity, droplet impact, and arc pressure mainly change the weld depth. The buoyancy force and the outer wall support force also have an effect on the weld depth, and the effect is opposite to the above three forces. The influence of gas shear stress and surface tension on the weld is reflected in the increase in the melt width. Electromagnetic force promotes the flow of metal inside the pool and has a greater effect on the heat and mass transfer between the top and bottom in the center of the pool.

The combined influence of the magnetic and electric fields accelerates the charged particles in the arc to bombard the molten pool, which produces an impact on the pool. From the study of droplet transfer behavior, the introduction of an EMF can change the droplet morphology and transfer form, so an EMF can change the size of the droplet impact force. The action of the LMF causes the arc to rotate, and the molten pool is subjected to tangential force. According to the principle of electromagnetic induction, eddy currents are formed at the center of the pool, further promoting the cyclic flow of the pool, which is influenced by the direction of the magnetic field. The EMF also accelerates the transfer frequency of the droplet, which periodically impacts the surface of the pool and promotes the width enhancement on a macroscopic level.

The application of an LMF will generate eddy currents in a plane perpendicular to the magnetic lines of force as well as additional eddy current forces through the interaction with the EMF. The high density of the magnetic lines in the upper part of the molten pool makes the metal motion more complex and agitation more intense than in conventional CO₂ welding, especially the electromagnetic and eddy current forces that drive the molten pool downward and toward the center of the pool. Meanwhile, the weld depth of fusion increases significantly due to arc contraction and increased arc pressure, and the weld depth-to-width ratio eventually increases.

3.4. Microstructure Characteristics

The complex heat and mass transfer behaviors in magnetron high-current CO₂ welding hinder the microstructure growth mechanism. To investigate the microstructure of the joint under different parameters, the weld microstructure was examined with and without EMF conditions, and the observation positions are shown in Figure 11.

From Figure 11c, it was found that the interior of the weld consists of columnar crystals with an obvious orientation. The microstructure of the weld mainly consists of pre-eutectic ferrite (PF), ferrite with a side plate (FSP), and acicular ferrite (AF). The red dashed line in the middle of the weld zone and the heat-affected zone in Figure 11d is the fusion zone, which is usually narrow and difficult to observe. The direction of the yellow arrow indicates the growth orientation of the columnar crystals inside the weld. The size of the molten pool during welding is small, and the cooling rate is fast. The temperature of the liquid metal within it is high and shows a certain gradient. The solid–liquid phase interface and the center temperature gradient are the largest, and the temperature gradient in the molten pool leads to a certain direction of nucleation growth of grains to form columnar crystals. Without the magnetic field, in the austenite grain boundaries at the growth of the PF presents a thicker long strip, which is sparsely distributed in the weld interior. FSP occurs along the long strip of pre-eutectic ferrite growth to the grain interior, with the distribution of pickaxe teeth. AF is irregularly distributed inside the grain. The amount of granular bainite is relatively small. Grain distribution in the HAZ is not uniform, the size in the region near the weld is relatively coarse, and the region near the base material is a relatively small size.

According to Figure 11e, the internal microstructure of the weld after applying the EMF is still columnar crystal organization, which is especially obvious in the region near the fusion line. The long PF in the weld zone is broken and transformed into a small-sized reticulation, and the size and number of FSPs are relatively reduced, while the size of AF is reduced, and the number of granular bainite around the ferrite grain boundaries is significantly increased compared with that in the absence of an EMF. The above phenomenon is mainly caused by the stirring effect of the ELMF on the liquid metal, which makes the coarser-sized PF and FSP break up and form more crystalline cores to promote nucleation. The EMF can have an effect on the flow state of the metal, promoting the radial flow of liquid metal. The metal flow rate increases, accelerating the cooling rate of the pool and making the internal microstructure of the weld finer and denser. The temperature gradient near the weld fusion zone is large, and the electromagnetic stirring effect of the EMF on the liquid metal is not obvious. The crystallization to the center of the weld increases, the temperature gradient decreases, and the electromagnetic stirring effect of the EMF is greater than the temperature gradient, so that the role of the EMF has different effects on the weld interior of different regions.

According to Figure 12a,d, the grain size of the weld under an EMF is improved, and the microstructure is obviously refined compared with the no-EMF condition. The grain size is smaller without an EMF. This means that the number of grains per unit area is more, and more grains can be dispersed in the external force generated by the plastic deformation, thus reducing the stress concentration. Grain refinement can improve the plasticity and significantly improve the overall mechanical properties. It can also be seen in Figure 12b–d,f that the percentage of large-angle grain boundaries under the EMF rises significantly. The grain boundary area is larger and zigzagged, and the dislocation motion is improved by the hindering effect of the grain boundary. The average KAM decreases under the EMF, as can be seen in Figure 12g–j.

3.5. Mechanical Properties

Two hardness picking paths were chosen to study the hardness change in the weld area: one path was taken from the base material below the weld centerline in the longitudinal direction until the center of the weld near the surface position; another path was 1 mm from the upper surface of the welded substrate, from the left side of the weld to the right side of the base material to take points in turn, until the right side of the base material. The specific location of the point is shown in Figure 11b.

The hardness measurements of the transverse and longitudinal paths were recorded, and the weld hardness distribution curve was plotted, as shown in Figure 13. It can be visualized on the weld cross-section that the hardness distribution is not uniform, and the weld hardness and the point location taken for the microstructure distribution are related to the welded joint and can be divided into the weld (WM), fusion zone (FZ), heat-affected zone (HAZ), and the base material (BM); the heat-affected zone can be divided into the superheated zone, normalized zone, and part of the tissue transformation zone. The superheated zone grain size is coarse, and it is the region of poor plastic toughness and high hardness. The normalizing zone metal after normalizing treatment shows better mechanical properties, and the grain size distribution is uniform and small. Part of the microstructure of the transformation zone did not cause phase transformation, and the grain size is not uniform. It can be found that when the hardness test point is approached from the BM to the WM, the hardness shows a phenomenon of increasing and then decreasing, and the hardness of the HAZ is the highest. The high temperatures to which the heat-affected zone is subjected to during welding result in rapid grain growth in areas close to the weld, which can lead to an increase in the hardness. The weld metal, which is formed after the solidification of the molten areas in the weld, undergoes high temperatures and rapid cooling and has a relatively fine grain size. Fine grains usually have good toughness and ductility but relatively low hardness.

When transverse hardness measurements are performed, the hardness in the WM is slightly lower than in the BM. However, the hardness of the WM varies under the action of different parameters of the EMF. The effect of the excitation current on the weld hardness is more obvious, with a fixed excitation frequency of 60 Hz, and the hardness is higher than that of the weld without the EMF, except that the hardness is slightly lower when the excitation current is 5 A. The maximum hardness is lower than that of the BM when the magnetic field is applied. The maximum hardness without the EMF is 226.6 HV0.5, while the maximum hardness of the weld at 9 A/60 Hz reaches 248.8 HV0.5, which is an increase of 9.80%. By fixing the excitation current at 7 A, it is found that the change in the frequency has little effect on the hardness, and there is little difference between the weld hardness under different excitation frequencies and that without an EMF.

The longitudinal hardness distribution under different excitation parameters is studied, and Figure 13 shows that the hardness of the WM is slightly higher than that of the BM. The maximum value of weld hardness without a magnetic field is 233.4 HV0.5 (95% confidence interval is [230, 240] HV), and the maximum value of weld hardness can reach 258.8 HV0.5 (95% confidence interval is [255, 265] HV) when the excitation parameter is 9 A/60 Hz, which is a 10.88% increase in the weld hardness in comparison.

In summary, the change in the weld hardness under an EMF is mainly related to the excitation current, and the magnetic field intensity increases with the increase in the excitation current, which, in turn, affects the internal microstructure of the weld and influences the weld cross-section hardness.

On the other hand, tensile sampling of the specimen was carried out, as shown in Figure 11a, and a fracture occurred in the WM for different parameters. The appearance and tensile strength of the specimen after pull-off are shown in Figure 14. It follows that the tensile strength of the weld changes after the ELMF is applied, showing an overall increasing trend. The tensile strength without an magnetic field is 567.67 MPa (95% confidence interval is [559, 579] MPa), and the tensile strength reaches a maximum value of 607.83 MPa (95% confidence interval is [595, 615] MPa) at an excitation parameter of 6 A/60 Hz, which is an increase of 7.07% in comparison. Under the condition of a constant excitation current, the weld tensile strength shows a single peak distribution with the increase in excitation frequency. The weld tensile strength is highest at a frequency of 60 Hz, after which the tensile strength gradually decreases with the increase in frequency. At the same frequency, the higher tensile strength is obtained at a medium excitation current, and the tensile strength obtained at a 7 A excitation current is better than other parameters.

4. Conclusions

In this paper, a three-dimensional, transient, multi-energy field coupled CO₂ welding droplet–molten pool model is established. Using a combination of numerical simulation and experimental verification, the heat and mass transfer mechanisms and microstructural characteristics of LMF on the droplet–molten pool are systematically investigated.

The radial flow of metal in the pool is enhanced under an LMF, the heat and mass transfer behaviors are accelerated, and the overall morphology is wide and shallow. Applying an LMF can reduce the short-circuit time, lower the energy accumulated in the short-circuited liquid bridge, and promote droplet transfer, which, in turn, reduces the generation of spatter.

Compared with conventional high-current CO₂ welding, EMF-assisted welding can significantly enhanced welding stability. The LMF has an obvious refining effect on the microstructure of the weld zone and reduces stress concentration. At the same time, it improves the plasticity and strength and significantly improves the comprehensive mechanical properties of the weld.

Through these studies, our understanding of the heat and mass transfer behaviors of magnetically controlled high-current CO₂ welding was comprehensively enhanced. This is important for improving the performance of welded joints by optimizing the machining process. The models constructed and insights gained can optimize the application of manufacturing techniques, which, in turn, enables the precise regulation of the microstructure of production components and the enhancement of their mechanical properties.

The EMF-assisted CO₂ welding technology studied holds great promise not only for the welding industry but for other industries and fields as well. Increased welding speeds and reduced heat-affected zones mean faster production lines and higher-quality welds, resulting in cost savings and increased competitiveness for manufacturers. Compared to conventional methods, EMF-assisted welding reduces energy consumption, which makes it a more environmentally friendly alternative that is in line with global sustainability goals. This technology can serve as a basis for further developments in welding technology, stimulating new research into EMF-enhanced processes. The interdisciplinary nature of the technology combines welding, electromagnetism, and materials science, which paves the way for future innovations at the intersection of these fields.

Author Contributions

Conceptualization, J.-Y.M. and H.-L.Z.; methodology, J.-Y.M. and Y.-L.C.; investigation: J.-Y.M., Y.-W.L. and B.-W.R.; data curation: Y.-W.L., B.-W.R., H.-L.Z. and S.-Y.Z.; writing—original draft: J.-Y.M., Y.-W.L. and H.-L.Z.; writing—review and editing: Y.-L.C. and Q.W.; funding acquisition: Y.-L.C. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

The original contributions presented in the study are included in the article; further inquiries can be directed to the corresponding authors.

Conflicts of Interest

The authors declare no competing interests.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Table

Figure 1. EMF-assisted CO2 welding platform.

Figure 2. Schematic diagram of welded seam section and measurement position, where d, w, and h represent the weld depth, weld width, and weld height, respectively.

Figure 3. Droplet model and physical properties.

Figure 4. High-speed camera for droplet–molten pool.

Figure 5. The change in the weld size under different magnetic field parameters. The red dotted line and the blue dotted line represent the weld depth and the weld height, and the bar graph represents the weld width, where the same color of the bar graph represents the same excitation frequency. In particular, the leftmost black bar represents the weld parameters without the EMF.

Figure 6. Comparison between numerical simulation results of weld.

Figure 7. Temperature and flow field distributions in the longitudinal section of the molten pool at different moments (no magnetic field on the left, an EMF applied on the right, with Y = 0 mm). (a,f) t = t0 + 0.05 s, (b,g) t = t0 + 0.30 s, (c,h) t = t0 + 0.50 s, (d,i) t = t0 + 0.60 s, and (e,j) t = t0 + 1.00 s.

Figure 8. Force analysis of short-circuit liquid bridge and schematic diagram of molten pool flow. r0 is the radius of the welding wire; r1 is the minimum interface radius after necking of the liquid bridge; r2 is the radius of the contact surface between the liquid bridge and the molten pool.

Figure 9. Temperature field and flow field distribution on the surface of the workpiece at t = t0 + 0.80 s (Z = 0 mm). (a) No EMF; (b) EMF.

Figure 10. LMF acting on the molten pool.

Figure 11. Schematic diagram of the position. (a) Overall view. (b) SEM sampling position. (c) Microstructure of the weld without an EMF. (d) Microstructure of the weld–fusion junction without an EMF. (e) Microstructure of the weld with an EMF. (f) Microstructure of the weld–fusion junction with an EMF.

Figure 12. EBSD analysis of weld zone: (a) no EMF IPF, (b) no EMF size angle grain boundary distribution, and (c) no EMF orientation angle distribution; (d) EMF IPF, (e) EMF size angle grain boundary distribution, and (f) EMF orientation angle distribution; (g) no EMF KAM and (h) no EMF KAM distribution; and (i) EMF KAM and (j) EMF KAM distribution.

Figure 13. Distribution of weld hardness for different excitation parameters (longitudinal on the left and transverse on the right).

Figure 14. Tensile strength of welds with different excitation parameters (excitation current and excitation frequency).

Table 1

The chemical compositions of Q235B mild steel and ER50-6 welding wire (wt.%).

	C	Mn	Si	S	P	Ni	Cr	Cu	Fe
Q235B	0.150	0.450	0.230	0.028	0.020	---	---	---	Bal.
ER50-6	0.110	1.450	0.870	0.013	0.012	0.017	0.031	0.125	Bal.

References

1. Yun, T.J.; Park, M.H.; Oh, W.B. A Study on Spatter Tracking Algorithm for a Vertical GMA Welding Process. J. Weld. Join.; 2019; 37, pp. 347-356. [DOI: https://dx.doi.org/10.5781/JWJ.2019.37.4.9]

2. Yamazaki, K.; Suzuki, R.; Shimizu, H. Spatter and fume reduction in CO₂ gas-shielded arc welding by regulated globular transfer. Weld. World; 2012; 56, pp. 12-19. [DOI: https://dx.doi.org/10.1007/BF03321376]

3. Wu, C.S.; Yang, F.Z.; Gao, J.Q. Effect of external magnetic field on weld pool flow conditions in high-speed gas metal arc welding. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf.; 2016; 230, pp. 188-193. [DOI: https://dx.doi.org/10.1177/0954405414555591]

4. Ke, W.; Zeng, Z.; Oliveira, J.P. Heat transfer and melt flow of keyhole, transition and conduction modes in laser beam oscillating welding. Int. J. Heat. Mass Transf.; 2023; 203, 123821. [DOI: https://dx.doi.org/10.1016/j.ijheatmasstransfer.2022.123821]

5. Gou, Q.Z.; Zhang, Z.; Xu, L. Heat and mass transfer behavior in CMT plus pulse arc manufacturing. Int. J. Mech. Sci.; 2024; 281, 109638. [DOI: https://dx.doi.org/10.1016/j.ijmecsci.2024.109638]

6. Chang, Y.L.; Liu, X.L.; Lu, L. Research status and prospect of short-circuit transition CO₂ welding. Weld. Tech.; 2013; 42, pp. 1-5.

7. Yang, Z.; Fang, C.; Wu, M. The mechanisms of arc coupling and rotation in cable-type welding wire CO₂ welding. J. Mater. Process. Tech.; 2018; 255, pp. 443-450. [DOI: https://dx.doi.org/10.1016/j.jmatprotec.2017.12.032]

8. Wang, L.; Chen, J.; Wu, C.S. Numerical analysis of arc and droplet behaviors in gas metal arc welding with external compound magnetic field. J. Mater. Process. Tech.; 2020; 282, 116638. [DOI: https://dx.doi.org/10.1016/j.jmatprotec.2020.116638]

9. Tan, C.; Liu, Y.; Gong, Q. Numerical and experimental study of thermal fluid flow and keyhole dynamic in laser welding of aluminum alloy assisted by electromagnetic field. Opt. Las. Tech.; 2023; 157, 108718. [DOI: https://dx.doi.org/10.1016/j.optlastec.2022.108718]

10. Agrawal, M.K.; Singh, R.P. Effect of external magnetic field on impact strength and hardness of weld of shielded metal arc welding process. Mat. Tod. Proc.; 2021; 45, pp. 3638-3641. [DOI: https://dx.doi.org/10.1016/j.matpr.2020.12.1219]

11. Singh, R.P.; Raghuvanshi, D.; Pal, A. Effect of external magnetic field on weld width and reinforcement height of shielded metal arc welded joints. Mat. Tod. Proc.; 2021; 38, pp. 112-115. [DOI: https://dx.doi.org/10.1016/j.matpr.2020.06.107]

12. Shoichi, M.; Yukio, M.; Koki, T. Study on the application for electromagnetic controlled molten pool welding process in overhead and flat position welding. Sci. Tech. Weld. Join.; 2013; 18, pp. 38-44. [DOI: https://dx.doi.org/10.1179/1362171812Y.0000000070]

13. Gou, Q.Z.; Zhang, Z. Simulation and analysis of the heat transfer mechanism of arc plasma with CMT plus pulse composite heat source. Weld. World; 2024; 68, pp. 525-541.

14. Liu, Y.B.; Sun, Q.J.; Liu, J.P. Effect of axial external magnetic field on cold metal transfer welds of aluminum alloy and stainless steel. Mater. Lell.; 2015; 152, pp. 29-31. [DOI: https://dx.doi.org/10.1016/j.matlet.2015.03.077]

15. Jiang, S.Y.; Zheng, X.F.; Chen, H.M. Control mechanism of CO₂ welding spatter by external magnetic field. J. Weld.; 2004; 03, pp. 65-67.

16. Jiang, S.Y.; Wang, X.W.; Chen, H.M. The impact of adscititious longitudinal magnetic field on CO₂ welding process. Adv. Mater. Res.; 2012; 538, pp. 1447-1450. [DOI: https://dx.doi.org/10.4028/www.scientific.net/AMR.538-541.1447]

17. Hua, A.B.; Yin, S.Y.; Chen, S.J. Control effect of longitudinal magnetic field on MAG welding arc and droplet transfer. J. Mech. Eng.; 2010; 46, pp. 95-100. [DOI: https://dx.doi.org/10.3901/JME.2010.14.095]

18. Wang, L.; Wu, C.S.; Yang, Z.F. Active control effect of external magnetic field on arc and molten pool behavior in high-speed GMAW. J. Mech. Eng.; 2015; 51, pp. 95-100.

19. Wang, L.; Wu, C.S.; Gao, J.Q. Suppression of humping bead in high speed GMAW with external magnetic field. Sci. Tech. Weld. Join.; 2016; 21, pp. 131-139. [DOI: https://dx.doi.org/10.1179/1362171815Y.0000000074]

20. Cho, D.W.; Kiran, D.V.; Song, W.H. Molten pool behavior in the tandem submerged arc welding process. J. Mater. Process. Tech.; 2014; 214, pp. 2233-2247. [DOI: https://dx.doi.org/10.1016/j.jmatprotec.2014.04.032]

21. Ogino, Y.; Hirata, Y.; Murphy, A.B. Numerical simulation of GMAW process using Ar and an Ar-CO₂ gas mixture. Weld. World; 2016; 60, pp. 345-353. [DOI: https://dx.doi.org/10.1007/s40194-015-0287-3]

22. Ogino, Y.; Hirata, Y.; Asai, S. Numerical simulation of metal transfer in pulsed-MIG welding. Weld. World; 2017; 61, pp. 1289-1296. [DOI: https://dx.doi.org/10.1007/s40194-017-0492-3]

23. Ogino, Y.; Asai, S.; Hirata, Y. Visualization of arc plasma and molten wire behavior in CO₂ arc welding process by three-dimensional numerical simulation. Weld. World; 2020; 64, pp. 1789-1797. [DOI: https://dx.doi.org/10.1007/s40194-020-00958-4]

24. Xiao, L.; Fan, D.; Huang, J. Numerical Study on Arc Plasma Behaviors in GMAW with Applied Axial Magnetic Field. J. Phys. Soc. Jpn.; 2019; 88, 074502. [DOI: https://dx.doi.org/10.7566/JPSJ.88.074502]

25. Xiao, L.; Fan, D.; Huang, J. Numerical study on arc-droplet coupled behavior in magnetic field controlled GMAW process. J. Phys. D Appl. Phys.; 2020; 53, 115202. [DOI: https://dx.doi.org/10.1088/1361-6463/ab6020]

26. Zhao, W.Y.; Tashiro, S.; Murphy, A.B. Deepening the understanding of arc characteristics and metal properties in GMAW-based WAAM with wire retraction via a multi-physics model. J. Manuf. Process.; 2023; 97, pp. 260-274. [DOI: https://dx.doi.org/10.1016/j.jmapro.2023.05.008]

27. Zhao, W.Y.; Wei, Y.H.; Tashiro, S. Numerical investigations of arc plasma characteristic parameters evolution and metal properties in GMAW-based WAAM of Al alloy with an integrated model. J. Manuf. Process.; 2023; 99, pp. 321-337. [DOI: https://dx.doi.org/10.1016/j.jmapro.2023.05.047]

28. Zhao, W.Y.; Jin, H.; Du, X. A 3D arc-droplet-molten pool integrated model of Al alloy GMAW process: Heat transfer, fluid flow and the effect of external magnetic field. Vacuum; 2022; 202, 111129. [DOI: https://dx.doi.org/10.1016/j.vacuum.2022.111129]

29. Han, J.; Han, Y.; Sun, Z. Effect of plasma welding current on heat source penetration ability of plasma-GMAW hybrid welding. Int. J. Heat Mass Transf.; 2022; 123, pp. 1835-1844. [DOI: https://dx.doi.org/10.1007/s00170-022-10280-6]

30. Wang, Y.; Jiang, P.; Geng, S. Influence of the wrinkle surface structures on the vapor flow and keyhole stability in 20 kW high power laser welding. Int. J. Heat Mass Transf.; 2022; 193, 122958. [DOI: https://dx.doi.org/10.1016/j.ijheatmasstransfer.2022.122958]

31. Li, J.; Geng, S.; Wang, Y. Mitigation of porosity in adjustable-ring-mode laser welding of medium-thick aluminum alloy. Int. J. Heat Mass Transf.; 2024; 227, 125514. [DOI: https://dx.doi.org/10.1016/j.ijheatmasstransfer.2024.125514]

32. Li, L.; Liu, W.; Gong, Q. Numerical analysis of the dynamic behavior of arc by rotating laser-GMAW hybrid welding of T-joints. Opt. Las. Tech.; 2023; 167, 109802. [DOI: https://dx.doi.org/10.1016/j.optlastec.2023.109802]

33. Shi, L.; Jiang, L.; Gao, M. Numerical research on melt pool dynamics of oscillating laser-arc hybrid welding. Int. J. Heat Mass Transf.; 2022; 185, 122421. [DOI: https://dx.doi.org/10.1016/j.ijheatmasstransfer.2021.122421]

34. Geng, S.; Yang, W.; Jiang, P. Numerical study of keyhole dynamics and porosity formation during high-power oscillating laser welding of medium-thick aluminum alloy plates. Int. J. Heat Mass Transf.; 2022; 194, 123084. [DOI: https://dx.doi.org/10.1016/j.ijheatmasstransfer.2022.123084]

35. Zhang, S.; Ma, G.; Peng, X. Numerical simulation of the effects of bypass current on droplet transfer during AZ31B magnesium alloy DE-GMAW process based on FLUENT. Int. J. Heat Mass Transf.; 2017; 90, pp. 857-863. [DOI: https://dx.doi.org/10.1007/s00170-016-9438-y]

36. Ke, W.; Liu, Y.; Teshome, F.B. Numerical study on multiphase evolution and molten pool dynamics of underwater wet laser welding in shallow water environment. Int. J. Heat Mass Transf.; 2024; 220, 124976. [DOI: https://dx.doi.org/10.1016/j.ijheatmasstransfer.2023.124976]