Renewable and Sustainable Energy Reviews

Contents lists available at ScienceDirect
Renewable and Sustainable Energy Reviews
journal homepage: www.elsevier.com/locate/rser
A review of vanadium electrolytes for vanadium redox flow batteries
Chanyong Choia
, Soohyun Kima
, Riyul Kima
, Yunsuk Choia
, Soowhan Kimb
, Ho-young Jungc
,
Jung Hoon Yangd
, Hee-Tak Kima,⁎
a Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology, 291 Daehak-ro, Yuseong-gu, Daejeon 305-
701, Republic of Korea
b OCI R & D center, OCI Company Ltd, 61 Sagimakgol-ro, 62 beon-gil, Jungwon-gu, Seongnam-si, Gyeonggi-do 462-807, Republic of Korea
c Department of Environment & Energy Engineering, Chonnam National University, 77 Yongbong-ro, Buk-gu, Gwangju 500-757, Republic of Korea
d Energy Storage Department, Korea Institute of Energy Research, Daejeon 305-343, Republic of Korea
ARTICLE INFO
Keywords:
Vanadium redox flow battery
Vanadium electrolyte
Solubility
Stability
Electrochemical analysis
ABSTRACT
There is increasing interest in vanadium redox flow batteries (VRFBs) for large scale-energy storage systems.
Vanadium electrolytes which function as both the electrolyte and active material are highly important in terms
of cost and performance. Although vanadium electrolyte technologies have notably evolved during the last few
decades, they should be improved further towards higher vanadium solubility, stability and electrochemical
performance for the design of energy-dense, reliable and cost-effective VRFBs. This timely review summarizes
the vanadium electrolyte technologies including their synthesis, electrochemical performances, thermal
stabilities, and spectroscopic characterizations and highlights the current issues in VRFB electrolyte development. The challenges that must be confronted to further develop vanadium electrolytes may stimulate more
researchers to push them forward.
1. Introduction
Renewable energy is regarded as one of the important means of
providing energy with sustainability. With increasing energy consumption and limited fossil fuels, it is considered by many an inevitable
choice. However, for efficient implementation of renewable energies
including solar and wind energy in electric grid applications, an energy
storage system (ESS), which provides storage of the electric energy
from renewable sources and its on-demand release, is required. An ESS
should be energy-efficient, safe, reliable, and cost-effective. In this
regard, redox flow batteries (RFBs) have gained increasing attention
for ESS applications. RFBs are largely characterized by their spatial
separation of energy storage and energy conversion function, which
cannot be attained in other secondary batteries based on solid state
active materials. Furthermore, this concept can be realized by using
mobile active materials dissolved in electrolytes. The electrolytes,
which deliver system energy, are stored in tanks, and are supplied to
a stack for charging and discharging. Therefore, power and energy
capabilities are independently tailored towards a more economical
system.
Among the RFBs suggested to date, the vanadium redox flow
battery (VRFB), which was first demonstrated by the Skyllas-Kazacos
group [1], is the most advanced, the only commercially available, and
the most widely spread RFB. In contrast with other RFBs such as Zn-Br
and Fe-Cr batteries, VRFBs exploit vanadium elements with different
vanadium oxidation states as positive and negative active materials,
and thus are free from cross-contamination problems [2,3]. They have
evolved for three decades in the industry sector, but in academic circles
are considered a new area that requires in-depth understanding and
further developments. VRFBs can be characterized by their multi-scale
and dynamic nature based on interdisciplinary technologies. In the
material science and technology sector, the developments of highly
efficient and durable electrodes, membranes, electrolytes, and bipolar
plates are of interest. The charge, mass, and heat transport, and cell/
stack design are the major issues in the battery design sector.
Furthermore, system integration to combine stacks and balance of
plants (BOPs) including tanks, pumps, sensors, electrical components,
and control units is highly important in employing VRFBs for the
electric grid. All the technologies collectively influence the performance
of VRFBs in complex manners. The interdisciplinary nature of VRFBs
has been well documented in recent reviews [4–6].
In this review, we describe the vanadium electrolyte technologies
from the view point of VRFB design, and summarize recent issues and
approaches regarding the electrolyte design for an advanced VRFB. The
vanadium electrolytes, which usually include vanadium ions, counter
anions, acids, and water, are the key component for VRFBs because
http://dx.doi.org/10.1016/j.rser.2016.11.188
Received 8 March 2016; Received in revised form 20 September 2016; Accepted 12 November 2016
⁎ Corresponding author.
E-mail address: [email protected] (H.-T. Kim).
Renewable and Sustainable Energy Reviews 69 (2017) 263–274
Available online 18 November 2016
1364-0321/ © 2016 Elsevier Ltd. All rights reserved.
MARK
they are the highest cost factor [7] and dominantly influence VRFB
performance. In particular, the electrochemical activity and the concentration and stability of vanadium ions determine the energy density
and the reliability of VRFBs [8]. Owing to the pacesetting work from
the pioneers of VRFBs [9], the vanadium electrolyte technology has
been notably improved, and is under evolution towards a denser, more
reliable, and more cost-effective system. Some reviews on vanadium
redox flow batteries listed the recent issues on the vanadium electrolytes [10,11], however, these did not provide an intensive and
diversified description regarding the vanadium electrolyte compositions, properties, influences on battery performances, and electrochemical/spectroscopic analysis. Very recently, an intensive review on the
physical properties and characteristics of the vanadium battery electrolyte under different conditions was published [6]. Compared to the
recently published reviews, the current contribution paid more attention to the interplay of the electrolyte properties and battery performances, and to the electrochemical and spectroscopic characterizations, which have not been collectively described yet but are highly
important in the electrolyte design. Also we avoided simply listing
various case studies of the electrolytes but tried to describe a linkage
among the previous publications. The objective of this review is to
provide a logical understanding on how the electrolyte design influences the battery performances, to deliver an organized information on
the analysis platforms for an advanced electrolyte design, and to inform
unresolved issues and unexplored area in this technology sector. In this
regard, we will first briefly summarize the functions of the VRFB
electrolyte, and after that, describe specific issues related to the
synthesis, solubility and stability, electrochemical performance, and
spectroscopic analysis.
2. Basic principles of VRFB electrolytes
2.1. Electrochemical reactions and composition
The VRFB consists of a stack and two electrolyte tanks, as shown in
Fig. 1. Positive and negative vanadium electrolytes are separately
stored in the tanks, and individually circulated through the stack and
the corresponding tanks. In the stack, the two electrolytes are
separated by a proton exchange membrane. The carbon-felt based
electrodes placed on the both sides of the membrane provide reaction
sites for the electrochemical reactions of the electrolytes. However, the
electrodes themselves do not participate in the redox reactions.
Therefore, the durability of the electrodes is quite high when operation
conditions are properly controlled [12].
At the positive electrode, the electrochemical reaction between
VO2+(vanadium oxidation state of +4) and VO2
+ (+5) is as given in
Eq. (1). At the negative electrode, the electrochemical reaction between
V3+(+3) and V2+(+2) occurs as expressed in Eq. (2). In this review,
VO2
+
, VO2+, V3+, and V2+ are denoted as V(V), V(IV), V(III), and V(II),
respectively, to show their oxidation states. The net reaction having a
standard cell voltage of 1.25 V can be described in Eq. (3).
Positive : VO + 2H + e ↔ VO + H O E = 1. 00 2+ + − 2 o V + 2 (1)
Negative : V + e ↔ V E = −0.25V o 3+ − 2+ (2)
Net : VO + H O + V ↔ VO + 2H + V E = 1. 25V o 2+ 2 3+ 2
+ + 2+ (3)
If V(III) is included in the positive electrolyte, V(III) is oxidized to
V(IV) prior to the oxidation of V(IV) to V(V) during charging. Similarly,
V(IV) in the negative electrolyte is reduced to V(III) during charging
prior to the reduction of V(III) to V(II). These reactions are described
in Eq. (4)
VO + 2H + e ↔ V + H O E = 0. 34 2+ + − 3+ 2 o V (4)
In its most common configuration, the positive VRFB electrolyte
consists of V(IV) and V(V) ions in a H2SO4 aqueous solution, and the
negative VRFB electrolyte consists of V(II) and V(III) ions in a H2SO4
aqueous solution [13]. During cycling, protons are transferred between
the negative and positive electrode surfaces across the positive and
negative electrolyte phases and proton-conducting membrane for
charge balance.
2.2. Solvation structure of vanadium ions
The solvation structures and dynamics of V2+, V3+, VO2+ and VO2
+
cations have been investigated by various techniques to understand the
chemistry of the vanadium electrolyte, and they are summarized in
Table 1. Both V2+ and V3+ cations are hydrated with 6 water molecules,
which can be written as [V(H2O)6]
2+ [14–16] and [V(H2O)6]
3+ [14–
17], forming an octahedral structure. Because of the higher charge
density of V3+ than V2+, the trivalent vanadium complex has shorter
vanadium-ligand bond length and a distorted hydration structure
compared to the divalent vanadium complex [15,16]. The VO2+ cation
has 5 water molecules in the first hydration shell, and has an octahedral
structure composed of 5 water molecules and a vanadyl oxygen [16,18–
20]. As a result of geometry optimization, the octahedral structure of
[VO(H2O)5]
2+ is distorted by repulsion of oxy-oxygen [16,18], and the
axial V-O bond length is longer than the average equatorial V-O bond
length. For the VO2
+ cation, two diffferent hydration structures were
suggested, octahedral [VO2(H2O)4]
+ [16,21,22] and bipyramidal
[VO2(H2O)3]
+ [16,21,23]. However, a bipyramidal structure is favored
over an octahedral structure as a result of geometry optimization and
thermodynamic calculations [16,21,23], and therefore the bipyramidal
structure is the most stable and dominant for the hydrated VO2
+
complex.
2.3. Influence of the electrolytes on VRFB performances
From the system design perspective of the VRFB electrolytes, the
energy density of the VRFB is significantly influenced by their electroFig. 1. Schematic of vanadium redox flow batteries. (a) Charging, (b) discharging. chemical activities and vanadium ion concentrations. The size of the
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
264
electrolyte tanks and that of the stack are reduced with an increase of
the vanadium concentration and the electrochemical activity of the
vanadium electrolytes, respectively [24]. The vanadium solubility of the
positive electrolyte, which is generally lower than that of the negative
electrolyte, mainly limits the energy density [4,5]. The V(V) ion is
known to be stable even at high concentration of up to 3 M in 6 M
sulfate/bisulfate below 30 °C [25,26]. However, above 50 °C, the V(V)
ions readily precipitate in the form of V2O5 at 1.8 M [26]. The early
commercial version of the VRFB called Generation 1 exhibited an
energy density of 20–35 W h kg−1 due to the low vanadium solubility.
In order to address the limit, replacing the positive redox couple with a
higher energy density couple has been suggested. Candidates include
Br-
/Br3- based vanadium-bromine RFB, Mn2+/Mn3+ magnesium-vanadium RFB, and Ce3+/Ce4+ based vanadium-cerium RFB. In spite of
their enhanced energy densities, they suffer from cross-contamination.
Another approach to increase energy density of the VRFB is to enhance
the stability of highly concentrated vanadium ions by modifying the
electrolyte, which will be described in greater detail in Section 4.
The other importance factor affecting energy density is cell voltage.
The combination of two vanadium ion redox couples provides an
equilibrium cell voltage of 1.25 V. It is higher than that (1.18 V) of FeCr RFB, an easiler version of RFB technology, which makes VRFB more
energy dense than Fe-Cr RFB [11]. The equlibrium cell voltage can be
raised by adopting a redox cople with higher redox potential for
positive electorde and/or a negative redox couple with lower redox
potential for negative electrode as exampled by Zn-Br battery. It has a
higher open circuit voltage (1.85 V) and higher energy density than
VRFB. However, it suffers from hydrogen evolution reaction at the
negative electrode during charging which consumes electrolyte and
lowers coulombic efficiency. It is attributed to the low redox potential
of Zn/Zn+ redox couple (−0.76 V vs. SHE). Although, hydrogen
evolution happens for VRFB, lowing coulombic efficiency, it is not as
significant as for Zn-Br RFB [11,12].
The most unique advantage of VRFB is an immunity from crosscontamination. Diffusion of the positive and negative active materials
across separator causes a mixing of the two active materials and creates
a capacity imbalance between the positive and negative electrolyte,
resulting in not only an efficiency loss but also an irreversible loss of
capacity. Such crossover of active species cannot be avoidable in the
current technology platforms. In this regard, the use of the same
vanadium element for both positive and negative electroltye is highly
beneficial; the vanadium crossover causes an efficiency loss, however,
capacity loss can be mitigated by appropriate rebalancing techniques.
The issues on vanadium ion crossover will be described in more detial
in Section 5.
3. Synthesis of the VRFB electrolytes
Vanadium electrolytes are prepared from VOSO4 or V2O5 (Fig. 2).
In the early stage of VRFB research, VOSO4 was adopted as a starting
material due to its more than 10 times higher solubility in an aqueous
H2SO4 solution compared to that of V2O5 [13,27]. Based on the
VOSO4-based route, the preparation of VO2
+ electrolytes with various
vanadium and H2SO4 concentrations has been reported in the literature [26,28]. Another method starting from V2O5 employs electrolytic
dissolution [9,25,29–31] or chemical reduction [25]. In the electrolytic
dissolution method, V2O5 powder is suspended in a H2SO4 solution at a
negative half-cell, and a H2SO4 solution of the same concentration is
placed at a positive half-cell. By oxidation of H2O, oxygen evolution
occurs on the positive pole of the electrolysis cell. On the negative pole,
V2O5 is reduced to V3.5+(same amount of VO2+ and V3+) according to
Eq. (5) [29]. On the other hand, in the chemical reduction method,
V2O5 in H2SO4 is reduced to form VO2+ by adding reducing agents such
as oxalic acid, as described in Eq. (6) [25].
V O +8H +3e →V +VO +4H O 2 5 + − 3+ 2+ 2 (5)
V O +(COOH) +4H →2VO +2CO +3H O 25 2 + 2+ 2 2 (6)
In order to operate the VRFB with a VO2+ electrolyte or V3.5+
electrolyte, a pre-charging step is required. For the use of a VO2+
electrolyte, the electrolyte is distributed to the positive and negative
tanks at a ratio of 2/1 (positive/negative) and the cell is pre-charged.
During the pre-charging process, VO2+ in the positive compartment is
oxidized to VO2
+ and that in the negative compartment is reduced to
V2+. After reaching a 100% SOC, half of the fully charged positive
electrolyte (VO2
+
) is removed for cell balancing [32]. Similarly, starting
Table 1
Hydration structures and structural bond lengths for vanadium cations. (a: axial, e:
equatorial).
Ion Hydration structure V-O (H2O) (Ã…) V-O oxo (Ã…) Ref.
V2+ 2.201 – [14]
2.146 – [15]
2.21 – [16]
V3+ 2.063 – [14]
2.039 – [15]
2.03 – [17]
2.09 – [16]
V4+ 2.17 (a), 2.03 (e) 1.57 [19]
2.269 (a), 2.116 (e) 1.568 [18]
2.323 (a), 2.212 (e) 1.572 [20]
2.36 (a), 2.21(e) 1.57 [16]
V5+ 2.283 (a), 2.126 (e) 1.614 [21]
2.224 (a), 2.004 (e) 1.628 [22]
2.32 (a), 2.09 (e) 1.61 [16]
2.104 1.612 [21]
2.114 1.638 [23]
2.11 1.60 [16]
Fig. 2. Synthesis of VRFB electrolyte.
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
265
from a V3.5+ electrolyte, equal amounts of the electrolyte are distributed to the positive and negative tanks, and the cell is pre-charged until
the V3.5+ electrolyte is oxidized to V(IV) in the positive half-cell and
reduced to V(III)the negative half-cell, which corresponds to a 0% SOC
[29].
From a cost persepective, the synthesis of a vanadium electrolyte
using low purity V2O5 is important. Currently, high purity (99.8%)
V2O5 is used as a starting material, while low purity (98%) V2O5,
typically recycled from slag or catalysts, has also been tested in the
industrial sector due to its lower price. However, the effect of
impurities included in low purity V2O5 on the properties of the
vanadium electrolyte are not clearly understood. Considering the
strong demand for lowering the vanadium electrolyte cost, this issue
requires more intensive research.
4. Solubilites and stabilities of the VRFB
The vanadium concentration of the VRFB electrolyte, which is a
decisive factor for the energy density of the VRFB, is limited by the
vanadium ion solubility and temperature stability. Also, the temperature stability significantly influences the VRFB cost, because the
requirement of heat management to control the operation temperature
in a narrow range increases the cost. In these regards, research on
vanadium electrolytes has focused on increasing the vanadium concentration and temperature stability.
Vanadium ion solubility and temperature stability vary depending
on the oxidation state of vanadium ions. At 2 M vanadium concentration, V(II), V(III), and V(IV) ions are precipitated in a 5 M sulfuric acid
solution below 10 °C. At the same concentration V(V) ions are rather
stable at the low temperature, but are unstable above 40 °C. These
solubility characteristics limit the operation temperature in a range of
10–40 °C. For more reliable operation, the vanadium concentration is
therefore usually lowered to 1.5 M in practical VRFBs [9].
Recenly, Xiao et al. [33] reported that, at 1.5 M vandium and
3.875 M total sulfate concentration, V(II), V(III), V3.5+, V(IV) and
V(V) could be stable in a temperature range of −25 °C to 30 °C. Studies
on the temperature stabilities of the vanadium electrolytes with
different oxidation states are summarized in Table 2.
Sulfuric acid concentration strongly influences vanadium solubility.
As the sulfuric acid concentration increases, the solubility decreases for
V(II), V(III), and V(IV) ions [34,35] due to the common ion effect; with
increasing H2SO4 concentration, the total sulfate concentration increases, which shifts the equilibrium to lower dissociation of vanadium
sulfate. Rahman [34] measured the V(IV) ion solubility with varying
H2SO4 concentration and temperature, and suggested a solubility
prediction model using the Debye-Huckel functional model. He
reported that the rate of the decrease of V(IV) solubility with H2SO4
concentration is higher for the low acid concentration regime than that
for high acid concentration regime due to a higher dissociation of
bisulfate ions (HSO4
−
) to sulfate ions (SO4
2−
). Also, at the higher
temperature, the solubility increases with H2SO4 concentration because the second dissociation constant of sulfuric acid decreases [36].
However, for V(V) ions, its stability increase with increasing H2SO4
concentration. This is due to the presence of more H+ ions, which
favors the backward reaction in the following reaction [9].
2VO +H O↔V O +2H 2
+ 2 25 + (7)
The main limitation for the vanadium electrolyte concentration is
the thermal precipitation of the V(V) ions at an elevated temperature.
The high temperature instability of V(V) originates from the easy
decomposition of the hydrated structure of VO2(H2O)3
+ to V2O5, as
described in Eqs. (8) and (9) [23,33].
(2VO (H O) ) → VO(OH) +H O 223 +
∆ 3 3 +
(8)
2VO(OH) → V O ∙3H ↓ 3 25 2O (9)
A DFT study shows the above process is endothermic and is
accelerated with increasing temperature [23]. More detailed effects of
vanadium and H2SO4 concentrations were investigated by Rahman
et al. [26] They reported that a 3.5 M V(V) at 6 M of total sulfate/
bisulfate was stable at 30 °C. But at 40 °C and above, precipitaion of
V(V) occurred so that total concentration of V(V) decreased. In order to
resolve this problem, high H2SO4 concentration is needed. However, at
these high H2SO4 concentrations, power densities are significantly
reduced due to increased viscosity. Such viscosity rise is one of the
important issues in the electrolyte design towards a higher vanadium
concentration. Although a high H2SO4 concentration improves the
solubility and thermal stability of V(V) ions, the diffusivities of
vanadium ions are reduced due to increased viscosity, consequently
causing a large electrode polarization. Furthermore, the parasitic
power loss from electrolyte pumping becomes larger with a more
viscous electrolyte, resulting in a reduction of the total energy efficiency
of the VRFB system. In this regard, a highly concentrated vanadium
electrolyte with low viscosity is strongly demanded by industry.
In designing the electrolytes, the solubility changes with the SOC
should be carefully considered. A positive electrolyte with a lower SOC
Table 2
Temperature stability of vanadium electrolytes.
Vanadium species Concentration (mol/l) Temperature (°C) Precipitation time Ref.
Vanadium (V) Sulfate (SO4
2−
) Additive
V(II) 1.5 5.0 – − 5 > 90 days [41]
2.0 5.0 419 h
V(III) 1.5 5.0 – − 5 > 90 days
2.0 5.0 634 h
V(IV) 1.5 5.0 – −5 > 90 days
2.0 5.0 18 h
3.0 3.0 6.0 M HCl −5 > 10 days [37]
V(V) 1.5 5.1 – 40 > 90 days [41]
2.0 5.0 < 95 h
2.0 5.0 3 wt% CH3SO3H 40 149 h
2.0 5.0 3 wt% Na2SO4 40 70 h
2.0 5.0 3 wt% Al2(SO4)3 40 120 h
2.0 6.0 0.1–1.0 wt% Polyacrylic acid 40 147–166 h
2.0 3.0 3% Trishydroxymethyl aminomethane 40 > 48 h [42,43]
3.0 3.0 6.0 M HCl 40 > 10 days [37]
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
266
of 85% exhibited a higher degree of stability than that with a higher
SOC of 95% due to its lower V(V) concentration [26]. Above a
vanadium concentration of 2 M, the SOC should be controlled to be
lower than 60% to prevent precipitation because of the low stability of
V(V) ions; however, this inevitably causes a significant loss in the
energy storage capacity. One of the practical methodologies to mitigate
the precipitation problem while minimizing the energy loss is modulation of the SOC according to the climate change.
With an aim to improve the stabilities of the vanadium electrolytes,
a few strategies have been suggested. Li et al. [37] reported that the
addition of hydrochloric acid to V(V) sulfuric acid solution can improve
the high temperature stability of the V(V) solution. For a sulfuric acid
supporting electrolyte, V(V) ions form a hydrated structure of
VO2(H2O)3
+ that is easily decomposed to V2O5 at elevated temperature,
as indicated by the aforementioned process. Meanwhile, for a mixed
supporting electrolyte of sulfuric and hydrochloric acid, V(V) ions form.
Neutral VO2Cl(H2O)2 is more stable than VO2(H2O)3
+ owing to
prohibition of the formation reaction from VO2(H2O)3
+ to VO(OH)3
(Fig. 3). Moreover, Kim et al. [38]. suggested an aqueous solution of
vanadium chloride and of hydrochloric acid. For the chloride based
electrolyte, thermally stable ions such as the vanadium dinuclear form
of [V2O3·4H2O]4+ and dinuclear-chloro complex of [V2O3ClH2O]3+ are
formed. In addition, the reaction kinetics was also improved owing to a
lowered electrolyte viscosity. Other acids including phosphoric acid and
EDTA were also tested; however, these were not as effective as HCl
[37,38]. Recenctly, Roe et al. [39] suggested that a 3 M vanadium
electrolyte could be stabilized by adding 1 wt% H3PO4 +2 wt% ammonium sulfate at 30 °C. Kausar et al. [40] demonstrated that the V(V)
mixed with 1 wt% H3PO4 +2 wt% ammonium sulfate is more stable at
50 °C than those with sodium hexametaphosphate, ammonium phosphate and ammonium sulfate additives. It was suggested that NH4
+
,
PO4
3−
, HPO4
2− and H2PO4
− could enhance high temperature stability
of V(V) ion by (1) either complexing or ion-pairing with the VO2
+ ions
or (2) being adsorbed on nucleation sites of V2O5 and consequently
preventing the precipitation of V2O5.
Besides strong acids, organic and inorganic additives were suggested to improve temperature stability. Zhang et al. [41] conducted an
intensive survey on the effects of various organic and inorganic
additives on the stability of V(II), V(III), V(IV), and V(V) ions ions in
sulfuric acid solutions. Among various additives, polyacrylic acid and
its mixture with CH3SO3H are the most promising candidates for VRFB
electrolyte stabilizing agents. Peng’s group [42,43] demonstrated that
3% trishydroxymethyl aminomethane (Tris) can make 2 M V(V) in a
3MH2SO4 solution stable even at 40 °C, lowering solution resistance
and consequently improving voltage efficiency. However, stabilizing
mechanisms for these additives were not clearly elucidated.
The low temperature stability of the V(IV) electrolyte was improved
by adding stabilizing agents such as sodium hexametaphosphate
(SHMP), K2SO4, or urea [24] and by adding 3 wt% SHMP, 2–5 wt%
K2SO4, or 5 wt% urea, the solubility of VOSO4 in 3 M H2SO4 was
increased up to 4 M. It was explained that these additives retard the
precipitation of V(IV) ions by being adsorbed on the surface of the
nuclei and reducing the rate of crystal growth. In addition, A. Mousa
et al., [44] conducted the stability test of V(II), V(III) ions with several
additives. It indicated that ammonium phosphate, ammonium sulfate
and sodium pentapolyphosphate are suitable to stabilize 2 M V(II),
V(III) ions ions with high sulfuric acid concentration.
5. Electrochemical performance of VRFB electroytes
This section describes the electrochemical performances of the
VRFB electrolyte. The effect of the VRFB electrolytes on energy density
of VRFB is straightforward as described in Sections 2 and 4. However,
their influences on VRFB efficiency and cycling stability are interrelated
with electrode and membrane material properties. Therefore, the
interplays among the electrolytes, electrodes, and membrane are highly
important. Avoiding from an overlap with the previous reviews on the
electrode and membrane materials for VRFB [45,46], the current
section focus on the electrolyte aspect of the interplays.
Energy efficiency (εenergy) of VRFB, which corresponds to the ratio
between the electric energy released from the cell during discharge and
the electric energy supplied to the cell during charge, the result of
combined processes, and can be expressed as a product of voltage
efficiency (εvoltage) and coulombic efficiency (εcoulombic) as given in Eq.
(10).
εenergy voltage coulombi = × ε ε c (10)
εvoltage and εcoulombic are defined as the ratio of the average discharging
voltage and charging voltage and that of discharge capacity and charge
capacity, respectively. The VRFB electrolytes affect εenergy of VRFB in
the following aspects; (1) the ionic conductivity of VRFB electrolytes
influences ohmic polarization of VRFB and consequently voltage
efficiency, (2) shunt current via vanadium electrolyte phases in a
bipolar stack lowers energy efficiency, (3) the redox kinetics of the
VRFB electrolytes at electrode surfaces and the mass transport
processes of the vanadium electrolytes in the porous electrodes
determine kinetic and transport polarizations, respectively, (4) side
reactions such as hydrogen evolution reaction lower coulombic efficiency, and (5) vanadium crossover and water crossover lead to a
capacity imbalance, resulting in a lowered coulombic efficiency and
capacity decay with a repeated cycling.
5.1. Ionic conduction
From the perspective of the battery performance of the VRFB
electrolytes, the ionic conductivities of the vanadium electrolytes,
which are usually provided by H2SO4, should be high for attaining
low ohmic polarization and high rate capability. The conductivity is
dependent on H2SO4 and vanadium ion concentrations and state of
charge (SOC). The ionic conductivity of the V(V) electrolyte was found
to decrease with vanadium ion concentration in a range of 2–5 M at
constant sulfate/bisulfate concentrations of 5, 6, and 7 M because of a
lowered proton concentration [26]. At a vanadium ion concentration of
2 M, the conductivites of V(II), V(III), V(IV), and V(V) electrolytes were
slightly increased with increasing the total sulfate concentration from
4 M to 5 M due to an increased proton concentration [47]. With an
increasing SOC, the ionic conductivities of the positive and negative
electrolyte increase because of their increased proton concentrations
[47]; at the positive electrolyte, two protons are generated with the
oxidation of a single V(IV) ion, and at the negative electrolyte, a proton
is delivered from the positive electrolyte through the polymer electrolyte membrane for charge balance.
Fig. 3. Geometry-optimized structure of VO2(H2O)3
+ and VO2(H2O)Cl based on DFT.
Reprinted with permission from [37].
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
267
5.2. Shunt current
In a bipolar VRFB stack, the cells are connected in series, and the
ionic current flows between a positive an adjacent negative electrode
which are separated by membrane. However, in VRFB stack, the liquid
electrolyte phases of each cells are connected via the manifolds and
channels as shown in Fig. 4, and the ionic current can flow from one
cell to another through the electrolyte phase in the manifolds and
channels [48–50].
Shunt current, in principle, is caused by the high ionic conductivity
of VRFB electrolytes. For fuel cells, shunt current can occur when a
liquid fuel or liquid coolant is used. However, the ionic conductivities
of liquid fuels and coolants are much lower than those of VRFB
electrolytes [51].
Since a shunt current lowers εenergy of VRFB stack and shorten cycle
life, the minimization of shunt current is one of the challenges of VRFB
stack design. Recently, the effect of shunt current on the operation of
VRFB was studied by using a circuit model [52] and dynamic model
[53]. Shunt current can be reduced by increasing the ionic resistance of
the flow path; the resistance is increased with an increased path length
in the manifold and a reduced cross-section of the port. However, these
approaches accompany an increased flow resistance, resulting in
parasitic power to circulate the VRFB electrolyte.
5.3. Electrode reaction
The electrochemical properties of vanadium electrolytes have been
analyzed via various electrochemical measurement techniques.
According to cyclic voltammetry measurements for 2 M V(V) electrolytes, the potential difference between the oxidation peak of V(IV) and
the reduction peak of V(V) decreased with increasing H2SO4 concentration, indicating an increased reversibility [54]. However, the limiting
current measured by linear sweep voltammetry with a rotating disc
electrode was monotonically decreased with increasing H2SO4 concentration for both 2 M V(IV) and V(V) electrolytes, which correlates well
with the viscosity increase with H2SO4 concentration [54]. The effect of
the vanadium concentration was also analyzed by cyclic voltammetry:
at a total sulfate concentration of 6 M, the redox peaks of the V(IV)/
V(V) couple were more intensified with vanadium concentration up to
3.5 M; however, above 3.5 M, the peaks were reduced, which was
explained by the sharp viscosity increase from 3.5 M vanadium
concentration.
Few quantitative investigations of the reaction kinetics for VRFB
redox couples have been made. For VO2+/VO2
+ redox couple, Zhong
and coworkers [55,56] reported a charge transfer coefficient of 0.71
and a rate constant of 2.47×10−4 Acm−2 with a carbon composite
electrode in 3 M H2SO4. Gattrell and coworkers [57,58] reported a high
asymmetry in the polarization of VO2+/VO2
+ redox reaction with a
graphite electrode as shown in Fig. 5. These results suggested that the
VO2+/VO2
+ has a kinetic limitation and is problematic. These electrochemical characteristics of VO2+/VO2
+ redox reaction was explained as
such that this redox reaction was not a simple one-electron transfer
reaction, but is a multi-step reaction in which oxygen transfer may
precede or follow an electron-transfer step. In this case, polarization
curves differ from the ideal Butler-Volmer form.
V2+/V3+redox reaction is relatively faster than VO2+/VO2
+ reaction
according to Gattrell and coworkers [58]. As shown in Fig. 5, the
exchange current for V2+/V3+ redox reaction appears to be much higher
than that of VO2+/VO2
+ redox reaction. Also, the high symmetry of the
V2+/V3+ redox reaction suggests its simple one electron transfer
reaction.
Recently, Aaron et al. [59] reported a comparative investigation of
the kinetic of the positive and negative electrolytes by employing a
dynamic hydrogen electrode. Interestingly, the exchange current
density of the positive electrolyte (V(IV)/V(V) couple) was 44 times
greater than that of the negative electrolyte (V(III)/V(II)) (Fig. 6). This
indicates that the redox reaction of the positive electrolyte is more
kinetically facile than that of the negative electrolyte and that the VRFB
performance is dominated by the slower negative electrode kinetics.
Fig. 4. (a) Simulated components domains in a 5-cell short stack model. (b) Half-cell structure in the stack where blue arrows indicate shunt current in manifold and internal current
through membrane, respectively.
Reprinted with permission from [50].
Fig. 5. Polarization curves showing the two reactions used in a VRB (graphite electrode,
1M H2SO4, 0.1 mV s−1
, 4000 rpm, 20 C. Left curve 16 mM V(II) and 36 mM V(III).
Right curve 31 mM V(IV) and 19 mM V(V)).
Reprinted with permission from [58].
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
268
This behavior is contrary to the result from Gattrell (Fig. 5). Since the
redox reaction of the positive electrolyte is more complex than that of
the negative electrolyte, as described in Eqs. (1) and (2), the observation is not easily understood.
More recently, Ventosa et al. [60] conducted an operando study of a
VRFB with a silver/silver sulfate reference electrode. According to a
galvanostatic intermittent titration technique experiment that can
separate the positive and negative polarizations, the kinetics of the
positive electrolyte was found to be slower than that of the negative
electrolyte, contrary to the conclusion drawn by Aaron et al.
However, Choi et al. [61] recently compared the polarizations of the
positive and negative electrodes by using a dynamic hydrogen electrode
and found that the negative electrode has a larger polarization than the
positive electrode, which supports the conclusion by Aaron et al. As
described in the recent review on VRFB electrodes, the surface
structure of the electrode significantly influences the reaction kinetics
[45]. In this regard, a comparative study of positive and negative redox
reactions for various electrode materials and a more detailed electrochemical analysis is necessary to elucidate the kinetics of the positive
and negative vanadium electrolyte.
A deeper understanding on the electrochemical reactions of the
VRFB electrolytes has been pursued by modeling. Shah and coworkers
suggested a two dimensional transient model to account for the
performance variations with concentration, flowrate, and electrode
porosity [62]. The influence of side reactions of H2 and O2 evolution on
the VRFB performance was also analyzed by dynamic modeling [63–
65]; The gas bubbles formed in the electrodes result in a partial
occlusion of the electrolyte flow, reducing charge transport and mass
transport polarizations. You et al. built a two-dimensional stationary
model to describe a single VRB flow cell [66]. They found the decrease
in the mass transfer coefficient almost has no effect on the distribution
of V3+ concentration and overpotential. Li and Hikihara developed a
model considering the transient behavior in a VRB and the model was
also examined based on the tests of a micro-RFB [67]. They found that
the chemical reaction rate is restricted by the attached external electric
circuit and the concentration change of vanadium ions depends on the
chemical reactions and electrolyte flow [11,62–66,68–70].
5.4. Additives
Since the pacesetting work by the Skyllas-Kazacos group [13], few
works on electrolyte design have been conducted for enhancing the
energy efficiency of VRFBs. The use of methane sulfonic acid
(CH3SO3H), which has higher acidity than H2SO4, was found to be
effective in enhancing power performance; Peng et. al. reported that a
mixed acid of 1.5 M CH3SO3H and 1.5 M H2SO4 improves the redox
reaction kinetics and the mass transport rate of the V(IV)/V(V) redox
couple [71]. However, the reason for the enhancement was not
provided in this work. Several OH containing compounds have been
employed as additives for improving redox reaction rates. Fructose,
mannitol, glucose, and D-sorbitol were tested as additives by Li and
coworkers [72]. D-sorbitol exhibited an improvement in energy efficiency from 79.8% to 81.8%. This was ascribed to the increase of active
hydroxyl groups on the electrode. Later, the effects of glycerol and npropyl alcohol additives on positive electrode performance were
reported [73]. A positive effect was exhibited for both additives, and
it was more pronounced for glycerin. It was suggested that the
vanadium ion concentration at the electrode is increased as glycerin
connects the hydroxyl group of the electrode and vanadium ions.
Some inorganic ions such as Sb3+, Bi3+, and In3+ also have been
added into the electrolyte to utilize their electo-catalytic activity. Shen
et al. [74] reported that the energy efficiency was increased from 57.5%
to 67.1% at a current density of 120 mA cm−2 by introducing 5 mM
Sb3+ into the negative electrolyte. Li et al. [75] employed electrolytes
containing Bi3+ to improve the performance of a VRFB. When adding
10 mM Bi3+, the energy efficiency was significantly improved by ~11%.
Both Sb3+ and Bi3+ were synchronously reduced and electrodeposited
onto the felt electrode surface during the charge and were oxidized
back to ions during the discharge of the flow cells. The reduced metal
particles acted as an electro-catalyst to improve the reaction kinetics.
The metal particles were formed only on the negative electrode side,
not on the positive side, because the standard potentials of Sn/Sn3+
and Bi/Bi3+ are located between those of V(II)/V(III) and V(IV)/V(V)
as −0.03 V vs SCE and 0.046 V vs Ag/AgCl, respectively. Therefore,
their electro-catalytic effect was confined to the V(II)/V(III) reaction.
The effect of In3+ ions on the kinetics of the V(IV)/V(V) redox reaction
was investigated by He et al. [76]. They suggested that In3+ ions
changed the hydration state of vanadium ions and promoted the charge
transfer rate. The energy efficiency at a current density of 20 mA cm−2
was increased by 1.9% at the optimal concentration of 10 mM using a
single static cell. Huang and co-workers [77,78] also studied the
influence of Mn2+ and Cr3+, which are commonly contained in
vanadium electrolytes as impurities. Mn2+ and Cr3+ were effective in
improving the reversibility of V(IV)/V(V) and increasing the diffusion
coefficient value of the vanadium ions at concentrations of 0.7–2.4 mM
and 1.9–5.8 mM, respectively. However, excessive amounts of both
ions caused an increase in the interfacial resistance between the
electrolyte and the electrode surface.
5.5. Vanadium ion and water crossover
Among various transport phenomena happening in VRFB, vanadium and water crossover through membrane would be the most
important issue because of their huge impact on the performance and
cycling stability of VRFB. The vanadium crossover can be regarded as a
self-discharge process; vanadium ions cross over though the ionexchange membrane and chemically react with the other vanadium
ions, resulting in efficiency loss and capacity decay [79]. Compare to
Zn-Br and Fe-Cr batteries, however, VRFBs exploit vanadium elements
with different vanadium oxidation states as positive and negative active
materials, which means VRFBs are free from the cross-contamination
problems [2,3].
Since the crossover behavior is mostly determined by the membrane characteristics, fabricating a membrane that can suppress the
Fig. 6. Kinetic region polarization data for VRFB charging and discharge behaviors
obtained from symetric cells and full cell with dynamic hydrogen electrode. The
electrolyte was 0.1 M vanadium in 5.0 M H2SO4.
Reprinted with permission from [59]..
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
269
crossover is an important issue in this technology sector. From the
point of vanadium electrolytes, the crossover influences the distribution of vanadium ion and water between a positive and negative
electrolyte, resulting in a variation in amount and composition of the
electrolytes with cycling or idling. The transferred vanadium ion across
the membrane reacted with the native ions in electrolyte each other,
the following self-discharge reaction [80–82].
At the positive electrolyte:
V +2VO +2H →3VO +H 2+ 2 O + + 2+ 2 (11)
V +VO →2VO 3+ 2
+ 2+ (12)
V +VO +2H →2V + H 2+ 2 O + + 3+ 2 (13)
At the negative electrolyte:
VO +V +2H →2V + H 2+ 2+ + 3+ 2O (14)
VO +2V +4H →3V +2H 2 O + 2+ + 3+ 2 (15)
VO +V →2VO 2
+ 3+ 2+ (16)
The vanadium ions crossing over the membrane from one half-cell
react with the vanadium ions in the other half-cell, altering positive and
negative electrolyte composition. Luo et al. [82] reported that an
imbalance in the amount of vanadium ions between positive and
negative electrolyte caused by vanadium and water crossover during
cycling leads to a capacity decay. The vanadium crossover through ionexchage membrane in VRFB can be caused by diffusion and migration
of vanadium ions. Diffusion flux of vanadium ion is driven by an ion
concentration gradient across negative and positive electrolyte [79,83].
Compared to diffusion, migration, which is driven by an electric
potential gradient between positive and negative half-cell, has a
relatively weak contribution to the net flux of vanadium ion species
for Nafion and S-Radel membranes [79]. Therefore, in order to prevent
vanadium ion crossover, the diffusion flux across membrane has to be
reduced.
Currently, cation exchange membrane and anion exchange membrane are employed for VRFB. For Nafion 115, a typical cation
exchange membrane, vanadium ion permeability decreases in this
order: V2+ > VO2+ > VO2
+ > V3+, according to Sun et al. [81]. Due to
the difference in membrane permeability for these vanadium ions, total
mass of the V2+ and V3+ ion transferred from the negative to positive
half-cell is larger than that of VO2+ and VO2
+ ion from the positive to
negative half-cell. Therefore, the net vanadium crossover directs from
negative to positive electrolyte, resulting in a vanadium imbalance
between the two electrolytes and consequent capacity loss with cycling.
Kazacos et al. [84] also demonstrated via mathematical modeling that
the capacity loss with cycling is dependent on the mass transfer
coefficient of the vanadium ions. Meanwhile, anion exchange membranes can effectively suppress the vanadium ion crossover and
consequently enhance coulombic efficiency. Tang et al. [80] reported
that the diffusion coefficients of the vanadium ions for Selemion AMV
anion exchange membrane are order of magnitude lower than those for
cation exchange membranes and it decreases in this order: V2+ > VO2
+
> V3+ > VO2+. The lowered vanadium ion crossover for anion exchange
membrane is attributed to Donnan exclusion effect; the positively
charged vanadium ions are thermodynamically rejected from the
membranes having fixed positive charges [85,86].
In addition to vanadium ion crossover, water crossover is also
highly important for successful operation of VRFB, because preferential water transfer across membrane can be a problem in VRFB because
one half-cell is flooded and more diluted while the other becomes more
concentrated, negatively impacting on the performance of the cell. The
water crossover is caused by the transfer of vanadium ions and proton
with the bound water and the transfer of water driven by osmosis
[87,88]. The amount and net direction of water crossover and the
relative contributions from the two processes are dependent on the
types of membrane, SOC, and operation mode in a complex manner.
Mohammadi et al. [89,90] studied the water transfer behavior during
self-discharge for anion- and cation-exchange membranes in the VRFB.
In their report, a significant amount of water was transferred across
cation exchange membranes from the negative to the positive half-cell
during self-discharge starting from 50% SOC. This behavior was
explained by the hydration shells of V2+ and V3+ ions which carry a
large amount of water and can easily permeate through cation
exchange membranes due to their relatively high charge numbers.
However, for anion-exchange membranes, the net water crossover
during self-discharge is mainly governed by osmotic water transfer due
to their low permeabilities for various hydrated vanadium ions. Sukkar
et al. [29] studied the water crossover behavior for the various cationic
exchange membrane during self-discharge process at different initial
SOC with a static dialysis cell. It showed that, at an initial SOC of 100%
and 50%, net water transfer directed toward positive electrolyte.
During a discharge from 50% to 0%, the direction of water transfer
reversed toward negative electrolyte.
Later, Sun et al. [81] conducted an intensive study on water
crossover during self-discharge and cycling. In self-discharge process,
the water crossover includes the transfer of vanadium ions with the
bound water and the corresponding transfer of protons with the
dragged water to balance their charge, and the transfer of water by
osmosis. In this case, 75% of the net water transfer is caused by
osmosis. They also monitored the water crossover behavior for Nafion
membrane during a repeated charge and discharge cycling and
observed that the volume of positive electrolyte was gradually increased
with cycling. The net water crossover direction was from positive to
negative electrolyte during charge and from negative to positive
electrolyte during discharge. Since the amount of water crossover
during discharge was larger than that during charge, a net water flow
from negative to positive electrolyte during cycling was resulted in. In a
cycling process, the water drag by proton across the membrane
(electro-osmosis) plays a key role in the water transfer, which is the
most important difference between the self-discharge and chargedischarge cycling. However, the directions of water drag by proton
for charge and discharge are opposite, diminishing their effects on the
net water crossover as schematically shown in Fig. 7. Therefore, the net
water crossover towards positive electrolyte is caused by the transfer of
vanadium ions with the bound water and the transfer of water driven
by osmosis.
6. Spectroscopic analysis of VRFB electrolytes
The electrochemical process of the VRFB is quite complex because
of the side reactions and processes that include (i) air oxidation of the
V(II) ions in the negative half-cell, (ii) hydrogen evolution at the
Fig. 7. Illustration of transfer of water and vanadium ions in charge–discharge cycles.
Reprinted with permission from [81].
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
270
negative electrode, (iii) differential transfer of vanadium ions from one
half-cell to the other, and (iv) volumetric transfer of the electrolyte
from one half-cell to the other due to pressure difference. These result
in variations of the vanadium ion concentrations and capacity-imbalance between the positive and negative electrolyte. Therefore, quantitative measurements of the oxidation states of vanadium ions and their
concentration are important in operating VRFBs. Furthermore, the
coordination structure of vanadium ions significantly influences their
solubility and temperature stability, as described in Section 3. Upon
this background, spectroscopic tools such as UV, Raman, and NMR
have been exploited for quantitative and qualitative analyses of VRFB
electrolytes.
6.1. UV-spectroscopy
The vanadium electrolytes exhibit color changes according to the
oxidation state: V(II) (violet), V(III) (green), V(IV) (blue), and V(V)
(yellow). Owing to the distinctive color differences, UV spectroscopy
can provide a fingerprint for the vanadium ions. Typical UV spectra for
the vanadium electrolytes measured by our group are displayed in
Fig. 8. For the V(IV) electrolyte, absorbance at 765 nm [91] or 750 nm
[47] was selected for a quantitative analysis. For V(II), V(III), and V(V)
ions, absorbance at 855 nm, 610 nm, and 390 nm was used, respectively [91]. Below a concentration of 0.1 M, all the vanadium electrolytes follow proportionality between concentration and absorbance.
Monitoring of the SOC by UV was attempted for the negative electrolyte
and a linear relationship between the absorbance at 750 nm and the
SOC was found in a SOC range of 5–100% [47]. A difference in
concentration between carbon felt in the cell and the tank was also
elucidated by using UV spectroscopy [92].
6.2. NMR spectroscopy
Nuclear magnetic resonance (NMR) spectroscopy is a powerful tool
to study the chemical identity and dynamics in ionic solutions. 1
H, 17O,
and 51V NMR spectroscopy have been used to study the coordination
structure and dynamics of the molecules in VRFB electrolytes in
relation to their thermal stability [20,23,93]. Chemical shift and line
width are two major parameters used in the analysis, and they are
directly related to the electronic atmosphere and mobility of atoms,
respectively. These collectively reveal the structure and dynamics of
water, vanadium ions, and counter anions. As the density of the
electrons in the molecular orbitals decreases, the chemical shift
increases due to a reduced shielding effect. The line width of the
NMR peak, which is mainly associated with spin-spin relaxation time
(T2), increase with decreasing molecular mobility; T2 becomes shorter
as molecular reorientation rates slow down.
The only oxidation state of vanadium ions active in 51V NMR is V5+
owing to its diamagnetism. Indeed, the other oxidation states (V(II),
V(III), and V(IV)) are paramagnetic and give rise to an extreme
broadening of the vanadium NMR lines [94]. Typical 51V NMR spectra
of the V(V) electrolyte are shown in Fig. 9. With a highly viscous
electrolyte such as high vanadium (5 M) and sulfate/bisulfate concentration (7 M), the 51V NMR does not show any peaks due to significant
peak broadening [23,26]. However, a 51V NMR peak was exhibited for
the diluted electrolytes (0.5 V(V)-4MS and 0.5 V(V)-7MS), and it was
sharper for the higher dilution.
Vijayakumar and coworkers [20] conducted 1
H and 17O NMR
analyses for a V(IV) electrolyte with varying vanadium concentration
and temperature. In the 1
H NMR spectra, only one peak appeared in
spite of the presence of different proton sites such as bulk water,
coordinated water (VO(H2O)5
2+), and HSO4

, indicating fast proton
exchange among them. The 17O NMR spectra exhibited two peaks. The
peak at higher chemical shift originates from HSO4
– or SO4
2-. The more
pronounced peak at a lower chemical shift is the result of a mixed
contribution from solvent water and water-coordinated VO2+. The
single peak from solvent water indicates rapid water exchange between
them. From the changes of the chemical shift and line width with
temperature and vanadium concentration, the stability of VO(H2O)5
2+
ions up to 3 M vanadium concentration in a temperature range of 240–
340 K was confirmed and weak coordination of sulfate anions to the
hydrated vanadyl ions was deduced. 17O and 51V NMR were also used to understand the instability of
VO2
+ ions at high temperatures [23]. As shown in Fig. 10, the chemical
shift of the 51V NMR peak for a 2 M V(V) electrolyte exhibited an
irreversible change at 335 K, indicating variation of the chemical
environment. Abrupt changes in the chemical shift in 17O and 51V
NMR spectra before precipitation and the DFT analysis collectively
suggest deprotonation of VO2(H2O)3
+ to form neutral VO(OH)3 prior
to the precipitation in the form of V2O5. In the study on thermal
stability of V(III) ions in a mixed acid of H2SO4 and HCl, 35Cl NMR
revealed an increased chlorine exchange between the first and second
co-ordination sphere of V(III) ions with increasing vanadium concentration [17]. Very recently, S. Kim et al., [93] reported the existence of
VO(OH)3 in room temperature with 51V NMR spectroscopy by induFig. 8. UV spectra of various vanadium electrolytes (2 M vanadium, 3 M sulfate diluted
with DI water by 20 times).
Fig. 9. 51V NMRspectra of various V(V) solutions in different supporting electrolyte.
5 M V(V)-7MS: 5 M VO2
+ with 7 M total sulfate; 5 M V(V)-7MS+1 wt% DW: 1 wt% water
is added to the 5 M (V)-7MS; 0.5 V 7S-4S: 5 MV(V)−7MS is diluted to 2 M V(V) with
7MH2SO4 solution; 0.5 V 7S-7S: 5 M V(V)-7MS is diluted to 2 M V(V) with 4 M H2SO4
solution.
Reproduced under permission from [26].
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
271
cing paramagnetic dipolar broadening of VO2
+ signal.
6.3. Raman spectroscopy
Raman spectroscopy is a useful tool to monitor changes in the V-O
bond because of various signals from characteristic peaks including VO terminal stretching (800–1000 cm−1
), V-O bridging mode (400–
800 cm−1
), and V-O bending and lattice modes (below 400 cm−1
). For
V(V) electrolytes, the broad peaks from V-O-S bridging stretching at
660–680 cm−1 and V-O-V stretching in the dimer (770 cm−1
) increased
with vanadium concentration, in good agreement with the high
temperature instability at high vanadium concentrations for the V(V)
electrolyte [95].
7. Conclusion
VRFB technology has gained increasing attention for ESS applications and experienced commercialization. Throughout this process, the
vanadium electrolyte has played a crucial role. Owing to the pacesetting
works from the pioneers, the vanadium electrolyte technology has been
notably improved, and is under evolution towards a denser, more
reliable, and more cost-effective system. In these regards, issues and
challenges for VRFB electrolytes are summarized and explained
throughout the review. The following points can be drawn from the
review;.
1) Various routes to fabricate the vanadium electrolytes from VOSO4
or V2O5 has been developed, however, a more cost-effective
preparation route and a way to utilize low purity-V2O5 would be
worth researching in consideration of the need for low cost
vanadium electrolytes.
2) Vanadium concentration of VRFB electrolyte, which is a decisive
factor for the energy density of VRFB, is currently limited by the
vanadium ion solubility and temperature stability. The acid in the
electrolyte profoundly influences the vanadium solubility and
stability. For H2SO4 and HCl based electrolytes, intensive studies
by a few groups have deepened the molecular scale understanding
of the acid effect. To enhance the performance and stability of the
vanadium electrolyte, various additives have been suggested based
on phenomenological observations. A mechanistic understanding is
necessary for the additive approaches.
3) The electrochemical performance of VRFB electrolyte is one of the
unexplored areas in this technology sector. In particular, the
interplays between the electrolyte and electrode for the redox
reactions are not fully understood, because of the lack of detailed
and comparative works on the vanadium redox reaction kinetics.
4) The vanadium and water crossover significantly influence the
volume and composition of the vanadium electrolytes during
operation or idle mode, and have been matter of interest in this
field. Although the crossover problems are mainly associated with
the characteristics of membrane, the role of vanadium electrolytes
has not been fully understood. For an advanced engineering to
address crossover problem, more detailed analyses on transport
phenomena of the vanadium electrolytes are necessary.
5) In pursue of more advanced electrolytes, spectroscopic techniques
such as UV, Raman, and NMR are quite effective in gaining
molecular-scale understanding as exampled in this review.
The current challenges and approaches that the timely review
summarizes may stimulate more researchers to push them forward.
Acknowledgements
This work was supported by the Korea Institute of Energy
Technology Evaluation and Planning (KETEP) and the Ministry of
Trade, Industry & Energy (MOTIE) of the Republic of Korea (No:
20142010102930 and 20142020103710).
References
[1] Rychcik M, Skyllas-Kazacos M. Characteristics of a new all-vanadium redox flow
battery. J Power Sources 1988;22:59–67.
[2] Sum E, Rychcik M, Skyllas-Kazacos M. Investigation of the V (V)/V (IV) system for
use in the positive half-cell of a redox battery. J Power Sources 1985;16:85–95.
[3] Singh P, Jonshagen B. Zinc ∙ bromine battery for energy storage. J Power Sources
1991;35:405–10.
[4] Wang W, Luo Q, Li B, Wei X, Li L, Yang Z. Recent progress in redox flow battery
research and development. Adv Funct Mater 2013;23:970–86.
[5] Skyllas-Kazacos M, Kazacos G, Poon G, Verseema H. Recent advances with UNSW
vanadium-based redox flow batteries. Int J Energy Res 2010;34:182–9.
[6] Skyllas‐Kazacos M, Cao L, Kazacos M, Kausar N, Mousa A. Vanadium electrolyte
studies for the vanadium redox battery—a review. ChemSusChem 2016;9:1521–43.
[7] Kear G, Shah AA, Walsh FC. Development of the all-vanadium redox flow battery
for energy storage: a review of technological, financial and policy aspects. Int J
Energy Res 2012;36:1105–20.
[8] Skyllas‐Kazacos M, Menictas C, Kazacos M. Thermal stability of concentrated V (V)
electrolytes in the vanadium redox cell. J Electrochem Soc 1996;143:L86–L88.
[9] Kazacos M, Cheng M, Skyllas-Kazacos M. Vanadium redox cell electrolyte optimization studies. J Appl Electrochem 1990;20:463–7.
[10] Parasuraman A, Lim TM, Menictas C, Skyllas-Kazacos M. Review of material
research and development for vanadium redox flow battery applications.
Electrochim Acta 2013;101:27–40.
[11] Weber AZ, Mench MM, Meyers JP, Ross PN, Gostick JT, Liu Q. Redox flow
batteries: a review. J Appl Electrochem 2011;41:1137–64.
[12] Ponce de León C, Frías-Ferrer A, González-García J, Szánto DA, Walsh FC. Redox
flow cells for energy conversion. J Power Sources 2006;160:716–32.
[13] Skyllas-Kazacos M, Grossmith F. Efficient vanadium redox flow cell. J Electrochem
Soc 1987;134:2950–3.
[14] Akesson R, Pettersson LGM, Sandstrom M, Wahlgren U. Ligand-field effects in the
hydrated divalent and trivalent metal-ions of the first and 2nd transition periods. J
Am Chem Soc 1994;116:8691–704.
[15] Benmelouka M, Messaoudi S, Furet E, Gautier R, Le Fur E, Pivan J-Y. Density
functional investigation of hydrated V (II) and V (III) ions: influence of the second
coordination sphere; water exchange mechanism. J Phys Chem A
2003;107:4122–9.
[16] Sepehr F, Paddison SJ. The solvation structure and thermodynamics of aqueous
vanadium cations. Chem Phys Lett 2013;585:53–8.
[17] Vijayakumar M, Li L, Nie Z, Yang Z, Hu J. Structure and stability of hexa-aqua
V(III) cations in vanadium redox flow battery electrolytes. Phys Chem Chem Phys:
PCCP 2012;14:10233–42.
[18] Larsen SC. DFT calculations of proton hyperfine coupling constants for [VO (H2O)
5] 2+: comparison with proton ENDOR data. J Phys Chem A 2001;105:8333–8.
Fig. 10. Variable temperature chemical shift analysis of 51V NMR of 2 M V(V)
electrolyte solution. The open circles and the closed circles in the figure represent the
heating cycle and subsequent cooling cycle, respectively.
Reproduced under permission from [23].
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
272
[19] Grant CV, Cope W, Ball JA, Maresch GG, Gaffney BJ, Fink W, et al. Electronic
structure of the aqueous vanadyl ion probed by 9 and 94 GHz EPR and pulsed
ENDOR spectroscopies and density functional theory calculations. J Phys Chem B
1999;103:10627–31.
[20] Vijayakumar M, Burton SD, Huang C, Li L, Yang Z, Graff GL, et al. Nuclear
magnetic resonance studies on vanadium(IV) electrolyte solutions for vanadium
redox flow battery. J Power Sources 2010;195:7709–17.
[21] Bühl M, Parrinello M. Medium effects on 51v NMR chemical shifts: a density
functional study. Chem- Eur J 2001;7:4487–94.
[22] Krakowiak J, Lundberg D, Persson I. A coordination chemistry study of hydrated
and solvated cationic vanadium ions in oxidation states+ III,+ IV, and+ V in
solution and solid state. Inorg Chem 2012;51:9598–609.
[23] Vijayakumar M, Li L, Graff G, Liu J, Zhang H, Yang Z, et al. Towards understanding
the poor thermal stability of V5+ electrolyte solution in Vanadium Redox Flow
Batteries. J Power Sources 2011;196:3669–72.
[24] Skyllas‐Kazacos M, Peng C, Cheng M. Evaluation of precipitation inhibitors for
supersaturated vanadyl electrolytes for the vanadium redox battery. Electrochem
Solid-State Lett 1999;2:121–2.
[25] Skyllas-Kazacos M, Kazacos M, McDermott R. Vanadium compound dissolution
processes. Patent Application PCT/AU1988/000471; 1988.
[26] Rahman F, Skyllas-Kazacos M. Vanadium redox battery: positive half-cell electrolyte studies. J Power Sources 2009;189:1212–9.
[27] Oei D-G. Permeation of vanadium cations through anionic and cationic membranes. J Appl Electrochem 1985;15:231–5.
[28] Zhao P, Zhang H, Zhou H, Chen J, Gao S, Yi B. Characteristics and performance of
10 kW class all-vanadium redox-flow battery stack. J Power Sources
2006;162:1416–20.
[29] Sukkar T, Skyllas-Kazacos M. Water transfer behaviour across cation exchange
membranes in the vanadium redox battery. J Membr Sci 2003;222:235–47.
[30] Chang F, Hu C, Liu X, Liu L, Zhang J. Coulter dispersant as positive electrolyte
additive for the vanadium redox flow battery. Electrochim Acta 2012;60:334–8.
[31] Ngamsai K, Arpornwichanop A. Analysis and measurement of the electrolyte
imbalance in a vanadium redox flow battery. J Power Sources 2015;282:534–43.
[32] Skyllas-Kazacos M, Rychick M, Robins R. All-vanadium redox battery. Google
Patents; 1988.
[33] Xiao S, Yu L, Wu L, Liu L, Qiu X, Xi J. Broad temperature adaptability of vanadium
redox flow battery—Part 1: electrolyte research. Electrochim Acta
2016;187:525–34.
[34] Rahman F, Skyllas-Kazacos M. Solubility of vanadyl sulfate in concentrated sulfuric
acid solutions. J Power Sources 1998;72:105–10.
[35] Cheng M. Electrolyte optimization and studies for the vanadium redox battery [MS
thesis]. Sydney, Australia: University of New South Wales; 1991.
[36] Marshall WL, Jones EV. Second dissociation constant of sulfuric acid from 25 to
350° evaluated from solubilities of calcium sulfate in sulfuric acid solutions1, 2. J
Phys Chem 1966;70:4028–40.
[37] Li L, Kim S, Wang W, Vijayakumar M, Nie Z, Chen B, et al. A stable vanadium
redox-flow battery with high energy density for large-scale energy storage.
advanced energy. Materials 2011;1:394–400.
[38] Kim S, Vijayakumar M, Wang W, Zhang J, Chen B, Nie Z, et al. Chloride supporting
electrolytes for all-vanadium redox flow batteries. Phys Chem Chem Phys: PCCP
2011;13:18186–93.
[39] Roe S, Menictas C, Skyllas-Kazacos M. A high energy density vanadium redox flow
battery with 3 M vanadium electrolyte. J Electrochem Soc
2016;163:A5023–A5028.
[40] Kausar N, Mousa A, Skyllas‐Kazacos M. The Effect of Additives on the High‐
Temperature Stability of the Vanadium Redox Flow Battery Positive Electrolytes.
ChemElectroChem; 2015.
[41] Zhang J, Li L, Nie Z, Chen B, Vijayakumar M, Kim S, et al. Effects of additives on
the stability of electrolytes for all-vanadium redox flow batteries. J Appl
Electrochem 2011;41:1215–21.
[42] Peng S, Wang NF, Gao C, Lei Y, Liang XX, Liu SQ, et al. Stability of positive
electrolyte containing trishydroxymethyl aminomethane additive for vanadium
redox flow battery. Int J Electrochem Sc 2012;7:4388–96.
[43] Peng S, Wang NF, Gao C, Lei Y, Liang XX, Liu SQ, et al. Influence of
trishydroxymethyl aminomethane as a positive electrolyte additive on performance
of vanadium redox flow battery. Int J Electrochem Sc 2012;7:4314–21.
[44] Mousa A, Skyllas‐Kazacos M. Effect of additives on the low‐temperature stability of
vanadium redox flow battery negative half‐cell electrolyte. ChemElectroChem
2015;2:1742–51.
[45] Kim KJ, Park M-S, Kim Y-J, Kim JH, Dou SX, Skyllas-Kazacos M. A technology
review of electrodes and reaction mechanisms in vanadium redox flow batteries. J
Mater Chem A 2015;3:16913–33.
[46] Schwenzer B, Zhang J, Kim S, Li L, Liu J, Yang Z. Membrane development for
vanadium redox flow batteries. ChemSusChem 2011;4:1388–406.
[47] Skyllas-Kazacos M, Kazacos M. State of charge monitoring methods for vanadium
redox flow battery control. J Power Sources 2011;196:8822–7.
[48] Katz M. Analysis of electrolyte shunt currents in fuel cell power plants. J
Electrochem Soc 1978;125:515–20.
[49] Zhou G, Chen L-D, Seaba JP. Modeling of Shunt Current in Liquid Electrolyte Fuel
Cells.ASME 2006 In: Proceedings of the 4th International Conference on Fuel Cell
Science, Engineering and Technology: American Society of Mechanical Engineers.
p. 863-869; 2006.
[50] Yin C, Guo S, Fang H, Liu J, Li Y, Tang H. Numerical and experimental studies of
stack shunt current for vanadium redox flow battery. Appl Energy
2015;151:237–48.
[51] Zhou H, Zhang H, Zhao P, Yi B. A comparative study of carbon felt and activated
carbon based electrodes for sodium polysulfide/bromine redox flow battery.
Electrochim Acta 2006;51:6304–12.
[52] Xing F, Zhang H, Ma X. Shunt current loss of the vanadium redox flow battery. J
Power Sources 2011;196:10753–7.
[53] Tang A, McCann J, Bao J, Skyllas-Kazacos M. Investigation of the effect of shunt
current on battery efficiency and stack temperature in vanadium redox flow battery.
J Power Sources 2013;242:349–56.
[54] Oriji G, Katayama Y, Miura T. Investigation on V(IV)/V(V) species in a vanadium
redox flow battery. Electrochim Acta 2004;49:3091–5.
[55] Zhong S, Kazacos M, Burford R, Skyllas-Kazacos M. Fabrication and activation
studies of conducting plastic composite electrodes for redox cells. J Power Sources
1991;36:29–43.
[56] Zhong S, Skyllas-Kazacos M. Electrochemical behaviour of vanadium (V)/vanadium (IV) redox couple at graphite electrodes. J Power Sources 1992;39:1–9.
[57] Gattrell M, Park J, MacDougall B, Apte J, McCarthy S, Wu CW. Study of the
mechanism of the vanadium 4+/5+ redox reaction in acidic solutions. J
Electrochem Soc 2004;151:A123.
[58] Gattrell M, Qian J, Stewart C, Graham P, MacDougall B. The electrochemical
reduction of VO 2+ in acidic solution at high overpotentials. Electrochim Acta
2005;51:395–407.
[59] Aaron D, Sun C-N, Bright M, Papandrew AB, Mench MM, Zawodzinski TA. In situ
kinetics studies in all-vanadium redox flow batteries. ECS Electrochem Lett
2013;2:A29–A31.
[60] Ventosa E, Skoumal M, Vázquez FJ, Flox C, Morante JR. Operando studies of allvanadium flow batteries: easy-to-make reference electrode based on silver–silver
sulfate. J Power Sources 2014;271:556–60.
[61] Choi C, Choi Y, Kim S, Jung H-y , Kim H-T. Resistor design for the use of dynamic
hydrogen electrode in vanadium redox flow batteries. Electrochim Acta
2016;213:490–5.
[62] Shah A, Watt-Smith M, Walsh F. A dynamic performance model for redox-flow
batteries involving soluble species. Electrochim Acta 2008;53:8087–100.
[63] Al-Fetlawi H, Shah A, Walsh F. Modelling the effects of oxygen evolution in the allvanadium redox flow battery. Electrochim Acta 2010;55:3192–205.
[64] Shah A, Al-Fetlawi H, Walsh F. Dynamic modelling of hydrogen evolution effects in
the all-vanadium redox flow battery. Electrochim Acta 2010;55:1125–39.
[65] Shah A, Tangirala R, Singh R, Wills R, Walsh F. A dynamic unit cell model for the
all-vanadium flow battery. J Electrochem Soc 2011;158:A671–A677.
[66] You D, Zhang H, Chen J. A simple model for the vanadium redox battery.
Electrochim Acta 2009;54:6827–36.
[67] Minghua L, Hikihara T. A coupled dynamical model of redox flow battery based on
chemical reaction, fluid flow, and electrical circuit. IEICE transactions on fundamentals of electronics, communications and computer sciences. 2008;91:1741-7.
[68] Al-Fetlawi H, Shah A, Walsh F. Non-isothermal modelling of the all-vanadium
redox flow battery. Electrochim Acta 2009;55:78–89.
[69] Knehr K, Kumbur E. Open circuit voltage of vanadium redox flow batteries:
discrepancy between models and experiments. Electrochem Commun
2011;13:342–5.
[70] Vynnycky M. Analysis of a model for the operation of a vanadium redox battery.
Energy 2011;36:2242–56.
[71] Peng S, Wang NF, Wu XJ, Liu SQ, Fang D, Liu YN, et al. Vanadium species in
CH3SO3H and H2SO4 mixed acid as the supporting electrolyte for vanadium redox
flow battery. Int J Electrochem Sc 2012;7:643–9.
[72] Li S, Huang K, Liu S, Fang D, Wu X, Lu D, et al. Effect of organic additives on
positive electrolyte for vanadium redox battery. Electrochim Acta 2011;56:5483–7.
[73] Jia Z, Wang B, Song S, Chen X. Effect of polyhydroxy-alcohol on the electrochemical behavior of the positive electrolyte for vanadium redox flow batteries. J
Electrochem Soc 2012;159:A843.
[74] Shen J, Liu S, He Z, Shi L. Influence of antimony ions in negative electrolyte on the
electrochemical performance of vanadium redox flow batteries. Electrochim Acta
2015;151:297–305.
[75] Li B, Gu M, Nie Z, Shao Y, Luo Q, Wei X, et al. Bismuth nanoparticle decorating
graphite felt as a high-performance electrode for an all-vanadium redox flow
battery. Nano Lett 2013;13:1330–5.
[76] He Z, Chen L, He Y, Chen C, Jiang Y, He Z, et al. Effect of In3+ ions on the
electrochemical performance of the positive electrolyte for vanadium redox flow
batteries. Ionics 2013;19:1915–20.
[77] Gui-Xin HFW, Dong-Mei YK-PL. Influence of Mn~(2+) concentration on the
electrochemical behavior of the anolyte for vanadium redox flow batteries. Chin J
Inorg Chem 2012;5:005.
[78] Huang F, Zhao Q, Luo C, Wang G, Yan K, Luo D. Influence of Cr3+ concentration on
the electrochemical behavior of the anolyte for vanadium redox flow batteries. Chin
Sci Bull 2012;57:4237–43.
[79] Agar E, Knehr K, Chen D, Hickner M, Kumbur E. Species transport mechanisms
governing capacity loss in vanadium flow batteries: comparing Nafion® and
sulfonated Radel membranes. Electrochim Acta 2013;98:66–74.
[80] Tang A, Bao J, Skyllas-Kazacos M. Dynamic modelling of the effects of ion diffusion
and side reactions on the capacity loss for vanadium redox flow battery. J Power
Sources 2011;196:10737–47.
[81] Sun C, Chen J, Zhang H, Han X, Luo Q. Investigations on transfer of water and
vanadium ions across Nafion membrane in an operating vanadium redox flow
battery. J Power Sources 2010;195:890–7.
[82] Luo Q, Li L, Wang W, Nie Z, Wei X, Li B, et al. Capacity decay and remediation of
nafion‐based all‐vanadium redox flow batteries. ChemSusChem 2013;6:268–74.
[83] Agar E, Benjamin A, Dennison C, Chen D, Hickner M, Kumbur E. Reducing
capacity fade in vanadium redox flow batteries by altering charging and discharging
currents. J Power Sources 2014;246:767–74.
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
273
[84] Skyllas-Kazacos M, Goh L. Modeling of vanadium ion diffusion across the ion
exchange membrane in the vanadium redox battery. J Membr Sci 2012;399:43–8.
[85] Sata T. Studies on anion exchange membranes having permselectivity for specific
anions in electrodialysis—effect of hydrophilicity of anion exchange membranes on
permselectivity of anions. J Membr Sci 2000;167:1–31.
[86] Chen D, Hickner MA, Agar E, Kumbur EC. Selective anion exchange membranes for
high coulombic efficiency vanadium redox flow batteries. Electrochem Commun
2013;26:37–40.
[87] Knehr K, Agar E, Dennison C, Kalidindi A, Kumbur E. A transient vanadium flow
battery model incorporating vanadium crossover and water transport through the
membrane. J Electrochem Soc 2012;159:A1446–A1459.
[88] Knehr K, Kumbur E. Role of convection and related effects on species crossover and
capacity loss in vanadium redox flow batteries. Electrochem Commun
2012;23:76–9.
[89] Mohammadi T, Chieng SC, Skyllas-Kazacos M. Water transport study across
commercial ion exchange membranes in the vanadium redox flow battery. J Membr
Sci 1997;133:151–9.
[90] Mohammadi T, Kazacos MS. Evaluation of the chemical stability of some
membranes in vanadium solution. J Appl Electrochem 1997;27:153–60.
[91] Choi NH, Sk Kwon, Kim H. Analysis of the oxidation of the V(II) by dissolved
oxygen using UV-visible spectrophotometry in a vanadium redox flow battery. J
Electrochem Soc 2013;160:A973–A979.
[92] Lee HJ, Choi NH, Kim H. Analysis of concentration polarization using UV-visible
spectrophotometry in a vanadium redox flow battery. J Electrochem Soc
2014;161:A1291–A1296.
[93] Kim S, Choi C, Kim R, Kim HG, Kim H-T. Temperature-dependent 51 V nuclear
magnetic resonance spectroscopy for the positive electrolyte of vanadium redox
flow batteries. RSC Advances. 2016;6:96847-52
[94] Weckhuysen BM, Voort P, Catana G. Spectroscopy of transition metal ions on
surfaces. Leuven University Press; 2000.
[95] Kausar N, Howe R, Skyllas-Kazacos M. Raman spectroscopy studies of concentrated vanadium redox battery positive electrolytes. J Appl Electrochem
2001;31:1327–32.
C. Choi et al. Renewable and Sustainable Energy Reviews 69 (2017) 263–274
274

Don't use plagiarized sources. Get Your Custom Essay on
Renewable and Sustainable Energy Reviews
Just from $13/Page
Order Essay
Still stressed from student homework?
Get quality assistance from academic writers!
error: Content is protected !!
Open chat
1
Need assignment help? You can contact our live agent via WhatsApp using +1 718 717 2861

Feel free to ask questions, clarifications, or discounts available when placing an order.

Order your essay today and save 30% with the discount code LOVE