Effects of intermetallic phases on electrochemical properties of powder metallurgy Mg-6%Al-5%Pb anode alloy used for seawater activated battery

Figure 1 shows the experimental preparation flowchart. Mg powders, Al powders and Pb powders were mixed uniformly in a three-dimensional mixer for 5 h. The mixed powder was then placed in a prepared mold (Φ60 mm × 45 mm) with argon protection. Firstly, in the hot pressing sintering procedure, the mixed powder was held at 380 °C for 30 min under a pressure of 300 MPa. Secondly, the sintered Mg-Al-Pb alloy was heated to 250 °C for 30 min and then extruded into sheets 5 mm in thickness and 50 mm in width. The extruded Mg-Al-Pb sheets were solution-treated at 350 °C for 24 h. Subsequently, the solution-treated Mg-Al-Pb alloys were hot-rolled to sheets with a thickness of 2 mm. During each hot-rolling pass, the Mg-Al-Pb alloy was held at 380 °C for 30 min to maintain the deformability. Finally, the as-rolled Mg-Al-Pb alloy sheets were annealed at 300 °C for 1 h. The chemical composition of the sintered Mg-Al-Pb alloy are shown in table 2.

Zoom In Zoom Out Reset image size

The microstructure of Mg-Al-Pb alloys was observed by optical microscopy (OM, DM4 M) and scanning electron microscopy (SEM,Quanta-200) with energy-dispersive x-ray spectroscopy (EDS). The crystalline phases were identified by x-ray diffraction (XRD, D Max 2550).

The electrochemical performances of the Mg-Al-Pb alloy were measured by a CHI660E electrochemical workstation using a three-electrode system. The working electrode was Mg-Al-Pb alloy, with a platinum plate serving as the counter electrode and a saturated calomel electrode as the reference electrode. For the potentiodynamic polarization curves, the scanning voltage ranges from −2. 4 V to −0. 8 V to ensure that corrosion potentials of each state of Mg alloys are included, and the scanning rate is 1 mV s⁻¹. In hydrogen evolution corrosion test, the anode samples with a diameter of 10 mm and thickness of 2 mm for immersion tests were ground with 1000 grit paper and cleaned with absolute ethanol. Galvanostatic discharge curves were obtained at current densities of 4. 5 mA cm⁻² and 45 mA cm⁻² for 1 200 s. Electrochemical impedance spectra (EIS) were measured after 20 min immersion in electrolyte, and the perturbation amplitude was 5 mV.

Finally, the prototype battery was assembled by using the solution-treated alloy as the anode and PbCl₂ as the cathode. The battery for comparison used a commercial AZ61 sheet as the anode. The anode sheets were 75 mm × 60 mm × 2 mm in size and the current density was 45 mA cm⁻² which mean that the working current of the prototype battery was 2 A. The whole discharge process was evaluated by a battery testing system called Neware Technology Limited (Battery charge and discharge detection equipment manufacturer). The electrochemical tests were carried out using a 3.5 wt% NaCl solution at (25 ± 0.5) °C. Three parallel experiments were conducted to minimize the error in each electrochemical test.

Figure 2 shows the XRD results for the studied Mg-Al-Pb alloys under different states. It is clear that second phases (Mg₁₇ Al₁₂ and Mg₂ Pb) formed during the hot press sintering. The extrusion procedure did not cause the dissolution or precipitation of the second phases. In the solution-treated alloy, the amount and strength of the peaks of the Mg₁₇ Al₁₂ phases and Mg₂Pb phases significantly decrease. This indicates that solution treatment promoted the dissolution of Mg₁₇ Al₁₂ phases and Mg₂ Pb phases. In the rolled state and annealed state, the Mg₁₇ Al₁₂ and Mg₂ Pb phases did not precipitate from the Mg matrix.

Zoom In Zoom Out Reset image size

To further confirm the specific microstructures, optical micrographs and SEM images of Mg-Al-Pb alloys in different states were obtained. Figure 3 shows the microstructures of the sintered Mg-Al-Pb alloy. As indicated by SEM images and the EDS results, the pure Mg₁₇ Al₁₂ phases were found embedded in matrix in the form of single block (point A). The size of Mg₂ Pb phases was much smaller, these intermetallic phases could gather together (point B) or diffused into Mg₁₇ Al₁₂ phases (point C). When too many Mg₂ Pb phases diffuse into one Mg₁₇ Al₁₂ phase, a mixture (point D) of Mg₁₇ Al₁₂ phases and Mg₂ Pb phases was formed. In figure 3(f), it was observed that sintered alloy composed of intermetallic phases, irregular coarse grains and some micropores existing around grain boundaries. The volume shrinkage of micropores was generated by different diffusion rates of metal elements during sintering [19]. The sizes of Mg₁₇ Al₁₂ phases were about 20 μm to 70 μm. In figure 3(g), the grain size of Mg matrix (about 20 μm) was much smaller than Mg₁₇ Al₁₂ phases. Gradient changes of Mg₂ Pb phases could be found in figure 3(g), which indicates that Pb powder produced fine Mg₂ Pb phases (with the size of 100 nm scale) with Mg powder. These fine Mg₂ Pb phases have strong diffusion ability to diffuse into Mg₁₇ Al₁₂ phases, which eventually resulted in the formation of the microstructure corresponding to point D.

Zoom In Zoom Out Reset image size

As showed in figure 4(a), the Mg₂ Pb aggregation stretched into a line shape distributing along the extrusion direction, and some Mg₁₇ Al₁₂ phases crushed into small pieces after extrusion. In addition, extrusion produced a special microstructure in figure 4(a). According to the line swept result, the contents of Al and Pb increased in this microstructure, which means that it was an aggregation of Mg₁₇ Al₁₂ phases and Mg₂ Pb phases. The extrusion stretched the Mg₂ Pb aggregation to contact Mg₁₇ Al₁₂ phases, which caused more Mg₁₇ Al₁₂ phases that were wrapped by Mg₂ Pb phases appearing in the extruded Mg-Al-Pb alloy. During the extrusion, these Mg₂ Pb phases diffused into Mg₁₇ Al₁₂ phases thus resulting in the stress concentration during extrusion and made those Mg₁₇ Al₁₂ phases crushed. Figure 4(b) showed that the extruded alloy exhibited significantly fine and elongated grains, which were caused by partial dynamic recrystallization during the hot extrusion [20]. According to figure 4(c), Mg₁₇ Al₁₂ phases, which were not affected by Mg₂ Pb phases, did not crushed during extrusion. It also can be found in figure 4(c) that the size of crushed Mg₁₇ Al₁₂ phase was affected by the amount of Mg₂ Pb phases.Excessive diffusion of Mg₂ Pb phases could result in the crushing of Mg₁₇ Al₁₂ phase, tailoring fine parts with similar size to those Mg₂ Pb phases. The interdiffusion happened between the fine Mg₁₇ Al₁₂ phases and Mg₂ Pb phases during extrusion, which result the formation of the aggregation of Mg₁₇ Al₁₂ phases and Mg₂ Pb phases in figures 4(a) and (c).

Zoom In Zoom Out Reset image size

The volume fractions of Mg₁₇ Al₁₂ phases and Mg₂ Pb phases in sintered alloy (figure 3(f)) and extruded alloy (figure 4(b)) were determined by ImageJ software. The results show that the volume fraction of Mg₁₇ Al₁₂ phases in sintered alloy is about 2.4%, while the volume fraction of Mg₂ Pb phases is about 0.8%. In extruded alloy, the volume fraction of Mg₁₇ Al₁₂ phases in sintered alloy is about 8.3% and the volume fraction of Mg₂ Pb phases is about 14.1%. This indicates that extruding procedure promoted the precipitation of Mg₁₇ Al₁₂ phases and Mg₂ Pb phases.

As depicted in figure 5(a), the Mg₁₇ Al₁₂ phases and Mg₂ Pb phases formed a mixed microstructure (with size of 50 μm) after solution treatment (figure 5(c)). This mixed microstructure was Mg₁₇ Al₁₂ phase containing Mg₂ Pb phases inside. Moreover, some white spots were found to be distributed homogeneously in the Mg matrix. The EDS result of area E in figure 5(b) shows that these spots are fine Mg₂ Pb phases. This indicates that solution treatment made Mg₂ Pb phases homogeneously distributed in the Mg matrix. From figure 5(c) and figure 5(d), the grain size of Mg matrix increased after solution treatment. The single form of Mg₁₇ Al₁₂ phases and most of Mg₂ Pb phases dissolved into the Mg matrix after solution treatment.

Zoom In Zoom Out Reset image size

Figure 6 shows the optical micrographs of the rolled Mg-Al-Pb alloy and annealed Mg-Al-Pb alloy. Twin crystals were found in rolled Mg-Al-Pb alloy (figure 6(a)). Under the combined action of heat treatment (380 °C) and rolling, the grain size of rolled Mg-Al-Pb alloy slightly increased and remaining intermetallic phases were dissolved. Figure 6(b) showed that rolling procedure generated microcracks along the rolling direction. Figure 6(d) indicated that the amount of the crack reduced after annealing. From figures 6(c) and (d), twin crystals disappeared and grain size obviously increased by annealing treatment.

Zoom In Zoom Out Reset image size

Figure 7(a) gives the potentiodynamic polarization curves of the Mg-Al-Pb alloys, and the corresponding parameters are summarized in table 2. Polarization tests could reflect the corrosion behavior [21]. The corrosion potential (E_corr ) and corrosion current (J_corr ) are common electrochemical parameters that reflect the behavior of corrosion occurrence [22]. Table 3 shows that the sintered alloy exhibit the highest J_corr (773.3 μA cm⁻²) among the Mg-Al-Pb alloys. It is believed to be caused by the excessive intermetallic phases that acted as the cathode during the micro-galvanic corrosion [23–26]. Compared with sintered alloy, extruded alloy with more intermetallic phases showed more negative E_corr . While the J_corr (300.3 μA cm⁻²) of extruded alloy was lower as the finer grain could create more grain boundaries which acted as corrosion barriers thus enhancing corrosion resistance [27, 28]. The E_corr of solution-treated alloy was more positive than those of sintered alloy and extruded alloy, but the J_corr was much lower. It was was attributed to the significant reduction of the intermetallic phases. However, rolled alloy showed more negative E_corr (−1.521 V) and higher J_corr (84.0 μA cm⁻²) than the solution-treated alloy. This indicated that the mixed microstructure in solution-treated alloy could enhance the corrosion resistance. The J_corr of annealed alloy increased sharply since the annealing treatment obviously increased the grain size which decreased the corrosion resistance of grain.

Zoom In Zoom Out Reset image size

The hydrogen evolution corrosion test can reveal the anode efficiency of different anodes [29]. In figure 7(b), hydrogen evolution rate after 10 h of immersion in 3.5 wt% NaCl solution increased in the order of solution-treated alloy < rolled alloy < extruded alloy < annealed alloy < sintered alloy, which is consistent with the order of J_corr of potentiodynamic polarization test. This indicated that the anode efficiency of solution-treated alloy is the highest.

Figure 8 shows the galvanostatic discharge curves of the Mg-Al-Pb alloys at the current densities of 4.5 mA cm⁻² and 45 mA cm⁻². The average discharge potentials of different anodes are shown in table 4. The activation times (time used for discharge potential to reach the most negative) of the sintered and the extruded Mg-Al-Pb alloys are shorter than those of the solution-treated, rolled and annealed Mg-Al-Pb alloys. This could be explained by the improved discharge activity by joint effect of the Mg₁₇ Al₁₂ and Mg₂ Pb phases, which is consistent with the results of the potentiodynamic polarization test. However, the discharge potential suddenly shifted to more positive direction at much higher rates, which caused the average discharge potentials of these two Mg-Al-Pb alloys to be more positive than those of the solution-treated, rolled and annealed Mg-Al-Pb alloys. The average discharge potential of extruded alloy was more positive than that of sintered alloy (table 4), indicating that extrusion decreased the discharge property.

Zoom In Zoom Out Reset image size

At low current density (figure 8(a)), no obvious difference could be found in the curves of the solution treated the as-rolled and the annealed alloys, since most of intermetallic phases in those alloys had dissolved. As seen from figure 8(b), the discharge curve of Mg-Al-Pb alloy is more stable after solution treatment. Previous discussions indicate that solution-treated alloy contains more second phases than rolled alloy and annealed alloy. Thus, the activation time of the solution-treated alloy is relatively shorter than that of the rolled alloy and annealed alloy at a current density of 45 mA cm⁻². According to table 4, the average discharge potential of solution-treated Mg-Al-Pb alloys is the most negative at both low and high current densities, showing the highest discharge properties. This means that the mixed microstructure in solution-treated alloy could further increase the discharge potential.

To further understand the evolution of the microstructure of the studied Mg-Al-Pb alloys during discharge and explain the difference in the electrochemical properties of Mg-Al-Pb alloys, figure 9 characterizes the corrosion morphologies after discharging at 45 mA cm⁻² for 10 s. Many bubble-shape corrosion products can be found in figures 9(a) and (b). The distribution of bubble corrosion products is very close to that of the intermetallic phases. Thus, the bubble corrosion products were from the microbattery. The process of chemical reaction is shown as follows:

$$\begin{eqnarray}&&{\rm{Anodic}}\,{\rm{reaction}}\,{\rm{Mg}}\to {{\rm{Mg}}}^{2+}+2{{\rm{e}}}^{-}\end{eqnarray} \tag{ 1 }$$

$$\begin{eqnarray}&&{\rm{Cathodic}}\,{\rm{reaction}}\,2{{\rm{H}}}_{2}{\rm{O}}+2{{\rm{e}}}^{-}\to {{\rm{H}}}_{2}+2{{\rm{OH}}}^{-}\end{eqnarray} \tag{ 2 }$$

$$\begin{eqnarray}&&{\rm{Product}}\,{\rm{formation}}\,{\rm{Mg}}+2{{\rm{H}}}_{2}{\rm{O}}\to {\rm{Mg}}{\left({\rm{OH}}\right)}_{2}+{{\rm{H}}}_{2}\end{eqnarray} \tag{ 3 }$$

Zoom In Zoom Out Reset image size

As shown in the EDS result (figure 9(g)), the atomic ratio of Mg:O is 1:2. Combining equation (3), it is obvious that the bubble products are Mg(OH)₂ . This indicates that coarse Mg₁₇ Al₁₂ phases and the gathered Mg₂ Pb phases will result in preferential corrosion of the surrounding Mg matrix and corrosion products will accumulate on those phases. The accumulation of Mg(OH)₂ impedes the subsequent discharge behavior and rapidly shifts the discharge potential toward the positive direction. Grain refinement in extruded alloy slowed down the accumulation of corrosion products. Therefore, the amount of Mg(OH)₂ corrosion products in the extruded alloy (figure 9(b)) is less than that in the sintered alloy (figure 9(a)). Fewer corrosion products also mean fewer obstacles to the discharge behavior, which made the discharge potential of the extruded alloy shift to the positive direction at a lower speed than that of the sintered alloy (figure 8).

In figure 9(c), the amount of Mg(OH)₂ further decreases, and some black dots can be found in the solution-treated alloy. Figure 9(d) gives the enlarged images of the red box area in figure 9(c),the black dots (marked by red arrows) in this figure can be identified as corrosion holes. These corrosion holes are from the homogeneously distributed Mg₂ Pb phases. The homogeneously distributed Mg₂ Pb phases will fall off with the dissolution of Mg to form the corrosion holes in figure 9(d). As mentioned earlier, the gathered Mg₂ Pb phases will accelerate the discharge corrosion of surrounding Mg matrix and then accumulate corrosion products to impede the discharge. The homogeneously distributed Mg₂ Pb phases can also promote the discharge corrosion of surrounding Mg matrix, while these fine Mg₂ Pb phases will fall off before the accumulation of Mg(OH)₂ . With the dissolution of Mg matrix and Mg₂ Pb phases shedding, other fine Mg₂ Pb phases and surrounding Mg matrix will contact the electrolyte and participate in discharge, which indicates that those homogeneously distributed Mg₂ Pb phases could continuously promote the corrosion of Mg matrix thus making discharge curve more stable. As shown in figure 9(e), the amount of bubble Mg(OH)₂ increases in the rolled alloy. These newly increased corrosion products are mainly from the dislocations that are introduced by the rolling procedure. These dislocations act as nucleation sites to create Mg(OH)₂ corrosion products and reduce the active electrode area [30]. Figure 9(f) proved that annealing could eliminate dislocations, which is consistent with the optical micrographs.

In conclusion, the discharge curve of the solution-treated alloy is the most stable and the discharge potential is the most negative among all states of Mg-Al-Pb alloys due to the joint effect of the mixed microstructure and homogeneously distributed Mg₂ Pb phases. Thus, the solution-treated alloy exhibits the best comprehensive discharge properties at the galvanostatic discharge of 1200 s.

The electrochemical impedance spectra of the Mg-Al-Pb alloys are shown in figure 10. The Nyquist plots of all Mg-Al-Pb alloys were obtained after soaking in 3.5% NaCl electrolyte for 20 min to obtain a stable open circuit potential. Figure 11 exhibits the equivalent circuit of the Mg-Al-Pb alloys. L₁ is related to hydrogen evolution corrosion [31]. R_s is the solution resistance. R_t represents the charge transfer resistance of the Mg-Al-Pb alloys, and CPE_dl is the double-layer capacitance of the alloy surface [31, 32]. R_L and L₂ are the parameters related to localized corrosion [33]. R_f and CPE_f represent the constant phase angle element and the resistance of corrosion products, respectively [34].

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The equivalent parameters of the studied Mg-Al-Pb alloys under different states are listed in table 5. The R_t of Mg-Al-Pb alloys is arranged as follows: solution treated (21.1 Ω cm²) > rolled (17.6 Ω cm²) > annealed (15.1 Ω cm²) > sintered (14.7 Ω cm²) > extruded (9.9 Ω cm²). The R_t of Mg-Al-Pb alloys was arranged as the same as their E_corr (table 3). Homogeneously distributed Mg₂ Pb phases improved the uniform corrosion of the solution-treated alloy, which made the R_L of the solution-treated alloy (158.5 Ω cm²) the highest in Mg-Al-Pb alloys. A smaller R_f means a lower resistance of the corrosion product film, the annealed alloy without Mg₁₇Al₁₂ phases and dislocations has the lowest R_f (1.8 Ω cm²).

The above electrochemical tests show that solution-treated alloy has the best comprehensive discharge properties among Mg-Al-Pb alloy that were prepared by powder metallurgy. To further comprehend the discharge behavior of solution-treated Mg-Al-Pb alloy compared with other commercial magnesium anode materials, Mg/PbCl₂ battery tests were performed at the current density of 45 mA cm⁻² (2 A of discharge current). Figure 12 shows the Voltage-Time curves of the Mg/PbCl₂ battery using the solution-treated alloy and a commercial AZ61 sheet as anodes.

Zoom In Zoom Out Reset image size

The battery based on the solution-treated alloy anode shows higher voltage and its curve is as stable as that of the battery based on commercial AZ61 anode. This indicates that solution-treated alloy shows much better anode properties than the commercial AZ61 sheet. Thus, the solution-treated Mg-Al-Pb alloy has the potential to serve as an excellent anode in seawater-activated batteries.