Aluminum-copper alloy anode materials for high-energy aqueous aluminum batteries

Aqueous aluminum batteries are promising post-lithium battery technologies for large-scale energy storage applications because of the raw materials abundance, low costs, safety and high theoretical capacity. However, their development is hindered by the unsatisfactory electrochemical behaviour of the Al metal electrode due to the presence of an oxide layer and hydrogen side reaction. To circumvent these issues, we report aluminum-copper alloy lamellar heterostructures as anode active materials. These alloys improve the Al-ion electrochemical reversibility (e.g., achieving dendrite-free Al deposition during stripping/plating cycles) by using periodic galvanic couplings of alternating anodic α-aluminum and cathodic intermetallic Al2Cu nanometric lamellas. In symmetric cell configuration with a low oxygen concentration (i.e., 0.13 mg L−1) aqueous electrolyte solution, the lamella-nanostructured eutectic Al82Cu18 alloy electrode allows Al stripping/plating for 2000 h with an overpotential lower than ±53 mV. When the Al82Cu18 anode is tested in combination with an AlxMnO2 cathode material, the aqueous full cell delivers specific energy of ~670 Wh kg−1 at 100 mA g−1 and an initial discharge capacity of ~400 mAh g−1 at 500 mA g−1 with a capacity retention of 83% after 400 cycles.


REVIEWER COMMENTS
Reviewer #1 (Remarks to the Author): The manuscript studies a new type of Al anode used for aqueous aluminium ion batteries (AIBs). The problem addressed in this work is essential and worth exploring as AIBs show high potential for largescale energy storage applications. In general, the manuscript is well prepared, and the English used in the manuscript is acceptable. The abstract is fine and covers essential elements of the work. The structure of the introduction is fine. The motivation of the research, research gap and research hypothesis are properly covered. In short, this reviewer believes that the manuscript provides a significant contribution to the research field. Publishing the manuscript will benefit the community. However, several points should be addressed before the manuscript can be accepted for publication.
-In Fig. 4a, at the voltage above ~1.8 V, increasing of the oxidation current for E-Al82Cu18//AlxMnO2 was observed. However, in the case of Al//AlxMnO2, such an increase of oxidation current was not observed. Could you clarify this point if there are any parasitic reactions? -The manuscript states that CVs were carried out at scan rates from 0.1 to 3 mV s−1. However, this review can not find the results of CVs at different scan rates. Inclusion of these results with proper discussion is expected.
-The inclusion of parameters of equivalent circuit models (EIS) for each case is expected. Moreover, all EIS spectra should be plotted using symmetrical scale and axis.
-For specific capacity mAh/g and specific rate A/g, please clarify g of what components.
Reviewer #2 (Remarks to the Author): The article "Highly reversible aluminum-copper alloys for sustainable aqueous aluminum batteries" describes a novel aqueous aluminum battery using a lamellar heterostructure of aluminum and aluminum-copper alloy in combination with an AlxMnO2 cathode. The authors demonstrate the advantages of such a microstructured anode alloy in terms of long cycle life with high capacity retention, high specific capacity, and high specific energy, and no dendrite formation. Extensive experimental work and various methods were combined.
The outstanding features of the work concern: -Contribution in particular to aluminum-ion batteries and not to aluminum-graphite batteries.
-An aluminum battery with long cycle life, high capacity retention, and high specific energy. -Introduction of a novel strategy to implement microstructured metal anodes in combination with aqueous electrolytes. -Introduction of special metal alloys (eutectics) as anode materials; eutectics are of particular interest because they form at the lowest temperatures in a phase diagram, enabling to save production energy.
The article describes a thorough and comprehensive investigation using appropriate experimental methods. I cannot see any shortcomings that would prohibit its publication. However, in my opinion, the authors should comment on already existing literature (https://doi.org/10.1016/j.joule.2019.01.005, https://doi.org/10.1038/s41467-020-15478-4). In addition, the authors should explain how they can be sure that Al3+ is an intercalated/de-intercalated species and not an Al-X species. Here, XPS study data from before and after intercalation could contribute to the conclusion. It is also not clear whether new cells/samples were always prepared for each measurement, whether the same cell/sample was used, and how many cells were prepared for each configuration. This is of interest because if a single cell is used, there may be problems with assembly and thus bias in the data and conclusions. A comment on the theoretical capacities of the Al-Cu alloys would also be of interest. I also miss a comparison with the state of the art for aluminumgraphite batteries. In general, there is no commentary/comparison on the state of the art of aluminum batteries and lithium-ion batteries. Also, in my opinion, adding copper to such a battery makes little sense and is not really "sustainable" as copper is widely used in our lives today, especially as green energy and electromobility increase and cables and current collectors are needed. For this reason, the price of copper is already rising. So copper is not the best solution. Moreover, copper reduces the specific energy of the battery in general. Nevertheless, eutectics seem to be advantageous due to their electrochemical performances and the fact that they can be produced at the lowest temperature in a phase diagram. Therefore, the research presented here is of high interest.
In my opinion, both the work and the conclusions are original. Since the aluminum battery is a promising concept with high energy densities and specific energies expected at the cell level, benefiting from large aluminum deposits and an already established infrastructure, progress on this battery is of great interest to a broad community (automakers, policy makers, scientists). Looking at both citation rates and article views, the topic of "aluminum battery" continues to be of growing interest. Since this article describes experimental work, it fills the large gap between theory and application. Therefore, it is timely and of great importance to the field of aluminum-based batteries.
The methods used are appropriate and the quality of the data is convincing. The reporting of data and methodology is, for the most part, sufficiently detailed and transparent to ensure its reproducibility.
The presentation of all data is very clear and aesthetic.
In my opinion, the conclusions and interpretation of the data are robust, valid, and reliable. It would improve the overall presentation if the authors would more fully address the type of ion that is intercalated/de-intercalated.
The references provided are current, appropriate, and balanced in terms of authors, topics, and relation to the research.
The manuscript is written in a clear and focused manner in good English with almost no spelling errors. Therefore, it was a pleasure to read it.
On a more subjective level, I find the article convincing, to the point, very interesting, and well presented. Its scientific quality is very high and the comparison with the existing literature is also given. However, the technology of "eutectic alloys for electrodes for batteries" is not new and was already published (especially by members of the group of authors of this report): https://doi.org/10.1016/j.joule.2019.01.005, https://doi.org/10.1038/s41467-020-15478-4 The supplementary information is detailed and contributes to a better understanding of the article.
Page 2 (line 26), Page 5 (line 86), Page 17 (line 353): "energy density of ~670 Wh kg−1" --> This is the unit of "specific energy" (energy per mass). What is the value for the energy density (energy per volume)? How was the specific energy calculated and at which level? At the cell level or at the electrode level?   How many cells were tested? Was this corrosion always observed? Or could it be that the cell was not carefully assembled? By the way, did you test the stainless steel foil for its electrochemical behavior to rule out influences of corrosion on the electrochemical data?  The manuscript entitled "Highly reversible aluminum-copper alloys for sustainable aqueous aluminum batteries" describes the use of Cu-Al alloy to facilitate the deposition/stripping of Al in an aqueous electrolyte. The reported results are interesting. However, the use of alternative substrates to facilitate Al metal deposition in the aqueous electrolyte has been already proposed in ref.14. Additionally, some significant questions need to be answered before publication: 1) All the study is performed in a two-electrode configuration. The authors should also perform the symmetrical stripping deposition process in three-electrode cells with a reference electrode to evaluate at which potential the process is taking place. 2) A second main point to be addressed is to exclude that the main electrochemical process taking place is not water decomposition, for example, performing the stripping deposition process in a beaker cell to evaluate any bubbling at the electrode. Eventually, the analysis of the generated gas can further indicate a possible side reaction taking place. Please check 10.1002/aenm.202100077.
3) The author should exclude the possibility of copper dissolution in the system. 4) The electrode mass loading of the electrochemical test performed in coin cells is extremely low. Please consider that low mass loading electrodes are not suitable to extrapolate gravimetric capacity values, energy, and power densities (see : "True Performance Metrics in Electrochemical Energy Storage", Y. Gogotsi and P. Simon, Science 2011, 334, 917-918).

Reviewer #1 (Remarks to the Author):
The manuscript studies a new type of Al anode used for aqueous aluminum ion batteries (AIBs). The problem addressed in this work is essential and worth exploring as AIBs show high potential for large-scale energy storage applications. In general, the manuscript is well prepared, and the English used in the manuscript is acceptable.
The abstract is fine and covers essential elements of the work. The structure of the introduction is fine. The motivation of the research, research gap and research hypothesis are properly covered. In short, this reviewer believes that the manuscript provides a significant contribution to the research field. Publishing the manuscript will benefit the community. However, several points should be addressed before the manuscript can be accepted for publication.

Reply:
We thank the reviewer for finding interest and significance of our work. We also appreciate the reviewer for his/her comments and suggestions. Following these valuable and insightful comments/suggestions, we have completely revised the manuscript. The detailed corrections are listed below.
(1) In Fig. 4a, at the voltage above ~1.8 V, increasing of the oxidation current for E-Al82Cu18//AlxMnO2 was observed. However, in the case of Al//AlxMnO2, such an increase of oxidation current was not observed. Could you clarify this point if there are any parasitic reactions?
Reply: We appreciate the reviewer for the constructive comment. Following this comment, we have double-checked and re-performed electrochemical measurements on the basis of full AR-AMB cells of E-Al82Cu18//AlxMnO2 and Al//AlxMnO2. There indeed observes slightly increasing oxidation current in the CV curve of E-Al82Cu18//AlxMnO2 at the voltage above ~1.8 V, compared with that of Al//AlxMnO2 (Figure 4a). This probably results from less polarization of Al82Cu18 alloy anode, which triggers further Al 3+ extraction from AlxMnO2nH2O after the general Al 3+ insertion/extraction processes, i.e., AlxMnO2nH2O + 3(y-x)e  + (y-x)Al 3+  AlyMnO2nH2O. This is attested by XPS and ICP-OES analysis when charged to 1.9 V. According to ICP-OES and XPS analysis, the x value decreases to ~0.9 from the initial ~0.12 (Table R1-1) and the ratio of Mn 3+ and Mn 4+ changes to 26.8: 73.2 from the initial 36.7: 63.3 ( Figure R1-1).
Supplementary Table R1-a. ICP analysis of AlxMnO2nH2O after charged to 1.9 V.
(2) The manuscript states that CVs were carried out at scan rates from 0.1 to 3 mV s 1 . However, this review cannot find the results of CVs at different scan rates. Inclusion of these results with proper discussion is expected.

Reply:
We thank the reviewer for the constructive suggestion, according to which we have supplemented the CV curves of full batteries at scan rates from 0.1 to 3 mV S 1 in Supplementary Figure 16. Based on these CV curves, we have presented proper discussion in text. At the scan rate of 0.1 mV s 1 , the anodic and cathodic peaks of E-Al82Cu18//AlxMnO2 can reach 1.647 and 1.491 V, respectively, with the voltage difference of ~156 mV. Whereas the voltage difference of anodic and cathodic peaks increases to ~673 mV when increasing the scan rate to 3 mV s 1 (Supplementary Figure 16a), it is still smaller than that of Al//AlxMnO2 cell at the scan rate of 0.2 mV s 1 (~863 mV) (Supplementary Figure 16b). These observations indicate the superior rate capability of E-Al82Cu18//AlxMnO2 cell. As shown in Supplementary Figure 16c, it achieves a specific capacity of as high as ~478 mAh g 1 at 0.1 mV s 1 and retains ~249 mAh g 1 at 3 mV s 1 (i.e., the discharge time of 467 s), higher than that of Al//AlxMnO2 cell (208 mAh g 1 ) even at 0.2 mV s 1 (7000 s).
(3) The inclusion of parameters of equivalent circuit models (EIS) for each case is expected. Moreover, all EIS spectra should be plotted using symmetrical scale and axis.
3 Reply: According to this suggestion, we have added the parameters of equivalent circuit models for each EIS spectrum in Supplementary Table 3, 4 and 6. All EIS spectra have been corrected with symmetric scale and axis. These corrections include (4) For specific capacity mAh/g and specific rate A/g, please clarify g of what components.
Reply: Following this suggestion, we have clarified g in specific capacity mAh g 1 and specific rate A g 1 to the loading mass of AlxMnO2 in the cathode. 4

Reviewer #2 (Remarks to the Author):
The article "Highly reversible aluminum-copper alloys for sustainable aqueous aluminum batteries" describes a novel aqueous aluminum battery using a lamellar heterostructure of aluminum and aluminum-copper alloy in combination with an AlxMnO2 cathode. The authors demonstrate the advantages of such a microstructured anode alloy in terms of long cycle life with high capacity retention, high specific capacity, and high specific energy, and no dendrite formation. Extensive experimental work and various methods were combined.
The outstanding features of the work concern: -Contribution in particular to aluminum-ion batteries and not to aluminum-graphite batteries.
-An aluminum battery with long cycle life, high capacity retention, and high specific energy.
-Introduction of a novel strategy to implement microstructured metal anodes in combination with aqueous electrolytes.
-Introduction of special metal alloys (eutectics) as anode materials; eutectics are of particular interest because they form at the lowest temperatures in a phase diagram, enabling to save production energy.

Reply:
We thank the reviewer for his/her positive, insightful and encouraging comments. We also appreciate this reviewer for the constructive suggestions.
Following his/her suggestions and comments, we have comprehensively revised the manuscript. The details can be found below.
( (2) In addition, the authors should explain how they can be sure that Al 3+ is an intercalated/de-intercalated species and not an Al-X species. Here, XPS study data from before and after intercalation could contribute to the conclusion.

Reply:
We appreciate the reviewer for his/her insightful and constructive suggestion. According to this suggestion, we have carried out additional XPS characterizations on AlxMnO2 electrodes after charge/discharge processes. Supplementary Figure 18a,  ( 3)  Therefore, the research presented here is of high interest.

Reply:
We appreciate the reviewer for his/her insightful and constructive comments. We agree with the reviewer that copper has been widely used in our lives today, especially as cables and current collectors. Considering the price of copper is increasing, we are also exploring other alloys of Al with low-cost and abundant elements. This will be presented in the next work. Nevertheless, in this paper we would like to present a model system, i.e., eutectic-composition alloying of Al and Cu (E-Al82Cu18), to demonstrate the concept that periodically aligned metallic/intermetallic Al/Al2Cu galvanic couples to lower the reversible Al stripping/plating overpotential. As a result of the more-noble Al2Cu pairs with the constituent less-noble -Al to form localized galvanic couples to trigger the Al stripping and serves as 2D nanopattern to guide the subsequent Al plating, the E-Al82Cu18 alloy electrode to exhibit exceptional rate capability and long-term stability during Al stripping/plating cycles. Furthermore, the eutectic Al82Cu18 alloy sheet can directly employed as anode of Al-ion batteries, which does not lead to additional use of copper as the current collectors. Reply: We appreciate the reviewer for his/her positive and insightful comments. We also thank him/her for the constructive suggestion, following which we carried out additional XPS characterization on AlxMnO2 cathode after charge/discharge processes.
The detailed results are shown in Supplementary Figure 18,19. As shown in Supplementary Figure 18a, to Mn 2+ . Supplementary Figure 19a,b shows high-resolution Mn 2p, Al 2p XPS spectra of AlxMnO2 after charging to 1.8 V, wherein Al 3+ ion extracted from AlyMnO2 in terms of AlyMnO2nH2O  AlxMnO2nH2O + 3(y-x)e  + (y-x)Al 3+ , with x = ~0.11. The chemical state of Mn becomes Mn 3+ and Mn 4+ with a ratio of 30:70. While for the presence of trace F and S, they are due to the physical adsorption of OTF (CF3SO3) on the surface of electrode during the charge/discharge processes. This is further confirmed by the fact that the contents of F and S do not change remarkably after discharge (Supplementary Figure 18d,e) and charge (Supplementary Figure 19 d,e). These observations demonstrate that only Al 3+ indeed is intercalated/de-intercalated species.  -020-15478-4). Different from these progresses, in this paper we design periodically aligned metallic/intermetallic Al/Al2Cu galvanic couples to circumvent poor rechargeability of aqueous Al-ion batteries, which is essentially impeded by inherent oxide layer of Al anode. By making use of eutectic engineering, we indeed achieve ordered lamellar nanostructure composed of alternating -Al and intermetallic Al2Cu in eutectic Al82Cu18 alloy. Owing to their different corrosion potentials, the less-noble -Al thermodynamically prefers to work as the electroactive material to supply Al 3+ charge carriers, and the more-noble Al2Cu pairs with the constituent -Al to form localized galvanic couples to trigger the Al stripping and 9 serves as 2D nanopattern to guide the subsequent Al plating. This enables exceptionally high Al reversibility at low potentials especially in N2-purged aqueous Al(OTF)3 electrolyte with ultralow oxygen concentration of 0.13 mg L 1 . As a result, the E-Al82Cu18 electrodes exhibit outstanding Al stripping/plating behaviors, with the overpotential of as low as ~53 mV and the Coulombic efficiency of as high as ~100%, for more than 2000 hours. In addition, these two papers have been listed in references.
(9) The supplementary information is detailed and contributes to a better understanding of the article.

Reply:
We appreciate the reviewer for his/her positive comment.  Reply: We appreciate the reviewer for the comment. To identify the production of additional Al2O3 during Al stripping/plating processes, we performed Raman and XPS characterizations on monometallic Al electrode after Al stripping and plating. As shown in Supplementary Figure 9 for Raman spectra, the Al electrode after stripping/plating for 40 cycles displays more intensive Raman bands of Al2O3. This is in sharp contrast with the as-prepared Al electrode, where there is too little Al2O3 on the surface to be detected by Raman spectroscopy. The remarkable change in intensity of characteristic Raman bands of Al2O3 implies that there produces additional or more Al2O3 on Al electrode after Al stripping/plating. This is also verified by XPS analysis in Supplementary Figure 10. Different from the as-prepared Al electrode (Supplementary Figure 10a), in which there observes small characteristic peak of Al 3+ at binding energy of 74.6 eV, the Al electrode has a dominant characteristic peak of Al 3+ after Al stripping/plating cycles due to the formation of additional Al2O3 in both a larger area and a greater thickness (Supplementary Figure 10b). This is different from the monometallic Al immersed in electrolyte for 2 h (Supplementary Reply: We thank the reviewer for the suggestion, according to which we have corrected them in text. (g) Page 13, line 276: "pre-intercalation of hydrated Al 3+ cation." --> How do you know about the "hydrated Al 3+ cation"?
Reply: We appreciate the reviewer for the insightful comment. Following this comment, we have carried out additional O 1s XPS characterization and thermogravimetric analysis (TGA) on as-prepared AlxMnO2nH2O. O 1s XPS analysis demonstrates that there mainly exist three oxygen-containing species, i.e., the O2  in MnO6 lattice, the OH  and the H2O, to correspond to the peaks at the binding energies 11 of 529.8, 530.9 and 533.0 eV (Supplementary Figure 15d). Therein, the latter is assigned to both crystal water and constitution water, which are identified by TGA profile at the temperature below 510 C. As shown in Supplementary Figure 15e, the weight loss below 120 C is attributed to the removal of the crystal water. When increasing temperature from 120 C to 510 C, the corresponding weight loss is ascribed to the constitution water due to the formation of hydrated Al 3+ with a high enthalpy.
Reply: The Al foils were polished by 7000-mesh sandpaper for removing surface oxide.

Reply:
The 0 h N2-purged electrolyte means the as-prepared one in the ambient surrounding.
(j) Figure 1a: Where do the values for this figure come from (references)? What do the lines perpendicular to the axes represent?
Reply: According to this suggestion, we have listed references in supplementary   Table 1. In addition, we have added the units of axes in Figure 1a for a better readability.
(k) Figure 1e: The colored elements are not really visible. Please modify.

Reply:
We thank the reviewer for the suggestion, according to which we have modified them in Figure 1e. Reply: Symmetric battery of monometallic Al usually undergoes poor rechargeability because of parasitic passivating oxide layer and concomitant hydrogen side reactions.
To measure effective EIS spectrum of Al symmetric battery after Al stripping/plating, in this paper we only performed Al stripping/plating for 12 cycles, i.e., 24 h. In addition, "f" has been added in Figure 3.
(q) Figure 4a: After how many cycles? Which peak refers to intercalation, which to deintercalation?
Reply: Figure 4a presents the CV curves of E-Al82Cu18//AlxMnO2 and Al//AlxMnO2 full cells after 9 cycles. Therein, the anodic and cathodic peaks refer to de-intercalation and intercalation of Al 3+ , respectively.
(r) Figure 4b: Which cycle is that?
13 Reply: Figure 4b presents the charge/discharge voltage profiles of E-Al82Cu18//AlxMnO2 and Al//AlxMnO2 full cells at the tenth cycle.
(s) Supplementary Figure 2. XPS analysis of as-prepared E-Al82Cu18 alloy sheets. --> How was the surface cleaned? Please note the "red" shade in (b), which is not red.

Reply:
After cutting E-Al82Cu18 into alloy sheets, they were polished by 7000-mesh sandpaper for direct use as electrodes in symmetric batteries and full cells. Therefore, the surface of E-Al82Cu18 alloy sheets was cleaned only by sandpaper polishing. In addition, we have corrected the shade color in Supplementary Figure 2b.  Figure 8). This is due to hydrogen production, which not only leads to battery bulge and electrolyte leak but also increases the pH value of electrolyte to facilitate the oxidation of Al metal and thus aggravate side reactions. In our symmetric batteries of monometallic Al, we do not use stainless steel foils. Therefore, there is not any corrosion influence of stainless steel on the electrochemical data. This is confirmed by CV measurement of cell with stainless steel and Al foils. As shown in Figure R2-1, there does not observe any evident redox peaks in the voltage window from 0.5 to 1.9 V.
(v) Supplementary Figure 7: What information do the peaks provide?
Reply: In this plot (Supplementary Figure 9)  Reply: In this plot and its caption, we have corrected "Energy density" as "specific energy". It was calculated according to the loading mass of AlxMnO2nH2O in cathode.
In addition, we have listed references in Supplementary  which not only lower their reversible Al stripping/plating overpotential, but also enable the less-noble -Al to work as the electroactive material to supply Al 3+ charge carriers and the more-noble Al2Cu to serve as 2D nanopattern to guide the subsequent Al plating. These regulated Al stripping/plating processes have been demonstrated by ex-situ SEM and EDS mapping in Figure 3a. As a result, the E-Al82Cu18 electrodes exhibit outstanding Al stripping/plating behaviors, with the overpotential of as low as ~53 mV and the Coulombic efficiency of as high as ~100%, for more than 2000 hours. Furthermore, the E-Al82Cu18 electrode still keeps the initial lamella nanostructure even after more than 1000 cycles of Al stripping/plating (2000 h) (Supplementary   Figure 13a). In view of the novelty, excellent performance of eutectic Al-Cu alloys, we wish the reviewer could share our confidence and belief that the work reported in this paper deserves to be published in high-impact Nature Communications.
(1) All the study is performed in a two-electrode configuration. The authors should also perform the symmetrical stripping deposition process in three-electrode cells with a reference electrode to evaluate at which potential the process is taking place.

Reply:
We appreciate the reviewer for his/her constructive suggestion. According to this suggestions, we have additionally performed cycling voltammetry measurements to demonstrate the symmetrical stripping/plating processes in three-electrode cells, in which the E-Al82Cu18, Al2Cu and Al sheets are used as the working and counter electrodes, respectively, and the Al wire as the reference electrode. The detailed results are shown in Supplementary Figure 4. The E-Al82Cu18 electrode exhibits remarkably enhanced symmetric Al stripping/plating behaviors, with an onset potential of as low as 0 V versus Al/Al 3+ and a dramatically enhanced current density. This is in sharp contrast to the intermetallic Al2Cu with strong Cu-Al covalent bonds and the monometallic Al with native oxide layer, which have the onset potentials of Al stripping to reach ~96 and ~172 mV, respectively, along the low current densities.
(2) A second main point to be addressed is to exclude that the main electrochemical process taking place is not water decomposition, for example, performing the stripping deposition process in a beaker cell to evaluate any bubbling at the electrode. Eventually, the analysis of the generated gas can further indicate a possible side reaction taking place. Please check 10.1002/aenm.202100077.

Reply:
We appreciate the reviewer for his/her insightful and constructive suggestions. According to these suggestions, we have carried out additional Al stripping/plating measurements of symmetric E-Al82Cu18 electrodes or monometallic Al electrodes in Swagelok-type cells. On monometallic Al electrodes, there generate many bubbles during the Al stripping/plating at 1 mA cm 2 (Supplementary Figure 7a). Gas chromatography demonstrates that the gas products are mainly H2 due to water decomposition (Supplementary Figure 7c). While for the E-Al82Cu18 electrodes, there does not observe any bubbles (Supplementary Figure 7b). This indicates the main electrochemical process of Al stripping/plating to take place on E-Al82Cu18, not water decomposition. In addition, the literature (10.1002/aenm.202100077) has been listed in references.
(3) The author should exclude the possibility of copper dissolution in the system.

Reply:
We appreciate the reviewer for the suggestion. Following this suggestion, we have carried out additional Al stripping/plating measurements on symmetric E-Al82Cu18 battery, in which 2 M Al(OTF)3 is used as aqueous electrolyte. After Al stripping/plating for 50 cycles, we performed ICP analysis of Al(OTF)3 electrolyte.
There is only 0.0143 mg/L Cu 2+ (2.8610 8 g in the tested symmetric battery) to be detected. This concentration is almost in agreement with the as-prepared 2 M