Redox-active compounds and uses thereof

Abstract

To provide a ferrocene-based redox-active compound to be a constituent of an electrolyte composition for use in an electrodialytic battery cell.SOLUTION: For example, tetrakis((3-trimethylammonio)propyl)ferrocene tetrachloride (TTMAP-Fc) synthesized by the following scheme II is illustrated.SELECTED DRAWING: Figure 1A

Description

[Technical Field]

[0001] FIELD OF THE DISCLOSURE The present disclosure relates generally to redox-active compounds, compositions containing same, and methods of using the compounds. [Background technology]

[0002] The deployment of grid-scale electrical energy storage will enable deep penetration of energy generation from intermittently available renewables. While today's batteries offer load-shifting capabilities, high prices still inhibit widespread integration of storage, potentially slowing the adoption of renewables. At the same time, growing water scarcity is forcing the installation of energy-intensive desalination technologies to meet growing water demands. For example, as a result of climate change and the rapid pace of global population growth, the pressure on freshwater supplies continues to grow. In communities located in areas without easy access to freshwater, such as the Persian Gulf and other desert regions, freshwater is produced through the desalination of seawater. This process is highly energy-intensive, whether driven hydraulically (e.g., via reverse osmosis [RO]), thermally (e.g., via flash distillation), or electrochemically (e.g., through electrodialysis). Elsewhere, all of these methods are routinely used to treat contaminated wastewater from industrial activities.

[0003] Furthermore, the price of electricity generation from renewable sources has fallen rapidly, driven primarily by technological improvements in solar and wind power. For example, in October 2017, Saudi Arabia received a tender to provide electricity from solar power at a price of $17.90 / MWh for a 300 MW plant. The ready availability of cheap electrons presents an opportunity for electrochemical methods of water desalination (or treatment) to play a larger role in meeting the growing demand for water. Summary of the Invention

[0004] The present disclosure describes redox-active compounds, compositions containing redox-active compounds, and methods of using redox-active compounds.

[0005] In one aspect, the disclosure describes a redox-active compound according to Formula I. The redox-active compound has a ferrocene core and two cyclopentadienyl ligands associated with the ferrocene core. One or more -XY substituents are covalently bonded to each cyclopentadienyl ligand. Each X is independently a C1 to C6 alkyl group. 12 Each Y is a solubilizing group independently selected from oligo(ethylene glycol), hydroxyl, trialkylammonio, alkylimidazolio, sulfonate, sulfate, carboxyl, phosphate, phosphonate, ammonium, or a nitrogen-containing heterocycle. The total number of cyclopentadienyl ligand substitutions per ferrocene core is 3 or greater.

[0006] According to one embodiment, the solubilizing group is a trialkylammonio or alkylimidazolio, and the alkyl substituents on the solubilizing group are C1 to C12 in length.

[0007] According to one embodiment, the solubilizing group is a trialkylammonio, and at least one of the alkyl groups is methyl.

[0008] According to one embodiment, the total number of substituents attached to the cyclopentadienyl ligands per ferrocene core is 3, 4, 5, 6, 7, 8, 9, or 10.

[0009] In another aspect, the present disclosure describes an electrolyte composition comprising a redox-active compound of Formula I and a salt.

[0010] According to one embodiment, the redox-active compound is present in the electrolyte composition at a concentration of 2% by weight or more, 3% by weight or more, 4% by weight or more, or 5% by weight or more, or up to 25% by weight.

[0011] According to one embodiment, the salt of the electrolyte composition is LiCl, LiBr, CaCl2, MgCl2, or a combination thereof.

[0012] According to one embodiment, the salt of the electrolyte composition is present in a concentration of 1% by weight or more, 5% by weight or more, or 10% by weight or more, or up to 60% by weight.

[0013] According to one embodiment, the electrolyte composition has a pH of 3 to 11 or 5 or higher.

[0014] In another aspect, the present disclosure describes a system. The system includes a device having a first reservoir having a first input and a first output, a second reservoir having a second output and a second input, a first electrolyte chamber having a first electrode and a first solution, and a second electrolyte chamber having a second electrode and a second solution. The first reservoir and the second reservoir are in electrolytic communication through a first type of membrane disposed between the first reservoir and the second reservoir. The first reservoir and the first chamber are in electrolytic communication through a second type of membrane disposed between the first chamber and the first reservoir. The second reservoir and the second chamber are in electrolytic communication through a second type of membrane disposed between the second reservoir and the second chamber. One or both of the first solution and the second solution contains a redox-active compound according to Formula I.

[0015] According to one embodiment, the solubilizing group of the redox-active compound in the system is a trialkylammonio or alkylimidazolio, and the alkyl substituents on the solubilizing group are C1 to C12 in length.

[0016] According to one embodiment, the total number of substituents attached to the cyclopentadienyl ligands is 3, 4, 5, 6, 7, 8, 9, or 10 per ferrocene core. [Brief explanation of the drawings]

[0017] The following discussion will refer to the following figures, wherein the same reference numbers may be used to identify similar / identical components in multiple figures. The drawings are not necessarily to scale. [Figure 1A] FIG. 1 is a schematic diagram of an electrodialysis system having a positively charged redox shuttle, according to certain embodiments. [Figure 1B] FIG. 1 is a schematic diagram of an electrodialysis system having a negatively charged redox shuttle, according to certain embodiments. [Figure 2A] FIG. 1 is a schematic diagram of a charging cycle of an electrodialysis battery having a pair of positively charged reactants, according to certain embodiments. [Figure 2B] FIG. 2B is a schematic diagram of a discharge cycle of the electrodialysis battery of FIG. 2A, according to certain embodiments. [Figure 3] 1 is a proposed reaction scheme for the decomposition of ferrocene. [Figure 4] Possible substitution patterns for polysubstituted cyclopentadienyl ligands. [Figure 5] A representative experimental setup for testing the stability of redox-active compounds.

[0018] All scientific and technical terms have meanings commonly used in the art unless otherwise specified. The definitions provided are to facilitate understanding of certain frequently used terms and are not meant to limit the scope of the present disclosure.

[0019] The singular forms "a," "an," and "the" include embodiments with plural referents unless the content clearly dictates otherwise.

[0020] The term "or" is generally used in its inclusive sense unless the context clearly dictates otherwise. The term "and / or" refers to one or all of the listed elements or a combination of any two or more of the listed elements.

[0021] As used herein, the words "have," "having," "include," "including," "comprise," "comprising," and the like are intended to be interpreted broadly and without limitation, generally meaning "including but not limited to." It will be understood that "consisting essentially of," "consisting of," and the like are encompassed by "comprising," and the like.

[0022] The phrase "consisting essentially of," when referring to a composition, device, system, method, etc., means that the components of the composition, device, system, method, etc. are limited to the recited components and any other components that do not materially affect the basic and novel property(ies) of the composition, device, system, method, etc.

[0023] The terms "preferred" and "preferably" refer to embodiments that may offer certain benefits, under particular circumstances. However, other embodiments may also be preferred, under the same or other circumstances. Furthermore, the recitation of one or more preferred embodiments does not imply that other embodiments are not useful and is not intended to exclude other embodiments from the scope of the present disclosure, including the claims.

[0024] The recitation of numerical ranges by endpoints includes all numbers subsumed within that range (e.g., 1 to 5 includes 1, 1.5, 2, 2.75, 3, 3.80, 4, 5, etc.; or 10 or less includes 10, 9.4, 7.6, 5, 4.3, 2.9, 1.62, 0.3, etc.). When a range of values ​​is "up to" a particular value, that value is included within the range. All upper and lower limits may be combined in any combination to form, for example, a range for a particular component or property.

[0025] The use of "first," "second," etc. in the above description and in the following claims is not intended to necessarily indicate that a recited number of steps are present. For example, a "second" step is merely intended to distinguish it from another step (such as a "first" step). The use of "first," "second," etc. in the above description and in the following claims is not intended to necessarily indicate an earlier time than another.

[0026] When a group occurs more than once in the formulae described herein, each group is "independently" selected, whether or not specifically stated. For example, one or more R 1 When a group is present in the formula, each R 1 The groups are independently selected, and the subgroups within these groups are also independently selected.

[0027] The term "alkyl" refers to a monovalent group that is a radical of an alkane and includes straight-chain alkyl groups, branched-chain alkyl groups, cyclic alkyl groups, and bicyclic alkyl groups, and combinations thereof, including both unsubstituted and substituted alkyl groups. The term "n-alkyl" (e.g., n-propyl) refers to a monovalent group that is a radical of an alkane in which the chain is straight.

[0028] The term "heterocyclic group" means a cyclic aliphatic group having at least one C atom replaced with a heteroatom such as N, O or S.

[0029] The term "carbonyl" means a divalent group of formula -(CO)- where a carbon atom is attached to an oxygen with a double bond.

[0030] The term "quaternary nitrogen" refers to a tetravalent nitrogen atom that is bonded to four groups and has a positive charge. The positively charged quaternary nitrogen group is associated with a counterion (anion).

[0031] A "phosphonate" group has the formula -PO3 2- or PO3H - The charge of the anion is balanced by the charge of the cation.

[0032] A "phosphate" group has the formula -OPOH -1 or -OPO3 2- The charge of the anion is balanced by the charge of the cation.

[0033] "Oligo(ethylene glycol)", also known as polyethylene glycol, has the formula -[OCH2CH2] n - group, where n is the number of repeat units.

[0034] "Hydroxyl" refers to a group of the formula -OH. At certain pH values, the hydroxyl group may be deprotonated and bears a negative charge.

[0035] "Trialkylammonio" means a group of the formula -NR1R2R3. Unless otherwise specified, each R group independently ranges in length from C1 to C 12 The alkyl may be selected from the following:

[0036] "Alkyl imidazoline" means the following formula:

[0037] [ka] In the formula, R has a length of C1 to C unless otherwise specified. 12 is an alkyl group.

[0038] A "sulfonate" group is an anion of formula -R-SO3-, where the charge of the anion can be balanced by a cation.

[0039] A "sulfate" group has the formula -R-OSO3 - and the charge of the anion may be balanced by the cation.

[0040] A "carboxyl" group is of the formula --COOH, and at certain pH values, the carboxyl may be deprotonated to have a formal negative charge that can be balanced with a cation.

[0041] An "ammonium" group, unless otherwise specified, is a cation of the formula NR1R2R3R4, each independently being H or C1 to C6 in length. 12 is an alkyl of.

[0042] Ferrocene refers to (η-Cp)Fe, which can be represented by the formula:

[0043] [ka] where two cyclopentadienyl ligands are associated with the iron atom. The iron atom is in the +2 oxidation state. The overall formal charge of the molecule is neutral. Ferrocenium has a (η 5 -Cp)2Fe + Ferrocenium is the cationic form of ferrocene, as described in. The iron atom in ferrocenium is in the +3 oxidation state.

[0044] The term "redox active compound" means a compound that can be adjusted to exist in different oxidation states. Redox active compounds can also be referred to as redox shuttles.

[0045] The term "electrolytic communication" refers to the ability of an ion or ions to pass between two or more chambers or reservoirs, for example, through a membrane to affect a chemical or electrochemical reaction. DETAILED DESCRIPTION OF THE INVENTION

[0046] The present disclosure generally relates to redox-active compounds. Redox-active compounds may be used in electrochemical desalination systems and, if desired, corresponding simultaneous energy storage. Research efforts in grid storage benefit from a single approach that minimizes cost per kWh, with the U.S. Department of Energy's cost target of $125 / kWh by 2022. An alternative strategy is to increase the revenue associated with each kWh of storage, an alternative unavailable to conventional energy storage technologies. However, this alternative is possible because the electrochemical batteries described in this disclosure produce a valuable secondary product, desalinated water, during charging and discharging. Because aqueous flow batteries share many capital requirements (e.g., pumps, membranes, piping) with electrochemical desalination technologies, a system combining the two can lead to significant capital cost savings compared to two separate systems. The revenue from the desalinated water compensates for the additional capital costs and leverages process intensification to break down below the $ / kWh barrier encountered with conventional energy storage technologies. Electrochemical approaches to desalination have the potential to easily scale modulus generation and lamp production while maintaining high energy efficiency and the ability to handle high salinity feeds.

[0047] The current state-of-the-art in electrochemical water desalination is electrodialysis; however, it consumes relatively more energy for salt removal (e.g., approximately 0.26–0.30 kWh / kg NaCl) than other desalination techniques such as reverse osmosis (e.g., 0.06–0.08 kWh / kg NaCl), but less energy than thermal techniques such as vapor compression (e.g., 0.6–1.0 kWh / kg NaCl). Capacitive deionization uses electrical energy, but is also energy-intensive at approximately 0.22 kWh / kg NaCl, and is best suited to removing trace amounts of dissolved salts from water because the electrodes are solid by definition. Unlike reverse osmosis, electrodialysis is a technology that can be used to treat brines at any salinity, but has seen limited use due to its high specific energy consumption for salt removal.

[0048] Energy consumption in electrodialysis is proportional to the applied voltage, and reducing (or minimizing) the voltage applied to the cell reduces the specific energy consumption of the electrodialysis stack. In conventional electrodialysis, ions are driven from or into seawater by faradaic reactions at the anode and cathode. In most cases, the faradaic reaction is simply water splitting, where water is oxidized to oxygen at the anode and reduced to hydrogen at the cathode. This creates a charge imbalance at the electrodes that is balanced by the movement of ions through strategically placed ion-selective membranes. However, water splitting requires energy, and therefore carries an energy penalty. This problem is exacerbated by the fact that significant overpotentials are associated with both the oxidation and reduction of water. Furthermore, oxygen and chlorine gases generated at the anode are highly toxic, necessitating the use of platinum / iridium-plated electrodes.

[0049] In California alone, if solar replaces fossil fuel sources of electricity generation in the state (0.40 quads, 41% of all sources) and pairs with desalination batteries for storage as described herein, an additional 0.63 quads of solar power could be brought online based on a cost-optimized round-trip energy efficiency of 64%. At the same time, the desalination batteries would provide water resources equivalent to 30% of the state's consumption.

[0050] The redox-active compounds of the present disclosure may be used in electrochemical cells designed to perform electrodialysis in an energy-efficient manner by circulating the redox-active compounds from the anode to the cathode, dissolving them in water, and then back again. The redox-active compounds have rapid kinetics for reduction or oxidation, significantly reducing the high operating voltages required for conventional electrodialysis, where water splitting drives salt transport across membranes such as ion-selective membranes. Lowering the operating voltage reduces specific energy consumption, since specific energy consumption is proportional to the operating voltage. The system may also be reversibly adapted for combined electrical energy storage by pairing two different redox-active compounds at the anode and cathode. For example, two different redox-active compounds described in the present disclosure may be paired where each redox-active compound is separately located at the anode or cathode. Furthermore, the redox-active compounds described in the present disclosure may be paired with redox-active compounds already available in the art, where the redox-active compounds are separately located at the anode and cathode.

[0051] In certain embodiments of energy-efficient, lower potential electrodialysis systems, redox-active compounds (also referred to as redox carriers or redox shuttles) dissolved in water are reduced at the cathode, then shuttled to the anode, where they are reoxidized, and subsequently redelivered to the cathode to complete the cycle.

[0052] Energy efficiency of an electrodialysis system is achieved through the selection of redox-active compounds. Effective redox-active compounds possess as many of the following properties as possible. For example, the redox-active compound should be chemically stable in the oxidized and reduced forms, remain highly water-soluble in the oxidized and reduced forms, and not be oxygen-sensitive in the oxidized and reduced forms. The redox-active compound should not be proton-binding, have rapid redox kinetics, and be chemically compatible with any components present in the water being treated. The redox-active compound should be stable and functional at the pH of the water being treated. Furthermore, the redox-active compound should have low permeability through ion-selective membranes and be non-toxic.

[0053] The redox-active compounds of the present disclosure should be chemically stable in the oxidized and reduced forms, remain highly water-soluble in the oxidized and reduced forms, and not be oxygen-sensitive in the oxidized and reduced forms. The redox-active compounds of the present disclosure are not proton-bound, have rapid redox kinetics, and are chemically compatible with any components present in the water being treated. The redox-active compounds of the present disclosure are stable and functional at the pH of the water being treated. The redox-active compounds of the present disclosure have low permeability through ion-selective membranes and are non-toxic.

[0054] The most common redox-active compound reported to date is iron(II) (Fe 2+ ) / iron(III)(Fe 3+ ), ferrocyanide / ferricyanide (Fe(CN)) bond, and ethylenediaminetetraacetoferrate(2-) (Fe(II)-EDTA) / ethylenediaminetetraacetoferrate(1-) (Fe(III)-EDTA). All three allow for reverse electrodialysis, i.e., the generation of electricity from the salinity gradient instead of the consumption of electricity to generate it. Unfortunately, Fe 3+The ion is soluble only at low pH and forms insoluble oxides or hydroxides at neutral pH, Fe(CN) forms highly insoluble Prussian blue species compounds in contact with many transition metals (especially iron), and Fe-EDTA complexes exhibit limited electrochemical stability.

[0055] Systems with Fe(CN) as the redox-active compound have been successfully demonstrated in laboratories using only NaCl as simulated seawater, but the ubiquitous presence of calcium and iron in seawater (typically at levels of 400 parts per million (ppm) and 1-3 parts per billion (ppb)) rapidly causes membrane fouling when these ions cross into the reservoir. They form insoluble precipitates of potassium, calcium, and iron when in contact with Fe(CN). Furthermore, even the neutral to slightly acidic conditions generated at the anode cause the release of highly toxic hydrogen cyanide.

[0056] However, ferrocene derivatives offer an alternative to already available iron redox-active compounds. Ferrocene has an iron core with two cyclopentadienyl ligands. The cyclopentadienyl ligands may be substituted with various chemical moieties. Each cyclopentadienyl substituent in the redox-active compound generally has a linker and a solubilizing group. The linker connects the solubilizing group to the cyclopentadienyl ligand. The linker may be an alkyl group.

[0057] An example of a redox-active ferrocene derivative useful as a redox-active compound is bis(trimethylammoniopropyl)ferrocene (BTMAP-Fc), of the following structure:

[0058] [ka]

[0059] BTMAP-Fc has all of the attributes listed above and may be a suitable redox-active compound. Properties of BTMAP-Fc compared to various iron-containing redox-active conjugates in Table 1 below.

[0060] [Table 1] † On platinum ‡ pH 14

[0061] In particular, each of the redox bonds other than BTMAP-Fc has at least one property that is inconsistent with the properties of the effective redox-active compounds for the cells described above. For example, Fe 2+ / Fe 3+ is insoluble at pH 7, Fe(EDTA) has low chemical and electrochemical stability, and Fe(CN) forms insoluble precipitates with other ions and releases toxic HCN at pH 7 or below. In principle, any water-soluble redox-active compound can be used in the cell embodiments of Figures 1A-1B, not just those listed in Table 1. Other redox couplings may be preferred, depending, for example, on the desired pH of the desalted and salted streams. An example of such an energy-efficient, redox-assisted electrodialysis system is further described below.

[0062] Ferrocene derivatives having an increased number of charged bulky substituents may have improved chemical stability relative to ferrocene derivatives having a single substituent on each cyclopentadienyl ring. Ferrocene derivatives having an increased number of charged bulky substituents may have improved electrochemical stability relative to ferrocene derivatives having a single substituent on each cyclopentadienyl ring. Ferrocene derivatives having an increased number of charged bulky substituents may have improved water solubility relative to ferrocene derivatives having a single substituent on each cyclopentadienyl ring. Ferrocene derivatives having an increased number of charged bulky substituents may have improved water solubility over a wider range of pH levels. Ferrocene derivatives having an increased number of charged bulky substituents may have improved stability over a wider range of pH levels.

[0063] Quaternary nitrogen linkers dramatically improve the oxygen stability of ferrocene, especially in its oxidized (ferrocenium) state. The mechanism of decomposition in unsubstituted ferrocene is attributed to a bimolecular, second-order reaction between two ferrocenium ions (Figure 3, reproduced from Singh et al., "A Kinetic Study of Ferrocenium Cation Decomposition Utilizing an Integrated Electrochemical Methodology Composed of Cyclic Voltammetry and Amperometry," 139 Analyst 5747 (2014)). Increasing the number of substituents on the cyclopentadienyl ligand on ferrocene is also associated with improved stability. Furthermore, charged quaternary nitrogen alkyl substituents on the ferrocene center increase the steric and Coulombic repulsions between different ferrocenium ions, thereby significantly improving stability.

[0064] The redox-active compounds of the present disclosure include a ferrocene core and two cyclopentadienyl ligands. A total of three or more independently selected -XY substituents are covalently bonded to the cyclopentadienyl ligands. Each X is independently selected from a straight-chain alkyl or a branched-chain alkyl. Each X is covalently bonded to one of the cyclopentadienyl ligands and one Y group. Each Y is a solubilizing group independently selected from oligo(ethylene glycol), hydroxyl, trialkylammonio, alkylimidazolio, sulfonate, sulfate, carboxyl, phosphate, phosphonate, ammonium, or a nitrogen-containing heterocycle.

[0065] The disclosed redox-active compounds may also be represented by general formula I:

[0066] [ka]

[0067] In Formula I, a first cyclopentadienyl ligand and a second cyclopentadienyl ligand are associated with the iron core. The first cyclopentadienyl ligand may have a total of n1 independently selected -XY substituents. The second cyclopentadienyl ligand has a total of n2 independently selected -XY substituents.

[0068] The compounds of general formula I may have a neutral charge (ferrocene) or may be a ferrocenium cation. The iron core is Fe +2 or Fe +3 may have an oxidation state of

[0069] The first cyclopentadienyl ligand and the second cyclopentadienyl ligand are interchangeable. The first cyclopentadienyl ligand and the second cyclopentadienyl ligand are distinguished herein to enhance understanding of the number of -X,Y substituents on each cyclopentadienyl ligand (n1 and n2). For example, the first cyclopentadienyl ligand has n1 -X,Y substituents, and the second cyclopentadienyl ligand has n2 -X,Y substituents.

[0070] The total number of -XY cyclopentadienyl substituents on the first and second cyclopentadienyl ligands n1+n2 is 3 or greater.

[0071] In some embodiments, the number of -X,Y substituents on the first cyclopentadienyl ligand, n1, and the number of -X,Y substituents on the second cyclopentadienyl ligand, n2, are the same. In some embodiments, n1 is 2 and n2 is 2. In some embodiments, n1 is 3 and n2 is 3. In some embodiments, n1 is 4 and n2 is 4. In some embodiments, n1 is 5 and n2 is 5.

[0072] In some embodiments, the number n1 of -X,Y substituents on the first cyclopentadienyl ligand and the number n2 of -X,Y substituents on the second cyclopentadienyl ligand are different. In some embodiments, n1 is 1 and n2 is 2. In some embodiments, n1 is 1 and n2 is 3. In some embodiments, n1 is 1 and n2 is 4. In some embodiments, n1 is 1 and n2 is 5. In some embodiments, n1 is 2 and n2 is 3. In some embodiments, n1 is 2 and n2 is 4. In some embodiments, n1 is 2 and n2 is 5. In some embodiments, n1 is 3 and n2 is 4. In some embodiments, n1 is 3 and n2 is 5. In some embodiments, n1 is 4 and n2 is 5.

[0073] Any substitution pattern of -XY substituents (n1 or n2 is 1 or greater) around the polysubstituted cyclopentadienyl ring is permitted, as shown in Figure 4. In some embodiments where n1 is 2, the -XY substitution pattern around the cyclopentadienyl ring can be 1, 2, or 1, 3. In some embodiments where n1 is 3, the -XY substitution pattern around the cyclopentadienyl ring can be 1, 2, 3, or 1, 2, 4. In some embodiments where n1 is 4, the -XY substitution pattern around the cyclopentadienyl ring can be 1, 2, 3, 4. In some embodiments, the substitution patterns of the -XY substituents n1 and n2 are the same. In some embodiments, the substitution patterns of the -XY substituents n1 and n2 are different.

[0074] In the cyclopentadienyl substituent -XY, X is generally a linker such as an alkyl group and Y is a solubilizing group.

[0075] In Formula I, each X is independently a branched or straight chain alkyl.

[0076] In some embodiments, each X is independently C1 to C 12In some embodiments, each X is independently a linear alkyl of C1-C3, C1-C6, or C1-C9 length. In some embodiments, each X is independently a linear alkyl of C2-C6 length. In some embodiments, each X is independently a linear alkyl selected from n-methyl, n-ethyl, n-propyl, n-butyl, n-pentyl, n-hexyl, n-heptyl, n-octyl, n-nonyl, n-decyl, n-undecyl, or n-dodecyl. In some embodiments, X is n-propyl.

[0077] In Formula I, each X may independently be a branched alkyl. A branched alkyl has a central alkyl chain and one or more branches covalently attached to the central alkyl chain. In some embodiments, the central alkyl chain is independently C1 to C6. 12 In some embodiments, the central alkyl chain of the branched alkyl is C1-C3, C1-C6, or C1-C9 in length. In some embodiments, the central alkyl chain of the branched alkyl is C2-C6 in length. In some embodiments, the central alkyl chain of the branched alkyl is propyl. In some embodiments, each X can independently be a single-branched central alkyl chain. Examples of single-branched alkyls include, but are not limited to, (1-methyl)ethyl, (3-ethyl)propyl, (3-butyl)pentyl, (4-pentyl)hexyl, (3-hexyl)heptyl, (5-heptyl)octyl, (3-octyl)nonyl, (6-nonyl)decyl, (2-decyl)undecyl, and (5-undecyl)dodecyl. In some embodiments, the central alkyl chain is such that the total number of possible branches is C n-1where n is the total number of carbons in the central alkyl chain. For example, if n is 5, then the total number of possible branches is 4. Examples of multiply branched alkyls include, but are not limited to, (1,2-dimethyl)ethyl, (1,2,3,4-tetramethyl)octyl, and (3,6,9-triethyl)dodecyl. In some embodiments, the alkyl branches of a multiply branched alkyl central chain may be the same length. In some embodiments, the alkyl branches of a multiply branched alkyl central chain are of different lengths. Examples of multiply branched alkyl central chains in which the branched alkyls are of different lengths include, but are not limited to, (2-methyl-4-butyl)octyl, (1-methyl-5-hexyl)dodecyl, and (2-methyl-butyl)hexyl.

[0078] In Formula I, each Y is independently a solubilizing group. In some embodiments, each Y is a solubilizing group independently selected from trialkylammonio, dialkylimidazolio, ammonium, oligo(ethylene glycol), hydroxyl, phosphonate, carboxyl, nitrogen-containing heterocycle, sulfonate, sulfate, or phosphate.

[0079] In some embodiments, Y is a trialkylammonio or dialkylimidazolio. The trialkylammonio or dialkylimidazolio is C1 to C6 in length. 12 Examples of trialkylammonio alkyl groups or dialkylimidazolio alkyl groups include n-methyl, n-ethyl, n-propyl, n-butyl, n-pentyl, n-hexyl, n-heptyl, n-octyl, n-nonyl, n-decyl, n-undecyl, and n-dodecyl. Each alkyl group in the trialkylammonio or dialkylimidazolio can be the same or different. In some embodiments, two of the alkyl groups in the trialkylammonio are the same. In some embodiments, one of the alkyl groups in the trialkylammonio is methyl, and the other two alkyl groups are independently C1 to C6 in length. 12In some embodiments, two of the alkyl groups of the trialkylammonio are methyl and the third alkyl group is C1 to C 12 In some embodiments, three of the alkyl groups of the trialkylammonio are methyl.

[0080] In some embodiments, Y is a nitrogen-containing heterocycle. The nitrogen-containing heterocycle may be fully unsaturated, partially unsaturated, or fully saturated. The nitrogen-containing heterocycle may have a formal charge. Examples of nitrogen-containing heterocycles with a formal charge include, but are not limited to, aziridinium, azetidinium, pyrrolidinium, piperidinium, azepanium, pyrrolium, and pyridinium. In some embodiments, the nitrogen-containing heterocycle may contain two or more nitrogen atoms. Examples include, but are not limited to, imidazolium, pyrimidinium, pyrazinium, tetrazolium, and triazolium. In some embodiments, the nitrogen-containing unsaturated heterocycle or nitrogen-containing saturated heterocycle may contain other heteroatoms such as O and S. Examples include, but are not limited to, thiazolium, isoxazolium, and oxazolium. In some embodiments, the nitrogen-containing heterocycle may be a fused ring system. Examples include, but are not limited to, indolium, quinolinium, and soquinolinium.

[0081] In some embodiments, Y is an oligo(ethylene glycol) (OEG). The oligo(ethylene glycol) may have 1 to 20 OEG repeat units. In some embodiments, the oligo(ethylene glycol) may have 1 to 5 OEG repeat units. In some embodiments, the oligo(ethylene glycol) may have 1 to 10 OEG repeat units. In some embodiments, the oligo(ethylene glycol) may have 1 to 15 OEG repeat units. In some embodiments, the oligo(ethylene glycol) may have 5 to 10 OEG repeat units. In some embodiments, the oligo(ethylene glycol) may have 10 to 15 OEG repeat units. In some embodiments, the oligo(ethylene glycol) may have 15 to 20 OEG repeat units.

[0082] The redox-active compounds of Formula I may include a counterion or counterions. Examples of counterions include Cl. - , Br - , I - , Li + , Ca + , Mg 2+ , BF4 - , and Na + These include, but are not limited to:

[0083] According to one embodiment, the redox-active compound of Formula I may be used in an electrodialysis system such as that depicted in FIGS. 1A-1B. According to such an embodiment, an electrochemical cell 100 is shown that provides energy-efficient electrodialysis. The cell 100 has four chambers 102, 104, 106, 108 in series. Each chamber is separated from its neighbor by a suitable membrane 110, 112, 114 (FIG. 1A) or 116, 118, 120 (FIG. 1B). The two central chambers 104, 106 contain a salted stream 130 and a desalted stream 132, and the two outer chambers 102, 108 contain a cathode and an anode (FIG. 1A) or an anode and a cathode (FIG. 1B), respectively. The chambers 102 and 108 may contain electrolytes. For example, the chamber containing the cathode may also contain a catholyte, and the chamber containing the anode may also contain an anolyte. The redox-active compound of Formula I, as well as other already available redox-active compounds, may be part of the anolyte or catholyte composition. For example, the redox-active compound of Formula I may be in chamber 102 as part of the catholyte composition or in chamber 108 as part of the anolyte composition. In some embodiments, the redox-active compound of Formula I is only part of the anolyte composition or the catholyte composition. In some embodiments, the redox-active compound of Formula I is in both the anolyte composition and the catholyte composition. The membrane may be an ion-selective membrane, such as a cation-exchange membrane or an anion-exchange membrane, depending on the cell design. If the redox-active compound has a sufficiently high molecular weight (e.g., is dendrimeric or polymeric in nature), the membrane may be a microporous membrane. The membrane may also incorporate several ion-selective elements and several microporous elements within the same membrane. In certain embodiments, the membrane may also be a composite membrane.

[0084] For example, Figure 1A shows a redox-assisted electrodialysis system having a positively charged redox-active compound 101. Migration of the redox-active compound from the anode 108 to the cathode 102 is indicated by arrow 128, and migration from the cathode 102 to the anode 108 is indicated by arrow 126. The cathode chamber 102 and the salting chamber 104 are separated by an anion exchange membrane 110, and the anode chamber 108 and the desalting chamber 106 are also separated by an anion exchange membrane 114. However, the membranes 110 and 114 do not necessarily have to comprise the same material or be of similar dimensions. The salting chamber 104 is also separated from the desalting chamber 106 by a cation exchange membrane 112. As can be seen, chloride and sodium ions cross membranes 110 and 112 into salified stream 130 in chamber 104 but cross membranes 114 and 112 to exit desalted stream 132 in chamber 106 .

[0085] In various embodiments, any number of pairs of alternating salting and desalting chambers may be used, however, water splitting may begin to occur in many chambers when the applied voltage exceeds 1.26V.

[0086] According to one embodiment, the redox-active compound of Formula I can be used in an energy storage device. For example, the four-chamber cell design described above in connection with FIGS. 1A-1B may also be adapted for use as an energy storage device (i.e., an electrodialysis battery). Performance is improved over conventional electrodialysis when the cell uses a small number of saltation / demineralization chamber pairs (e.g., one pair). For energy storage applications where the preferred operating current density is also lower than that for electrodialysis, a smaller number of chamber pairs is also advantageous. The above-described cell may be adapted for energy storage by using two different redox-active reactants as separate anolytes and catholytes instead of shuttling the same compound between the anode and cathode. In some embodiments, two different redox-active compounds of Formula I may be used. In some embodiments, a redox-active compound of Formula I and an already available redox-active compound may be used. Unlike three-chamber cell designs (e.g., U.S. Pat. Nos. 9,340,436, 9,670,077, and 9,673,472, each of which is incorporated herein by reference), the four-chamber design is capable of continuously producing demineralized water all the time during operation, instead of only half the time. Also, when crossover occurs, the four-chamber design does not suffer from precipitation of insoluble solids.

[0087] An embodiment of an electrodialysis battery with four chambers is shown in Figures 2A-2B. A desalination battery is a flow battery with multiple chambers. The reduction of the anolyte and the oxidation of the catholyte during the charging half-cycle transfers NaCl through suitable ion-selective membranes into or out of the intervening chambers holding seawater. + and Cl - The reverse process occurs during the discharge half-cycle. At every point during the cycle, one of the water chambers experiences a net salt inflow, while the other sees a net outflow. The energy required to affect desalination is simply the difference between the energy input during charging and the energy recovered during discharge.

[0088] 2A shows the charge cycle of a four-chamber battery using a positively charged pair of redox-active compounds, one of which is of Formula I and the other is zinc. The battery includes four chambers 202, 204, 206, and 208 and three membranes 210, 212, and 214. During the charge cycle, salt stream 230 is located in chamber 204 between anolyte chamber 202 and chamber 206, which contains desalted stream 232. Catholyte chamber 208, which contains the redox-active compound of Formula I, is separated from chamber 206 and desalted stream 232 by anion exchange membrane 214, while anolyte chamber 202 is separated from chamber 204 and salt stream 230 by another anion exchange membrane 210. As noted above, membranes 210 and 214 do not necessarily have to comprise the same material or be similarly sized. Salting chamber 204 is also separated from desalting chamber 206 by cation exchange membrane 212. During a charging cycle, chloride and sodium ions enter chamber 204 across membranes 210 and 212 forming salting stream 230, but exit chamber 206 across membranes 214 and 212 forming desalting stream 232.

[0089] FIG. 2B shows the same battery of FIG. 2A during a discharge cycle. Thus, chloride and sodium ions exit chamber 204 across membranes 210 and 212 to form desalted stream 232, but enter chamber 206 across membranes 214 and 212 to form salted stream 230. Notably, if the reactants become negatively charged, the membranes are inverted; however, membranes 210 and 214 are cation exchange membranes and membrane 212 is an anion exchange membrane. As previously mentioned, membranes 210 and 214, if they are cation exchange membranes, do not necessarily have to comprise the same material or be similarly sized. In the embodiment shown, the anolyte is zinc and the catholyte is a redox-active compound of Formula I. However, in other embodiments, the anolyte and catholyte may contain different redox-active compounds of Formula I.

[0090] In some embodiments, a redox-active compound already available in the art may be paired with a redox-active compound of Formula I. In some embodiments, a redox-active compound already available in the art is part of the anolyte and a redox-active compound of Formula I is part of the catholyte. In some embodiments, a redox-active compound already available in the art is part of the catholyte and a redox-active compound of Formula I is part of the anolyte.

[0091] The anolyte and catholyte of already available redox active compounds are not limited to the above-mentioned embodiments. The already available redox active components of the anolyte and / or catholyte can be aqueous solutions of any combination of the following in one or more of their oxidation states, as their ions or oxocations or oxoanions, and / or complexed with ligand(s): titanium(III), titanium(IV), vanadium(II), vanadium(III), vanadium(IV), vanadium(V), chromium(II), chromium(III), chromium(VI), manganese(II), manganese(III), manganese(VI), manganese(VII), iron(II), iron(III), iron(VI), cobalt(II), cobalt(III), nickel(II), copper(I), copper(II), zinc(II), ruthenium(II), ruthenium(III), sulphur ... and / or polymers incorporating complexing or covalent moieties of any of the foregoing species.

[0092] The anolyte and catholyte may also contain aqueous solutions of pH buffer components that may or may not be redox-active under typical operating conditions. Thus, the redox-active compounds in the anolyte and catholyte should be stable over a wide range of pH values. The redox-active compounds of Formula I are stable and can function over a wide range of pH values.

[0093] In certain aqueous embodiments, the pH of the anolyte and catholyte matches the pH of the electrolyte in the central chamber, which may be, for example, approximately neutral (pH 5-9) for water desalination, acidic (pH 0-5) for treating acidic wastewater, or alkaline (pH 9-14) for treating alkaline wastewater. In some embodiments, it may be advantageous for the anolyte pH to be slightly lower than the other chambers, such as when the anolyte is zinc / zinc chloride. In further embodiments, the pH of each of the electrolytes in the system is substantially the same within the electrochemical cell. In still further embodiments, the anolyte, catholyte, and water each have a pH between, inclusive, 3 and 10. Accordingly, the cell may include a pH monitoring and regulation system for periodic and / or continuous pH monitoring and regulation. In some embodiments, the pH of the electrolyte is between 3 and 11. In some embodiments, the pH of the electrolyte is between 5 and 9.

[0094] In further embodiments, electrodialysis battery cells such as those described in FIGS. 2A-2B may be coupled to an energy storage system when designed as a flow battery. Flow batteries are attractive for energy grid storage because they allow the battery's energy storage capacity to be decoupled from the power it can deliver. Aqueous flow batteries may be incorporated into electrochemical desalination systems because they share many common desalination capital requirements, such as pumps, piping, and cell stack design, but they may also be utilized for both desalination and energy storage. In principle, incorporating electrical energy storage into a desalination battery could further reduce costs by facilitating load shifting on the electric grid, enabling electricity arbitrage, and / or deferring investment in transmission and distribution infrastructure. A desalination battery with a high cell potential would function as a viable energy storage device. By leveraging revenue streams available for energy storage technologies, the system may defray the costs of desalination while simultaneously enabling increased adoption of renewables.

[0095] The electrolyte (i.e., the anolyte and catholyte) can be of various compositions. In some embodiments, the electrolyte composition includes a redox-active compound of Formula I and an inorganic salt. In some embodiments, the redox-active compound of Formula I is present in a concentration of 2% to up to 25% by weight. In some embodiments, the redox-active compound of Formula I is present in a concentration of 2% by weight or greater, 3% by weight or greater, 4% by weight or greater, or 5% by weight or greater. In some embodiments, the electrolyte composition may also include some amount of dissolved inorganic salt. In certain embodiments, the dissolved inorganic salt is the same material as the liquid desiccant being regenerated. In some embodiments, the inorganic salt is LiCl, LiBr, CaCl, MgCl, or any combination thereof. In some embodiments, the inorganic salt of the electrolyte composition is present in a concentration of 1% to 60% by weight. In some embodiments, the inorganic salt of the electrolyte composition is present in a concentration of 1% by weight or greater, 5% by weight or greater, or 10% by weight or greater. [Example]

[0096] Example 1A The following examples describe synthetic routes to access redox active compounds of formula I, and Scheme 1 shows synthetic routes to access redox active compounds of formula I.

[0097] [ka]

[0098] First, an alkylation reaction is carried out. In this reaction, X covalently bonds to a cyclopentadienyl ligand to produce an intermediate compound. It is understood that X may initially contain a leaving group or a group that can be converted to a leaving group, such as a halogen (e.g., Cl) or another suitable group (e.g., hydroxyl). The leaving group is replaced by Y in reaction 2. The alkylation reaction may be repeated to provide the desired number of n1 and n2 substituents. The stoichiometry of ferrocene relative to X in the alkylation reaction may be manipulated to better control n1 and n2 (described below).

[0099] Second, a reaction is carried out in which the intermediate compound reacts with Y. In this reaction, Y becomes covalently bonded to X to give the redox-active compound of formula I.

[0100] The first and second reactions may be carried out at elevated or reduced temperatures to control the reaction rate. The first and second reactions may be carried out in an organic solvent such as dichloromethane, acetonitrile, or tetrahydrofuran. The intermediate products of the alkylation reaction and the redox-active compound of Formula I may be isolated and purified via liquid chromatography or recrystallization. The intermediate products of the alkylation reaction and the redox-active compound of Formula I may be characterized by NMR, IR, mass spectrometry, or other characterization techniques.

[0101] Example 1B The following example describes a synthetic route to access the redox-active compound tetrakis((3-trimethylammonio)propyl)ferrocene tetrachloride (TTMAP-Fc). The method described in this example may be used to synthesize other redox-active compounds of general formula I.

[0102] Scheme II shows a synthetic route to access the redox-active compound TTMAP-Fc, where n1 and n2 are each 2, each X is propyl, each Y is trimethylammonio, and 4Cl - It is written in terms of the counterion, Formula I. Other counterions may also be used.

[0103] [ka]

[0104] The synthesis of TTMAP-Fc may be carried out in three steps. The first step is acylation of the cyclopentadienyl ligand to give an acyl intermediate. The acylation reaction may be carried out by combining 1 molar equivalent of ferrocene, 2 molar equivalents of 3-chloropropionyl chloride, and a stoichiometric amount of aluminum trichloride (AICl) in a reaction vessel.

[0105] The second step is the in situ reduction of the acylated intermediate formed from the acylation reaction to give the corresponding alkane intermediate. Reduction of the acyl functionality may be accomplished by combining the acyl intermediate (without workup or isolation) in a reaction vessel with a reaction mixture from a previous reaction containing sodium borohydride. Alternatively, lithium borohydride may be used to reduce the acyl functionality.

[0106] For better control of the substituents on n1 and n2, the acylation and reduction reactions may be carried out in an iterative manner, varying the stoichiometry of ferrocene (or ketone intermediate) and 3-chloropropionyl chloride. For example, to synthesize TTMAP-Fc using this method, the alkylation and reduction reactions are repeated twice, with the stoichiometry of the first alkylation reaction being 1:2 ferrocene to 3-chloropropionyl chloride and the stoichiometry of the second alkylation reaction being 1:2 ferrocene to 3-chloropropionyl chloride.

[0107] To synthesize compounds of general formula I in which n1 and n2 are greater than 2, the alkylation and reduction reactions may be repeated an appropriate number of times with an appropriate stoichiometric ratio of ferrocene (or the ferrocene product from the first alkylation and reduction steps) to 3-chloropropionyl chloride. For example, if the desired compound has three n1s and three n2s, then the alkylation and reduction reactions are carried out a total of three times, with a 1:2 ratio of ferrocene to 3-chloropropionyl chloride in each alkylation reaction. For example, if the desired compound has two n1s and one n2, then the alkylation and reduction steps are carried out twice, with a 1:2 stoichiometry of ferrocene to 3-chloropropionyl chloride in the first alkylation reaction and a 1:1 stoichiometry of ferrocene to 3-chloropropionyl chloride in the second alkylation reaction.

[0108] The alkylation and reduction reactions may be carried out at elevated or reduced temperatures to control the reaction rate. The alkylation and reduction reactions may be carried out in organic solvents such as dichloromethane, acetonitrile, or tetrahydrofuran. The products of the alkylation and reduction reactions may be isolated and purified via liquid chromatography or recrystallization. The products of the alkylation and reduction reactions may be characterized by NMR, IR, or mass spectrometry.

[0109] The third step in the synthesis of TTMAP-Fc is the quaternization of the alkane intermediate, which is mixed with excess trimethylamine and ethanol in a sealed reaction vessel and heated to 60-80°C.

[0110] The product of the quaternization step may be isolated and purified via liquid chromatography or recrystallization. The product of the amination step may be characterized by NMR, IR, or mass spectrometry.

[0111] Example 2 The degradation of TTMAP-Fc, or any other redox-active compound of general formula I, can be tested by exposing TTMAP-Fc to a continuous voltage supply and monitoring the decomposition and polymerization of the cyclopentadienyl ligand via mass spectrometry. Figure 5 shows an example of an experimental setup. In this setup, 5% by weight of the redox-active compound (TTMAP-Fc) is mixed in a solution of 5% by weight of lithium chloride. The solution is exposed to a continuous voltage. Aliquots of the solution may be taken at various times and examined by mass spectrometry. To test the stability of the redox-active compound in different environments, experiments may be performed with different salt concentrations, different redox-active compound concentrations, different pH levels, and different salts.

[0112] Example 3 An energy-efficient redox-assisted electrodialysis system may be constructed in accordance with that shown in Figure 1B, in which a redox-active compound of general formula I is used in place of Fe(CN) as the redox-active compound. The use of a redox-active compound of general formula I is expected to avoid the formation of insoluble solids formed from the reaction of Fe(CN) with certain metal ions. The anode and cathode may be constructed of three sheets of porous carbon fiber paper (SGL39AA, available from SGL Carbon in Wiesbaden, Germany) compressed onto a pyrosealed graphite block with serpentine flow channels (available from Entegris in Billerica, MA) and separated from the central desalting / salting chamber by a VITON® gasket (available from Atlantic Gasket Corporation, Philadelphia, PA). The anion exchange membrane may be FUMASEP FAS-30 (available from the Fuel Cell Store in College Station, TX) and the cation exchange membrane may be FUMASEP E630 (available directly from FuMATech).

[0113] In typical operating modes, TTMAP-Fc of 0.05M and TTMAP-Fc of 0.05M +A solution of 0.6 M NaCl and 1% NaCl (50 mL) can be flowed across the anode / cathode, 0.6 M NaCl can be in both the salting and desalting chambers (50 mL each), and the high-frequency ASR of the system is measured. The current density is directly proportional to the salt transport rate. Depending on how the cell is operated, the specific energy consumption can be approximately 40 mA / cm. 2 The current density is lower than that of conventional electrodialysis, about 10 mA / cm 2 Although lower than that of reverse osmosis, the specific energy consumption for reverse osmosis rises sharply with increasing uptake salinity. If desired, the cell can produce more desalination at a higher specific energy consumption. Improvements to the cell's specific energy consumption are expected as a result of the reduced ASR. With this system, direct production of water at potable salinity (<0.5 ppt NaCl) is possible in a single step from seawater (e.g., 35 ppt NaCl). A constant voltage of 0.5 V may be applied to the cell, increasing the salinity of the water in the salting chamber and decreasing the salinity of the water in the desalination chamber.

[0114] In another possible electrodialysis system, several stages of desalination may be performed, but redox-active compounds are used, optionally with a supporting electrolyte containing a salt in the salting / desalting step, at concentrations appropriate to the salt concentration in the salting / desalting step to minimize water transport in each stage. This reduces or minimizes energy inefficiencies resulting from water transport from the salting / desalting chambers to or from the anolyte / catholyte chambers.

Claims

1. A redox-active compound comprising: a ferrocene core containing two cyclopentadienyl ligands; One or more substituents each bonded to said cyclopentadienyl ligand independently selected from -X-Y, wherein X is C 1 ~C 12 a linear or branched alkyl of, Y is an oligo(ethylene glycol), hydroxyl, trialkylammonio, alkylimidazolio, sulfonate, sulfate, carboxyl, phosphate, phosphonate, ammonium, or a soluble group containing one or more of a nitrogen-containing heterocyclic ring, and the total number of substituents bonded to the cyclopentadienyl ligand per ferrocene core is 3 or more, one or more substituents, and a redox-active compound.

2. The soluble group is a trialkylammonio or alkylimidazolio containing an alkyl substituent, and the alkyl substituent on the soluble group is independently C 1 to C 12 alkyl, and the redox active compound according to claim 1.

3. The redox-active compound according to claim 1, wherein Y is a trialkylammonio and at least one of the alkyls is methyl.

4. The redox-active compound according to claim 1, wherein the total number of substituents bonded to the cyclopentadienyl ligand is 4 or more per ferrocene core.

5. X is C 2 ~C 6 The redox active compound according to claim 1, wherein it is such.

6. The redox-active compound according to claim 1, further comprising one or more counterions.

7. A redox-active composition comprising the redox-active compound according to claim 1.

8. An electrolyte composition comprising: an aqueous solution of the redox-active compound according to claim 1 and a salt.

9. The electrolyte composition according to claim 8, wherein the redox-active compound is present at a concentration of 2% by weight or more.

10. The electrolyte composition according to claim 8, wherein the redox-active compound is present at a concentration of at most 25% by weight.

11. wherein the salt is LiCl, LiBr, CaCl 2 , MgCl 2 , or a combination thereof, the electrolyte composition according to claim 8.

12. The electrolyte composition according to claim 8, wherein the salt is present at a concentration of 1% by weight or more.

13. The electrolyte composition according to claim 8, wherein the salt is present at a concentration of at most 60% by weight.

14. The electrolyte composition according to claim 8, wherein the aqueous solution has a pH of 3 to 11.

15. A system comprising: a device comprising: a first reservoir having a first input portion and a first output portion; a second reservoir having a second input portion and a second output portion; a first electrolyte chamber having a first electrode; a second electrolyte chamber having a second electrode; a first type of membrane disposed between the first reservoir and the second reservoir, wherein the first reservoir and the second reservoir are in electrolytic communication with each other through the first type of membrane; a second type of membrane different from the first type disposed between the first electrolyte chamber and the first reservoir, wherein the first electrolyte chamber and the first reservoir are in electrolytic communication with each other through the second type of membrane, and disposed between the second electrolyte chamber and the second reservoir, wherein the second electrolyte chamber and the second reservoir are in electrolytic communication with each other through the second type of membrane; a first solution disposed within the first electrolyte chamber; a second solution disposed within the second electrolyte chamber, and a system comprising an apparatus comprising one or both of the first solution and the second solution, the one or both comprising a redox active compound according to claim 1. **Claim 16** The soluble group is a trialkylammonio or alkylimidazolio containing an alkyl substituent, and the alkyl substituent on the soluble group is independently C 1 to C 12 alkyl of, the system according to claim 15. **Claim 17** The system according to claim 15, wherein Y is trialkylammonio and at least one of the alkyls is methyl. **Claim 18** The system according to claim 15, wherein the total number of substituents attached to the cyclopentadienyl ligand is 4 or more per ferrocene core. **Claim 19** X is C 2 ~C 6 The system according to claim 15, wherein it is so.

Images