Method and device for storing electrical energy in chemical redox compounds - Efficient redox flow battery

Abstract

Device for storing electrical energy in chemical redox compounds in the form of a redox flow battery, based on at least two liquid redox electrolytes (catholyte and anolyte (4)), comprising at least one electrochemical flow cell consisting of two half-rows, each with an electrode compartment through which a redox electrolyte flows, and a collector electrode made of a chemically inert, electron-conducting material (2) and circulating pumps (3), wherein the two half-cells are separated by an impermeable, electrically non-conducting wall (1), characterized in that a) the electrochemical conversion of the redox electrolytes in question is quantitatively controlled during a single half-cell flow and discharged redox electrolytes leave the flow cell, b) during the discharge process, the discharged catholyte and anolyte fluids or parts thereof form an uninterrupted electrolytic current key within a defined mixing zone of both, which electrolytically connects the two half-cells at the outlet, and during the charging process, a corresponding uninterrupted current key is formed by a defined flow division of discharged catholyte and anolyte before the inlet of both half-cells, both of which replace a semipermeable membrane. c) Catholyte and anolyte contain more than one redox system, d) the discharged redox electrolytes exiting the flow cell and combined according to the current key are separated again into catholyte and anolyte fractions and stored separately from their still charged fraction, e) the discharged redox electrolytes are temporarily stored in containers that can act as a pump.

Description

[0001] The invention relates, according to the preamble of claim 1, to a device for storing electrical energy in chemical redox compounds in the form of a redox flow battery (RFB) based on at least two liquid redox electrolytes, comprising at least one electrochemical flow cell consisting of two half-cells, each with an electrode compartment through which a redox electrolyte flows, and each half-cell having an electrode made of a chemically inert, electron-conducting material. The conventional semipermeable, ion-conducting separating membrane between the two half-cells is, according to the invention, comprised of a novel current-key construction formed from the redox electrolytes reacted within the cell, which is positioned after the cell outlet during discharge and before the cell inlet during charging. Separate storage of these reacted electrolytes offers further advantages based on quantitative conversion.

[0002] Regarding the economic significance of the technical field of the present invention, it should be noted that the increasing demand for so-called renewable energies has led to a steep rise in alternative energy conversion technologies. Wind and solar energy are among the renewable energy sources with the highest growth rates. However, a disadvantage of both is that they cannot be used continuously, i.e., they occur intermittently. Therefore, their full and efficient utilization necessarily requires acceptable energy storage (in terms of both price and environmental impact) to enable the use of these alternative energies even during periods of windlessness or at night. Wind and solar energy have the significant advantage of being usable in a decentralized manner, which, with the help of emerging "smart grid" technology, makes large, costly, and increasingly controversial overhead power lines less urgent.To achieve environmental protection goals and a decentralized energy supply, a strong expansion of practical and inexpensive storage technologies is absolutely necessary in parallel, to which the present invention contributes.

[0003] Of the various known types of energy storage, electrochemical energy storage is distinguished by the fact that it does not require a change in energy type for the crucial process of electricity generation, thus avoiding undesirable conversion losses (efficiencies). Among electrochemical energy storage devices, the now well-known redox flow batteries possess decisive advantages that make them particularly suitable for decentralized applications. Like all conventional batteries, the redox flow battery uses chemical energy to store electrical energy. The unique aspect here is the energy storage in the form of two redox compounds dissolved in electrolyte solutions, each with a distinct redox potential. These compounds react with each other via a load and electron-conducting electrodes.In redox flow batteries, the redox reactions and their electron exchange do not occur, as in a classical, chemical-analytical redox titration, directly in a homogeneous solution from a redox-active species dissolved in a solvent to a redox partner also dissolved in the same solution. Instead, they occur indirectly via two collector or electrodes and a load in the circuit. While one electrode accepts electrons from a suitable redox system in an electrolyte solution (i.e., acts as an oxidizer), the second electrode in the circuit releases these electrons back to a suitable redox partner in an electrolyte solution after use by a load (i.e., acts as a reducer). The positive electrode is called the cathode, the negative the anode; the electrolyte solutions in contact with them are called the catholyte and anolyte.To ensure that the two reacting redox systems, when charged, exchange electrons only via the "detour" of the two electrodes and an external load, and not react directly with each other in solution (which would reduce the Coulombic efficiency of the electrochemical cell), conventional redox flow batteries prevent direct mixing of the two charged redox electrolytes by means of an ion-conducting, semipermeable separating membrane between the two collector electrodes. This separating membrane must fulfill tasks in all known redox flow batteries that require significant compromises. On the one hand, it should prevent electrochemically active substances from passing through; on the other hand, it should allow non-redox-active substances, such as protons and / or hydroxide ions, to pass through as freely as possible in order to keep the cell's internal resistance low.This internal, purely electrolytic ion current is of the same order of magnitude as the external current transported by electrons and should therefore take place as unhindered as possible, which causes major problems in practice.

[0004] Within the scope of the present invention, the term "redox flow battery" is used in its usual sense, utilizing two liquid redox electrolytes that are pumped through an electrochemical flow cell. The basic structure of a conventional redox flow battery, comprising two liquid redox electrolytes, two collector electrodes in an anolyte and catholyte half-cell separated by an ion-conducting separator, two redox electrolyte circulation pumps, and two electrolyte reservoirs, is known to those skilled in the art. For the sake of simplicity, the terms "half-cell" or "cell" are sometimes used instead of "electrochemical half-cell" or "electrochemical cell," and the redox electrolytes contained in the redox system are referred to simply as "electrolytes."

[0005] The object of the present invention is to significantly improve upon the disadvantages of all conventional redox flow batteries based on two circulating redox electrolyte liquids, thereby increasing their practicality and achieving greater market success. The performance improvements include: a) reducing manufacturing costs by eliminating the need for an expensive, ion-conducting semipermeable separating membrane; b) increasing reliability by eliminating the vulnerable separating membrane and thus reducing maintenance requirements; c) significantly reducing internal resistance; d) maintaining a more constant battery terminal voltage during discharge; and e) effectively increasing electrical efficiency by reducing pump energy losses. This object is achieved by a method according to the main claim and devices according to the dependent claims.The present invention discloses, in general terms, methods for constructing membraneless redox flow batteries that offer significant advantages compared to the prior art. In accordance with a conventional redox flow battery, the redox flow battery according to the invention has a positive half-line and a negative half-line, but instead of a semipermeable membrane separating the two half-lines, it features a novel electrolytic current key according to the invention – formed from a special mixing zone of the two reacted electrolyte solutions, as well as at least two electrolyte reservoirs. According to the invention, this current key is located downstream of the electrochemical flow cell during the discharge process and combines the redox electrolytes discharged in the cell, and during the charging process, it is located upstream of the discharged redox electrolytes entering the cell.The prerequisite is a quantitative or near-quantitative redox conversion during each single flow through the electrochemical cell.

[0006] As a direct consequence of the necessary quantitative turnover of the redox systems involved, further constructive possibilities arise to save large amounts of pumping energy, which significantly impair the efficiency of conventional redox flow batteries.

[0007] A particular advantage of redox flow batteries is that the complete cell formed from the two half-cells can be easily connected in series (so-called stacks) using bipolar electrodes, thus multiplying its power output (maximum voltage). The maximum deliverable current then depends on the size of the active electrode surface and the efficient transport of the electrochemically active components (reactants) to this surface. The maximum battery voltage of a battery stack is calculated by multiplying the single-cell voltage by the number of cells connected in series. In contrast to conventional batteries (e.g., lead-acid or metal hydride), the storage capacity in watt-hours (Wh) is determined by the quantity of reacting redox systems used.Since these redox electrolyte solutions are stored separately and only pumped through the electrochemical flow cells when needed, the significant design advantages of easy scalability arise: The two redox electrolytes are typically stored in two tanks outside the electrochemical battery flow cells. Because of this galvanic isolation, there is no self-discharge, unlike with other batteries. The redox flow battery is the only type of electrochemical energy storage device in which the energy quantity (determined by the tank size) and power (determined by the electrochemical flow cell) can be increased or decreased independently of each other.

[0008] Commercially, the so-called "all-vanadium redox flow battery" (VRFB) is the most widely used redox flow battery. Its basic principles were developed as early as the 1980s. The VRFB utilizes the fact that vanadium possesses two stable redox pairs that react with each other redoxically through comproportionation. An advantage is that only a single chemical element is used for both the catholyte and the anolyte. This means that if the two redox electrolyte circuits accidentally mix across the separating membrane, both redox electrolytes do not need to be completely reconstituted and replaced. However, this process, also known as "cross-over," results in a loss of stored energy.The redox electrolyte fluid in VRFBN consists predominantly of dilute sulfuric acid, which supplies the necessary protons for electrolytic reaction across the semipermeable membrane and in which the vanadium species are dissolved. The energy is stored in the redox pairs V. 2+ / V 3+ (negative half-cell) and VO 2+ / VO2 + (Positive half-cell) stored in sulfuric acid solution, which is located in separate storage tanks. The prior art and the basic principle in the field of redox flow batteries will be explained in more detail below using the vanadium redox flow battery as an example, since this redox pair is particularly suitable for the present invention. However, the disclosed basic principles can be applied analogously to other electrochemical cell types or redox flow batteries with other redox pairs.

[0009] In VRFB, the following electrochemical reactions take place in the relevant electrochemical half-cells in simplified terms, without details of the exact molecular mechanism: E0=1.00 V E0=−0.25V

[0010] On the cathode side, during the charging cycle, vanadium cations of oxidation state +4 are oxidized to vanadium cations of oxidation state +5 by the release of an electron to a collector electrode. A voltage source connected to charge the redox electrolytes reduces the vanadium cations of oxidation state +3 to divalent, positively charged vanadium cations on the anode side during charging. During discharge (current draw), the pentavalent vanadium compound oxidizes the divalent ion in the other half-cell back to its trivalent form, while the vanadium ion of oxidation state five is reduced back to its tetravalent starting state. Electron transport from the divalent state to the reduction of the pentavalent ion occurs via the two electrodes and an external load that completes the circuit.In the electrochemical flow cell, the circuit is closed electrolytically by the transport of protons through the semipermeable separating membrane. Without current being drawn, a voltage of 1.25 V is present between the positive (+) and negative (-) terminals of a completed cell, according to the half-cell potentials (1) and (2). When initially constructing a VRFB, a sulfuric acid solution with a potential of approximately 1.7 M V is used. 3+ and V4 + -compounds, which corresponds exactly to the situation in a completely discharged state. Both oxidation states can coexist in a redox-stable manner. The starting electrolyte, which contains equimolar amounts V 3+ - and V 4+ -compounds, is then electrochemically converted into catholyte (V) when the electrolyte flows through both half-cells under applied external voltage (usually galvanostatic charging cycle). 4+ becomes V 5+ oxidized) and anolyte (V 3+ becomes V 2+(reduced) transformed.

[0011] To increase power output by increasing the voltage, the individual cells are typically connected in series using bipolar electrodes. These combined cells are called a cell stack. The individual electrochemical cells are usually connected in parallel to each other by the redox electrolytes, while the individual cells are electrically connected in series. In this configuration, the charge state of the redox electrolytes in each of the half-cells of the cell stack is the same. Generally, in conventional redox flow batteries, a quantitative redox conversion is not targeted during a single flow through the cell. However, due to circulation and mixing with unreacted electrolytes in the reservoirs, the initial concentration of both reacting redox systems decreases continuously as the battery discharges.If the state of charge (SOC) is too low, concentration polarization occurs in the half-cells, which reduces the clamping voltage and, due to the resulting reduced mass transfer, unnecessarily wastes pumping energy. Therefore, discharges near SOC = 0 are avoided.

[0012] Unlike many other redox flow batteries with different redox pairs, in the VRFB, where a noticeable crossover of the electroactive components through the separating membrane is practically unavoidable, the redox electrolyte solutions do not need to be completely replaced. Instead, they are simply combined when efficiency drops too significantly after a certain number of charge and discharge cycles, and a charging process is performed as during initial commissioning. For example, it has been shown that the number of cycles up to a capacity loss of 50% increased from 145 to 598 when, upon first reaching this loss, the fully discharged redox electrolytes were mixed and, after a 1:1 distribution, the procedure was repeated as during the initial charging [Y. Zhang et al. / Applied Energy 204 (2017) 373-381]. However, this requires an additional process and leads to undesirable downtime, which is avoided by the present invention.

[0013] For an overview of the numerous redox systems used to date, we refer here only to two recent review articles [P. Leung, RSC Adv., 2012, 2, 10125-10156; J. Noack et al.; Angew. Chem. Int. Ed. 2015, 54, 9776-9809]. Since a higher cell voltage correlates with a higher power density, numerous RFBs have also been introduced that operate with non-aqueous solvents, as aqueous systems hardly allow voltages higher than 1.7 V due to the decomposition of water [P. Leung et al., Recent developments in organic redox flow batteries: A critical review, Journal of Power Sources 360 (2017) 243-283]. However, these systems usually involve larger quantities of flammable solvents. This potential hazard must be taken into account. Furthermore, the high cost of organic solvents hinders wider use. Resource-efficient redox flow batteries based on water-soluble, purely organic redox systems are also of interest.

[0014] Despite their significant advantages over traditional batteries, redox flow batteries have not yet achieved a major commercial breakthrough because—contrary to theoretical considerations and descriptions in review articles—they have encountered problems with their stability and reliability in routine applications. Generally, there are too many service intervals due to high material stress caused by strong acids or alkalis, or by elevated temperatures. Additionally, cycle stability decreases over the operating period due to undesirable side reactions and cross-over effects. Cost estimates have shown that over 60% of the total cost of a redox flow battery system is attributable to the electrochemical flow cell stack alone, with the often very fragile semipermeable membrane separating the anolyte and catholyte compartments again accounting for the largest share of the cost.To clarify the problem that this invention claims to solve, the often unknown disadvantages of all conventional redox flow batteries with a separating membrane must be described in more detail.

[0015] A major problem is the electrical resistance of the semipermeable separating membrane, which usually accounts for the largest share of the total internal resistance of a conventional redox flow battery cell. These cells generally exhibit a decrease in terminal voltage with increasing current draw, which is primarily caused by the ohmic effect (U = i × R) of the internal resistance R of the electrochemical cell as the current flow i increases. In some types, increasing the acid concentration does increase the membrane's conductivity, but it decreases the ionic conductivity of the electrolyte solutions outside the membrane phase.While the use of thinner membranes reduces their ohmic resistance, it leads to more fragile membranes that are easily damaged mechanically and to a strong increase in the unwanted permeation of the redox-active molecules or ions involved (“cross-over”), which reduces the Coulomb efficiency and also lowers the cycle stability.

[0016] Currently, the most commonly used semipermeable separator membranes in conventional redox flow batteries are perfluorinated sulfonic acid ion exchange membranes from DuPont (Nafion®). The undesirable "cross-over" effect of electrochemically active substances is primarily based on the following mechanisms: a) diffusion due to concentration gradients, b) migration due to potential gradients, and c) convection due to pressure gradients. All three mechanisms have been studied in detail in the literature, but satisfactory solutions for completely preventing this undesirable "cross-over" effect of electrochemically active species have not yet been found. Furthermore, voltage-reducing electro-osmotic effects also occur at the semipermeable separator membranes, increasing with rising current density.The contributions of migration and electro-osmotic ion transport are independent of the membrane thickness, while pure diffusion is inversely proportional to the membrane thickness and independent of the current density.

[0017] Despite numerous attempts to develop an ideal separation membrane, the disadvantages that have so far hindered the greater commercial success of redox flow batteries have not been completely overcome. Even in redox flow types where the "cross-over" effects do not necessitate a complete exchange of the two redox electrolytes, such as VRFBs containing only vanadium species, further undesirable disturbances occur in practical operation. For example, water can also pass through the semipermeable separation membrane and thereby alter the vanadium concentrations in the two redox electrolytes differently, resulting in a so-called "cell imbalance" that has received little theoretical consideration and is often omitted from review articles.In a potential "cross-over" of vanadium cations and sulfate anions in VRFB, the diffusion rate determines the net transport of vanadium species from one half-cell to the other. For example, the higher-valent vanadium ions (V. 5+ and V 4+ ) in the existing electric field (corresponding approximately to the clamping voltage of a cell across the membrane thickness) faster than the lower valence vanadium ions (V 2+ and V 3+Water passes through the semipermeable membrane, thus altering the initial concentrations in the respective redox electrolytes. Water can also pass through the membrane via several simultaneously occurring phenomena: firstly, by carrying along the hydration shell of the net transported ions, and secondly, through the effect of osmosis. Asymmetrical water transport leads to the precipitation of the less soluble redox system and to additional undesirable Donnan potentials across the membrane. In practice, contrary to theoretical assumptions, this results in a decrease in battery capacity over the operating period (decreasing cycle stability). Water transport is strongly dependent on the membrane properties and the composition of the redox electrolytes.For example, in cation exchange membranes (Nafion®), there is a net water transport from the negative to the positive half-cell, while in anion exchange membranes and nanoporous separators, it proceeds in the opposite direction. The water transport is also determined by the state of charge (SOC) of the pumped redox electrolytes, which changes continuously in conventional redox flow batteries, unlike in the present invention.

[0018] The commonly used Nafion®-based membranes exhibit not only generally excessively high crossover rates but also very high costs. Therefore, significant efforts have been made worldwide to develop alternative membranes. For example, SPEEK (sulfonated polyetheretherketone) membranes, polybenzimidazole membranes, and nanofiltration membranes have been tested as separators in redox flow batteries. However, most alternative membrane materials exhibited insufficient chemical stability, particularly for VRFBs with their corrosive electrolytes, which is why the search for more suitable membranes continues. Additionally, many semipermeable separator membranes swell upon contact with anolyte or catholyte fluids, and this can lead to mechanical stresses in the cell structure, resulting in leaks.This leads to increased, undesirable mixing of the two charged electrolyte solutions (electrochemical short circuit), necessitating a complete replacement with freshly prepared redox electrolytes and resulting in unwanted maintenance. Current research on redox flow batteries therefore focuses on further improving their performance, particularly regarding reliability, maintenance-free operation, energy density, efficiency, and cycle stability.

[0019] The present invention solves these central and important problems of all conventional redox flow batteries, which require a semipermeable separating membrane, by disclosing alternative electrochemical half-cell arrangements and devices that, individually, in combination, or together, eliminate the need for the fragile and expensive semipermeable separating membrane and, through other measures and design arrangements that only then become possible, significantly improve the performance, efficiency, and overall performance of redox flow batteries. According to the invention, some disclosed modifications to the design of redox flow batteries are essential to achieve the desired positive effect. Only if highly discharged redox electrolytes are available after a single flow through the electrochemical cell, as required by the novel current-key design, and are stored separately, are storage tanks in the form of double-acting piston pumps possible.In tandem and hourglass configurations, such devices according to the invention require hardly any energy to pump the electrolytes. For the first time, they also make it possible to pump highly viscous electrolytes through the redox flow battery without significant energy losses. These devices represent particularly advantageous subsequent developments, which, however, are only made possible by the main claim and also serve the general goal of building better and more cost-effective redox flow batteries.

[0020] Membrane-free mini redox flow batteries with laminar flow of anolyte and catholyte were first presented in 2002 by R. Ferrigno et al. in a short, two-page article [R. Ferrigno et al., J. AM. CHEM. SOC. 2002, 124, 12930-12931]. They are now being described with increasing frequency in the literature and also through patent applications. This underscores the significant technical need to develop membrane-free redox flow batteries to avoid the known disadvantages of membrane-based designs. Of the numerous publications dealing with mini redox flow batteries with co-laminar electrolyte flow, only two will be mentioned here as examples. One of the first studies deals with a membraneless mini fuel cell for hydrogen, methanol and formic acid using the oxidizing agents potassium permanganate or oxygen dissolved in sulfuric acid [Microfluidic fuel cell based on laminar flow, ER Choban et al.[Journal of Power Sources 128 (2004) 54-60]. Another example uses porous graphite flow-through electrodes in a mini-VRFB with co-laminar electrolyte flow, achieving a power output of 20 mW / cm² at low flow rates with a redox electrolyte conversion (fuel utilization) of over 90% in a single flow [A Microfluidic Fuel Cell with Flow-Through Porous Electrodes, E. Kjeang et al., J. AM. CHEM. SOC. 2008, 130, 4000-4006]. This, along with Example 1, provides further confirmation that near-quantitative conversions are possible in a single flow through the half-cells. A recent overview of membraneless redox flow batteries based on this was published in 2017 [MO Bamgbopa et al., Renewable and Sustainable Energy Reviews, Volume 70, April 2017, Pages 506-518].

[0021] Following the initial feasibility study of electrochemical micro-laminar flow cells by R. Ferrigno et al., several patent applications were filed in the USA, e.g.: US 2006 / 0003217 A1 dated June 10, 2005, US 2006 / 0228622 A1 dated November 21, 2005, US 2010 / 0112391 A1 dated October 30, 2009, US 2011 / 0123902 A1 via PCT application February 25, 2009, US 2015 / 0099199 A1 dated October 2, 2014, all of which are based on the principle of co-laminar flow of catholyte and anolyte and underline their economic significance. The University of Illinois (IL, USA) holds two US patents in the field of electrochemical membraneless mini-flow cells, which avoid undesirable electrochemical mixing of reactants by means of co-laminar flow: US 6713206 B2 was filed on January 14, 2002, even before the first publication by R. Ferrigno on June 2, 2002, and was granted on March 30, 2004. The second patent, US 7252898 B2, was filed on June 27, 2003, and granted on August 7, 2007.

[0022] The methods and arrangements based on membraneless electrochemical cells with co-laminar flow contact, disclosed in all previously published applications, patent applications, and granted patents, exhibit several disadvantages. The most obvious disadvantage is their low performance, which severely limits their applicability because they must compete with the highly developed lithium-based batteries. In contrast to the present invention, all previous publications based on laminar flow are based on so-called "microfluidic cells," that is, highly miniaturized redox flow batteries on a lower centimeter scale. A semipermeable membrane is only unnecessary if mixing of the charged analyte and catholyte solutions during the contact time can be avoided.In all the studies and patents described, this is achieved by establishing the necessary electrolytic connection between catholyte and anolyte through a co-laminar liquid flow over a distance of only a few centimeters. Such laminar liquid flows in the low Reynolds number range exhibit no turbulence and thus prevent significant undesirable mixing ("cross-over") of the reactants. However, to achieve strictly laminar liquid flows, the flow velocity, the adjacent liquid interfaces, and vibrations must not be too high to avoid turbulence. This can only be achieved using very small, smooth-walled micro-flow reactors. These contact zones, often measuring only a few centimeters, and their correspondingly small electrode surfaces naturally only deliver currents in the lower mA range and are, due to further limitations, hardly practical.A significant amount of the charged redox electrolytes is also lost, as they mix after the short laminar flow path due to the design. Large diffusion overpotentials occur during current draw because of the limited electrode surface area. The electrochemical reaction volume is often less than 1 mm³. 3 The contact zone that forms the current key is only a few millimeters wide, resulting in noticeable internal resistance. Even more serious, however, are further disadvantages of this method for avoiding the problematic separating membrane in redox flow batteries: such mini-redox flow batteries do not allow for simple regeneration of the redox electrolytes using the same arrangement, nor do they permit high-performance designs. The more powerful of these mini-redox flow batteries are based on porous electrodes through which current flows, which then require higher pumping capacities.

[0023] The current state of the art for higher-performance, membrane-free redox flow batteries describes only so-called hybrid flow batteries with solid metal electrodes or gas diffusion electrodes. In these batteries, two redox systems dissolved in electrolytes do not react with each other via electrodes and external loads. With solid metal electrodes, which either dissolve during discharge or where metal deposition occurs during charging, this redox reaction takes place locally at a phase boundary and is associated with phase regeneration. Metal deposition from the solution can lead to inhibition of crystallization and undesirable whisker formation. Furthermore, the capacity of such hybrid redox flow battery types is limited by the size of the metal electrode in the respective half-cell. A major advantage of true redox flow batteries—their independent scalability—is lost in this design.In electrochemical flow cells with gas diffusion electrodes, atmospheric oxygen, similar to fuel cells, typically serves as the oxidizing agent, thus eliminating the need for an oxidizing electrolyte. Here, the less-than-optimal kinetics of oxygen reduction limit higher current densities. Even the use of more expensive catalysts has not significantly altered this limitation. Such configurations can be considered a special type of redox flow battery. While they theoretically do not require a separating membrane, in practice they often employ one to prevent short circuits caused by unwanted dendrite growth or excessive wetting of the gas diffusion electrodes.

[0024] In WO 2014 / 026728 A1 (priority date 14.08.2012), an attempt is made to reduce the electrical resistance of the separating membrane between the anolyte and catholyte, which is to be achieved by a size-exclusion membrane. According to the invention, water-soluble, high-molecular-weight compounds, such as redox-active polymers or oligomers, are used as redox-active components, and a size-exclusion membrane is used as the separating membrane to separate the high-molecular-weight redox-active components in the two half-cells. Nevertheless, the narrow ion channels of the membrane still represent a considerable electrolytic resistance in the nanometer range, which, at a larger current flow according to i × R, results in a certain detrimental voltage drop in the electrochemical cell – caused solely by this factor. Additionally, undesirable osmotic phenomena or Donnan potentials occur across the membrane.However, the increased viscosity of these polymer solutions is even more detrimental, requiring additional pumping energy and significantly reducing efficiency. If radicals occur as intermediates in organic redox compounds, there is a high risk of electrode fouling due to radical polymerization. These polymerization reactions must be prevented to avoid blocking the electrode surfaces. For this reason, this theoretically advantageous design has not yet achieved commercial success.

[0025] Besides the principle of co-laminar micro-contact between the two redox electrolytes, the phase boundary between two immiscible liquids can also be used as a membrane substitute if it can be ensured that the partition coefficients of the electroactive compounds prevent significant crossover. Navalpotro et al. [Angew. Chem. Int. Ed. 2017, 56, 12460-12465] describe basic concepts of membraneless redox flow batteries based on two immiscible liquids. Here, an acidic aqueous solution of hydroquinone and a hydrophobic ionic liquid of 1-butyl-1-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide with dissolved benzoquinone were used. Both redox electrolytes form a stable phase boundary upon contact with the heavier ionic liquid at the bottom. Stabilization of the labile phase boundary under more turbulent flow conditions was not described.A further disadvantage is that the cost savings from the absence of a semipermeable membrane are lost due to the more expensive solvents, and here too, a continuous cycle occurs during discharge, leading to unstable cell voltages and increased pumping energy. Bamgbopa et al. [Electrochimica Acta 267, February 2018, pp. 41-50 DOI 10.1016 / j.electacta.2018.02.063] describe an "all-iron" redox flow battery with the same ionic liquid as Navalpotro as the organic phase, as well as a concept in which the mixing of the electrolytes is reduced by the laminar flow of the electrolytes. In one embodiment, the combination of the electrolytes is described. The aqueous anolyte contains iron(II) sulfate as the redox reagent, and the organic catholyte contains iron(III) acetylacetonate. However, a corresponding redox flow battery lost 40% of its original capacity after 25 cycles.The cycle stability is therefore significantly lower than that of conventional VRFBs! The cost of the ionic fluid must also be considered. According to the Merck catalog, 50 g costs approximately €350. For medium-sized battery capacities with storage volumes in the cubic meter range, such redox flow batteries cannot compete with conventional batteries in terms of cost. In the event of a fire, corrosive hydrofluoric acid can be produced from the ionic fluid described above. The "cross-over" effect is controlled by the respective distribution equilibrium and should not be disregarded. The resulting additional, non-beneficial redox systems in the other electrolyte can be electrochemically detrimental because they are reduced or oxidized at different potentials than the main redox system and, unlike specifically selected "auxiliary redox systems," do not increase the conversion rate of the main redox system.

[0026] The solution revealed here for avoiding the known separation membrane problems in conventional redox flow batteries with two liquid, circulating redox electrolytes consists in detail of pumping the catholyte and anolyte only once through the relevant, but appropriately designed, electrochemical half-cells of a redox flow battery, whereby both redox electrolytes must be completely or almost completely redox-converted; that is, the catholyte and anolyte solutions must exit a single redox flow battery cell discharged or almost discharged. Fig. Figure 1 schematically and in a simplified manner illustrates the different design of a redox flow battery according to the invention, without a semipermeable separating membrane, using a traditional, conventional redox flow battery setup as an example during discharge. The fundamental difference lies in replacing the otherwise usual semipermeable separating membrane between the two half-cells with an impermeable, electrically non-conductive wall (1) and in arranging the necessary purely electrolytic connection between the two half-cells either upstream or downstream of the respective collector electrodes in the half-cells. According to the invention, the collector electrodes (2) are also designed differently than in conventional redox flow batteries in order to enable a nearly quantitative conversion in a single flow within the defined time. The redox electrolytes can be pumped through the electrochemical cell by pumps (3), as usual.However, according to the invention, it is more advantageous not to return the redox electrolytes, after quantitative conversion in the catholyte and anolyte compartments, to the storage containers (4), where they would mix with the unreacted portion. This dilution of the concentration of electrochemically active material leads to an undesirable reduction in clamping voltage at levels of charge (SOC) below 50% due to concentration polarization in the half-cells and inefficient pumping, since the pumping energy requirement remains constant and does not depend on the SOC. The core of the invention is a special, particularly low-resistance current-key design (5). Immediately after or simultaneously with exiting the respective half-cells, the discharged catholyte and anolyte solutions are turbulently and over a large area combined in the respective catholyte and anolyte compartments at the shortest possible distance between the two collector electrode ends (“jet-stream” principle).According to the invention, the mixing zone of the two discharged redox electrolytes forms a kind of novel electrolytic current key or electrolyte bridge, which is particularly low-resistance and thus leads to a very advantageously low internal resistance of the redox flow battery in question. The discharged redox electrolytes thus combined are again divided 1:1 or empirically determined differently (compensating for cell imbalance effects) by a volume flow controller (6) and are advantageously stored separately from the still undischarged portion in the storage tanks for the purpose of recharging. In a less advantageous conventional arrangement with continuous pumping, the circulation pumps (3) are controlled according to the invention by measuring electrodes (7) so that an almost quantitative electrochemical conversion takes place in the respective half-cells and undesirable electrolysis of the solvent is avoided.

[0027] Fig. Figure 1 shows, in a simplified schematic way, the principle of the invention when, in order to illustrate the significant differences, as little as possible is to be changed in the underlying, conventional structure, which, however, precludes crucial further advantages. Here, a novel current key construction (5) is located behind the two electrochemical flow half-cells as a replacement for the susceptible semipermeable separating membrane - from an electrolyte flow perspective. Fig. Figure 2 shows a redox flow battery arrangement according to the invention during the charging process. The components shown correspond to those of the Fig. 1. The only difference: Here, the novel electrolytic current key (5), which replaces the semipermeable separating membrane, is located, from a fluid dynamics perspective, upstream of the two flow-through half-cells. In a redox flow battery according to the invention, both current keys (upstream and downstream of the cell) are arranged in parallel, and the electrolyte flows are controlled by valves according to whether the cell is being discharged or charged.

[0028] The most significant advantages of a membrane-free redox flow battery design according to the invention only become apparent when the conventional mixing of the electrochemically reacted redox electrolytes in the cell with the remaining tank supply of charged electrolytes is eliminated. This includes a group of further dependent claims based on this principle, thereby realizing a single general inventive idea: the construction of significantly improved redox flow batteries with regard to manufacturing costs, reliability, voltage stability, and higher efficiency due to reduced pump energy losses.

[0029] The mixing of the two redox electrolytes contradicts a well-known misconception that catholyte and anolyte must be strictly separated to avoid undesirable "cross-over" effects. Initial attempts to construct membrane-free, experimental redox flow batteries, as described in the literature, avoid this undesirable mixing by establishing a very short, strictly co-laminar flow of both redox electrolytes in direct contact. This is only possible on a microscale, over a few millimeters or centimeters, and precludes the construction of high-performance redox flow batteries as well as recharging based on the same principle. However, the solution disclosed here according to the invention necessarily requires a nearly quantitative conversion of both redox systems during a single flow through the half-cell during the discharge process.This, in turn, can best be achieved in practice by the further addition of various auxiliary redox systems to the catholyte and anolyte, which preferably enable more quantitative recharging, and by means of a novel design (arrangement) of the electrochemical RFB flow half-cells. According to this invention, the discharged redox electrolytes combined in a novel current key are preferably not added back to the still-charged portion of the catholyte or anolyte supply, as is usual in conventional redox flow batteries, which systematically reduces the state of charge (SOC) of the circulating redox electrolytes, but are stored separately, according to a previously selectable distribution ratio. This results in further advantageous arrangements that lead to extremely stable cell voltages, since fully charged redox electrolytes are always pumped through the cells.This inventive elimination of continuous cycling of the redox electrolytes, which wastes large amounts of pumping energy in conventional redox flow batteries when the state of charge (SOC) is below 50%, enables further inventive measures that save a significant proportion of the energy required for pumping. This achieves a further objective: improved efficiency compared to the prior art.

[0030] The central solution to the problem of avoiding the expensive, fragile, and vulnerable semipermeable separating membrane in redox flow batteries with two flowing redox electrolytes is the replacement of this separating membrane between the catholyte and anolyte half-cells by special, novel electrolytic current keys, as disclosed in this invention. These keys essentially consist of the mixing zone of the catholyte and anolyte liquid flows reacted in the half-cells. Although redox flow batteries with relatively high conversion rates are already in use with a single flow of the redox electrolytes through the half-cells, a literature and patent search has so far yielded no information regarding the replacement of the semipermeable separating membrane by a significantly lower-resistance electrolytic current key formed by a forced, turbulent, and large-area mixing of both reacted redox electrolytes.

[0031] Redox flow battery devices, which also rely on a quantitative conversion of both electrochemically reacting redox electrolytes, are described in US patent application US 2011 / 0223450 A1 entitled "CASCADE REDOX FLOW BATTERY SYSTEMS," filed on January 7, 2011, and published on September 15, 2011, using a Cr-Fe redox flow battery as an example. While the trivial solution disclosed therein—a cascade-like series connection of several conventional redox flow batteries with a semipermeable membrane—does allow for complete discharge of both catholyte and anolyte fluid, the design complexity, pressure loss (losses due to increased pumping energy for the last battery in the series, which delivers hardly any power), and membrane costs are too high to be practical. The electrolyte transport revealed here, solely through gravitational forces acting on solutions, is also unrealistic.This patent application does not teach how to completely dispense with a separating membrane or how to achieve quantitative conversion ("fuel efficiency") of the reactants within a single electrochemical half-cell at acceptable flow rates and times. US 2011 / 0223450 A1 also describes a "tank separator," which can be a deformable bladder.

[0032] In DE 10 2013 225 159 B4, granted on February 25, 2016, cascade-like arrangements of conventional redox flow battery cells with a semipermeable membrane are also described, which are intended to enable a near-quantitative conversion and therefore allow the electrolytes to be collected in additional, separate tanks. However, the design of a novel current key that replaces the membrane and has other crucial advantageous properties is not disclosed.

[0033] The method disclosed herein and the exemplary devices described herein are possible with many redox system combinations, provided that all participating redox systems react with each other in a conventional redox titration in homogeneous solution without precipitation or other side reactions. A particular advantage of this invention is that proven electrolyte compositions and electrode materials can be adopted when designing a novel, membrane-free redox flow battery according to the invention. Analogous conditions apply when using reversible, purely organic redox-active compounds or redox polymers as the electrolyte solution. The large number of redox system combinations makes it virtually impossible to explicitly and completely list all possible combinations in the claims.For experts, it is not difficult to develop alternative designs based on the approach revealed here, which does without a membrane.

[0034] The technical solution disclosed here to the aforementioned serious problems of redox flow batteries with separating membranes lies in several fundamental changes to the process (of generating electricity from chemical compounds) and to the general arrangement (device) of the flow electrodes of redox flow batteries. The most important further change concerns the electrochemical redox conversion of catholyte and anolyte at the collector electrodes in the two half-cells during a single flow through these half-cells. According to the invention, this conversion should be close to 100% (near-quantitative redox conversion = near-coulometric operation).Further improvements with the same technical goal, namely to significantly increase the performance and maintenance-free time of all redox flow batteries, concern the design of the electrochemical flow half-cells, more precisely the design of the crucial collector electrode surfaces, constructive details of the critical contact zone of both reacted redox electrolyte solutions (low-resistance current key), as well as sensor-controlled monitoring of the electrochemical reactions, which, through variable flow rates of catholyte and anolyte, is intended to guarantee a coulometric mode of operation in each half-cell.

[0035] The physicochemical basis of the invention originates from the field of electrochemical analysis: The fundamental electrochemical operation of a redox flow battery can formally be compared to a redox titration of the two reacting redox systems; the only difference being that in the battery, the chemical reaction (transfer of electrons) between the two redox partners does not take place in a homogeneous solution, but rather the electrons are transferred from the reducing redox system to the oxidizing redox system via two collector electrodes and an external circuit with loads. Fig. Figure 3 provides a simplified graphical representation of how the redox potential of two exemplary redox systems (Fe 2+ / Fe 3+ and Ce 3+ / Ce 4+) during the flow through the relevant half-cells between inlet and outlet, if a complete turnover takes place during this flow contact time. Fig. Figure 3A shows a real recorded titration curve (half-cell voltage plotted against the degree of conversion) of a redox system using a reducing agent (Fe 2+ The reaction effect ranges from 0% conversion to 200% (2:1 excess of oxidant compared to the reactant). In the chosen example (iron(II) salt solution), the potential at the relevant half-cell electrode of a corresponding redox flow battery in the fully charged state (100% reduced form) is approximately +0.3 V relative to the normal hydrogen electrode (NHE). Fig. Figure 3B shows the same thing for the half-cell with the oxidizing redox system (Ce). 4+ ).

[0036] In the initial state (0% conversion with the reducing agent in the other half-cell = fully charged), a potential of approximately +1.6 V (NHE) is established at this electrode in the chosen example of a Ce(IV) salt solution. The effective voltage of this exemplary redox flow battery between the two collector electrodes in the half-cells is then approximately 1.3 V in the resting state (no current draw). The terminal voltage profile of the redox flow battery during discharge and charge can be easily determined from the difference in the individual electrode potentials at the respective conversion level. Such a voltage profile is typical for conventional redox flow batteries and is unavoidable because it obeys the Nernst equation.Due to the external current flow (current draw or connection of both electrodes via a load), in a redox flow battery, both individual electrode potentials act during the discharge cycle relative to the normal hydrogen electrode analogously to electrolysis at a constant potential in the two half-cells through which the different redox systems flow. In analytical chemistry, this is commonly referred to as internal electrolysis.

[0037] If a specific spatial liquid segment in the flow-through half-cell through which the redox electrolyte flows is observed under steady-state conditions over the course of the flow time, then the reducing agent in this segment ( Fig. 3A) is successively oxidized by electron release at the collector electrode, as in a direct titration with the oxidizing agent in homogeneous solution. At a conversion rate of 0.5 (SOC 50%), the condition for this redox system is: equal concentration or activity of the reduced and oxidized forms ([Red] = [Ox]). According to the Nernst equation, the so-called formal standard potential E0 of the redox system in question is established at this point at this electrode. These standard potentials are known and tabulated for many redox systems.

[0038] Fig. Figure 3B shows the analogous course of the half-cell potential of the oxidizing redox system compared to a standard hydrogen electrode (NHE) during a discharge (current supply) in which this redox system accepts electrons from the collector electrode. At a conversion level of 0.5 (SOC = 50%), equal amounts are present in the oxidized and reduced forms. When [Ox] = [Red], the corresponding standard potential E0 of the respective redox system is established at the electrode. For the solution to the problems revealed here, it is important to know that in an analytical-chemical titration, the other redox partner is always present in both [Ox] = [Red] situations. Therefore, at the SOC of the reducing redox system, equal amounts of Fe are present in this example. 2+ and Fe 3+ also equivalent quantities and Ce 3+an equivalent amount of Fe is produced side-by-side before and during the SOC by the oxidizing redox system. 3+ , which cannot be further oxidized even with a stronger oxidizing agent. Analogous considerations apply to the charging process. If a solution with equivalent amounts of Fe is used 3+ and Ce 3+ , which corresponds to discharged redox systems of an iron-cerium redox flow battery, is electrochemically reduced at one electrode, so only the Fe can 3+ are reduced. Conversely, if such a solution is electrochemically oxidized, only the Ce is reduced. 3+ to Ce 4+ oxidized because the Fe 3+Under these conditions (aqueous sulfuric acid solution), it cannot be further oxidized! These facts are important for the revealed solution to the problems of membrane redox flow batteries: The combination of both reacted (discharged) redox electrolytes after an almost quantitative conversion in the redox flow battery before recharging does not prevent subsequent recharging. The oxidation and reduction power (E0 values) are hardly affected by other redox-inert ions present, apart from changes in the activity coefficients. If such mixed redox electrolyte solutions are used from the outset, "cross-over" effects are negligible, with the exception of a slightly reduced Coulomb yield.

[0039] The arrows in Fig. Figure 3 illustrates the voltage curves of both half-cells against an NHE reference electrode during discharge and charge. The small, right-pointing arrows integrated into the lines indicate how the overall redox flow battery voltage gradually decreases during discharge until collapse at full conversion, while the larger, left-pointing arrows show the voltage increase during charging. These voltage changes during discharge and charge are inherent to the system and not caused by the isolation membrane.

[0040] The redox systems most commonly used in redox flow batteries are assumed to exhibit reversible electrochemical behavior, meaning they can be described by the Nernst equation. For example, with a 90% electrochemical conversion of the reducing agent at one electrode, the [Red] / [Ox] ratio is 1:10. This means that at 25 °C, and in the case of a one-electron transition, the half-cell potential is approximately 60 mV more positive than the standard potential. At a ratio of 1:100 (conversion rate 0.99), it is approximately 120 mV more positive, and at a conversion rate of 0.999, approximately 180 mV more positive. This means that these potential differences exist between the inlet and outlet of the electrochemical half-cell and can be used for control purposes. An analogous situation exists in the other half-cell with the oxidizing redox system. Here, the small black arrow points in Fig. 3B states that the half-cell potential relative to a standard hydrogen electrode becomes approximately 60 mV more negative at 90% conversion and approximately 120 mV more negative at 99% conversion in this example of a one-electron transition. Therefore, if one were to aim for 99% conversion in both cases with a single flow through the half-cells, the redox flow battery voltage at the output of both half-cells would be approximately 0.24 V lower than at the input under these conditions. With a two-electron transition of the anolyte and catholyte redox systems, this drop is halved. Due to the generally good electronic conductivity of the electrode materials, these different redox potentials of the respective redox electrolyte between the inlet and outlet of the electrochemical flow cells only manifest as locally differing overvoltages, which have also been proven to lead to acceptable inhomogeneous current density distributions in RFB applications in practice.

[0041] This also applies to the total electrochemical conversion when – as with conventional redox flow batteries – the entire volumes of both redox electrolytes in the external reservoirs are pumped through the half-cells. Here, too, the terminal voltage of a redox flow battery changes by approximately 240 / n mV (where n = number of electrons exchanged) between fully charged and nearly discharged, according to the Nernst equation. This voltage drop is unavoidable in current redox flow batteries with continuous circulation and complete conversion. In practice, however, such batteries are not operated until completely discharged because, at excessively low active redox system concentrations, an undesirable concentration polarization occurs that increases with the current load, further reducing the battery terminal voltage. Additionally, pumping already highly converted redox systems leads to undesirable pumping energy losses.

[0042] The individual electrode potential relative to a reference electrode at the end, just before exiting the flow-through half-cell, will correspond to that at the inlet, except for a very small ohmic drop along the electron-conducting electrode. Here, according to the laws governing mixed potential formation, the electrode zone with the highest exchange current density determines the currently established electrode potential. This zone is located at the inlet of the electrochemical half-cells due to the higher depolarizer concentrations. The outlined redox potential changes at the collector electrodes during flow through the half-cells are not dependent on the initial redox system concentrations but only on the ratio [Red] to [Ox].To prevent excessive changes in the redox potential of the two systems between initial contact with the collector electrodes and exiting the flow cell, and to avoid excessive concentration polarization, a conversion rate well above approximately 95-99% is less advantageous. While a slight loss in electrochemical yield does occur in RFBs without a separating membrane and with direct contact of the redox electrolytes, which are converted to over 90% in the flow half-cells, this is more than compensated for by other advantages.

[0043] Extensive experience has been gained in advance regarding an optimal flow rate for achieving a quantitative conversion with a single cell flow. The invention discloses arrangements and designs of collector electrodes that reduce concentration polarization phenomena at the end of the flow paths. The concentration changes of the redox systems involved along the electrode surfaces during flow through the two electrochemical half-cells also lead to corresponding changes in the local electrolysis current of a considered liquid segment along the contacted electrode surface. The temporal sequence of the electrochemical reactions at the current collectors in the anolyte and catholyte compartments when the redox electrolyte flow is stopped or within a considered flow segment with the respective redox systems corresponds to that of a so-called..."Internal electrolysis" or coulometry at constant potential. It was found that with a constant redox electrolyte volume under consideration and a constant electrode potential of a working electrode relative to a reference electrode, and with a constant relative motion of the electrochemically reacted substance relative to a working electrode on whose surface the oxidation or reduction of the redox system takes place, the current I is given by: I(t)=l0 e−kt The constant k decreases exponentially with time (here, the flow path along the electrode surface). To enable quantitative conversion in the shortest possible time, the constant k must be made as large as possible. The cell constant k is determined according to k=15.8 DA / δV with: D = diffusion coefficient A = Electrode surface V = Volume of the measuring solution δ = thickness of the Nernst diffusion layer

[0044] Since the diffusion coefficient can only be increased within a small range (e.g., by heating), this electroanalytical method aims to maximize the cell constant k by making the electrochemically active electrode surface A as large as possible and keeping the electrolyte volume V in contact with it as small as possible. The thickness of the Nernst diffusion layer δ, which is crucial for concentration polarization, should also be kept as small as possible. It is reduced when the electrolyte flow along the electrode surface is increased. According to the invention, these relationships dictate that the half-cells should be designed with shallow electrolyte flow channels with a large ratio of electrode surface area to contacted electrolyte volume.Based on extensive measurements, the following approximately valid mathematical relationship exists between the height H of the flow channel and the flow velocity F of the electrolyte solution to be electrochemically converted: δ≈3HF

[0045] If only a specific segment of the catholyte or anolyte solutions flowing through the electrochemical half-cells of a redox flow battery (or the current profile at a flow pause with a completely filled half-cell volume) is considered in isolation, the inventor was able to observe an analogous exponential decline due to the increasing depletion of the electrochemically reacted components over time. To enable the electrochemical reaction at the collector electrodes to be as quantitative and rapid as possible, the invention adheres to the formulas (3) to (5) above, which allows for a wide variety of design possibilities. US Pat. No. US 6,686,082 B2 and US Patent Application Publication No.US 2006 / 0234107A1 describes a fuel cell in which the reactants flow in progressively narrower channels to increase the flow rate; however, it does not describe a simultaneous and significantly more effective increase in the active collector electrode surface area for the purpose of quantitative conversion within a short contact time. Nor does it mention the elimination of a semipermeable separating membrane or the advantages of separately collecting both reacted electrolytes, such as a more stable clamping voltage and reduced energy consumption for pumping the redox electrolytes.

[0046] The inventive method consists of enabling the electrochemical reaction of catholyte and anolyte in a redox flow battery to proceed almost quantitatively even at high power output (high current draw). According to the invention, the electrochemical flow cell is designed such that, firstly, the electrochemically active surface area increases in the flow direction, while, secondly, the flow channel height is reduced disproportionately to the cross-sectional change. This design feature results in a corresponding reduction of the contact volume and a corresponding increase in the relative flow velocity. The latter, in turn, further reduces the thickness of the Nernst diffusion layer, all of which leads to an increase in the cell constant k in the flow direction. Ideally, this would perfectly compensate for the exponential current decay.This counteracts concentration polarization towards the end of the contact area between the redox electrolytes and the electrode. According to the invention, this can be achieved through various suitable arrangements and different shapes of the electrochemical flow cells of the redox flow battery, which cannot all be listed exhaustively here. New findings in microfluidics and nanotechnology for surface area enlargement can also be advantageously utilized. However, to achieve high performance, extensive parallelization is necessary.

[0047] Increasing the surface area of ​​the collector electrodes in the half-cells of redox flow batteries can be achieved in a variety of ways. For example, the redox electrolyte can be allowed to flow past the collector electrodes, and the electrode surface area can be increased by means of suitable spacers between two electrode surfaces. These spacers, cut to define the flow channel layer thickness and width, increase the latter in the direction of flow, thus counteracting the exponential current drop in the liquid segment under consideration. A spacer thickness that decreases in parallel with the flow direction also reduces the volume segment of the redox electrolyte solution in the flow channel and increases the flow velocity towards the end of the flow path. This reduces the Nernst diffusion layer and correspondingly increases the cell constant.This inventive method allows the electrochemical conversion of the respective redox electrolytes used to be maximized within a predetermined contact time (flow time) of the redox systems with the electrodes. A redox potential measurement at the inlet and outlet of the two flow-through half-cells then allows, according to the relationships described above with the Nernst equation, individual, automatic control of the flow rate of each half-cell with the aim of achieving a nearly quantitative conversion. The signal transmitters are two platinum single-rod electrodes at the inlet and outlet, and suitable flow controllers or valves perform the control.

[0048] The revealed solution to the problem of realizing high-performance redox flow batteries without a vulnerable separating membrane can easily inspire experts to develop other embodiments, in which the fundamental basic requirements – electrode surface area enlargement, narrowing of the effective flow channels and increase of the relative flow along the flow path, as well as corresponding low-resistance current key designs – can be realized.

[0049] In the disclosed application of the aforementioned design measures to redox flow battery half-cells, the disclosed method, with this specific design of the flow channels, prevents undesirable excessive concentration polarization with a corresponding reduction in the battery terminal voltage and allows quantitative conversions within short flow times. It is advantageous in a novel membrane-free redox flow battery according to the invention if the recharging of the redox electrolytes also proceeds quantitatively. Fig. Figure 3, with its thicker arrows pointing to the left, illustrates the terminal voltage curve of a redox flow battery with the example redox systems during charging. It can be seen that the terminal voltage is the difference between the two individual potentials as the charge approaches 100% (in the Fig. 3 (near zero conversion) the voltage rises rapidly and exponentially, leading to undesirable electrolysis of the solvent (usually water). The subsequent evolution of hydrogen or oxygen must be prevented for various reasons. Firstly, gas bubbles in the catholyte or anolyte compartment disrupt mass transfer; secondly, the electrode materials can be chemically attacked. To avoid entering this voltage range for safety reasons and to prevent the crucial concentration polarization (only a small amount can be reduced or oxidized in the electrolyte flow, requiring disproportionately high pumping energy), a fully quantitative charging process is often avoided. However, to achieve a more quantitative charging of the main redox systems in a redox flow battery according to the invention, further chemically compatible, so-called "auxiliary redox systems" can be added to the catholyte and anolyte.

[0050] To facilitate the crucial quantitative charging of the catholyte and anolyte in a single flow through the redox flow battery half-cells of a novel redox flow battery disclosed herein, without side reactions (e.g., gas evolution during charging), additional auxiliary redox systems are added to both main redox electrolytes. For example, if a further redox system with a more negative formal redox potential E0 is added to the reducing electrolyte and one with a more positive potential E0 is added to the oxidizing electrolyte, then towards the end of the charging cycle these two auxiliary redox systems preferably react at the two collector electrodes (functioning as working electrodes during the charging process), whereby the respective redox potentials of these auxiliary redox systems are established at the latter relative to a reference electrode, neglecting the transfer and concentration overpotential.For the terminal voltage (the difference between the two individual electrode potentials) of the redox flow battery in question, this means a corresponding increase during charging until these auxiliary redox systems are quantitatively converted. This manifests as a voltage step in the charging characteristic, the magnitude and duration of which depend on the concentration of the auxiliary redox systems used. Only then does the terminal voltage rise exponentially due to electrolysis of the solvent (usually water), and the charging of the redox flow battery must be stopped. This voltage step towards the end of the charging process can be used for control purposes, for example, to reduce the charging current and thus prevent disruptive gas evolution at the electrodes.

[0051] If the terminal voltage of a redox flow battery with suitable auxiliary redox systems increases towards the end of the charging process, and if the auxiliary redox systems are also reduced or oxidized in the process, they react in this form in a homogeneous solution with any remaining unreacted amounts of the main redox system, thus increasing its conversion rate. Depending on the amount of auxiliary redox systems added, they also contribute to an increase in energy density during discharge, representing a further improvement over conventional designs and redox system combinations. If more than one auxiliary redox system is added to the catholyte and anolyte, a redox cascade is approached – analogous to the biochemical respiratory chain. It is also possible to use higher concentrations of auxiliary redox systems.Then, an auxiliary redox system can become a primary system, contributing significantly to a higher power density. This counteracts the limited solubility of the selected redox compounds. The result is more or less pronounced steps in the charging and discharging characteristics, which superimpose the continuous Nernst characteristic.

[0052] The present invention of membrane-free redox flow batteries dispenses with the known principle of a difficult-to-maintain co-laminar flow or unstable phase boundary between two immiscible liquids in the contact area of ​​catholyte and anolyte, and instead achieves the necessary electrolytic contact by means of a forced mixing of both solutions in the form of a novel, particularly low-resistance current-key design. The present invention also describes a novel, advantageous design of the electrochemical flow cells, which are the basis of all redox flow batteries, in order to achieve the most quantitative conversion possible in a single flow.The present invention contradicts a pre-existing misconception that the elimination of separating membranes is only possible in miniaturized, low-power redox flow batteries using small, immiscible co-laminar redox electrolyte currents over a length of a few centimeters. Contrary to existing misconceptions, the present invention requires even a partial or complete mixing of both redox systems after a single pass through the respective half-cells—albeit after a nearly complete electrochemical conversion with current generation in the respective redox flow battery. Subsequent partial or complete collection in mixing vessels allows this method according to the invention to separate the combined redox electrolytes into catholyte and anolyte components for regeneration.According to the invention, this portion can then be returned to the corresponding storage tanks, with or without mixing with the portions that have not yet been discharged. By foregoing the mixing with redox electrolytes that have not yet flowed through the cells (recirculation principle), which is only practical under these conditions (quantitative conversion), the invention opens up the possibility of further significant improvements: Firstly, only fully charged redox electrolytes then flow through the two half-cells of the redox flow battery, which means a constant battery clamping voltage until the supply of charged redox electrolytes is exhausted.Secondly, this results in the further possibility, according to the invention, of collecting the discharged redox electrolytes after they have passed through the cell in the same reservoir – but with material and electrical separation from the still-charged portion after appropriate distribution in the mixer according to the novel current key. In contrast to US 2011 / 0223450 A1, however, the invention does not use a passively easily movable separating membrane between the still-charged and the converted redox electrolyte, but rather a fixed piston that can be moved externally by means of a piston rod. The latter also enables the electrolytes to be pumped when the push rod is moved by other mechanical drive techniques in the sense of a double-acting piston pump.When using gravitational energy or pneumatic energy from the already necessary 200-bar inert gas cylinders for oxygen removal, particularly low losses result due to the low pumping energy. The design conditions revealed here for a near-quantitative conversion with a relatively short flow time also apply to the charging cycle.

[0053] The significantly modified arrangement of components and the design of redox flow batteries revealed here allows even known asymmetric disturbances or different Coulom's efficiencies during discharge and charge in the respective half-cells to be automatically corrected by a corresponding, suitable, empirically determined distribution ratio of the discharged redox electrolyte solutions without requiring maintenance effort and downtime due to re-preparing the relevant redox electrolytes.

[0054] The present invention also describes a solution to prevent the disadvantage of the disclosed preferred flow channel design—namely, higher hydraulic resistance compared to conventional redox flow batteries with, for example, graphite felt electrodes—from reducing efficiency. This is achieved by directly utilizing the advantage of having almost quantitatively converted redox electrolytes still under pressure downstream of the circuit breaker in the redox flow battery. This is done by employing the principle of a double-acting piston pump to reduce energy consumption. By preventing mixing with the still-charged redox electrolyte volume in the storage tanks, which are designed as large piston pumps, constant redox system concentrations flow through the two redox flow battery half-cells until the charged supply is completely consumed. This results in a constant terminal voltage during the discharge process.This avoids another well-known disadvantage of typical previous redox flow battery designs. The advantages are most clearly described as reduced voltage drop under load, a more constant voltage during discharge, increased efficiency, and improved reliability. Fig. Figure 4 illustrates these advantages with an example of a typical charging and discharging characteristic of a redox flow battery: (single cell voltage versus current drawn / cm²) 2plotted); 1 = Course of the battery terminal voltage during charging with increasing current density (higher galvanostatic charging current); 2 = Open-circuit voltage (Nernst voltage at the terminals of a single cell without current flow); 3 = Course of the battery terminal voltage during discharging with different current densities (solid line = conventional redox flow battery; dashed line = redox flow battery without a high-resistance membrane and reduced concentration polarization due to optimal flow channel design according to the invention); A = Range of cross-polarization (kinetic inhibition, dependent on the exchange current density of the respective redox system); B = ixR ohmic voltage drop due to internal resistances in the redox flow battery cell, mainly caused by the separating membrane; C = Concentration polarization (diffusion of the electroactive components to the electrode surface and back is limiting).

[0055] To illustrate the general applicability of this invention to other redox flow battery types besides VRFB, only one example will be given: When combining redox systems other than those with different vanadium oxidation states, e.g., Cr 2+ , Ti 3+ , Fe 2+ or similar in combination with Ce 4+ During the charging process of the combined, discharged redox systems, one partner redox system passes through the respective half-cell unchanged because it cannot be further reduced or oxidized. For example, in a redox flow battery with a Ti 3+ - Ce 4+ Combination of the combined, discharged redox electrolyte solution with Ti 4+ and Ce 3+ - Ions only the Ti during the charging process 4+ back to the Ti 3+ be reduced because Ce 3+ which cannot be reduced further, on the other hand, only the CE can be used during the charging process. 3+ back to the Ce 4+be oxidized; Ti 4+ It cannot be further oxidized in this process. Electrochemically, the redox species, which remain unchanged during charging, do not directly interfere, but only through corresponding, small effects on the activity coefficient.

[0056] According to the invention, a quantitative operating mode (coulometric flow cells on the catholyte and anolyte sides) is monitored by easily performed redox potential measurements at the inlet and outlet of a half-cell. For a single-electron transition, for example, a potential difference of approximately 120 mV in the measured redox potential indicates an approximately 99% conversion rate. For multi-electron transitions, 120 mV / n (where n is the number of electrons exchanged per formal conversion) means the same. Extensive measurement series have shown that, due to the other advantages of a membrane-free redox flow battery, a conversion rate of over 95% (or a change of approximately 100 mV / n) can also be targeted in practice. To achieve these values, the flow rates of the catholyte and anolyte are adjusted accordingly until this difference in redox potentials between the inlet and outlet of the respective half-cell occurs.Since this is done separately for catholyte and anolyte flow, the different electrochemistry of both redox partners at the electrode surfaces can be corrected better and continuously according to the invention than in conventional redox flow batteries of traditional design.

[0057] Small platinum single-rod electrodes (with integrated reference electrode) that protrude into the electrolyte flow but do not touch the collector electrodes can be used for redox potential measurement. At optimally adjusted flow rates, they will produce potential profiles similar to those shown in Fig. Figure 3 shows the curves. Using suitable software, the relevant curves can also be displayed graphically, and the flow rates can be automatically controlled individually for each half-cell using a computer. According to the invention, such sensor-based monitoring of the electrochemical reactions in both half-cells allows for the easy and automatic compensation of asymmetric changes in the redox electrolytes (e.g., a reduction in the redox system concentration due to undesired side reactions – oxidation of the reducing agent by atmospheric oxygen), which benefits the long-term stability of the redox flow batteries.

[0058] The elimination of a separating membrane in redox flow batteries and the turbulent mixing of the two redox electrolytes reacted in the respective half-cells, as described in the invention, necessarily leads to a further challenge: the optimal design of this contact zone in the sense of a novel electrolytic current key with high electrical conductivity. Alternative current key designs also allow for advantageous alternative shapes of the two half-cells, particularly with regard to the disclosed optimal design of the flow channels. In the inventive arrangements of the two membrane-less half-cells, the distance between the respective collector electrode ends in the catholyte and anolyte compartments should be kept as short as structurally possible, and the cross-section of the contact area of ​​the two discharged redox electrolytes should be made as large as possible.If necessary, the collector electrode ends (at the output of the respective half-cell) are extended with an electronic conductor made of the same or a different chemically inert material, and these two electronic conductors are brought together as close as possible (without short-circuiting) in a catholyte and anolyte mixing zone. Increasing the contact area of ​​the two reacted electrolyte solutions during mixing further reduces the electrolytic resistance between the two half-cells (internal resistance). According to the invention, preferably inert, highly permeable meshes (coarse mesh) made of the relevant electronic conductors are used, which are held at a minimal distance by spacers. Short-circuiting must be avoided, and easy and rapid removal of the combined, discharged redox electrolytes must be ensured.In this arrangement according to the invention, the combined redox electrolytes can be further treated differently, which offers additional advantages over the conventional redox flow battery design. The use of the electrolyte in which at least both mutually reactive redox systems are dissolved in discharged form, according to the invention, offers further advantages as a current key: higher electrolytic conductivity, which correspondingly reduces the internal resistance of a cell. Due to the simultaneous presence of redox systems with a high charge number z, the conductivity κ is determined according to the known dependence of the latter on the concentration c0 and the ion mobility of cations u. + and anions increase proportionally to the charge number. κ=z×c0×F(u++u−)

[0059] In addition, the catholyte and anolyte also contain readily soluble, but electrochemically inactive at the electrodes, completely dissociated salts (e.g., sodium or potassium sulfate); due to their good solubility, these are usually present in higher concentrations. Since highly acidic redox electrolyte solutions are frequently used in redox flow batteries, the particularly high ion mobility of the protons is especially advantageous here. Ions with a high charge number can approach this proton mobility. This is a further advantage of the electrolyte composition according to the invention with fully reacted redox partners. The electrolytic resistance between the two redox flow half-cells is further reduced by the process disclosed here when the ion mobility is increased by the faster electrolyte flow at the end of the half-cell.This is also achieved by the fact that the reacted (fully discharged) catholyte and anolyte liquids meet in a kind of "jet stream" after passing through the electronic conductor.

[0060] In a preferred arrangement, possible only under the disclosed conditions of quantitative conversion with a single flow through the half-cells, the anolyte and catholyte fluid flows are combined in an external mixer downstream of the electrochemical cell with the collector electrodes (after flowing through the half-cells) during discharge. A condition is that no further undesired reactions (e.g., precipitation) occur between the two discharged redox systems, which is the case with most redox system combinations used in redox flow batteries. If the entire quantity of catholyte and anolyte solution is collected via a mixer, a sufficiently sized third reservoir is necessary.This approach prevents the fully charged redox electrolyte solutions in the storage tanks from being diluted by the discharged electrolytes, which would occur with conventional redox flow battery circulation. The disclosed solution thus avoids a terminal voltage drop according to the Nernst equation (change in the [Ox] / [Red] ratio due to continuous dilution) and due to concentration polarizations during discharge. This results in the solution described in [reference]. Fig. Figure 4 shows the discharge characteristic at increasing current densities, characterized by a dashed line. This third tank, which may have twice the capacity of the catholyte and anolyte storage tanks, can be eliminated by means of an alternative, pump-energy-saving device for conveying catholyte and anolyte, according to a further procedure disclosed herein. Example 1

[0061] That electrochemical flow cells, or current-supplying galvanic cells based on redox potential, can be constructed without a separating membrane or diaphragm and without any loss of efficiency was demonstrated by a method developed in Fig. Figure 5 schematically illustrates an exactly coulometrically operating electrochemical arrangement with a circulating electrolyte. The aim was to demonstrate an exact coulometrically operating, membrane-free method as simply as possible as a "proof of principle," without having to construct a complete redox flow battery with potential side reactions or other losses. By applying a voltage between a generator electrode (1) (platinum wire) and a second electrode (3) (activated carbon slurry with Pt network lead) with the positive terminal connected to electrode (1), elemental iodine was generated from a circulating electrolyte containing iodide. This elemental iodine dissolves completely in the circulating electrolyte. Only the resistor R2 (to simulate a load) was connected between electrode (2) (graphite threads as collector electrode) and the activated carbon slurry (3).Electrode (2) was thus negatively polarized via the bipolar electrode (3), and with a sufficiently large surface area and a moderate flow of the electrolyte, the iodine produced at electrode (1) could be quantitatively (coulometrically) reduced back to iodide. The current between the collector electrode (2) and electrode (3) flowed spontaneously through the resistor R2, acting as a galvanic cell. The crucial aspect of this experimental setup, however, was the simple proof of a 100% coulometric process: the galvanically generated current flow between the collector electrode (2) and electrode (3) corresponded exactly to the electrolysis current between electrode (1) and electrode (3) that produced free iodine.This provided convincing proof that, with optimized flow conditions and electrode surfaces, the absence of membranes or diaphragms between the individual electrochemical half-cells does not affect the current output, and simultaneously allows for the construction of a novel redox flow battery with only one circulating redox system. The electrolyte containing elemental iodine can, for example, be stored temporarily. In principle, this design corresponded to a zinc-ceramic battery. 4+A redox flow battery, with the activated carbon surface acting as the stationary capacity-limiting factor. The electrode reaction can be considered a transition of oxygen atoms from the water to the surface of the activated carbon, releasing electrons. The oxygen is bound there in the form of a quinone-like chemical sorbate. Even in small quantities, carbon nanotubes (CNTs) can represent a vast surface area on which the apparently reversible electrochemical reactions take place.

[0062] Fig. Figure 6 shows details of the experimental setup without a semipermeable membrane. The electrolyte consists of 3 M NaBr, 0.005 M Nal, and 0.1 M NaH₂PO₄ / Na₂HPO₄. This electrolyte is circulated past electrodes (1) and (2) by an airflow introduced at point 4. The rising air bubbles carry the iodine generated at electrode (1) to the large surface area of ​​the collector electrode (2). Electrode (2) was optionally made of graphite fabric or bundled graphite threads. The galvanic half-cell with the collector electrode (2) was electrolytically connected at the bottom to the half-cell of electrode (1) and, via a current key without a membrane or diaphragm, also to the bipolar electrode (3). Since no net reaction takes place in such an experimental setup, this coulometric experiment can be run for months without any drop in yield.To prove that the free iodine produced at electrode (1) can actually be measured with the galvanic cell with the collector electrode (2) according to Faraday's laws, a small, known amount of SO2 was added to the airflow used to circulate the electrolyte.

[0063] On its way to the collector electrode (2), this reacts with the iodine, is itself oxidized, and reduces the iodine back to iodide. Fig. Figure 7 shows a corresponding recorder log that shows the voltage drop across resistor R2 ( Fig. 6) records. The current yields achieved are plotted on this original diagram and provide convincing proof of a coulometric method of operation without the need for membranes or current keys with diaphragms. Measurements of the redox potential between point A and B ( Fig. 6) yielded a difference of approximately 120 mV, which, according to the Nernst equation, indicates a quantitative conversion of iodine to iodide. If the conversion at the collector electrode (2) were not quantitative, iodine would have been converted via the electrolytic connection (current key) with electrode (3), which would have reduced the Coulombic efficiency of the galvanic cell (between electrodes (2) and (3)). By simply checking whether the electrolysis current during iodine production was equal to that of the galvanic cell, the flow rate of the base electrolyte was previously adjusted to a suitable value. Example 2

[0064] As an example of redox flow batteries not based on vanadium compounds, a titanium-cerium-based battery is presented here. In this example, equimolar solutions of titanium(III) sulfate (reducing agent) and cerium(IV) sulfate (oxidizing agent) are used as the redox systems. The concentration is set as high as possible. The requirement is that all oxidation states remain in solution at this concentration. Methanesulfonic acid exhibits better solubility than sulfuric acid solutions and is therefore used. To increase the base conductivity of the electrolyte, neutral salts (e.g., sodium or potassium sulfate) are added to the solution at the maximum possible concentration.

[0065] The corresponding redox equations and the standard redox potential are shown in equations (7) and (8): Ti(CH3SO3)3 + +H2O → TiO(CH3SO3)2 + CH3HSO3 + H + + e - E0=+0.10 V 2 Ce(CH3SO3)4 + 2H ++ 2e - → 2 Ce(CH3SO3), + 2CH3HSO3 E0=1.71 V

[0066] When equimolar solutions of both redox systems are passed through a membraneless, nearly coulometrically operating electrochemical cell according to the invention, a voltage of approximately 1.6 V is established between the two electrodes of a single cell. The redox electrolytes are initially prepared as if a completely discharged mixture of catholyte and anolyte were present, e.g.:

[0067] 1100 g of Ce₂(CO₃)₂ × 5 H₂O are stirred into 1.5 L of water, and 2498 g of methanesulfonic acid are slowly added and made up to 4 L with water. The solution is then heated to 55°C, and 958 g of titanyl sulfate (TiOSO₄) are added while stirring vigorously. To increase the electrolytic conductivity, 200 g of potassium sulfate are added to the finished solution. Such a solution corresponds to a discharged mixture of redox electrolytes in a titanium-cerium redox flow battery (Ti₂O₆). 4+ and Ce 3+ The latter, equipped with semipermeable membranes, has been described extensively in the literature. This solution was divided 1:1 into catholyte and anolyte components and passed through the respective flow-through half-cells. In this case, platinized titanium was used as the anode and graphite felt as the cathode in a cell with a current key according to Fig. 10 instead of a membrane. A current was galvanostatically applied to this cell, ensuring that the resulting voltage did not exceed 1.5 V. The flow rate was adjusted so that the redox sensors at the inlet and outlet of the flow half-cell showed a difference of approximately 110 mV. The separately collected catholyte and anolyte solutions were analytically tested for their degree of conversion. The Ti 4+ -ions were over 90% concentrated in Ti under these conditions. 3+ -ions converted; the Ce 3+ -ions were converted to Ce at a rate of 99%. 4+ -ions converted, both of which correspond to a recharging of the redox electrolyte solutions combined according to the invention by means of the special current key. The lower yield at Ti 4+ - The reduction was caused by traces of dissolved oxygen that had not been removed.

[0068] Through electrochemical reactions at the collector electrodes in the flow cells, the Ti(III) compound is oxidized to Ti(IV) compound and the Ce(IV) compound is reduced to Ce(III) compound. In the nearly coulometric (almost quantitative redox conversion at the electrodes) operation according to the invention, at the corresponding half-cell outlets, only tetravalent titanium compounds or trivalent cerium compounds are present, with only a negligible residue remaining. When combined in a common solution, these compounds no longer react redoxically with each other. When the two reacted redox electrolyte streams are combined, this solution exhibits a measurable redox potential corresponding to the equivalence point of a corresponding potentiometrically indicated analytical-chemical titration curve, although this can vary slightly due to residual traces of unreacted compounds.After mixing the two discharged redox electrolytes, they are divided again 1:1 (or empirically determined, given different electrochemical efficiencies) and returned to their storage tanks in the advantageous form of a double-acting piston pump. Example 3

[0069] The inventive arrangement of a membrane-free redox flow battery functions particularly efficiently because it is more quantitative, using catholytes and anolytes containing more than one redox system. In this example, without claiming practicality (only to illustrate the principle of working with more than one redox system per catholyte and anolyte), a titanium / iron redox flow battery is to be constructed, which additionally contains the Cr(II) / Cr(III) redox system on the reducing electrolyte side alongside Ti(III) / Ti(IV) as the main component, and Ce(III) / Ce(IV) alongside Fe(II) / Fe(III) as the main component on the oxidizing redox electrolyte side. The titanium and iron compounds are present in equimolar compounds at their maximum soluble concentration. In order for the two additional redox systems Cr(II) / Cr(III) and Ce(III) / Ce(IV) to have their effect, they should be present in a lower concentration than the main redox systems.They are auxiliary reagents used to prevent gas evolution at the electrodes towards the end of the charging process. This is intended to make the regeneration of the relevant main redox electrolytes more quantitative.

[0070] The redox systems used in this example show, according to the simplified equations (9) - (12): Cr 2+ ⇄ Cr 3+ + e - E0=−0.41 V TiO 2+ +2H + + e - ⇄ Ti 3+ + H2O E0=+0.10 V Fe 3+ + e - ⇄ Fe 2+ E0=+0.77 V Ce 4+ + e - ⇄ Ce 3+ E0=+1.61 V

[0071] Standard redox potentials E0 range between -0.41 V and +1.61 V (NHE), meaning that if the iron-titanium system is potential-determining due to the higher concentration of its ions in the main redox electrolyte, a redox flow battery based on this will show a single-cell voltage of approximately 0.67 V (0.77 - 0.10).

[0072] During the discharge process, Ti flows on the reducing side of the redox electrolyte. 3+ - and Cr 2+ -The compounds pass together at the collector electrode in the relevant half-cell and are converted to Ti by the release of electrons at this electrode. 4+ - and Cr 3+ - Compounds are oxidized. On the oxidizing side, the existing Fe are oxidized. 3+ - and Ce 4+ -compounds formed by the uptake of electrons from the collector electrode to Fe 2+ - and Ce 3+-compounds reduced. At the output of the electrochemical flow cell (or cell stacks), with a quantitative electrochemical conversion, there are therefore Ti on one side. 4+ - and Cr 3+ -connections and on the other side Fe 2+ - and Ce 3+ - Connections are present side by side. When both discharged redox electrolytes are combined, they no longer react with each other redoxically. Therefore, according to the invention, they can be combined without a separating membrane or diaphragm, thereby forming an advantageous, particularly low-resistance current key. After the mixing of both redox electrolyte currents, which have undergone almost quantitative electrochemical conversion, following their passage through the electrochemical cell, they are again divided 1:1 (or empirically depending on the electrochemical efficiency) and separately routed through the half-cells of the redox flow battery for recharging, after passing through the input current key and divider.

[0073] In this process, a solution flows which contains Ti 4+ -, Cr 3+ -, Fe 2+ - and Ce 3+ -compounds are contained through the two half-cells for recharging. On the cathode side, under these conditions, acidic aqueous electrolyte, only Ti can be found. 4+ and Cr 3+ reduce back to the original product. The Fe 2+ - and Ce 3+ - Compounds do not interfere with the electrochemistry, as shown in Example 8. Similarly, on the anode side (electrode withdrawing electrons), only Fe is removed from all compounds. 2+ - and Ce 3+ -compounds are oxidized back into their original forms.

[0074] Afterwards, the separately collected redox electrolytes are recharged and available for another discharge cycle.

[0075] These additional redox systems can also serve to indicate the near end of the charging process through characteristic voltage changes. When the main redox electrolytes in the electrochemical cells of the redox flow battery can no longer be electrochemically converted, the auxiliary redox systems become the potential-determining factor, and the terminal voltage, neglecting other voltage losses (e.g., internal resistance, contact resistances, etc.), shows a small step before electrolyte decomposition. If the redox flow battery cell voltage continues to rise during charging, there is a risk of hydrogen or oxygen formation, which reduces the yield and also causes flow-related disadvantages (blocked electrode surfaces). As a side effect, redox reactions in the electrolyte solution also further reduce the remnants of unreacted main redox systems, making them even more reductive.The oxidative auxiliary redox systems are completely converted.

[0076] The previously described possibility of indirectly converting the main redox systems (due to their high concentration) electrochemically by means of auxiliary redox systems (electron transfer in solution) can, according to the invention, be further used to employ redox systems whose electron transfer is kinetically inhibited (high overpotential) at certain electrode materials, provided that a suitable auxiliary redox system with a correspondingly more negative or positive formal potential exhibits a higher exchange current density at this electrode material. According to this principle, even small amounts of added auxiliary redox systems can exhibit "catalytic effects" by acting as mediators.This allows many readily soluble redox systems with poor electrode kinetics to be used in redox flow batteries by adding small amounts of these redox systems with high exchange current densities, without generating a detrimentally large crossover voltage. It is impossible to list all possible combinations of redox systems and electrode materials here. Experts can easily select suitable redox system combinations that meet these requirements based on measurements of the exchange current density at the relevant electrode materials. Example 4

[0077] In an advantageous embodiment of the disclosed novel redox flow battery design, tubular half-cells are used, which are easier to seal than rectangular cell stacks. The required increase in electrode surface area, along with narrowing flow channels for catholyte and anolyte, is achieved by filling these cylinders with dense packings of electronically conductive filler material (e.g., spheres), the sphere diameters of which decrease continuously or section by section in the direction of flow of the redox electrolytes. Fig. Figure 8 shows an exemplary simplified scheme for the discharge process of a redox flow battery with tubular half-cells filled with electronically conductive packing material; 1 = packing material (spheres, etc.) with an electronically conductive surface; 2 = current key – not shown in detail here – and mixer / divider for transferring the discharged redox electrolytes either to a third storage tank or, after an empirically determined division, back to the respective storage tanks for mixing with the still-charged residue or, preferably, back into a novel area separated from the charged redox electrolyte fraction by means of an externally movable piston (see Example 5); 3a = input of charged anolyte electrolyte (to be quantitatively reduced in the column); 3b = completely electrochemically converted anolyte; 4a and 4b = analogous to 3a and 3b for the catholyte flow through the half-cell during the discharge process (current supply).

[0078] Fig. 9 shows the arrangement of Fig. 8 during the charging process. The discharged redox electrolytes are not directly fed back into the same half-cell inputs as during discharge, but are first separated using a second current key and divider. If they were temporarily stored in a common reservoir, they are separated there into catholyte and anolyte fractions according to their stoichiometry or "cell imbalance" and passed through the respective half-cells for regeneration, with a suitable charging voltage applied externally to the electrodes (here, the electronically conductive filling). The necessary, purely electrolytic connection between the two half-cells is established by a special current key, which also determines the exact ratio and is located upstream of the half-cells. The drain electrodes are in Fig. 9 (not shown here for clarity); 1 = packing material (spheres, etc.) with an electronically conductive surface; 2 = current key and divider; 3a = inlet of discharged anolyte electrolyte from the respective reservoir; 3b = completely electrochemically reduced anolyte; 4a and 4b = analogous to 3a and 3b for catholyte regeneration. Sufficient flow of the redox electrolytes to be converted prevents back-diffusion of the converted redox systems and an efficiency-reducing reaction in solution. Example 5

[0079] The basic structure of the novel current key construction according to the invention, or the construction of such an electrolyte bridge between the two half-cells, is described in Fig. 10 and Fig. 11 schematically simplified representation. Fig. Figure 11 shows, as an example, a possible application of the novel current key as a replacement for the semipermeable membrane in a conventional redox flow battery setup using cell stacks. In this case, it is advantageous to use only a portion of the redox electrolytes for the current key. For this specific application, hydraulic balancing must be performed, and the two electronic conductors (1, 1'), which are electrically connected to the respective ends of the collector electrodes via a conductor (2, 2') or in direct contact, must possess frit-like properties. Within these conductors, the redox electrolyte transport must proceed so slowly in certain sections that, upon mixing with the other redox electrolyte, they form a nearly discharged partial current. Such balancing can be easily achieved using a starch / iodide solution and an oxidizing agent.With the correct pressure setting, the latter must not pass through the frit electrode in a reduced quantity, in order to oxidize the iodide on the other side and thus indicate the iodine-starch reaction. With a small partial flow of the redox electrolytes, even conventional redox flow battery designs can tolerate the associated slight loss of efficiency if the other advantages of a membrane-free design outweigh it. However, the quantitative conversion of the redox electrolytes as they flow through the cell and the separate storage of both, even in cell-stack designs, are particularly advantageous.

[0080] In the other use of this type of power key ( Fig. 10) After the collector electrodes, meshes, sieves, coarse frits, or the like are sufficient. The conductors (1, 1') are protected from direct contact (short circuit) by a narrow spacer (3) with peripheral flow openings. During the discharge process, the discharged redox electrolyte solutions flow through these and, turbulently mixed, exit at the periphery, as indicated by the arrows (5). If such a current-key design is used as a membrane replacement in conventional redox flow batteries, the redox electrolyte mixture exiting at (5) can be added back to the reservoirs after suitable distribution. It is advantageous here to start with discharged redox-pair mixtures (see Example 8) for charging and generating the catholyte and anolyte, as these are generally less susceptible to crossover effects.

[0081] The in Fig. The inserted sketch (6) shows only one example of the many possibilities available to increase the combination cross-section of the two electrolyte solutions in order to minimize electrolytic resistance.

[0082] An example of a particularly simple and inexpensive to manufacture, membrane-free redox flow battery (flow cell) is in Fig. Figure 12 is sketched. The charged catholyte enters at A and the charged anolyte at B, each in a tubular half-cell with a downward bend. Before being joined to form a U-shape, both are filled with inert and electrically conductive packing material via a well-sealing flange connection. Segments 1–6 and 1'–6' show regions where the surface area of ​​the packing material is successively increased and the flow channels are reduced in size. Both legs terminate at the bottom with an electrically conductive sieve, frit, or similar element in close contact with the electrically conductive packing material (as the collector electrode). Reproducible packing densities are easily achieved if the packing material is weighed section by section and each half-cell is filled like a chromatographic column. The process is particularly simple when using a graphite felt packing.Here too, weighed quantities are filled into the tube, and then the desired surface area and flow channel dimensions for the relevant section are set at predetermined positions (snap-in) by varying the pressure applied using electrically conductive screens, which also ensure the even distribution of the flowing electrolytes. After identical and reproducible filling of the half-cells with the electrode material, both are flange-connected by means of an insulator ring with outlet openings C for the combined discharged electrolytes, whereby the current key according to the invention with a large cross-section is automatically formed at D. If a double-acting pump according to the invention is used as a reservoir, the higher hydraulic resistance of such half-cells can easily be generated without significant energy expenditure. Fig. The electrical connections are not shown in Figure 12. If the same cell is to be used for charging, a valve-like control must be used to prevent the electrolytes from mixing in area D and to create two outputs for regenerated electrolytes there. The current key is then located at the input and is, as shown in Fig. Figure 10 shows how to build it. Because of the very simple construction and the large surface area achievable even with small dimensions, a separate electrochemical cell can also be used solely for the charging process. This cell is identical but with the surface area increased in the opposite direction (from 6 and 6' towards 1 and 1'). The mixed, discharged electrolytes then flow in at the bottom and exit at the top, charged and separated into catholyte and anolyte.

[0083] The additional cell specifically for recharging doesn't have to be a disadvantage. It can also offer flexibility, such as simultaneous discharging and charging, for example, when solar energy is being used from the battery at night and wind energy is being stored in parallel. The absence of bipolar electrodes is compensated for by the significantly larger surface area that can be achieved, which improves performance. Connecting multiple cells of this type in series externally is easily accomplished, as is replacing individual cells. Stray currents can be avoided by appropriately dimensioning the inlet and outlet lines and by incorporating flow interruptions, similar to the design of a drip tray in an IV stand.

[0084] Unlike conventional cell stacks, which require an expensive membrane and are difficult to seal, U-shaped flow cells are based on inexpensive, easily sealable plastic pipes and are very easy to manufacture, modify, and maintain. Because of the absence of a fragile semipermeable membrane, very robust designs are possible. Example 6

[0085] Fig. Figure 13 schematically shows another preferred embodiment of the disclosed apparatus. The redox system to be converted at the coiled electrode 1, 1' flows centrally into the half-cell at point 4 and is guided through a perforated inner tube and a spacer 2, which opens to an increasing surface area from the inside out, through the spiral channel as indicated by arrow 5. Preferably, the spacer, which is also coilable, should gradually become thinner from the inside out in order to achieve a higher flow velocity by narrowing the flow channel. To obtain a complete electrochemical flow cell with coulometric operation, a second cell of analogous construction is brought into close contact with the first such that both outflowing and completely converted redox electrolytes meet across the entire width of their largest flow front at the end, thereby forming the necessary current key.During the charging process, a power key is used, as in . Fig. Figure 10 shows the electrolyte flow being used before the inflow (4), and the exiting charged redox electrolytes are then separately discharged and added to the storage tanks. Switching the electrolyte flows at this point, with a flat and wide channel end, can be accomplished by means of a suitable valve control. Example 7

[0086] Fig. Figure 14 schematically and in a simplified manner shows another advantageous arrangement in which a conventional cell-stack structure is possible during the discharge cycle. Here, catholyte and anolyte flow into the respective flow-through half-cell at points 1 and 1'. The electrodes are planar and, due to their structure and connection with the outer walls, force the redox electrolytes into a meandering or serpentine flow. The flow channels formed by the planar electrodes become narrower from inlet to outlet, so that the flow velocity increases according to the invention. These flow channels are in Fig. Figure 14 is drawn exaggeratedly long to illustrate the channel narrowing. If the channels are filled with conductive material (e.g., graphite felt), the number of meanders can be reduced by compressing this filling towards the end of the collector electrode. After this serpentine course through the half-cells, both redox electrolytes flow through a final perforated electrode and are discharged and combined in a very small space at (2). The advantage here is that the current-carrying area between the two half-cells is automatically particularly large. The distance between the two last perforated electrodes should be as close as technically possible. They must be insulated from each other. The electrodes at the inlet at 1 and 1' can thus also serve as bipolar electrodes. According to the invention, this also allows for a "stack" configuration without disruptive separating membranes, as indicated by arrows (3 and 3').The serpentine flow channel with narrowing channels can also consist of a non-electronically conductive material, while the channels can be filled with an electronically conductive material (e.g., graphite felt, graphite fabric, graphite threads, activated carbon, or carbon nanotubes (CNTs)). Redox sensors at the inlet and outlet for controlling the optimal flow rate are included. Fig. Figure 14 is not shown. The sensors at the outlet must be positioned upstream of the current key area, which is characterized by numerous flow arrows. The actuators for the redox sensors (so-called Pt single-rod measuring chains) are flow-regulating devices (e.g., valves) that are not shown here.

[0087] Fig. Figure 15 shows schematically and by way of example the arrangement of Fig. 14 during the charging cycle. The modified inlet and outlet arrangement can be changed by means of appropriate valves in the case of permanently installed lines, which is obvious to those skilled in the art. According to the solution disclosed here, which does not require a semipermeable separating membrane, the necessary electrolytic current key must be present before the two redox electrolytes enter the two half-cells for regeneration during the recharging process. In (1), the combined, discharged redox electrolytes initially flow together into the two half-cells, thereby forming the necessary electrolytic current key, which also divides them. In (2) and (2'), the regenerated (recharged) redox electrolytes flow out of the respective redox flow battery half-cells and can again be collected separately in their storage tanks.The fact that they may contain a further, redox-inactive compound according to the invention is not a problem, but rather advantageously increases the electrical conductivity of the electrolytes, which is particularly beneficial for the current key. In . Fig. At (3) and (3'), valves that may be required to regulate the flow rate are shown. These valves can compensate for stoichiometric or other imbalances (e.g., oxidation of the reducing redox system by atmospheric oxygen) and regulate the overall flow velocity in the channels to ensure a near-quantitative conversion in the half-cells. At (4) and (4'), electronic conductors are positioned in a network or other configuration suitable for easy flow, and can also be positioned upstream of (3) and (3'). They are electrically connected to the relevant electrodes at the half-cell inlet. The control redox sensors (not shown) can be positioned upstream of (4) and (4') and at the outlet at (2) and (2'). Example 8

[0088] In a particularly advantageous further embodiment, the main claim of this invention, namely the quantitative redox conversion during a single cell flow to form an electrolytic current key that replaces a membrane, is optimally used to maintain a particularly stable terminal voltage of a redox flow battery during the discharge process. This is achieved, according to the invention, by preventing the discharged redox systems from being recirculated and mixed with the still-charged electrolyte in the storage tank. This can be accomplished either by using a sufficiently large intermediate storage tank or, even more advantageously, by using special storage tanks in the embodiment of a cylinder with a movable piston connected to a push rod, which can function as a pump. Fig. Figure 16 illustrates the principle of a so-called double-acting piston pump, which is also commercially available in acid-resistant versions and various sizes. The connection to the actual redox flow battery is made via flexible pressure lines. Discharged redox electrolyte flows into the cylinder through opening (1). Through opening (2), the relevant, still fresh, charged redox electrolyte is pumped through the respective half-cell by the movable piston (3) with its sealing ring (4) as it moves in the direction of the arrow. This principle allows for a higher pressure build-up compared to centrifugal or peristaltic pumps. At the end of the piston stroke, the relevant charged redox electrolyte, which up to this point has maintained a constant battery terminal voltage, is depleted, and the charging process is initiated by a piston movement in the opposite direction.At the same time, a suitable valve control on the redox flow battery ensures that the electrolytes are now supplied via the charging current key.

[0089] This principle also works if only one push rod (5) or (5') is present. This type of redox electrolyte transport through the redox flow battery is particularly advantageous when the flow channel dimensions in the respective half-cells require higher pressures. For example, a significant portion of the pressure generated on the other side of the piston is still present on the side with the inlet opening (1), through which the electrolyte discharged in the half-cell flows back. Such a pressurized backflushing of the piston with liquid considerably reduces the energy expenditure for pumping the redox electrolytes through the redox flow battery, which reduces efficiency. This is based on a preferred and proven pumping technology used in reverse osmosis. However, a seawater desalination plant based on this principle requires significantly higher pressures.For practical purposes, it should be noted that, for example, with a medium cylinder size of 150 cm in length and a diameter of 40 cm, at least 4 kWh of electricity can be stored using a vanadium redox flow system.

[0090] The inventive separate storage of discharged and still charged redox electrolyte solution in energy-saving, double-acting piston pumps can be further optimized if the piston pumps for anolyte and catholyte are connected by means of a common push rod. Then, a single mechanical push assist based on electrical, pneumatic, or potential energy is sufficient to pump both electrolytes simultaneously through the half-cells, even under high pressure. Example 9

[0091] The principle of a double-acting piston pump for the efficiency-reducing transport of the two redox electrolytes through the electrochemical half-cells of a redox flow battery can be used in a further exemplary preferred arrangement to save even more pumping energy. This example is also based on claim 1 with the generation of nearly quantitatively converted redox electrolytes. Fig. Figure 17 schematically illustrates the principle of the discharge process, which utilizes the potential energy of a sufficiently heavy weight for both storage tanks, represented by vertically operating double-acting piston pumps. The weight is movably positioned on one side (end) of an isosceles lever (rocker arm), which connects both pistons in the respective electrolyte tanks via a pushrod. At the end of each piston stroke, the weight is moved to the end of the other lever arm. Energy is required only for this movement. This can be achieved, for example, by a small electric motor and gear drive along the lever surface, or pneumatically using the pressure of an inert gas from a pressurized cylinder. An inert gas, such as nitrogen or argon, is typically required in redox flow batteries with aqueous electrolytes to remove dissolved oxygen. Fig. Figure 17 shows, in simplified form, a redox flow battery cell with two electrochemical flow electrodes (1) and (1') and the two associated charged redox electrolytes (2) and (2'). These electrolytes are pumped through the half-cells by a corresponding movement of the piston (3) and (3'). The disclosed current-key construction with integrated mixer and electrolyte divider (4) then pressurizes the rear of the piston, thus returning the electrolytes to the tank when discharged. An advantage is that the potential energy of a weight is used to pump both redox electrolytes. The positioning of the redox sensors before and after the half-cell flow is not shown, nor is the control device for the optimal flow rate of the redox electrolytes per unit time, as it is obvious to those skilled in the art. Fig. 18 shows the arrangement of Fig. Figure 17, simplified diagram, shows the recharging situation. To charge, the weight is shifted to the other side of the lever by the only energy-consuming action. The labels correspond to those of Fig. 17. The position of the flow key (4) with the fluid flow divider and regulator is important here, in front of the electrolyte flowing into the relevant half-cell. Example 10

[0092] The electrolyte pump device according to the invention, in tandem configuration, can be further optimized in terms of energy efficiency by replacing the moving parts. The pumping and suction action of a double-acting piston pump also occurs when the pistons are fixed and the two mechanically connected housings are moved accordingly. In this case, the potential energy corresponding to the weight of both redox cylinders can also be utilized. Fig. Figure 19 shows redox electrolyte reservoirs according to the invention in the form of double-acting piston pumps in a particularly preferred tandem arrangement, wherein the common push rod is rotatably fixed in the center and vertically displaceable by a few centimeters. Such a reservoir and pump arrangement is operated like an hourglass. Cylinder A is filled with catholyte and cylinder B with anolyte, respectively, with the respective proportions of discharged and still charged redox electrolyte on the other piston side. The tandem arrangement rests on the base C with one foot (6), and the two cylinders, rigidly connected to each other (not shown), move downwards by their own weight, with the locked pistons performing a stroke that results in the expulsion of electrolyte at the top and the intake of electrolyte at the bottom. Once both pistons have reached their end point, the entire tandem arrangement is rotated by 180°, and the charging or discharging process begins, depending on the fill level on both piston sides.The only energy required for pumping consists of a small lifting motion followed by a 180° tilting movement. For this, the two pump housings are first locked onto the push rod, then – due to the rounded feet – only lifted a few centimeters, rotated, lowered, and released from the locking position. External energy is required only for this brief process, and this contributes to the overall efficiency. Depending on the pressure required to circulate the fluid through the redox flow battery, the housings can also be weighted down with additional external weights.

[0093] This device for pumping redox electrolytes through the respective half-cells is also advantageously applicable to redox flow batteries in general, which do not achieve 100% redox conversion in a single flow. In such cases, after the piston stroke has reached its end and the movement in the opposite direction continues to pump slightly less charged redox electrolytes through the half-cells, and the discharge characteristic occurs in corresponding clamping voltage stages. The same principle applies to charging. After switching the valve with the current key before the electrolytes enter the half-cells, charging can also be carried out in several stages by means of multiple piston strokes (rotations of the device).

[0094] Although the invention has been further illustrated and described in detail by the preferred embodiments, the invention is not limited by the disclosed examples and other variations of the disclosed principles can be derived by the person skilled in the art without leaving the scope of protection of the invention. List of characters Fig. 1. Basic structure of a membraneless redox flow battery in conventional design during the discharge process; 1 = Replacement of the conventionally used semipermeable membrane with a dense insulator; 2 = Collector electrodes; 3 = Circulation pumps; 4 = Redox electrolyte; 5 = Special current key for the half-cell output; 6 = Redox electrolyte divider for the discharged electrolytes; 7 = Sensors for controlling the electrolyte flow rate. Fig. 2. Basic structure of a membraneless redox flow battery in conventional design during the charging process: 1 = Replacement of the conventionally used semipermeable membrane with a dense insulator; 2 = Collector electrodes; 3 = Circulation pumps; 4 = Redox electrolyte; 5 = Special current key for the half-cell inlet; 6 = Mixer of the discharged redox electrolytes; 7 = Sensors for controlling the electrolyte flow rate. Fig. 3 Single-electrode potentials of an iron(II) - cerium(IV) system. Potential profiles during a redox titration in homogeneous solution; A: Titration of an iron(II) sulfate solution with ce(IV) sulfate solution; B: Titration of a ce(IV) sulfate solution with an iron(II) sulfate solution; small arrows pointing to the right show the single-electrode potential profile of the two half-cells (A and B) during current generation during a discharge of a correspondingly constructed redox flow battery; the large arrows pointing to the left show the single-electrode potential profile of the two half-cells (A and B) during a recharge of completely discharged electrolytes (here: only Fe(III) and Ce(III) are present). Fig. 4 Typical charging and discharging characteristics of a redox flow battery: (single cell voltage versus current / cm²) 2 applied); 1 = Battery terminal voltage curve during charging; 2 = Open-circuit voltage of a single cell; 3 = Battery terminal voltage curve during discharging with different current densities (solid line = conventional redox flow battery; dashed line = redox flow battery according to the invention; A = Region of cross-polarization; B = ohmic voltage drop due to internal resistance; C = concentration polarization due to insufficient mass conversion. Fig. 5 Principle of an experimental redox flow battery setup without a separating membrane and coulometric operation 1 = Generator electrode; 2 = Collector electrode; 3 = Bipolar electrode (activated carbon slurry) ; R1 = resistance for galvanostatic current flow via (1) and (3); R2 = resistance of an external load and current measurement using voltage U. Fig. 6 Details of the experimental setup of Fig. 5 1 = Generator electrode; 2 = Collector electrode; 3 = Bipolar electrode (activated carbon slurry); 4 = Air inlet for electrolyte circulation: 3 m NaBr, 0.005 m Nal and 0.1 m NaH2PO4 / Na2HPO4; 5 = Air outlet; A and B = Positions of the redox electrodes for conversion measurement. Fig. 7 Schreiber protocol for proving the coulometric operation of an experimental iodine-activated carbon redox flow battery upon introduction of traces of SO2. Fig. 8 Simplified scheme of a membraneless redox flow battery according to the invention with tubular half-cells filled with electronically conductive packing materials during the discharge process; 1 = packing material (spheres, etc.) with electronically conductive surface; 2 = special current key; 3a = input of charged (reduced in the half-cell) anolyte; 3b = completely electrochemically converted anolyte; 4a and 4b = analogous to 3a and 3b for the catholyte. Fig. 9 Simplified scheme of a membraneless redox flow battery according to the invention with tubular half-cells filled with electronically conductive packing material during the charging process; collector electrodes are not shown here for clarity); 1 = packing material (spheres, etc.) with electronically conductive surface; 2 = mixer and current key; 3a = inlet of discharged anolyte from the respective reservoir; 3b = completely electrochemically reduced anolyte; 4a and 4b = analogous to 3a and 3b for catholyte regeneration; during the discharge process, the mediator-electrolyte currents occur as in Fig. 2a shown. Fig. 10 General principle of the current key according to the invention as a membrane replacement 1 = electronic conductor in mesh or frit form on the anolyte side; 1' = electronic conductor in mesh or frit form on the catholyte side; 2, 2' electrical connection from 1 and 1' to the nearest collector electrode; 3 = insulator spacer; 4 = discharged analyte flow; 4' = discharged catholyte flow; 5 = peripheral outflow of the now mixed redox electrolytes; 6 = example of a cross-sectional enlargement; during charging, the mixed and discharged electrolytes flow into the respective half-cells at 5 in the opposite direction. Fig. 11 Example of a current key according to the invention as a membrane replacement in “cell-stack” designs 1 = frit-shaped electronic conductor on the anolyte side; 1' = frit-shaped electronic conductor on the catholyte side; 2, 2' electrical connection from 1 and 1' to the nearest collector electrode; 3 = insulator spacer; 4 = discharged analyte partial flow; 4' = discharged catholyte partial flow; 5 = peripheral outflow of the now mixed redox electrolytes; during charging, a partial current of the mixed and discharged electrolytes flows in reverse direction at 5 into the respective half-cells. Fig. 12. Example of an easily and inexpensively produced redox flow battery cell during the discharge process A = Catholyte inlet; B = Anolyte inlet; C = Insulator; D = Current key area, electrically conductive sieves or meshes are in contact with the electrically conductive filler materials; 1-6 and 1'-6' illustrate areas of increasing filler density with electronically conductive fillers (spheres, meshes, sieves, graphite felts with increasing pressure, etc.). Fig. 13 Example of a redox flow battery half-cell in wound form 1 and 1' wound collector electrode; 2 = spacer for generating the channels according to the invention; 3 = housing; 4 = electrolyte inlet channel Fig. 14 Example of a redox flow battery according to the invention in a “cell-stack” arrangement during the discharge process 1 and 1' Entry of the charged electrolytes into a cell with 2 half-cells and meandering flow channels and current key formed from a perforated final collector electrode on each side; 2 = Outflow opening of the combined, discharged electrolytes; 3 and 3' bipolar electrodes for attaching further cells of the same design. Fig. 15 Example of a redox flow battery according to the invention in a “cell-stack” arrangement during the charging process 1 = Entry of the mixed, discharged redox electrolytes into the current key as an electrolyte divider; 2 = Outlet of the charged and separated electrolytes; 3 and 3' = Flow rate controller; 4 and 4' Control sensors for 3 and 3' (not shown: sensor pair at outputs 2 and 2'). Fig. 16 Principle of the redox electrolyte storage tanks according to the invention in the form of a so-called double-acting piston pump during the discharge process 1 = Entry of the relevant, discharged electrolyte into the left cylinder area; 2 = Exit of the relevant charged electrolyte from the right cylinder area; 3 = movable piston; 4 = sealing ring; 5 and 5' Pushrod for moving the piston in the direction of the arrow, at the end in the opposite direction = loading process: Fig. 17 Example of a membraneless redox flow battery according to the invention, powered by gravitational energy during the discharge process (flow diagram) 1 and 1' = Collector electrodes; 2 and 2' Catholyte and anolyte; 3 and 3' Pistons for separately storing discharged electrolytes; 4 = Current key with divider; Fig. 18 Example of a membraneless redox flow battery according to the invention, powered by gravitational energy during the discharge process (flow diagram) 1 and 1' = Collector electrodes; 2 and 2' Catholyte and anolyte; 3 and 3' Pistons for separately storing discharged electrolytes; 4 = Current key with divider; the weight is moved to the other side of the centrally mounted rocker at the end of the respective piston stroke by means of electrical or pneumatic energy. Fig.19 Example of a membraneless redox flow battery according to the invention with storage tanks in the form of double-acting piston pumps in tandem arrangement driven by gravitational energy A = anolyte; B = catholyte; 1 = inlet of the respective discharged electrolyte into the upper cylinder region; 2 = outlet of the respective charged electrolyte from the lower cylinder region; 3 movable piston, 4 = sealing ring; 5 and 5' combined pushrods for moving the piston; 6 = base; both cylinder housings are rigidly connected to each other (not shown) and represent the weight that sinks downwards; at the end, the entire tandem arrangement is rotated 180 degrees, like an hourglass, for which only a brief external energy is required.

Claims

[1] Device for storing electrical energy in chemical redox compounds in the form of a redox flow battery, based on at least two liquid redox electrolytes (catholyte and anolyte (4)), comprising at least one electrochemical flow cell consisting of two half-rows, each with an electrode compartment through which a redox electrolyte flows, and a collector electrode made of a chemically inert, electron-conducting material (2) and circulating pumps (3), wherein the two half-cells are separated by an impermeable, electrically non-conductive wall (1), characterized by , that a) the electrochemical conversion of the redox electrolytes in question is quantitatively controlled during a single half-cell flow and discharged redox electrolytes leave the flow cell, b) during the discharge process, the discharged catholyte and anolyte fluids or parts thereof form an uninterrupted electrolytic current key within a defined mixing zone of both, which electrolytically connects the two half-cells at the outlet, and during the charging process, a corresponding uninterrupted current key is formed by a defined flow division of discharged catholyte and anolyte before the inlet of both half-cells, both of which replace a semipermeable membrane. c) Catholyte and anolyte contain more than one redox system, d) the discharged redox electrolytes exiting the flow cell and combined according to the current key are separated again into catholyte and anolyte fractions and stored separately from their still charged fraction, e) the discharged redox electrolytes are temporarily stored in containers that can act as a pump. [2] Device according to claim 1, characterized by, that a) the electrochemical half-cells contain collector-electrode arrangements that allow the electrochemically active electrode surface and the relative movement of the redox electrolytes to the electrode surface to increase continuously or discontinuously in the flow direction, and that a quantitative electrochemical conversion is ensured in a single flow by appropriate variation of the flow rate, b) the electrolytic current keys, during battery discharge in a narrow mixing range and during battery charging in a narrow division range, still have closely adjacent but mutually insulated, permeable electronic conductors that are electrically conductively connected to their nearest collector electrode.c) the redox electrolytes reacted in the half-cells, after their mixing in the current key and re-separation into catholyte and anolyte components, are returned to their respective, special cylindrical storage tanks, whereby mixing with still charged redox electrolyte components is avoided by separating discharged and still charged redox electrolyte in the cylinder by a materially and electrolytically sealing, movable and insulating piston with piston rod, and this storage tank device functions as a double-acting piston pump, which also takes over the pumping of the respective redox electrolytes through the electrochemical cell. [3] Device according to one of the preceding claims, characterized by, that the required surface area increase of the collector electrodes and flow velocity increase in the flow direction of the half-cell is achieved by means of electronic conductors made of a chemically inert material that are in contact with each other, wherein the size of the packing particles decreases continuously or discontinuously in the flow direction, or already structured material, such as graphite felt, shows an analogous behavior due to different pressure. [4] Device according to any one of the preceding claims, characterized by , that the electrode surface area enlargement and flow velocity increase in the direction of flow in the half-cells are generated by means of suitably cut, chemically inert, and in the direction of flow thinning spacers and coilable electrodes. [5] Device according to any one of the preceding claims, characterized by, that the surface area increase of the redox electrolytes towards the end of the flow path in "cell-stack" designs is achieved by a meandering or serpentine fluid flow with channels that widen towards the end, and the increase in the redox electrolyte flow velocity is achieved by reducing the channel height through correspondingly different pressure on the graphite felt often used as an electrode, and the electrolytic current key is constructed by means of a collector electrode, which - equipped with narrowly tapered tubular channels - allows time for the electrochemical reaction in a stagnation area and then releases the electrolyte in question as a jet current into the current key area. [6] Device according to any one of the preceding claims, characterized by, that in both redox electrolyte storage tanks of a redox flow battery with the function of a liquid-pumping, double-acting piston pump via a common push rod in tandem design, only one common drive source is required for pumping both redox electrolytes. [7] Device according to claim 6, characterized by , that the mechanical movement of the pushrod of the storage tanks designed as a double-acting piston pump for the purpose of electrolyte transfer is caused by the potential energy of this tandem arrangement, the latter being built vertically and resting on a pushrod and being rotated by 180° at the end of the piston stroke. [8] Device according to any of the above claims, characterized by, that the pushrod of the two redox electrolyte tanks in the form of a double-acting piston pump is pushed downwards by the potential energy of a sufficiently heavy weight and the pushrod in the other redox electrolyte tank is thereby pulled upwards via a rocker-like connection between the two and the weight is transported at the respective end point of movement back to the highest position above the other redox electrolyte tank by means of a motor carriage. [9] Device according to any of the above claims, characterized by that the surface area increase and flow velocity increase are generated by microtechnology using appropriate microfluidics, and that greater electrical power is generated through massive parallel operation. [10] Device according to any of the above claims, characterized by, that the electrochemical conversion in the two redox flow battery flow half-cells during a single flow is individually controlled during the catholyte and anolyte flow by varying the flow rate of the respective redox electrolyte, such that a potential difference, measured with separate platinum single-rod electrodes in the electrolyte, exists between the redox potential of the respective redox electrolyte at the inlet to the respective flow half-cell and at the outlet, according to the Nernst equation for the desired conversion rate, and that the conversion rate should be above 50%, but preferably between 80 and 99%.

Images