Redox Flow Battery For Energy Storage - Sumitomo Electric ...

Copy and paste this link to your website, so they can see this document directly without any plugins.


power, battery, system, output, energy, batteries, Power, Battery, Redox, with, development, wind, storage, Energy, Wind, that, which, Flow, generation, Electric, ions, such, redox, capacity, used, cell, Storage, voltage, electrolyte, practical


4 · Redox Flow Battery for Energy Storage
1. Introduction
To realize a low-carbon society, the introduction of
renewable energies, such as solar or wind power, is increasingly being promoted these days worldwide. A major challenge presented by solar and wind power generators is
their fluctuation in output. If they are introduced in large
numbers to the power system, problems, such as voltage
rises, frequency fluctuations and surplus of the generated
power, are predicted to occur. As a solution to these problems, energy storage technologies are attracting attention,
amongst which energy storage batteries are expected to
become indispensable for use. Various energy storage batteries are being developed and many application verification projects using such batteries are currently being
promoted. Thus, expectations are growing for their practical use to the power system in near future. The redox flow
(RF) battery, a type of energy storage battery, has been
enthusiastically developed in Japan and in other countries
since its principle was publicized in the 1970s(1). Some such
developments have been put into practical use. This paper
reviews the history of the RF battery’s development, along
with the status quo of its use.
2. The Necessity of Energy Storage and
Its Technologies in Practical Use(2)
As an energy storage technology that has long been
used in the power system, pumped hydro energy storage is
widely known. Occupying about 10% of the total power
generation capacity, it functions as a load leveler; namely,
it levels the power load by storing power during off-peak
hours and discharging it during peak hours. Variable-speed
pumped hydro energy storage, which can vary the rotating
speed of a pump, is currently in practical use. Some
pumped hydro systems have a sophisticated power system
stabilization function of frequency regulation or others. As
other energy storage technologies, energy storage batteries, superconducting magnetic energy storage (SMES), flywheels, compressed air energy storage (CAES), and electric
double-layer capacitors (EDLC) are known. They are being
developed for various applications to make effective use of
their individual characteristics, and some of the developments have been put into practical use.
An advantage of SMES is that they are high in energy
storage efficiency and can discharge a large amount of
power instantaneously, since they store electric energy as
it is. They are expected to be put into practical use in the
near future, as electric power companies and national projects are conducting their verification tests. Chubu Electric
Power Co., Inc. is field-testing a 5 MVA SMES at a liquidcrystal factory. This SMES, used for instantaneous voltage
sag compensation, is among the world’s largest. In other
countries, such as the United States, SMES is already commercialized for the power system stabilization and for instantaneous voltage sag compensation. Regarding flywheel
technologies, the Fusion Institute of the former Japan
Atomic Energy Research Institute (currently the Japan
Atomic Energy Agency) has a flywheel power generator
with the world’s largest energy storage capacity (8 GJ or
2,200 kWh). The generator is used as a magnetic field coil
power supply. The Okinawa Electric Power Co., Inc, has a
23 MW flywheel for frequency regulation. The CAES is a
technology that compresses air and stores it in an underground hollow space, and generates power in combination
with a gas turbine generator where necessary. It is in practical use at some power stations in Germany and the United
States. EDLC technology has a characteristic of instantaneous large output because, like capacitors, it charges and
discharges electricity by absorbing and desorbing electric
charge without any chemical reaction taking place. The
technology is also advantageous in that it is maintenancefree. Recently, EDLC products with large capacities that
can be used for electric-power facilities are commercially
available, and are also being used in such applications as
instantaneous voltage sag compensation, absorption of
regeneration energy and voltage regulators for electricrailway, and natural-energy-generation output fluctuation
Battery energy storage technology is superior in technical integrity to the above energy storage technologies
and has excellent practicality because it can be installed
and distributed in suburban areas. It is thus a highly promising technology.
Renewable energies, such as solar and wind power, are increasingly being introduced as alternative energy sources on
a global scale toward a low-carbon society. For the next-generation power system, which uses a large number of these
distributed power generation sources, energy storage technologies will be indispensable. Among the energy storage
technologies, battery energy storage technology is considered to be most viable. In particular, a redox flow battery,
which is suitable for large scale energy storage, has currently been developed at various organizations around the
world. This paper reviews the technical development of the redox flow battery.
Keywords: redox flow battery, energy storage, renewable energy, battery, vanadium
Redox Flow Battery for Energy Storage
3. Battery Energy Storage
Table 1 shows the varieties of energy storage batteries
and their individual characteristics(3). Among them, lead
acid batteries have the longest history and are extremely
common for use in automobiles. In some overseas countries, such as the United States, Germany and Puerto Rico,
lead acid batteries have been used as energy storage facilities as several aged application examples indicate. In
Puerto Rico, 20 MW (40 minutes) lead acid batteries were
introduced to regulate frequency and to provide spinning
reserve. With the recent increase in demand for energy
storage batteries, not only lead acid batteries but also various other types of batteries are being enthusiastically developed for practical applications worldwide, to make the
best use of the characteristics of the individual batteries.
Among them, sodium sulfur (NaS) batteries have excellent
features, such as high energy density and superior charge/
discharge efficiency, and have been used in many practical
applications in Japan and in other countries. They are used
not only for load leveling in power substations and industrial plants but also for use in combination with solar and
wind power generation. At the Futamata Wind Power Station in Aomori Prefecture, Japan, 34 MW NaS batteries are
installed along with 51 MW wind power generation facilities.
4. Redox Flow Battery for Energy Storage
The word redox is a combination of, and thus stands
for, reduction and oxidation. A redox battery refers to an
electrochemical system that generates oxidation and reduction between two active materials, forming a redox system,
on the surface of inactive electrodes (the electrodes themselves do not change). A redox flow (RF) battery has the
electrolyte including these active materials in external containers, such as tanks, and charges and discharges electricity by supplying the electrolyte to the flow type cell by
pumps or other means.
4-1 Principle, configuration and characteristics of RF
(1) Principle and configuration of an RF battery
As shown in Fig. 1, an RF battery consists mainly of a
cell where the redox reaction occurs, positive and negative
electrolyte tanks in which active material solution is stored,
pumps and piping that circulate the electrolyte from the
tanks to the cell. It is interconnected with the AC power
system via an AC/DC converter.
Metal ions that change valence can be used in a redox
system; however, in light of such factors as energy density
and economy, the iron (Fe2+/Fe3+)–chromium (Cr3+/Cr2+)
system and the vanadium (V2+/V3+–VO2+/VO2+) system are
considered feasible redox systems. The V–V system is especially advantageous because it uses the same metal ions at
both the positive and negative electrodes and the battery
capacity does not decrease even when the positive and negative electrolytes are mixed each other through the membrane, while in the case of a system using two different
metal ions, such as iron and chromium, the battery capacity decreases if the two electrolytes are mixed. The V–V system is thus currently widely developed around the world.
VO2+ V3+
Electrode Separating
Cell+ −
ePump Pump
AC/DC converter
Discharge LoadPower
Fig. 1. Principle and Configuration of an RF Battery
Table 1. Energy Storage Batteries
Battery variety Redox flow NaS Lead acid Lithium ions Nickel hydride Zinc bromide
Active material
(positive/negative) V ions/V ions S/Na
Metal compound oxides
containing Li ions/Carbon
Nickel oxyhydroxide/
Hydrogen-storing alloy
Theoretical energy
density (Wh/kg) 100 786 167 392~585 225 428
voltage/cell (V) 1.4 2.1 2.1 3.6~3.8 1.2 1.8
temperature (°C)
temperature About 300
Major accessory Circulation pump Heater Not in particular Not in particular Not in particular Circulation pump
• Independently
designable output
and capacity
• Easily measurable
state of charge
• Reusable electrolyte
• Most common
storage batteries
• Useful under
high temperatures
• Many records
of use such as
in cars and
• Many records
of use as small
• Many records
of use as small
The electrode reaction of the vanadium system can be
expressed by the following formulae:
• Positive electrode
VO2+ (tetravalent) + H2O ⇆ VO2+ (pentavalent) +
2H+ + e−: E0 = 1.00 V (1)
• Negative electrode
V3+ (trivalent) + e− ⇆ V2+ (bivalent): E0 = −0.26 V (2)
In these formulae, the reaction from left to right represents the reaction during charging. At the positive electrode in the cell, tetravalent V ions (VO2+) are oxidized to
pentavalent V ions (VO2+) while at the negative electrode,
trivalent V ions (V3+) are reduced to bivalent V ions (V2+).
The hydrogen ions (H+) generated at the positive electrode
during charging move to the negative electrode through
the membrane to maintain the electrical neutrality of the
electrolyte. Supplied electric power is thus stored in the
form of the transformation of V ions of differing valence.
During discharging, the stored power is delivered by the
reverse reaction. The RF battery’s electromotive force is dependent on the redox system used. In the case of the vanadium redox system, the standard electromotive force
calculated based on the standard oxidation reduction potential (E0) is 1.26 V. However, when the electrolytes and
cells are prepared practically, the electromotive force is
about 1.4 V.
A structure of two or more cells stacked is called a cell
stack. Figure 2 shows a representative cell stack structure,
and Fig. 3, the cross-section structure of such a cell stack.
The voltage of a single cell is only 1.4 V at its highest, and to
realize high voltage for practical use, many battery cells need
to be connected in series. As to the connection method, the
serial stacking method using bipolar plates, which resemble
the method used in fuel cells, is employed. The role of the
cells is to realize the efficient oxidation and reduction reaction of vanadium ions in the electrolyte. As in an electric circuit element, the cell should preferably have a low internal
resistance. In addition, it is desirable that the oxidation and
reduction in the cell do not involve side reactions. The cell
components include the electrodes, a membrane, the bipolar plates, and a frame that houses these components to
form a cell, as shown in the figure. Because an acidic solution is used as the electrolyte, the materials in contact with
the electrolyte should be resistant to acid.
(2) Characteristics of RF batteries
RF batteries have the following characteristics and can
be used in various applications.
(1) The battery reaction principle is simply the change
in valence of the metal ions in the electrolyte, realizing a long charge/discharge cycle service life.
(2) The output section (cells) and capacity section
(tanks) are independent of each other and can be
optimally designed according to application needs.
(3) Maintenance is easy mainly because the same electrolyte is supplied to individual cells and therefore
the state of charge (SOC) in each cell does not
need to be monitored, and because heat can be
controlled easily based on the flowing electrolyte.
Since the SOC can be easily monitored by measuring the potential of the electrolyte, the SOC can be
monitored continually during operation.
(4) The electrolyte is stored in the positive and negative
tanks separately, so that no self-discharge occurs during stand-by and stoppage, except in the cell section.
(5) RF batteries are useful to absorb irregular, shortcycle output fluctuations, such as in natural energy
generation, because they have the characteristic of
instantaneous response in an order of milliseconds
and can charge and discharge at an output rate a
few times larger than the designed rating for a short
period of time(4).
(6) The electrolyte is friendly to the environment, because it scarcely changes during normal operation
except for changes in ion valance, and it can be
used virtually permanently and reused.
The disadvantages of RF batteries are as follows:
(7) Because an RF battery uses metal ions dissolved in
a solution as its active material, solubility is limited,
and hence the volume of the tank section is by necessity large, and the energy density is relatively
small compared with other energy storage batteries.
(8) Pumping power is necessary to circulate electrolyte
to the cells.
(9) Shunt-current losses may occur through electrolyte.
4-2 History of the development of RF batteries
Table 2 summarizes the history of the development of
RF batteries. Full-scale development of the batteries started
6 · Redox Flow Battery for Energy Storage
Intermediate frame
(vinyl chloride)
(carbon felt)
Separating membrane
(ion exchange
Bipolar plate
(plastic carbon)
(Fluid outlet)
(vinyl chloride)
(Fluid inlet)
End plate (iron)
End frame
(Conductive terminal side)
Fig. 2. Typical Cell Stack Structure (Sumitomo Electric)
1 cell
Positive electrode
Serial connection
1 cell 1 cell
Separating membrane
Bipolar plate
Fig. 3. Cross-Section Structure of Cell Stack (Sumitomo Electric)
in the 1970s. The principle of the RF battery system was
presented by L. H. Thaller of the National Aeronautics and
Space Administration (NASA) of the United States in 1974(1).
NASA mainly conducted research on the Fe/Cr system,
discontinuing it in 1984 with the publication of the Final
Report(5). At the same time in Japan, the Electrotechnical
Lab. (ETL; currently the National Institute of Advanced
Industrial Science and Technology) was conducting basic
research, and the development of the Fe/Cr system made
progress as a project of the New Energy and Industrial
Technology Development Organization (NEDO). In about
1985, in Australia, the University of New South Wales
(UNSW) proceeded to develop the V system.
Concerning the practical use of RF batteries, electric
power companies and manufacturers in Japan jointly conducted research with enthusiasm, and in about 2001, the
V system was partially put into practical use. The summary
of the development of the Fe/Cr system and V system is
explained below:
(1) Iron–Chromium (Fe/Cr) system
Around 1980 in Japan, expectations grew regarding
the development of large-capacity energy storage batteries
that would complement pumped hydro energy storage to
improve the load factor, which was getting lower at that
time, by load leveling. In NEDO’s Moonlight Project, the
development of four advanced batteries, including RF,
sodium/sulfur, zinc/bromine, and zinc/chlorine batteries,
started. Among the batteries, research into RF batteries was
conducted mainly by the ETL(12). The laboratory conducted basic research on many possible redox pairs, and
proceeded with practical research into the Fe/Cr system
using hydrochloric acid solution(6)-(10).
Along with these elemental technology developments,
Mitsui Engineering and Shipbuilding Co., Ltd. (MES)
manufactured and tested 10 kW and 60 kW system prototypes as part of the NEDO project during 1984 to 1987(11).
Kansai Electric Power Co., Inc. (KEPCO) and Sumitomo
Electric Industries, Ltd. (Sumitomo Electric) also started
to develop RF batteries in 1985 on their own, and tested a
60 kW class Fe/Cr system RF battery in 1989(12), (13).
The Fe/Cr system has the following problems: the Cr
ions’ electrode reaction is slow; because the different metal
ions are used in positive and negative reactions, each ion
is mixed through the membrane and thus gradually decrease the battery capacity; the Cr ions’ redox potential is
close to the hydrogen gas generation potential and a small
amount of hydrogen gas is generated from the negative
electrode near the end of the charge, thereby reducing the
battery capacity because of differences in the SOC between
the positive and negative electrodes. KEPCO and Sumitomo Electric theoretically solved the problem of the mixture of redox ions between the positive and negative
electrodes by using a single-fluid Fe/Cr system(14) in which
Fe ions and Cr ions are mixed in both the positive and negative electrodes. To solve the problem of the generation of
hydrogen gas, the electrode characteristics were improved,
and various types of accessories known as rebalancing systems,
which adjust the SOC for both the positive and negative
electrodes in the long run, were proposed.
With the aim of improving the energy density of the
Fe/Cr system, MES replaced Fe ions with Br ions for the
positive electrode, and researched into Cr/Br systems(15).
Likewise, the ETL and Ebara Corp. jointly investigated the
feasibility of the Cr/Cl system(16). Furthermore, the V/O2
system, in which air is used on the positive electrode, was
(2) Vanadium system (V/V system)
In Australia, where vanadium resources are abundantly available, Prof. Maria Kazacos of the UNSW proposed V system RF batteries, which use V ions at both the
positive and negative electrodes, around 1985(18)-(20), and
applied for a basic patent in 1986(21). In Japan, which has
no natural vanadium resources, V system RF batteries were
not researched into enthusiastically for economic reasons.
However, Kashima Kita Electric Power Corp. (Kashima
Kita) and the ETL. developed the technology of recovering
vanadium included in petroleum and heavy fuels from the
soot of the fuels burned at thermal power plants. Thus the
economic value of V system RF batteries was reviewed and
their development started in the country(22). The electromotive force of the V system was approximately 1.4 V,
which was 1.4 times as large as that of the Fe/Cr system, so
that, provided that the cells and energy efficiency were the
same, the output was double. Because the electrode reaction of V ions was comparatively fast in practical use, the
output was found to be several times as large. The system
used V ions at both the positive and negative electrodes;
therefore, even if ions were mixed between the positive and
negative electrodes through the membrane, the battery capacity did not decrease, in contrast to the Fe/Cr system.
Table 2. History of RF Battery Development
1949 Kangro (German patent): Cr/Cr and other systems
1974 Battelle: Cr/Cr, Fe/Cr, V, Mo, Mn and other systems
1974 NASA released the principle of the RF battery—U.S. basic
patent (’75)
• Fe/Cr system 1 kW (’78), Final Report (’84)
ETL started the research and development of RF Battery.
1980 NEDO (Moonlight Project) established the project
“Advanced Battery Electric Power Storage System.”
• RF (ETL./Mitsui Engineering and Shipbuilding [MES]),
NaS (Yuasa Battery), Zn/Br (Meidensha), and Zn/Cl2
(Furukawa Electric)
• ETL, Fe/Cr system, 1 kW (’82); MES, 60 kW (’84 to’87)
NEDO (Sunshine Project)
• RF battery for solar power generation (MES and Ebara)
1985 University of New South Wales (UNSW; Australia) released
the V system RF battery and applied a basic patent (’86).
1989 ETL. and Kashima Kita Electric Power developed V system
RF battery for the use of vanadium from the soot
• V system, 1 kW (Ebara, ’90); 10 kW (MES, ’91); 200 kW
(Kashima Kita, ’97)
KEPCO and Sumitomo Electric
• Fe/Cr system, 60 kW (’89); V system (450 kW, ’96)
1998 ETL. and Kashima Kita
• 10 kW Redox Super Capacitor on-vehicle test
2001 Sumitomo Electric put V system RF battery into practical use
(for load leveling, instantaneous voltage sag compensation
and emergency use).
NEDO verified the RF battery for stabilizing the wind power
output fluctuation. Sumitomo Electric: 170 kW (’00), 6 MW
2011 The development of RF batteries is proceeding worldwide,
including in the U.S., Europe and China.
The redox potential of V ions at the negative electrode was
higher than that of Cr ions, so that hydrogen gas generation was extremely small, which did not need the rebalance
system in practical use, this was also a great advantage. The
development of V system RF batteries started in Japan in
earnest in about 1989 because of these advantages and due
to the applicability of the battery technology of the Fe/Cr
In 1997, Kashima Kita manufactured a 200 kW/800
kWh system on trial(23). KEPCO and Sumitomo Electric
manufactured a 450 kW/900 kWh system in 1996(24).
Thereafter, the development of small-capacity system for
installation at consumers was proceeded(25), (26), and in
2000, a 100 kW/800 kWh system developed for buildings
was actually installed in an office building and verification
operations were conducted(27), (28). Sumitomo Electric developed practical products in 2001, and supplied products
for various uses, such as for load leveling, instantaneous
voltage sag compensation, and emergency power supply(29).
4-3 Application cases of an RF battery
The applications of an RF battery include not only
load leveling, which was the initial aim of the development,
but also instantaneous voltage sag compensation and emergency power supply at the sites of consumers; stabilization
of output fluctuation for natural energy sources such as
wind and solar power generation, which is recently becoming increasingly common; and frequency regulation in the
power system for high-quality electric power supply. Table 3
shows the test systems and practically used systems for
which Sumitomo Electric supplied products, along with
their application purposes. Following the table, the distinctive applications are explained.
(1) Load leveling system
At first when the development of RF batteries started,
the purpose was to develop large-capacity energy storage
batteries for installation at power substations, to level the
load and improve the load factor; however, the first example of their practical use was a system installed at a consumer. Consumers can reduce the contract electric power
and use inexpensive nighttime power by storing power during nighttime when the demand is low and discharging
power from the battery during daytime peak hours to accommodate the peak power demand, thus reducing electric charge cost and, in some cases, reducing the size of
power-receiving facilities. From the perspective of power
suppliers, RF batteries enable power supply facilities to be
used more efficiently as they level the electric power load,
benefitting both the consumer and the supplier. Photo 1
and 2 show a case of RF battery application at a university(30). Photo 1 shows battery cubicles that house 12 battery
cell stacks installed on the first floor of the storehouse.
Photo 2 shows the electrolyte tanks and pumps installed in
the basement. Each electrolyte tank, made of rubber, is
about 4 m in height and has a net capacity of 31 m3, and
each is installed in an iron frame. The system consists of
three banks, one of which includes four cell stacks of AC
168 kW × 10 hours in capacity, and has an output of AC
500 kW and a capacity of 5,000 kWh.
(2) Case of instantaneous voltage sag compensation system
At semiconductor plants and other factories, an instantaneous voltage sag may damage products in process. Con8 · Redox Flow Battery for Energy Storage
Photo 1. 500 kW System for Load Leveling (Battery Cubicles)
Customer or owner Application Output capacity Year of delivery
Electric power company Research and development 450 kW × 2H 1996
Office building Research and development(load leveling) 100 kW × 8H 2000
Electric power company Research and development 200 kW × 8H 2000
Wind power output
fluctuation stabilizing
verification (single unit)
170 kW × 6H 2000
Constructor Research and development(combination with solar power) 30 kW × 8H 2001
Instantaneous voltage
sag compensation,
peak-cut control
3 MW × 1.5sec
(1.5MW × 1H) 2001
Electric power company Research and development 250kW × 2H 2001
University Load leveling 500 kW × 10H 2001
Laboratory Research and development 42 kW × 2H 2001
Electric power company Research and development 100 kW × 1H 2003
Office building Load leveling 120 kW × 8H 2003
Instantaneous voltage
sag compensation,
load leveling
55 kW × 5H 2003
Railway company
Research and development
(load leveling, instantaneous
voltage sag compensation)
30 kW × 3H 2003
Office building Research and development 100 kW × 2H 2003
Data center
Instantaneous voltage
sag compensation,
emergency power supply
300 kW × 4H 2003
Laboratory Load leveling 170 kW × 8H 2004
Office building
Load leveling, emergency
power supply for firefighting equipment
100 kW × 8H 2004
Load leveling, emergency
power supply for firefighting equipment
125 kW × 8H 2004
Electric power company Research and development 152 kW × 2.6H 2005
Load leveling, emergency
power supply for firefighting equipment
120 kW × 8H 2005
Electric power company Research and development(combination with solar power) 100 kW × 4H 2005
Wind power output
fluctuation stabilizing
verification (wind farm)
4 MW × 1.5H 2005
Table 3. RF Battery Application Examples
sidering the loss of business opportunities resulting from
a facility reset, the loss might be huge. When used in such
an application, an RF battery system is required to quickly
respond to instantaneous voltage sags and supply electric
power to important loads during the moments of such sags.
Because an RF battery system has an immediate high output characteristic and its tank capacity can be designed
flexibly according to the required capacity, economical
design is possible according to the requirements. The load
leveling function and the peak-cut function can also be
provided where necessary.
Photo 3 and 4 show an application at a liquid crystal
factory(31). The major specifications of the RF battery system
are shown in Table 4. The cell stacks are installed in the
battery cubicle on the second floor of the building, and the
electrolyte tanks, which are made of polyethylene (30 m3 ×
8 units), are installed on the first floor. Normally they perform peak-cut operation at 1,500 kW, and when a voltage
sag occurs, they discharge 3,000 kW for 1.5 seconds.
(3) Application of RF batteries in combination with wind
power generation
Energy storage batteries are expected to be a good
solution when renewable energy, such as solar and wind
power generation, is introduced in large amounts to the
power system, and various verification projects are currently underway. Expectations are becoming higher for the
introduction of energy storage batteries in near future.
Regarding RF batteries, NEDO performed verification
tests by installing energy storage batteries to wind power
generator facilities to see if their output fluctuations could
be smoothed as expected. Generally speaking, wind power
output fluctuations vary periodically, ranging from milliseconds to hours. RF batteries can be designed to either reduce or increase the battery capacity by adjusting the
amount of electrolyte, thus satisfying the needs for large or
small capacity. In particular for short-frequency fluctuation, RF batteries are expected to improve economic efficiency through design, taking advantage of their high-rate
output characteristics. In the following, the summary of
verification tests is explained.
(a) Application to single unit of wind power generation
In the fiscal 2000 NEDO project entitled “Investigation
for Introducing Battery Energy Storage System to a Wind
Power Generation(32),” three types of energy storage batteries (RF batteries, NaS batteries, and lead acid batteries)
were installed along with wind generators for testing.
Among the batteries, the RF batteries were tested by the
Institute of Applied Energy (IAE), which was entrusted by
NEDO. A system of 170 kW (maximum 275 kW) in output
ratings and 1,020 kWh in capacity was installed in the
Horikappu Power Station of the Hokkaido Electric Power
Co., Inc.. This system enabled a test that used the immediate
high-rate output characteristics of RF batteries by adjusting
the AC/DC converter output to the rating of the wind
power generator of 275 kW.
Figure 4 shows a conceptual configuration diagram of
the system. The battery was installed at the wind power generation interconnection point, and smoothed the wind
Photo 2. 500 kW System for Load Leveling (Electrolyte Tank)
Table 4. Specifications of Instantaneous Voltage Sag Compensation System
Output capacity
During peak shift operation 1,500 kW × 1 h
During instantaneous voltage
sag compensation operation 3,000 kW × 1.5 s
Cell configuration (100 cells × 4 stacks in series × 3 banks in parallel) × 3 systems
Electrolyte Sulfuric acid aqueous solution including vanadium at 1.7 mol/L
Electrolyte tanks Polyethylene tank 30 m3 × 8 units
Photo 3. Instantaneous Voltage Sag Compensation 3 MW System
(Battery Cubicles)
Photo 4. Instantaneous Voltage Sag Compensation 3 MW System
(Electrolyte Tanks)
power output fluctuation by absorbing their fluctuations.
As a smoothing method, the actual wind power output was
passed through a low-pass filter characterized by a given
time constant, and short-period components were thus removed. The obtained output value was set as the reference.
An output equal to the difference between this reference
and the actual wind power output was delivered from the
battery. The output combining the battery output and the
actual wind power output was sent to the power system.
Figure 5 shows an example of data obtained when the
smoothing time constant is one hour. While wind power
output fluctuates little by little in the order of seconds, the
combined output sent to the power system is favorably
smoothed. Figure 6 shows the output data of one day,
which was obtained with the smoothing time constant set
to one hour(33). The changes in the remaining battery
capacity during the period are also shown in the space at
the bottom of the diagram. The larger the time constant,
the better the smoothing level of the output fluctuation.
However, the battery output needs to be larger when the
time constant is greater, and thus a greater battery capacity
is necessary. The remaining battery capacity shown in the
diagram is calculated based on the data of the electrolyte
potential monitor cell, obtained by using the advantage of
the RF battery, which enables the online monitoring of the
remaining battery capacity.
(b) Application to wind farm power generation
As part of a NEDO project entitled the “Development
of Technologies for Stabilization of Wind Power in Power
Systems(34),” the Electric Power Development Co., Ltd. (JPower) built an RF battery system (rated output: AC 4,000
kW/6,000 kWh, maximum output: 6,000 kW) as an annex
to Hokkaido Tomamae Wind Villa Power Plant (output:
30,600 kW, wind power generators: 19 units, start of operation: December 2000) in fiscal 2003, to test and verify
wind power output fluctuation smoothing(35)–(40).
Photo 5 and 6 show the battery cubicles, cell stacks,
electrolyte tanks, and piping. This system consists of four
banks, each of which comprising four modules. Each module includes electrolyte tanks (15 m3 each for a positive and
a negative electrolyte), six cell stacks, and one heat exchanger. Each cell stack in turn includes six sub stacks each
consisting of 18 cells, totaling 108 cells. The rated DC output per cell stack is 45 kW. In each bank, cell stacks are
connected with six in parallel and four in series, and the
rated DC output per bank is 1,000 kW (maximum 1,500 kW).
This system aims at short-period output fluctuation
smoothing (a few seconds to less than ten minutes). The
smoothing method is basically the same as that for the single unit wind power generator described previously. The
combined output is the sum of the wind farm output and
10 · Redox Flow Battery for Energy Storage
Photo 5. 6 MW RF Battery Annexed to Wind Farm
(Battery Cubicle and Cell Stack)
Wind power
generation output


RF battery
battery system
Fig. 4. Basic Configuration of RF Battery System Combined
with Wind Power Generator
Ou tp ut (k W
2:00 2:02 2:04 2:06 2:08 2:10 2:12 2:14 2:16 2:18 2:20 2:22 2:24 2:26 2:28 2:30
Wind power generation Combination
(wind power generation + RF battery)
RF battery
Time • Wind power generation rating: 275 kW
• RF battery rating: 170 kW × 6 h
Fig. 5. Example of Wind Power Output Smoothing Characteristics
(Time Constant: 1 h, May 22, 2001)
-300 0
Ou tp ut (k W
Ba tte
ry c ap ac ity
0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00
(wind power generation + RF battery)
RF battery
Time • Wind power generation rating: 275 kW
• RF battery rating: 170 kW × 6 h
Wind power generation
Battery capacity
Fig. 6. Example of Wind Power Output Smoothing Characteristics (All Day)
(Time Constant: 1 h, May 22, 2001)
the battery output, and the reference of the combined output is set to a value obtained by smoothing wind farm output using the first-order lag filter having a given time
constant. Practically, the output required of the RF battery
is the sum of (1) the difference between the combined output reference and the wind farm output, and (2) the correction amount required for the supplementary charge or
discharge. This is because the battery capacity is limited,
and if the SOC reaches the lowest limit by battery losses or
if the charge amount exceeds the discharge amount due
to wind conditions and the SOC can reach the upper limit,
the SOC should be retained at an appropriate range to
avoid these situations by supplementary charge or discharge with smoothing operation. This control is named
SOC feedback control.
An example of results of the verification test is shown
below. Figure 7 is an example of verification test data using
variable time-constant control. The basic time constant is
set to 30 minutes, but in this test, it is variable depending
on the conditions. When wind power output changes suddenly, the battery may not compensate for it because the
battery capacity is limited. It is considered effective to provisionally reduce the time constant to decrease the load on
the battery and to perform smoothing operation as far as
possible. The test data shows that optimal variable timeconstant control and smoothing operation can be performed without overloading the battery, according to wind
power output conditions.
Figure 8 shows an example of data of a test using bank
control. The optimal number of banks is provided for control according to the required battery output, thus improving system efficiency. The test data indicates that the
number of banks required for operation changes from four
to three and from three to two, depending on the output
required of the battery.
As explained so far, the RF battery system annexed to
the wind farm is proven to perform the desired smoothing
operation. It is also verified that the SOC feedback control
and variable time-constant control for smoothing, both of
which are designed not to exceed the battery capacity, operate effectively. Control technologies including bank control, which realize efficient operation at the optimal
number of banks according to the output required of the
battery, have also been verified to operate effectively. The
system including these controls has thus been verified to
operate stably and efficiently during the test period.
(4) Expectations for applications to secondary control in
the power system(41)–(43)
Large-scale battery systems are expected to be used for
secondary control in the power system. Currently, frequency regulation is performed by adjusting those outputs
while keeping the output of thermal power generation,
pumped hydro energy storage, hydraulic power generation
in good balance, based on load frequency control (LFC)
signals from the power control center. When the LFC operation in such a current power generation is replaced with
one using an RF battery system, the battery capacity required for the same LFC capacity is expected to be small,
because the battery can vary charge/discharge output instantaneously (at high response speed), and because the
battery have the characteristic of supplying high-rate outSEI TECHNICAL REVIEW · NUMBER 73 · OCTOBER 2011 · 11
Photo 6. 6 MW RF Battery Annexed to the Wind Farm
(Electrolyte Tanks)
0 1800
0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00
0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00
Ou tp ut [k W
Ba tte
ry o ut pu t [
kW ]
(b an k 1)
Ti m e co ns ta nt [s ec .]
Wind speed [m/s]
Wind power generation
output [kW]
Wind power output
after smoothing [kW] Battery output [kW]
Bank output [kW] Time constant [s] SOC feedback signal [kW]
Positions where time
constant was varied
Wind power
generation stop
Wind power
generation start
Nov. 22, 2006
Fig. 7. Example of Test Results Using Variable Time Constant Control
(Basic Time Constant: 30 minutes)
ut pu t [
kW ]
Ba tte
ry o ut pu t (
ea ch b an k) [k W
Wind power generation
output [kW]
Wind power output
after smoothing [kW]
Battery output
(all banks total) [kW]
0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00
0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 0:00
Ba nk 4
Ba nk 3
Ba nk 2
Ba nk 1
Jun. 12, 2007
Fig. 8. Example of Test Results using Bank Control
put if it is only for a short duration. In such a case, economic benefit can be expected. For example, if an RF battery can produce output three times as large as the rating
for a very short time, the RF battery’s LFC capacity is 300%
compared with the general thermal power generation’s
LFC capacity of 5%. This means that even a battery with a
markedly small rating has the potential to be applied.
4-4 Overseas developments concerning RF batteries
The development of RF batteries initially started at the
National Aeronautics and Space Administration (NASA)
in the United States and the Electrotechnical Lab. in Japan.
Since then, development for its practical use has been
accelerated in Japan, exemplified by the research and development projects led by the Electrotechnical Lab. and
NEDO and the joint development projects by electric
power companies and manufactures. Recently, because
smart-grid technologies are becoming increasingly common worldwide since the introduction of renewable energy
in great amounts, energy storage batteries are expected to
play a more important role, and hence the development
of energy storage batteries is being promoted. RF batteries
are no exception. They are attracting public attention anew
as large-capacity storage batteries, and their development
is being promoted worldwide. The main system of their development is the V system. The following is an update on
their development overseas(44).
In Australia, Prof. M. Kazacos developed the V system
jointly with V-Fuel Pty Ltd., and further developed the technology realizing the V/Br system, which is higher in energy
In the United States, Deeya Energy, applying NASA’s
technologies, developed a few kilowatt-order products
using the Fe/Cr system for wireless base stations. Ashlawn
Energy, LLC was entrusted by the U.S. Department of Energy (DoE) to perform a project, and released the plan of
verifying the 1 MW/8 MWh class V system. Primus Power
Corp. also received a budget from the DoE, and plans to
develop a verification plant of the flow battery of 25
MW/75 MWh using Zn/Cl2 system.
In Canada, VRB Power Inc. commercialized a few kilowatt class V system RF battery system for the use of combination with independent energy supply or renewable
energy. The technology was purchased by Prudent Energy
Corp. based in China, and its business is promoted to be
expanded. Recently, in the United States and in China, the
company plans to supply MW class RF battery facilities
which will be used with solar power generation and wind
power generation.
In Europe, an RF battery called Regenesys, which uses
the Na/Br system (sodium polysulfide/sodium bromide),
was developed on a large scale as a new redox system. However, the development has been discontinued. In Austria,
Cellstrom GmbH developed a 10 kW/100 kWh V system
RF battery, and has been working for the commercialization in combination with independent power supplies and
solar power generation. In the United Kingdom, RE-Fuel
technology Ltd. is developing a V system RF battery and
has a concept that it is applicable to electric automobiles
and their charge stations. In Germany, FraunhoferGesellschaft is researching into non-aqueous electrolyte
that is capable of realizing high energy density. In South
Africa, CSIR is studying the Cr/Br system.
In Asia, Cellennium Company Ltd. is enthusiastically
promoting the V system in Thailand. In South Korea, Samsung Electronics Co., Ltd. is developing the battery using
non-aqueous electrolyte. Extremely enthusiastic development is observed in China: in 2009, the renewable energy
introduction target was drastically raised (solar power 20
GW and wind generation 150 GW in 2020), and with this
as an opportunity, the development of energy storage batteries is speeding up. As a major development, the State
Grid Corporation of China (SGCC), a major power transmission company, plans a verification project, and battery
manufacturers are competing in development. As for RF
batteries, the previously mentioned Prudent Energy Corp,
as well as research institutes, such as the Dalian Institute of
Chemical Physics, the Chinese Academy of Sciences
(DICP)(46), and Chengde Wanlitong Industrial Group, are
promoting the development.
For zinc/bromine batteries, Australian ZBB Energy
Corp. and RedFlow Ltd. and American Premium Power
Corp. are committed to their development. Plurion Ltd.,
a British company, is developing Zn/Ce-system flow batteries to realize high energy density.
5. Conclusion
In Japan, most RF batteries that have been put into
practice use at the sites of consumers comprise several hundreds of kilowatts class facilities. In other countries, on the
other hand, relatively small systems of a few kilowatts to several tens of kilowatts have been commonly used for independent power supplies. The biggest facility among them
is the 6 MW facility installed in the wind farm as NEDO’s
verification project. In the near future, such applications
as smoothing of a larger scale wind power output fluctuations or secondary control in the power system are expected. In consideration of these applications, largercapacity RF batteries will be necessary. Such facilities must
be safe, reliable, durable and cost efficient at a level equivalent to conventional power systems. We expect that further development of the RF battery, including system up
and improvement of cell materials, will be promoted along
with various verification tests by end-users, and that the RF
battery will play an important role in the power system in
the near future.
(1) L. H. Thaller, “Electrically Rechargeable Redox Flow Cells,” Proc. of
the 9th IECEC, P.924 (1974)
(2) T. Ise, T. Tanaka, “Advanced Technologies for Electric Energy Storage Systems,” CMC books (2006)
(3) JESC, “Battery for power storage system,” JEAC 5006-2010
(4) T. Sasaki, T. Shigematsu, “Maintaining Electrical Power Quality using
New Energy System -Evaluation for Applying Redox Flow Battery-,”
IEEJ Kansai Technical Meeting (2001)
(5) N. H. Hagedorn, “NASA Redox Storage System Development Project
Final Report,” DOE/NASA/12726-24, NASA TM-83677 (1984)
12 · Redox Flow Battery for Energy Storage
(6) H. Kaneko, K. Nozaki, T. Ozawa, “Characteristics of Redox Systems
Used in Redox Flow Cells for Electric Energy Storage Criteria, Measuring Methods and Results of Chromium, Titanium, and Iron Systems,” Bul. Electrotechnical Laboratory, vol.41, No.11, P.877 (1977)
(7) K. Nozaki, H. Kaneko, T. Ozawa, “Prospect of Electric Energy Storage
by Secondary Batteries,” Circulars of the Electrotechnical Laboratory,
No.201, P.51 (1979)
(8) K. Nozaki, H. Kaneko, A. Negishi, K. Kanari, T. Ozawa, “Performance
of ETL New 1kW Redox Flow Cell System,” 19th IECEC P.844 (1984)
(9) K. Nozaki, I. Nagashima, H. Kaneko, T. Ozawa, “Relation between
Cell Resistance and Electrolyte Composition of Redox Flow Battery,”
Denki Kagaku, vol.57, No.10 (1989)
(10) A. Negishi, K. Nozaki, H. Kaneko, Y. Sekita, “Electrode Materials for
Redox Flow Cells in PV Power Generating Systems,” Bul. Electrotechnical Laboratory, vol.54, No.11, P.1295 (1990)
(11) O. Hamamoto, M. Takabatake, M. Yoshitake, H. Misaki, “Research
and Development of 10kW class Redox Flow Battery,” 20th IECEC,
No.2, P.98 (1985)
(12) T. Tanaka, T. Sakamoto, N. Mori, K. Mizunami, T. Shigematsu, “Development of a Redox Flow Battery,” SEI Technical Review, vol.137,
P.191 (1990)
(13) T. Shigematsu, T. Ito, T. Sakamoto, “New Energy Storage System with
Redox Flow Battery,” Electrical Review, vol.78, No.9 (1993)
(14) Randall F. Gahn, Norman H. Hagedorn, Jerri S. Ling, “Single Cell
Performance Studies on the Fe/Cr Redox Energy Storage System
Using Mixed Reactant Solutions at Elevated Temperature,” NASA
TM-83385 (1983)
(15) O. Hamamoto, M. Takabatake, M. Yoshitake, “Redox Battery Designing for the Practical Applications,” Proc. of the 28th Battery Symposium in Japan, 3A19, P.183 (1987)
(16) K. Nozaki, Y. Wada, “Feasibility of Chromium-Chlorine Redox Flow
Battery,” Proceedings of the 28th Battery Symposium in Japan, 3A20,
P.185 (1987)
(17) K. Nozaki, H. Kaneko, A. Negishi, K. Kurokawa, K. Sato, M. Nakajima,
“Performance of All Vanadium Redox Flow Battery and Vanadium
Air Battery,” Proc. of the 59th Battery Symposium in Japan, 1J30
(18) B. Sun, M. Skyllas-Kazacos, “A Study of the V(II)/V(III) Redox Couple for Redox Cell Application,” J. of Power Sources,15, P.179 (1985)
(19) B. Sun, M. Rychik, M. Skyllas-Kazacos, “Investigation of the
V(V)/V(IV) System for Use in the Positive Half-Cell of a Redox Battery,” J. of Power Sources,16, P.85 (1985)
(20) M. Rychik, M. Skyllas-Kazacos, “Characteristics of a new all-vanadium
redox flow battery,” J. of Power Sources, 22, P.59 (1988)
(21) M. Skyllas-Kazacos, M. Rychik, R. Robins, “All Vanadium redox battery,” JP2, 724, 817(1997)
(22) M. Nakajima, M. Sawahata, S. Yoshida, K. Sato, H. Kaneko, A.
Negishi, K. Nozaki, “Vanadium redox flow battery with resources saving recycle ability I. Production of electrolytic solution for vanadium
redox flow battery from boiler soot,” J. of Electrochemistry, vol.66,
No.6, P.600 (1998)
(23) “200kW Vanadium Redox Battery,” J. of OHM, No.2, P.73 (1998)
(24) N. Tokuda, T. Kumamoto, T. Shigematsu, H. Deguchi, T. Ito, N.
Yoshikawa, T. Hara, “Development of a Redox Flow Battery,” SEI
Technical Review, vol.151, P.95 (1997)
(25) T. Hara, T. Shigematsu, “Development of a Redox Flow Battery,” J.
of JABMEE, No.10, P.42 (1998)
(26) N. Tokuda, T. Kanno, T. Hara, T. Shigematsu, Y. Tutsui, A. Ikeuchi,
T. Ito, T. Kumamoto, “Development of a Redox Flow Battery,” SEI
Technical Review, vol.156, P.73 (2000)
(27) N. Tokuda, K. Motoi, T. Hara, H. Sato, Y. Tsutsui, K. Emura, T. Shinzato, T. Kanno, “Development of a Redox Flow Battery for the office
building,” SEI Technical Review, vol.158, P.19 (2001)
(28) N. Tokuda, S. Kikuoka, M. Furuya, S. Kondou, T. Kanno, “Development of a Redox Flow Battery,” Technical Meeting of IEEJ-B, (2001)
(29) T. Shigematsu, “Redox Flow Battery for Energy Storage Closing to
Practical Use,” MATERIAL STAGE vol.1, No.10, P.40 (2002)
(30) H. Morii, S. Miura, M. Minamiura, K. Motoi, “Application of Redox
Flow Battery for Load Leveling,” J.of JECA, No.10 (2001)
(31) T. Shinzato, K. Emura, T. Hara “Development of Vanadium Redox
Battery for Voltage-sag Compensation,” Technical Meeting of IEEJB, No.254 (2001)
(32) NEDO-NP-0004, “Investigation for Introducing Battery Energy Storage System to a Wind Power Generation,” (2002)
(33) T. Shigemitsu, T. Kumamoto, H. Deguchi, T. Hara, “Application of
a Vanadium Redox-flow Battery to Maintain Power Quality,” IEEE
T&D Conference, OR-25 (2002)
(34) NEDO Report, “Development of Technologies for Stabilization of
Wind Power in Power Systems,” (2008)
(35) K. Kobayashi, “Efforts of Electric Power Companies to Cope with
Power Storage Problems 3) Wind Power Stabilization Technology
Development Project,” J. of Energy & Resources, vol.25, No.6, P.394
(36) N. Tokuda, “Technology for Smoothing Fluctuations of Wind Farm
Power Output,” IEEJ Journal, vol.125, No.11 (2005)
(37) T. Shibata, “Application of Redox Flow Battery for Stabilization of
Wind Power,” J.IEIE Jpn, vol.25, No.10, P.800 (2005)
(38) K. Yoshimoto, T. Nanahara, G. Koshimizu, “Analysis of Data Obtained in Demonstration Test about Battery Energy Storage System
to Mitigate Output Fluctuation of Wind Farm,” IEEE PES CALGARY
(39) G. Koshimizu, T. Numata, K. Yoshimoto, H. Hasuike, T. Shibata,
“Subaru Project Analysis of Field test results for Stabilization of 30.6
MW Wind Farm with Energy Storage,” EESAT (2007)
(40) NEDO Report, “The Technical Development for Stabilization of
Wind Power in Power Systems - Analytical study for the Aged Deterioration –,” (2009)
(41) T. Kaizuka, T. Sasaki, “Evaluation of Control Maintaining Electrical
Power Quality by use of Rechargeable Battery System,” IEEE PES
Winter Meeting Columbus (2001)
(42) K. Enomoto, T. Sasaki, T. Shigematsu, H. Deguchi, “Evaluation Study
about Redox Flow Battery Response and its Modeling,” T. IEEJ,
vol.122-B, No.4 (2002)
(43) T. Sasaki, K. Enomoto, T. Shigematsu, H. Deguchi, “Evaluation Study
about Redox Flow Battery Response and its Modeling,” CEPSI
Fukuoka (2002)
(45) M. Skyllas-Kazacos, “Novel vanadium chloride / polyhalide redox
flow battery,” J. of Power Sources, 124, P.299 (2003)
(46) Ping Zhao, Huamin Zhang, Hantao Zhou, Jian Chen, Sujun Gao,
Baolian Yi, “Characteristics and performance of 10kW class all-vanadium redox-flow battery stack,” J. of power sources, 162, P.1416
• Senior Specialist
Manager, Power System R&D Laboratories
He has been engaged in the development
of redox flow batteries.

PDF Document reader online

This website is focused on providing document in readable format, online without need to install any type of software on your computer. If you are using thin client, or are not allowed to install document reader of particular type, this application may come in hand for you. Simply upload your document, and will transform it into readable format in a few seconds. Why choose

  1. Unlimited sharing - you can upload document of any size. If we are able to convert it into readable format, you have it here - saved for later or immediate reading
  2. Cross-platform - no compromised when reading your document. We support most of modern browers without the need of installing any of external plugins. If your device can oper a browser - then you can read any document on it
  3. Simple uploading - no need to register. Just enter your email, title of document and select the file, we do the rest. Once the document is ready for you, you will receive automatic email from us.

Previous 10

Next 10