A Long-Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical

Article abstract of DOI:10.1016/j.chempr.2019.04.021

With the rapidly falling cost of photovoltaic and wind energy generation, grid-scale integration of renewable energy becomes inevitable. This, however, is impeded by the intrinsic intermittency of these renewable energy resources. Safe, cost-effective storage could solve this problem. Aqueous organic redox flow batteries (AORFBs), which store energy externally in low-cost electroactive water-soluble organic molecules, have emerged as promising for this application. Current AORFBs are limited by short lifetimes and the limited choices for redox-active material for the positive electrolyte. We report a long-lifetime TMAP-TEMPO/BTMAP-Vi all-organic aqueous flow battery, the capacity retention rate of which is among the highest of all-organic AORFBs. We discuss the potential cause of the stabilization of the free radical posolyte molecule. The results we report here constitute an important step toward massive-scale intermittent renewable energy penetration into the future electric grid. TMAP-TEMPO represents an extremely stable redox-active radical organic for an AORFB posolyte. An all-organic AORFB based on TMAP-TEMPO and BTMAP-Vi exhibits an OCV of 1.1 V and a long lifetime, featuring a concentration-independent temporal capacity retention rate of >99.974% per h, or a capacity retention rate of 99.993% per cycle over 1,000 consecutive cycles. The massive-scale integration of renewable electricity into the power grid is impeded by its intrinsic intermittency. The aqueous organic redox flow battery (AORFB) rises as a potential storage solution; however, the choice of positive electrolytes is limited, and the aqueous-soluble organic positive redox-active species reported to date have short lifetimes. Here we report a stable organic molecule for the positive terminal, 4-[3-(trimethylammonio)propoxy]-2,2,6,6-tetramethylpiperidine-1-oxyl (TMAP-TEMPO) chloride, exhibiting high (4.62 M) aqueous solubility. When operated in a practical AORFB against a negative electrolyte comprising BTMAP-viologen at neutral pH, the flow cell displayed an open-circuit voltage of 1.1 volts and a Coulombic efficiency of >99.73%. The capacity retention rate is among the highest of all-organic AORFBs reported to date, at 99.993% per cycle over 1,000 consecutive cycles; the temporal capacity fade rate of 0.026% per h is independent of concentration.

Full text of DOI:10.1016/j.chempr.2019.04.021

Article
A Long Lifetime All-Organic Aqueous Flow
Battery Utilizing TMAP-TEMPO Radical
Yahua Liu, Marc-Antoni Goulet,
Liuchuan Tong, ..., Michael J.
Aziz, Zhengjin Yang, Tongwen
Xu
maziz@harvard.edu (M.J.A.)
yangzj09@ustc.edu.cn (Z.Y.)
twxu@ustc.edu.cn (T.X.)
HIGHLIGHTS
Highly soluble and long lifetime
TMAP-TEMPO for pH 7 aqueous
flow batteries
The posolyte features a
concentration-independent
capacity retention rate
Temporal capacity retention rate
of >99.974% per h for over 1,000
cycles
Rational molecular design leads to
improved stability
TMAP-TEMPO represents an extremely stable redox-active radical organic for an
AORFB posolyte. An all-organic AORFB based on TMAP-TEMPO and BTMAP-Vi
exhibits an OCV of 1.1 V and a long lifetime, featuring a concentration-
independent temporal capacity retention rate of >99.974% per h, or a capacity
retention rate of 99.993% per cycle over 1,000 consecutive cycles.
Liu et al., Chem 5, 1–10
July 11, 2019 ª 2019 Elsevier Inc.
https://doi.org/10.1016/j.chempr.2019.04.021

Please cite this article in press as: Liu et al., A Long Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical, Chem (2019),
https://doi.org/10.1016/j.chempr.2019.04.021
Article
A Long Lifetime All-Organic
Aqueous Flow Battery Utilizing
TMAP-TEMPO Radical
1
2
3
1
3
1
3
Yahua Liu, Marc-Antoni Goulet, Liuchuan Tong, Yazhi Liu, Yunlong Ji, Liang Wu, Roy G. Gordon,
Michael J. Aziz, * Zhengjin Yang,1,4,* and Tongwen Xu1,*
2,
SUMMARY
The Bigger Picture
The massive-scale integration of renewable electricity into the power grid is
impeded by its intrinsic intermittency. The aqueous organic redox flow battery
With the rapidly falling cost of
photovoltaic and wind energy
generation, grid-scale integration
of renewable energy becomes
inevitable. This, however, is
impeded by the intrinsic
(
AORFB) rises as a potential storage solution; however, the choice of positive
electrolytes is limited, and the aqueous-soluble organic positive redox-active
species reported to date have short lifetimes. Here we report a stable organic
molecule for the positive terminal, 4-[3-(trimethylammonio)propoxy]-2,2,6,6-
tetramethylpiperidine-1-oxyl (TMAP-TEMPO) chloride, exhibiting high (4.62 M)
aqueous solubility. When operated in a practical AORFB against a negative elec-
trolyte comprising BTMAP-viologen at neutral pH, the flow cell displayed an
open-circuit voltage of 1.1 volts and a Coulombic efficiency of >99.73%. The ca-
pacity retention rate is among the highest of all-organic AORFBs reported to
date, at 99.993% per cycle over 1,000 consecutive cycles; the temporal capacity
fade rate of 0.026% per h is independent of concentration.
intermittency of these renewable
energy resources. Safe, cost-
effective storage could solve this
problem. Aqueous organic redox
flow batteries (AORFBs), which
store energy externally in low-cost
electroactive water-soluble
organic molecules, have emerged
as promising for this application.
Current AORFBs are limited by
short lifetimes and the limited
choices for redox-active material
for the positive electrolyte. We
report a long lifetime TMAP-
TEMPO/BTMAP-Vi all-organic
aqueous flow battery, the capacity
retention rate of which is among
the highest of all-organic
INTRODUCTION
Because of the emissions from fossil fuel combustion, efforts have been devoted to
creating a sustainable supply of energy for our modern lifestyle with a minimum
1
,2
carbon footprint. Facilitated by technological breakthroughs, more electricity is
generated from sunlight and wind with ever-increasing efficiencies and rapidly fall-
3
–6
ing cost; the massive-scale integration of energy from these renewable resources
into the power grid becomes inevitable. However, if directly connected to the grid,
the intrinsically intermittent nature of solar and wind energy will cause undesirable
power fluctuations, which could potentially be solved by energy storage systems.
To this end, numerous methods have been proposed to store electrical energy,
among which flow batteries are garnering interest for longer discharge duration
AORFBs. We discuss the potential
cause of the stabilization of the
free radical posolyte molecule.
The results we report here
7
–10
applications.
The energy content of flow batteries is stored externally in their
electrolytes (the posolyte, i.e., the positive electrolyte, and the negolyte, i.e., the
negative electrolyte). Therefore, their capital cost approaches the cost of these
electrolytes at sufficiently long discharge duration. If sufficiently inexpensive electro-
lytes can be developed, flow batteries may reach costs below those of conventional
battery technologies, such as lead-acid and lithium-ion. Within this framework,
AORFBs, which store energy in water-soluble redox-active organic compounds
constitute an important step
toward massive-scale intermittent
renewable energy penetration
into the future electric grid.
1
1–13
composed of earth-abundant elements, may enable low electrolyte costs.
Nevertheless, the lifetime of current AORFBs cannot currently meet the expected
9
,14
2
0 years of service life for widespread market penetration.
Capacity fade of
1
4
AORFBs is often the result of molecular decomposition of the organic reactants.
Chem 5, 1–10, July 11, 2019 ª 2019 Elsevier Inc.
1

Please cite this article in press as: Liu et al., A Long Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical, Chem (2019),
https://doi.org/10.1016/j.chempr.2019.04.021
Progress has been made in recent years, especially for the negolyte. For the partic-
ular case of neutral-pH systems, the focus of the current work, the first neutral-pH
1
5
AORFB employed a negolyte based on methyl viologen (MV). MV was later shown
1
6
to have a low capacity retention rate due to bimolecular annihilation. By intro-
ducing positively charged functional groups to enhance Coulombic repulsion, the
lifetime of an MV derivative, bis-(trimethylammonio) propyl viologen (BTMAP-Vi),
was extended to a multi-year timescale (for more details on the stability of MV-
1
6
derivatives see Supplemental Information). An example of an MV derivative
with intermediate stability is the 4,4ʹ-(thiazolo[5,4-d]thiazole-2,5-diyl)bis(1-(3-(trime-
1
1
thylammonio) propyl)pyridin-1-ium).
Despite the advancements achieved for synthesizing stable organic molecules for
aqueous negolytes, little progress has been made in designing stable organic
high-potential posolytes. In addition, the concentration-dependent cycling stability
1
1,15–18
of electrolyte in AORFBs remains a long-standing obstacle,
and only a
few studies conducted rigorous long-term cycling experiments (e.g., >500 cycles
1
9–21
and >1 week).
Here, we report an extremely stable variant of the radical posolyte molecule
4
-[3-(trimethylammonio)propoxy]-2,2,6,6-tetramethylpiperidine-1-oxyl (TMAP-TEMPO)
À1 9
chloride (Figures 1A, S1, and S2) synthesized from low-cost ($5–6 kg ) 4-hydroxy-
2
,2,6,6-tetramethylpiperidin-1-oxyl (4-OH-TEMPO) by adding Coulombic repulsion
to suppress bimolecular interactions and by eliminating functionalities that could
potentially lead to molecular decomposition. TMAP-TEMPO chloride features a
solubility of 4.62 M in water and a highly reversible single-electron oxidation potential
at 0.81 (V) versus the standard hydrogen electrode (SHE), approaching the thermody-
namic limit of water stability at neutral pH. When paired with BTMAP-Vi (Figures S3–
S5), the TMAP-TEMPO/BTMAP-Vi battery has an open-circuit voltage (OCV) of 1.1 V
and exhibits an extremely high capacity retention rate, at 99.993% per cycle over
1
,000 consecutive cycles; the temporal capacity fade rate of 0.026% per h is indepen-
dent of concentration.
RESULTS AND DISCUSSION
À1
A cyclic voltammetry (CV) study of TMAP-TEMPO at a scan rate of 100 mV s gave a
reversible oxidation peak at 0.81 V versus SHE (E1/2) in 1 M NaCl (Figure 1B). A peak
separation of 58 mV was observed in good accordance with the theoretical value of a
reversible single-electron redox reaction at pH 7 (59 mV), and the oxidation peak
does not shift when the scan rate varies (Figure S5). Regardless of the electrode
overpotential, the oxidation potential of TMAP-TEMPO, 0.81 V versus SHE, ap-
proaches the thermodynamic upper limit of water stability at pH 7 (oxygen evolution
reaction, 0.816 V versus SHE). No oxygen generation was detected with a ppm-level
sensitive oxygen detector when the experiment was conducted in a properly sealed
environment.
1CAS Key Laboratory of Soft Matter Chemistry,
iChEM (Collaborative Innovation Center of
Chemistry for Energy Materials), School of
Chemistry and Material Science, University of
Science and Technology of China, Hefei 230026,
P.R. China
2Harvard John A. Paulson School of Engineering
and Applied Sciences, 29 Oxford Street,
Cambridge, MA 02138, USA
Rotating disk electrode measurements of TMAP-TEMPO gave an oxidation rate con-
À2
À1
stant (k
0
) of 1.02 3 10 cm s (Figure S6). The rate constants of both TMAP-TEMPO
3
Department of Chemistry and Chemical Biology,
Harvard University, 12 Oxford Street, Cambridge,
MA 02138, USA
and BTMAP-Vi (Table 1) are much faster than those of most other inorganic, organic,
or organometallic species (Table S1).
4Lead Contact
*
Correspondence: maziz@harvard.edu (M.J.A.),
The promising electrochemical characteristics suggest that TMAP-TEMPO might
make a suitable posolyte for an AORFB. We then assembled a neutral-pH AORFB,
with a posolyte comprising TMAP-TEMPO chloride and a negolyte comprising
yangzj09@ustc.edu.cn (Z.Y.),
twxu@ustc.edu.cn (T.X.)
https://doi.org/10.1016/j.chempr.2019.04.021
2
Chem 5, 1–10, July 11, 2019

Please cite this article in press as: Liu et al., A Long Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical, Chem (2019),
https://doi.org/10.1016/j.chempr.2019.04.021
Figure 1. Syntheses of TMAP-TEMPO and BTMAP-Vi and Their Potential Use as Electrolytes in AORFB
(
(
A) Synthetic routes for TMAP-TEMPO and BTMAP-Vi.
B) Cyclic voltammograms of TMAP-TEMPO (solid red trace), 4-OH-TEMPO (dashed light red trace), and BTMAP-Vi (solid blue trace). The electroactive
À1
compounds were tested at 1 mM in 1 M NaCl solution at a scan rate of 100 mV s , on a glassy carbon working electrode.
(
C) Schematic of a neutral flow battery assembled with TMAP-TEMPO in the posolyte and BTMAP-Vi in the negolyte, with chloride ions passing through
a Selemion AMV anion-selective membrane.
BTMAP-Vi tetrachloride, separated by a Selemionꢀ AMV anion-exchange mem-
brane. In a typical charge process, electrons are withdrawn from the TMAP-TEMPO,
À
flow along the external circuit, and combine with the negolyte, BTMAP-Vi, while Cl
ions serve as the internal charge carrier, migrating across the Selemionꢀ AMV mem-
brane (Figure 1C). TMAP-TEMPO (the posolyte side) is the capacity-limiting side,
and as a reference, the 4-OH-TEMPO/BTMAP-Vi cell with the same electrolyte
concentration was also assembled (Figure 2, for details see the Supplemental
Information).
The total discharge capacity was 25.7 mAh, which corresponded to 95.9% of the
theoretical capacity (26.8 mAh). The deviation between the theoretical and the
measured value is attributed to the loss during material transfer and the electrode
overpotential during the charge-discharge process. We then stepwise charged the
cell at constant voltage (1.5 V) with a 10% increment in the state-of-charge (SOC)
and measured the polarizations of the cell. The OCV at 50% SOC rests at around
1
.10 V (Figure S7), which is slightly lower than expected from CV studies (1.19 V,
Figure 1B).
Chem 5, 1–10, July 11, 2019
3

Please cite this article in press as: Liu et al., A Long Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical, Chem (2019),
https://doi.org/10.1016/j.chempr.2019.04.021
Table 1. Electrochemical and Physicochemical Properties of TMAP-TEMPO, 4-OH-TEMPO, and BTMAP-Vi
2
À1
À1
À1
2
À1 a
s )
Electrolyte
D (cm
s
)
k
0
(cm s
)
E
1/2 (V versus SHE)
Water Solubility (mol L
)
Permeability (cm
Reference
À6
À5
À6
À2
À4
À2
À10
TMAP-TEMPO
3.48 3 10
2.95 3 10
3.60 3 10
1.02 3 10
2.60 3 10
2.20 3 10
0.81
0.80
-0.36
4.62
2.1
6.40 3 10
(AMV)
This work
À9
6
4
-OH-TEMPO
1.34 3 10 (AMV)
À10
7
BTMAP-Vi
2.0
6.70 3 10
5.20 3 10
(DSV)
À11
-
0.38
(AMV)
This work
a
Membrane names are included in the parentheses. Abbreviations are as follows: D, diffusion coefficient; k
0
, electron-transfer rate constant; and E1/2, redox
potential.
À2
At ꢀ100% SOC, the peak galvanic power density of the cell reaches 99.03 mW cm
À2
at a current density of 162.7 mA cm (Figure 2D), which is nearly the same as that of
À2
17
the neutral FcNCl/MV cell (100 mW cm ) with comparable OCV (1.05 V). The
modest peak power density of the TMAP-TEMPO/BTMAP-Vi cell is due to high
membrane resistance. The polarization area specific resistance of the entire cell at
2
5
0% SOC is 3.17 U cm , while high-frequency area specific resistance is 2.20 U
2
cm (Figure S8). The high contribution from the membrane is expected to decrease
as highly anion-conductive membranes become available.
The TMAP-TEMPO/BTMAP-Vi cell was initially cycled at constant current densities of
À2
1
0, 20, 40, and 60 mA cm (Figure 2E). For each current density, 20 consecutive
galvanostatic cycles were performed and we observed almost 100% Coulombic
efficiency during 20 consecutive cycles. A capacity utilization of 97% and a round-
À2
trip energy efficiency of 93.41% were reached at the current density of 10 mA cm
.
Prolonged galvanostatic cycling was performed to probe the cycling lifetime of the
cell and the stability of the posolyte. Both the TMAP-TEMPO/BTMAP-Vi cell and
À2
the reference cell were cycled at 40 mA cm for 1,000 cycles (Figures 2B and 2F).
The TMAP-TEMPO/BTMAP-Vi cell displayed a Coulombic efficiency of >99.73%
and a significantly long cycling lifetime. The cell retained 94% of the initial capacity
after 1,000 cycles, projecting a capacity retention rate of 99.993% per cycle or a tem-
poral capacity retention rate of 99.974% per h (Figure 2F) or a loss rate of roughly
0
1
0
.026% per h. In contrast, the reference cell lost 41% of its original capacity over
,000 cycles (Figures 2B and S9–S12), projecting a temporal capacity loss rate of
.22% per h, which is almost one order of magnitude higher than that of the
TMAP-TEMPO/BTMAP-Vi cell.
The extended cycling lifetime of a flow battery could be the result of less membrane
crossover of electroactive species or higher stability of the electroactive molecules.
To understand the possible contribution from crossover, the permeabilities of
4
-OH-TEMPO, TMAP-TEMPO, and BTMAP-Vi across a Selemionꢀ AMV membrane
were measured using a two-compartment diffusion cell (for details see the
Supplemental Information; Figures S13–S15) and were determined to be 1.34 3
À9
2
À1
À10
2
À1
À11
2
À1
1
0
cm s , 6.40 3 10
cm s , and 5.20 3 10
cm s , respectively (Table 1).
The lowered permeability for TMAP-TEMPO is consistent with its larger molecular
radius and positive charge. The capacity-limiting TEMPO species show higher
membrane permeabilities than that of the BTMAP-Vi. A CV study of the electrolyte
after 1,000 consecutive cycles confirms the crossover of TMAP-TEMPO to the nego-
lyte solution and BTMAP-Vi to the posolyte solution (Figure S16), even though the
membrane crossover rate of either is very low.
Based on these measurements, we estimate that crossover contributed 39.6% of
capacity lost in the TMAP-TEMPO cell, and only 9.4% of the capacity lost in the
4
Chem 5, 1–10, July 11, 2019

Please cite this article in press as: Liu et al., A Long Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical, Chem (2019),
https://doi.org/10.1016/j.chempr.2019.04.021
Figure 2. Performance of a Neutral-pH AORFB Assembled with Either 4-OH-TEMPO (0.1 M, A–B) or the Chloride Salt of TMAP-TEMPO (0.1 M, C–F) in
À
Posolyte (10 mL) and BTMAP-Vi Tetrachloride (0.1 M) in Negolyte (15 mL, 1.5 Times e Excess)
(
(
A) Structures of 4-OH-TEMPO and BTMAP-Vi.
B) Galvanostatic cycling of the reference 4-OH-TEMPO/BTMAP-Vi cell at 40 mA cm for 1,000 consecutive cycles. Charge-discharge capacity, round-
À2
trip energy efficiency, and Coulombic efficiency (EE and CE, respectively) were plotted as functions of the cycle number. The entire 1,000 cycles
occurred over a period of 191 h.
(
(
(
(
C) Structures of TMAP-TEMPO and BTMAP-Vi.
D) Polarization curves of the TMAP-TEMPO/BTMAP-Vi cell at varied SOC.
À2
E) Charge-discharge capacity, CE, and EE of the cell when galvanostatically cycled at current densities of 10, 20, 40, and 60 mA cm , respectively.
À2
F) Galvanostatic cycling of the TMAP-TEMPO/BTMAP-Vi cell at 40 mA cm for 1,000 consecutive cycles. Charge-discharge capacity, EE and CE were
plotted as functions of the cycle number. The entire 1,000 cycles occurred over a period of 220 h. (Insets in (B) and (F): the normalized discharge capacity
of the cell at the 1st, the 100th, the 200th, the 500th, and the 1,000th cycle, respectively.)
Chem 5, 1–10, July 11, 2019
5

Please cite this article in press as: Liu et al., A Long Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical, Chem (2019),
https://doi.org/10.1016/j.chempr.2019.04.021
reference cell because the capacity loss of 4-OH-TEMPO is so obvious. Hence, the
slower crossover of TMAP-TEMPO accounts for merely a small fraction of the
ꢀ
103 slower capacity fade rate, namely 5.12% of the total savings of capacity loss
see the Supplemental Information, ‘‘Calculation of capacity loss due to electrolyte
crossover’’). This indicates that TMAP-TEMPO is structurally more stable than
-OH-TEMPO.
(
4
The rest of the capacity loss is attributed to the chemical decomposition of
TMAP-TEMPO. We hypothesize that the ring-opening side reactions on the ni-
troxide radicals of TMAP-TEMPO during galvanostatic charging or discharging
2
2
is responsible for the loss of redox-active material. The attack of water on
the oxidized nitroxide radicals would open the ring but would not change the
aliphatic hydrogens; this is consistent with the absence of new proton peaks
1
in the HNMR spectrum of the TMAP-TEMPO after cycling for 1,000 consecutive
cycles (Figure S17).
To further investigate the source of capacity loss and eliminate the effect of cross-
over, we employed the unbalanced compositionally symmetric-cell cycling method
1
4
by putting the same 50% SOC TMAP-TEMPO on both sides. Over 25 days, 1,850
cycles, the cell lost roughly 14% of its initial capacity, equating to roughly 0.023%
per h (Figures S18–S20), which is in reasonable agreement with the fade rate
observed in the previously mentioned full-cell data.
Permeability analysis and the symmetric-cell results combine to establish the low
fade rate of TMAP-TEMPO. Encouraged by the high capacity retention rate of
TMAP-TEMPO/BTMAP-Vi at 0.1 M and the high water solubility of the electrolytes,
we constructed a cell with a higher active concentration of 0.5 M for both sides. The
concentrated TMAP-TEMPO/BTMAP-Vi cell showed a peak galvanic power density
À2
of 134 mW cm (Figure 3A, EIS of the cell is presented in Figure S21) and exhibited
Coulombic efficiencies of >99.9% at all current densities (Figure 3B). The cell was
À2
cycled continuously for 200 cycles at 100 mA cm (Figures 3C and S22). After
2
00 cycles, the cell retained 95% of its original capacity, projecting a temporal ca-
pacity retention rate of 99.973% per h or a temporal capacity loss rate of 0.027%
per h. The temporal capacity loss rate is nearly the same as that obtained from the
TMAP-TEMPO/BTMAP-Vi cell at 0.1 M concentration, indicating a concentration-in-
dependent temporal capacity loss rate, which is in contrast to the results obtained
1
1,15,18
for other radical posolytes.
Even at a higher electrolyte concentration of
1
.5 M, the cell manifested a capacity retention rate of 99.985% per cycle, projecting
a temporal capacity retention rate of 99.977% per h or a temporal capacity loss rate
of 0.023% per h (Figure 4).
TMAP-TEMPO has a water solubility of 4.62 M (Supplemental Experimental Proced-
À1
ures), which corresponds to a theoretical posolyte capacity of 120 Ah L . We expect
the solubility of the oxidized form to be even higher because when oxidized, the
molecule will bear one more positive charge and the molecular weight or motif
remains unchanged. The solution has a viscosity of ꢀ580 mPa$s at 4.5 M (a homoge-
neous solution as shown in Figure S23), which is higher than that of a polymeric
1
2
electrolyte and would compromise the overall efficiency of the system because
of pumping loss. The viscosity at 0.5 M is only 3.58 mPa$s (Figure S21B).
When taken together, the long-term galvanostatic cycling results, symmetric-
cell cycling results, and permeability measurements suggest that TMAP-TEMPO is
exceptionally stable. This stability might be due to any of several possible causes,
6
Chem 5, 1–10, July 11, 2019

Please cite this article in press as: Liu et al., A Long Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical, Chem (2019),
https://doi.org/10.1016/j.chempr.2019.04.021
Figure 3. Performance of Concentrated TMAP-TEMPO/BTMAP-Vi Cell Assembled with
TMAP-TEMPO Chloride (0.5 M) in Posolyte (10 mL) and BTMAP-Vi Tetrachloride (0.5 M) in
Negolyte (15 mL)
(
(
A) Polarizations of the cell at varied SOC.
B) Charge-discharge capacity, CE and EE of the cell when galvanostatically cycled at current
À2
densities of 100, 80, 60, and 40 mA cm , respectively.
À2
(
C) Galvanostatic cycling of the cell at 100 mA cm for 200 consecutive cycles. Charge-discharge
capacity, EE, and CE are plotted as functions of the cycle number. Inset: the normalized discharge
capacity of the cell at the 1st, the 50th, the 100th, the 150th, and the 200th cycle, respectively. Total
discharge capacity of the cell on the first discharge is 446.1 C, which is 93% of the theoretical value.
The entire 200 cycles occurred over a period of 80.6 h.
namely steric hindrance, Coulombic repulsion, and electronic isolation of the
charged functional groups. The plausible mechanisms are provided as follows.
The four peripheral methyl groups provide steric hindrance, which could suppress
possible interactions between the free radicals. Additionally, the Coulombic repul-
sion between the positively charged TMAP-TEMPO molecules might also reduce
1
5
bimolecular or multi-molecular interaction. Prior work with 4-OH-TEMPO and
1
8
Me
11
TEMPO-4-ammonium (TEMPTMA, also known as N -TEMPO ) and TEMPO-4-
2
3
sulfate exhibited lifetimes that decreased with increasing concentration, whereas
in our system, concentration-independent lifetime with TMAP-TEMPO was
observed. This suggests that the TMAP-TEMPO has suppressed bimolecular or
multi-molecular annihilation and may follow a decomposition mechanism that is
only first order on TMAP-TEMPO concentration. Finally, attaching a charge directly
to the 6-membered ring might compromise the stability of the molecules because
the electronic effect, especially a strong one, could make the joint weak and suscep-
tible to ring-opening side reactions. This is more consistent with the longer lifetime
1
8
of TMAP-TEMPO than with that of TEMPTMA, to which the positively charged
2
3
ammonium is directly attached, or with that of TEMPO-4-sulfate, to which the
negatively charge sulfate is directly attached. The use of aliphatic spacers may miti-
gate ring-opening side reactions caused by the electronic effect of the charged
functionalities.
Chem 5, 1–10, July 11, 2019
7

Please cite this article in press as: Liu et al., A Long Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical, Chem (2019),
https://doi.org/10.1016/j.chempr.2019.04.021
Figure 4. Performance of High Concentration TMAP-TEMPO/BTMAP-Vi Cell Assembled with
TMAP-TEMPO Chloride (1.5 M) in Posolyte and BTMAP-Vi Tetrachloride (1.5 M) in Negolyte.
(
(
A) Polarizations of the cell at varied SOC.
B) Representative cell voltage and charge-discharge current density versus time during cycling
À2
at 100 mA cm
.
À2
(
C) Galvanostatic cycling of the cell at 100 mA cm for 250 consecutive cycles. Charge-discharge
capacity, EE, and CE are plotted as functions of the cycle number. Inset: the normalized discharge
capacity of the cell at the 1st, the 50th, the 100th, the 150th, the 200th, and the 250th cycle,
respectively. Total discharge capacity of the cell on the first discharge is 625.0 C. The entire
2
50 cycles occurred over a period of 171.7 h.
TMAP-TEMPO/BTMAP-Vi is among the most stable all-organic AORFB chemistry to
date, but we note that there are longer-lifetime AORFBs at high pH in which the pos-
2
1
olyte utilizes the inorganic redox couple ferricyanide-ferrocyanide, and at neutral
1
7
pH in which the posolyte comprises the organometallic redox couple FcNCl, and
1
6
an even longer-lifetime functionalization, BTMAP-Fc. A critical question is whether
there is a fundamental reason that voltage trades off against lifetime, or whether it is
possible to enhance both. This question is apparent if one examines OCVs and life-
times among ferrocene-based batteries and does the same for TEMPO-based ones.
The FcNCl/MV cell (OCV 1.06 V) has a temporal capacity fade rate of 0.078% per h,
while the low-OCV version, BTMAP-Fc/BTMAP-Vi (OCV 0.75 V) shows greatly
enhanced lifetime with a temporal capacity fade rate of 0.0042% per h. Similarly,
TMAP-TEMPO/BTMAP-Vi (OCV 1.19 V) has a temporal capacity fade rate that is
Me
half that of TEMPTMA/MV (OCV 1.42 V) and is less than one-fourth that of N
TEMPO/[(NPr) TTz]Cl
fade mechanisms will likely be necessary before this question can be answered.
-
2
4
(OCV 1.44 V) (Table S2). Better understanding of capacity
In summary, we have synthesized an extremely stable redox-active radical organic
for an AORFB posolyte, TMAP-TEMPO, which has >4.5 M solubility in water. An
all-organic AORFB based on TMAP-TEMPO and BTMAP-Vi exhibits an OCV of
1
.1 V and exhibits exceptionally long lifetime. It features a concentration-indepen-
dent temporal capacity retention rate of >99.974% per h, or a capacity retention
rate of 99.993% per cycle over 1,000 consecutive cycles. Symmetric-cell testing
8
Chem 5, 1–10, July 11, 2019

Please cite this article in press as: Liu et al., A Long Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical, Chem (2019),
https://doi.org/10.1016/j.chempr.2019.04.021
established the low fade rate of TMAP-TEMPO. We anticipate that by improving the
ion selective membrane, the entire cell resistance might be greatly reduced, thereby
increasing the power density, and the membrane crossover rate of electroactive spe-
cies might be reduced, thereby further increasing the capacity retention rate. Our
results constitute an important step toward massive-scale intermittent renewable
energy penetration into the future electric grid.
SUPPLEMENTAL INFORMATION
Supplemental Information can be found online at https://doi.org/10.1016/j.chempr.
2
019.04.021.
ACKNOWLEDGMENTS
Financial support received from the National Science Foundation of China (Nos.
9
1534203, 21720102003, 21878281, 21506201) and the International Partnership
Program of Chinese Academy of Sciences (No. 21134ky5b20170010) is gratefully
acknowledged. Research at Harvard was supported by the U.S. DOE award
DE-AC05-76RL01830 through PNNL subcontract 304500, the Massachusetts Clean
Energy Technology Center, and the Harvard School of Engineering and Applied
Sciences.
AUTHOR CONTRIBUTIONS
Z.Y., M.J A., and T.X. designed the project. Y.L., Y.J., L.T., and Z.Y. synthesized the
compounds. Y.L. and M.-A.G. collected the experimental data. Z.Y., Y.L., M.-A.G.,
L.T., R.G.G., M.J.A., L.W., Y.L., and T.X. analyzed the experimental results and
helped with discussions. Z.Y., Y.L., M.-A.G., T.X., and M.J.A. wrote the paper, and
all authors contributed to revising the paper.
DECLARATION OF INTERESTS
The authors declare no competing interests.
Received: November 20, 2018
Revised: March 27, 2019
Accepted: April 23, 2019
Published: May 13, 2019
REFERENCES AND NOTES
1
. Armand, M., and Tarascon, J.M. (2008).
Building better batteries. Nature 451, 652–657.
management in formamidinium-lead-halide–
based perovskite layers for efficient solar cells.
Science 356, 1376–1379.
active materials. Angew. Chem. Int. Ed. 56,
686–711.
2. Obama, B. (2017). The irreversible
10. Noack, J., Roznyatovskaya, N., Herr, T., and
Fischer, P. (2015). The chemistry of redox-flow
batteries. Angew. Chem. Int. Ed. 54, 9776–
momentum of clean energy. Science 355,
6
. Bush, K.A., Palmstrom, A.F., Yu, Z.J., Boccard,
M., Cheacharoen, R., Mailoa, J.P., McMeekin,
D.P., Hoye, R.L.Z., Bailie, C.D., Leijtens, T., et al.
126–129.
9809.
3
4
5
. Arora, N., Dar, M.I., Hinderhofer, A., Pellet, N.,
Schreiber, F., Zakeeruddin, S.M., and Gr a¨ tzel,
M. (2017). Perovskite solar cells with CuSCN
hole extraction layers yield stabilized
(2017). .23.6%-efficient monolithic perovskite/
11. Luo, J., Hu, B., Debruler, C., and Liu, T.L. (2018).
A pi-conjugation extended viologen as a two-
electron storage anolyte for total organic
aqueous redox flow batteries. Angew. Chem.
Int. Ed. 57, 231–235.
silicon tandem solar cells with improved
stability. Nat. Energy 2, 17009.
efficiencies greater than 20%. Science 358,
7
8
. Wang, W., Luo, Q.T., Li, B., Wei, X.L., Li, L.Y.,
and Yang, Z.G. (2013). Recent progress in
redox flow battery research and development.
Adv. Funct. Mater 23, 970–986.
768–771.
. Hou, Y., Du, X., Scheiner, S., McMeekin, D.P.,
Wang, Z., Li, N., Killian, M.S., Chen, H., Richter,
M., Levchuk, I., et al. (2017). A generic
interface to reduce the efficiency-stability-
cost gap of perovskite solar cells. Science 358,
12. Janoschka, T., Martin, N., Martin, U., Friebe, C.,
Morgenstern, S., Hiller, H., Hager, M.D., and
Schubert, U.S. (2015). An aqueous, polymer-
based redox-flow battery using non-corrosive,
safe, and low-cost materials. Nature 527,
78–81.
. Soloveichik, G.L. (2015). Flow batteries: current
status and trends. Chem. Rev. 115, 11533–
11558.
1
192–1197.
. Yang, W.S., Park, B.W., Jung, E.H., Jeon, N.J.,
Kim, Y.C., Lee, D.U., Shin, S.S., Seo, J., Kim,
E.K., Noh, J.H., et al. (2017). Iodide
9. Winsberg, J., Hagemann, T., Janoschka, T.,
Hager, M.D., and Schubert, U.S. (2017). Redox-
flow batteries: from metals to organic redox-
13. Lin, K., Chen, Q., Gerhardt, M.R., Tong, L., Kim,
S.B., Eisenach, L., Valle, A.W., Hardee, D.,
Gordon, R.G., Aziz, M.J., et al. (2015).
Chem 5, 1–10, July 11, 2019
9

Please cite this article in press as: Liu et al., A Long Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical, Chem (2019),
https://doi.org/10.1016/j.chempr.2019.04.021
Alkaline quinone flow battery. Science 349,
529–1532.
flow battery (AORFB) toward sustainable and
safe energy storage. J. Am. Chem. Soc. 139,
near-neutral pH with record capacity retention
rate. Adv. Energy Mater. 9.
1
1
207–1214.
1
4. Goulet, M.A., and Aziz, M.J. (2018). Flow
battery molecular reactant stability determined
by symmetric cell cycling methods.
2
2
2. Ma, Y., Loyns, C., Price, P., and Chechik, V.
(2011). Thermal decay of TEMPO in acidic
media via an N-oxoammonium salt
intermediate. Org. Biomol. Chem. 9, 5573–
5578.
1
8. Janoschka, T., Martin, N., Hager, M.D., and
Schubert, U.S. (2016). An aqueous
redox-flow battery with high capacity
and power: the TEMPTMA/MV system.
Angew. Chem. Int. Ed. 55, 14427–14430.
J. Electrochem. Soc. 165, A1466–A1477.
1
5. Liu, T.B., Wei, X.L., Nie, Z.M., Sprenkle, V., and
Wang, W. (2016). A total organic aqueous
redox flow battery employing a low cost
and sustainable methyl viologen anolyte
and 4-HO-TEMPO catholyte. Adv. Energy
Mater 6.
3. Winsberg, J., Stolze, C., Schwenke, A.,
Muench, S., Hager, M.D., and Schubert, U.S.
(2017). Aqueous 2,2,6,6-
1
9. Murali, A., Nirmalchandar, A.,
Krishnamoorthy, S., Hoober-Burkhardt, L.,
Yang, B., Soloveichik, G., Prakash, G.K.S.,
and Narayanan, S.R. (2018). Understanding
and mitigating capacity fade in aqueous
organic redox flow batteries. J. Electrochem.
Soc. 165, A1193–A1203.
Tetramethylpiperidine-N-oxyl catholytes for a
high-capacity and high current density
oxygen-insensitive hybrid-flow battery. ACS
Energy Lett. 2, 411–416.
1
6. Beh, E.S., De Porcellinis, D., Gracia, R.L., Xia,
K.T., Gordon, R.G., and Aziz, M.J. (2017). A
neutral pH aqueous organic–organometallic
redox flow battery with extremely high capacity
retention. ACS Energy Lett. 2, 639–644.
2
1. Kwabi, D.G., Lin, K., Ji, Y., Kerr, E.F., Goulet,
M.-A., DePorcellinis, D., Tabor, D.P., Pollack,
D.A., Aspuru-Guzik, A., Gordon, R.G., et al.
(2018). Alkaline quinone flow battery with long
lifetime at pH 12. Joule 2, 1894–1906.
20. Ji, Y., Goulet, M.A., Pollack, D.A., Kwabi, D.G.,
Jin, S., Porcellinis, D., Kerr, E.F., Gordon, R.G.,
and Aziz, M.J. (2019). A phosphonate-
1
7. Hu, B., DeBruler, C., Rhodes, Z., and Liu, T.L.
(2017). Long-cycling aqueous organic redox
functionalized quinone redox flow battery at
10
Chem 5, 1–10, July 11, 2019