The effect of pulsed electric current on the structural and mechanical behavior of 6016 aluminium alloy in different states of hardening

This study presents the effect of high current pulses on the structural and mechanical behavior of the 6016 aluminium alloy in three different states of hardening: naturally aged, super saturated, and annealed. The 6016 aluminium alloy was used for the first time in terms of electrically-assisted forming. The influence of the application of different current parameters on the material behavior was conducted. The study of electrically-assisted tensile tests showed that the application of current pulses results in a distinct response of the material, depending on the hardening state. Although in a hardened state, the mechanical properties and plasticity are deteriorated, in the solution treated state, they are improved. For the changes of the material properties is responsible the interaction of the flowing current with the precipitates and the aging process. The new parameters were proposed to describe the distinctions in the material properties between the different states of hardening of the aluminium alloy during the electrically-assisted tension. The material examination was conducted using light and scanning electron microscopy, using also electron backscattered diffraction methods. The application of, for example, the grain orientation spread parameter demonstrated the presence of recrystallized grains, in electrically-assisted specimens.


Introduction
Sheet metal forming processes are commonly used to create car body parts from metallic materials [1]. In the recent years, lightweight materials, such as aluminium alloys, have been widely used in forming processes, and their application has led to the increase in the fuel efficiency of cars [2]. However, despite the many advantages of using aluminum alloys as construction elements, their formability at low temperatures is insufficient, and therefore, warm and hot forming processes are applied. Hot-forming processes provide high formability of the material but also low mechanical properties, and so, additional artificial aging is necessary to improve them. On the other hand, warm forming processes make it possible to maintain high mechanical properties, but, in the case of complicated-shape parts, the formability can be unsatisfactory [3]. The above-mentioned problems, as well as, e.g., low energy efficiency and the material's heterogeneous temperature distribution of these processes, prompt to look for alternative solutions [4]. A promising alternative can be Electrically-Assisted Forming (EAF).
In many studies, it has been reported that the application of electric current (especially pulsed current) to the metal during its deformation can significantly increase its formability [5]. The nonthermal effects that could be responsible for this phenomenon have been called the electroplastic effect [6]. However, there is still no consensus as to whether the electroplastic effect exists or whether the above-mentioned changes are caused only by the thermal effect, such as Joule heat. Nevertheless, Conrad [6] claimed that the electric current flowing through the material generates an electron wind, which affects the dislocations and accelerates their movement. However, it is the case only when the threshold value of current density is achieved. According to Molotskii et al. [7], the magnetic field generated by the current during Electrically-Assisted (EA) deformation facilitates dislocations depinning from the paramagnetic obstacle present in the materials. On the other hand, thermal theories focus on the dissolution of metallic bonds of a lattice structure due to the increased number of electrons and microscale Joule heat [5], as well as on the electron stagnation theory [8]. The microscale Joule heat is caused by the current flowing through the crystal lattice and its defects, such as dislocations and grain boundaries, where the electrical resistivity is lower. After all, there are many examples of electricallyassisted deformation. Roth et al. [9] showed that the elongation of the 5754 aluminium alloy can be increased more than 4 times during pulsed EA tension. Hong et al. [10] observed an increase in strain and a decrease in the applied load during compression of the 6061-T6 aluminium alloy. Many advantages have also been shown during the application of electricity to other plastic deformation processes, such as rolling [11] or wire drawing [12]. In the case of studies on the influence of electricity on the material's behavior in different states of hardening, Simonetto et al. [13] demonstrated that the application of direct current has a greater impact on the 1050 aluminium alloy in the H24 state than in the annealed state. Other studies have also proved various effects of the current on the material's behavior in different states of hardening in the case of the 5xxx [14], 6xxx [15], and 7xxx aluminium alloy [16].
The aim of this work was to determine the influence of pulsed electric current on the mechanical behavior of the 6016 aluminium alloy in different states of hardening. A 6016 aluminium alloy was applied for the first time in terms of electrically-assisted forming. A wide range of completely different states of hardening was applied, in the case of the precipitation hardening alloy, i.e. Naturally Aged (T4), Super Saturated (SS), and Annealed (AN). New parameters, i.e. stress ratio and elongation, for selected points on the stress strain curve were proposed to describe the differences between the states of hardening during the EA tension. Furthermore, the applied strain rate of 0.01 s −1 is higher than the one usually used and closer to the values used in industry. In addition, not only one, which is the usual case, but also a few different electric current parameters were validated. Light and scanning microscopy were applied to register the changes in the microstructure and explain them.

Materials and methods
A commercial 6016 aluminium alloy (AA) sheet with the thickness of 1 mm in a naturally aged state (T4 temper) was selected for the tests. The chemical composition of the as-received material is listed in Table 1 and was measured using a chemical composition analyser emission spectrometer LECO GDS500A.
The testing samples were cut in a dog-bone shape by milling, along the rolling direction of the as-received sheet, with the gage length of 75 mm and the gage width of 12 mm. The following heat treatment was carried out to obtain different states of hardening: heating at 525 ºC for 30 min in the oven and immediate water quenching at 25 ºC were applied to obtain the supersaturated solid solution state and heating at 420 ºC for 120 min and cooling at the rate of 20 ºC h −1 in the oven were applied to obtain the annealed state [2].
Uniaxial tensile tests were performed using an INSTRON 3369 tensile machine at the strain rate of 0.01 s −1 until fracture; however, in the case of the SS state, no later than 20 min after supersaturation. Electric pulses were provided by a self-constructed high current pulse generator, based on the bank of supercapacitors, through the copper electrodes attached to the sample. A function generator AXIOMET AX-DG2010AF and a digital oscilloscope RIGOL DS1052E were applied to create the pulse shape and monitor the current parameters, respectively. The current was measured by Rogowski Coil Power Electronic Measurements UK. To verify the repeatability of the results, three samples were tested for each parameter set. A FLIR T440 infrared thermal imaging camera was used to monitor the sample temperature throughout the whole test. To avoid reflection, the sample surface was coated with black paint. The following current parameters, in the form of t d /t p, where t d is the pulse duration time and t p is the time period, were applied to conduct the electrically-assisted tension: 20 ms/1 s, 50 ms/2.5 s, 100 ms/5 s, 200 ms/10 s, and 300 ms (only one pulse during the whole test). Tensile tests, for each set of current parameters, were carried out in such a way that the first pulse was applied after the time of t p from the beginning of the test. The 2.52 V voltage was applied. The experimental setup containing the above devices can be seen in Fig. 1. Uniaxial tensile tests at the temperature of 200 ºC were performed using a ZWICK 1478 tensile machine at the strain rate of 0.01 s −1 until fracture. On the contrary to the tests conducted on the INSTRON tensile machine, samples with the gage length of 42 mm and the gage width of 8 mm were deformed at room and elevated temperature.
The specimens for the Light Optical Microscopy (LOM) analysis were cut near the fracture surface and etched with 0.5% of hydrofluoric acid reagent and 99.5% of water. The microstructure was observed using an Olympus GX51 optical microscope. The Vickers microhardness was measured by the LECO LM100AT hardness tester under the load of analysis was performed using the above-mentioned SEM equipped with an EBSD detector (Oxford). The applied step size was 0.25 µm. The specimens for the EBSD analysis were mechanically grinded and electrochemically polished using A2 Struers reagent at 24 V and 12 ºC. The confidence index standardization, as well as single-iteration grain dilation cleanup procedures were carried out. The clean-up procedure, performed only one time for each analysis, affected less than 3% of the measurement pixels.

Current − thermal relationships
The application of the above-mentioned high current pulse generator does not allow one to set up the current value. The value of the current flowing through the sample is the result of the resistivity of the material, the dimension of the sample, the voltage of the generator and the current parameters (especially the pulse duration). Therefore, Table 2 presents the influence of the current parameters on the nominal current density (defined as the current value, measured by the Rogowski coil, divided by the transverse section area of the tested sample) and the maximum registered temperature. However, the maximum temperature was generally registered at the end of the test (at the last pulse) and in the middle of the sample, and the average temperature of the sample during the test was definitely lower. The ambient temperature was 25 ºC. Generally, in the case of an electrically-assisted tensile test, the temperature distribution along the sample is nonuniform and an example of it is shown in Fig. 2a for the T4 state.

Tensile properties
As mentioned above, for each set of current parameters, the test was repeated three times. The obtained repeatability is quite good for all the current parameter sets selected in the present study, as shown in Fig. 2b. Some small differences are visible on the stress − strain curves, also because the tensile test and the current generator were launched separately. Figure 3a, b shows the engineering stress − strain curves of AA6016-T4 tested under different current conditions. Figure 3a presents the engineering stress − strain curves with the application of current pulses no longer than 100 ms, while Fig. 3b with the current pulses longer than 100 ms. Simultaneously, in Fig. 3a, b indicated also temperatures occurred during the tests in some characteristic points. In the case of the T4 state, the results show that the application of current pulses does not lead to an increase of the material plasticity, in the form of its elongation, especially when the current pulses are no longer than-100 ms. When current pulse is applied, the flow stress immediately decreases and the temperature increases. However, after the duration of the electric current, the strain hardening behavior manifests itself again, but at a lower level of stress.
The engineering stress − strain curves of AA6016-AN with the pulsed current application are demonstrated in Fig. 4a, b. In this case, the value of elongation after the application of the pulsed current also remained at a lower level than in the case of no such application. However, due to the material behavior after the annealing process, the strain hardening behavior did not occur after the duration of the current pulses (Fig. 4b). It should be noted that, independently of the state of the material, an increase in the material's elongation is observed with the increasing pulse duration, while the ratio of t d /t p remains unchanged. Figure 5a, b shows the engineering stress − strain curves of AA6016-SS with the application of different electrical current pulses. In contrast, in the case of the SS state, after applying the 300 ms current pulse, the material's elongation  increases compared to that without it. In the case of the SS state, after the duration of the electric current, the strain hardening behavior manifests itself again; however, unlike the T4 state, the stress returns to its original value. In addition, not only the Portevin-Le Chatelier (PLC) phenomenon but also suppression of the PLC effect by the pulsed current  is observed. The PLC effect manifests itself as a serration in the tensile test curve and is the result of the interaction between the moving dislocations and the clusters of the solute atoms [16].
In order to better compare the influence of the application of pulsed electric current on the properties of three different states of AA6016, the introduction of two parameters was proposed. The first parameter is the ratio of σ pc / σ wc , where σ pc is the stress value, taken from the engineering stress − strain curve (marked with a black cross on the curves in (Figs. 3b, 4b and 5b), for the current parameters of 200 ms/10 s, 7 s after the application of the first pulse, while, σ wc is the stress value, taken from the engineering stress − strain curve, without the application of current, at the same point as σ pc . The above-mentioned point (value) should be chosen immediately before the next pulse in order to compare the mechanical properties at the lowest possible temperature. However, it was chosen 7 s after the first pulse, due to the limited plasticity of AA6016-T4 (Fig. 3b). The second parameter is the ratio of ε pc /ε wc , where ε pc is the total elongation value, taken from the engineering stress − strain curve, for the current parameters of 200 ms/10 s, while, ε wc is the value of the total elongation, taken from the engineering stress − strain curve, without the application of current.
The above-mentioned parameters have been plotted, as a function of the state of the material, in Fig. 6a (σ pc /σ wc ) and Fig. 6b (ε pc /ε wc ). The correlation between these parameters and the state of the material is clearly seen. In both cases, the lowest value of the parameters is reached in the T4 state, in contrast to the SS state, for which the values of the parameters are the highest. The AN state takes the intermediate values. A similar correlation was obtained for the abovementioned parameters for the EA tensile samples and the 300 ms pulse duration.
To better study the effect of temperature and the effect of pulsed electric current, tensile tests at 200 ºC were conducted (Fig. 7a). For comparison, the temperature distributions throughout the tensile tests of 6016-T4 aluminium alloy, for selected current parameters, are visible in Fig. 7b. However, because the tensile tests at elevated temperature were conducted on a different tensile machine and the shape and dimensions of the tensile specimens were also different, control tests at room temperature were performed. It is clearly seen that there is no elongation increase in the case of the T4 state, however, the flow stress value was dramatically decreased. This decrease in elongation (Fig. 7a) under warm forming conditions of the aluminium alloy was also reported by Camberg [17] and Abolhasani [18]. This behaviour is still unclear, but could be correlated with a switch of the dominant deformation mechanism (dislocation slipping versus climbing) that occurred under warm conditions.
Owing to the microhardness measurements, it is possible to obtain a deeper understanding of the changes that occurred in the tensile properties caused by the pulsed current application. Figure 8a, b show the microhardness profiles of the samples after no current and the EA tensile tests for different current parameters of 200 ms/10 s and 300 ms of a naturally aged as well as solution treated state, respectively. The parameter d in Fig. 8, is the distance from the edge of the fracture, taken from the tensile test, along the rolling direction. Independently of the material state, the hardness close to the edge of the fracture is higher than in the other areas. This behavior is the result of a higher material deformation near the edge. In the case of the naturally aged state (Fig. 8a), the highest hardness was reached in the specimens without the application of current pulses. A completely different situation is observed in the case of the solution treated state (Fig. 8b), where the lowest hardness was measured in the specimens without the application of current pulses.    surfaces of the tested samples, regardless of the hardening state of the material (Figs. 9, 10 and 11 ). In the case of the annealed specimens, strained without current, brittle intergranular separation was observed, in small shells or river line patterns, as well as many short tear ridges (Fig. 9a). A different surface pattern is presented in Fig. 9b for specimens strained with the application of current, where the tear ridges are much longer and the intergranular separation is less frequent. Similar results were obtained in the case of naturally aged specimens (Fig. 10a, b); however, the differences between the specimens received from the EA and non-EA tensile tests are smaller. Moreover, it was observed that, under pulsed electric current, the number of large and deep dimples on the specimen surface is higher than in the case without current. Unlike the previous results, in the case of the solution treated state, the EA specimens (Fig. 11b) present a more significant intergranular separation and smaller tear ridges than those of the non-EA ones (Fig. 11a). In addition, a more regular pattern and a small number of big dimples are observed in specimens strained with the pulsed current application.

Morphology and microstructure results
A microstructure analysis using LOM was conducted in order to verify the differences in the hardness measurements. Figure 12 presents micrographs of naturally aged specimens, taken near the edge of the fracture, when no current was applied (Fig. 12a) and the EA tensile tests for different current parameters of 200 ms/10 s and 300 ms (Fig. 12b, c,  respectively). It is clearly seen that, in the case of EA specimens, there are much smaller precipitates in comparison to the non-EA specimens. In addition, the previously large precipitates look more broken up and divided, as a result of the tensile deformation, in the case of the EA specimens. The platelet-like precipitates visible in Fig. 12 are generally considered to be β phases.
EBSD methods were used for the quantitative identification of the microstructural changes. The presence of three phases was taken into account in the EBSD analysis: aluminium (FCC), β phase (Mg 2 Si) and β'' phase (Mg 5 Si 6 ) [19]. Figure 13a, b present the Inverse Pole Figure (IPF) maps of the naturally aged and solution treated specimens, respectively. The grain size of the T4 state specimen is 38 ± 5 µm and that of the SS state specimen is 41 ± 11 µm. The Grain Orientation Spread (GOS) parameter, which is defined as the average misorientation between all pixels within a grain, was used to identify the newly recrystallized grains [20], as an effect of the electro-pulse application during tension [21]. The β and β'' phases were cut off from the GOS maps. For aluminium alloys, it can be assumed that if the GOS value is less than 2º [22], then the grains can be identified as recrystallized [23]. Figure 14 shows the area (sum of pixels) of the recrystallized grains obtained from the GOS maps for four specimens. It is distinctly seen that, in both cases, i.e. the naturally aged and the solution treated specimens, the area of recrystallized grains increases by about 50% when current is applied during tension. In correspondence to the results from Fig. 14, the GOS maps (with the Band Contrast maps) are shown in Fig. 15(a−d). It is worth mentioning that the specimens for the EBSD analysis were cut from the middle of the tensile samples, which were stopped during the tension immediately after the application of the last current pulse. The phase maps' analysis did not show any unequivocal effect on the number, size, and distribution of the β and β'' precipitates.
The distributions of the misorientation angle of the solution treated specimens are presented in Fig. 16. One can clearly see the increase in misorientation angles of over 15º in the case of the EA specimen with the 300 ms pulse duration (Fig. 16b), when compared with the non-EA specimen (Fig. 16a). The High Angle Grain Boundaries (HAGB) make up 21.3% of all the angles in the case of the EA specimen, while in the case of the non-EA specimen-only 8.01%. At the same time, the mean value of the Geometrically Necessary Dislocation (GND) density parameter was lower in the case of the EA specimen (2.45*10 14 m −2 ) than in the case

Discussion
As mentioned above, the current parameters, especially the pulse duration, play an important role in the EAF processes. For all the states of the material, with an increasing pulse period and a constant t d /t p ratio, an increase in elongation was observed (Figs. 3-5). Simultaneously, even if the total time of the pulses applied, with the same current density, was very similar, an increase of temperature was observed with an increasing pulse duration. This demonstrates that a single pulse must have the right amount of energy. A sufficient value of the applied energy is able to initiate the microstructural changes needed to modify the mechanical properties of the material.
Some other interesting observations were presented in Fig. 6. The application of pulses with approximately the same current parameters in different states of hardening leads to a distinct response of the material. As can be seen, the response of these different materials is caused by the distinct influence of the flowing current on the microstructure in a given state. The ratio of σ pc /σ wc reached the lowest value in the case of the T4 state. In this state, after the application of the long current pulse (Fig. 3b), the stress did not return to its original value, even when the temperature dropped significantly. This behavior clearly shows a change in the microstructure of the naturally aged aluminium alloy as a result of the current application. Microhardness measurements (Fig. 8) were conducted to confirm the above and show that the hardness of the EA stretched specimens is approximately 10% lower as compared to the non-EA specimens. The hardness drop can testify to precipitates dissolution/smash in the naturally aged aluminium. Even when the maximum registered temperature was only 239 ºC, the microscale temperature (as a result of the microscale Joule heat) could be definitely higher, especially the near grain boundaries and the precipitates [24]. The LOM analysis (Fig. 12) of the naturally aged EA tensioned specimens proved that the β precipitates look more smashed and divided. Even if their volume is the same as in the case of the non-EA specimens,  their new condition results in lower mechanical properties. The LOM observations could not reveal the presence of the small, coherent and needle shaped β'' precipitates, which are also responsible for the material strength. In this case, the flowing current and the accompanying rise in temperature could also lead to dissolution and transformation of the β'' precipitates and their clusters.
On the other end of the graph (Fig. 6a) is the highest value of the ratio of σ pc /σ wc for the SS state. In this state, the stress reached its original value despite the fact that the temperature did not drop to the ambient value (Fig. 5b). Furthermore, the microhardness measurement demonstrated, contrary to the T4 state, that the EA samples reached higher values of hardness. The obtained results are consistent with the previous studies [16], in which the increase in current density, during the EA tension, was responsible for the increase in the mechanical properties. The above results attest to the fact that the application of pulsed current, in the solution treated state, improves the aging process, resulting in higher mechanical properties [25]. A result of this aging process improvement could also be the decrease in the dimple size and the length of the tear ridges visible on the fracture surface of the EA tensioned sample (Fig. 11b). However, the LOM analysis (not shown here) did not reveal any microstructural changes in the range of Mg 2 Si precipitates (SS state). Therefore, the applied current pulses had to improve the aging process of the smaller precipitates such as Mg 5 Si 6 . As indicated previously, the application of pulsed electric current during the tension of the solution treated state can improve the aging process. Kim et al. [15] suggested that this phenomenon could be responsible for the disappearance of the PLC effect. However, as can be seen in Fig. 5b, when the current pulse is removed and the temperature drop is progressing, the serrations become visible again on the stress − strain curve. It is commonly known that temperature has a significant impact on the PLC effect. The increase of temperature leads to enhancement of the diffusion processes and thus reduction of the ability to form clusters of the solute atoms (Cottrell atmosphere), which are responsible for the serrations. The presence of the flowing current in the material can improve the above temperature effect and lead to an early disappearance of the serrations [26].
The lowest value of the ε pc /ε wc parameter was reached for the T4 state and the highest for the SS state. The fracture surface analysis revealed (Fig. 10), in the case of the T4 state, an increase in the dimple size and tear ridges length, confirming the increase in the material ductility [27]. These results are consistent with the hardness measurement. The EBSD analysis, especially the GOS parameter, also suggests that the plasticity of the T4 state during the EA tension should be improved; however, its elongation was reduced (Fig. 3). The solution treated state exhibits the opposite properties; regardless of the current parameters, the elongation is higher, especially when the pulse duration is higher than 100 ms. Some factors could be responsible for this state of affairs. Kim et al. [15] reported that in the case of an EA tensioned, artificially aged aluminium alloy, small microvoids occurred around the precipitates. This presence of microvoids, whose number is accelerated by the electric current during deformation, is responsible for the earlier fracture of the sample. Secondly, even if the increase in the area of recrystallized grains is at a similar level in both states (Fig. 14), in the SS state, a variation of the two other parameters was observed in the EBSD analysis. An approximately 10% decrease in the GND density and the KAM parameter was observed in the case of the EA tension specimen compared to that without the current's application. The GND density parameter is a measure of the density of defects and is correlated with the GOS parameter [23]. The fraction of the misorientation angle, in the angle range above 15º, increased about two times in the case of the EA tension specimen compared to that without the current's application. The new recrystallized grains are defined by HAGB with the misorientation angle over 15º [28]. In the case of the T4 state, a change of these three above-mentioned parameters was not observed. Finally, the microscale Joule heat can improve the material's recovery process through a decrease in the dislocation entanglement and an increase in its movement, which leads to its annihilation. However, this process is hindered in the case of precipitation hardening alloys, because the local increase of temperature cannot move the precipitates and the movement of the dislocations is impeded by many obstacles on their way. For the same reason, the effect of electroplasticity can be limited, since the electron wind force acting on the dislocation line will be too low. However, there is still no clear consensus on whether there is an electroplastic effect [29] or not [30].

Conclusions
In the present work, the effect of high current pulses on the mechanical and microstructural behavior of the 6016 aluminium alloy in different states of hardening was studied. The main conclusions are as follows: 1. The current parameters decide about the mechanical properties and material plasticity during the EAF processes. For each state of hardening, for an increasing pulse duration, an increase in the elongation was observed; however, in most cases, the total elongation was still lower than in the tests without the current application. 2. The current flowing through the material generates the microscale Joule heat, which interacts with the precipi-tates present in the material, leading to their destruction and dissolution. Finally, the hardness and strength of the naturally aged aluminium alloy were reduced, as well as its plasticity. However, in the case of the solution treated state, the applied current led to the improvement of the ageing process, without the reduction of its plasticity for the selected current parameters. The application of current pulses led to a decrease in material plasticity in the annealed state. 3. The study proved that the new proposed parameters, i.e. the ratios of σ pc /σ wc and ε pc /ε wc are great in describing the differences between the distinct states of hardening during EA tension. These parameters can be successfully used in other studies regarding the EAF processes of different states of hardening. 4. The EBSD analysis, especially the one using the GOS parameter, showed that the application of high current pulses during tension can lead to recovery and recrystallization processes of the microstructure of the 6016 aluminium alloy. Data availability The data will be available from the corresponding author upon reasonable request.

Competing interests
The authors have no competing interests to declare that are relevant to the content of this article.
Ethics approval This article does not contain any studies with human participants or animals performed by any of the authors.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.