Thermoelectric coupling deep drawing process of ZK60 magnesium alloys

In this work, thermal deep drawing and thermoelectric coupling drawing tests were performed on an annealed ZK60 magnesium alloy sheet to investigate the effect of the pulse current on the formability of the ZK60 magnesium alloy during the drawing process. The effects of thermal deep drawing and the thermoelectric coupling on the forming of drawn parts were studied, and the changes of the thinning rate and hardness were obtained. The results showed that the using of pulse current improved the formability of the ZK60 magnesium alloy and increased the ultimate depth of drawing. The hardness value of the drawn part showed an overall increasing trend, followed by thermoelectric coupling deep drawing, which was significantly higher than the sheet after annealing without deep drawing. The deep-drawn parts had different grain sizes in different places, and dynamic recrystallization occurred to varying degrees in each region. Furthermore, the degree of dynamic recrystallization in the punch fillet and cylinder wall was significantly higher than in other areas.


Introduction
Attempts are being made to reduce energy consumption and pollution by reducing the weight of vehicles and electronic devices, as the aerospace, automotive manufacturing, and consumer electronics industries rapidly increase. Magnesium has a wide range of applications as a lightweight metal structural material, including in aerospace, car transportation, electronic communication, and in other fields [1][2][3]. The forming performance of the material will be poor at room temperature, owing to the characteristics of the magnesium alloy itself, and defects such as wrinkling or cracking will occur during the deep drawing process. This can cause significant problems for actual production and seriously affect the forming accuracy and product quality, preventing the widespread use of magnesium alloys in industrial applications [4][5][6]. Electroplasticity is a process in which the deformation resistance of a material diminishes due to the action of moving electrons after the application of a pulse current, considerably enhancing plasticity [7,8]. Conrad et al. [9] provided a theoretical basis for the electroplasticity effect mechanism by explaining the theory of electroplasticity with the mechanism of electrons and dislocation interactions. Li et al. [10] researched the effect of pulse current on the tensile deformation of SUS304 stainless steel, and showed that air cooling eliminated the impact of current heating. The pulse current reduced the flow stress and dislocation density. It increased the plasticity of SUS304 stainless steel by 72.5%, which was higher than the plasticity of the samples evaluated without the pulse current. Lv et al. [11] assessed how the electroplastic effect affected the deep drawing of high-strength steel plates of various strengths, and discovered that the electroplastic effect could improve the formability and mechanical characteristics. The study also found that the high-strength steel microstructure was unaffected by the Joule heat created by the pulsed current. Li et al. [12] conducted an experimental study on the electric pulse-assisted bulging of AZ31 magnesium alloy, they showed that the bulging performance of the alloy could be improved by increasing the peak current density. The height-diameter ratio of the non-electrically bulging shape was 0.4 at the same temperature, while the height-diameter ratio of the electrically assisted bulging shape could be raised to 0.48 when the peak current density was 45 A/mm 2 . Xie et al. [13] studied the effect of current on an AZ31B magnesium alloy, and showed that the drawing limit increased by 1 mm when the current or frequency increased at the same temperature, indicating that the magnesium alloy material exhibited a pure electroplastic effect during drawing. Li et al. [14] investigated the electroplasticity of magnesium alloys and believed that the enhanced plasticity of the magnesium alloys at low temperatures was due to the dynamic recrystallization induced by electrical pulses, which could promote dislocation movement within the material and speed up the rate of dynamic recrystallization nucleation. Liu et al. [15] further studied the electroplasticity of the magnesium alloy effect mechanism. According to research on AZ31B magnesium alloys, the pulse current accelerated dislocation annihilation, and dynamic softening before necking was caused by the pulse current rotating the grains from a hard orientation to a soft orientation, rather than resulting from dynamic recrystallization, which confirmed the existence of a pure electroplastic effect. Thus, combining traditional techniques with electric-assisted forming technology, adding pulse current into the forming process to improve the forming properties of ZK60 magnesium alloy sheets has shown to be feasible [16], and it would be beneficial to obtain magnesium alloy products with high quality and excellent performance. As a result, the flow law of ZK60 magnesium alloy plates during the process of thermoelectric coupling deep drawing was investigated to generate new ideas for the forming of magnesium alloys and to explore the feasibility of using electroplastic processing technology in real-world manufacturing.
In this work, a pulse current was added to the hot drawing process of a ZK60 magnesium alloy, allowing a sheet to be drawn with it, and compared the experimental results of hot drawing at the same temperature. Through the analysis of the thinning rate, hardness and microstructure change law of deep drawing samples, the theory that pulse current can help to improve the deformation ability of materials was verified, and the feasibility of the application of electro plastic processing technology in deep drawing of magnesium alloy plates was verified.
2 Thermoelectric coupling deep drawing process design

Thermoelectric coupling deep drawing tool
The thermoelectric coupling deep drawing device was mainly composed of a YS9000D-3050 square wave DC pulse power supply, heating ring, punch, die, thermocouple, blank holder, temperature control box, and digital thermometer, as shown in Fig. 1. To avoid the influence of electrical parameters on the experimental results, the drawing speed was constant, the pulse current was constant at 50 A, the frequency was 800 Hz, and the duty cycle was 90% during thermoelectric coupling drawing. The reverse deep drawing was used in the forming process, and the die's periphery was surrounded by a heating ring, which was connected to a temperature control box for heating. A through hole was drilled into one end of the die's side wall, allowing the thermocouple to be connected to a digital thermometer via the through hole. The temperature range was − 50 °C to 450 °C, which was suitable for the temperature measurement range in this experiment. Using a chip thermocouple in contact with the sheet, the temperature change of the sheet could be measured in real time. A groove with the same diameter as the blank and a depth of 1.0 mm was provided on the upperend surface of the blank holder, which was used to position the sheet material. In addition, a through hole embedded electrode was provided in the blank holder, and the diameter of the circumference where the through hole was located was not bigger than the diameter of the deep-drawn flange part. The non-hole end of the electrode was in contact with the sheet material, and a blind hole with a certain depth was opened at one end of the electrode, which was used to connect to the wire. Conveniently, the electrode and the wire are connected and fixed with a soldering pen and solder. In addition, the electrode jacket contained insulating rubber and interference was fit with the blank holder through the hole, which prevented current from flowing through the rest of the die, ensuring that a current loop was formed during the entire drawing and deformation process, allowing the current and heat to be coupled.

Thermoelectric coupling deep drawing process
The plate was annealed and cut into a circular plate with a diameter of 78 mm before testing. Finally, 400 grit sandpaper was used to lightly polish the sample surface until it was smooth, and no lubricating measures were used in the deep drawing process. The deep drawing die was installed on the H1F80-11 servo press, the servo press was opened, and the upper die was slid down until it was just closed with the lower die in the thermoelectric coupling deep drawing process. After reaching a certain temperature, the temperature control box and pulse power supply were powered on, and the deep drawing process was started (at this time, it was in a powered-on state until the deep drawing was complete). Then, the power supply and temperature control box were powered off once the deep drawing was finished. Before deep drawing, the circuit was disconnected, and the temperature control box and the pulse power supply were powered on when the upper die made contact with the lower die. Thus, a circuit formed, and the sheet was deep drawn at this point. After the deep drawing process was complete, the temperature control box and the power supply were powered off.

Experimental materials and methods
The experimental material consisted of a commercial ZK60 rolled magnesium alloy with a thickness of 1 mm, and its main chemical components were Zn 5.0, Zr 0.31, Mn 0.10, and Al 0.05, while the remainder was Mg (mass fraction, %). The die size parameters and test conditions are shown in Table 1.
The etchant used in metallographic analysis was picric acid etchant, and the composition of the etchant was 1 g picric acid, 1 ml acetic acid, 20 ml ethanol and 4 ml distilled water. The mirror surface of the sample was corroded for 5 s ~ 10 s during corrosion. Used ZEISS A1 microscope to observe and analyze the sample metallographically, and used image processing software to calculate the grain size. Starting from the bottom center of the cylinder, cut the cylinder along the central axis to the edge of the flange. The schematic diagram of positions for metallographic observation, including the bottom of the cylinder, the corner of the punch, the wall of the cylinder, the corner of the die and the flange, is shown in Fig. 2a. The sampling direction of the tensile sample was parallel to the RD direction of the plate, and the tensile tests were conducted at room temperature with a tensile rate of 1 mm/min on a Z010 Allround tabletop microtensile testing machine. The thickness and hardness of the 10 node positions in the measurement results are shown in Fig. 2b. The measuring tool selected for measuring the sample thickness was the double-tipped digital micrometer. The equipment used to measure the hardness was FM-700FM-ARS Vickers hardness tester. During the test, the test load used was 50 gf and the holding time was 10 s. Take the average value as the hardness value of the measuring point.

Annealing treatment
The samples were annealed at different temperatures, with the temperature gradually increasing in the furnace before they were cooled to room temperature after holding for 1 h. The magnesium alloy sheets were then annealed and marked at 300 °C, 340 °C, 380 °C, and 420 °C. The annealed  sample's mechanical properties were tested, and the annealing plan was finally determined (according to ASTM E8/ E8M-08). According to the stress-strain curve in Fig. 3, when the temperature was in the range of 300 °C to 420 °C, the tensile strength of the sample gradually decreased with the increase in temperature. The sheet underwent recrystallization and softening during the annealing process, and the internal texture was weakened, reducing the strength of the sheet and increasing its elongation. The elongation of the ZK60 magnesium alloy sheet reached its maximum value when it was annealed at 380 °C. When the annealing temperature was increased to 420 °C, the strength and elongation of the material decreased significantly. When the elongation was only 23.4%, the tensile strength was 273.5 MPa, and the strength and plasticity decreased at this time. This occurred because the temperature was too high, the grains grew, and the grains became too large, causing the mechanical properties to deteriorate [17]. Annealing treatment at a suitable temperature range could significantly improve the plasticity and forming ability of the ZK60 magnesium alloy. As shown in Table 2, after 1 h of annealing at 380 °C, the elongation of the sheet increased from 15 to 28%, while the tensile strength and hardness decreased to varying degrees. In addition, the hardness value dropped from 81.4 to 62.9 HV, and the tensile strength dropped from 332.4 to 295.4 MPa. The sheet metal underwent recrystallization and softening during the annealing process, which reduced its strength and hardness of the sheet metal. The tensile strength and hardness of metal materials will generally have a positive correlation, meaning that the tensile strength and hardness of the annealed material would both decrease at the same time [18]. Considering the ability of deep drawing to form deeper cylindrical parts and improve the deep drawing depth and yield, the samples to be drawn were chosen from sheets that were annealed at 380 °C for 1 h. Figure 4 shows that the deep drawing depth of the ZK60 magnesium alloy at room temperature was very small, and the fillet radius of the punch had not been pulled out, resulting in a large crack. Because of the close-packed hexagonal crystal structure of magnesium alloys, there were only two slip systems at room temperature: the basal plane slip and the critical shear slip. The required critical shear stress was large, which was relatively high compared to basal slip. Thus, it was difficult to open, as polycrystalline materials require at least five separate slip systems for free  deformation, and the plastic deformation ability was poor [19,20]. Figure 5 shows the macroscopic morphology of the ZK60 magnesium alloy after drawing at 120 °C, where the three drawn parts (a), (b), and (c) had drawing depths of 4.0, 4.5, and 4.5 mm, respectively. Thermally drawn samples (a) and (b) were among these, where sample (c) was the result of drawing while a pulse current was passed at 120 °C. The surfaces of samples (a) and (c) were intact and free of cracks, whereas sample (b) had a crack at the rounded corner of the punch, as shown in the figure. The maximum drawing depth of the ZK60 magnesium alloy was 4.0 mm at 120 °C; however, with the addition of the pulse current, the drawing part with a depth of 4.5 mm was successfully punched out, with no surface cracks and good drawing quality. Figure 6 shows the macroscopic morphology of the ZK60 magnesium alloy after drawing at 200 °C, where the three drawing parts (a), (b), and (c) had drawing depths of 7.7, 8.4, and 8.4 mm, respectively. Samples (a) and (b) were thermally drawn, while sample (c) was the result of drawing under the condition of passing a pulse current at 200 °C. Sample (a) showed that a cylindrical part with a depth of 7.7 mm could be punched out successfully at 200 °C, and the depth of hot drawing at 200 °C increased by 3.7 mm compared to 120 °C. Thus, the temperature had a significant impact on the formability of magnesium alloys. At temperatures below 200 °C, the plasticity of the ZK60 magnesium alloy did not improve significantly, and its deep drawing ability was limited. After the temperature reached 200 °C, the deep-drawn parts were drawn to a certain depth. The corner of the punch in sample (b) had cracks, while the surface quality of sample (c) was good, with no cracks on the surface. The maximum drawing depth increased by 0.7 mm compared to hot drawing at 200 °C. Thus, using a pulse current to achieve thermoelectric coupling during the hot drawing process improved the formability of the ZK60 magnesium alloy, increased the deformation ability of the sheet metal during drawing, and increased the limit drawing depth. Figure 7 shows the change in the thinning rate of each node of the deep-drawing part. During deep drawing, thinning and cracking can easily occur due to the action of unidirectional tensile stress in the cylinder wall area, and the thinning will gradually decrease from the fillet area of the punch to the bottom end of the cylinder arm. Thus, the die fillet area consisted of a transition zone, and the deformation of the material was more complicated, and the thickness of the die gradually increased as it progressed from the fillet to the flange. As shown in Fig. 7a, the maximum thinning rate of the punch fillet area was 3.1% at 120 °C of hot drawing, while the maximum thinning rate of the punch fillet area was 4.1% at 120 °C of thermoelectric coupling drawing. Only a slight thickness reduction phenomenon was observed at the bottom of the deep-drawn part, and sheet metal thinning was not obvious. The fillet area of the punch was affected by two types of tensile stress, radial and tangential, causing the material to thin significantly and cracks most likely occur. As shown in Fig. 7b, when the temperature reached 200 °C for hot drawing, the maximum thinning rate of the punch fillet area was 8.0%, and at 200 °C of thermoelectric coupling deep drawing, the maximum thinning rate of the punch fillet area reached 9.6%. Compared with the hot drawing process at the same temperature, the thermoelectric coupling drawing process could increase the drawing depth of the part to a certain extent, and improve the thinning rate of the drawing part. Figure 8 shows the hardness distribution curves for each node of the deep-drawing part, where the stress and deformation degrees of each position were not uniform after deep drawing of the ZK60 magnesium alloy. At different temperatures, the regions where the hardness peaks appeared were different. The hardness of each area improved after deep drawing, and the maximum hardness values appeared in the die fillet area and the flange, respectively. The amount of material that had to be transferred to the flange increased as well, where the thicker the flange area, the more serious the material hardening. The bottom of the cylindrical part was slightly harder than the hardness of the annealed sheet, and the cylinder wall and the punch fillet area were obviously thinned and deformed severely. The hardness value increased from the bottom end of the cylinder wall to the die fillet area due to work hardening. As shown in Fig. 8a, b, the hardness value reached its maximum in the die fillet area when drawing occurred at 120 °C. When thermally drawn at 120 °C, the maximum hardness was 67.2 HV, and when thermoelectrically coupled at 120 °C, the value was 67.7 HV. The maximum hardness appeared in the flange area when the temperature reached 200 °C. When thermally drawn at 200 °C, the maximum hardness was 69.8 HV, and when thermoelectrically coupled at 200 °C, the maximum hardness was 70.6 HV. At the same temperature, thermoelectric coupling deep drawing improved the deformation ability of the material to a certain extent. The increase in drawing depth led to an increase in the degree of deformation, and the degree of work hardening also increased. The hardness of the material also increased, and the hardness values of the deep-drawn parts improved to varying degrees, ensuring the performance of the formed product.

Microstructure
The drawn part after the thermoelectric coupling deep drawing at 200 °C was chosen as an example to clarify the microstructural changes in each area during thermoelectric coupling deep drawing of the ZK60 magnesium alloy, and the microstructure was observed in different areas of the cross-section. The metallographic structure of each region is shown in Fig. 9, and the microstructure of each region showed different degrees of recrystallization during the deep drawing process. The bottom of the stretching part was almost an undeformed area, which could be regarded as the structure of the sheet without deep drawing after direct annealing treatment. At this time, some recrystallized grains have appeared. In the punch fillet area, the number of dynamic recrystallized grains with smaller size increases significantly. The grain orientation along the fillet direction was elongated compared to other regions. This area was severely deformed and dynamic recrystallization occurred. Dynamic recrystallization occurs based on the original recrystallization structure, which further improves the degree of recrystallization, and the crystal grains become smaller. The tensile stress transmitted by the punch made the grains elongate in the cylinder wall area, the orientation of the large grains was obvious, and the degree of recrystallization was second only to the punch fillet area. The degree of deformation in the concave die fillet area was lower than that of the punch die fillet and cylinder wall area, and the degree of dynamic recrystallization was lower, which was higher than that of the bottom of the cylinder. The flange area of the deepdrawing part was affected by the tangential pressure, limiting sheet deformation in the circumferential direction and resulting in a few dynamic recrystallization grains. In each area of the deep-drawn part, no deformation twins were discovered. The current was found to play a role in not only promoting the slippage of the magnesium alloy, but also effectively inhibiting the formation of twins during the deformation process. At the same time, in order to reflect the significant influence of thermoelectric coupling process on the sheet structure, the microstructure of untreated sheet was also selected for comparison, and the untreated sheet structure is shown in Fig. 9f. Compared with the untreated plate, although the deformation of the tube bottom was very small, the grain size was obviously refined after the thermoelectric coupling treatment, and the fine grain strengthening phenomenon appeared.
The frequency distribution of grain size in each area of ZK60 magnesium alloy drawing was also calculated to further analyze the changes in microstructure in each area of the ZK60 magnesium alloy after drawing. As shown in Fig. 10, the grain size at the bottom of the cylinder was concentrated in the vicinity of 4 ~ 8 μm, accounting for about 59% of the total. This area had the smallest deformation degree, with an average grain size of 6.6 μm. The degree of deformation of the punch fillet and cylinder wall was relatively high, and the proportions of punch fillet and cylinder wall grain size near 0 ~ 4 μm were about 39% and 32%, respectively, while the grain size was about 4-8 μm. This accounted for about 51% and 41%, respectively. It was observed that there were more and finer grains at the punch fillet, and the degree of dynamic recrystallization was high, where the average grain size was only 5.0 μm. In the cylinder wall, the dynamic recrystallization degree was second only to the punch fillet area, and the average grain size was 6.0 μm. The degree of deformation of the die fillet and flange area was significantly lower than the punch fillet and cylinder wall area, and the grain size was primarily concentrated around 4 ~ 8 μm, accounting for about 47 ~ 52% of the total. The proportion of grain size in the vicinity of 0 ~ 4 μm decreased significantly, the deformation amount was small, with fewer dynamic recrystallization grains. The die fillet and flange had average grain sizes of 6.4 and 6.2 μm, respectively.

Discussion
The enhancements in hardness and drawing formability of the ZK60 magnesium alloy for the case of thermoelectric coupling drawing were not only the consequence of work hardening and temperature, but also directly related to the non-thermal effect of the electroplastic effect. The Joule heating effect, pure electroplastic effect, and skin effect consist of all types of electroplastic effect, and the skin effect can be ignored in most cases [21]. Among these, pure electroplasticity consists of the interactions between the moving electrons and dislocations that cause atoms to collide with floating electrons, accelerating dislocation movement, reducing flow stress, and further improving the plastic deformation ability of the material [22]. Kim et al. [23] assessed how the pulse current affected the static recovery behavior of magnesium alloys, and found that after electrification, the deformation resistance of the material decreased. This demonstrated that the current had a non-Joule heating effect in addition to a Joule heating effect on the material. At present, research on the pure electroplastic effect is still in the exploratory stage, and research on the additional effect of current is still not sufficient; thus, more in-depth exploration is urgently needed.
In the thermoelectric coupling deep drawing experiment, the pulse current could soften the material and reduce the deformation resistance, and the thermal effect made the magnesium alloy recover and recrystallize, which provided energy for dislocation movement, and promoted the initiation of the slip system. The non-thermal effect was dominated by the electromigration effect caused by the electron wind [24], which improved the dislocation mobility, allowed the dislocations to recombine, and promoted the dynamic recrystallization of the deformed region. In the process of deep drawing, the dislocation density in the deformation region became higher, and the dislocation entanglement became intensive. The essence of plastic deformation was the movement and proliferation of dislocations. The introduction of pulse current could reduce dislocation entanglement, promote the movement of dislocations, and release the accumulated deformation energy, thereby decreasing dislocation density. Both the non-thermal effect and thermal effect temperature were very important factors. The combined actions could more easily activate the slip system in magnesium alloys, promoting dynamic recrystallization, and improving material deformability. Compared to traditional hot drawing, the mechanism that relied only on increasing the temperature  Fig. 10 Frequency distribution histogram of the grain size in each region of the ZK60 magnesium alloy after deep drawing to promote dynamic recrystallization was incomparable. Thus, thermoelectric coupling deep drawing, based on ensuring the formability of the ZK60 magnesium alloy, could also ensure its mechanical qualities.

Conclusions
Under ordinary drawing conditions at room temperature, cylindrical parts of ZK60 could not be formed by deep drawing. Thus, increasing the temperature and electric current could effectively increase the ultimate drawing depth of the ZK60 sheet, as well as increase the maximum thinning rate of the drawn parts.
(1) The workpiece with thermoelectric coupling deep drawing at 120 ℃, compared with the hot-drawn workpiece at the same temperature, the limit drawing depth is increased by 0.5 mm, and the maximum thinning rate is increased by 1.1%. When the temperature reaches 200 ℃, compared with the workpiece thermally deepdrawn at the same temperature, the limit drawing depth is increased by 0.7 mm, and the maximum thinning rate is increased by 1.6%. The thermoelectric coupling deep drawing technique could improve the formability of the ZK60 magnesium alloy sheet. (2) After the sheet metal was deep-drawn, the hardness value of the deep-drawn part as a whole showed an increasing trend, with the hardness value increasing as it approached the flange. And each region is significantly higher than the sheet hardness value without deep drawing annealing. (3) After the sheet metal was deep-drawn and deformed by thermoelectric coupling, the grain sizes in the different regions were different, and each region had different degrees of recrystallization. The degree of recrystallization in the punch fillet and cylinder wall was significantly higher than in the other areas.

Data availability
The raw/processed data required to reproduce these findings cannot be shared at this time due to technical or time limitations.
Code availability Not applicable.

Declarations
Ethics approval Not applicable.

Consent to participate Not applicable.
Consent for publication Not applicable.