Inhibition of Hydrogen Evolution by a Bifunctional Membrane between Anode and Electrolyte of Aluminum–Air Battery

The hydrogen evolution reaction of the anode is a severe barrier that limits the further commercial application of Al–air batteries. Therefore, this study introduces a bifunctional membrane for the inhibition of hydrogen evolution in Al–air batteries. The reference to Al2O3@PAN as “bifunctional” means that it has both hydrophobic and anti-corrosion functions. Al2O3 can effectively inhibit the migration of hydroxide ions, and PAN is an excellent hydrophobic material. The bifunctional membrane is placed between the aluminum anode and the electrolyte, which can prevent the invasion of excess water and hydroxide ions, thereby inhibiting the hydrogen evolution corrosion of the anode. Electrochemical tests have confirmed that the corrosion inhibition rate of a bifunctional membrane containing 1.82 wt. % Al2O3@PAN is as high as 89.24%. The specific capacity of Al–air batteries containing this membrane can reach 1950 mAh/g, and the utilization rate of the aluminum anode has reached 61.2%, which is helpful in reducing the waste of aluminum resources. The results prove that the bifunctional membrane has excellent anti-corrosion properties. Bifunctional membranes can also be used to prevent the corrosion of metals in other fields.


Introduction
With the need to reduce traditional fossil fuel usage and environment pollution around the world, the development of clean energy has become extremely important. As a new type of environmentally friendly energy storage device, metal (such as Mg, Zn, Al, etc.)air batteries have received extensive attention from scientists [1,2]. The improvement in catalytic efficiency has improved the electrochemical performance of metal-air batteries and further broadened their application scope [3,4]. As a special fuel cell, the Al-air battery uses oxygen in the air as the material to activate the cathode and aluminum as the anode reactant. It has the advantages of being lightweight, very safe, and having high specific energy [5][6][7]. Al-air batteries have been widely used in the fields of electric vehicles [8], telecommunications [9], medical equipment [10,11], etc. In order to obtain a high energy density, Al-air batteries mostly use alkaline electrolytes. However, alkaline electrolytes can cause strong hydrogen evolution reactions in the anodes [12]. Due to these serious hydrogen evolution reactions, the actual specific capacity of traditional Al-air batteries is less than 60% of the theoretical value [13]. Therefore, inhibiting the hydrogen evolution reaction of the aluminum anode and increasing the specific capacity of the battery are urgent issues that must be addressed.
The main reason for the hydrogen evolution reaction is that excessive hydroxide ions and water molecules undertake side reactions with the aluminum anode [14]: To date, scientists have performed a lot of work on inhibiting the attack of hydroxide ions on the aluminum anode. For example, researchers have used an aluminum alloy instead of pure aluminum to increase the overpotential of the hydrogen evolution reaction [15,16], added corrosion inhibitors to the electrolyte to restrict the migration rate of hydroxide ions [17,18], and coated an anti-corrosion film on the surface of the aluminum anode to prevent direct contact between the hydroxide ions and the anode [19,20]. The above methods do inhibit the corrosion of the aluminum anode to a certain extent. However, according to the equation of the side reaction, it is not only the excess hydroxide ions, but also, and more importantly, the excess water molecules that cause hydrogen evolution corrosion. Most current research is aimed at the inhibition of hydroxide ions, and few studies have addressed the restriction of water molecules.
In relevant anti-corrosion studies in other fields, the importance of inhibiting both water and hydroxide ions simultaneously in order to prevent corrosion has been realized. Cui et al. [21] fabricated a hydrophobic micro-oxidation/zinc stearate composite coating on a Mg alloy through electrodeposition. The hydrophobic coating showed excellent corrosion resistance and could be used for the corrosion protection of mechanical micro-cracks. A hydrophobic Mg(OH) 2 /calcium myristate film was prepared by Zheng et al. [22] to prevent the corrosion of Mg metal parts, and this film could significantly extend the service life of metal parts. However, few studies have been undertaken on bifunctional membranes with both anti-corrosion and hydrophobicity properties for the corrosion protection of the anode in Al-air batteries. In our previous research [23], we prepared a Janus-type membrane with anti-corrosion and hydrophobic functions on the surface of an aluminum anode used in an Al-air battery. The Janus-type membrane showed excellent corrosion resistance, which greatly improved the specific capacity of the Al-air battery. The membrane was composed of two layers of membranes with different functions, and the preparation process was complicated.
Here, we prepare an innovative bifunctional Al 2 O 3 @Polyacrylonitrile (PAN) membrane with both hydrophobic and anti-corrosion functions, which will be used to inhibit the hydrogen evolution reaction. Al 2 O 3 is a commonly used corrosion inhibitor in Al-air batteries [24,25], and is generally added to an electrolyte to inhibit hydrogen evolution corrosion. PAN is a hydrophobic polymer that is often used in the hydrophobic layer of mask filters [26]. Adding a bifunctional membrane with both hydrophobicity and corrosion resistance properties between the anode and the electrolyte could prevents the invasion into the aluminum anode of excessive hydroxide ions and water molecules simultaneously.

Materials
The reagents used to prepare the bifunctional membrane, including Al 2 O 3 nanodispersion (the mass fraction of Al 2 O 3 is 20 wt. %), PAN, N,N-dimethylformamide (DMF), and KOH flakes, were all purchased from Shanghai Macklin Biochemaical Co., Ltd. (Shanghai, China) Nickel foam and aluminum foil were used as the current collectors and metal anode, respectively. For the air cathode, the activated carbon, ether black, MnO 2 nanoparticles, and organic solvent N-methyl-2-pyrrolidone (NMP) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China).

Membrane Preparation
The preparation of the precursor solution is illustrated in Figure 1a. The PAN was added to the DMF solution, and then heated and stirred at 60 • C for 12 h until the solution was uniform (Figure 1c). Subsequently, the Al 2 O 3 nanodispersion was added to the above solution, and stirred at 60 • C for 30 min to obtain a uniform white solution as the precursor  (Figure 1d). In this study, the mass ratios of the Al 2 O 3 nanodispersion, the PAN and the DMF were (0.5, 0.7, 0.9, 1.1, 1.3):1:10, and the mass fractions of Al 2 O 3 were 0.77 wt. %, 1.13 wt. %, 01.48 wt. %, 1.82 wt. %, and 2.14 wt. %, respectively. We filled the precursor solution into the syringe pump at a flow rate of 2.0 mL/h for electrospinning. The substrate was aluminum foil, and the distance between the nozzle and the substrate was 14 cm. The voltage was fixed at 12 kV. The electrospinning process is shown in Figure 1b. After electrospinning, the membrane was dried under natural conditions for more than 12 hours, so that the solvent in the membrane could be completely volatilized. Finally, an Al 2 O 3 @PAN bifunctional membrane was formed. A photo of the prepared bifunctional membrane is shown in Figure 1e. 14 wt. %, respectively. We filled the precursor solution into the syringe pump at a flow rate of 2.0 mL/h for electrospinning. The substrate was aluminum foil, and the distance between the nozzle and the substrate was 14 cm. The voltage was fixed at 12 kV. The electrospinning process is shown in Figure 1b. After electrospinning, the membrane was dried under natural conditions for more than 12 hours, so that the solvent in the membrane could be completely volatilized. Finally, an Al2O3@PAN bifunctional membrane was formed. A photo of the prepared bifunctional membrane is shown in Figure 1e.

Characterization and Electrochemical Tests
The morphological structure of the bifunctional membrane was investigated using a scanning electron microscope (SEM, Su-8010, Hitachi, Tokyo, Japan). The crystal phase of the membrane was analyzed by X-ray diffraction (XRD) (Bruker D8 Advance, Bruker, Karlsruhe, Germany) with a Cu-Kα radiation of 0.1541 nm as the X-ray source. The hydrogen evolution experiment was undertaken in a purpose-built experimental device, as shown in Figure 2a. The aluminum foil with bifunctional membranes on both sides was placed in a 4M KOH solution, and the hydrogen evolution rate was tested at room temperature. The wettability of the membrane was examined by water contact angle (CA) measurements (DSA30, Kruss, Hamburg, Germany) with 3 μL water droplets. The N2 adsorption-desorption isotherms of different membranes were characterized by specific surface area and pore size analysis, using ASAP 2020 (Micromeritics, Norcross, GA, USA).
The electrochemical test was carried out with a three-electrode system on an electrochemical workstation (RST 5000, RST, Suzhou, China). Potentiodynamic polarization was measured at a scan rate of 1 mV/s. Hg/HgO and a platinum wire were used as the reference and counter electrodes, respectively. Electrochemical impedance spectroscopy (EIS) measurements were performed in a frequency range from 100 kHz to 0.01 Hz, and at an AC amplitude of 10 mV. The Al-air batteries were assembled with a sandwich structure; the schematic diagram is shown in Figure 2b. The constant current was discharged using a battery testing system (CT2001A, LAND, Wuhan, China).

Characterization and Electrochemical Tests
The morphological structure of the bifunctional membrane was investigated using a scanning electron microscope (SEM, Su-8010, Hitachi, Tokyo, Japan). The crystal phase of the membrane was analyzed by X-ray diffraction (XRD) (Bruker D8 Advance, Bruker, Karlsruhe, Germany) with a Cu-Kα radiation of 0.1541 nm as the X-ray source. The hydrogen evolution experiment was undertaken in a purpose-built experimental device, as shown in Figure 2a. The aluminum foil with bifunctional membranes on both sides was placed in a 4 M KOH solution, and the hydrogen evolution rate was tested at room temperature. The wettability of the membrane was examined by water contact angle (CA) measurements (DSA30, Kruss, Hamburg, Germany) with 3 µL water droplets. The N 2 adsorption-desorption isotherms of different membranes were characterized by specific surface area and pore size analysis, using ASAP 2020 (Micromeritics, Norcross, GA, USA).
The electrochemical test was carried out with a three-electrode system on an electrochemical workstation (RST 5000, RST, Suzhou, China). Potentiodynamic polarization was measured at a scan rate of 1 mV/s. Hg/HgO and a platinum wire were used as the reference and counter electrodes, respectively. Electrochemical impedance spectroscopy (EIS) measurements were performed in a frequency range from 100 kHz to 0.01 Hz, and at an AC amplitude of 10 mV. The Al-air batteries were assembled with a sandwich structure; the schematic diagram is shown in Figure 2b. The constant current was discharged using a battery testing system (CT2001A, LAND, Wuhan, China).

Characterization of the Al2O3@PAN Membrane
Figure 3a-c show the XRD patterns of the pure PAN membrane, the pure Al2O3 pow der, and the Al2O3@PAN nanofibers. For the pure PAN, the peaks appeared at 17.2° an 30°, which is consistent with the research of Sabetzadeh et al. [27]. For the pure Al2O3, t diffraction peaks appeared at 37.6°, 39.5°, 45.8°, 60.9°, 67.1°, and 76.8°, and can be assigne to the (3 1 1), (2 2 2), (4 0 0), (5 1 1), (4 4 0) and (6 2 0) planes of Al2O3, respectively. T locations of the peaks adhere to the standard JCPDS (79-1558) [28]. When the PAN w mixed with Al2O3, the characteristic peaks of the two did not change significantly, b their sizes were reduced. This implies that Al2O3 was successfully embedded in the pol mer, reducing the crystallinity of the PAN. Similar results were also reported by Pendy al. [29]. Figure     The locations of the peaks adhere to the standard JCPDS (79-1558) [28]. When the PAN was mixed with Al 2 O 3 , the characteristic peaks of the two did not change significantly, but their sizes were reduced. This implies that Al 2 O 3 was successfully embedded in the polymer, reducing the crystallinity of the PAN. Similar results were also reported by Pendy et al. [29]. Figure

Characterization of the Al2O3@PAN Membrane
Figure 3a-c show the XRD patterns of the pure PAN membrane, the pure Al2O3 p der, and the Al2O3@PAN nanofibers. For the pure PAN, the peaks appeared at 17.2° 30°, which is consistent with the research of Sabetzadeh et al. [27]. For the pure Al2O diffraction peaks appeared at 37.6°, 39.5°, 45.8°, 60.9°, 67.1°, and 76.8°, and can be assi to the (3 1 1), (2 2 2), (4 0 0), (5 1 1), (4 4 0) and (6 2 0) planes of Al2O3, respectively locations of the peaks adhere to the standard JCPDS (79-1558) [28]. When the PAN mixed with Al2O3, the characteristic peaks of the two did not change significantly their sizes were reduced. This implies that Al2O3 was successfully embedded in the p mer, reducing the crystallinity of the PAN. Similar results were also reported by Pen al. [29]. Figure    Surface wettability is an important factor affecting corrosion resistance. In this study, the functional membrane was prepared by electrospinning it onto the surface of the aluminum anode. Figure 4a shows the surface wettability values of the aluminum anodes with different coatings. It can be seen from the figure that the surface of pure aluminum is hydrophilic, and its contact angle is about 85 • . After adding the pure PAN membrane to the surface of pure aluminum, it gained obvious hydrophobicity, and the contact angle became about 116 • , mainly due to the high hydrophobicity of PAN. When the membrane was supplemented with a small amount (0.77 wt. %) of Al 2 O 3 , the contact angle reached 133 • . The morphology of the droplet on the surface of the membrane is shown in Figure 4b. As the Al 2 O 3 content in the composite membrane increased, the contact angle gradually decreased. This may be due to the fact that Al 2 O 3 nanoparticles were embedded in the PAN nanofibers, which increased the roughness of the membrane and made the penetration of water easier. Surface wettability is an important factor affecting corrosion resistance. In this study, the functional membrane was prepared by electrospinning it onto the surface of the aluminum anode. Figure 4a shows the surface wettability values of the aluminum anodes with different coatings. It can be seen from the figure that the surface of pure aluminum is hydrophilic, and its contact angle is about 85°. After adding the pure PAN membrane to the surface of pure aluminum, it gained obvious hydrophobicity, and the contact angle became about 116°, mainly due to the high hydrophobicity of PAN. When the membrane was supplemented with a small amount (0.77 wt. %) of Al2O3, the contact angle reached 133°. The morphology of the droplet on the surface of the membrane is shown in Figure  4b. As the Al2O3 content in the composite membrane increased, the contact angle gradually decreased. This may be due to the fact that Al2O3 nanoparticles were embedded in the PAN nanofibers, which increased the roughness of the membrane and made the penetration of water easier. The specific surface area is here obtained using the N2 adsorption-desorption isotherms, which are calculated according to the Brunner-Emmet-Teller (BET) model. The results are shown in Figure 5a. The specific surface area of the pure PAN membrane was 8.57 cm 3 /g, and this increased after adding a small amount of Al2O3. As the Al2O3 content increased, the specific surface area of the membrane decreased. The pore size distribution has been determined based on the Barrett-Joyner-Halenda (BJH) test, and the results are shown in Figure 5b. It can be seen from the figure that the pore size distributions of different membranes were homogeneous. The pore size of the pure PAN membrane was larger. When a small amount of Al2O3 was added to the membrane, the pore size of the membrane was significantly reduced. The pore size of the membrane did not change significantly when the Al2O3 content was 1.13-1.82 wt. %. However, as the Al2O3 content continued to increase, the pore size increased significantly. The surface areas and most common pore sizes of different samples are listed in Table 1. The specific surface area is here obtained using the N 2 adsorption-desorption isotherms, which are calculated according to the Brunner-Emmet-Teller (BET) model. The results are shown in Figure 5a. The specific surface area of the pure PAN membrane was 8.57 cm 3 /g, and this increased after adding a small amount of Al 2 O 3 . As the Al 2 O 3 content increased, the specific surface area of the membrane decreased. The pore size distribution has been determined based on the Barrett-Joyner-Halenda (BJH) test, and the results are shown in Figure 5b. It can be seen from the figure that the pore size distributions of different membranes were homogeneous. The pore size of the pure PAN membrane was larger. When a small amount of Al 2 O 3 was added to the membrane, the pore size of the membrane was significantly reduced. The pore size of the membrane did not change significantly when the Al 2 O 3 content was 1.13-1.82 wt. %. However, as the Al 2 O 3 content continued to increase, the pore size increased significantly. The surface areas and most common pore sizes of different samples are listed in Table 1.
The aluminum anode with a bifunctional membrane was placed in a 4 M KOH solution, and the amount of hydrogen precipitated by corrosion was tested via the air exhaust method. Figure 6 shows the curve of the hydrogen evolution rate in the aluminum anodes. Compared with the pure aluminum anode, the hydrogen evolution rate of the aluminum anode with the functional membrane was significantly reduced. The pure PAN hydrophobic membrane somewhat suppressed the hydrogen evolution corrosion rate. When Al 2 O 3 was added to the membrane, the hydrogen evolution rate was greatly reduced. When the Al 2 O 3 content increased from 0.77 to 1.82 wt. %, the hydrogen evolution rate decreased accordingly. In an alkaline solution, Al 2 O 3 can react with OH − to form (AlO 2 ) − , which will consume part of the OH − and thus reduce its concentration near the surface of the aluminum anode [30,31]. Therefore, although the hydrophobicity of the membrane decreased and the pore size increased, the addition of Al 2 O 3 effectively reduced the level of excess OH − near the aluminum anode. It is worth noting that when the Al 2 O 3 content in the membrane increased to 2.14 wt. %, the hydrogen evolution rate increased. Excessive hydroxide ions became attached to the PAN fibers, and the pore sizes of the membranes thus increased significantly. Under these conditions, too many water molecules, and possibly even hydroxide ions, can penetrate the membrane and reach the surface of the aluminum anode, which will result in a reduction in the corrosion inhibition performance of the membrane.  The aluminum anode with a bifunctional membrane was placed in a 4M KOH solution, and the amount of hydrogen precipitated by corrosion was tested via the air exhaust method. Figure 6 shows the curve of the hydrogen evolution rate in the aluminum anodes. Compared with the pure aluminum anode, the hydrogen evolution rate of the aluminum anode with the functional membrane was significantly reduced. The pure PAN hydrophobic membrane somewhat suppressed the hydrogen evolution corrosion rate. When Al2O3 was added to the membrane, the hydrogen evolution rate was greatly reduced. When the Al2O3 content increased from 0.77 to 1.82 wt. %, the hydrogen evolution rate decreased accordingly. In an alkaline solution, Al2O3 can react with OHto form (AlO2) -, which will consume part of the OHand thus reduce its concentration near the surface of the aluminum anode [30,31]. Therefore, although the hydrophobicity of the membrane decreased and the pore size increased, the addition of Al2O3 effectively reduced the level of excess OHnear the aluminum anode. It is worth noting that when the Al2O3 content in the membrane increased to 2.14 wt. %, the hydrogen evolution rate increased. Excessive hydroxide ions became attached to the PAN fibers, and the pore sizes of the membranes thus increased significantly. Under these conditions, too many water molecules, and possibly even hydroxide ions, can penetrate the membrane and reach the surface of the aluminum anode, which will result in a reduction in the corrosion inhibition performance of the membrane.    Figure 7a shows the potentiodynamic polarization curves of aluminum electrodes with different functional membranes in a 4 M KOH solution. The corrosion reactions of aluminum in alkaline solutions include cathodic and anodic processes, as described in   Figure 7a shows the potentiodynamic polarization curves of aluminum electrodes with different functional membranes in a 4 M KOH solution. The corrosion reactions of aluminum in alkaline solutions include cathodic and anodic processes, as described in Equations (2) and (3) [32,33]:

Corrosion Resistance of the Al 2 O 3 @PAN Membrane
The corresponding corrosion potential (E coor ) and corrosion current density (I coor ) obtained by fitting the curves are listed in Table 2. The Tafel curve shows that the corrosion potential drops after adding the corrosion-resistant membrane. The E coor of pure aluminum was −1.77 V, the E coor of the aluminum electrode with a pure PAN membrane was −1.79 V, and the E coor of the aluminum electrode with a bifunctional membrane (1.48 wt. % Al 2 O 3 @PAN) reached −1.95 V. In the aluminum electrode with a 2.14 wt. % Al 2 O 3 @PAN membrane, the E coor slightly decreased to −1.87 V. This explains why the addition of a functional membrane increases the overpotential of the hydrogen evolution corrosion of the aluminum electrode. Figure 7b shows the dependence of I coor on Al 2 O 3 content. The polarization curve of the aluminum electrode with a pure PAN membrane is suppressed significantly, especially in the cathodic part, indicating that PAN acted as a cathodic inhibitor, and mainly affected the efficiency of the cathodic reaction. When the content of Al 2 O 3 increased from 0.77 to 1.82 wt. %, the I coor decreased from 8.23 to 3.23 mA/cm 2 . The negative shift in the corrosion voltage and the lower corrosion current indicate that the bifunctional membrane suppressed both the cathodic and the anodic process. When the Al 2 O 3 increased to 2.14 wt. %, the I coor increased to 6.41 mA/cm 2 . The main reason for this phenomenon is that more Al 2 O 3 nanoparticles were attached to the PAN nanofibers in the 2.14 wt. % Al 2 O 3 @PAN membrane. As a result, the hydrophobicity of the membrane decreased, and the inhibitory effect against water molecules was limited, resulting in a reduced ability to suppress the hydrogen evolution corrosion of the membrane. This result is consistent with the change in the E coor , and is corroborated by the conclusion of the hydrogen evolution experiment. The corrosion inhibition rate can be calculated by the following equation [34]: The corrosion inhibition rate of the aluminum electrode with a pure PAN membrane was 69.9%, and that of aluminum with a 1.82 wt. % Al 2 O 3 @PAN membrane reached the highest value of 89.24%. The corrosion inhibition rate of the aluminum electrode with a pure PAN membrane was 69.9%, and that of aluminum with a 1.82 wt. % Al2O3@PAN membrane reached the highest value of 89.24%.    Situated between the electrode and the electrolyte, the functional membrane can effectively prevent the direct contact of the aluminum electrode and the electrolyte. The surface of a traditional pure aluminum anode is not modified, and the corrosion damage it permits is thus serious, as shown in Figure 8a. A PAN membrane with hydrophobic properties can significantly limit the invasion of water molecules, and has a certain corrosion inhibition effect, as shown in Figure 8b. A bifunctional membrane can not only inhibit the invasion of hydroxide ions, but can also block excess water molecules, as shown in Figure 8c. Through the above polarization curve analysis, we found that the two components (PAN and Al 2 O 3 ) are indispensable to a bifunctional film. The experimental results show that when the membrane only contains PAN, the corrosion reaction efficiency of the cathodic part can be reduced. After the addition of Al 2 O 3 , the corrosion inhibition rate was significantly improved, indicating that Al 2 O 3 effectively hindered hydroxide ion entry. In the study of Lee et al. [35], Al 2 O 3 was adhered to the surface of a zinc electrode to inhibit its hydrogen evolution corrosion in an alkaline solution. Here, after adding PAN to the anti-corrosion membrane, we achieved a higher corrosion inhibition rate due to the hydrophobic property of PAN.
Membranes 2022, 12, x FOR PEER REVIEW 9 of 14 the study of Lee et al. [35], Al2O3 was adhered to the surface of a zinc electrode to inhibit its hydrogen evolution corrosion in an alkaline solution. Here, after adding PAN to the anti-corrosion membrane, we achieved a higher corrosion inhibition rate due to the hydrophobic property of PAN. The electrochemical behavior of an aluminum electrode with a protective membrane in an alkaline electrolyte was further tested by EIS, in order to obtain more detailed information on the mechanisms and the electrode membrane's structure [36]. Figure 9a shows the Nyquist curve of the aluminum anode in 4M KOH. The impedance spectrum of the aluminum electrode in an alkaline solution shows characteristics of capacitance, indicating that the charge transfer resistance is dominant in the corrosion process [37,38]. Figure  9b shows an equivalent circuit diagram, which is used to fit EIS. In the equivalent circuit, s R is the solution resistance, f R and c Q are the resistance and capacitance of the protective membrane, respectively, ct R is the charge transfer resistance, and dl Q is the electric double layer capacitance. The impedance parameters by fitting the EIS can be seen in Ta  The electrochemical behavior of an aluminum electrode with a protective membrane in an alkaline electrolyte was further tested by EIS, in order to obtain more detailed information on the mechanisms and the electrode membrane's structure [36]. Figure 9a shows the Nyquist curve of the aluminum anode in 4 M KOH. The impedance spectrum of the aluminum electrode in an alkaline solution shows characteristics of capacitance, indicating that the charge transfer resistance is dominant in the corrosion process [37,38]. Figure 9b shows an equivalent circuit diagram, which is used to fit EIS. In the equivalent circuit, R s is the solution resistance, R f and Q c are the resistance and capacitance of the protective membrane, respectively, R ct is the charge transfer resistance, and Q dl is the electric double layer capacitance. The impedance parameters by fitting the EIS can be seen in Table 3. According to the equivalent circuit, the values of R f and Q c express the characteristics of the functional membrane on the surface of the aluminum electrode. The R f of the aluminum electrode with a Al 2 O 3 @PAN membrane is significantly greater than that of the single-functional PAN membrane, indicating that the coexistence of Al 2 O 3 nanoparticles and PAN increases the membrane's resistance and makes it more protective. When the content of Al 2 O 3 ranges 0.77 to 1.82 wt. %, as the Al 2 O 3 fraction increases, the R f value of the composite membrane increases. When the Al 2 O 3 content reaches 2.14 wt. %, due to the reduced hydrophobicity of the membrane, more water molecules can pass through (see the previous discussion of hydrophobicity), and the R f value of the composite membrane will decrease. The charge transfer resistance accurately reflects the protective effect of the functional membrane on the electrode. After adding a functional membrane onto the surface of the aluminum anode, the R ct value became significantly higher than that of the pure aluminum anode. The R ct of the aluminum anode with a bifunctional membrane showed clear improvements. This proves that the Al 2 O 3 @PAN membrane has the best protective effect on an aluminum electrode by effectively hindering the erosion effect of the large number of hydroxide ions and water molecules on the aluminum surface, and its anti-corrosion effect is better than that of a PAN membrane with a single hydrophobic function. The R ct values of the aluminum electrodes with bifunctional membranes rank in the following order: 0.77 wt. % Al 2 O 3 @PAN < 1.13 wt. % Al 2 O 3 @PAN < 1.48 wt. % Al 2 O 3 @PAN < 2.14 wt. % Al 2 O 3 @PAN < 1.82 wt. % Al 2 O 3 @PAN. Its greater charge transfer resistance proves that the composite Al 2 O 3 @PAN membrane can effectively reduce the size of the reaction sites [39] on the surface of the aluminum anode. The R ct value of 1.82 wt. % Al 2 O 3 @PAN is the largest, indicating that its corrosion effect is the best, which is consistent with the results of the polarization test. wt. % Al2O3@PAN is the largest, indicating that its corrosion effect is the best, which is consistent with the results of the polarization test.  Considering the excellent anti-corrosion properties of the bifunctional membrane, we placed it between the aluminum electrode and the electrolyte to protect the anode from self-corrosion. Figure 10a,b show the discharge performance of the assembled Al-air batteries at 3 and 10 mA/cm −2 . It can be seen from the figure that after adding an anti-corro-  Considering the excellent anti-corrosion properties of the bifunctional membrane, we placed it between the aluminum electrode and the electrolyte to protect the anode from selfcorrosion. Figure 10a,b show the discharge performance of the assembled Al-air batteries at 3 and 10 mA/cm −2 . It can be seen from the figure that after adding an anti-corrosion membrane, the discharge time of the battery was significantly prolonged-most notably that of the battery with a bifunctional anti-corrosion membrane. It is worth noting that even when a bifunctional anti-corrosion membrane is added to the battery, its discharge voltage platform remains very similar to that of a battery with a pure aluminum anode. This proves that when a bifunctional membrane is added, the supply of hydroxide ions can still meet the normal discharge requirements of the battery. When the content of Al 2 O 3 in the bifunctional membrane is 1.82 wt. %, the discharge time of the Al-air battery is the longest, indicating that the corrosion inhibition effect of the 1.82 wt. % Al 2 O 3 @PAN composite membrane is the best. tional membrane increases the specific capacity of Al-air batteries effectively. The specific capacity of the Al-air battery with a 1.82 wt. % Al2O3@PAN membrane reached 1950 mAh g −1 at a current density of 10 mA cm −2 . This specific capacity offers clear advantages compared to a traditional Al-air battery with liquid electrolyte. Compared with previous studies [17], this research has obtained higher specific capacities under the same discharge current density. Figure 10d compares the anode utilizations of Al-air batteries with different membranes. The results are consistent with those of the Tafel, EIS and constant current discharge tests. The utilization rate of an aluminum anode with a bifunctional 1.82 wt. % Al2O3@PAN membrane was the highest, at 61.2%. The anode utilization rate obtained when using a bifunctional membrane is comparable to that achieved by organicinorganic inhibitors [17]. This proves that the bifunctional membrane effectively reduces the degree of unnecessary consumption displayed by the aluminum electrode due to selfcorrosion. Adding the bifunctional membrane can significantly increase the anode utilization rate of the Al-air battery, which helps to reduce the waste of aluminum resources, and is thus of great significance to environmental protection and energy saving.    Table 4 summarizes the values. The anode utilization rate of the Al-air battery is determined by the following equation [40,41]: where I represents the current (in A), t is the discharge time (in h), ∆m is the weight loss (in g) and F is the Faraday constant. It can be seen from Figure 10c that the bifunctional membrane increases the specific capacity of Al-air batteries effectively. The specific capacity of the Al-air battery with a 1.82 wt. % Al 2 O 3 @PAN membrane reached 1950 mAh g −1 at a current density of 10 mA cm −2 . This specific capacity offers clear advantages compared to a traditional Al-air battery with liquid electrolyte. Compared with previous studies [17], this research has obtained higher specific capacities under the same discharge current density. Figure 10d compares the anode utilizations of Al-air batteries with different membranes. The results are consistent with those of the Tafel, EIS and constant current discharge tests. The utilization rate of an aluminum anode with a bifunctional 1.82 wt. % Al 2 O 3 @PAN membrane was the highest, at 61.2%. The anode utilization rate obtained when using a bifunctional membrane is comparable to that achieved by organic-inorganic inhibitors [17]. This proves that the bifunctional membrane effectively reduces the degree of unnecessary consumption displayed by the aluminum electrode due to self-corrosion. Adding the bifunctional membrane can significantly increase the anode utilization rate of the Al-air battery, which helps to reduce the waste of aluminum resources, and is thus of great significance to environmental protection and energy saving.

Conclusions
In this study, a bifunctional anti-corrosion membrane for Al-air batteries was prepared by electrospinning. The components of the bifunctional membrane are PAN and Al 2 O 3 nanoparticles. The PAN displays excellent hydrophobicity, and Al 2 O 3 nanoparticles can prevent the attack of hydroxide ions on the aluminum anode. Our composite membrane is placed between the aluminum electrode and the electrolyte, which can effectively inhibit self-corrosion. Using it in Al-air batteries can significantly improve the anode utilization and specific capacity. We found that a higher corrosion inhibition rate can be obtained by increasing the Al 2 O 3 content in the bifunctional membrane appropriately, but too many Al 2 O 3 nanoparticles will affect the hydrophobicity of the membrane, resulting in a decrease in the corrosion inhibition rate. The corrosion inhibition rate of the 1.82 wt. % Al 2 O 3 @PAN membrane reached 89.24% in an alkaline electrolyte, while the specific capacity was 1950 mAh g −1 and the utilization rate of the aluminum anode reached 61.2%. The bifunctional membrane proposed in this research has a low cost, significant effectiveness, simple preparation requirements, and a minimal impact on the discharge of the battery. The membrane can effectively inhibit the hydrogen evolution reaction of the aluminum anode, increase the specific capacity, improve the utilization rate of the aluminum anode, and reduce the waste of aluminum resources. The results of this study offer a new approach to metal corrosion inhibition under alkaline conditions.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.