Metal-free hybrid seawater fuel cell with an ether-based electrolyte

Abstract

In this work, the design of a new metal-free hybrid seawater fuel cell consisting of a flowing seawater cathode and a hard carbon anode was proposed. The electrochemical performance of the cell was investigated with two different electrolytes, i.e., 1 M NaClO4 in ethylene carbonate (EC)/propylene carbonate (PC), and 1 M NaCF3SO3 in tetraethylene glycol dimethyl ether (TEGDME). The TEGDME-based electrolyte showed a good cycle performance for 100 cycles, whereas EC/PC showed poor cycle stability after 30 cycles. Our results showed that a low conducting solid-electrolyte interphase (SEI) was formed with a thick layer, and the PVdF binder was degraded during the redox reaction when the EC/PC-based electrolyte was used. In contrast, the TEGDME-based electrolyte induced the formation of a more efficient SEI layer without degradation of the binder.

Full text

Journal of
Materials Chemistry A
Materials for energy and sustainability
www.rsc.org/MaterialsA
ISSN 2050-7488
Volume 2 Number 46 14 December 2014 Pages 19569–19922
COMMUNICATION
Jae-Kwang Kim, Youngsik Kim et al.
Metal-free hybrid seawater fuel cell with an ether-based electrolyte

Metal-free hybrid seawater fuel cell with an ether-
based electrolyte†
Hyojin Kim, Jeong-Sun Park, Sun Hye Sahgong, Sangmin Park, Jae-Kwang Kim*
and Youngsik Kim*
In this work, the design of a new metal-free hybrid seawater fuel cell
consisting of a flowing seawater cathode and a hard carbon anode
was proposed. The electrochemical performance of the cell was
investigated with two different electrolytes, i.e.,1MNaClO
4
in
ethylene carbonate (EC)/propylene carbonate (PC), and 1 M NaCF
3
SO
3
in tetraethylene glycol dimethyl ether (TEGDME). The TEGDME-based
electrolyte showed a good cycleperformance for 100 cycles, whereas
EC/PC showed poor cycle stability after 30 cycles. Our results showed
that a low conducting solid-electrolyte interphase (SEI) was formed
with a thick layer, and the PVdF binder was degraded during the redox
reaction when the EC/PC-based electrolyte was used. In contrast, the
TEGDME-based electrolyte induced the formation of a more efficient
SEI layer without degradation of the binder.
Lithium–air batteries have received much attention owing to
their high-energy density, which is 10 times higher than that of
lithium-ion batteries.
1
However, the costs of lithium raw
materials have roughly doubled since the rst applications of
the lithium-ion batteries. The costs may continue to increase
through large-scale commercialization of lithium-battery
storage systems because of their increasing demand and
limited supply. In contrast, sodium resources are virtually
inexhaustible and easily accessible. Moreover, sodium is placed
below lithium in the periodic table, i.e., the two metals have a
similar electrochemical equivalent and standard potential. In
this respect, sodium–air batteries have the potential to meet
large-scale energy storage needs. In sodium–air batteries,
sodium is typically used in its metal state as the anode; for this
purpose, sodium is generally extracted from seawater.
2,3
Although the use of sodium–metal anodes ensures high-energy
densities, it may cause safety issues; in addition, costs may rise
as additional processing steps may be required for sodium
extraction.
To address these issues, we have designed a novel hybrid
seawater fuel cell, which makes direct use of seawater as the
cathode material.
4,5
The cell proposed in this work is unique in
that the cathode-active material, e.g., seawater, is not stored in
the battery. During the battery-charge process, the Na
+
ions
generated from NaCl dissolved in owing seawater transfer to
the negative electrode, while gaseous Cl
2
is released. During
discharge, the Na
+
ions are transferred back to seawater and the
oxygen dissolved in seawater is reduced, resulting in the
formation of NaOH in the presence of water and sodium ions.
The charge–discharge reactions for the hybrid seawater fuel cell
can be written as
Charge: 4NaCl /4Na
+
+ 2Cl
2
+4e

,E¼4.07 V vs. Na
Discharge: 4Na
+
+2H
2
O+O
2
+4e

/
4NaOH, E¼3.11 V vs. Na
Notably, the use of hard carbon as the anode makes our
hybrid seawater fuel cell a metal-free system; moreover, the
electrochemical properties mainly depend on the characteris-
tics of the components (e.g., electrolyte and solid membrane).
Several studies have shown the important role played by the
nature of the electrolyte in enhancing the electrochemical
properties of sodium batteries;
6–10
in particular, ethylene
carbonate (EC)/propylene carbonate (PC) solvent formulation
showed good results in alloy compound-based sodium batteries
with both NaClO
4
and NaPF
6
salts.
9,11
Similarly, tetraethylene
glycol dimethyl ether (TEGDME) showed good electrochemical
properties with the NaCF
3
SO
3
salt in sodium–metal
batteries.
12,13
This suggests that these systems may be used as
electrolytes for sodium ion-based batteries.
In this study, 1 M NaClO
4
in EC/PC and 1 M NaCF
3
SO
3
in
TEGDME were chosen as electrolytes, and their cycle perfor-
mance evaluated in our metal-free hybrid seawater fuel cell with
School of Energy & Chemical Engineering, Ulsan National Institute of Science and
Technology (UNIST), Ulsan 689-798, Republic of Korea. E-mail: jaekwang@unist.ac.
kr; ykim@unist.ac.kr
†Electronic supplementary information (ESI) available: Schematic illustration of
the seawater fuel cell, ionic conductivity and XRD of the solid electrolyte,
charge–discharge curves and cycle performance, and EIS. See DOI:
10.1039/c4ta04937c
Cite this: J. Mater. Chem. A,2014,2,
19584
Received 19th September 2014
Accepted 7th October 2014
DOI: 10.1039/c4ta04937c
www.rsc.org/MaterialsA
19584 |J. Mater. Chem. A,2014,2,19584–19588 This journal is © The Royal Society of Chemistry 2014
Journal of
Materials Chemistry A
COMMUNICATION

an optimised cell construction based on the NASICON (Na
3
-
Zr
2
Si
2
PO
12
) solid electrolyte to ensure compatibility. Notably,
this is the rst study that explores the inuence of the electro-
lyte on the performance of a metal-free hybrid seawater fuel cell,
showing an improvement of the cycle life with the use of
TEGDME.
The conguration of the metal-free hybrid seawater fuel cell,
based on a hard carbon/liquid electrolyte/solid electrolyte/
owing seawater system, is shown in Fig. 1 (S1†).
The NASICON solid electrolyte at the interface between the
liquid electrolyte and owing seawater showed a monoclinic
structure (C2/cspace group) and an ionic conductivity of 7.0 
10
4
Scm
1
at room temperature (Fig. S2†). The practical
potential of the oxidation reaction in the Na/seawater cell was
3.6 V; the reduction potential was 2.8 V (Fig. S3†). In addition,
we investigated the charge–discharge performance of the hard
carbon anode in the two electrolytes (Fig. S4†). The potentio-
gram sloped from 1.18 to 0.1 V during the initial reduction,
followed by a long at region between 0.1 and 0 V, reaching
290 mA h g
1
and 340 mA h g
1
for EC/PC and TEGDME,
respectively. During the following oxidation, a capacity of 120–
170 mA h g
1
was observed close to 0 V vs. Na/Na
+
; then, the
potential gradually increased to 1.2 V, suggesting that the hard
carbon anode underwent a reversible sodiation; its irreversible
capacity in EC/PC and TEGDME was determined to be 47 mA h
g
1
and 37 mA h g
1
, respectively (Fig. S4†).
In addition, the EC/PC and TEGDME-based electrolytes were
evaluated in the metal-free hybrid seawater fuel cell in terms of
charge–discharge performances at room temperature; typical
at curves at 3.46 and 2.26 V, and at 3.55 and 2.47 V, for EC/PC-
and TEGDME-based electrolytes, respectively, were obtained for
the charge and discharge reactions with a voltage separation
(DV) of 1.2 V and 1.08 V, respectively (Fig. 2a). This indicates
that the cell resistance of the latter is lower, and it performs
better than the former, considering the higher-charge delivery
and discharge capacities tested under the same current densi-
ties. The initial charge and discharge capacities at 0.2 C
(0.05 mA cm
2
) were determined to be 174 and 115 mA h g
1
,
respectively, for 1 M NaClO
4
in EC/PC; 176 and 126 mA h g
1
,
respectively, for 1 M NaCF
3
SO
3
in TEGDME. TEGDME showed a
higher active-material utilization rate during discharge, and its
irreversible capacity was lower than that of EC/PC. Both elec-
trolytes showed a stable cycle performance up to 30 cycles
(Fig. 2b). However, aer 30 cycles, the EC/PC cycle performance
decreased rapidly to 80 cycles, whereas TEGDME was still stable
aer 100 cycles. A comparison of the cycle properties also
showed a better retention of the initial property by TEGDME
compared with that of EC/PC. Aer 100 cycles, TEGDME
retained 90% of the rst cycle discharge capacity at 0.1 C-rate (to
compare with 39% for EC/PC). TEGDME showed instability
during 3 cycles. The gradual increase of discharge capacity may
be due to a slow penetration of the electrolyte, which leads to a
higher viscosity in the hard carbon anode. Thus, the evaluation
of the electrochemical properties showed a higher performance
of the cell when the TEGDME-based electrolyte was used. These
results are extremely encouraging, considering that the
seawater cathode combined with NaCF
3
SO
3
of the TEGDME-
based electrolyte performed well even at 25 C.
Na
2
CO
3
, sodium alkyl carbonate (NaOCO
2
R), and polymers,
such as polycarbonate and polyethylenoxide, are some of the
discharge compounds to form a solid electrolyte interphase
(SEI) that have been reported for sodium-ion batteries
Fig. 1 Schematic illustration of the designed metal-free seawater fuel
cell with a diagram of the charge–discharged states.
Fig. 2 (a) Charge–discharge curves and (b) cycling performance of
the metal-free seawater fuel cells with the two different electrolytes
(1 M NaClO
4
in EC/PC and 1 M NaCF
3
SO
3
in TEGDME, room
temperature, and 0.05 mA cm
2
).
This journal is © The Royal Society of Chemistry 2014 J. Mater. Chem. A,2014,2,19584–19588 | 19585
Communication Journal of Materials Chemistry A

containing alkyl carbonate-based electrolytes.
8,14
TEGDME also
forms a SEI layer with the decomposition products of AOCH
2
R
and A
2
CO
3
(A ¼Li or Na).
15
X-ray photoelectron spectroscopy
(XPS) measurements were carried out to investigate the surface
layers formed by the electrochemical reactions of the two elec-
trolytes (Fig. 3). The strong peak at 284.5 eV in the pristine hard
carbon anode spectra was assigned to the sp
2
carbon in the C–C
bond of graphene of the hard carbon anode. The intensity of the
sp
2
carbon peak was signicantly reduced upon sodiation due
to the formation of the SEI on the hard carbon anode surface; as
a result, its position was shied towards a lower binding energy
because of the low electronegativity of sodium. Other
peaks originated from the poly(vinylidene uoride) (PVdF)
binder (CF
2
at 290.5 eV) and the functional groups on the hard
carbon electrodes.
16
Aer 50 cycles, the sp
2
carbon peak at 284.5
eV almost disappeared, indicating that the hard carbon anode
in both cells is covered by the decomposition products of the
electrolyte; this result is conrmed by SEM (Fig. S5†). The peaks
at 290.0 eV were assigned to compounds such as Na
2
CO
3
and
sodium alkyl carbonates; peaks at 288.5, 286.6, and 284.9 eV
were assigned to O]C–O, the ester linkage CO, and CH
2
,
respectively. The difference between the two electrolytes
becomes evident at 284.9 eV and 290.0 eV. In particular, the
peak intensities associated with CH
2
and CO
3
in EC/PC are
stronger than those in the TEGDME electrolyte. Since CH
2
originated from the alkyl/alkylene groups and polymer species,
a larger amount of hydrocarbon compounds was found in the
surface lm formed by the EC/PC-based electrolyte. In addition,
the CO
3
intensity originated from alkali and alkyl carbonates of
SEI was higher in EC/PC than in TEGDME, suggesting a thicker
SEI layer.
The thick SEI layer in the EC/PC electrolyte showed a
higher interface resistance than that in TEGDME (Fig. S6†). In
addition, because some electrolytes may induce the decom-
position of the PVdF binder,
17
the intensity of the CF
2
peak
was monitored. Our results showed that the intensity of the
CF
2
peak in EC/PC decreased, in contrast to that in TEGDME.
The decomposition of the binder was conrmed by F 1s
spectra (Fig. 4). The pristine hard carbon anode showed a
single peak, which was assigned to the C–FbondofthePVdF
binder. The NaCF
3
SO
3
salt led to a small shiof the C–Fpeak
to a higher binding energy in TEGDME. The cycled hard
carbon showed two peaks, e.g., one with a higher binding
energy (688.0 eV, which originated from C–F of the binder)
and one with a lower binding energy (684.1 eV, which orig-
inated from NaF formed during the redox reaction in the
SEI).
15
Since there is no uorine source in the EC/PC-based
electrolyte, the lower-binding energy peak indicates the
decomposition of the binder during the electrochemical
reaction, suggesting that uorine partially reacted with
sodium when using the EC/PC electrolyte. In addition, the
binder decomposition might cause an electrochemical reac-
tion of the hard carbon anode with insulating electron
transfer.
In contrast, the degradation of the binder was not observed
in the TEGDME-based electrolyte; NaCF
3
SO
3
partially decom-
posed to form NaF, which caused the C–F peak to shitowards
a lower binding energy.
12
The F 1s spectra of the hard carbon
anode cycled in 1 M NaClO
4
/TEGDME clearly suggests that NaF
originated from the NaCF
3
SO
3
salt (Fig. S7†). Also, some NaF
SEI compound improved the cycle stability in the metal-free
Fig. 3 XPS C 1s spectra of the surface of the hard carbon anode cycled
in 1 M NaClO
4
in EC/PC and 1 M NaCF
3
SO
3
in TEGDME (after
50 cycles).
19586 |J. Mater. Chem. A,2014,2,19584–19588 This journal is © The Royal Society of Chemistry 2014
Journal of Materials Chemistry A Communication

seawater fuel cell, comparing with the non-NaF formation in
1 M NaClO
4
/TEGDME. Therefore, we concluded that the high
cycle performance of the metal-free hybrid seawater fuel cell
with 1 M NaCF
3
SO
3
in TEGDME is due to an efficient SEI layer
and stability of the binder.
Conclusions
In summary, in this work we designed a metal-free hybrid
seawater fuel cell based on a hard carbon/liquid electrolyte/
solid electrolyte/owing seawater system. In order to maximise
the cycle stability of the cell, two different electrolytes were
employed, namely 1 M NaClO
4
in EC/PC and 1 M NaCF
3
SO
3
in
TEGDME; these have recently been shown to improve sodium
batteries. In contrast to the EC/PC-based system, the metal-free
hybrid seawater fuel cell cycled using the TEGDME-based elec-
trolyte showed a stable cycle performance (even aer
100 cycles). XPS and EIS clearly showed that the formation of
the SEI layer is more efficient with the TEGDME-based electro-
lyte; in addition, the binder decomposition was observed only in
the metal-free hybrid seawater fuel cell with the EC/PC-based
electrolyte. Thus, the results presented in this work prove that
the electrolyte containing 1 M NaCF
3
SO
3
in TEGDME ensures a
stable electrochemical performance of the proposed metal-free
hybrid seawater fuel cell.
Experimental section
Seawater was used as a positive electrode material. The hard
carbon negative electrode was fabricated from a 80 : 10 : 10
(wt%) mixture of hard carbon (MeadWestvaco Co. U.S.A.),
Super-P carbon black (TIMCAL) as the current conductor, and
PVdF (Sigma Aldrich) as the binder. For the hybrid multi-layer
electrolyte, liquid electrolyte and NASICON ceramic plate as the
solid electrolyte were used. The NASICON solid electrolyte was
prepared by a solid-state reaction method: Na
3
PO
4
$12H
2
O,
SiO
2
, and ZrO
2
(obtained from Sigma Aldrich) were mixed and
then calcined at 400 and 1100 C; aer several mixing and
calcination steps, the powder was pressed into a pellet, which
was subsequently sintered at 1230 C.
The carbon paper (Fuel Cell Store, Inc.,) employed as a
current collector was placed in the owing-seawater positive
electrode. The assembled cell was exposed to seawater and
connected to a testing station. A battery cell tester (WBCS3000)
was employed to perform the charge and discharge tests at
different current densities. XPS spectra of the hard carbon
anode were obtained with a PHI 5500 spectrometer, and Al Ka
(1485.6 eV) was used as the X-ray source at an anode voltage of
13.8 kV. The electrolytes on the surface of the hard carbon
anode were removed by sotissue for XPS measurements.
Acknowledgements
This work was supported by the Research Fund (1,140083,01) of
Ulsan National Institute of Science and Technology (UNIST) and
Basic Science Research Program through the National Research
Foundation of Korea (NRF) funded by the Ministry of Education
(NRF-2014R1A1A2A16053515).
Notes and references
1 K. M. Abraham and Z. Jiang, J. Electrochem. Soc., 1996, 143,
1–5.
Fig. 4 XPS F 1s spectra of the surface of the hard carbon anode cycled
in 1 M NaClO
4
in EC/PC and 1 M NaCF
3
SO
3
in TEGDME (after 50
cycles).
This journal is © The Royal Society of Chemistry 2014 J. Mater. Chem. A,2014,2,19584–19588 | 19587
Communication Journal of Materials Chemistry A

2 P. Hartmann, C. L. Bender, M. Vracar, A. K. D¨
urr, A. Garsuch,
J. Janek and P. Adelhelm, Nat. Mater., 2013, 12, 228–232.
3 J. Kim, H. D. Lim, H. Gwon and K. Kang, Phys. Chem. Chem.
Phys., 2013, 15, 3623–3629.
4 J. K. Kim, E. Lee, H. Kim, C. Johnson, J. Cho and Y. Kim,
ChemElectroChem., 2014, submitted.
5 J. K. Kim, F. Mueller, H. Kim, J. S. Park, D. H. Lim, G. T. Kim,
D. Bresser, S. Passerini and Y. Kim, NPG Asia Mater., 2014,
accepted.
6 S. Komaba, W. Murata, T. Ishikawa, N. Yabuuchi, T. Ozeki,
T. Nakayama, A. Ogata, K. Gotoh and K. Fujiwara, Adv.
Funct. Mater., 2011, 21, 3859–3867.
7 X. Xia and J. R. Dahn, J. Electrochem. Soc., 2012, 159, A515–
A519.
8 A. Ponrouch, R. Dedryv`
ere, D. Monti, A. E. Demet,
J. M. A. Mba, L. Croguennec, C. Masquelier, P. Johansson
and M. R. Palacin, Energy Environ. Sci., 2013, 6, 2361–2369.
9 A. Ponrouch, A. R. Go˜
ni and M. R. Palacin, Electrochem.
Commun., 2013, 27,85–88.
10 D. J. Lee, J. W. Park, I. Hasa, Y. K. Sun and B. Scrosati, J.
Mater. Chem. A, 2013, 1, 5256–5261.
11 A. Ponrouch, E. Marchante, M. Courty, J. M. Tarascon and
M. R. Palacin, Energy Environ. Sci., 2012, 5, 8572–8583.
12 P. Hartmann, C. L. Bender, J. Sann, A. K. D¨
urr, M. Jansen,
J. Janek and P. Adelhelm, Phys. Chem. Chem. Phys., 2013,
15, 11661–11672.
13 R. Younesi, M. Hahlin, M. Treskow, J. Scheers, P. Johansson
and K. Edstr¨
om, J. Phys. Chem. C, 2012, 116, 18597–18604.
14 S. Tsubouchi, Y. Domi, T. Doi, M. Ochida, H. Nakagawa,
T. Yamanaka, T. Abe and Z. Ogumi, J. Electrochem. Soc.,
2012, 159, A1786–A1790.
15 S. Xiong, Y. Diao, X. Hong, Y. Chen and K. Xie, J. Electroanal.
Chem., 2014, 719, 122–126.
16 D. Aurbach, B. Markovsky, I. Weissman, E. Levi and Y. Ein-
Eli, Electrochim. Acta, 1999, 45,67–86.
17 M. Manickam, Electrochim. Acta, 2003, 48, 957–963.
19588 |J. Mater. Chem. A,2014,2,19584–19588 This journal is © The Royal Society of Chemistry 2014
Journal of Materials Chemistry A Communication