Search in book...
Toggle Font Controls
Create new playlist

Name your new playlist

Playlist description (optional)
Sign In

Email address

Password

Forgot Password?

or

Continue with Facebook

Continue with Google
Sign Up

Full Name

Email address

Confirm Email Address

Password

or

Continue with Facebook

Continue with Google

Chapter 12

Vanadium Redox Flow Batteries

Christian Doetsch

Jens Burfeind Fraunhofer Institute for Environmental, Safety, and Energy Technology UMSICHT, Oberhausen, Germany

Abstract

A redox flow battery (RFB) is a special type of electrochemical storage device. Electric energy is stored in electrolytes which are in the form of bulk fluids stored in two vessels. Power conversion is realized in a stack, made of electrodes, membranes, and bipolar plates. In contrast to conventional lead–acid or lithium-ion batteries, the energy conversion unit and energy storage unit are separate devices. From this point of view RFBs are more like reversible fuel cells in that the stack is the power conversion unit and is built in a filter press design; or, compared with other storage technologies, like pumped hydro systems, where power conversion is realized by pumps and turbines, but capacity by higher and lower elevation reservoirs. This advantage leads to applications where higher or various ratios of capacity to power (kilowatt-hours per kilowatt) are needed or advantageous—usual are ratios from 5:1 to 10:1. The most common and mature RFB is the vanadium redox flow battery (VRFB) with vanadium as both catholyte (V²⁺, V³⁺) and anolyte (V⁴⁺, V⁵⁺). No cross-contamination from anolyte to catholyte is possible and hence this is one of the most simple electrolyte systems known. Other electrolyte systems could be cheaper (Fe/Cr) or more efficient, but currently these are not available. Other important advantages of the VRFB include long-lasting operation time, long lifecycle; good stability; ease of regeneration or recycling of the electrolyte; and very low flammability. Disadvantages include low energy density both gravimetric and volumetric so that mobile applications are excluded, and currently they are relatively expensive when compared to lead–acid or lithium-ion batteries. Their higher cost is largely due to lower maturity and thus they do not have the advantage that mass production would bring about.

Keywords

flow battery

vanadium redox flow battery

hybrid flow battery

current development of flow batteries

application for flow batteries

1. Introduction and historic development

The redox flow battery was first developed in 1971 by Ashimura and Miyake in Japan [1]. In 1973 the National Aeronautics and Space Administration (NASA) founded the Lewis Research Center at Cleveland, Ohio (USA) with the object of researching electrically rechargeable redox flow cells. The Exxon Company (USA), Giner Ind. (USA), and Gel Inc. (USA) were awarded contracts to develop a hybrid redox flow battery [2] and in the following 6 years research was done on different redox couples, membrane development, electrodes, etc. [3] Iron chloride (FeCl₃) and titanium chloride (TiCl₂) were proposed as electrolytes. In 1975 a patent (US Patent 3 996 064) was filed by Lawrence H. Thaller. The description of present-day systems is identical to that described in the original patent which involved a two-tank system and a cell with a separator and two graphite electrodes. With the present-day application of flow batteries to store sustainable electricity, they appear to have foreseen future problems: “Because of the energy crisis… and due to economic factors within the electric utility industry, there is a need for storing bulk quantities of electrical power […] be produced intermittently […] by devices such as wind-driven generators, solar cells or the like” (US Patent 3 996 064). Later Thaller [4] replaced titanium with chromium. This electrolyte couple is in the spotlight today with one plant being set up by Enervault (USA) (www.enervault.com) and discussions taking place about further developments. In 1978 an all-vanadium system was proposed for the first time. In this special case both electrolytes consist of the same metal, but at different stages of oxidation. Therefore, there is no problem with cross-contamination through the separating membrane. This technology was further developed in the 1980s by Maria Skyllas-Kazakos at the University of New South Wales (Australia). This vanadium-based redox flow battery is today the most developed and popular flow battery and its sales exceed those of other flow batteries. Also in the 1980s the Japanese company, Sumitomo, was very active in filing patents and developing new membranes and electrolytes. This activity stopped at the end of the 1990s and was restarted 5 years ago. The Canadian company, VRB Power (CA), was another very active company from 2000 to 2008. At the end of 2008 it filed for bankruptcy and was bought out by Prudent Energy VRB Systems. In Austria, Martha Schreiber developed electrochemical storage systems, with special emphasis on redox flow; she founded the company Cellstrom GmbH (Austria), which today, together with of the German company Gildemeister Energy Solutions, is now part of Deckel-Maho-Gildemeister (DMG) Mori Seiki AG of Japan.

The development of scientific interest in redox flow technology can be seen in Fig. 12.1.

Figure 12.1 Number of publications (research with Google Scholar; allintitle “redox flow” July 2015).

The 1970s witnessed the start of a small number of publications each year (5–12). Since 2004 the number of publications has risen significantly and has almost doubled in number every 3 years. Currently, there are nearly 300 publications per year.

The current strong interest in flow batteries can also be seen in the overview of patents involving redox flow batteries (RFBs): over the past 5 years (2010–14) there have been roughly twice the number of patents filed over the previous 40 years (1970–2009) (Fig. 12.2).

Figure 12.2 Development of filed patents regarding redox flow in the world [research with depatisnet; »redox flow« and (»batter*« or »cell« or »cell«) in claims, title, abstract; May 2015].

The inventors responsible for these patents come mainly from Japan, China, USA, South Korea, Europe (especially Germany and Austria), Canada, Australia, Taiwan, and India (see Fig. 12.3).

Figure 12.3 Filed patents regarding redox flow in different countries [research with depatis net; »redox flow« and (»batter*« or »cell« or »cell«) in claims, title, abstract; May 2015].

Today, the companies working with RFBs include large companies such as Sumitomo (Japan), DMG Mori Seiki AG (former Gildemeister) (Germany/Japan), and Prudent Energy VRB Systems (USA and Canada); medium-sized enterprises such as UET-Uni Energy Technologies (USA) in cooperation with Dalian Rongke Power (China); and a few startup companies with new ideas such as Enervault (Fe/Cr) (USA) and Volterion (compact welded stacks) (Germany).

2. The function of the VRFB

The electrochemical redox flow cell consists of two half-cells which are separated by a separator which can be an anionic exchange membrane, a cationic exchange membrane, or a porous membrane. The liquid electrolyte stores electrical energy in the form of chemical ions which are soluble in liquid aqueous or nonaqueous electrolytes. The electrolytes of the negative half-cell (anolyte) and the positive half-cell (catholyte) are each circulated by a pump in separate circuits. Both electrolyte recirculation circuits are separated by a separator. The function of the separator is to prevent electrical short circuits, prevent cross-mixing of the electrolyte, and to insure the ion exchange across the separator balances the electrical charge of the anolyte and catholyte (Fig. 12.4).

Figure 12.4 Working principle of vanadium redox flow batteries.

The ions that are exchanged depend on the kind of redox flow battery; the most common types are cationic exchange membranes such as NAFION®. These perfluorinated and sulfonated membranes have been used for decades and are very stable against chemical attack and oxidative corrosion caused by high potentials. These membranes are mostly used in acid electrolyte systems for vanadium redox flow cells or iron chromium cells. Charge balancing is easily done by the transport of hydrated protons (hydronium-ion) through the membrane. These membranes are available worldwide through several commercial suppliers and because they are fluorinated membranes the price is quite high.

A second type of membrane is the anionic exchange membrane. In this case the counter ion of the active species is responsible for the charge balance. These anionic-type membranes are in general cheaper, but chemical stability has to be carefully checked.

A third class of membranes is the so-called micro- or nanofiltration membranes. The working principle of these membranes is quite different because the function is based on ion exclusion. Small ions such as hydronium (H₃O⁺) or sulfate

(S O_{4}^{2 -})

$(S O_{4}^{2 -})$

ions can freely cross the membrane for charge balancing, while much larger ions such as vanadyl ions are too bulky to cross the membrane so cross-mixing of the electrolyte is prevented.

A number of variations and modifications to such membranes have been made during the past few years. One notable development involves a class of nonfluorinated membranes known as SPEEK membranes—sulfonated poly(tetramethyldiphenyl)etheretherketone—which show promising chemical stability and proton conductivity.

Some attempts have been made to implement inorganic fillers such as SiO₂ and ZrO₂ into the membrane, with the purpose of preventing vanadium ions from crossover by inducing a charge in the membrane which excludes positive charge vanadium ions. The advantage of ionic conducting membranes is that they are effective at preventing the crossover of electrolytes, and hence make for an effective and efficient cell.

Microporous ion exclusion membranes are an order of magnitude cheaper than cationic ion conducting membranes, but because of their relatively poor efficiency they have not been produced commercially.

The membrane separates only the anolyte and catholyte and allows for charge balance. The chemical oxidation and reduction that result from electron transfer take place at the electrode. In most cases the reaction takes place at a graphitic carbon felt. In the VRFB no extra catalyst is necessary, but activation of the graphite felt does help to accelerate the reaction. This is done by adding carboxcylic groups to the graphite surface using either heat treatment or chemical and electrochemical oxidation. Graphite is the material of choice because of its chemical resistance and low cost. The graphite felt must have a high surface area and good electrical conductivity. High surface area is necessary to provide enough reaction sites. Due to its operation in flowthrough mode the graphite felt must have a very open structure (95% void volume) with a thickness of (3–5) mm. The open structure of the felt is necessary to achieve a low-pressure drop between the electrodes. On the other hand, the open structure leads to a relatively high inner resistance of the overall stack in comparison with cells which are constructed in flowby mode as seen in fuel cells and electrolyzers.

As shown in Fig. 12.5 (left and right), respectively, compound-based bipolar plates separate the anolyte and the catholyte from each other. This arrangement gives the opportunity for electrical connection in series. The bipolar plates are often made of a compound-based material, which could be processed by hot pressing or injection molding. Hot pressing is often done with duroplastic resins; injection molding uses thermoplastic material such as polyethylene and polypropylene which are also low-cost materials. To achieve sufficient electrical conductivity a very high amount of graphite and carbon black up to 85% by mass is necessary. This, however, limits the ability to process and limits mechanical stability. In a flowthrough design no flow field is necessary.

Figure 12.5 Left, schematic view of a single redox flow cell; right, schematic view of a redox flow stack 60 cell.

The contact resistance between graphite felt and compound-based bipolar plate dominates the overall inner resistance of the stack; as a result much work has been done to lower contact resistance by gluing the graphite felt to the bipolar plates with conducting glue.

The role of the gaskets in a stack is often underestimated and several gaskets are usually needed to properly seal the stack. Gaskets on each side of the separator are used to insure a good seal toward the outside; furthermore, additional gaskets are needed to seal the bipolar half-plates and internal manifolds.

The thickness of the graphite felt, frames, and gaskets must be properly adjusted to insure good compression of the felt and sufficient strength given to the gaskets.

The gaskets are made from elastomeric materials such as ethylene propylene diene monomer (EPDM) rubber or fluoroelastomeric materials.

Single cells are stacked together to achieve higher voltages. The stack itself is compressed by strong end-plates and tension rods.

The vanadium electrolyte consists of vanadium salts which are dissolved in aqueous sulfuric acid. The liquid electrolyte corresponds to the active mass in a conventional battery. The amount of liquid electrolyte which is stored in tanks determines the capacity of the RFB. The big advantage of RFBs is that power and capacity can be scaled independently.

To operate an RFB additional pumps, piping, valves, and storage tanks are necessary. All such equipment must be able to withstand the harsh conditions of sulfuric acid and strong oxidizing power of vanadium ions in the valence state of 5.

To insure an efficient system, each vanadium redox flow system has a simple battery management program which controls the flow rate of pumps with respect to load requirements and state of charge.

The nominal charge voltage of each single cell is usually limited to 1.6 V to avoid the potential at which water is decomposed into oxygen and hydrogen. This is known as the oxygen evolution reaction (OER) which causes carbon corrosion that rapidly destroys the graphite electrodes [5–12]. The end of discharge voltage is kept to (0.9–1.0) V to reach a reasonable efficiency of the cell. It is worth mentioning that an electrochemical cell could be discharged down to 0 V without destroying the cell.

3. Electrolytes of VRFB

A vanadium-based electrolyte is widely used in flow batteries. This is due to the simplicity and stability of the electrolyte system in the aqueous phase. In an aqueous solution, four different but stable valence states of vanadium exists (V²⁺, V³⁺, V⁴⁺, and V⁵⁺). In anolyte vanadium (+2 and +3) ions exist as V²⁺, V³, while the +4 and +5 valence states of vanadium exist only as oxo-complexes

(V O^{2 +}, V O_{2}^{+})

$(V O^{2 +}, V O_{2}^{+})$

. By changing the valence states of vanadium species, energy could be stored electrochemically. These basic redox reactions are:

$V^{2 +} ⇌ V^{3 +} + e^{-} E^{O ―} = - 0 .255 V$ $V^{2 +} ⇌ V^{3 +} + e^{-} E^{O ―} = - 0 .255 V$

(12.1)

and

$V O^{2 +} + H_{2} O ⇌ V O_{2}^{+} + e^{-} + 2 H^{+} E^{O ―} = +1 .004 V$ $V O^{2 +} + H_{2} O ⇌ V O_{2}^{+} + e^{-} + 2 H^{+} E^{O ―} = +1 .004 V$

(12.2)

The oxidation of V²⁺ releases one electron and V³⁺ is formed. This creates a standard potential of 0.255 V. The oxidation of V⁴⁺ to V⁵⁺ by simultaneous splitting of a water molecule releases a proton and one oxygen atom which form the oxo complex. This delivers a standard potential of +1.004 V [13]. The overall standard potential for the reaction is 1.259 V.

In an aqueous electrolyte the vanadium salts in all the four different valence states—V⁺², V⁺³, V⁺⁴, and V⁺⁵—must be soluble in concentrations which should be as high as possible. The more vanadium salts held in a stable solution without precipitation the higher the volumetric energy density of the electrolyte. The vanadium salt in valence state 5 has the lowest solubility. The following equation describes the reaction equilibrium between solid vanadium pentoxide and vanadium +5 in solution [14]:

$\frac{1}{2} V_{2} O_{5} (s) + H^{+} ⇌ V O_{2}^{+} + \frac{1}{2} H_{2} O$ $\frac{1}{2} V_{2} O_{5} (s) + H^{+} ⇌ V O_{2}^{+} + \frac{1}{2} H_{2} O$

(12.3)

The higher the proton concentration (acid concentration) the more the equilibrium is shifted to the right side (principle of le Chatellier) and the more the V⁺⁵ vanadium (in the form of VO₂⁺) can be kept in the solution. Furthermore, the high proton concentration of the electrolyte results in high electrolyte conductivity which in turn leads to good cell performance. The sulfuric acid is usually at a concentration of between 2 mol L^–1 and 6 mol L^–1 [15].

Very recently mixed electrolytes of sulfuric and hydrochloric acids have been used as electrolytes. The addition of hydrochloric acid and therefore chlorine ions allows for very high vanadium concentrations in the solution at even higher temperatures. The higher stability of the electrolyte is caused by the formation of a chloro–oxo complex, and this stable complex prevents the condensation reaction and precipitation of vanadium pentoxide [16]. This mixed electrolyte has the advantage of higher energy density and temperature stability but, on the other hand, can result in possible release of poisonous chlorine gas at the positive electrode during the charge process. Furthermore, the presence of chlorine could compromise material stability in the cell.

4. VRFB versus other battery types

VRFBs like all other flow batteries are in competition with batteries such as lead–acid batteries. This competition is driven by technoeconomic needs for different applications. These storage device needs could be for:

• an uninterruptable power supply (UPS) for which total efficiency and cycle lifetime are not crucial, but capital expenditure (CAPEX), lifetime, and response time are the most important criteria

• home applications coupled with photovoltaics for which efficiency and cycle time are very important issues

• grid operators for which CAPEX, cycle lifetime, and efficiency are important criteria.

In Table 12.1 a list of the main specifications of major electrochemical storage systems are shown.

Table 12.1

Main Specs of VRFB Compared with Other Battery Types [17]

Battery type	Cycle lifetime (cycles)	Energy efficiency/(%)	CAPEX/ [€ (kW h)^–1]	Status
VRFB	20 000	70–85	500–650	Under development; near to market launch
Lead–acid	300–2 000	75–90	300–600	Commercial
NiCd	1 000	60–65	–	Commercial
NiMh	1 400	70	–	Commercial
Li-Ion	500–15 000	90–95	1 000–1 500	Commercial (consumer size), demonstration (huge, stationary)
NaS (high temperature)	3 000–7 000	70–85	130–230	Commercial
Zn–air	500	60–70	250–300	Under development

So far, we have discussed only a few applications and their specifications. To insure the best battery type for a particular application, all specifications must be considered. In the following section the VRFB is compared with other batteries for particular applications.

From an energy density (gravimetrical and volumetrical) point of view, the VRFB is low compared with zinc–air and lead–acid batteries. As a result the VRFB is more suitable for stationary applications. Mobile applications of VRFB could only be possible in niche areas such as on ferries.

Vanadium flow batteries have the highest cycle lifetime of all presently available batteries including lithium-ion batteries.

One big advantage of VRFBs is that they have a long life, because the liquid electrolyte does not degenerate to any great extent and can be used for decades without replacement. Electrodes made of graphite felt are also very stable, and furthermore membrane failure occurrences are extremely rare.

Another advantage of VRFBs is that self-discharging is extremely low due to the fact that self-discharging could only occur in the reaction chamber (cell) and not in separated storage tanks.

The energy efficiency of VRFBs is high and its mean value is better than that of Zn–air and NiCd batteries and is in the same range as that of NaS batteries, but less than that of Li-Ion batteries.

Two unique characteristics of flow batteries are: capacity is dependent on the size of storage tanks; and power is dependent on stack size. As a result any existing flow battery system can be improved and extended by adding additional electrolyte tanks.

These characteristics lead to flow batteries being used for stationary applications (low energy density) with high cycling rates (up to 365 full cycles per year) with a long-lasting lifetime and the capacity for long storage times. In short, flow batteries have high storage capacities in relation to power.

5. Application of VRFB

Different researchers have proposed a large number of different possible applications for VRFBs [18]. In Table 12.2 there are nine important properties, with the advantages/disadvantages of RFBs compared with other types of batteries [19].

Table 12.2

Assessment of Redox Flow and Batteries and Other Storage Devices Matched with Different Battery Properties

Application	Redox flow	Lead–acid	Sodium–sulfur	Lithium–ion	Supercapacitor
(1) Time shift	+	+	+	+	–
(2) Renewable integration	+	+	+	+	–
(3) Network investment deferral	+	±	+	+	–
(4) Primary control power	+	+	+	+	–
(5) Secondary control power	+	+	+	+	–
(6) Tertiary control power	+	+	+	+	–
(7) Power system startup	±	+	+	+	–
(8) Voltage support	±	+	+	+	+
(9) Power quality	±	±	±	±	+

From Table 12.2 it can be seen that RFBs are promising in all fast and bulk applications (1–6), but less useful in applications with fewer cycles (7—power system startup) and lower capacity (8 and 9), where supercapacitors are very promising. There is however serious competition with other batteries such as lead–acid, NaS and Li-Ion.

5.1. Applications

The most promising applications for redox flow batteries are:

1. Time shift applications

a. Economics-driven systems which charge the storage plant with inexpensive electric energy purchased from the grid during low-price periods and discharge the electricity back to the grid during periods of high price.

b. Technology-driven systems which charge the storage plant with surplus energy from the grid during low demand periods and discharge the electricity back to the grid (or island grid) during periods of high demand.

2. Renewable integration systems which assist in wind- and solar-generation integration by reducing output volatility and variability, reducing congestion problems, providing backup for unexpected generation shortfalls, and reducing minimum load violation.

3. Network investment deferral systems which postpone or avoid the need to upgrade the transmission and/or distribution infrastructure.

4. Primary, (5) Secondary, and (6) Tertiary control power: selling positive (discharging) and negative (charging) control power to grid operation in a timeframe of less than 30 s (primary), less than 5 min (secondary), or less than 30 min (tertiary control power).

5.2. Current Large-Scale Applications

In Japan the Tomamae wind farm on Hokkaido Island is linked to a 4 MW VRFB storage system which is used to smooth out wind-generated energy peaks and valleys. In addition to this renewable integration application, there is a showcase at Sumitomo’s head office in Osaka, which provides 3 MW peak shaving.

In China the Zhangbei National Wind and Solar Energy Storage and Transmission Demonstration Project is attached to a 2 MW VRFB provided by Prudent Energy.

In the United States, Prudent Energy is also responsible for the current largest VRFB storage system in the United States (0.6 MW) which is at Gills Onions, California. It is used to store energy produced from the methane generated from a biowaste plant [20].

In Germany a 2 MW RFB coupled to a wind turbine is being developed by Fraunhofer at Pfinztal [21]. Fraunhofer [22] is also developing a 0.5 m² RFB stack (planned for 25 kW) (Fig. 12.6).

Figure 12.6 Redox flow battery with 0.5 m² cell area at Fraunhofer UMSICHT [23].

These developments have led to discussions on the future of flow battery systems in terms of “numbering up” versus “scale up”. On the one hand, mass production (numbering up) will rapidly reduce the cost of battery modules while, on the other hand, scale up of stack size will also lead to lower cost per kilowatt and less installation demand. Currently, most stacks are at the one-digit kilowatt level and high capacities are realized by linking hundreds of stacks. With increasing market share the size of each stack will increase as demand for larger units grows.

Most present applications involve large-scale systems because VRFBs are getting cheaper as a result of upscaling than other systems. Moreover, they have in most cases a high capacity so that the time for charging or discharging takes a few hours (4–10 h). For VRFBs this power-to-capacity ratio is therefore between 1:4 and 1:10. A listing of current worldwide installations is given schematically in Fig. 12.7.

Figure 12.7 Redox flow battery projects in operation worldwide.
Here HH and MM refer to hours and minutes, respectively. (Data from Sandia [24].)

Despite this there are developments aimed at using VRFBs for small-scale applications. In Germany, for example, where the home solar market is becoming increasingly important, small batteries rated at (1–2) kW and (5–10) kW h are being developed for home applications by VOLTERION, for example (Fig. 12.8).

Figure 12.8 Stack for home applications designed by VOLTERION [25].

6. Recycling, environment, safety, and availability

One of the most important advantages of RFBs is that the electrolyte could be regenerated while operating and recycled after the lifetime of storage systems. While the stack consists mostly of uncritical material like graphite and plastic, which does not need recycling, the electrolyte (anolyte and catholyte) consists of vanadium and sulfuric acid and does need recycling. Vanadium is a high-priced material that can be almost 100% reclaimed, as can sulfuric acid [26]; from the economic point of view vanadium is the important component. Reclaimed vanadium can be used to produce new electrolyte for RFBs or for other purposes (steel industry).

From the environmental protection point of view, only VRFB electrolyte has to taken into account. This is because sulfuric acid is corrosive and vanadium is a heavy metal. As a result, double-wall storage vessels/catch basins and splash guards have to be provided for the whole system. In this respect the electrolytes of VRFBs can be compared with the electrolytes of lead–acid batteries.

From a safety point of view, VRFBs are safer than many other types of batteries and there is almost no risk of fire because of the larger amount of water present in the system. Furthermore, in case of a short circuit [27] or mixing of anolyte and catholyte (comparable with the “nail test” for lithium-ion batteries) there is only a minor exothermic reaction with less than a 1 °C temperature rise.

In light of being a crucial component the availability of vanadium has often been discussed. Vanadium is an important byproduct of a number of mining operations and is used almost exclusively in ferrous and nonferrous alloys. Vanadium consumption in the iron and steel industry represents about 85% of the vanadium-bearing products produced worldwide [28]. The global supply of vanadium originates from primary sources such as ore feedstock, concentrates, metallurgical slags, and petroleum residues. The main supplier countries are South Africa, China, Russia, but supplies are also exported from Canada, USA, Argentina, etc. From the German/European point of view, vanadium has been assessed as a medium critical metal by Erdmann et al. (2011) [29] (Fig. 12.9).

Figure 12.9 Criticality and vulnerability of some metals in Germany. (Data from Ref. [29].)

7. Other flow batteries

During the past 40 years nearly every chemically possible electrolyte combination has been evaluated as a suitable electrolyte for flow batteries. Due to limitations in chemical stability, energy density, poisoning, or radioactivity only a few electrolyte systems can be considered for practical flow battery application. These applications include hybrid flow batteries; in these batteries the anode is in a fully charged state and is usually a solid metal which dissolves during discharge to form the corresponding salt. Frequently used anode materials are zinc, iron, and possibly copper. In the following section the most important variations are briefly described, and in Table 12.3 the most investigated flow batteries are listed.

Table 12.3

Most Investigated Flow Batteries

System type/active material	Cell voltage/V	Chemistry	Electrolyte
Redox flow		Anode/cathode	Anode/cathode
Vanadium VRB	1.4	V²⁺/VO₂⁺	H₂SO₄/H₂SO₄
Vanadium–bromine	1.3	V²⁺/1/2Br₂	VCl₃/NaBr (HCl)
Polysulfide–bromine PSB	1.5	2S₂^2–/Br₂	NaS₂/NaBr (NaOH)
Iron–chromium	1.2	Fe²⁺/Cr³⁺	HCl/HCl
Hydrogen–bromine	1.1	H₂/Br₂	NaBr (NaOH)
Hybrid flow		Anode/cathode	Anode/cathode
Zinc–bromine	1.8	Zn/Br₂	Zn/ZnBr₂ (NaOH)
Zinc–cerium	2.4	Zn/2Ce⁴⁺	CH₃SO₃H/CH₃SO₃H

7.1. Iron–Chromium Flow Battery

One of the first flow battery electrolyte chemistries studied was the iron–chromium flow battery (ICB). It has been extensively studied by NASA (USA) and Mitsui (Japan). The iron–chromium battery is a real RFB with energy stored in Fe²⁺/Fe³⁺ and Cr²⁺/Cr³⁺ couples which are dissolved in hydrochloric acid. During discharge Fe²⁺ is oxidized to Fe³⁺ and simultaneously Cr³⁺ is reduced to Cr²⁺. To keep the overall charge in balance a proton is exchanged through the separator which separates the anolyte and catholyte.

In the original iron–chromium system, cross-mixing of the electrolyte was a serious problem. Over time the iron and chromium ions diffuse through the membrane, so an irreversible capacity loss occurs. To avoid this cross-mixing effect, expensive ion exchange membranes were used.

Modern iron–chromium batteries work with a mixed electrolyte which uses iron and chromium on both sides. This allows the use of inexpensive porous separators. The optimal working temperature of the iron–chromium flow battery is (40–60) °C, which is quite high for a battery and thus makes this battery suitable for hot climates. The electrolyte is cheap and nonflammable. One disadvantage is the possibility of hydrogen evolution, which causes a loss in efficiency.

7.2. Polysulfide Bromine Flow Battery

The polysulfide bromine (PSB) redox flow battery is a well-investigated battery type. The great advantages of this type of battery is the very cheap and abundant electrolyte and the high voltage of 1.5 V. The electrolyte is an alkaline solution of sodium polysulfide and sodium bromide as anolyte and catholyte, respectively. During charging and discharging, sodium ions exchange through an ionic exchange membrane to keep the charge in balance. The efficiency of the battery is about 75%. As in every bromine-based electrolyte the bromine must be dissolved in the electrolyte with the aid of a complexing agent. The reactions are:

$\begin{array}{l} Positive ele ctrod e : NaB r_{3} + 2 N a^{+} + 2 e^{-} \to 3 NaBr (discharge) \\ Negative electrode : 2N a_{2} S_{2} \to N a_{2} S_{4} + 2 N a^{+} + 2 e^{-} (discharge) \end{array}$ $\begin{array}{l} Positive ele ctrod e : NaB r_{3} + 2 N a^{+} + 2 e^{-} \to 3 NaBr (discharge) \\ Negative electrode : 2N a_{2} S_{2} \to N a_{2} S_{4} + 2 N a^{+} + 2 e^{-} (discharge) \end{array}$

The active species are highly soluble in aqueous electrolyte and therefore the electrolyte has a relative high energy density at low cost.

In 2002 Regenesys built a 15 MW, 120 MW h PSB flow battery system at Little Barford in the United Kingdom, but the project was never fully commissioned. The business was owned by RWE Power and it left the project before final commissioning.

7.3. All-Organic Redox Flow Battery

Very recently a new type of flow battery has been under development which involves organic molecules that are soluble in aqueous phase and could easily be oxidized and reduced.

A benefit of these organic flow batteries is that the electrolyte can be very cheap and not based on limited resources like vanadium. A promising candidate is the sulfonated anthraquinone redox couple. In 2013 researchers suggested the use of 9,10-anthraquinone-2,7-disulfonic acid (AQDS), a quinone, as a organic redox molecule in metal-free flow batteries [30]. AQDS easily undergoes rapid and reversible two-electron two-proton reduction at a carbon electrode in sulfuric acid. Each of the carbon-based molecules holds two functional groups which can be oxidized and reduced. This is a promising research area as it has the potential to offer a low-cost flow battery electrolyte. By modifying the chemical structure of the basic anthraquinone molecule, solubility in the aqueous phase could be increased. Moreover, the potential could be shifted even higher.

7.4. Hybrid Flow Batteries

7.4.1. Zinc–Bromine Flow Battery

The zinc–bromine flow battery is a so-called hybrid flow battery because only the catholyte is a liquid and the anode is plated zinc. The zinc–bromine flow battery was developed by Exxon in the early 1970s. The zinc is plated during the charge process. The electrochemical cell is also constructed as a stack. Storage capacity is determined by the size and thickness of the plated zinc plate and of the catholyte storage reservoir, and as a result the power rating and capacity correspond to each other.

The catholyte contains an organic complexing agent to keep the generated bromine in solution during the charging process. A microporous separator is used in most cases. During charging, zinc is plated on a carbon composite plate. The morphology of the plated zinc is strongly related to current density, temperature, and flow velocity. At high current densities, zinc tends to dendritic growth which might cause short circuits through the separator. During charging, bromine is generated. Bromine is highly oxidative and is a poison. Its solubility in water is limited, so to increase the solubility an organic complexing amine is added; this interacts with the bromine to keep it in solution. The organic dense phase behaves like oil and forms a separate phase; this has to be considered for a system layout. An important issue is the toxicity of the bromine. Its high oxidative power necessitates the use of chemically resistant parts for the flow battery, which are expensive. Temperature stability of the complexed bromine is also an issue, since temperature must be kept below 50 °C.

7.4.2. Zinc–Cerium Flow Battery

The zinc–cerium battery is a nonaqueous battery. It is an important battery because of its high potential. The electrolyte used is methanesulfonic acid. The high potential of the catholyte cerium requires the use of very expensive electrode materials (titan electrodes and precious metal coatings) for the cathode. Graphite felt cannot be used for the reaction as the cathode side. It would be oxidized because of the high potential. At the anode, zinc is electroplated on and stripped off the carbon polymer electrode during charge and discharge, respectively [31–33]:

$Z n^{2+} \underset{(aq)}{} + 2 e^{-} \leftrightarrow Z n_{(s)} (- 0.76 V vs . SHE)$ $Z n^{2+} \underset{(aq)}{} + 2 e^{-} \leftrightarrow Z n_{(s)} (- 0.76 V vs . SHE)$

At the positive electrode (cathode), Ce(III) oxidation and Ce(IV) reduction take place during charge and discharge, respectively:

$C e^{3 +} \underset{(aq)}{} - e^{-} \leftrightarrow C e^{4 +} \underset{(aq)}{} (ca .+1 .44 V vs . SHE)$ $C e^{3 +} \underset{(aq)}{} - e^{-} \leftrightarrow C e^{4 +} \underset{(aq)}{} (ca .+1 .44 V vs . SHE)$

Because of the large cell voltage, hydrogen (0 V vs. SHE) and oxygen (+1.23 V vs. SHE) could evolve theoretically as side reactions during battery operation (especially on charging). The positive electrolyte is a solution of cerium(III) methanesulfonate.

Due to the high standard electrode potentials of both zinc and cerium redox reactions the open-circuit cell voltage is as high as 2.43 V. Methanesulfonic acid is used as electrolyte, as it allows high concentrations of both zinc and cerium; the solubility of the corresponding methanesulfonates is 2.1 mol for Zn, 2.4 mol for Ce(III), and up to 1.0 mol for Ce(IV). Methanesulfonic acid is particularly well suited for industrial electrochemical applications and is considered to be a green alternative to other support electrolytes.

7.4.3. Iron/Iron Flow Battery

One simple approach is the all-iron hybrid flow battery which uses a very cheap electrolyte: $7 (kW h)^–1. Iron is plated at the anode and the Fe²⁺/ Fe³⁺ is in the form of a complex in alkaline solution. The iron electrode is well known and has been used for decades in the nickel–iron battery; the reaction is highly reversible and very stable. As the catholyte, ferro/ferricyanide can be used; this is also a well-investigated reaction and known as a stable redox system. The kinetics of the reaction are fast, so high current densities up to 200 mA cm^–2 could be achieved. The cost should be about $150 kW^–1. The reactions involved are:

$\begin{array}{l} F e^{2 +} \to F e^{3 +} + e^{-} + 0.77 V \\ F e^{2 +} + 2 e^{-} \to F e^{0} - 0.41 V \end{array}$ $\begin{array}{l} F e^{2 +} \to F e^{3 +} + e^{-} + 0.77 V \\ F e^{2 +} + 2 e^{-} \to F e^{0} - 0.41 V \end{array}$

A challenge is hydrogen evolution as a side-reaction; this reduces the efficiency of the system [34].

7.4.4. Copper/Copper Flow Battery

A relatively simple approach is the use of an all-copper hybrid flow battery. The idea is to stabilize Cu (I) as a chloro complex in solution using suitable anions such as chloride or amine. There are three different valence states of copper available for use with this battery.

One interesting effect is that this battery could in principle be recharged by applying higher temperatures in which case the Cu(I) complex becomes unstable and disproportionate to metallic copper and Cu(II), which is the starting material of the charged battery.

The energy density (20 W h L^–1) achieved is comparable with traditional VRFBs. This is due to the high solubility of copper (3 M), which offsets the relatively low cell potential (0.6 V). The electrolyte is cheap, simple to prepare, and easy to recycle since no additives or catalysts are used. The system can be operated at 60 °C eliminating the need for a heat exchanger and delivers energy efficiencies of 93, 86 and 74% at 5, 10, and 20 mA cm^–2, respectively [35].

7.4.5. Hydrogen–Bromine Battery

The hydrogen–bromine battery works with sodium–bromine in alkaline solution, which is a low-cost and well-known electrolyte. The combination with the hydrogen evolution reaction (HER) and hydrogen reduction reaction (HRR) has the advantage of being very fast with low overpotential in combination with the high oxidative power of bromine. These advantages are tempered by disadvantages such as on both sides catalysts are needed to enhance the reaction. New developments include working with non-noble catalyst systems, but the state of the art does involve precious metal catalysts at the anode as well as at the cathode. To insure reasonable storage capacity the hydrogen must be stored under pressure. This could partly be achieved by hydrogen evolution in the stack of up to 1 MPa (10 bar) to 2 MPa (20 bar). For higher pressures a compressor is needed, making the overall system both complex and expensive [36].

Very recently, the Israeli company EnStorage has developed such a system with a target capacity of 150 kW and 900 kW h.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.

Table of Contents for Chapter 12: Vanadium Redox Flow Batteries

Create new playlist

Sign In

Sign Up