Double-membrane triple-electrolyte redox flow battery design

Abstract

A redox flow battery is provided having a double-membrane (one cation exchange membrane and one anion exchange membrane), triple-electrolyte (one electrolyte in contact with the negative electrode, one electrolyte in contact with the positive electrode, and one electrolyte positioned between and in contact with the two membranes). The cation exchange membrane is used to separate the negative or positive electrolyte and the middle electrolyte, and the anion exchange membrane is used to separate the middle electrolyte and the positive or negative electrolyte. This design physically isolates, but ionically connects, the negative electrolyte and positive electrolyte. The physical isolation offers great freedom in choosing redox pairs in the negative electrolyte and positive electrolyte, making high voltage of redox flow batteries possible. The ionic conduction drastically reduces the overall ionic crossover between negative electrolyte and positive one, leading to high columbic efficiency.

Claims

1. A redox flow battery comprising: a) a cation exchange membrane having a first surface and a second surface; b) an anion exchange membrane having a first surface and a second surface; c) a first electrolyte positioned between and in contact with the first surface of the cation exchange membrane and the first surface of the anion exchange membrane; d) a second electrolyte in contact with the second surface of the cation exchange membrane and a first electrode; and e) a third electrolyte in contact with the second surface of the anion exchange membrane and a second electrode, wherein: the first electrolyte is not in contact with any electrode in its complete flow path; the first electrolyte and second electrolyte are different in terms of at least one species of anion; the first electrolyte and third electrolyte are different in terms of at least one species of cation; and the redox flow battery is a rechargeable battery that is capable of both generating and storing electrical energy.

2. The redox flow battery of claim 1, wherein the second electrolyte is more basic than the first electrolyte, and the first electrolyte is more basic than the third electrolyte.

3. The redox flow battery of claim 2, wherein the second electrolyte comprises an anion-based redox pair or a cation-based redox pair.

4. The redox flow battery of claim 2, wherein the second electrolyte comprises an Al(OH).sub.4.sup./Al redox pair, a Zn(OH).sub.4.sup.2/Zn redox pair or a Co(CN).sub.6.sup.3/Co(CN).sub.6.sup.4 redox pair.

5. The redox flow battery of claim 2, wherein the third electrolyte comprises an anion-based redox pair or a cation-based redox pair.

6. The redox flow battery of claim 2, wherein the third electrolyte comprises a Co.sup.3+/Co.sup.2+ redox pair or a Ce.sup.4+/Ce.sup.3+ redox pair.

7. The redox flow battery of claim 2, wherein the second electrolyte comprises an Al(OH).sub.4/Al redox pair and the third electrolyte comprises a Co.sup.3+/Co.sup.2+ redox pair.

8. The redox flow battery of claim 2, wherein the second electrolyte comprises a Zn(OH).sub.4.sup.2/Zn redox pair and the third electrolyte comprises a Ce.sup.4+/Ce.sup.3+ redox pair.

9. The redox flow battery of claim 2, wherein the second electrolyte comprises a Co(CN).sub.6.sup.3/Co(CN).sub.6.sup.4 redox pair and the third electrolyte comprises a Co.sup.3+/Co.sup.2+ redox pair.

10. The redox flow battery of claim 2, wherein the second electrolyte comprises a Zn(OH).sub.4.sup.2/Zn redox pair, and the third electrolyte comprises a Fe.sup.3+/Fe.sup.2+ redox pair.

11. The redox flow battery of claim 2, wherein the first electrode is a negative electrode and the second electrode is a positive electrode; or wherein the first electrode is a positive electrode and the second electrode is a negative electrode.

12. The redox flow battery of claim 1, wherein the first electrode is a negative electrode and the second electrode is a positive electrode; or wherein the first electrode is a positive electrode and the second electrode is a negative electrode.

13. The redox flow battery of claim 1, wherein the second electrolyte comprises an anion-based redox pair or a cation-based redox pair.

14. The redox flow battery of claim 1, wherein the second electrolyte comprises an Al(OH).sub.4.sup./Al redox pair, a Zn(OH).sub.4.sup.2/Zn redox pair or a Co(CN).sub.6.sup.3/Co(CN).sub.6.sup.4 redox pair.

15. The redox flow battery of claim 1, wherein the third electrolyte comprises an anion-based redox pair or a cation-based redox pair.

16. The redox flow battery of claim 1, wherein the third electrolyte comprises a Co.sup.3+/Co.sup.2+ redox pair or a Ce.sup.4+/Ce.sup.3+ redox pair.

17. The redox flow battery of claim 1, wherein the second electrolyte comprises an Al(OH).sub.4.sup./Al redox pair and the third electrolyte comprises a Co.sup.3+/Co.sup.2+ redox pair.

18. The redox flow battery of claim 1, wherein the second electrolyte comprises a Zn(OH).sub.4.sup.2/Zn redox pair and the third electrolyte comprises a Ce.sup.4+/Ce.sup.3+ redox pair.

19. The redox flow battery of claim 1, wherein the second electrolyte comprises a Co(CN).sub.6.sup.3/Co(CN).sub.6.sup.4 redox pair and the third electrolyte comprises a Co.sup.3+/Co.sup.2+ redox pair.

20. The redox flow battery of claim 1, wherein the second electrolyte comprises a Zn(OH).sub.4.sup.2/Zn redox pair, and the third electrolyte comprises a Fe.sup.3+/Fe.sup.2+ redox pair.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 shows a conventional single-membrane double-electrolyte RFB.

(2) FIG. 2 shows an illustration of a double-membrane triple-electrolyte (DMTE) RFB concept.

(3) FIG. 3 shows a double-membrane triple-electrolyte RFB.

(4) FIG. 4 shows a standard electrode potential for some redox pairs and cell voltage for different batteries.

(5) FIG. 5 shows the open circuit voltage of an Al/Co-DMTE-RFB.

(6) FIG. 6 shows the charge and discharge curves of an Al/Co-DMTE-RFB.

(7) FIG. 7 shows the open circuit voltage of a Zn/Ce-DMTE-RFB.

(8) FIG. 8 shows the charge and discharge curves of a Zn/Ce-DMTE-RFB.

(9) FIG. 9 shows a continuous charge-discharge test of a Zn/Ce-DMTE-RFB for 10 cycles.

(10) FIG. 10 shows the efficiency calculation of a Zn/Ce-DMTE-RFB for each cycle.

DETAILED DESCRIPTION OF THE INVENTION

(11) The present invention will now be described in detail with reference to a few preferred embodiments thereof. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. It will be apparent, however, to one skilled in the art, that the present invention may be practiced without some or all of these specific details. In other instances, well known process steps and/or structures have not been described in detail in order to not unnecessarily obscure the present invention.

(12) The double membrane, triple electrolyte (DMTE) RFB systems described herein can dramatically increase cell voltages and decrease ionic crossover simultaneously by involvement of a double-membrane arrangement (one piece of CEM and one piece of AEM) that divides the RFB cell into three compartments filled with triple-electrolyte (one in contact with a negative electrode, one in contact with a positive electrode, and the third one in between the two membranes). By introducing one more ion-selective membrane, the two electrolytes in the negative side and the positive side can be substantially separated but still remain ionically conductive by the third electrolyte positioned between the membranes. This particular three-compartment design favorably brings great freedom in selecting redox pairs as well as their supporting electrolytes for both the negative side and the positive side, making high cell voltage RFBs possible. Besides providing the function of ionic conduction, the middle electrolyte in between also serves as a great buffer that can significantly reduce the overall counter-ion crossover between the negative side and the positive side, fundamentally solving the electrolyte contamination problem and providing great convenience for electrolyte separation and rebalance.

(13) Specifically, the double-membrane triple-electrolyte design described herein allows for a strongly basic negative electrolyte (high pH, e.g. at least 8, at least 9, at least 10 or higher) and a strongly acidic positive electrolyte (low pH, e.g., not more than 6, not more than 5, not more than 4 or lower) to be used at the same time in the same redox flow battery, where a neutral middle electrolyte is in between. As a result, very negative redox pairs that are usually only stable in basic electrolytes and very positive ones that are usually only stable in acidic electrolytes can be simultaneously incorporated into the DMTE-RFB, providing very high cell voltage and very low ionic crossover at the same time. Not only can the cell voltage be increased, the side reaction of hydrogen evolution in negative electrode can also be suppressed, as the standard electrode potential for hydrogen evolution reaction is very negatively shifted in a basic electrolyte in comparison with an acidic electrolyte (e.g., from 0 V at pH=0 to 0.828 V at pH=14), thermodynamically extending the operational window of cell voltage. Such a low counter-ionic crossover design overcomes the very challenging electrolyte contamination problems that hamper the commercial use of most of the current batteries.

(14) FIG. 2 shows a DMTE-RFB system 200 wherein first electrolyte 60 may be partially surrounded by a second electrolyte 50 and a third electrolyte 70, wherein first electrolyte 60 may be separated from second electrolyte 50 by a first membrane 80, such as a cation exchange membrane (CEM) 80, and wherein first electrolyte 60 may be separated from third electrolyte 70 by a second membrane 90, such as an anion exchange membrane (AEM) 90. Second electrolyte 50 may be partially surrounded by a first electrode 35, such as a negative electrode (anode) 35. Third electrolyte 70 may be partially surrounded by a second electrode 45, such as a positive electrode (cathode) 45.

(15) FIG. 3 shows a DMTE-RFB 300 herein, and similar to FIG. 2, first electrolyte 60 may be partially surrounded by a second electrolyte 50 and a third electrolyte 70, wherein first electrolyte 60 may be separated from second electrolyte 50 by a first membrane 80, such as a cation exchange membrane (CEM) 80, and wherein first electrolyte 60 may be separated from third electrolyte 70 by a second membrane 90, such as an anion exchange membrane (AEM) 90. Second electrolyte 50 may be partially surrounded by a first electrode 35, such as a negative electrode (anode) 35. Third electrolyte 70 may be partially surrounded by a second electrode 45, such as a positive electrode (cathode) 45. Second electrolyte 50 is flowed from second electrolyte source 51 via pump 17. Third electrolyte 70 is flowed from third electrolyte source 71 via pump 19. First electrolyte 60 is flowed from first electrolyte source 61 via pump 18. DMTE-RFB 300 may be connected to a grid input/output processor 11.

(16) It will be understood to those skilled in the art that elements 35 and 45 are referred as electrodes, but they may also include current collectors (not shown). The current collectors may be the same or different material as the electrodes. It will be understood to those skilled in the art that electrodes/current collectors 35, 45 may have high specific surface area (e.g., be highly porous).

(17) First, second and third electrolytes 60, 50, 70 are not particularly limited and may comprise any suitable electrolyte or salt, such as those based on cations of hydronium, sodium, magnesium, potassium or calcium, or anions based on hydroxide, perchlorate, sulfate, phosphate, acetate, chloride, bromide or carbonate. First and second electrodes 35, 45 are not particularly limited and may comprise any suitable electrode material, such as Al, Zn, Cu, Cd, Pb and C.

(18) The DMTE-RFB systems described herein have great advantages over conventional single-membrane, double-electrolyte RFBs and offers high OCV, low ionic crossover, and suppressed hydrogen evolution. The materials used to construct the DMTE battery systems described herein are not particularly limited and may be a myriad of materials, for example, any materials selected from conventional or otherwise known materials used for similar purposes in the energy arts. Such materials include, but are not limited to, cation exchange membranes, anion exchange membranes, electrolyte solutes and solvents, compounds capable of providing the desired redox pairs, acids, bases, negative electrodes, positive electrodes, and the like. The DMTE-RFB systems described herein have a wide range of applications, especially for high voltage and low ionic crossover RFBs.

(19) For example, an attractive candidate for a RFB system is an aluminum-cobalt DMTE-RFB system (e.g., Al/Co-DMTE-RFB), configured as [(Al/Al(OH).sub.4.sup.)//(CO.sup.3+/CO.sup.2+)]. When compared to FIG. 2 or 3, the Al portion of the Al/Co-DMTE-RFB is comprised in second electrolyte 50 and the Co portion of the Al/Co-DMTE-RFB is comprised in third electrolyte 70. The Al/Co-DMTE-RFB system offers a very high cell voltage (4.29 V standard cell voltage), as it successfully combines the very negative redox pair of Al/Al(OH).sub.4.sup. (2.337 V standard electrode potential) in base and the very positive redox pair of Co.sup.3+/Co.sup.2+ (+1.953 V standard electrode potential) in acid. Such a high standard cell voltage (4.29 V) is believed to be the highest one reported among all known RFB systems, which value is 1.7 times that of Zn/Ce-RFB systems (2.50 V), 3.2 times that of Polysulfide-bromide S/Br-RFB systems (1.36 V), 3.4 times that of All-Vanadium RFB systems (1.26 V), and 3.6 times that of Iron-Chromium Fe/Cr-RFB systems (1.18 V), as shown in FIG. 4. The very high standard cell voltage of the Al/Co-DMTE-RFB system is even higher than that of lithium ion batteries (around 3.5 V), suggesting a great potential of the design to rival other RFB technologies.

(20) The test of open circuit voltage (OCV) for the Al/Co-DMTE-RFB system is shown in FIG. 5. As can be seen, a high initial OCV (3.54 V) and a stable OCV (3.28 V) after 20 min is realized. Considering the very sluggish kinetics observed for the Co.sup.3+/Co.sup.2+ redox pair previously and the completely non-optimized electrodes used, these OCV data are believed to be fairly consistent with the standard cell voltage (4.29 V). Additionally, the Al/Co-DMTE-RFB system is functional, as shown by the charge/discharge curves of FIG. 6. The expected discharge reactions and charge reactions are shown in Eq. 1 and Eq. 2, respectively.
Co.sup.3++e.sup..fwdarw.Co.sup.2+Eq. 1 (a)
Al+4OH.sup..fwdarw.Al(OH).sub.4.sup.+3e.sup.Eq. 1 (b)
Co.sup.2+.fwdarw.Co.sup.3++e.sup.Eq. 2 (a)
Al(OH).sub.4.sup.+3e.sup..fwdarw.Al+4OH.sup.Eq. 2 (b)

(21) For the discharge process (the lower curve in FIG. 6), the cell voltage slightly and smoothly decreases from the initial 3.11 V to 2.84 V for 10 min of discharge. An even longer time of discharge is also possible although only 10 min of discharge operation is shown here as a preliminary experiment. The charge process has also been tried, and the cell voltage increases from the initial 3.25 V to 3.59 V for 10 min of charge (upper curve of FIG. 6). Clearly, the experimental demonstration of the Al/Co-DMTE-RFB system confirms that such a design is feasible and successful.

(22) For example, another attractive candidate for a RFB system is a zinc-cerium DMTE-RFB system (Zn/Ce-DMTE-RFB), configured as [(Zn/Zn(OH).sub.4.sup.2)//(Ce.sup.4/Ce.sup.3+)]. When compared to FIG. 2 or 3, the Zn portion of the Zn/Ce-DMTE-RFB is comprised in second electrolyte 50 and the Ce portion of the Zn/Ce-DMTE-RFB is comprised in third electrolyte 70.

(23) The Zn/Ce-DMTE-RFB system offers a standard cell voltage of 2.96 V, as it combines the negative electrode potential (1.216 V) from the Zn/Zn(OH).sub.4.sup.2 redox pair and the positive one (+1.743 V) from the Ce.sup.4+/Ce.sup.3+ redox pair. Such a high standard cell voltage is also higher than those of all conventional aqueous RFB systems, e.g., higher than that of AV-RFB system (1.26 V) and that of Zn/Ce-RFB system (2.50 V, in spite of the strong concern of hydrogen evolution in negative electrode for Zn/Ce-RFB system). The discharge and charge reactions are represented in Eq. 3 and Eq. 4, respectively.
Ce.sup.4++e.sup..fwdarw.Ce.sup.3+Eq. 3 (a)
Zn+4OH.sup..fwdarw.Zn(OH).sub.4.sup.2+2e.sup.Eq. 3 (b)
Ce.sup.3+.fwdarw.Ce.sup.4++e.sup.Eq. 4 (a)
Zn(OH).sub.4.sup.2+2e.sup..fwdarw.Zn+4OH.sup.Eq. 4 (b)

(24) During charge process, the zincate anions are reduced to zinc metal and the sodium cations are balanced from the middle compartment to the negative compartment. In the meanwhile, cerium(III) cations are oxidized into cerium(IV) and the perchlorate anions are balanced from the middle compartment to the positive compartment. During the discharge process, the opposite reactions and ion transfer directions will apply.

(25) After being charged to reach a state of charge of 90%, the OCV is monitored for 15 minutes. As seen in FIG. 7, it shows an initial OCV of 3.14 V and quickly stabilizes to 3.10 V. These OCVs are higher than the standard one (2.96 V), which is reasonable since the cell is in charged state (90% of state of charge). Clearly, such a high observed OCV again confirms and verifies that the DMTE-RFB system is feasible and successful.

(26) Equally important, both a discharge operation and a charge operation have been successfully achieved with a constant current (60 mA current or 5 mA/cm.sup.2 current density). As seen in FIG. 8, the cell voltage, in the charge operation (upper curve of FIG. 8), slightly and very smoothly increases from initial 3.08 V to 3.17 V after 30 min of charge, indicating very high voltage efficiency (93%-96%). Different from the conventional Zn/Ce-RFB system where hydrogen evolution was found to be a strong concern (760 mV over-potential when acidic electrolyte used), the hydrogen evolution reaction is greatly suppressed in the Zn/Ce-DMTE-RFB system of the present invention as its over-potential drops from 760 mV in Zn/Ce-RFB to 388 mV in the inventive Zn/Ce-DMTE-RFB system (1.216 V of Zn/Zn(OH).sub.4.sup.2 redox pair vs. 0.828 V of OH.sup./H.sub.2 at pH=14). Indeed, the hydrogen evolution phenomenon has not been found during the whole discharge operation as well as during the charge operation. In the discharge operation (lower curve of FIG. 8), the cell voltage decreases from the initial 2.88 V to 2.72 V after nearly 4 hours of discharge, showing a very steady voltage region, and then drops sharply after available species being mostly consumed. The voltage efficiency for discharge ranges from 92%-97%, almost equivalent to this number for charge.

(27) The discharge duration lasts for 3 hours and 56 minutes, very close to the charge duration 4 hours, indicating high Coulombic efficiency. The overall Coulombic efficiency, voltage efficiency and energy efficiency are calculated in Table 1. Combining the increased cell voltage, decreased ionic crossover, and suppressed hydrogen evolution, the Zn/Ce-DMTE-RFB system of the present invention is clearly superior to the conventional Zn/Ce-RFB system.

(28) TABLE-US-00001 TABLE 1 Efficiency calculation for one charge-discharge cycle Discharge Charge Average discharge Average charge Columbic Voltage Energy time (s) time (s) Voltage (V) voltage (V) efficiency efficiency efficiency 13777 14400 2.82 3.12 96% 90% 86%

(29) The charge-discharge voltage curve at 5 mA/cm.sup.2 is shown in FIG. 9, showing 10 successful continuous cycles without obvious Coulombic efficiency, voltage efficiency and energy efficiency change (as shown in FIG. 10 of efficiency for each cycle).

Example 1

(30) An aluminum-cobalt DMTE-RFB system (Al/Co-DMTE-RFB), configured as [(Al/Al(OH).sub.4.sup.)//(Co.sup.3+/Co.sup.2+)] was constructed. A three-compartment cell made up of three plastic jars was designed and used as follows. Three 50 ml plastic jars were put in series with a hole (a quarter inch of diameter) opened between adjacent two jars. The three jars, based on half-reaction inside, were assigned as negative, middle and positive compartments. One piece of Nafion 212 membrane (DuPont, 50 m thickness) and one piece of Fumasep FAA membrane (FuMa-Tech, 70 m thickness) were used as the CEM and AEM, respectively. The CEM is put between the negative compartment and the middle compartment while the AEM is put between the middle compartment and the positive compartment, along with an O-ring to seal the conjunction part. Two clamps were used to compress three jars tightly to avoid electrolyte leakage. A potentiostat/galvanostat (Solartron 1287A) was used in both OCV and discharge-charge cycle tests.

(31) A solution that contained 3.76 M NaOH, 0.24 M NaAlO.sub.2 and 0.05 M NaSnO.sub.2 was used as the negative electrolyte. A solution that contained 0.1 M Co(ClO.sub.4).sub.2 and 2 M HClO.sub.4 was used as the positive electrolyte, which was prepared by dissolving CoCO.sub.3 into perchloric acid. A 4 M NaClO.sub.4 solution was used as the middle supporting electrolyte. A small piece of Al strip (ESPI Metals, 2 cm by 3 cm, 5N grade) and a small piece of graphite felt (SGL Group, 2 cm by 3 cm, Sigracell GFA5 EA type) were used as the negative electrode and the positive electrode, respectively. The cell was first charged at 50 mA (or 8.3 mA/cm.sup.2 of current density) for 2.5 hours and the OCV was tested for 20 min. The discharge-charge cycle is then carried out for 20 min by setting current constant at 5 mA (or 0.83 mA/cm.sup.2 of current density).

Example 2

(32) A zinc-cerium DMTE-RFB (Zn/Ce-DMTE-RFB), configured as [(Zn/Zn(OH).sub.4.sup.2)//(Ce.sup.4+/Ce.sup.3+)] was constructed. A three-compartment cell made up of three acrylic flow channels was designed and used as follows. Three 5 cm by 6 cm rectangular channels were put in series with membranes in between. The three channels, based on half-reaction inside, were assigned as negative, middle and positive compartments. One piece of Nafion 1135 membrane (DuPont, 87.5 m thickness) and one piece of Fumasep FAA membrane (FuMa-Tech, 70 m thickness) were used as the CEM and AEM, respectively. The CEM is put between the negative compartment and the middle compartment while the AEM is put between the middle compartment and the positive compartment, along with silicone gasket to seal the conjunction part. The positive electrode and negative electrode are each put next to its corresponding compartment, respectively. Two clamps were used to compress the three channels and electrodes tightly to avoid electrolyte leakage. Electrolytes are stored outside the channel in three tanks and delivered by peristaltic pump (Masterflex L/S 100 RPM). The working flow battery set-up is a potentiostat/galvanostat (Solartron 1287A) and was used in both OCV and discharge-charge cycle tests.

(33) The negative electrolyte contained 3 M NaOH and 0.5 M Na.sub.2[Zn(OH.sub.4)]. A solution that contained 0.5 M Ce(ClO.sub.4).sub.3, 2 M HClO.sub.4 was used as the positive electrolyte, which was prepared by dissolving Ce.sub.2(CO.sub.3).sub.3 into perchloric acid. The middle electrolyte used was 4 M NaClO.sub.4 solution. The volume for each electrolyte used in test is 30 ml. A rectangular copper plate (ESPI Metals, 5 cm by 6 cm, 3N grade) was used as negative current collector. Before the experiment, the copper was rinsed with acetone and deposited with a layer of cadmium according to the method in reference. Graphite based bipolar plate (SGL group, 5 cm by 6 cm, Sigracet TF6 type) was used as positive current collector. Graphite felt (SGL Group, 3 cm by 4 cm, Sigracell GFA5 EA type) was used as positive electrode and compressed by plastic frame to contact bipolar plate. The cut-off voltage for charge and discharge are 3.24 and 1.8 respectively. The discharge-charge cycle was carried out at constant current density at 60 mA (or 5 mA/cm.sup.2 of current density) with flow rate for all three electrolytes at 20 ml/min.

(34) The DMTE-RFB systems described above may have other configurations besides those of acid/neutral/base configurations and are not limited thereto. Hydroxide ions play the role of ligand to tune the electrode potential of, for example, a Zn(II)/Zn or Al(III)/Al system, more negative. Various ligands and central ions may be used for a similar role and may be combined and implemented in a DMTE-RFB system. Table 2 lists some possible candidates and combinations of negative, middle and positive electrolyte.

(35) TABLE-US-00002 TABLE 2 Negative, middle and positive electrolyte choices (electrode potential from reference) Negative electrolyte Middle electrolyte Positive electrolyte [Co(CN).sub.6].sup.3/[Co(CN).sub.6].sup.4 (.sub.N = 0.83 V) Na.sub.2SO.sub.4 Co.sup.3+/Co.sup.2+ (.sub.P = 1.95 V) [Co(EDTA)].sup./[Co(EDTA)].sup.2 (.sub.N = 0.37 V) NaClO.sub.4 Ce.sup.4+/Ce.sup.3+ (.sub.P = 1.74 V) [Fe(CN).sub.6].sup.3/[Fe(CN).sub.6].sup.4 (.sub.N = 0.36 V) VO.sub.2.sup.+/VO.sup.2+ (.sub.P = 1.00 V) [Cr(CN).sub.6].sup.3/[Cr(CN).sub.6].sup.4 (.sub.N = 1.28 V) Fe.sup.3+/Fe.sup.2+ (.sub.P = 0.77 V) Zn(OH).sub.4.sup.2/Zn (.sub.N = 1.22 V) Al(OH).sub.4.sup./Al (.sub.N = 2.34 V)

(36) While this invention has been described in terms of several preferred embodiments, there are alterations, permutations, and substitute equivalents, which fall within the scope of this invention. It should also be noted that there are many alternative ways of implementing the methods and apparatuses of the present invention. It is therefore intended that the following appended claims be interpreted as including all such alterations, permutations, and substitute equivalents as fall within the true spirit and scope of the present invention.