DE102021001501A1 - Process and device for regenerating anolyte and catholyte in redox flow batteries - Google Patents

Abstract

The invention discloses a novel electrochemical method for increasing the energy density of redox flow batteries (RFB), which can also be used by means of a novel device for regenerating anolyte - and catholyte - liquids of traditional RFB. Organic redox polymers in solid electron exchange resin form or as dissolved polymers as so-called main redox systems that determine the RFB storage capacity are preferably used for this purpose. The new electrochemical technology of "bi-polar internal electrolysis" together with special equipment arrangements avoids the build-up of high hydrostatic pressure when the anolyte and catholyte flow through the corresponding main redox systems in the RFB storage tanks. Specially designed flow channels with electric Conductive walls and special electrolyte bridges avoid direct physical contact of the redox systems in the anolyte and catholyte with the main redox systems and, in addition to an exchange of redox-inactive ions via the electrolyte bridges, only allow an electron exchange. The flow channels for anolyte and catholyte are preferably arranged in such a way that the regeneration device behaves like an ion exchange column, ie electrochemically “fully charged” anolyte or catholyte always escape until the end of the storage capacity. The electron transfer in the reservoirs with the stationary main redox systems is brought about by surface-enlarging electronic conductors in contact with the electrically conductive flow channel walls.

Description

The invention lies in the field of electrochemical energy storage using the known redox flow batteries. Since the redox-active substances are dissolved in an electrolyte here and the energy is no longer stored in the battery electrodes - as in many traditional batteries - but in two different chemically oxidizable or reducible compounds external to the battery cells, one wins constructive freedoms. The electrical energy that can be stored now depends on the amount of redox-active compounds (redox systems) stored externally in storage tanks and no longer affects the volume or weight of the actual battery with the current-supplying electrochemical flow-through cells. The redox systems used are usually dissolved in a good, high-conductivity electrolyte. The redox electrolyte, which is oxidized during the discharging process at the collector electrode, is called "anolyte" and that which is reduced during discharging is called "catholyte". The present invention discloses a novel method and a novel regeneration device based on known redox flow batteries with the aim of decisively increasing the still insufficient energy density of redox flow batteries. Novel organic-based redox compounds that are known or still to be synthesized should preferably be used for this purpose. Energy density is determined by the amount of redox groups per volume or weight. Since this amount is usually limited by the solubility of the relevant redox compound in the electrolyte, constantly synthesizing new compounds that contain several redox groups in an organic macromolecule. An overview of these redox polymers is given by P. Novák, K. Müller, KSV Santhanam, O. Haas, in "Electrochemically Active Polymers for Rechargeable Batteries", Chem. Rev. 1997, 97, 207-281 and - more recently - J. Winsberg, T. Hagemann, T. Janoschka, M.D. Hager and U.S. Schubert in "Redox Flow Batteries: From Metal-Based to Organic Active Materials", Angew. Chem. 2017, 129, 702 - 729. The object of the present invention is to transfer all of these newly developed energy-dense, preferably organic redox compounds, which will be synthesized in the future with a targeted redox potential, to a solid redox polymer resin Both the base and the one based on dissolved and highly viscous redox polymer electrolytes can be used practically in redox flow batteries without suffering correspondingly large pumping energy losses, which greatly reduce efficiency. This task is to be solved with a new process that uses the newly developed technology of bi-polar internal electrolysis and with the help of a new device for regenerating anolyte and catholyte. The new device and regeneration arrangement consists of special, preferably column-shaped storage tanks, which are filled with these new energy-dense organic so-called main redox systems, and through which the anolyte or catholyte to be regenerated flows. The electron transfer required for regeneration between the redox systems in anolyte and catholyte and the electrochemically precisely matching main redox systems is accomplished by different arrangements, depending on whether a solid so-called electron exchange resin or a dissolved organic redox polymer present. What they both have in common is a hydraulically optimised, line-bound “flow through” of these storage capacity-determining main redox systems in various embodiments, which preferably show a behavior analogous to an ion exchange column. The solution to a technical problem disclosed here of decisively improving the energy density of redox flow batteries (factor > 10), the new method has other advantages, which are listed below.

After fossil energies are to be replaced worldwide by environmentally friendly regenerative energies - in particular by solar and wind energy - in order to protect the climate, there is a huge need for reliable and inexpensive energy storage technologies. The current electrochemical energy storage systems, which were primarily developed on the basis of inorganic chemical elements, are not very suitable for global requirements, since the raw materials are becoming scarce with global demand, they cannot be obtained or disposed of in an environmentally friendly manner, and because they are or will become too expensive , if they are to be increasingly used worldwide. Disadvantages of the classic battery types, which also include metal hydride batteries, are - in addition to their high acquisition costs - their limited service life (limited number of cycles) and the use of toxic or aggressive chemicals with the associated disposal problems. A total change away from fossil energy sources to the use of renewable energies - as is now politically stipulated worldwide - should also be able to cover the energy requirements of an average family for several days in the area of "storage of regenerative energies in the private domestic area. Taking into account electrical consumers and the ban on operating heating systems that emit CO ₂ , in such an exemplary stationary domestic application, around 300 kWh of storage capacity would have to be stored for around a week without wind and sufficient sunshine. In order to store such amounts of energy in a decentralized manner, the redox flow batteries, which have been commercially available for a long time and are now known to every specialist, are particularly suitable.

the 3 shows the principle very schematically, based on a traditional structure, but with an advantageous variant here: use of bi-polar internal electrolysis for the regeneration of anolyte and catholyte. In general, the following applies to redox flow batteries: The compounds required to set the potential at the so-called collector electrodes (2) of an electrochemical flow cell are redox-active species dissolved in a solvent, which during the charging or discharging process in an electrochemical Flow reactor (= electrochemical cell) are converted into their respective other redox stage. An electrochemical redox flow cell contains two polarity-specific chambers (4, 5) with electron-conducting collector electrodes, through which a solution (anolyte) with an electrochemically oxidizable chemical compound or a solution with an electrochemically reducible compound (catholyte) flows . Two circulating pumps (3) are used for this purpose, the energy consumption of which is included in the overall efficiency of the relevant redox flow battery. In order to increase the electrical conductivity, these solutions also contain salts, so they are good electrolytes (ion conductors). Both electrolytes (anolyte and catholyte) in traditional redox flow batteries are each connected to a larger external reservoir. In this way, two independent circuits are formed for the redox-active compounds dissolved, for example, in water or an organic solvent. In the electrochemical flow cell, they are separated between the polarity-specific chambers (= half-cells) by an ion-conducting membrane (1), which is intended to prevent direct material mixing of the two redox compounds. The redox short circuit to be avoided means electron exchange from molecule to molecule without the "detour" via the collector electrodes and the external consumer. However, according to Faraday's law, an ion exchange of non-redox-active ions between the two chambers must take place via this membrane in accordance with the current drawn. The ease of this ion transport through the separating membrane decisively determines the internal resistance of the relevant electrochemical cell, which according to U = ix R reduces the battery voltage with increasing current strength. The battery voltage depends on the difference in standard redox potentials of anolyte and catholyte. In this application, the term standard redox potential is used, although strictly speaking standard formal potentials are meant, which are practically measured in the electrolyte in question when the concentrations of both redox forms of the system in question are the same (]Ox} = [ Red] in the Nernst equation regardless of activity coefficients, solvation effects or complexations, etc.).

The present invention discloses a novel regeneration technique for the anolytes and catholytes of most redox flow batteries using a new method and apparatus. The traditional storage tanks for anolyte and catholyte of this type of battery are filled according to the invention with an energy-dense electrochemically adapted main redox system and anolyte or catholyte flow through the corresponding main redox system present in a large stoichiometric excess for the purpose of regeneration. This new design (arrangement and structure) allows for the first time numerous new, particularly energy-dense. organic redox systems can be used in conjunction with redox flow batteries without great loss of efficiency. In contrast to inorganic redox flow batteries, preferably based on metallic elements, which only allow a very limited choice of redox potential, organic redox molecules can be synthesized in a targeted manner towards a specific standard redox potential by varying the various substituents . They can be obtained easily and in large quantities from inexpensive raw materials that are not subject to shortages.

Redox flow batteries, which work with completely organic, low molecular weight redox systems dissolved in an electrolyte, have been increasingly described in the literature and in patents in recent years, but in principle they are not convincing in terms of their energy density. The energy density in terms of ampere hours per kilogram or liter (Ah/kg or Ah/L) or watt hours per kilogram or liter (Wh/kg or Wh/L) depends on the per weight or per volume of the complete battery (including storage tanks ) chemical redox equivalents to be accommodated. It is limited by the solubility of the organic redox compounds in a liquid electrolyte, which is necessary because of the necessary high ion conductivity (influence on the important internal resistance). However, this packing density of redox molecules can be increased by orders of magnitude by using polymeric redox compounds with numerous active, electron-exchanging redox centers per macromolecule. When using such redox polymers, very high energy densities can be achieved in redox flow batteries, which can exceed those of lithium-based rechargeable batteries.

Particularly promising organic redox polymers in this context are the solid so-called electron exchange resins that were specifically synthesized in the middle of the last century and were used in columns like classic ion exchange resins. Another promising class that has been synthesized in recent years is water-soluble organic redox polymers, which are neutral, harmless Salt solutions are dissolved. Both classes of fabric are very environmentally friendly; both in production and in disposal, where they enable simple "thermal recycling". In addition, with the water-soluble organic redox polymers - they are used in traditional redox flow batteries - the use of large amounts of flammable organic solvents can be dispensed with. Both redox polymer classes can be rated as extremely sustainable in terms of environmental protection, even if they are widely used. There is therefore also a great need to be able to use these organic redox substance classes in redox flow batteries without major losses in efficiency due to high pump energies.

With these redox polymer classes (organic solid resins and water-soluble polymers), which are interesting for reasons of energy density, a technical problem has to be solved: Such redox polymers can no longer be used, as is usual with classic redox flow batteries, as Pump low-viscosity solutions in a circular flow through the reservoir and the current-supplying electrochemical cells (or connected in series - battery-cell stacks). The electron exchange resins would require a solid suspension, which leads to abrasion and line blockages over long periods of operation. In addition, current-yielding electron transfer to or from the collector-electrode surfaces in the electrochemical flow-through cells is severely hampered because of the short contact time of the solid particles with redox-active groups. In the case of the organic redox polymers that can be dissolved in a solvent, the viscosity of the two electrolytes (anolyte and catholyte), which increases sharply with increasing energy density (degree of polymerization) and/or concentration, prevents energy-saving pumping, so that the efficiency can approach zero. These existing problems hinder the practical, economical application of these new promising classes of substances, with which the energy density of redox flow batteries can be significantly increased. The present invention solves these problems with a new method that always pumps the anolytes and catholytes of the redox flow batteries through the “battery stacks” in a nearly fully charged state by means of a new type of regeneration device. As a result, this invention can also be used in the future developments still to be expected in the field of novel organic redox polymers. Therefore, no specific redox-active redox polymer substance classes are listed in the claims. The regeneration device disclosed here is used particularly advantageously in solid granular redox systems (electron exchange resins) and in liquid high-polymer redox systems, each with specific different embodiments of the regeneration device.

The increasing demand for so-called renewable energies has led to a sharp rise in alternative energy conversion technologies, with the conversion of solar and wind energy into electricity becoming the most important because it is almost ubiquitous. Wind and solar energy are among those with the greatest growth rates. The disadvantage of both is the fact that they cannot be used continuously, i.e. they occur intermittently. For their full and efficient use, therefore, an acceptable (in terms of price and environmentally sustainable) energy storage is essential, in order to be able to use these alternative energies even in windless times or at night (including dark doldrums). They also have the great advantage of being able to be used decentrally without loss of efficiency. Here, the first commercially available redox flow batteries of traditional design have been used to particular advantage. With such decentralized energy storage, together with the smart grid technology that is just emerging, large, complex and contentious overland power lines are becoming less urgent. Lithium-based rechargeable batteries have so far been used less as stationary storage with high capacity for reasons of price and safety. In addition, widespread use as domestic energy storage would lead to significant raw material shortages, which are reflected in the price. Many traditional redox flow batteries also pose a significant environmental risk because toxic compounds and strong acids or alkalis are present in large quantities in the reservoirs due to the limited energy density to date. The novel design of redox flow batteries with high energy density according to the invention solves these raw material and environmental problems by using the electrochemical technology of “internal electrolysis” as a new method for regenerating anolyte and catholyte in principle of all redox flow batteries with a wide variety of chemistries and uses environmentally friendly and mass-producible organic compounds. This also opens up a new redox flow battery regenerator business model with a large market.

Redox flow technology is not new. The first redox flow batteries based on metal ions were proposed as energy storage devices as early as 1949 [W. Kangro, "Method for storing electrical energy", DE914264C (1949)]. In 1970, NASA began researching the redox systems of iron and titanium. Iron(III) chloride (FeCl ₃ ) in acidic solution was researched as an oxidizing agent and chromium(II) chloride (CrCl ₂ ) as a reducing agent in an alkaline environment. late 70s Years up to the beginning of the 80s, vanadium compounds in different oxidation states were used for the first time as a redox pair and an American patent was applied for in 1988 [ U.S. 4,786,567 ]. The one-element vanadium redox flow battery is the most common type to date. Although the vanadium redox flow batteries were developed in the 1980s, they have not had any major commercial success until recently. Other well-known examples of commercially available batteries are the zinc bromide, polysulfide bromide and iron-chromium "flow" batteries.

Unfortunately, in the course of the discharging process when supplying electricity, the terminal voltage of conventional redox flow batteries decreases in accordance with the progressing total conversion in the storage tank and - following the Nernst equation - until it reaches full conversion of both redox systems (chemical discharge of anolyte and catholyte in the storage tanks) falls rapidly to zero. This happens because the anolyte and catholyte, more or less discharged in the flow cell, are pumped in a circular flow through the storage tanks. For recharging, a reversed polarity and higher voltage is applied galvanostatically to the redox flow battery terminals (collector electrodes) and the two redox electrolyte solutions that determine the storage capacity are again pumped in a circular flow through the electrochemical cell until the entire catholyte - and anolyte - amount is completely recharged, i.e. has been returned to its original state. With a normal galvanostatic charge, this can be seen from a steep rise in voltage towards the end. There is a risk that hydrogen and oxygen will form, especially with aqueous electrolytes, which reduces the so-called coulombic efficiency. Due to the separate storage of anolyte and catholyte in appropriately sized storage containers, there is no self-discharge. They are not consumed in the process. Theoretically, the discharging-charging cycles can be repeated indefinitely, which represents a huge sustainability advantage of redox flow batteries compared to other battery types.

In traditional redox flow batteries, as long as the anolyte and catholyte are pumped through the electrochemical cells while they are charged, it supplies electricity. The battery voltage depends on the difference in standard redox potentials between the redox systems in question in catholyte and anolyte. The current intensity that can be drawn depends on the concentration of the electro-active compounds, the collector-electrode surfaces and the electrode kinetics and the mass transport from and to the collector-electrode surfaces. The total amount of electricity that can be delivered (capacity in Ah), on the other hand, depends on the material amount (g equivalents) of the anolyte or catholyte redox systems stored in the two storage tanks. This means that the amount of energy that can be stored in a redox flow battery is directly proportional to the stored amount of redox systems and can be described by Faraday's law. The maximum power that can be drawn in watts, on the other hand, is a function of the size of the clamping voltage, the size of the electrochemical reactor (surface of the collector electrodes) and the electrochemical kinetics and mass transport at the electrodes. In contrast to the well-known traditional batteries (lead/acid, metal/hydride, lithium/polymer, etc.), the amount of electricity to be stored (Ah capacity) is separate from the maximum current that can be supplied (maximum current in amperes without a complete collapse of the cell voltage ) adjustable. For stationary applications in the kWh range, the reservoirs of the previously standard redox flow batteries can also easily hold several m ³ (similar to a heating oil tank). Unfortunately, the energy density of all redox flow batteries known to date is only a fraction of that of heating oil. However, the latter is consumed when energy is used, while the electrochemical storage works without consumption due to the reversible electrochemical processes. The usual sizes of redox flow batteries have so far been in the range of around 2 to 50 kW and usual energy densities between 5 and around 40 Ah/L.

Numerous variants of the redox flow system have been described in the literature with regard to the redox chemicals used, all of which have their specific advantages and disadvantages. For an overview, please refer to relevant review articles [e.g. P. Alotto, M. Guarnieri, F. Moro, Renewable and Sustainable Energy Reviews, Volume 29, Pages 325-335 (2014)]. In the case of redox flow batteries based on transition metals, the solubility and the redox potential of the relevant metal complexes can be changed within limits by varying the metal ligands. Small molecule organometallic compounds have also been described [MH Chakrabartia, RAW Dryfe, EPL Roberts: "Evaluation of electrolytes for redox flow battery applications", Electrochimica Acta, 52, 2189-2195 (2007)]. Here, organic ligands are used that complex redox-active inorganic metal ions. Such ligands are, for example, dipyridyl, terpyridyl, phenanthroline, or imidazoles ( U.S. 2012/0171541 A1 ). Due to the formation of the complex, the redox potential of the relevant metal ions is shifted compared to a situation in a purely aqueous or acidic or alkaline solution. Compromises have to be made when it comes to the question of electrochemical kinetics and energy density. Not all readily soluble redox compounds inorganic nature show sufficient electrochemical electrode kinetics. The present invention also shows ways of avoiding these kinetically related disadvantages.

Numerous inorganic, small-molecule redox systems have been explored in order to reach higher energy densities. However, many of these systems are based on strongly acidic or strongly alkaline redox electrolytes, which are toxic, expensive and harmful to the environment, making disposal and acceptance difficult. With all of the innovations described and/or patented up to now, one is forced to make compromises if one wants to achieve maximum energy density at a favorable price and with maximum electrochemical kinetics. For this reason, attempts to circumvent the limited solubility of redox-active compounds based on metals or metal ions by using organic-based metal-free redox systems have recently been increasingly described. Purely organic redox compounds have been increasingly used in redox flow batteries since 2011. For example, low molecular weight 2,2,6,6-tetramethyl-piperidinyloxyl (TEMPO) and N-methylphthalimide were used in a traditional redox flow battery with an ion-conducting separating membrane [Z. Li, Li S, Liu SQ, Huang KL, Fang D, Wang FC, Peng S Electrochem. Solid State Lett. 14, 2011, A171-A73]. Pyrazine-based cyanoazacarbons have been reported in US 8,080,327 B1 used both as an anolyte and as a catholyte, with the electrode spaces separating ion-conducting membranes based on low-impedance cation exchangers and anion exchangers. In 2014, Yang et al. the first fully organic aqueous redox flow battery. In this, the water-soluble 1,2-benzoquinone-3,5-disulfonic acid was used as the organic catholyte and anthraquinone-2-sulfonic acid and anthraquinone-2,6-disulfonic acid were used as the organic anolyte [B. Yang, L. Hoober-Burkhardt, F. Wang, GKS Prakash, SR Narayanan, J. Electrochem. society 2014, 161, A1371-A1380.128]. Water as a solvent offers a number of advantages over organic solvents: it is non-flammable, inexpensive and allows high ionic mobility, which leads to lower ohmic internal resistances. The greatest advantage of purely organic redox systems, however, is that the synthesizer can work towards a desired standard redox potential. A 2017 reference [J. Winsberg, T Hagemann, T Janoschka, MD Hager, US Schubert; Angew.Chem. 2017, 129, 702-729] shows examples of successful and stable organic redox compounds whose redox potential in aqueous solution can be varied between -1.39 V and +1.34 V compared to a standard hydrogen electrode. The physicochemically limited solubility in a suitable solvent is also problematic in the case of anolytes and catholytes consisting entirely of organic redox systems, which ultimately limits the energy density. The energy density depends on the number of redox active groups per unit volume. The redox group density can in turn be increased by synthetically incorporating several of these groups per molecule. Then there are macromolecules with multi-redox centers. which then have a correspondingly higher energy density, whereby the molecular accessibility for an electron exchange with a redox partner or an electrode surface can become important, which in turn can be problematic with the electron exchange resins..

Polymeric redox systems are of particular importance with regard to the present invention, because the polymerization creates many active redox centers within a macromolecule (in a very small space, so to speak), which increases the stoichiometric energy density. The synthesis of oxidation-reduction polymers, which are also called redox polymers or electron exchange polymers or electron-transferring polymers, has been intensively researched since the middle of the last century [G. Manecke "Oxidation-Reduction Polymers" in Macromolecular Microsymposia-XII and XIII, Prague, 1973, Book 1974, 181-199, Pure and Applied Chemistry, Vol. 38, Nos. 1-2 (1974); https://doi.org/10.1016/B978-0-408-70639-1.50014-1]. Numerous oxidation and reduction processes of organic or inorganic substances both in an aqueous environment and in other solvents were described at that time. Such redox polymers in resin form were at that time - analogous to ion exchange resins - packed in columns (columns) - called electron exchange columns - in order to carry out quantitative oxidations or reductions without introducing another substance (as an impurity) into the product. In this way, contamination by the redox partner converted in the process can be avoided. In the reference "Synthesis, Electrochemistry and Application of some Oxidation Reduction Polymers", in "Charged Gels and Membranes", NATO Adv. Study Inst. Charged React. polym. 2; Forges Les Eaux; 1973; D Dordrecht; THERE. reidel; 1976; volume 2; pp. 173-189, the author Georg Manecke describes the synthesis of redox resins, which in this case are produced by classic methods of macromolecular chemistry, such as polymerization, condensation, polymer-analogous reactions. These are often products with a vinylic structure, obtained by polyaddition and not by oxidation. Most of the redox polymers proposed at the time are either vinyl polymers onto which quinone molecules have been grafted, or formaldehyde + phenol + hydroquinone condensation polymers. As a convincing example of quantitative electrochemical transformations The reduction of Fe ³⁺ on a polypyrrole electron exchange column was described using such electron exchange columns with redox resins in the particle diameter range of 0.2-0.5 mm, with a yield of 99%. As a further example, the oxidation of tin(II) chloride using a polyacene-benzoquinone electron exchange column was described, with a yield of 98.5% being able to be achieved. In this context, it is important that the degree of oxidation or reduction of these electron exchangers after exhaustion of the electron exchanger column (breakthrough of unreacted starting material - analogous to ion exchanger columns - by passing through dissolved oxidizing or reducing agents suitable redox potential could be completely regenerated [Myer Ezrin and HG Cassidy, Annals New York Academy of Sciences, 1749-6632 (1953)].The regenerability of such redox polymers is therefore well known and is also the basis of the present invention 1976 (DE 2625443 C2), for example, an electrochemical regeneration of redox exchange resins is described. In addition, various indigo sulfonates were listed as another example of planned syntheses towards a targeted standard redox potential [N. Grubhofer, Naturwiss. 12 , 557 (1955)].

The redox resins obtainable by polymerisation are insoluble, limitedly swellable high polymers which have reversible redox systems. Depending on whether they are in reduced or oxidized form, they act as regenerable, insoluble reducing or oxidizing agents. In the studies at that time, however, the focus was not on the amounts of substance converted per unit of time, which would be important for use in redox flow batteries, but only on reversible, quantitative oxidation or reduction without the introduction of another substance that contaminates the product. At that time, the compound to be reduced or oxidized was allowed to run through the relevant electron exchange column in dissolved form, driven only by gravity. This resulted in a throughput of a few milliliters per minute, which is too little for a medium-power redox flow battery. The use of such known electron exchange columns in connection with redox flow batteries has hitherto been prevented by the immense pressure build-up when flowing through the columns filled with an insoluble, granular material with a diameter of only a few micrometers. The pumping power required for this reduces the energy yield of a redox flow battery considerably.

In connection with the constantly newly synthesized organic redox systems and the sharp increase in the number of publications on this subject, the interposition of a so-called redox mediator for "energy transfer" between a main redox system in the storage tank and of the electrochemical cell [WO 2013/012391 A1 and provisional US application US 2014/0178735 A1 of Jun. 26, 2014 (Pub. Date)]. However, these patent applications do not disclose how the unavoidable hydraulic pressure build-up when the redox mediator solution flows through large layers of granular solid material (organic polymer resins with redox-active groups) can be avoided. For this purpose, high-pressure pumps must be used, similar to what is known as high-pressure liquid chromatography, which - because of its energy requirements - allows the efficiency of such a "mediator-supported" redox flow battery to approach zero.

The use of redox mediators as “electron shuttles” has been well known in biosensors for over 30 years [page 534 in K. Cammann, U. Lemke, A. Rohen, J. Sander, H. Wilken, B. Winter in “ Chemo- and Biosensors - Basics and Applications”, Angew. Chem. 103 (1991) 519-541]. Because of these hydraulic problems, mediator-assisted redox flow batteries with a solid, granular main redox system have not been commercialized to date.

Organic redox polymers can be used both in aqueous systems [T. Janoschka, N Martin, U Martin, C Friebe, S Morgenstern, H Hiller, MD Hager & U S Schubert "An aqueous, polymer-based redox-flow battery using noncorrosive, safe, and low-cost materials", NATURE, Vol. 527, 5 November 2015, p. 78-81] as well as in non-aqueous solvents, allowing even higher voltages in the latter case because water electrolysis cannot occur.

In WO2014026728 A1 from February 20th, 2014, new, very promising water-soluble redox polymer compounds for anolyte and catholyte are described, in which the separating membrane in the electrochemical flow cells consists of inexpensive size exclusion membranes (e.g. dialysis membranes). They allow the smaller, non-redox-active ions of the electrolyte to pass through more easily and therefore lead to a very advantageous lower internal resistance of the battery. When choosing a suitable "cut-off", they avoid the passage of the dissolved redox polymers so well that more than 10 ⁴ discharge and charge cycles became possible.

With regard to the present invention, the dissolved, high molecular weight, redox-active components used can be any compounds, provided they can be present in at least two different stable oxidation states and have a molecular weight such that they cannot pass through the size exclusion membrane used. Compared to solid electron exchange resins, dissolved redox polymers have the great advantage that the redox groups are more accessible for electron transfer. A previously unsolved problem with highly concentrated organic redox polymer electrolytes is the high viscosity, which firstly results in a high loss of effectiveness due to increased pump energy and secondly in poorer mass transport (high diffusion overvoltages) in the two half-cells. Because the viscosity of anolyte and catholyte increases sharply with the degree of polymerization and the increase in concentration, it has so far not been possible to achieve outstanding energy densities on this basis. The inventors report an energy density of about 10 Wh/L, although theoretically (redox equivalents) a much higher one should be possible. The selected examples illustrate the urgent need for functional technical solutions to solve the problems described in the most important developments in the field of organic redox polymers. The object of the invention is to make the large number of novel redox polymers with high energy density generally usable in redox flow batteries, for which purpose the special regeneration device disclosed is intended to serve.

So-called semi-redox flow batteries (“Cambridge Crude”) have also been proposed in the literature, in which an inorganic suspension of the electro-active lithium batteries together with graphite powder in an organic solvent is conducted past the electrodes, which reduces the energy density of Redox flow batteries should increase almost tenfold [M. Duduta, B Ho, VC Wood, P Limthongkul, VE Brunini, WC Carter, Y Chiang Adv Energy Mater. XX, 1-6 (2011)]. However, the conversion into a product was delayed, presumably because abrasion losses and particle size changes occur during long-term pumping of suspensions, which in practice lead to capacity-reducing deposits or blockages.

In WO 2013/012391 A1 and a US provisional application US 2014/0178735 A1 from June 26, 2014 (Pub. Date) describes a redox flow battery which uses redox mediators, which are already known from biosensors. However, the latter are always used together with the capacity-providing main redox systems in a jointly circulated electrolyte solution. Lithium compounds are preferably mentioned as the most important electro-active compounds. An example of the state of the art in so-called mediator-supported redox flow batteries can be found in: Feng Pan and Qing Wang [Molecules 2015, 20, 20499-20517; doi:10.3390/ molecules 201119711]. the 5 in this publication clearly shows that a mediator-electrolyte has to be pressed through the solid-particle layer in direct molecular contact. Likewise in US 2014/0178735 A1 , Pub. Date: Jun. 26, 2014 there is no indication that the redox mediators used must not come into direct molecular contact in order to exploit the numerous advantages of this arrangement. Suspensions of electro-active compounds with special redox mediators are also described here. The selection of the compounds to be used, electro-active components, mediators and electrolyte salts is closely defined in this application. No further details are given about a particularly advantageous structural arrangement which prevents sedimentation of the suspension or clogging and clogging of the lines. Precautions that lead to the battery voltage remaining constant over a longer discharge time are also not mentioned. The present invention wants to eliminate all these disadvantages and discloses more advantageous arrangements also for future developments in the field of redox polymers and mediators. The use of redox mediators as “electron shuttles” is not new due to numerous previous publications, especially for biosensors, and is not the core of the present invention.

A large number of redox polymers have now been produced. Targeted syntheses of these redox resins, which are familiar to every specialist, allow redox exchange columns with the desired properties for use in redox flow batteries, such as different standard redox potentials and high energy densities, to be easily and inexpensively created manufacture. However, this is not the subject of the invention, which is limited to the apparatus arrangement of the storage containers. Relevant reviews underline the achievable high energy densities [P. Novák, K. Müller, KSV Santhanam, O. Haas, Chem. Rev., 97, 207-281 (1997)]. Nonylbenzohexaquinone, for example, has a theoretical storage capacity of 488 Ah/kg. Benzoquinone, for example, has a theoretical power density of 1400 Wh/kg based on its redox potential of 2.8 V in non-aqueous media and its theoretical capacity of 496 Ah/kg. It is thus higher than with the usual inorganic materials of corresponding Li batteries [Z. Song, H. Zhou, Energy Environment. science 6, 2280-2301 (2013)]. A highly concentrated, highly viscous solution of such a redox system could theoretically be used in e-mobility leads to car ranges of around 800 km per 100 liters of redox main systems. With appropriate standardization, such discharged organic main redox systems can be exchanged for charged ones in a few minutes at new filling stations. The discharged ones are then recharged on site using alternative energies and are again available for "filling up". By means of such energy-dense, water-soluble redox polymers, environmentally friendly redox flow batteries with inexpensive raw materials that are not becoming scarce could also be used non-stationarily for the first time. Lengthy charging times, as required by Li-based batteries, can thus be avoided. Only two "discharged", highly viscous liquids have to be exchanged for "charged" ones at the same time. Despite their ideal electrochemical requirements (high energy density), the novel organic redox polymers have not been used in redox flow batteries to date because the energy losses when pumping anolyte and catholyte through a column filled with these materials are exorbitantly high. These caveats are changed by the invention disclosed herein.

The present invention primarily solves the efficiency problem caused by pump energy losses that are too high if mediators are generally used between a main redox system that determines the storage capacity and the electrochemical flow cells. The technical solution to the problems outlined above is based on the electrochemical principle of internal electrolysis, but with a preferred novel bi-polar structure, and leads to the possibility of being able to regenerate the anolytes and catholytes of many different redox flow batteries, which leads to a greatly increased energy density. The decisive difference to well-known mediator-supported redox flow batteries, the small-volume redox mediator electrolytes as an "electron shuttle" between the storage capacity-determining main redox systems in the storage tanks and the electrochemical flow cell , or the "battery cell stack" is to avoid a high hydraulic pressure build-up and a larger molecular material contact of the redox systems in anolyte and catholyte when flowing through the corresponding main redox systems. The invention solves these problems by means of a novel, special device (apparatus arrangement) of the bipolar internal electrolysis disclosed here.

The problem of an excessively high hydraulic pressure build-up is solved by a hydraulically optimized line or flow channel-bound “flow” through the main redox systems involved, which are preferably mounted in columns. A hydraulically advantageous liquid flow through the solid or highly viscous main redox systems, which are stationary in the storage tanks, is new. The process of bi-polar internal electrolysis according to the invention requires electrically conductive walls (only electrons are conductive). The invention can advantageously be combined with a further innovation in that the passage of anolyte and catholyte through the storage capacity-determining main redox systems with the high energy density is carried out in such a way that the redox reactions between the two remain locally limited and runs in layers or zones in the direction of flow through a preferably columnar device (storage tank). For this purpose, according to the invention, the electrically conductive wall of the anolyte and catholyte flow channels is interrupted several times in the flow direction in an insulating manner. In addition, the outer flow channel wall surfaces are preferably only brought into contact with surface-enlarging electrical conductors perpendicular to the direction of flow. Under such conditions, anolyte and catholyte leave storage tanks constructed in this way at suitably selected flow speeds until they are completely exhausted (complete discharge of the entire main redox system) always almost fully regenerated when the battery is discharged and electrochemically almost fully converted when the main redox systems are recharged .

The invention discloses at least two possibilities for avoiding such an efficiency-reducing pressure build-up in preferably column-shaped storage tanks. In a preferred arrangement of the equipment, the direct material molecular contact between the two anolyte and catholyte solutions pumped around and the corresponding main redox systems in the storage tanks is largely achieved by a cable sheathing (metallic pipe or electrically conductive flexible hose) that only conducts electrons avoided. According to the invention, this hydraulically optimized anolyte or catholyte flow channel with electron-conducting walls is interrupted several times by insulating sections during the constructive passage through the corresponding electron exchange resins in order to avoid a continuous equipotential surface. In this way, the last section of the flow channel before the exiting anolytes or catholytes are transported further into the electrochemical cell until the so-called breakthrough (complete exhaustion of the amount of energy stored in the main redox system) always has the same, constant redox -Potential that of a charged main redox system, or that of the not yet discharged last column section. A chemically required reorientation of non-redox-active ions and closure of the internal electrolysis circuit is accomplished by electrolyte bridges or current keys, which is described in more detail in the examples. It is advantageous to use the necessary isolating interruptions in the flow channel line as an electrolyte bridge. They can have an ion-selective membrane coating at these points (see Fig. 5 ). As a result, only the electrochemically inactive ions (e.g. OH- or H+ or other ions such as ligands, etc.) that close the circuit purely electrolytically during the internal electrolysis can pass through there.

To explain how it works: The quantitatively (stoichiometrically) much larger amount of the storage capacity-determining main redox system in the external exchange column (storage tank as a regenerator device) characterizes the electron-conductive envelope (flow channel wall) of the relevant , the anolyte or catholyte flowing through this main redox system has its characteristic and electrochemically appropriate redox potential (same or slightly more negative for the anolyte or slightly more positive for the standard redox potential for the catholyte) and the relevant coating (metallic or more conductive, more flexible tube) transports the electrons further to the redox system dissolved in the catholyte in the flow channel in the case of reducing properties and in the opposite direction to the redox system dissolved in the anolyte in the case of oxidizing properties. From an electrochemical point of view, the conditions are analogous to those of the tried and tested so-called "internal electrolysis". The electrically conductive areas of the flow channel walls - separated from each other by insulating sections - become longer and longer towards the end of the flow in order to increase the electrochemically active inner surface and thus to account for the exponential depletion of reactive redox species in the anolyte and catholyte solution wear. As a further measure to avoid concentration polarization, the relative flow velocity can be increased by narrowing the cross section, which reduces the Nernstian diffusion layer thickness. Under optimal conditions, anolyte and catholyte always leave the storage tanks designed according to the invention almost quantitatively regenerated until the two main redox systems are completely discharged, which leads to a very advantageous constant battery voltage during discharge. Conversely, when the main redox systems are recharged, the anolyte and catholyte always emerge almost quantitatively discharged from such a regeneration device until the relevant main redox systems are fully charged and remain with the galvanostatic charging characteristics of the redox flow battery cells the charging voltage is constant.

The charging and discharging processes can be easily measured by simple redox potential measurements using so-called platinum single-rod measuring chains at the entrance and exit of the relevant flow channel line. If, for example, the redox reaction under consideration is one with only one transfer electron, both platinum electrodes would show a voltage difference of approx. 120 mV with a 99% redox conversion. In order to make these two electrochemical processes (discharging and recharging of anolyte and catholyte) quantitative in a single flow through the device, the flow rates are adjusted accordingly empirically, using the redox potential measurements for control. Depending on the efficiency (electrochemical yield with a single column flow), the flow, diameter and number of flow channels as well as the total electrochemically active length are also determined empirically. Snake-shaped constructions analogous to a snake cooler in distillation have proven themselves; but many other shapes (eg, spiral, meander, etc.) that provide the required inner surface are also possible. To enlarge the inner surface and to compensate for mass transport limitations in anolyte and catholyte lines, various measures that are familiar to the person skilled in the art can be taken. For example, the free diameter of these lines can be filled with additional conductive intermediate walls and thus multi-chamber lines/multi-lumen hoses can be obtained. Corrugation of the inner surface in the longitudinal direction is also hydrodynamically advantageous. In order to avoid limitations due to mass transport effects on the outside of these mediator-electrolyte lines - since no convection is possible there - this area is covered according to the invention with additional electronically conductive particles, nets or other materials (e.g. graphite felt, graphene paper, -threads or fabric) additionally filled. For this purpose, the mediator lines can also be brought into the columns with metal spikes or devices similar to wire brushes. It is often sufficient if the outer surface - which is in direct electrical contact with the inner flow channel wall - is larger by a factor of >100 in order to compensate for the diffusion limitation there. the 3 indicates this surface enlargement schematically: on the one hand, metallic brushes with spikes perpendicular to the direction of flow of anolyte and catholyte and, as a further example, a corrugated metallic sheet serve to allow the electron transfer from the main redox system to these conductors to proceed preferably only perpendicular to the direction of flow . As a result, when the anolyte and catholyte flow through, there is a layered conversion similar to that in ion exchange columns.

Such a transport of the anolyte and catholyte solutions to be regenerated within electrically conductive pipelines that are materially sealed against redox-active substances - without larger hydraulic ones Restrictions - enables completely new electrochemical combinations. The arrangement according to the invention allows, inter alia, a problem-free change of solvent between anolyte and catholyte and the corresponding main redox system. For example, the anolyte and catholyte cycle can run on a non-aqueous basis in order to avoid efficiency-reducing side reactions (e.g. evolution of hydrogen or oxygen) at the collector electrodes, which enables higher clamping voltages than approx. 1.2 volts. In this way, higher terminal voltages can be achieved with non-aqueous electrolytes without having to use large amounts of low-viscosity and highly flammable organic solvents as the capacity-determining redox system, as is the case with a traditional redox flow battery design. Here, the main redox system can consist, for example, of a flame-retardant organic gel (e.g. diphenyl ether). The diffusion problems of the redox centers in the macromolecules then correspond to those in lithium polymer batteries, which function well even without mechanical electrolyte movement. The invention also allows the stable construction of devices that are in the electrochemical research field "Interface of two immiscible electrolytes (ITIES). It has already been shown [D. Henn, K. Cammann, Electroanalysis 2000, 12, no. 16, 1263-1271 (2000)] that these interfaces can be stabilized, for example, by dialysis membranes or membranes with small pores (eg diaphragms). These interfaces between two immiscible electrolyte solutions behave electrochemically like metal electrodes, ie one can control the transfer of charge carriers in both directions by means of an externally applied voltage difference at these phase boundaries. It is also possible to create an electrolytic "current key" with a solvent between two diaphragms in contact with the anolyte and catholyte that is immiscible with that of the anolyte and catholyte solution. In order to increase the conductivity, the known lipophilic salts are preferably used in this non-aqueous solution, which, because of their distribution coefficient, do not contaminate aqueous anolyte and catholyte solutions. In this case, selective complexing agents for non-redox-active ions to be transferred are used for the necessary charge transport in the case of electrolyte bridges.

Another advantage of the novel apparatus arrangement disclosed here is the unrestricted use of newer organic redox polymers with very high energy density. Organic redox solids (electron exchange resins) and highly viscous water-soluble redox polymers can be produced with molar concentrations well over 20 molar. This means that redox flow batteries on a more environmentally friendly basis and made from inexpensive, cheap organic precursors can even have higher energy densities than lithium-based batteries.

Solid, insoluble redox resins require a special embodiment of the process. You need a redox mediator in order to be able to exchange the electrons with the electrical conductors that are in contact with the outer conductive flow channel walls. The relevant redox monomer dissolved in a solvent can serve for this purpose. For example, when the electron exchange resins were first synthesized, the redox potential of the solid redox resin was measured [G. Manecke, Z.EI.Ch, 57, 89-94 (1953)]. This mediator solution only needs to wet the solid redox particles and the embedded electrical conductors.

It is particularly advantageous if the redox systems in the anolyte and catholyte, which are electrochemically adapted to the main redox system, take on this electron “shuttle” task. Then, according to the invention, the flow channel walls are used in an electrically conductive but porous manner, with pore sizes that are smaller than the diameters of the smallest redox polymer resins. This embodiment of the bi-polar internal electrolysis also requires areas of the electrically conductive flow channel walls that are insulated from one another in order to guarantee layered regeneration perpendicular to the direction of flow. but no additional electrolyte bridge. This is a constructive advantage.

Similarly advantageous embodiments are possible with the dissolved redox polymers as the main redox system if novel, electronically conductive size exclusion membrane hoses are used as the flow channel, which are guided through the storage tanks through small electrically insulating areas in the millimeter range. Here, the polymers identical to the relevant main redox system can be used as anolyte and catholyte in a much more dilute solution. In the longer term, it is better to design the collector-electrode surfaces in the battery larger when the current decreases more than to accept an increase in viscosity when the concentration increases. This construction of the regeneration device also has the structural advantage. no need to install separate size exclusion membranes in terms of production technology. However, by adding non-redox-active salts to the anolyte and catholyte, the same osmolarity must be set as that of the relevant main redox system.

By varying the surfaces of the electronic conductors outside the flow channels, the different diffusion and kinetics-related limitations are adjusted empirically. For example, a different electrochemical phase boundary kinetics and a different mass transport behavior of the redox systems in anolyte and catholyte can be compensated with the relevant main redox system in the storage tank by correspondingly larger surfaces of the outer electron conductors. The storage tank does not always have to have a columnar or barrel-like shape; Depending on the place of use, rectangular, flat tanks are also possible, for example in front of room walls. Through a constructive parallelism of many anolyte and catholyte flow channels with subsequent combination for transport through the power-generating electrochemical battery cells (see Fig. 3 ) the redox equivalents per time unit required to generate a certain current (battery power) can be easily adjusted.

According to the invention, the electrochemical arrangement for the construction of the relevant electron exchange column based on redox polymer resins is as follows: main redox electrolyte in insoluble resin form / liquid, stationary redox mediator / solid electronic conductor (conductive flow channel -wall) / anolyte or catholyte in the flow channel. In the case of liquid or highly viscous main redox systems, there is no need for an additional liquid mediator. It is advantageous for the flow channels to use known electrically conductive hoses, eg plastic hoses, the walls of which contain electrically conductive particles, e.g. B. in the form that the rubber mixtures forming their walls contain soot powder or mixed carbon short fibers or hoses, in the walls of which metallic conductors, such as metal threads, wires or strands, are incorporated [G. Batrinescu, LA Constantin, A. Cuciureanu, MA Constantin "Conductive Polymer-Based Membranes", InTech http://dx.doi.org/10.5772/63560]. Particularly good electrical conductivity is obtained through the incorporation of modern tear-resistant carbon threads, some of which have an electrical conductivity comparable to copper ( EP 0 882 924 B1 ). They can also be used excellently as external electrical conductors transverse to the direction of flow.

Another advantage of this invention is that all other components of a conventional redox flow battery (electrochemical flow cell or cell "stacks") as well as the pumping mechanism and the type of chemical species used) are not affected by a storage capacity increase must be affected, so the invention can represent an addition or capacity expansion of known redox flow batteries. This is possible without contamination of the anolyte and catholyte if the latter, while passing through the column, contains the main redox systems other than the exchange of electrons across the electrically conductive flow channel walls and the exchange of non-redox-active ions through an ion-selective electrolyte bridge have no physical contact.

The invention is of particular economic importance because it lays the foundation for a completely new business model when the novel storage tanks according to the invention are electrochemically connected to existing redox systems with regard to the standard redox potential of the energy-dense, organic main redox systems used. Flow batteries can be adapted to a wide variety of material bases. The manufacturer of these new "electrochemical in-situ regenerators" of anolyte and catholyte solution from conventional redox flow batteries can concentrate fully on optimizing the connected, newly designed storage tanks, filled with new redox systems with high energy density. The material separation from the chemistry of the conventional redox flow battery in question, which is as extensive as possible according to the invention, offers significantly more design freedom. Problems with the specific electrochemistry of the anolyte and catholyte solutions (now acting as redox mediators) in the electrochemical flow cell stacks are avoided since the anolyte and catholyte are not contaminated by foreign matter. The design task is limited to ensuring that the various anolytes and catholytes in question at a given concentration, the regeneration unit according to the invention until the final discharge (quantitative conversion of the entire main redox system amount) or charging of this unit either in the state fully charged or fully discharged before passing through the battery cells. This can easily be regulated by the contact time of the relevant redox systems in anolyte and catholyte with the electrochemically corresponding main redox system in the regeneration unit. Due to the possibility of a complete material separation, the novel apparatus structure of the storage tanks also enables combinations of inorganic and organic redox systems, which was previously difficult to achieve.

A regeneration behavior analogous to the operation of ion exchange columns is achieved in that the electrochemical redox reaction of the anolyte and catholyte with the electrochemically corresponding main redox system only within a localized expansion in the main redox system, preferably perpendicular to the Flow direction of anolyte and catholyte (layered) takes place. The flow channel wall areas, which are electrically isolated from one another, serve this purpose, as does a preferred one Alignment of the outer electrical conductors perpendicular to the direction of flow and, if necessary, additional insulating partitions in the main redox systems (see Fig. 6 ). The recharging (regeneration process) and discharging of the anolyte and catholyte solutions takes place in layers as they flow through the storage tanks with the two main redox systems until almost complete conversion occurs at the outlet. This automatically leads to a particularly advantageous, constant charging and discharging voltage of the relevant redox flow battery. This represents a novelty and offers special advantages for connected consumers. Because of the principle, recharging also takes place at a constant voltage until the main redox systems are fully charged, side reactions that reduce efficiency (e.g. water electrolysis), which can occur with classic redox flow battery designs, are prevented.

The aim of producing regeneration devices for any redox flow batteries with a wide variety of material electrochemistry is achieved according to the invention in that there is significant direct material (molecular) contact between the redox systems contained in the anolyte and catholyte and the relevant electrochemically adapted main redox Systems in the storage tanks is largely avoided and preferably only electronic contact (electron shuttle) is enabled via the electrically conductive flow channel walls. Only through these innovative, constructive, technical measures can such mediator (or redox shuttle) supported redox flow batteries be used in practice without enormous pumping energy losses. In addition to this decisive, efficiency-increasing, hydraulic advantage, the device according to the invention has further technological advantages, which are explained further below.

The present invention does not describe or preferably describe any specific type of redox flow battery (on a purely inorganic, metal-organic or purely organic redox system basis). The arrangement disclosed herein provides advantages for all previously described redox flow batteries regardless of the particular redox chemistry employed. It is also independent of the type of electrochemical flow cell. The wide range of applications is an advantage in terms of construction, because it can then be manufactured in larger quantities and in a modular manner (see Fig. 6 ).

1. a) drastische Erhöhung der Energiedichte (Faktor > 10) in den Haupt-Redox-Elektrolyt-Reservoirs (hier Elektronenaustauscher-Säulen auf Basis von Redox-Polymeren in fester, unlöslicher Harz-Form oder gelöst in Form von hochviskosen Lösungen), verglichen zum Stand der Technik bei Redox-Flow-Batterien;
2. b) stabile Batterie-Klemm-Spannung im Verlauf des Entlade-Vorgangs und bei der Wiederaufladung durch Einhaltung von Bedingungen analog denen von lonen-Austauscher-Säulen in der Regenerationsvorrichtung;
3. c) Trennung der wichtigen elektrochemischen Elektroden-Kinetik (Austauschstromdichte pro Kollektor-Elektroden-Oberfläche) von den elektrochemisch-stofflichen Eigenschaften der Haupt-Redox-Systemen, die die Energiedichte bestimmen;
4. d) Weitgehende Vermeidung eines direkten stofflichen Kontaktes - falls nötig - von Anolyt und Katholyt mit dem Speicher-Kapazitäts-liefernden Haupt-RedoxSystemen in den speziellen Vorratsbehältern, was den konstruktiven und planerischen Spielraum erhöht;
5. e) Ermöglichung von nicht-wässrigen Anolyten und Katholyten zur Erzielung höherer Klemm-Spannungen (Leistungen) bei verminderter Verwendung brennbarer organischer Lösungsmittel nur für die klein-volumigen Mediator-Kreisflüsse niedriger Viskosität, während als Hauptspeicher Lösungsmittel in den Vorratstanks ein organisches, schwer entflammbares, hoch-viskoses verwendet werden kann;
6. f) Einfache Ermöglichung eines Wechsels der Lösungsmittel (wässrig - nichtwässrig oder umgekehrt) für den Anolyt und Katholyt-Kreislauf und die Haupt-Redox-Systeme;
7. g) Eine direkte molekulare, stoffliche Trennung der Reaktionspartner in der Regenerationssäule ermöglicht Elektronenaustauscher-Säulen, die bei völlig unterschiedlichen pH-Werten arbeiten, was die Verwendung ein und derselben Redox-Verbindung - pH-mäßig auf das optimale Standard-Redox-Potential eingestellt - in Anolyt- und Katholyt-Tank ermöglicht;
8. h) Basis einer neuen Geschäftsidee als in situ Anolyt - und Katholyt - Regenerator bei vielen klassischen Redox-Flow-Batterien unterschiedlichster Elektrochemie (Redox-Systeme) mit dem Vorzug: höhere Energiedichte, modulare Bauweise mit leichter Veränderung der Speicher-Kapazität;
9. i) Die Kosten für die Chemikalien fallen weniger ins Gewicht, weil das Volumen der umgepumpten Anolyt- und Katholyt-Lösungen und die darin gelöste Menge an elektrochemisch besonders reversiblen Redoxsystemen nur einen kleinen Bruchteil (ca. 1/10 bis 1/100.000) der bisher in den Vorratsbehältern gespeicherten Redox-Chemikalien ausmacht. Als Anolyt und Katholyt Redox-Systeme können daher auch teurere anorganische Übergangsmetalle, metallorganische Verbindungen (wie beispielsweise Ferrocen) mit hoher Austauschstromdichte an den Kollektor-Elektroden Einsatz finden. Auch können Edelmetalle bei den Kollektor-Elektroden in der Durchfluss-Zelle verwendet werden, wenn die Mediatorlösung insgesamt nicht aggressiv ist und Membran und Zelle nicht angegriffen wird;
10. j) Bei den großen Mengen an Haupt-Redox-Systemen in den Vorratstanks (Elektronenaustauscher-Säulen) hat der organische Chemiker eine große Freiheit, nach besonders preiswerten und stabilen Verbindungen mit maximal unterschiedlichem Standard-Redox-Potential zu suchen und die Frage einer ausreichenden Elektrodenkinetik weniger Bedeutung zuzumessen, da die größere Oberfläche der elektrischen Leiter in Kontakt mit den elektrisch leitfähigen äußeren Fließkanal-Wandungen im Vergleich zur elektrochemisch aktiven inneren Fließkanal-Oberfläche in dieser Hinsicht kompensierend wirkt.

Summary: The present invention thus avoids all known disadvantages of classic redox flow batteries, which have been mentioned in more detail above, and in contrast has the following advantages:

1. a) Drastic increase in energy density (factor > 10) in the main redox electrolyte reservoirs (here electron exchange columns based on redox polymers in solid, insoluble resin form or dissolved in the form of highly viscous solutions) compared to the status quo the technology of redox flow batteries;
2. b) stable battery terminal voltage during the discharging process and during recharging by maintaining conditions analogous to those of ion exchange columns in the regeneration device;
3. c) Separation of the important electrochemical electrode kinetics (exchange current density per collector-electrode surface) from the electrochemical-material properties of the main redox systems, which determine the energy density;
4. d) Extensive avoidance of direct material contact - if necessary - of anolyte and catholyte with the storage capacity-delivering main redox systems in the special storage tanks, which increases the constructive and planning scope;
5. e) Enabling non-aqueous anolytes and catholytes to achieve higher clamping voltages (powers) with reduced use of flammable organic solvents only for the small-volume, low-viscosity mediator circuit flows, while the main storage solvent in the storage tanks is an organic, flame-retardant, high-viscosity can be used;
6. f) Easily enabling a change of solvents (aqueous - non-aqueous or vice versa) for the anolyte and catholyte circuit and the main redox systems;
7. g) A direct molecular, material separation of the reaction partners in the regeneration column enables electron exchange columns that work at completely different pH values, which means that one and the same redox compound can be used - pH-wise adjusted to the optimal standard redox potential - Allowed in anolyte and catholyte tank;
8. h) Basis of a new business idea as an in situ anolyte - and catholyte - regenerator for many classic redox flow batteries of different electrochemistry (redox systems) with the advantage: higher energy density, modular design with easy change of the storage capacity;
9. i) The costs for the chemicals are less significant because the volume of the anolyte and catholyte solutions pumped around and the amount of electrochemically particularly reversible redox systems dissolved therein is only a small fraction (approx. 1/10 to 1/100,000) of the previously redox chemicals stored in the reservoirs. Expensive inorganic transition metals and organometallic compounds (such as ferrocene, for example) with a high exchange current density at the collector electrodes can therefore also be used as anolyte and catholyte redox systems. Precious metals can also be used for the collector electrodes in the flow cell if the mediator solution is not aggressive overall and the membrane and cell are not attacked;
10. j) With the large amounts of main redox systems in the storage tanks (electron exchange columns), the organic chemist has a great deal of freedom to look for particularly inexpensive and stable compounds with a maximum of different standard redox potentials and the question of sufficient electrode kinetics less importance, since the larger surface area of the electrical conductors in contact with the electrically conductive outer flow channel walls compared to the electrochemically active inner flow channel surface has a compensating effect in this respect.

A further advantageous embodiment results from an implementation of the in 9 shown principle. This variant of the new process uses a total of at least six redox systems with different standard redox potentials. The mode of action is based on the fact that, to compensate for concentration and transition overvoltages, the anolytes and catholytes to be regenerated in the device according to the invention contain at least one additional, further redox system that has a more negative or more positive standard redox potential than that of the relevant main redox system. For example, using a traditional vanadium redox flow battery 9 considered, where the relevant standard redox potentials of anolyte and catholyte are marked with V(2/3) and V(4/5), the main redox systems to be used for regeneration should in the simplest case be exactly this redox - match potentials. A 99% charged main redox system for the anolyte V(2/3) has a redox potential of approx. 120 mV more negative than the standard redox potential (with [Ox] = [Red] in the Nernst equation) on. If, for example, only 50% of the anolyte has been reacted after flowing through the battery cells, then it flows through electrodes in the regeneration device according to the invention, which have a redox potential that is approx. 120 mV more negative and thus the trivalent vanadium ion reduce again to divalent. Since it constantly passes new flow channel sections with this negative overvoltage in the course of its flow through the main redox system until the main redox system is completely discharged, it is quantitatively regenerated. Then it shows the same redox potential as the main redox system in question. The same applies to the catholyte and the corresponding main redox system with a redox potential of V(4/5), which is approx. 120 mV more positive than the standard redox potential of V even in the state of 99% full charge (4/5) is.

If the overvoltages are not to collapse so quickly towards the end of the capacity of the main redox systems, the main redox systems that are suitable for the anolyte and catholyte are selected for the anolyte in such a way that their standard redox potential is more negative than that of the V(2 /3) system and which is more positive for the regeneration of the catholyte. The relevant standard redox potentials of the two main redox systems are in for this case 9 indicated by the arrows 3 and 3'. In this case, when these main redox systems are recharged, further redox systems are required whose standard redox potential should be around 2 or 2'. If they are chemically compatible, such auxiliary redox systems can also be dissolved in anolyte and catholyte or can be routed through the main redox systems by means of a separate flow channel (eg multi-lumen hose). A particularly advantageous example is described below. A regeneration device that also works as an anolyte and catholyte with water-soluble organic redox polymers results in very advantageous combination and construction options.

The following examples are only intended to illustrate the novel principle of bipolar internal electrolysis according to the invention. The principle together with the specification of a chromatographic working method (or analogous to an ion exchange column) can be implemented in a wide variety of flow channel guides through the reservoir with the main redox system and shapes. All possible variations cannot be fully listed here. Due to the technical novelty of the bi-polar internal electrolysis, the principle and the first measurement results should also be described in more detail in the following examples.

Example 1: Description of the "internal electrolysis" principle

The functional principle of the regeneration device according to the invention is to be explained with the help of a historical experimental setup for "internal electrolysis" in 1 to be shown. On a platinum mesh (1), an outer copper sulphate solution (5) is separated by means of a zinc rod (2) into a zinc sulphate solution (4) in an inner compartment - and by a diaphragm, here a ceramic filter candle (3). - Metallic copper is deposited on the latter network by means of an electrical connection between the zinc rod and the platinum network (electrical short circuit) without any further externally applied voltage. The baser redox system Zn/Zn ²⁺ compared to the Cu/Cu ²⁺ system supplies electrons via the electrical conductor in the direction of the platinum network, where they reduce Cu ²⁺ to metallic copper. As in a redox flow battery, electrons flow (without being used here for external work, however) from the metallic zinc to the platinum mesh, in order to reduce copper(II) ions there to metallic copper, while in the zinc -electrode an equivalent amount of zinc(II) ions go into solution. If the zinc is present in a stoichiometric excess, the copper is deposited quantitatively. This is also the basis of the analytical technique of electro-gravimetry. The stoichiometric excess Principle of one of the two redox systems reacting with each other is the chemical basis of the quantitative working method.

The historical experimental setup 1 for internal electrolysis can also be used with dissolved redox systems, for example to carry out any redox state changes without applying an external voltage. In this case, electrodes 1 and 2 (platinum mesh and zinc rod) are replaced by inert electrical conductors (eg non-corrosive precious metals, carbon, graphite, etc.). This internal electrolysis becomes interesting if the same redox system or a similar one with the same standard redox potential is present in both compartments. Due to the character of the redox system that is present in the higher concentration (stoichiometric excess), which influences the redox potential, the short circuit of both electrodes will cause electrons to flow from the electrode with the more negative redox potential to the other electrode until both have the same Have redox potential (adjustment of the electrochemical equilibrium). Therefore, in the run-up to this invention, an attempt was made to replace iron(II) ions, as they occur in some redox flow batteries with the redox system Fe(III)/Fe(II) when the catholyte is completely discharged, with an excess of Fe (III) ions to oxidize again, which in this example is equivalent to a regeneration.

In this test setup, instead of a zinc sulfate solution, a soluble redox system, here a sulfuric iron(II) sulfate solution (0.01 M), was introduced into the inner compartment (4). Such a solution could be, for example, the completely discharged redox electrolyte of a ferric sulfate electrolyte used in a Fe-Ti or Fe-Cr redox flow battery to convert Ti(III) sulfate to Ti(IV) to oxidize sulfate or Cr(II) to Cr(III). In these special redox flow battery types, the iron(III) is reduced to bivalent in the electrochemical flow cell. In order to oxidize the latter back to the trivalent iron by means of a suitable redox system, in the device from 1 in compartment (5) filled the same volume of iron (III) sulfate solution in a much higher concentration (0.1 M). In the stationary state (no flow-through conditions), the electrode (2) - this time a platinum electrode instead of the zinc rod - showed a redox potential in a pure sulfuric iron(II) sulfate solution (0.01 M) compared to a standard Hydrogen reference electrode of approx. +400 mV. The electrode (1) in a pure sulfuric iron(III) sulfate solution showed a redox potential of approx. +900 mV. Because of the short circuit between the two electrodes, electrons can now flow from electrode (2) to electrode (1). Equivalent amounts of bivalent positive iron ions are oxidized to trivalent ones in compartment (4), while an aliquot amount is reduced to iron(II) ions in compartment (5). In the present experiment, the amount of iron(II) present in compartment (4) was only 1/10 of the amount of iron(III) in compartment (5). Both solutions were stirred by bubbling nitrogen through them. Equilibrium was reached after some time and no more current flowed through the short-circuit bridge. The potentials of both electrodes compared to a standard hydrogen electrode have equalized and the concentration ratios of iron(III) ions to iron(II) ions are the same in both compartments. Chemical analysis revealed a ferric ion to ferrous ion ratio of 9 in both compartments. Inserting this value into the relevant Nernst equation resulted in a potential for both electrodes of 826 mV compared to the standard hydrogen electrode. 830 mV was measured. However, these experiments required a very dense diaphragm so that the two redox partners do not mix via this necessary current key and thus undermine the electron transport via the electrical conductor. A ceramic filter cartridge, which was additionally sealed with an agar-agar gel, was used here. Under stationary conditions, in a beaker as in 1 shown - despite the excess of the oxidizing redox system - only 90% of the iron(II) ions are oxidized. Therefore, by increasing the volume, a 100-fold stoichiometric Fe ³⁺ ion was obtained for the outer ferric sulfate solution surplus used. Due to this stoichiometric excess, 99% of the bivalent iron in compartment (4) could then be oxidized to trivalent iron. This proves the possibility of a quantitative regeneration of the redox system in compartment (4).

Example 2: Internal electrolysis with bipolar electrodes

In the experimental setup 1 the two working electrodes (1) and (2) were electronically short-circuited. The same effect can be achieved by using the two faces of a conductor, each in separate compartments connected with an electrolytic key, in tubular form. Then, from an experimental setup, 1 a build that in 2 is sketched, while at the same time cheaper electron-conductor was used as electrode-material. The cylindrical bi-polar electrode (1) consisted of an electron-conducting pyrolytic graphite tube such as is used, for example, in flameless atomic absorption spectroscopy. In the inner compartment (2) there was a weak sulfuric acid iron(II) sulfate solution (0.01 M) and in the outer compartment (3) there was also a weak sulfuric acid iron(III) sulfate solution (0.1 M) that was 10 times more concentrated. same volume as inside. Instead of a diaphragm, a membrane - selectively allowing anions through - was used, which seals off the graphite tube in a semipermeable manner. In addition, to accelerate the mass transfer to and from the two electrode surfaces (inside and outside of the graphite tube), stirring was also carried out using a magnetic stirrer (5). Identical results as in example 1 were obtained under stationary conditions, which proves that the experimental setup with bi-polar electrodes and semi-permeable membrane also worked like an internal electrolysis. In another experiment, the volume of the outer redox electrolyte solution (compartment 3) was increased by a factor of 10. In compartment (2) a 99 percent regeneration of the primary redox system (here iron(II) sulphate) could be achieved. This proved the fundamental possibility of regenerating a primary redox mediator with an identical redox system in a different oxidation state and in excess. Analogous experiments were also carried out with the reduction of Ti ⁴⁺ compounds using excess titanium ³⁺ compounds. They prove that regeneration by the same redox system in stoichiometric excess is possible. It turned out, however, that the time it took for the electrochemical equilibrium to be established was relatively long. This was due to impeded ion flow through the ion selective membrane and diffusion and reaction overpotentials at the two phase boundaries where electron transfer occurs. In order to compensate for the latter, an attempt was made to carry out this regeneration with another redox system with a more positive redox potential. Therefore, in a further experiment, a 0.1 M solution of cerium(IV) sulfate was used in compartment (2) as the regeneration redox system in compartment (3) in the case of iron(II) sulfate. In terms of redox, it behaved exactly like iron(III) sulfate, albeit with a faster establishment of equilibrium. This proves that other redox systems with a suitable redox potential can also be used to regenerate catholyte or anolyte. Such a bi-polar internal electrolysis has not previously been described and is the core of this patent application. The features of this technique are: At least two electrically interconnected electron-conducting surfaces, each in contact with at least one compartment, the compartments in question being connected via an electrolytic bridge (current key, permselective or ion-selective membrane) to complete the circuit. Permselectivity prevents the reacting redox-active compounds from mixing across the compartments via this electrolytic connection. Ions that are not redox-active, such as H ⁺ - or OH ^- ions, are allowed to pass through.

Example 3: Primary redox mediators and main redox systems in reservoirs

3 shows in a highly simplified (without current key) and schematic manner the basic structure of a version of the device according to the invention for a novel redox flow battery technology. It uses a small-volume anolyte and catholyte circulatory flow with predominantly material separation and preferably purely electronic contact (electron exchange predominantly only via electrical conductors) to the relevant main redox system that determines storage capacity (liquid redox polymers or solid redox polymer resins). The resemblance to typical redox flow battery arrangements and based on tube bundle heat exchangers is intentional here in order to clarify the principle. Note, however, that the flow circuit of anolyte and catholyte (4) and (5) shown here only represents the small-volume part, the redox systems on the anolyte and catholyte side and the electronically conductive flow channel line through the electron exchange columns (here storage tanks) in 3 only represents an exemplary and non-optimal embodiment without a predetermined scale with regard to length and inner diameter. The volume of the main redox systems (light colored) must be much larger in real terms than shown here. The two anolyte and catholyte electrolytes do not need to reach the maximum solubility of the relevant redox compounds because of their regeneration when flowing through the main redox systems be set. In the case of dissolved organic redox polymers, for example, they can be prepared in such a way that a high viscosity does not yet develop in order to keep the pump energy losses low. In general, the electrochemical capacity (expressed in Ah) of the main redox system is orders of magnitude higher than the redox capacity of the pumped anolyte and catholyte solutions. The circulating pumps (3) convey the anolyte and catholyte as traditionally through the reservoir, but here by means of tubular lines with electrically conductive walls and then to the collector electrodes (2) of the electrochemical flow cell with a separating membrane (1), which has a should prevent direct mixing of the redox partners, but does not allow redox-active ions to pass through. The electrochemical flow cell (or battery cell stack) corresponds to those traditionally used in redox flow batteries; it is not the subject of the invention disclosed here. Within the electrically conductive flow channels, the electron transfer from and to the surface of the channel walls is optimized by the flow in such a way that concentration polarization and cell voltage-reducing diffusion overvoltage are reduced. Outside of these channels, this is not the case and disadvantageous concentration polarizations could take place in the electrolyte in stationary redox systems. In order to minimize this electrochemical effect, the outer surface of the electrically conductive flow channels is increased by electrical contact to other electronic conductors - embedded in the main redox system. 3 (detail enlargement) shows an example with a wire brush-like surface enlargement by means of non-corrosive metallic conductors, which penetrate the main redox systems preferably perpendicular to the mediator flow direction in order to obtain chromatographic conditions and the equipotential surface does not long over a limited desired area to expand the direction of flow. Another example for directed electron-line to flow-canal-wall is indicated by folded metal-rim next to it. It is obvious for every expert to estimate the optimal surface enlargement of electronic conductors on the outside of the mediator flow channels and to find suitable material. Due to the immobile main redox system, a stationary current density is established there during the inner electrolysis, which is orders of magnitude lower than that on the inner surface, where anolyte and catholyte flow past with convection.

It is optimal if the outer surface of the conductor surface in electrical contact with the conductive flow channel wall reflects this current density ratio, i.e. the outer surface is larger by this ratio than the corresponding inner surface in the flow channel section under consideration.

A person skilled in the art will easily be able to implement other possibilities that are suitable for producing a large surface perpendicular to the direction of flow of anolyte and catholyte from a conductive material that is in intimate material contact with the main redox system (in resin or liquid form). Examples can be: Graphite fabric, graphite felt, graphene-based solid conductors (ideal when placing the best conductivity perpendicular to the flow direction for anisotropic conductors), carbon nanotubes (CNT) with extremely large surface area and similar materials, such as “polymers of intrinsic microporosity (or PIMs)” [https://pubs.acs.org/action/showCitFormats?doi=10.1021/acs.analchem.0c04554&ref=pdf]. The latter represent porous materials with a distorted and rigid macromolecular structure. They allow robust microporous coatings with a very large surface. With a typical micropore size range of about 1 nanometer, they can give an accessible surface area of 700-1000 m ² /g. Multilayer graphene (MLG) also has excellent properties, such as high electrical/thermal conductivity and a current-carrying capacity that exceeds that of copper [H. Murata, Y Nakajima, N Saitoh, N Yoshizawa, T Suemasu, K Toko "High-Electrical-Conductivity Multilayer Graphene Formed by Layer Exchange with Controlled Thickness and Interlayer", http://www.nature.com/ scientific reports, (2019) 9:4068 [https://doi.org/10.1038/s41598-019-40547-0].

It is advantageous for a procedure analogous to an ion exchange column with locally progressive conversion (regeneration of anolyte and catholyte) if the redox conversion preferably initially takes place in layers and perpendicular to the direction of flow. This requirement requires an electrical interruption of the electronically conductive flow channel walls in order to avoid an excessively large, disadvantageous equipotential surface. Advantageously, the conductive wall areas become longer in the direction of flow in order to counteract the depletion of the concentration of the relevant redox systems in the anolyte and catholyte. To support this requirement, it may also be necessary to place electrically insulating partitions perpendicular to the direction of flow of anolyte and catholyte. The flow channels for anolyte and catholyte can also have other shapes in order to increase the flow distance through the main redox system and at the same time contact a maximum amount of the main redox system. For example, they can also be serpentine, spiral (see 6 , 7 , 8th ) or be meandering.

Depending on the nature of the main redox systems, different versions of the device according to the invention are selected. In the case of solid, insoluble electron exchange resins, a small amount of the electrochemically corresponding anolyte or catholyte is preferably used as an electron intermediate carrier between the solid resin surface and the adjacent electrical conductors. In this case, the anolyte and catholyte flow channels can also be porous. Then there is no need to use special ion-conducting current keys between the two compartments (inside and outside the flow channels). In the case of the solid redox resins, the corresponding monomeric redox systems can also be used as anolyte and catholyte solution.

If polymeric and high-viscosity organic redox systems are used as the main redox systems, the anolyte and catholyte flow channels must be non-porous and permeable for all solutes. Here, a material separation of these mediator redox systems from the relevant main redox systems takes place, analogous to that in the electrochemical flow cell, either with an electrolytic current key (see Fig. 4 ), ion-selective membranes or size-exclusion membranes (e.g. dialysis membranes, etc.), in which case the anolyte and catholyte redox system and main redox system can be identical in the latter case, but are present in different concentrations. Anolyte and catholyte have lower concentrations of the water-soluble organic redox polymers whose viscosity does not yet require excessive pumping energy. The osmotic pressure between the highly polymeric circulating redox systems and the stationary, highly viscous main redox systems is balanced by the appropriate addition of highly conductive electrolytes (neutral salts). As an electrolyte bridge with a larger effective cross section, the individual electrically conductive flow channel sections can also be separated with dialysis tubing bundles connected in parallel. The use of ion-selective membranes for the passage of non-redox-active ions (preferably H ⁺ - or OH ^- - ions) is recommended when changing the solvent between anolyte or catholyte and the corresponding main redox system or when changing from an inorganic to one an organic redox system between the two.

The membrane-based power keys are in arrangements after 5 also hose-like or tubular and interrupt the electrically conductive flow channel sections in parts of different sizes. The purpose of this interruption of the electrical conductivity of the flow channel wall is to avoid equipotential surfaces that are too long, so that new sections with the redox potential of the charged main redox system are always obtained in the direction of flow. As the redox systems in the anolyte and catholyte are increasingly regenerated during the discharging process as they flow through within the reservoir with the relevant main redox system, and the amount of redox material that has not yet been converted in the direction of flow decreases, due to the taking place Concentration polarization the corresponding flow of electrons to or from the wall, which is undesirable as it retards quantitative regeneration. To compensate, according to the invention, the electrically conductive sections of the flow channels become larger in the direction of flow. These sections are separated by channel-shaped insulators. Tubular ion-selective or molecular filter membranes can preferably be used for this purpose. 5 shows schematically how the anolyte and catholyte flow channel (1) is constructed with electrically conductive wall areas and insulating (no flow of electrons possible) areas. The electrically conductive sections (3) become longer in the flow direction. The isolating areas (2) are formed by tubular ion-selective membranes or, in the case of aqueous, high-polymer main redox systems, by corresponding dialysis tubes that block the redox system. Depending on the requirements of this electrolyte bridge (necessary ion flow for current equalization of the electrochemical conversion), longer, only ion-conducting flow channel sections or dialysis tubing bundles with suitable permeability can also be used here.

The volume of the electron exchange column, which determines the total storage capacity, and the number and diameter of the mediator channels can be designed as required. The additional electronic conductors to be introduced in the main redox system with contact to the outer walls of the flow channel are, for the sake of clarity (and the large number of obvious possibilities), only schematic and exemplary for a mediator channel in 3 sketched. Likewise, for the sake of clarity, the isolating channel sections, which preferably consist of an ion-selective membrane, and the electronically conductive areas of the mediator flow channel walls, which become longer in the direction of flow, are not shown explicitly.

An electrochemical classic electrolyte bridge between anolyte or catholyte and the corresponding main redox system is also included for the sake of clarity 3 not shown. They can be prepared using standard current key electrolytes compatible with the adjacent phases in electrolytic contact. the 4 outlines a possibility in which a narrow-pored Frits limited space can be made between anolyte or catholyte and a resinous main redox system that closes the circuit electrolytically during regeneration. Anolyte and catholyte flow through parallel flow channels (1), which are shown here as ideal straight lines and without the electrically insulating sections, and are reunited after regeneration. The relevant main redox system is located in compartment (2), which in reality has a much larger volume and is provided with electrical conductors that are not drawn in (but are shown as an enlarged detail), here in the form of electron exchange resins and electrical conductors in contact with the conductive flow channel walls and wetted with a mediator (here the corresponding redox monomer that should not mix with anolyte or catholyte). A traditional electrochemical salt bridge (4) filled with a highly conductive salt solution that does not interfere with the redox systems involved. This current key electrolyte can be renewed or kept clean by means of external supply and discharge lines. Such a construction can also be used particularly advantageously in order to separate aqueous anolyte and catholyte solutions from the corresponding aqueous redox polymer main redox systems by a liquid layer of a water-immiscible solvent. In the interest of a low internal resistance, the layer thickness of the liquid organic matrix should be kept as small as possible. Viscous organic, water-immiscible electrolytes can also be stabilized within a single diaphragm section. In addition to good conductivity, which can be achieved with the help of known lipophilic salts, specific carrier molecules can also be added for the ions to be transported through this phase boundary (e.g. valinomycin if potassium ions are to be transferred or Aliquat S if anions are to be transferred via have to be transported over this limit). However, an electrolytic separation of two organic solvents that are not soluble in water is also possible with such an arrangement. An aqueous electrolyte bridge with highly conductive salts such as potassium chloride or lithium sulphate can do a good job here. If protons or hydroxyl ions have to flow through the electrolyte bridge during the electrochemical reaction during the discharging or charging process, an additional pH buffer can also be of good service there. Such a diaphragm current key is also well suited when the anolyte and catholyte are aqueous and the corresponding main redox systems are dissolved in an organic, water-immiscible organic solvent or vice versa. When using such an electrolyte bridge with an attached electrolyte reservoir, there is the possibility of rapid electrolyte exchange. The placement of at least one such electrolyte bridge is up to the designer, as is the shape (annular or planar). If, on the other hand, anolyte and catholyte themselves serve as redox mediators in compartment (2), the electrically conductive flow channel wall sections can also be porous and an electrolyte bridge can be completely dispensed with.

Example 4: Use of several redox systems in the anolyte or catholyte cycle

A particular advantage of the construction of mediator-supported redox flow batteries according to the invention is the possibility of regeneration of anolyte and catholyte without the electrochemically corresponding main redox systems chemically contaminating the former. In order to make the recharging of the main redox systems more quantitative and faster through the electrochemically adapted (same standard redox potential) redox systems of anolyte and catholyte, each of them is equipped with another so-called auxiliary redox system with a more negative one or more positive standard redox potential than that of the relevant main redox system. Small amounts of the auxiliary redox system (e.g. 10% of the respective primary redox systems in anolyte and catholyte) are sufficient.

A titanium/iron redox flow battery in a strongly acidic electrolyte is to be considered here as an example. In this battery, the trivalent positive titanium ions reduce the trivalent positive iron ions in the electrochemical flow cell to bivalent positive iron ions and are themselves oxidized to tetravalent positive titanium ions (simplified here, neglecting the water coordination). The chemical reaction outlines the following equation: $$\begin{array}{ccc}{\text{<figure-callout id="2" label="H" filenames="DE102021001501A1\_0002.png,DE102021001501A1\_0003.png" state="{{state}}">H</figure-callout>}}_2{\text{TiCl}}_6+{\text{<figure-callout id="2" label="FeCl" filenames="DE102021001501A1\_0002.png,DE102021001501A1\_0003.png" state="{{state}}">FeCl</figure-callout>}}_2\leftarrow \to & {\text{<figure-callout id="3" label="TiCl" filenames="DE102021001501A1\_0002.png,DE102021001501A1\_0003.png" state="{{state}}">TiCl</figure-callout>}}_3+{\text{<figure-callout id="3" label="FeCl" filenames="DE102021001501A1\_0002.png,DE102021001501A1\_0003.png" state="{{state}}">FeCl</figure-callout>}}_3+2\text{HCl}& \text{cell voltage}=0.74\text{V}\end{array}$$

In a regenerative version of this redox combination, a Fe(III) solution and a Ti(III) solution are passed through the battery cells where the ferric iron is reduced to Fe(II) and the ferric titanium to Ti(IV). is oxidized under power supply. After flowing through the electrochemical flow cells, the tetravalent titanium (Ti ⁴⁺ ions) and the divalent iron (Fe ²⁺ ions) are added to different regeneration columns according to the invention.

In this example, in a specific embodiment, the regeneration column (main redox system) consists of an organic electron exchange resin, which is matched in terms of its electrochemical properties to the Ti ⁴⁺ /Ti ³⁺ standard redox potential, which is synthetically easily possible . In the charged state, the redox potential of this exchange column corresponds to that of a pure Ti ³⁺ ion solution, which - if used in excess - reduces the Ti ⁴⁺ ions flowing through to Ti ³⁺ ions in electrochemical equilibrium and thus regenerates them. The same applies to the divalent iron emerging from the electrochemical flow cells. The Fe ²⁺ ions are also oxidized back to trivalent iron (Fe ³⁺ ions) when flowing through a corresponding redox exchange resin with the standard redox potential of the Fe ³⁺ /Fe ²⁺ system and are available again energy generation available.

In the present example (in the case of solid, insoluble electron exchange resins), the anolyte and catholyte do not have to be materially separated from the relevant electrochemically adapted main redox system. A porous electrically conductive flow channel line with insulated sections will suffice. The anolyte or catholyte electrolytes wet the stationary layered electron exchange resins together with electrical conductors and enable electron transfer there between the redox polymer resin via the electrical conductors to the outer flow channel wall. As in 3 (Enlargement of detail), electrical conductors are preferably located with their surface perpendicular to the direction of flow of the mediator electrolytes in order to limit the electrochemical reaction preferably locally and perpendicularly (in layers) to the direction of flow.

In order to allow the anolyte and catholyte regeneration to proceed more quantitatively, it is advantageous to add other so-called auxiliary redox systems, each with a more negative and a more positive standard redox potential than the primary anolyte or catholyte redox system in question has. In the present example, a small amount of a Cr ³⁺ compound is added to the Ti ⁴⁺ /Ti ³⁺ system and a small amount of a Ce ³⁺ compound is added to the Fe ³⁺ /Fe ^2* system. The added amount can be between 1% and maximum saturated. In the recharging process of the two main redox systems (redox exchanger resins) in the electrochemical flow cells, when the redox system of the primary anolyte or catholyte has been almost quantitatively returned to their charged state and during the galvanostatic charging process, there is a risk of a unwanted water electrolysis occurs in the auxiliary redox systems previously electrochemically converted. The trivalent chromium compound becomes a Cr(II) compound, which in turn has a more negative redox potential than that of the almost completely charged corresponding main redox system. It ensures a more quantitative regeneration of the electron exchange column during flow-through because of the compensation of flow-through and concentration overvoltages. The trivalent cerium has a similar effect, which when flowing through the battery cells during recharging after the quantitative oxidation of the divalent iron ions to Fe(III) is oxidized to Ce(IV), which quantitatively regenerates this main redox system when flowing through the corresponding regeneration column (recharges).

Another advantage of using at least one additional redox system in each anolyte or catholyte circuit with a more negative or more positive standard redox potential compared to the corresponding primary redox system is that towards the end of the recharge water decomposition is avoided if coulometrically 100% of the primary redox systems are converted during the charging process in the electrochemical flow cell. Of course, analogous electrochemical situations can also be achieved by adding different complexing agents, for example to inorganic metal-based redox systems, which partially stoichiometrically complex the relevant metal cation and thus shift the standard redox potential for this complexed part in the desired direction . The auxiliary redox systems come into action electrochemically only when the main redox systems are recharged. They play no part in the discharge process other than increasing the conductivity of the anolyte and catholyte, which is desirable.

In a further implementation of the device disclosed here, these so-called auxiliary redox systems, each with a more positive or more negative redox potential than the corresponding main redox systems in the storage tanks, are not mixed with the anolyte or catholyte, but separately and with a higher concentration with its own flow channel line analogous to the anolyte or catholyte line (interrupted electrically conductive walls) through the volume of the main redox systems, although these are also at least 10 mV more positive or are designed to be more negative than the relevant main redox systems in the storage tanks. This can preferably be done using multi-channel flow lines (multi-lumen hoses) that act like a single line to the outside. This circulatory flow of the auxiliary redox system only flows when the main redox systems are recharged and compensates, as in the above case - dissolved in anolyte and catholyte - occurring overvoltages (diffusion and reaction overvoltages).

the 9 is intended to illustrate the principle of using additional auxiliary redox systems by means of a representation of various standard redox potentials of inorganic and organic redox systems. It shows examples of standard redox potentials of various inorganic and organic redox systems on a volt scale versus a standard hydrogen electrode (NHE) as reference electrode. The arrows marked with numbers show possible standard redox potentials of the auxiliary redox system (1 and 1'), storage capacity-determining main redox system (2 and 2') and of anolyte and catholyte (3 and 3 ') on. The redox potential differences are only shown here as examples. By applying the charging voltage to the battery cells, the auxiliary redox systems only have an electrochemical effect on the main redox systems during their recharging. The current-yielding anolytes and catholytes are 3' and 3', respectively, with respect to their standard redox potential. Due to the difference to the standard redox potentials of the relevant main redox systems at 2 and 2', there is an overvoltage in both cases to compensate for kinetic and diffusion-related limitations. If more than one auxiliary redox system is dissolved per anolyte and catholyte, they are not electrochemically involved in the discharge of the electricity storage (main redox systems) except for increasing an electrolytic conductivity. However, when recharging the main redox systems, they can increase the amount of the main redox systems converted per time.

In the example shown, a change from an organic to an inorganic redox system could also be made possible if material contact was completely prevented. For example, a traditional vanadium redox flow battery, which still lies within the potential window 3-3' with regard to its respective redox potential, could be massively expanded in this way with regard to its energy density. A switch from an organic to an inorganic redox system when passing the current through the electrochemical flow cells can be advantageous if it can prevent electrode fouling.

Example 5: Arrangement in liquid organic main redox systems

A potential application in the promising invention is intended here as an example only to illustrate the device according to the invention WO2014026728 A1 be sketched. Redox flow batteries are described here, in which cost-effective and long-lasting redox flow batteries can be created with as little effort as possible by using new materials and membranes. Here, redox-active components of water-soluble organic high-molecular compounds are used as anolyte and catholyte and a size exclusion membrane (e.g. dialysis membrane) as membrane to avoid mixing ("chemical short circuit" of anolyte and catholyte) of the high-molecular redox-active components present in dissolved or dispersed form deployed. A major problem with all water-soluble organic redox polymers is the increase in viscosity of the relevant main redox systems, which increases with the degree of polymerization. The inventors paid particular attention to the rheological behavior in synthesizing the preferred anolytes and catholytes. The viscosity should be below 20 mPas in order not to generate excessive pumping energy losses. A pH-neutral sodium chloride solution was used as the basic electrolyte for both the anolyte and the catholyte. This new, very inexpensive and environmentally friendly type of redox flow battery, which was revealed a few years ago, could only be implemented with energy densities of around 10Ah/L due to the viscosity problem. With the device according to the invention, it can be installed particularly well in an in 6 Realize the exemplary construction version shown. When using a device according to the invention and using analyte and catholyte electrolytes with lower concentrations of the relevant redox polymers, the corresponding identical main redox systems can be highly concentrated and thus also highly viscous (syrupy). In this example of a device according to the invention, spiral-like flow channels (see Fig. 7 and 8th ) used. 6 is only intended to illustrate the principle schematically and not to scale. The anolytes or catholytes to be regenerated enter the spiral flow channel section (3) with interrupted electrically conductive walls at position (1) and exit or enter a next main redox system spiral module at position (5). In this example, the modules are filled with a highly viscous - very energy-dense - main redox system and with an electrically conductive network (2). The latter should preferably promote electron conduction in the direction perpendicular to the flow direction in contact with the flow channel wall. Anolyte and catholyte flow through the electrically conductive spiral channels (3) in 8th are only indicated flat (viewed from the side, not in the sense of a detailed technical drawing) and without drawn isolating sections. After two spiral passes, when the anolyte and catholyte flows are central again are, they exit at position (5) or into another regeneration module. After 2 spiral passages of anolyte and catholyte, the inlet and outlet channels are again vertically above one another, which leads to an easy connection with other double spiral modules. In the discussed example with identical, water-soluble redox polymer systems as anolyte and catholyte and main redox system, but in very different concentrations, no porous mediator flow channels may be used, in contrast to the systems with solid electron exchange resins, in order to achieve a direct To avoid concentration of anolyte and catholyte and dilution of the main redox systems or concentration of anolyte and catholyte. In each anolyte and catholyte regeneration module there is an electrolyte bridge in the form of a size exclusion membrane (4) which has the same specifications as the separation membrane in the electrochemical flow cell. However, at this point it is important to ensure that the anolyte and catholyte have the same osmolarity compared to the corresponding main redox system. A corresponding addition of salt to the more diluted system is used for this purpose.

If inorganic redox systems are used as the anolyte and catholyte, which are to be regenerated in this way with the main organic redox systems, the membranes (4) consist of ion-selective membranes which do not allow any redox-active substances to pass through. The ion-selective membrane is selected according to the electrochemical conversion and the ions flowing over a current key (cation- or anion-selective). Corresponding membranes are easy to produce (K. Cammann: Working with ion-selective electrodes, Springer-Verlag, 1996). For mechanical reinforcement, metal or fabric mesh or blotting paper can be used when constructing the membrane. If the configuration of the device according to the invention is operated with solid electron exchangers as the main redox system, the flow channel is kept porous and an additional electrolyte bridge is not required.

It is advantageous if, towards the end of the mediator-electrolyte regeneration, modules are used that contain more than 2 spiral flow channels with a reduced diameter, because this counteracts the depletion of the redox systems that are still to be converted with a reduced diffusion layer thickness. Since there is correspondingly little electrochemical conversion towards the end of the flow section until complete conversion (complete regeneration), the ion-selective membranes (as an electrolyte bridge) can also be kept smaller there.

With a standardization of production technology, any number of regeneration modules can be "flanged on" one behind the other. For the sake of clarity, the sections of the flow channel that are isolated from one another and have different lengths are not shown. In 6 the outer container walls are electrically non-conductive; as are the respective inlets (1) and outlets (5). In a cascaded arrangement of the main redox systems shown here as an example, the discharging and charging also takes place in modules from the first to the last module before the anolyte and catholyte re-enter the electrochemical flow cell. Only when the last module with the main redox system is also discharged does the battery voltage collapse. The advantage of this embodiment variant is that the total capacity of the main redox systems can be easily changed and there are also advantages in terms of maintenance.

Based on 6 other spiral designs of the anolyte and catholyte flow channel guide are also possible. the 7 shows an exemplary arrangement of a solid electronic conductor with spiral-shaped incisions, which are then sealed on one or both sides with a size exclusion or ion-selective membrane (glue or sealing pressure). This version of an electrolytic bridge is easier to manufacture. Another advantage is the large area for an ion exchange.

In a further embodiment, which is probably the optimal one, simply electronically made conductive size exclusion membrane tubing or corresponding tubing made of an ion-selective membrane are used as flow channel lines through the regenerators filled with the main redox system ( 8th ). In principle, it is also possible with the help of dialysis tubing bundles, such as those used in medical technology, but with internal diameters that are not too narrow and with electrically conductive size exclusion membranes, to build particularly efficient and high-revenue arrangements without an additional, separate electrolyte bridge. Corresponding electrically conductive size exclusion membranes have already been developed to avoid biofouling in water treatment plants [P. Formoso, E. Pantuso, Gi De Filpo, FP Nicoletta "Electro-Conductive Membranes for Permeation Enhancement and Fouling Mitigation: A Short Review", http://dx.doi.org/10.3390/membranes7030039]. Such membranes have also already been developed in medical technology [Shih-Chieh Yen, Zhao-Wei Liu, Ruey-Shin Juang, Sravani Sahoo, Chi-Hsien Huang, Peilin Chen, Yu-Sheng Hsiao, Ji-Tseng Fang “Carbon Nanotube/Conducting Polymer Hybrid Nanofibers as Novel Organic Bioelectronic Interfaces for Efficient Removal of Protein-Bound Uremic Toxins”, ACS Appl. mater Interfaces 2019, 11, 43843-43856]. With these electric lei tending size-exclusion membranes or ion-selective membranes as hose bundles saves the construction of separate electrolyte bridges in terms of production technology. When working with prefabricated conductive dialysis hose bundles, for example, the auxiliary redox system and the anolyte or catholyte can flow through a hose bundle in such a way that they are each 50% mixed (the same number but differently filled individual hoses) are connected . Such a structure then corresponds to one as described for solid resinous electron exchangers, which also does not require a special current key because of the porous flow channel wall.

The main redox systems, which are stationary in the regeneration units (storage tanks) according to the invention, can then be prepared in such a concentrated manner that highly viscous liquids are formed. Together with a higher degree of polymerization, extremely high energy densities can be achieved that surpass everything known.

Water-soluble organic redox polymers, which can just about be pumped, can also be used for non-stationary applications together with the corresponding main redox system structures based on the modular system. The battery cells and the connected pumped anolytes and catholytes and any other auxiliary redox systems remain stationary in the circuit through the regeneration unit. The individual regeneration modules receive combined filling and outlet openings only for the exchange of discharged main redox systems. At special filling stations with 4-channel taps, discharged redox flow battery reservoirs are replenished by replacing the used main redox systems with "fully charged" systems (in the form of viscous, non-toxic solutions according to the displacer principle; charged system pushes discharged from the device) reactivated. This process is analogous to conventional refueling with petrol or diesel. When using the aqueous organic redox polymers as the main redox system and as the anolyte and catholyte, there is no longer any flammability and there is no toxicity (no benzene, toluene, ethylbenzene, xylenes, ethyl tert-butyl ether (ETBE) and methyl tert -butyl ether (MTBE)) is no longer present.

According to the possible high energy density, a car with a 100 kg filling can drive over 1000 km before it has to be refueled or recharged. The gas station then regenerates the discharged green redox systems on-site using solar or wind energy. A transport of dangerous goods would be omitted. It is important that the cell stack and the primary anolyte and catholyte remain unchanged. With a clever process engineering design, reserve redox systems in standardized modular form are also conceivable, which correspond to the function of a traditional reserve petrol canister.

Example 6: Use of the new device as a domestic stationary solar power storage

Many of the new synthesized water-soluble redox polymers are particularly advantageous as local, decentralized energy storage with large storage capacity. The energy-dense, highly viscous main redox electrolytes remain stationary in reservoirs, which in the case of domestic solar power storage can also have a flat, rectangular basement wall-sized shape, in which the electrically conductive dialysis hoses are laid in such a way that regeneration conditions for anolyte and catholyte exist again , as with ion exchange columns. In this embodiment of the regeneration device, it is particularly advantageous that anolyte and catholyte do not react aggressively, since they only consist of common salt solution and dissolved organic redox polymer. For this reason, the very energy-saving circulating pumps that are common in hot water heaters or solar circuit pumps can also be used here to pump anolyte and catholyte. The energy consumption of such a solar circuit pump accounts for up to 5% of the battery efficiency. However, this loss is offset by the high energy density gained.

A redox flow battery with such regenerators with a volume of conventional heating oil tanks (e.g. 3000 liters of catholyte and 3000 liters of anolyte system) can temporarily store around 1,500 kWh of energy with the environmentally friendly, water-soluble redox polymers. This already corresponds to approx. 25% of the annual electricity consumption of a family of four. This also opens up the possibility of making older buildings completely self-sufficient using a solar system and heat pump. If wind energy is also used in the form of the innovative wind wheel tree [https://www.vattenfall.de/infowelt-energie/oekostrom-aus-demwindbaum-der-firma-newwind] or by means of a wind tunnel-supported house gable ( EP2513471A2 from October 24th, 2012), millions of old buildings that have been converted into passive houses in this way can even supply electricity to the grid and make e-mobility possible in the first place

It goes without saying that a precise and exhaustive list of the redox systems suitable as anolyte or catholyte and the main redox systems providing storage capacity in the device according to the invention is not possible, since new suitable systems are constantly being discovered . Likewise, process engineers can easily perform other details of the disclosed process: bi-polar internal electrolysis combined with chromatographic flow-through behavior (analogous to an ion exchange column) and low hydrostatic pressure build-up.

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• DE 914264 C [0009]
• US4786567 [0009]
• US 2012/0171541 A1 [0012]
• US8080327B1 [0013]
• WO 2014026728 A1 [0019, 0068]
• WO 2013/012391 A1 [0022]
• US 2014/0178735 A1 [0022]
• EP 0882924 B1 [0035]
• EP 2513471 A2 [0078]

Claims (9)

A method and apparatus for regenerating anolyte and catholyte liquids from redox flow batteries in which electrical energy is stored in redox chemical compounds, wherein an anolyte and catholyte recycle flow is pumped through electrochemical flow cells to generate electricity, thereby characterized in that for the electrochemical regeneration of the redox systems used in the anolyte and catholyte, the electrochemical process of "internal electrolysis" is applied in such a way that the redox systems to be regenerated in the anolyte and catholyte by a preferred only electronic contact using electronic conductors with the electrochemically exactly corresponding, and present in a stoichiometric excess, capacity-determining, energy-dense main redox system in special regeneration devices (exchange electrons) and are thereby regenerated; Method and device according to claim 1 ., characterized in that the electrochemical regeneration of anolyte and catholyte is carried out using the method of bi-polar internal electrolysis, with catholyte and anolyte after flowing through the electrochemical battery cell stack by means of hydraulically advantageous flow channels of variable shape with electrically conductive - and in Wall sections that become longer in the direction of flow - are conducted by line through the storage tanks filled with the electrochemically corresponding main redox system and the electrically conductive flow channel wall sections, depending on the type of main redox system with ion-selective or size-exclusion membrane sections - acting as an electrolyte bridge and for redox -active connections impermeable - are interrupted. and the electrically conductive flow channel wall sections are in electrical contact with further strongly surface-enlarging electron-conducting material, which preferably protrudes perpendicularly to the flow direction of anolyte and catholyte into the main redox systems and the flow channels for anolyte and catholyte through the electrochemically corresponding ones Main redox systems are performed in preferably columnar filled storage tanks that the electrochemical reaction between the redox systems in anolyte and catholyte and the main redox system in question preferably only from an electron transfer via electrical conductors and preferably only perpendicular to the Flow direction expires and thereby ion exchange column analogous conditions are maintained and anolyte and catholyte always leave the regeneration device almost fully charged or discharged; Method and device according to one of the preceding claims, characterized in that the energy-dense main redox systems used are preferably specifically synthesized organic redox polymers, which are either in in a solid resinous form or as a redox high polymer dissolved in a solvent; Method and device according to one of the preceding claims, characterized in that the anolyte and catholyte to be regenerated from any redox flow battery types to compensate for diffusion and passage overvoltages when recharging the relevant main redox systems in the device respectively at least one further auxiliary redox system in dissolved form with a more negative standard redox potential than the standard redox potential of the anolyte in question and a more positive standard redox potential than the standard redox potential of the catholyte; Method and device according to one of the preceding claims, characterized in that the effective, capacity-determining main redox system on the anolyte side has a more negative standard redox potential than the redox system in the anolyte and on the catholyte side has a more positive standard redox potential than that in the catholyte and at least one additional auxiliary redox system each in a flow channel system separate from the standard anolyte and catholyte - preferably in the form of multi-channel lines - is used to recharge the two main redox systems , which on the anolyte side has a standard redox potential which is more negative than the standard redox system of the main redox system in question and on the catholyte side has a standard redox potential which is more positive than that is the standard redox potential of the main redox system in question; Method and device according to one of the preceding claims, characterized in that, as flow channels for anolyte and catholyte and all auxiliary redox systems, electrically conductive multilumen hoses with ion-selective membrane properties or size-exclusion membrane properties (as integrated, redox-active compounds blocking Electrolyte bridge) are used and anolyte or catholyte, together with the corresponding auxiliary redox system, are passed through different lumens of the same hose through the relevant main redox systems. Method and device according to one of the preceding claims, characterized in that in the case of solid electron exchange resins as the main redox system, the relevant monomers are used as the anolyte or catholyte redox system, which also uses porous, electrically conductive flow channel walls to transfer the redox polymer Resins and the other electrical conductors wet, which makes additional electrolyte bridges unnecessary and, with liquid redox polymers as the main redox system, a less viscous, more dilute solution of the relevant redox polymer with a suitable standard redox potential as anolyte or catholyte together is used with appropriate size exclusion membranes and the osmolarity is adjusted to that of the main liquid redox system in question by means of salt addition; Method and device according to one of the preceding claims, characterized in that the special flow channels for anolyte and catholyte or other auxiliary redox systems are guided spirally through the main redox system and each have 2xn (n = 1, 2,3, etc.) with inlet and outlet vertical and central to an easily exchangeable column including electrolyte bridge, so that this regeneration column is designed as a completely exchangeable module which, when connected in series depending on the desired storage capacity, results in the total capacity; Method and device according to one of the preceding claims, characterized in that when using organic redox polymers dissolved in a solvent including water as the energy-dense main redox system, the electrochemically corresponding anolytes and catholytes from redox polymers of the same molecule class but in more dilute form exist and as auxiliary redox systems this redox polymer molecule class of the main redox system in a form with other substituents that change the standard redox potential in the desired way and more than one auxiliary redox system in form a redox cascade with at least 5 mV difference per cascade stage in the standard redox potential for both anolyte and catholyte.

Images