An Inverse Aluminum Battery: Putting the Aluminum as the Cathode

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Letter

An Inverse Aluminum Battery: Putting the Aluminum as the Cathode Leigang Xue, Sen Xin, John B. Goodenough, and Charles Austen Angell ACS Energy Lett., Just Accepted Manuscript • Publication Date (Web): 25 Apr 2017 Downloaded from http://pubs.acs.org on April 26, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Energy Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

An Inverse Aluminum Battery: Putting the Aluminum as the Cathode Leigang Xue,[a,b] Sen Xin,[b] John B. Goodenough, [b] and C. Austen Angell *[a] a

b

School of Molecular Sciences, Arizona State University, Tempe, AZ 85287;

Texas Materials Institute, The University of Texas at Austin, Austin, TX 78712

Email: *[email protected]

1 ACS Paragon Plus Environment

ACS Energy Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 14

ABSTRACT. Aluminum has long been regarded as a promising anode for energy storage due to its high energy density and low cost, but its application is hindered by the inability of cathodes to provide reversible Al3+ insertion. In contrast, we report the use of Al as cathode allows a rechargeable high-energy battery. The battery comprises a molten sodium anolyte and a molten NaAl2Cl7 catholyte, separated by a NaSICON solid Na+ electrolyte. It is operated at 200 °C to overcome the ceramic separator kinetics and to keep sodium and NaAl2Cl7 in the molten state. Due to the simple composition and trivalence of Al, the sodium anolyte and NaAl2Cl7 catholyte together shows a high energy density of 366 Wh kg-1 although its voltage is only about 1.55 V and only 60% of the capacity can be realized. The high energy density, low-cost and internal safety make this new cell chemistry applicable to the large scale energy storage market.

TOC GRAPHICS

2 ACS Paragon Plus Environment

Page 3 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

Aluminum is light-weight, low-cost, and trivalent. It has, therefore, promoted attempts to create a rechargeable battery with an aluminum anode since it is quite electropositive in aqueous solution (1.66 V vs. SHE).1, 2 On contact with an aqueous electrolyte, aluminum forms a protective oxide film that blocks transfer of Al3+ across it.1, 3 However, non-aqueous media such as ionic liquids4-7 and molten salts8 provide an alternative electrolyte in which aluminum does not form a surface oxide film, and might therefore be practical for developing rechargable batteries. The most promising indications so far come from the use of the room-temperature ionic liquid EMIAlCl4/EMIAl2Cl7 as the electrolyte for reversible electroplating and dissolution of Al. Different kinds of cathode materials have been investigated to match the Al anode in this ionic liquid electrolyte, to make batteries. The use of Mn2O4 as a cathode leads to a 2.65 V battery;4 the use of V2O5 cathode leads to a 0.6 V battery.5 The most recent and eye-catching battery is that reported by Lin et al.6 in which graphite was used as cathode because it can reversibly intercalate/deintercalate AlCl4-: the voltage varies between 2.2 and 1.5 V, but the stated capacity, based on the graphite cathode, is only about 70 mAh g-1. From these works we can see it is difficult to get a high-voltage battery with Al as anode. Morevover, the finding of a high performance cathode for Al3+ intercalation/deintercalation has a long way to go since the trivalent Al3+ ions bind anions that need either to be stripped, or to be accommodated on insertion into a host lattice. This leads us to consider the inverse use of Al, namely as a cathode, based on its relatively low electropositivity compared to alkali metal anodes. Based on aqueous solution potential data the voltage output of a Na-Al cell would be in the neighborhood of 1.05 V but, not surprisingly, we found a higher voltage, 1.55 V in the molten salt medium. This is closer to, but of course less than, the value calculated from the standard free energy change for the simple process:

namely, 1.80 V at 25°C.

ACS Paragon Plus Environment

3


Page 4 of 14

For Al to be useful as a cathode it should be capable of supporting reversible plating and stripping of Al metal and there must be a compatible electrolyte to support alkali ion migration, to and from the anode compartment to maintain charge balance. So far, the simplest and highest energy candidate we can propose is a combination of AlCl3 and NaCl as catholyte. NaAl2Cl7 (mole ratio of AlCl3 : NaCl = 2:1) is reported here as the start composition. By using NaAl2Cl7 as catholyte and molten sodium as anolyte, the battery configuration is as shown in Figure 1. A Na+ conductor, NaSICON, is used to separate the two liquids.

Figure 1. Schematic diagram of the Na-Al battery. Molten NaAl2Cl7 as catholyte, molten sodium as anolyte, and the two liquids are separated by a Na+ conductor separator. In the charged state, both the catholyte and the anolyte are liquid. During the discharge, the liquid volume decreases, as solid NaCl and Al metal are produced. In the recharge process, solid NaCl and Al metal return to the molten sodium and molten NaAl2Cl7 respectively. Valves on the top and bottom are to indicate the possibility of recharging the cell with fresh liquid catholyte and anolyte. ======================== According to the NaCl-AlCl3 phase diagram (Figure 2),9, 10 NaCl and AlCl3 can combine in the solid state up to 158°C as NaAlCl4 while the homogeneous liquid state can persist down to about 110°C, a eutectic temperature, near the composition NaAl2Cl7. Beyond the eutectic composition it is pure AlCl3 (solid at low temperatures and vapor above 185°C) that is the phase in equilibrium with the liquid. NaAl2Cl7 as catholyte obviously provides for a higher capacity than NaAlCl4 because of a


4

Page 5 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

higher Al content, so is preferable. To ensure that the aluminum in NaAlCl4, Tm = 158 °C,11 could also participate in the cell reaction we chose to test our cell at 200 ℃ in our initial experiments, as suggested by the line “abc” on the phase diagram. At this temperature, NaAl2Cl7 is kept in liquid and no AlCl3 gas will be produced.

Figure 2. NaCl-AlCl3 phase diagram, amended to show the composition and state change of the NaAl2Cl7 catholyte during discharge and charge at 200 °C. Overcharge, to the right of point a (NaAl2Cl7), could produce AlCl3 at pressures greater than 1 atm and should be avoided. ======================== The battery is assembled in a charged state thus the first stage of a cycle is a discharge. NaAl2Cl7 will transform to NaAlCl4 first (point a to point b in the phase diagram), as shown in equation (1).

after which the composition on the phase diagram is NaCl: AlCl3 = 4:1 or 20 mol % AlCl3. NaAlCl4 then becomes the catholyte which can continue the discharge as shown in equation (2).

Therefore, the total cell reaction is:


5


Page 6 of 14

The theoretical capacity of the NaAl2Cl7 catholyte can be expressed simply as the number of coulombs generated per g of active material C = nF/M (M the molecular weight in g), but is more commonly reported as the time needed to pass this charge as a current (1 amp = 1coulomb s-1) in some convenient units, usually hours of time at mA current, so that: C = nF/M coulombs g-1 = nF/M As g-1 = nF/M x 1000/3600 mAhg-1

(4)

Then substitution of n = 6, F = 96500, and M = 325.5 g, yields C = 494 mAhg-1 However, the “Al” in the NaAl2Cl7 catholyte must not be fully deposited because some must be left as liquid NaAlCl4 to serve as the electrolyte for Na+ flux in the subsequent re-charge process. That is, according to the phase diagram, this second step discharge can not reach pure NaCl, but has to stop at point c. Therefore, the practical capacity of this catholyte must be lower than 494 mAh g-1, and the final capacity value must be decided by experiment.

Figure 3. Cyclic voltammograms of NaAl2Cl7 catholyte. Pt as reference electrode, scan rate 10 mVs-1, T = 200 °C. ======================== The electrochemical behavior of the NaAl2Cl7 is displayed in the cyclic voltammogram, Figure 3. The scan covers the potential range that includes the reduction to metallic aluminum at the negative


6

Page 7 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

extreme and the oxidation of chloride ion to chlorine gas at the positive extreme. The redox process for the Al/Al(III) couple is simple and reversible. A battery, illustrated in Figure S1, was assembled to provide proof of concept. Sodium was placed on the top, and NaAl2Cl7 as catholyte was absorbed in an Al wool pellet which was placed at the bottom. Working at 200 °C, discharge and charge behaviors at 0.1C were recorded and are shown in Figure 4a. After a short initial high voltage stage of unknown origin, the discharge voltage dropped and stabilized at about 1.5 V. When 296 mAh g-1 had been delivered, corresponding to 60% of the available Al being used, the discharge voltage dropped to about 1.3 V. At this point, the composition in the cathode chamber was NaAlCl4, NaCl and Al. Then we recharged the battery and the voltage was about 1.63 V; however, only 67.5% of the capacity could be recovered, implying that the solid NaCl and Al produced could not fully return to the liquid NaAl2Cl7 state with this battery design. The solid NaCl that formed during discharge did not establish sufficient contact with the remaining NaAlCl4 for the charging process to regenerate all the initial liquid NaAl2Cl7. A second discharge curve yielded an even smaller capacity.


7


Page 8 of 14

Figure 4. Discharge and charge curves of the Na-Al battery at 0.1C with (a) NaAl2Cl7 catholyte; (b) NaAl2Cl7-EMIAlCl4 (4:1 mole ratio) catholyte. The capacity is calculated on the basis of the mass of NaAl2Cl7. ======================== This problem can be ameliorated in different ways, which we now describe. The first is to add a low-melting component such as the "ionic liquid" EMIAlCl4 to the electrolyte to increase the liquid volume and improve the wetting. EMIAlCl4 is inactive in the charge/discharge process and the Al in it cannot be deposited.4 It is always in the liquid state but causes the conductivity of the NaAl2Cl7EMIAlCl4 (mole ratio 4:1) mixture to become lower than for pure NaAl2Cl7 (Figure S2) due to an alkali cation trapping phenomenon.12 The much improved charge/discharge profile with this new catholyte is shown in Figure 4b. The voltage during the initial discharge is around 1.3 V, but in the following charge-discharge, the discharge and charge voltages are 1.56 V and 1.63 V, respectively, which are the same as with the pure NaAl2Cl7 battery (Figure 4a). The energy efficiency (product of Coulomb and voltage efficiencies) is high, about 95.7%, but there is still the trend to loss of capacity with increasing cycle number. A decrease in the charging rate, or an increase in operating temperature, might help with this problem. Of course, the addition of the extra supporting material EMIALCl4 (not optimized) lowers the practical capacity: a better battery design could decrease the quantity needed to maintain the contact between the precipitated solid NaCl, liquid NaAlCl4 and solid electrolyte separator. A second way is to limit the discharge to a very shallow value such that the cell process can be described by the equation

in which no NaCl is generated. However, this obviously would be associated with a drastic decrease in capacity (to 61.8 mAh g-1) so is not to be considered further.


8

Page 9 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

A third way would be to include, in the cathode compartment, a supply of an adduct of AlCl3 with some appropriate (molecular) base chosen to maintain the activity of AlCl3 at about the same value as that in liquid NaAl2Cl7. This would effectively buffer the electrolyte against any NaCl precipitation and remove the volume change in the electrolyte from cell design consideration. The cell design could be modified to use a minimum electrolyte volume, and the capacity would be determined by the amount of AlCl3 in the adduct. The AlCl3 adduct, unlike NaCl, would be a soft solid that would readily release AlCl3 to the electrolyte to maintain an optimum activity for Al deposition. To minimize the effect on the capacity, the base should be simple and low molecular weight, preferably inorganic. Neglecting the base molecular weight for the moment the cell reaction would reduce to the simple transfer of AlCl3 to the electrolyte as NaAl2Cl7 to accommodate the inmigration of Na+, effectively making the cell reaction 3Na + AlCl3 = Al + 3NaCl coupled with the acid-base process using AlCl3 from the bank, viz., 3NaCl + 6AlCl3 = 3NaAl2Cl7 to maintain the liquid state of the electrolyte at 110°C, and give an overall cell reaction 3Na +7AlCl3 = Al + 3NaAl2Cl7 Alternatively, it could be 3Na +4AlCl3 = Al + 3NaAlCl4 if a higher operating temperature, T> 158°C, would be used. Unfortunately, the mass of AlCl3 needed to keep the NaCl in the low temperature (chloroaluminate) liquid state at constant AlCl3 activity would need to be counted in the cell capacity calculation. On the other hand, using an AlCl3 storage bank would simplify the cell design because the volume change (formerly entirely residing in the electrolyte) would now be shared with the volume change in the AlCl3 bank. We have yet to explore this variant of the inverse aluminum cell. From the above considerations we may conclude that, if the cycling performance could be improved by future work on catholyte composition focused on optimized choice of inactive


9


Page 10 of 14

electrolyte additive and cell design, this battery could have many advantages for large scale energy storage,13-15 such as high energy density, low cost, and high safety etc. In Table S1, we provide a comparison with the lithium-ion battery and Na-NiCl2 (ZEBRA) battery. The Na-Al battery can realize a high energy density, about 766 Wh kg-1 (from simple product of the capacity calculated earlier and the steady experimental cell voltage - see details in Supporting Information), for the ideal case where all the Al in the NaAl2Cl7 catholyte can be utilized, and sodium mass is ignored. In the more rational case where the mass of sodium involved in the cell reaction (Eq.3) is included in the energy density assessment, then the theoretical energy density falls to 538 Wh kg-1. Our work (Figure 4) shows that at least 60% of the NaAl2Cl7 catholyte can be utilized, namely, 366 Wh kg-1 which is the value we claim for our battery. These values are close to those for the lithium ion battery (578 Wh kg-1, see SI), based on active cathode mass alone, and 398 Wh kg-1 based on anode and cathode masses combined. ZEBRA battery has a high theoretical capacity (787 Wh kg-1), but this value would be much lower if one were to take the liquid NaAlCl4 electrolyte into account in the calculation (We are not sure how much NaAlCl4 is needed for the NiCl2 solid cathode to function properly). Moreover, because of slow diffusion within the solid NiCl2 formed on the nickel surface, only a part of Ni can be utilized.13 For the Na-Al cell, both the sodium and aluminum are earth-abundant and inexpensive. It should therefore be much cheaper than the lithium-ion battery, and, in view of the greater cost of Ni over Al it should also be cheaper than the ZEBRA battery. In the charged state, both the catholyte and the anolyte are liquid, so they can in principle be replaced after the performance degrades without discarding the whole battery. This would offer a further cost advantage over other battery types. Concerning safety, if the Na+ conducting separator should crack, sodium would react with NaAl2Cl7, to form NaCl and Al, being in this respect as safe as the ZEBRA battery.16-19 In summary, we demonstrate here that molten NaAl2Cl7 can serve as a catholyte and can deliver a high capacity of 296.4 mAh g-1 at 60% depth of discharge; a 1.55 V discharge voltage, which is


10

Page 11 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

higher than expected from aqueous solution potential data, can be obtained with the molten salt medium. By using NaAl2Cl7 as catholyte and molten sodium as anolyte, we show a high energy efficiency Na-Al prototype battery of which we can find no previous record. Although the capacity fades undesirably in the present cell, its high energy density, low cost, high safety and replenishable reserves making the Na-Al battery particularly promising in the grid-storage market. The optimum working temperature, catholyte composition, and battery configuration need to be further researched.

ASSOCIATED CONTENT The supporting information is available. Experimental; The photo and schematic of the Na-Al prototype battery; Conductivities; Comparisons among three kinds of batteries AUTHOR INFORMATION [email protected] Notes The authors declare no competing financial interest. ACKNOWLEDGMENT We are grateful to the DOE/Office of Electricity's energy storage program for Award No 1111357 to Arizona State University where L.X. conceived the project and carried out the measurements, and to Ceramatec for the donation of the NaSICON separators used in our cells. We acknowledge helpful discussions with Dr. Sai Bhavaraju of Ceramatec, Dr. Kang Xu of ARL labs, Bethesda, MD, and with our collaborator Prof. S. W. Martin of Iowa State University.

REFERENCES


11


Page 12 of 14

(1) Li, Q. F.; Bjerrum, N. J. Aluminum as anode for energy storage and conversion: a review. J. Power Sources 2002, 110, 1-10. (2) Cho, Y. J.; Park, I. J.; Lee, H. J.; Kim, J. G. Aluminum anode for aluminum-air battery - Part I: Influence of aluminum purity. J. Power Sources 2015, 277, 370-378. (3) Nakayama, Y.; Senda, Y.; Kawasaki, H.; Koshitani, N.; Hosoi, S.; Kudo, Y.; Morioka, H.; Nagamine, M. Sulfone-based electrolytes for aluminium rechargeable batteries. Phys. Chem. Chem. Phys. 2015, 17, 5758-5766. (4) Brown, G.; Paranthaman, M.; Dai, S.; Dudney, N.; Manthiram, A.; Mclntyre, T.; Sun, X.; High energy density aluminum battery. U.S. Patent 0,082,904, April 5, 2012. (5) Jayaprakash, N.; Das, S. K.; Archer, L. A. The rechargeable aluminum-ion battery. Chem. Commun. 2011, 47, 12610-12612. (6) Lin, M. C.; Gong, M.; Lu, B. G.; Wu, Y. P.; Wang, D. Y.; Guan, M. Y.; Angell, M.; Chen, C. X.; Yang, J.; Hwang, B. J.; Dai, H. J. An ultrafast rechargeable aluminium-ion battery. Nature 2015, 520, 325-328. (7) Xia, S.; Zhang, X. M.; Huang, K.; Chen, Y. L.; Wu, Y. T. Ionic liquid electrolytes for aluminium secondary battery: Influence of organic solvents. J. Electroanal. Chem. 2015, 757, 167175. (8) Li, Q. F.; Hjuler, H. A.; Berg, R. W.; Bjerrum, N. J. Influence of substrates on the electrochemical deposition and dissolution of aluminum in NaAlCl4 melts. J. Electrochem. Soc. 1991, 138, 763-766.


12

Page 13 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Energy Letters

(9) Berg, R. W.; Hjuler, H. A.; Bjerrum, N. J. Phase-diagram of the NaCl-AlCl3 system near equimolar composition, with determination of the cryoscopic constant, the enthalpy of melting, and oxide contaminations. Inorg. Chem. 1984, 23, 557-565. (10) Mohandas, K. S.; Sanil, N.; Mathews, T.; Rodriguez, P., An electrochemical investigation of the thermodynamic properties of the NaCl-AlCl3 system at subliquidus temperatures. Metall. Mater. Trans. B 2001, 32, 669-677. (11) Dewing, E. W., thermodynamics of the system NaCl-AlCl3. Metall. Trans. B 1981, 12, 705-719. (12) Tucker, T. G.; Angell, C. A. Approaches to, and Problems with, Ionic Liquid Electrolytes for Alkali Metal Electrochemical Devices: The Case of Low-Melting Chloroaluminate Binary Solutions. J. Electrochem. Soc. 2014, 161, H796-H801. (13) Bones, R. J.; Teagle, D. A.; Brooker, S. D.; Cullen, F. L. Development of a ni,nicl2 positive electrode for a liquid-sodium (zebra) battery cell. J. Electrochem. Soc. 1989, 136, 1274-1277. (14) Wang, K.; Jiang, K.; Chung, B.; Ouchi, T.; Burke, P. J.; Boysen, D. A.; Bradwell, D. J.; Kim, H.; Muecke, U.; Sadoway, D. R. Lithium-antimony-lead liquid metal battery for grid-level energy storage. Nature 2014, 514, 348-350. (15) Li, H. M.; Yin, H. Y.; Wang, K. L.; Cheng, S. J.; Jiang, K.; Sadoway, D. R. Liquid Metal Electrodes for Energy Storage Batteries. Adv. Energy Mater. 2016, 6, 19. (16) Dustmann, C. H. Advances in ZEBRA batteries. J. Power Sources 2004, 127, 85-92. (17) Lu, X. C.; Li, G. S.; Kim, J. Y.; Mei, D. H.; Lemmon, J. P.; Sprenkle, V. L.; Liu, J. Liquid-metal electrode to enable ultra-low temperature sodium-beta alumina batteries for renewable energy storage. Nat. Commun. 2014, 5, 8.


13


Page 14 of 14

(18) Ellis, B. L.; Nazar, L. F. Sodium and sodium-ion energy storage batteries. Curr. Opin. Solid State Mater. Sci. 2012, 16, 168-177. (19) Benato, R.; Cosciani, N.; Crugnola, G.; Sessa, S. D.; Lodi, G.; Parmeggiani, C.; Todeschini, M. Sodium nickel chloride battery technology for large-scale stationary storage in the high voltage network. J. Power Sources 2015, 293, 127-136


14

An Inverse Aluminum Battery: Putting the Aluminum as the Cathode

Recommend Documents