Evaluation of Tris-Bipyridine Chromium Complexes for Flow Battery Applications: Impact of Bipyridine Ligand Structure on Solubility and Electrochemistry

Article abstract of DOI:10.1021/acs.inorgchem.5b01328

This report describes the design, synthesis, solubility, and electrochemistry of a series of tris-bipyridine chromium complexes that exhibit up to six reversible redox couples as well as solubilities approaching 1 M in acetonitrile. We have systematically modified both the ligand structure and the oxidation state of these complexes to gain insights into the factors that impact solubility and electrochemistry. The results provide a set of structure-solubility-electrochemistry relationships to guide the future development of electrolytes for nonaqueous flow batteries. In addition, we have identified a promising candidate from the series of chromium complexes for further electrochemical and battery assessment.

Full text of DOI:10.1021/acs.inorgchem.5b01328

Article
pubs.acs.org/IC
Evaluation of Tris-Bipyridine Chromium Complexes for Flow Battery
Applications: Impact of Bipyridine Ligand Structure on Solubility and
Electrochemistry
Pablo J. Cabrera,†,§ Xingyi Yang,#,§,‡ James A. Suttil,†,§,‡ Rachel E. M. Brooner,†,§ Levi T. Thompson,*,#,§
and Melanie S. Sanford*,†,§
†Department of Chemistry, University of Michigan, 930 North University Avenue, Ann Arbor, Michigan 48109, United States
#Department of Chemical Engineering, University of Michigan, 2300 Hayward Street, Ann Arbor, Michigan 48109, United States
§Joint Center for Energy Storage Research (JCESR), Argonne, Illinois, United States
S
* Supporting Information
ABSTRACT: This report describes the design, synthesis,
solubility, and electrochemistry of a series of tris-bipyridine
chromium complexes that exhibit up to six reversible redox
couples as well as solubilities approaching 1 M in acetonitrile.
We have systematically modified both the ligand structure and
the oxidation state of these complexes to gain insights into the
factors that impact solubility and electrochemistry. The results
provide a set of structure−solubility−electrochemistry rela-
tionships to guide the future development of electrolytes for nonaqueous flow batteries. In addition, we have identified a
promising candidate from the series of chromium complexes for further electrochemical and battery assessment.
INTRODUCTION1
(5) be accessible from inexpensive, earth abundant starting
■
materials.
Redox flow batteries (RFBs) represent an emerging technology
In this context, metal coordination complexes (MCCs) have
for use in grid scale energy storage applications.2 In these
emerged as particularly promising candidates for use in
systems, electrochemical energy is stored in solutions of redox
nonaqueous RFBs. Examples of MCCs that have been studied
active materials that are pumped between external electrolyte
previously are shown in Figure 1. These include acetylacetonate
tanks across structured electrodes. The energy density of a RFB
is dictated by eq 1, which has three main parameters: voltage
window (Vcell), solubility (Cactive), and number of electrons
transferred (n).3
̂
E ∝ 0.5 × n × V × Cactive × F
(1)
cell
Aqueous RFBs are commercially available,4 and the most
common electrolytes are solutions of vanadium salts.5 These
offer the advantages of reversible electrochemistry, excellent
cyclability, and high solubility (∼2 M in H2O). However, their
energy densities are limited by the small voltage window of
water (∼1.2 V).2a,b,3 As a result, significant research effort over
the last 15 years has focused on identifying redox active
materials that can be used in solvents with wider voltage
windows, such as acetonitrile or propylene carbonate.6−12
These nonaqueous solvents could enable large increases in Vcell
and n, and thereby ultimately yield RFBs with significantly
higher energy densities. Redox active materials for this
application should meet at least five criteria. They should (1)
undergo kinetically fast and chemically reversible redox
reactions; (2) exhibit multiple reversible redox events over a
large voltage window; (3) have high chemical, electrochemical,
and cell-component stabilities in multiple charge (oxidation)
states; (4) exhibit high solubility in multiple charge states; and
Figure 1. Representative examples of MCCs used in nonaqueous
RFBs.
Received: June 12, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b01328
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
complexes (A),10 metallocene derivatives (B),11 dithiolate
vanadium complexes (C),12a and metal bipyridine and
phenanthroline adducts (D).8,9 Complexes A−D meet many,
but not all, of the five criteria. For instance, an alkylammonium
ferrocene derivative (B)11a was shown to undergo reversible
redox reactions and exhibit high solubility (1.7 M in
alkylcarbonate solvent mixtures). However, this material is
limited to a single redox event (n = 1e−) at 3.49 V versus Li/Li+,
thus requiring the use of a low-potential redox active partner
(often Li-metal), which leads to an increase in the complexity
of the system and membrane design.
Scheme 1. Synthesis of Bipyridine Ligands (a) L3−L14 and
(b) L16 and L17
The majority of previous investigations of MCCs for RFB
applications have focused on metal complexes bearing a single,
commercially available ligand (e.g., acetylacetonate or 2,2′-
bipyridine). As such, there are very few systematic studies of the
impact of ligand structure and substitution on the solubility and
redox properties of MCCs.10a,h Such studies would provide
structure−solubility−electrochemistry relationships that could
be broadly useful for the design of new generations of MCCs
for this application. A key objective of our program is to
leverage our team’s complementary expertise in synthetic
inorganic chemistry and electrochemistry/battery engineering
to delineate these structure−function relationships in prospec-
tive battery materials.
We report herein the design, synthesis, solubility, and
electrochemical characterization of a series of bipyridine
chromium complexes for use as electrolytes in nonaqueous
RFBs. These complexes exhibit up to six redox couples over an
approximately 2 V window13 and thus offer the potential for
symmetric, multielectron cells (i.e., having the same active
species in both half-cells of the battery).14 We demonstrate that
the incorporation of substituents on the bipyridine ligands
impacts the number of redox couples observed within the
electrochemical window of acetonitrile as well as the position
and reversibility of these couples. Furthermore, we show that
changes to both ligand substituents and metal oxidation state
have dramatic effects on solubility.
followed by alkylation of the dihydroxybipyridine intermediate
with the appropriate alkyl halide (Scheme 1b). Ligands L16 and
L17 were obtained in 37% and 65% yield over these two steps,
respectively.
Overall, 17 tris-bipyridyl Cr3+ complexes and 7 tris-bipyridyl
Cr0 complexes were prepared in this study (Scheme 2). The
Cr3+ complexes were obtained in 45−99% yield by stirring the
appropriate ligand with [Cr(MeCN)4](BF4)2 in acetonitrile for
15 min at room temperature, followed by 1e− oxidation with
AgBF4 (Scheme 2a). The Cr0 complexes were prepared in 42−
99% yield by refluxing the appropriate ligand with Cr(CO)6 in
degassed mesitylene under a nitrogen atmosphere (Scheme
2b). All of the metal complexes were characterized via
elemental analysis and mass spectrometry. In addition, an X-
ray crystal structure of [Cr(L15)3]3+ was obtained, which
shows an octahedral geometry with all Cr−N bond lengths in
the range of 2.032−2.046 Å (Figure 2).
Solubility of Ester-bipyridine Cr3+ Complexes. We first
assessed the solubility of [Cr(L1)3]3+-[Cr(L14)3]3+ in
acetonitrile using UV−vis absorption spectroscopy. Acetonitrile
was selected as the solvent based on its wide electrochemical
window (∼5 V) and low viscosity. The solubilities of the Cr3+
complexes derived from the commercially available bipyridine
(L1) and 4,4′-dimethylbipyridine (L2) ligands were deter-
mined and used as benchmarks for comparison to the
synthesized analogues. Solubility is a complicated phenomenon
that involves the interplay of many, often competing, factors.17
Nonetheless, we initially hypothesized that three features of R
were most likely to favorably impact the solubility of these
complexes in acetonitrile: (1) longer chain length, (2) more
chain branching, and (3) replacing −CH2− groups with more
electronegative atoms in the chain. We reasoned that increasing
the chain length and branching could enhance solubility by
disrupting packing in the solid state.17 In addition, we reasoned
that replacing nonpolar −CH2− groups with more polar atoms
such as oxygen could increase solubility by enhancing favorable
dipole−dipole interactions between the solute and the
acetonitrile.
RESULTS AND DISCUSSION
■
Design and Synthesis of Metal Complexes. Our initial
work focused on the synthesis of a range of air-stable ester-
substituted bipyridine Cr3+ complexes. The ester substituent
can be easily modified, thus enabling fast and scalable
derivatization. We also synthesized the Cr0 analogues of several
of these ester-bipyridine compounds. These Cr0 complexes
represent another oxidation state accessed upon charging and
discharging the flow battery. A previous report from our groups
showed that the overall charge of a MCC has a large impact on
solubility.14 Thus, we pursued a detailed investigation of the
impact of ligand modifications on the solubility of both Cr3+
and Cr0 complexes. Finally, we synthesized and characterized a
series of alkoxybipyridine Cr3+ complexes. We targeted these
more electron-rich Cr3+ species to gain insights into the impact
of ligand electronic properties on the solubility and electro-
chemistry of these systems.
As summarized in Figure 3, the solubilities of the Cr3+
complexes of ligands L3−L14 range from 0.05 to 0.71 M.
Increasing the alkyl chain length from one carbon ([Cr-
(L3)3]3+) to four carbons ([Cr(L5)3]3+)18 resulted in a >5-fold
increase in solubility from 0.13 to 0.71 M. However, a further
increase in the alkyl chain length to a seven carbon chain
([Cr(L7)3]3+) led to a decrease in solubility relative to
[Cr(L5)3]3+ (0.39 M versus 0.71 M). The Cr complexes
bearing seven and eight carbon chains had the same solubility
(0.39 M for [Cr(L7)3]3+ and [Cr(L8)3]3+). We hypothesize
that the lower solubility of [Cr(L7)3]3+ and [Cr(L8)3]3+
compared to [Cr(L5)3]3+ may be due to the formation of a
lipophilic shell around the more hydrophobic complexes, which
The ester-bipyridine ligands were prepared via Jones
oxidation of 4,4′-dimethylbipyridine to afford the correspond-
ing dicarboxylic acid,15 which was further converted to the acid
chloride by reaction with oxalyl chloride. Esterification with the
corresponding alcohol afforded the desired ligands (L3−L14)
in 31−67% yield over the three-step sequence (Scheme 1a).
The 4,4′-alkoxy-substituted bipyridine ligands were prepared
by demethylation of 4,4′-dimethoxybipyridine with HBr,16
B
DOI: 10.1021/acs.inorgchem.5b01328
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 2. Synthesis of Ester-bipyridine and Alkoxy-bipyridine Cr3+ (a) and Cr0 Complexes (b)
length has a more significant impact on solubility than
branching in [Cr(L4)3]3+ versus [Cr(L3)3]3+.
A comparison of the Cr3+ complexes with ligands L9−L12
allowed us to assess the impact of electronegative groups on
solubility. For example, replacing two −CH2− groups in the
ester chains of [Cr(L7)3]3+ with electronegative oxygen atoms
in [Cr(L11)3]3+ resulted in enhanced solubility (0.39 M versus
0.54 M). [Cr(L7)3]3+ and [Cr(L11)3]3+ can also be compared
to [Cr(L9)3]3+, which contains a benzyl group with a seven
atom backbone. The low polarity of the phenyl ring along with
the possibility of stabilizing π−π interactions in the solid state
would be expected to render this complex significantly less
soluble than [Cr(L7)3]3+ and [Cr(L11)3]3+. Indeed, the
solubility of [Cr(L9)3]3+ was only 0.05 M in acetonitrile.
While the introduction of oxygen atoms into the seven atom
ester chain had some impact on solubility, an analogous effect
was not observed in compounds containing four atom ester
chains (i.e., complexes bearing ligands L5 and L10). For
instance, comparing [Cr(L5)3]3+ and [Cr(L10)3]3+, the
replacement of −CH2− with −O− had a minimal impact on
solubility (0.71 and 0.62 M, respectively). In addition, the
incorporation of the cyclic ether chain in [Cr(L12)3]3+ had
minimal impact on solubility when compared to the linear
ethylene glycol ether in complexes [Cr(L10)3]3+ and [Cr-
(L11)3]3+ (0.56 M versus 0.62 and 0.54 M, respectively).
Finally, we examined the incorporation of other polar
functional groups into the ester chains. Complexes containing
a cyano group and trifluoromethyl substituents on the esters
([Cr(L13)3]3+ and [Cr(L14)3]3+, respectively) showed sol-
ubilities in a range similar to the most soluble complexes (0.58
and 0.61 M, respectively). Overall, while some trends in
solubility can be discerned from this series, it appears that the
solubility of these Cr3+ complexes reaches a plateau at
approximately 0.6−0.7 M, and the complexes bearing four
atom ester chains ([Cr(L5)3]3+, ([Cr(L6)3]3+, [Cr(L10)3]3+,
and [Cr(L13)3]3+) have consistently high solubilities.
Figure 2. Solid-state X-ray crystal structure of [Cr(L15)3]3+. Thermal
ellipsoids at 50% probability. The hydrogen atoms and the counterions
are omitted for clarity.
offsets the effects of chain elongation and disorder on molecular
packing.
The complex bearing an isopropyl ester ([Cr(L4)3]3+)
showed a significant improvement in solubility relative to the
analogous methyl ester-containing compound [Cr(L3)3]3+
(0.34 M versus 0.13 M, respectively). This could be due to
either the increase in chain length or the increase in branching.
In an attempt to deconvolute these effects, we compared the
butyl and sec-butyl ester derivatives [Cr(L5)3]3+ and [Cr-
(L6)3]3+, which both contain a four carbon ester chain. The
complex bearing the more branched sec-butyl chain had a
slightly lower solubility than that with the linear butyl chain
(0.6 M versus 0.71 M, respectively). This suggests that chain
Solubility of Cr0 Complexes. Solubilities of the neutral
Cr0 complexes containing ligands L3, L5−L7, and L9−L11
C
DOI: 10.1021/acs.inorgchem.5b01328
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Solubility of [Cr(L)3]3+ in acetonitrile.
Figure 4. Solubility of [Cr(L)3]0 in acetonitrile.
were also assessed. In these systems, the solubility in
acetonitrile ranged from essentially insoluble (<0.0015 mM)19
to 210 mM (Figure 4). For all of the ligands examined, the
neutral Cr0 compounds were significantly less soluble than the
corresponding cationic Cr3+ species. In the neutral complexes,
moving from the one carbon chain in [Cr(L3)3]0 to the four
carbon butyl chain in [Cr(L5)3]0 improved solubility by almost
an order of magnitude (0.48 mM versus 3.9 mM, respectively).
However, upon further increasing the alkyl chain length to a
seven carbon chain ([Cr(L7)3]0), a significantly lower solubility
was observed (0.018 mM). Again, this may be due to the
lipophilic nature of the heptyl chain.
The incorporation of a branched sec-butyl chain ([Cr-
(L6)3]0) reduced the solubility of the complex as compared to
the linear butyl chain (0.089 mM versus 3.9 mM, respectively).
As discussed above, a similar decrease in solubility with
branching was observed in the Cr3+ complexes, although the
results for the neutral complex are more pronounced. The
incorporation of the benzyl chain ([Cr(L9)3]0) led to a
complex that is essentially insoluble in acetonitrile (<0.0015
mM).19 This is likely due to the nonpolar nature of this
substituent as well as stabilizing π−π stacking interactions of
the aromatic units in the solid state.
chain ([Cr(L5)3]0) to the methoxyethyl chain ([Cr(L10)3]0)
resulted in an ∼3-fold increase in solubility (3.9 versus 10 mM,
respectively). Even more dramatically, moving from a heptyl
chain ([Cr(L7)3]0) to a methoxy(ethoxy)ethyl chain ([Cr-
(L11)3]0) afforded an almost four order of magnitude
enhancement in solubility (0.018 versus 210 mM, respectively).
Solubility of Alkoxybipyridine Cr3+ Complexes. The
solubilities of the alkoxybipyridine Cr3+ complexes [Cr-
(L15)3]3+ to [Cr(L17)3]3+ were also determined in acetonitrile
and range from 0.29 to 0.61 M (Figure 5). Ethers bearing
methyl and butyl chains ([Cr(L15)3]3+ and [Cr(L16)3]3+)
yielded compounds with similar solubilities (0.61 and 0.58 M,
respectively). Somewhat surprisingly, the incorporation of the
methyl ether diethylene glycol chain ([Cr(L17)3]3+) resulted in
a complex with only moderate solubility in acetonitrile (0.29
M). In general, the alkoxy-derived Cr3+ molecules reach a
similar saturation point as the ester-bipyridine Cr3+ species.
Overall, these studies reveal two key points regarding
solubility:
(1) There are significant differences in solubility (up to 4
orders of magnitude) between active species bearing the same
supporting ligands but different oxidation states (i.e., Cr3+
versus Cr0). This must be accounted for in nonaqueous RFB
battery design and cycling.
The incorporation of electronegative atoms on the ester
chains led to dramatic improvements in solubility compared to
the aliphatic analogues. For example, moving from the butyl
(2) The solubility of these molecules is subject to a
complicated interplay of factors. The solubility of Cr3+
D
DOI: 10.1021/acs.inorgchem.5b01328
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. Solubility of functionalized alkoxy-bipyridine [Cr(L)3]3+ in
acetonitrile.
Figure 6. Cyclic voltammograms of a 10 mM solution of complex with
0.1 M TBABF4 in acetonitrile. Reference is Ag/Ag+ with AgBF4 (0.01
M); working electrode is glassy carbon disk; counter electrode is
platinum wire; scan rate is 100 mV s−1; temperature is 23 °C. (A)
[Cr(L5)3]3+, (B) [Cr(L8)3]3+, and (C) [Cr(L14)3]3+. Bar indicates
starting potential, and arrow indicates starting sweep direction.
complexes is relatively insensitive to substituent effects (with
values ranging from 0.2 to 0.7 M for most complexes). We
hypothesize that the overall 3+ charge dominates the solubility
properties of these molecules, rather than the specific
substituents on the ligands. In contrast, the solubilities of the
analogous neutral Cr0 species varies over 4 orders of magnitude
(∼0.02 to 200 mM). At Cr0, the largest solubility enhancement
was achieved by replacing nonpolar −CH2− groups with
electronegative oxygen atoms on the ester side chains of the
ligands.
redox potential (E1/2) of the first couple of [Cr(L1)3]3+ and
[Cr(L2)3]3+, we observe a significant shift toward a more
negative potential with the more electron-donating methyl-
substituted ligands (−0.56 V versus −0.72 V, respectively).
These values are in accordance with those reported in the
literature.21 When the complexes are cycled to potentials below
−2.5 V, a new shoulder at approximately −1.4 V is observed.
We attribute this shoulder to irreversible decomposition of the
complexes. This indicates that only the first four redox couples
are sufficiently stable to be considered for cycling in a RFB.
Nevertheless, this chemistry could afford a symmetrical cell
with 2e− transfers at each electrode.
Complexes bearing the electron-withdrawing ester-substi-
tuted bipyridine ligands generally show six couples between
−0.14 V and −2.05 V (Table 2 and Figure 6A). While the
electronic structure of Cr(bpy)3 complexes in the various
oxidation states is still the subject of significant debate,22
Wieghardt and co-workers have suggested that the redox events
of these complexes are primarily ligand-centered.23 Most
modifications to the ester substituent have minimal impact
on the electrochemistry (Table 2). For the majority of the ester
bipyridine complexes, the first three redox couples (first,
second, third) are spaced by approximately 0.5 V, whereas the
last three redox couples are closer in potential, with a spacing of
about 0.2 V.24 This is consistent with previous literature reports
Electrochemistry of Cr-Complexes. Cyclic voltammetry
for all Cr3+ and Cr0 complexes was performed in a three-
electrode cell in acetonitrile with TBABF4 as the supporting
electrolyte and Ag/Ag+ as the reference. Tables 1−4 list the
redox potentials (E1/2, average of the cathodic and anodic
red
potentials), the peak current ratios after deconvolution (ip
/
ipox, ratio of cathodic current to anodic current), and the peak
potential separation (ΔEp, difference between cathodic and
anodic potentials) for each of the complexes. The couple at the
most positive potential is designated as the first redox couple
(Figure 6A). A peak current ratio of 1 is indicative of a redox
couple that is chemically reversible on the time scale of the CV
experiment. It is important to note that these CV reversibility
measurements do not necessarily correlate with cyclability in
electrochemical evaluations over longer timeframes (e.g.,
charge−discharge). Nonetheless, CV serves as a useful method
to conduct a rapid initial analysis of MCC electrochemistries.20
The bipyridine [Cr(L1)3]3+ and dimethylbipyridine [Cr-
(L2)3]3+ complexes show six well-defined redox couples across
2.3 and 2.1 V, respectively (Table 1). When comparing the
Table 1. Potentials, Peak Current Ratios, and Peak Separations for the Bipyridine Cr3+ and Dimethylbipyridine Cr3+ Complexes
a
in Acetonitrile
compound
first
second
third
fourth
fifth
sixth
[Cr(L1)3]3+
E1/2 (V)
−0.56
1.03
−1.08
1.19
−1.65
0.91
−2.27
1.04
−2.58
1.02
−2.81
1.62
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.084
−0.72
0.94
0.077
−1.22
0.98
0.077
−1.75
0.94
0.077
−2.34
1.04
0.084
−2.62
1.05
0.084
−2.84
2.11
[Cr(L2)3]3+
ox
ipred/ip
ΔEp (V)
0.070
0.077
0.077
0.077
0.091
0.056
a
Reference is Ag/Ag+ with AgBF4 (0.01 M); supporting electrolyte is TBABF4 (0.1 M); working electrode is glassy carbon disk; counter electrode is
platinum wire; scan rate is 100 mV s−1; temperature is 23 °C.
E
DOI: 10.1021/acs.inorgchem.5b01328
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Table 2. Potentials, Peak Current Ratios, and Peak Separations for the Ester-bipyridine Cr3+ Complexes in Acetonitrile
compound
first
second
third
fourth
fifth
sixth
[Cr(L3)3]3+
E1/2 (V)
−0.20
1.02
−0.61
1.04
−1.14
1.04
−1.67
1.03
−1.86
0.94
−2.03
0.95
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.070
−0.17
0.99
0.070
−0.59
0.99
0.077
−1.22
1.00
0.063
−1.65
0.98
0.056
−1.85
0.97
0.056
−2.02
0.95
[Cr(L4)3]3+
[Cr(L5)3]3+
[Cr(L9)3]3+
[Cr(L10)3]3+
[Cr(L11)3]3+
[Cr(L12)3]3+
[Cr(L13)3]3+
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.060
−0.20
0.99
0.060
−0.61
0.96
0.065
−1.14
0.99
0.060
−1.67
0.94
0.045
−1.86
0.97
0.050
−2.04
1.02
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.072
−0.17
1.01
0.078
−0.57
0.91
0.080
−1.11
0.58
0.071
−1.64
0.24
0.066
−1.82
0.63
0.069
−1.99
0.87
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.078
−0.14
0.58
0.078
−0.57
1.11
0.070
−1.13
0.95
0.083
−1.67
1.29
0.066
−1.89
0.87
0.072
−2.07
0.52
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.13
0.11
0.11
0.12
0.12
0.14
−0.17
1.06
−0.58
0.90
−1.12
0.98
−1.63
0.98
−1.82
1.02
−2.00
1.06
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.084
−0.21
0.99
0.070
−0.61
0.96
0.084
−1.15
1.01
0.063
−1.68
0.95
0.056
−1.87
0.96
0.070
−2.04
0.99
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.093
−0.17
0.99
0.094
−0.59
0.96
0.10
0.093
−1.63
0.95
0.092
−1.83
0.96
0.096
−2.02
0.99
−1.12
1.01
ox
ipred/ip
ΔEp (V)
0.14
0.13
0.13
0.14
0.14
0.14
a
Reference is Ag/Ag+ with AgBF4 (0.01 M); supporting electrolyte is TBABF4 (0.1 M); working electrode is glassy carbon disk; counter electrode is
platinum wire; scan rate is 100 mV s−1; temperature is 23 °C.
a
Table 3. Potentials, Peak Current Ratios, and Peak Separations for the Alkoxy-bipyridine Cr3+ Complexes in Acetonitrile
compound
first
second
third
fourth
fifth
sixth
[Cr(L15)3]3+
E1/2 (V)
−0.92
0.70
−1.36
0.79
−1.74
0.65
−2.30
0.75
−2.57
0.92
−2.80
0.47
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.090
−0.94
0.98
0.095
−1.38
0.98
0.10
0.10
0.12
0.076
−2.79
0.25
[Cr(L16)3]3+
[Cr(L17)3]3+
−1.76
1.00
−2.31
0.97
−2.57
1.00
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.077
−0.93
0.87
0.063
−1.37
1.03
0.070
−1.76
0.90
0.070
−2.30
1.01
0.063
−2.55
0.73
0.042
ox
ipred/ip
ΔEp (V)
0.070
0.070
0.077
0.063
0.077
a
Reference is Ag/Ag+ with AgBF4 (0.01 M); supporting electrolyte is TBABF4 (0.1 M); working electrode is glassy carbon disk; counter electrode is
platinum wire; scan rate is 100 mV s−1; temperature is 23 °C.
of related systems.13 Widely spaced redox events are desirable
for RFB applications, since the potential window of the cell
(Vcell) is determined by the spacing between the highest and
lowest potential redox couple. In principle, the six redox
couples associated with these ester-substituted bipyridine Cr
complexes should enable a symmetrical 3e− RFB cell in which a
Cr0 complex is oxidized by 3e− on one side of the cell and
reduced by 3e− on the other side of the cell.25
A few ester substituents resulted in irreversible electro-
chemical behavior. For example, the cyclic voltammograms of
complexes bearing sec-butyl ([Cr(L6)3]3+), heptyl ([Cr-
(L7)3]3+), and octyl chains ([Cr(L8)3]3+) show large anodic
currents at approximately −2.0 V (see Figure 6B for an
example). These large increases in current could be due to
ligand dissociation, as noted for other Cr(bpy)3 complexes,21 or
may be due to the formation of side products with high
diffusivities. The cyclic voltammograms of the corresponding
free bipyridine ligands show quasi-reversible peaks between −2
V and −2.5 V. The perfluorinated ester-containing complex
[Cr(L14)3]3+ also shows multiple irreversible redox events
(Figure 6C). This is likely due to low stability of the fluorinated
alkyl chains at highly reducing potentials.26
For the alkoxy-substituted bipyridine Cr3+ complexes, the
first reduction wave occurs at −0.9 V, which is significantly
more negative than for the corresponding ester-derived
complexes (Table 3). This is consistent with the resonance
electron-donor properties of alkoxy substituents.27 The
complexes bearing alkoxy-substituted bipyridines show four
quasi-reversible redox processes by cyclic voltammetry (Figure
7A). There is a decrease in the voltage separation between each
redox couple relative to the analogous ester-substituted
derivatives. At strongly reducing potentials (V > −2.5 V), the
alkoxy-bipyridine complexes exhibit poor stability, which results
in the observation of small shoulders (−1.59, −1.19, 0.84 V,
Figure 7B). Additionally, the fifth peak in all of the alkoxy-
bipyridine complexes has a larger peak current compared to the
F
DOI: 10.1021/acs.inorgchem.5b01328
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
The sixth redox couple shows a large peak current as compared
to the others. Interestingly, analogous behavior is not observed
in the Cr3+ series. The reason for this increase in peak current is
unclear, but it might be due to the generation and reduction of
the free ester-bipyridine ligand.21 Overall, the neutral
complexes show less reversible peaks than those of the Cr3+
complexes based on their peak current ratios. Notably, the low
solubility of the neutral complexes in acetonitrile requires that
the CV experiments be performed near saturation limits. This
leads to low peak currents compared to the double layer
current (background current). As a consequence of their low
solubility in acetonitrile, the Cr0 compounds bearing sec-butyl,
heptyl, and benzyl ester substituents ([Cr(L6)3]0, [Cr(L7)3]0,
and [Cr(L9)3]0) could not be characterized by CV.
Overall, the cyclic voltammetry studies revealed two key
points for the development of new redox active MCCs for flow
battery applications:
(1) Alterations to the MCC ligand backbone can significantly
impact the reversibility of the redox events as well as the
stability of the metal complex toward reduction. Most ester-
substituted bipyridine complexes maintain six reversible redox
couples; however, the incorporation of long aliphatic chains or
perfluorinated substituents results in molecules with irreversible
redox behavior and poor electrochemical stability.
Figure 7. Cyclic voltammogram of a 10 mM solution of [Cr(L16)3]3+
with 0.1 M TBABF4 in acetonitrile. Reference is Ag/Ag+ with AgBF4
(0.01M); working electrode is glassy carbon disk; counter electrode is
platinum wire; scan rate is 100 mV s−1; temperature is 23 °C. (A) CV
in black is for a voltage window of −2.5 to 1.5 V. (B) CV in red is for a
voltage window of −3.0 to 1.5 V. Bar indicates starting potential and
arrow indicates starting sweep direction.
(2) The electronic properties of the ligand lead to significant
changes in the stability of the complexes, number of accessible
redox couples, and cell potential. The MCCs derived from
bipyridine ligands bearing electron-donating substituents (H,
alkoxy and methyl) show lower stability at highly reducing
potentials (below −2.5 V), and they provide access to just four
stable redox couples across 1.4 V. In contrast, most of the ester-
substituted complexes exhibit enhanced electrochemical
stability at low potentials and provide six reversible redox
couples across ∼2 V window.
other redox couples in the complex (Figure 7B). This increase
in current may be related to free ligand in solution, as the
voltammogram of free ligand shows a reversible redox couple at
approximately −2.6 V. In the case of the ethylene glycol-
substituted complex ([Cr(L17)3]3+), the sixth peak is outside
of the electrochemical window of acetonitrile (Table 3).
The cyclic voltammograms of the Cr0 complexes are similar
to those of the corresponding Cr3+ species (compare Figure 8A
and Figure 8B). Six 1e− couples are present at the same
potentials as their Cr3+ analogues (Table 4). The first three
redox couples are widely spaced (∼0.5 V), whereas the last
three peaks (fourth, fifth, sixth) are closer in potential (∼0.2 V).
CONCLUSIONS
■
This report describes a systematic study of the impact of
bipyridine ligand substitution on the solubility and cyclic
voltammetry of Cr(bpy)3 complexes. These studies show that
solubility in the Cr0 state is currently the limiting factor for
achieving high energy densities. However, this solubility can be
enhanced by 4 orders of magnitude through the incorporation
of ester substituents bearing polar oxygen atoms. These
substituents have minimal impact on the electrochemical
behavior (as measured by cyclic voltammetry). On the basis
of all of these investigations, complex [Cr(L11)3]0 was selected
as a particularly promising candidate to move forward to more
advanced battery testing.14 Our initial studies (reported
independently) have shown that an unoptimized H-cell
containing [Cr(L11)3]0 is capable of accessing four of the six
available redox couples (two redox couples at each electrode).14
As such, complex [Cr(L11)3]0 can be compared directly to
the alkylammonium ferrocene derivative B,11a which is among
the most promising MCC RFB active species reported to date.
Compound B is approximately 8-fold more soluble than
[Cr(L11)3]0 (1.7 M versus 0.2 M, respectively). Compound B
also has higher stability than [Cr(L11)3]0, as it can be cycled
over 100 times without significant capacity fade. In contrast,
[Cr(L11)3]0 begins to fade after ∼10 cycles in an H-cell.14
However, [Cr(L11)3]0 offers several key advantages relative
to B. Complex B participates in a single one-electron transfer
event, and as such, it can only be used as one side of an
asymmetric 1e− RFB (n = 1). In contrast, [Cr(L11)3]0 enables
Figure 8. (A) Cyclic voltammogram in black for a saturated solution
of [Cr(L5)3]0 (∼3.85 mM) with 0.1 M TBABF4 in acetonitrile.28 (B)
Cyclic voltammogram in red for a 10 mM solution of [Cr(L5)3]3+
with 0.1 M TBABF4 in acetonitrile. Reference is Ag/Ag+ with AgBF4
(0.01M); working electrode is glassy carbon disk; counter electrode is
platinum wire; scan rate is 100 mV s−1; temperature is 23 °C. Bar
indicates starting potential and arrow indicates starting sweep
direction.
G
DOI: 10.1021/acs.inorgchem.5b01328
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Table 4. Potentials, Peak Current Ratios, and Peak Separations for the Ester-bipyridine Cr0 Complexes in Acetonitrile
compound
first
second
third
fourth
fifth
sixth
[Cr(L3)3]0
E1/2 (V)
−0.18
1.56
−0.59
0.95
−1.12
1.02
−1.65
1.12
−1.83
0.34
−2.01
1.20
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.049
−0.18
0.99
0.051
−0.58
0.83
0.054
−1.13
0.98
0.052
−1.66
0.96
0.050
−1.85
1.08
0.049
−2.02
0.96
[Cr(L5)3]0
[Cr(L10)3]0
[Cr(L11)3]0
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.077
−0.17
0.92
0.047
−0.58
1.04
0.064
−1.12
0.91
0.059
−1.64
0.71
0.049
−1.82
1.00
0.059
−2.00
0.89
ox
ipred/ip
ΔEp (V)
E1/2 (V)
0.061
−0.18
1.3
0.061
−0.58
1.22
0.070
−1.11
1.13
0.078
−1.63
1.06
0.070
−1.82
0.98
0.085
−2.00
1.32
ox
ipred/ip
ΔEp (V)
0.065
0.062
0.067
0.072
0.059
0.078
a
Reference is Ag/Ag+ with AgBF4 (0.01 M); supporting electrolyte is TBABF4 (0.1 M); working electrode is glassy carbon disk; counter electrode is
platinum wire; scan rate is 100 mV s−1; temperature is 23 °C.
the construction of a symmetrical RFB in which two or three
electron transfers occur on each side of the cell (i.e., n = 2 or 3).
As shown in eq 1, a doubling or tripling of n results in a
doubling or tripling of the energy density compared to an RFB
with n = 1. The symmetric battery configuration (with the same
complex on both sides of the cell) offers the additional
advantage of mitigating irreversible performance erosion by
crossover of active species between the two sides of the cell.29
The advantages associated with complexes of general
structure [Cr(L11)3]0 should motivate future work aimed at
further enhancing the solubility and electrochemical stability of
these and related MCCs.30 With advances in these two areas,
MCCs that exhibit multielectron transfer events hold
considerable promise as active species for nonaqueous RFBs.
REFERENCES
■
(1) List of abbreviations: MCC = metal coordination complex; RFB
= redox flow battery; Vcell = voltage window; Cactive = solubility; n =
number of electrons transferred; F = Faraday’s constant; E = energy
̂
density; CV = cyclic voltammetry; V = voltage; i = current; TBABF4 =
tetrabutylammonium tetrafluoroborate; bpy = bipyridine; E1/2 = half-
wave potential; ipred/ipox = peak current ratio; ΔEp = peak potential
separation.
(2) (a) Yang, Z.; Zhang, J.; Kintner-Meyer, C. W.; Lu, X.; Choi, D.;
Lemmon, J. P.; Liu, J. Chem. Rev. 2011, 111, 3577. (b) Alotto, P.;
Guarnieri, M.; Moro, F. Renewable Sustainable Energy Rev. 2014, 29,
325. (c) Dunn, B.; Kamath, H.; Tarascon, J.-M. Science 2011, 334, 928.
(d) Weber, A. Z.; Mench, M. M.; Meyers, J. P.; Ross, P. N.; Gostick, J.
T.; Liu, Q. J. Appl. Electrochem. 2011, 41, 1137.
(3) (a) Ponce de Leon
Szanto, D. A.; Walsh, F. C. J. Power Sources 2006, 160, 716. (b) Huang,
Q.; Wang, Q. ChemPlusChem 2015, 80, 312.
(4) Wang, W.; Luo, Q.; Li, B.; Wei, X.; Li, L.; Yang, Z. Adv. Funct.
Mater. 2013, 23, 970.
́ ́
, C.; Frías-Ferrer, A.; Gonzalez-García, J.;
́
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.5b01328.
(5) Skyllas-Kazacos, M.; Chakrabarti, M. H.; Hajimolana, S. A.;
Mjalli, F. S.; Saleem, M. J. Electrochem. Soc. 2011, 158, R55.
(6) For all-organic redox active molecules for nonaqueous RFBs, see:
(a) Brushett, F. R.; Vaughey, J. T.; Jansen, A. N. Adv. Energy Mater.
2012, 2, 1390. (b) Li, Z.; Li, S.; Liu, S.; Huang, K.; Fang, D.; Wang, F.;
Peng, S. Electrochem. Solid-State Lett. 2011, 14, A171. (c) Kaur, A. P.;
Holubowitch, N. E.; Ergun, S.; Elliott, C. F.; Odom, S. A. Energy
Technology 2015, 3, 476. (d) Huang, J.; Cheng, L.; Assary, R. S.; Wang,
P.; Xue, Z.; Burrell, A. K.; Curtiss, L. A.; Zhang, L. Adv. Energy Mater.
2015, 5, 1401782. (e) Wei, X.; Xu, W.; Huang, J.; Zhang, L.; Walter,
E.; Lawrence, C.; Vijayakumar, M.; Henderson, W. A.; Liu, T.;
Cosimbescu, L.; Li, B.; Sprenkle, V.; Wang, W. Angew. Chem., Int. Ed.
2015, 54, 8684. (f) Oh, S. H.; Lee, C. W.; Chun, D. H.; Jeon, J. D.;
Shim, J.; Shin, K. H.; Yang, J. H. J. Mater. Chem. A 2014, 2, 19994.
(7) For metal−organic hybrid nonaqueous RFBs: (a) Wang, W.; Xu,
W.; Cosimbescu, L.; Choi, D.; Li, L.; Yang, Z. Chem. Commun. 2012,
48, 6669. (b) Wei, X.; Xu, W.; Vijayakumar, M.; Cosimbescu, L.; Liu,
T.; Sprenkle, V.; Wang, W. Adv. Mater. 2014, 26, 7649. For semisolid
redox active electrolytes for nonaqueous RFBs, see: (c) Duduta, M.;
Ho, B.; Wood, V. C.; Limthongkul, P.; Brunini, V. E.; Carter, W. C.;
Chiang, Y.-M. Adv. Energy Mater. 2011, 1, 511. (d) Hamelet, S.;
Larcher, D.; Dupont, L.; Tarascon, J.-M. J. Electrochem. Soc. 2013, 160,
A516. (e) Hamelet, S.; Tzedakis, T.; Leriche, J.-B.; Sailler, S.; Larcher,
D.; Taberna, P.-L.; Simon, P.; Tarascon, J.-M. J. Electrochem. Soc. 2012,
159, A1360. (f) Youssry, M.; Madec, L.; Soudan, P.; Cerbelaud, M.;
Guyomard, D.; Lestriez, B. Phys. Chem. Chem. Phys. 2013, 15, 14476.
For polyoxometalates for nonaqueous RFBs, see: (g) Pratt, H. D., III;
Hudak, N. S.; Fang, X.; Anderson, T. M. J. Power Sources 2013, 236,
259. (h) Pratt, H. D., III; Anderson, T. M. Dalton Trans. 2013, 42,
15650.
Experimental procedures and characterization of all new
compounds including spectroscopic and potentiometric
data (PDF)
X-ray crystallographic data of [Cr(L15)3]3+ (CIF)
AUTHOR INFORMATION
■
Corresponding Authors
*(L.T.T.) E-mail: ltt@umich.edu.
*(M.S.S.) E-mail: mssanfor@umich.edu.
Author Contributions
‡X.Y. and J.A.S. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported as part of the Joint Center for Energy
Storage Research, an Energy Innovation Hub funded by the
U.S. Department of Energy, Office of Science, Basic Energy
Sciences. We gratefully acknowledge Dr. Jeff Kampf for X-ray
crystallographic analysis of [Cr(L15)3]3+, as well as funding
from NSF Grant CHE-0840456 for X-ray instrumentation.
H
DOI: 10.1021/acs.inorgchem.5b01328
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(8) Nonaqueous systems MCCs bearing bipyridine or phenanthro-
line ligands: (a) Matsuda, Y.; Morita, T. M.; Tanaka, K.; Okada, M.;
Matsumura-Inoue, T. Denki Kagaku 1985, 53, 632. (b) Matsuda, Y.;
Tanaka, K.; Okada, M.; Takasu, Y.; Morita, M.; Matsumura-Inoue, T. J.
Appl. Electrochem. 1988, 18, 909. (c) Morita, M.; Tanaka, Y.; Tanaka,
K.; Matsuda, Y.; Matsumura-Inoue, T. Bull. Chem. Soc. Jpn. 1988, 61,
2711. (d) Mun, J.; Lee, M.-J.; Park, J.-W.; Oh, D.-J.; Lee, D.-Y.; Doo,
S.-G. Electrochem. Solid-State Lett. 2012, 15, A80. (e) Chakrabarti, M.
H.; Dryfe, R. A. W.; Roberts, E. P. L. Electrochim. Acta 2007, 52, 2189.
(f) Kim, J.-H.; Kim, K. J.; Park, M.-S.; Lee, N. J.; Hwang, U.; Kim, H.;
Kim, Y.-J. Electrochem. Commun. 2011, 13, 997. (g) Chakrabarti, M.
H.; Lindfield Roberts, E. P.; Saleem, M. Int. J. Green Energy 2010, 7,
445. (h) Xing, X.; Zhang, D.; Li, Y. J. Power Sources 2015, 279, 205.
(i) Park, M.-S.; Lee, N. J.; Lee, S.-W.; Kim, K. J.; Oh, D.-J.; Kim, Y.-J.
ACS Appl. Mater. Interfaces 2014, 6, 10729.
(9) Patents about nonaqueous RFBs MCCs bearing bipyridine based
ligands: (a) Maly-Schreiber, M.; Whitehead, A. H. W.O. Patent
101284, September 13, 2007. (b) Sun, H.; Park, J.; Lee, D.; Son, S.
(Samsung Electronics Co., Ltd., KR). U.S. Patent 0195283, August 11,
2011. (c) Lee, M. J.; Lee, D.; Oh, D.; Park, J.; Hwang, S. (Samsung
Electronics Co., Ltd., KR). U.S. Patent 0107661, May 3, 2012. (d) Lee,
M.; Oh, D.; Hwang, S. (Sansung Electronics Co., Ltd., KR). Redox
Flow Battery. U.S. Patent 0171530, July 5, 2012. (e) Park, J.; Lee, M.;
Lee, D.; Oh, D. (Samsung Electronics Co., Ltd., KR). Redox Flow
Battery. U.S. Patent 0171541, July 5, 2012. (f) Mun, J.; Hwang, S.; Lee,
D.; Kim, Y.; Kwon, O.; Yim, T. (KRKR, Seoul National University
R&DB FoundationSamsung Electronics Co., Ltd., KRKR, Seoul
National University R&DB FoundationSamsung Electronics Co.,
Ltd.). Electrolyte for Redox Flow Battery and Redox Flow Battery
Including the Same. U.S. Patent 004819, January 3, 2013. (g) Park, J.;
OH, D.; Lee, D.; Lee, M. (Samsung Electronics Co., Ltd., KR).
Organic Electrolyte Solution and Redox Flow Battery Including the
Same. U.S. Patent 196,206, August 1, 2013. (h) Park, J.; OH, D.; Lee,
D.; Lee, M. (Samsung Electronics Co., Ltd., KR). Redox Flow Battery.
U.S. Patent 193,687, July 10, 2014.
(10) Nonaqueous MCCs bearing acetylacetonate ligands: (a) Yama-
mura, T.; Shiokawa, Y.; Yamana, H.; Moriyama, H. Electrochim. Acta
2002, 48, 43. (b) Liu, Q.; Sleightholme, A. E. S.; Shinkle, A. A.; Li, Y.;
Thompson, L. T. Electrochem. Commun. 2009, 11, 2312. (c) Sleigh-
tholme, A. E. S.; Shinkle, A. A.; Liu, Q.; Monroe, C. W.; Thompson, L.
T.; Li, Y. J. Power Sources 2011, 196, 5742. (d) Shinkle, A. A.;
Sleightholme, A. E. S.; Thompson, L. T.; Monroe, C. W. J. Appl.
Electrochem. 2011, 41, 1191. (e) Zhang, D.; Liu, Q.; Shi, X.; Li, Y. J.
Power Sources 2012, 203, 201. (f) Shinkle, A. A.; Sleightholme, A. E. S.;
Griffith, L. D.; Thompson, L. T.; Monroe, C. W. J. Power Sources 2012,
206, 490. (g) Zhang, D.; Lan, H.; Li, Y. J. Power Sources 2012, 217,
199. (h) Suttil, J. A.; Kucharyson, J. F.; Escalante-Garcia, I. L.; Cabrera,
P. J.; James, B. R.; Savinell, R. F.; Sanford, M. S.; Thompson, L. T. J.
Mater. Chem. A 2015, 3, 7929.
(11) Nonaqueous systems MCCs based on metallocene ligands:
(a) Wei, X.; Cosimbescu, L.; Xu, W.; Hu, J. Z.; Vijayakumar, M.; Feng,
J.; Hu, M. Y.; Deng, X.; Xiao, J.; Liu, J.; Sprenkle, V.; Wang, W. Adv.
Energy Mater. 2015, 5, 1400678. (b) Hwang, B.; Park, M.-S.; Kim, K.
ChemSusChem 2015, 8, 310. (c) Zhao, Y.; Ding, Y.; Song, J.; Li, G.;
Dong, G.; Goodenough, J. B.; Yu, G. Angew. Chem., Int. Ed. 2014, 53,
11036. (d) Park, K.; Cho, J. H.; Shanmuganathan, K.; Song, J.; Peng,
J.; Gobet, M.; Greenbaum, S.; Ellison, C. J.; Goodenough, J. B. J. Power
Sources 2014, 263, 52.
(15) Hoertz, P. G.; Staniszewski, A.; Marton, A.; Higgins, G. T.;
Incarvito, C. D.; Rheingold, A. L.; Meyer, G. J. J. Am. Chem. Soc. 2006,
128, 8234.
(16) Hong, Y.-R.; Gorman, C. B. J. Org. Chem. 2003, 68, 9019.
(17) (a) Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic
Chemistry, 1st ed.; University Science Books: CA, 2005; p 153−202.
̀
(b) Palmer, D. S.; Llinas, A.; Morao, I.; Day, G. M.; Goodman, J. M.;
Glen, R. C.; Mitchell, J. B. O. Mol. Pharmaceutics 2008, 5, 266.
(18) With complex [Cr(L5)3]3+, the relatively large error resulted
from the high viscosity of the saturated solution.
(19) Solubility estimated based on the lower detection limit of the
Nandrop UV−vis spectrometer (2 ng/mL).
(20) In principle, differences in the chain length of the ligand (and
thus molecular size) could impact the electron transfer kinetics and
diffusion (and thus limit power densities of a battery). However, in
previous work with M(acac)3 complexes (see ref 10h), we found small
differences in the diffusion coefficient of complexes with varying ligand
chain lengths, which were all within an acceptable range for RFB
applications (see ref 2d). These results suggest that this is unlikely to
be a problem in the current systems.
(21) (a) Sato, Y.; Tanaka, N. Bull. Chem. Soc. Jpn. 1969, 42, 1021.
(b) Hughes, M. C.; Rao, J. M.; Macero, D. J. Inorg. Chim. Acta 1979,
35, L321.
(22) For a discussion of the electronic structure of [M(bpy)3]
complexes see: (a) Creutz, C. Comments Inorg. Chem. 1982, 1, 293.
(b) Konig, E.; Herzog, S. J. Inorg. Nucl. Chem. 1970, 32, 585. For IR
̈
studies on low oxidation states of [Cr(bpy)3] suggesting that the bpy
ligand is negatively charged see: (c) Saito, Y.; Takemoto, J.;
Hutchinson, B.; Nakamoto, K. Inorg. Chem. 1972, 11, 2003. For a
discussion of the electronic structure of [Cr(bpy)3]+2: (d) Cotton, F.
A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M. Advanced Inorganic
Chemistry, 6th ed.; Wiley & Sons: New York, 1999; p 822. For a
discussion of the formal oxidation state of the metal center in
[Cr(bpy)3]+3 and [Cr(bpy)3]+2: (e) Wilkinson, G.; Gillard, R. D.;
McCleverty, J. A. Comprehensive Coordination Chemistry, 1st ed.;
Pergamon Press: Oxford, 1987; Vol. 3, p 712. For the synthesis and
characterization of [Cr(bpy)3]+2 and [Cr(bpy)3]+1: (f) Hein, F. V.;
Herzog, S. Z. Anorg. Allg. Chem. 1952, 267, 337. For the synthesis and
characterization of [Cr(bpy)3]0: (g) Herzog, S.; Schon, W. Z. Anorg.
Allg. Chem. 1958, 297, 323. (h) Behrens, H.; Muller, A. Z. Anorg. Allg.
̈
Chem. 1965, 341, 124. (i) Quirk, J.; Wilkinson, G. Polyhedron 1982, 1,
209. For the synthesis and characterization of [Cr(bpy)3]x (x = −1,
−2, −3) see: Herzog, S.; Grimm, U.; Waicenbauer, W. Z. Chem. 1967,
7, 355.
(23) Scarborough, C. C.; Sproules, S.; Weyhermuller, T.; DeBeer, S.;
̈
Wieghardt, K. Inorg. Chem. 2011, 50, 12446.
(24) For RFB applications, it is important that these closely spaced
redox events be reversible. If one of them is not, decomposition can
occur via disproportionation.
(25) In practice, we have been able to demonstrate charge−discharge
in a symmetrical cell with n = 2e− with complex [Cr(L11)3]0. See ref
14.
(26) Barker, D. J.; Brewin, D. H.; Dahm, R. H.; Hoy, L. R. J.
Electrochim. Acta 1978, 23, 1107.
(27) (a) Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry:
Part A: Structure and Mechanisms, 3rd ed.; Plenum Press: New York,
1991; pp 197−208. (b) Tucker, J. W.; Stephenson, C. R. J. J. Org.
Chem. 2012, 77, 1617.
(28) The cyclic voltammogram of [Cr(L5)3]0 shown in Figure 8 is
the third full scan. See the SI for full details on the CV analysis.
(29) Currently, the most common membranes used for nonaqueous
RFBs are anion exchange membranes, such as Neosepta AHA and
AMI-7001. However, these are not very effective at preventing cross-
over. For membrane developments in nonaqueous RFBs see: Shin, S.-
H.; Yun, S.-H.; Moon, S.-H. RSC Adv. 2013, 3, 9095.
(12) (a) Cappillino, P. J.; Pratt, H. D., III; Hudak, N. S.; Tomson, N.
C.; Anderson, T. M.; Anstey, M. R. Adv. Energy Mater. 2014, 4,
1300566. (b) Anderson, T. M.; Anstey, M.; Tomson, N. C. (Sandia
Co., US) U.S. Patent 20140239906, August 28, 2014.
(13) McDaniel, A. M.; Tseng, H.-W.; Damrauer, N. H.; Shores, M. P.
Inorg. Chem. 2010, 49, 7981.
(30) Long-term stability of the redox active species is an important
factor for the development of RFBs. We have conducted a solution
stability study for complex [Cr(L11)3]0. A saturated solution of the
complex was diluted in acetonitrile and stored for 24 h. UV−vis
(14) Cabrera, P. J.; Yang, X.; Suttil, J. A.; Hawthorne, K. L.; Brooner,
R. E. M.; Sanford, M. S.; Thompson, L. T. J. Phys. Chem. C 2015, 119,
15882.
I
DOI: 10.1021/acs.inorgchem.5b01328
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
absorption spectroscopy demonstrated unchanged spectral data
suggesting that no decomposition occurs during that time-scale (see
Supporting Information for details).
J
DOI: 10.1021/acs.inorgchem.5b01328
Inorg. Chem. XXXX, XXX, XXX−XXX