Energy Technology
Generation, Conversion, Storage, Distribution

Accepted Article
Title: High Performance Aqueous Sodium-Ion Capacitors Enabled by
Pseudocapacitance of Layered MnO2
Authors: Yadi Zhang, Yufeng An, Jiangmin Jiang, Shengyang Dong, Langyuan
Wu, Ruirui Fu, Hui Dou, and Xiaogang Zhang
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication of the
final Version of Record (VoR). This work is currently citable by using the Digital
Object Identifier (DOI) given below. The VoR will be published online in Early
View as soon as possible and may be different to this Accepted Article as a
result of editing. Readers should obtain the VoR from the journal website shown
below when it is published to ensure accuracy of information. The authors are
responsible for the content of this Accepted Article.
To be cited as: Energy Technol. 10.1002/ente.201800157
Link to VoR: http://dx.doi.org/10.1002/ente.201800157

www.entechnol.de

10.1002/ente.201800157

Energy Technology

FULL PAPER
High Performance Aqueous Sodium-Ion Capacitors Enabled by
Pseudocapacitance of Layered MnO2
Abstract: Aqueous sodium-ion capacitors (ASICs) are becoming
increasingly important due to the remarkable advantages of aqueous
electrolyte about the excellent ionic conductivity, non-flammability and low
cost compared with organic systems. But, low capacitance of the electric
double-layer capacitive material and narrow potential window of aqueous
electrolyte both have negative effects on the enhancement of energy density.
Therefore, we employ typical pseudocapacitive material, layered
MnO2/CNTs composite as cathode to fabricate sodium ion capacitor. It needs
to be emphasized that the electrochemical process involves two kinds of
energy storage mechanisms, such as the reversible Na+ adsorption/desorption
onto the surface of each layer and fast Na+ (de)intercalation into the 2D
interlayer space. Thus, the composite delivers a high specific capacitance
(322.5 F g-1 at 0.5 A g-1) and an excellent cycle stability (5000 cycles with
capacitance retention of approximately 90%). By means of the synergistic
effects of the layered MnO2/CNTs cathode, sodium-ion water-in-salt
electrolyte (NaWiSE) and polyimide organic anode, the as-assembled ASIC
achieves a high energy density of 78.5 Wh kg-1, accompanied by high power
density of 11 000 W kg-1 and excellent cycle performance (even 77%
capacity retention after 10 000 cycles).

Introduction
Decreasing fossil fuel and increasing environmental concerns
powerfully stimulate people to explore sustainable and environmental
friendly energy storage devices.[1] Sodium ion capacitors (SICs) as one
kind of developing energy storage devices, are famous for high energy
density ascribed to battery materials and high power density resulted
from supercapacitor materials.[2] Up to now, nearly all of the SICs
consist of battery materials, electric double-layer capacitive (EDLC)
materials and organic electrolytes. In most of these systems, battery
materials that depends on ion (de)intercalation have high capacity, while
EDLC materials store charge fast, and organic electrolytes provide wide
cell voltages. However, it is still a formidable subject to well integrate
the virtues including high energy density, high power density, long
lifespan, low cost and high security into one device. Similar to
supercapacitors and batteries, the energy stored in a SIC is closely
related to its operating voltage window and capacity, which can be
potentially enhanced by simultaneously enlarging cell voltage and/or
improving capacity.[3] On the former, organic electrolytes are preferred
ones to provide large cell voltage, but they often cause safety incidents
and cost issues. Therefore, great efforts have been focused on aqueous
electrolytes due to the intriguing characteristics of high conductivity,
low cost, excellent thermal and chemical stability and safety. Among
them, water-in-salt electrolytes (WiSEs) based on lithium salt have been
popularly explored and introduced into energy storage devices. On

Y. D. Zhang, Y. F. An, J. M. Jiang, S. Y. Dong, L. Y. Wu, R. R. Fu, H. Dou, X. G.
Zhang
Jiangsu Key Laboratory of Electrochemical Energy Storage Technologies
College of Material Science and Engineering
Nanjing University of Aeronautics and Astronautics
Nanjing 210016, P.R. China.
E-mail: azhangxg@163.com

account of the reduced activity of free water molecules and the existence
of over-potential, the stable working voltage can be enlarged to 3.0 V,
and even 4.0 V when the gel electrolyte was employed.[4] Recently, some
researchers have utilized WiSEs in sodium/potassium ion devices.[5] In
particular, Nakamoto et al. reported NaClO4-based WiSE for aqueous
sodium ion batteries, which enables an electrochemical window of 2.8
V allowing the aqueous sodium-ion system to work reversibly.[5b]
Meanwhile, non-lithium and non-fluorinated NaClO4-based WiSE make
it much cheaper and greener than fluorinated-imide-based WiSE or
NaClO4-based organic electrolytes. Given this, NaClO4-based WiSE can
be regarded as alternative electrolyte used in aqueous sodium ion
capacitors (ASICs).
In matters of capacitance of SICs, low capacitances of carbon
materials are stumbling blocks for the enhanced capacitance for SICs.
Relatively, pseudocapacitive materials can deliver larger capacitances
than EDLC materials due to intrinsically different mechanisms of charge
storage, while they are rarely used to fabricate SICs. Thankfully,
increasing interests have been focused on the application of
pseudocapacitive materials for SICs in recent years.[6] For instance,
Wang et al. used layered pseudocapacitive material Ti2CTx to construct
organic SICs.[6b] Interestingly, the interlayer distance of Ti2CTx can be
distinctly expanded by the first Na + intercalation verified by
transmission electron microscopy (TEM) investigation, which
guaranteed barely significant structural change during the following
charge/discharge process. The as-assembled Ti2CTx//Na2Fe2(SO4)3 SICs
showed an ultrahigh energy density of 260 Wh kg-1 at power density of
1.4 kW kg-1 (based on the weight of Ti2CTx) as well as excellent cycle
stability. Beyond that, MnO2 is the most common pseudocapacitive
material by reason of the attractive features of high theoretical specific
capacitance (1370 F g-1), low cost, environmental pollution-free and
simple synthesis.[7] As electrode materials, MnO2 mainly stores charges
in the salt solution through a fast and reversible reaction: MnO2 + xM+
+ xe- = MxMnO2, where M+ represents Li+, Na+, K+ or NH4+, etc. It is
note that the reaction process involves an adsorption/desorption process
of cations at the material surface and/or (de)intercalation process of
cations into the bulk material.[8] However, the low conductivity of MnO2
seriously limits the its specific capacitance (200-300 F g-1). Therefore,
many researchers attempted to improve its conductivity by introducing
carbon materials, the enhanced specific capacitance can exceed 500 F g1
at present.[9] Recently, Zhang and co-workers reported a highperformance rechargeable zinc-manganese dioxide system.[10] In that
system, layered zinc-buserite was converted from tunnel structured
manganese dioxide after first discharging, which allowed for a fast zinc
(de)intercalation and delivered a long cycle span (94% capacity retention
after 2000 cycles) because of no-phase transformation after first
discharging. Instructively, Jabeen et al. designed and assembled an
asymmetric supercapacitor relied on layered Na 0.5MnO2 as positive
electrode, that was fabricated via in-situ conversion from spinel Mn3O4
by the Na+ intercalation after cycling in Na2SO4 electrolyte.[11] The
layered material also expressed an ultralong life span due to reasonable
interlayer spacing for the convenient cation intercalation and no-phase
change in electrochemical process. Similar to the above converted
resultants, birnessite MnO2 with layered structure comprised of edgesharing MnO6 octahedra, has attracted much attention due to the

This article is protected by copyright. All rights reserved.

Accepted Manuscript

Yadi Zhang, Yufeng An, Jiangmin Jiang, Shengyang Dong, Langyuan Wu, Ruirui Fu, Hui Dou, Xiaogang Zhang*

10.1002/ente.201800157

Energy Technology

extraordinary advantages of moderate interlayer distance (d=7.0 Å),
layer charge density and high ion-exchange activity.[12] What’s more, the
crystal water in the interlayer can effectively expand the interlayer
spacing and support the layered structure in the electrochemical process,
benefiting to outstanding cycle performance.[13] However, the
electrochemical performances were often investigated under a low
potential range (0-0.9 V), accompanied by Na+ adsorption/desorption
process without Na+ (de)intercalation process. In order to stimulate the
extra capacitance, wide potential range is employed to promote the Na +
(de)intercalation, but there is not sufficient evidence to confirm that so
far.
Based on the above considerations, we systematically investigate the
electrochemical performance of layered MnO2. As cathode materials,
Na+ could easily intercalate into MnO2 interlayer after first discharging,
which naturally leads to an expanded interlayer spacing. Under a wide
potential range, the energy storage process not only involves Na+
adsorption/desorption on the surface but Na+ (de)intercalation in the
interlayer. Moreover, the co-existence of Na+ and crystal water makes
the layered structure unbroken during a long cycle process. Combining
with organic materials, polyimide with sensitive structure and fast
kinetic process, and NaClO4-based WiSE with high ionic conductivity,
the ASIC exhibits a high energy density of 78.5 Wh kg-1 and an ultralong
lifespan (77% capacity retention after 10 000 cycles).

Results and Discussion
In the synthetic process of MnO2/CNTs composite, KMnO4 served as
oxidant and manganese source, CNTs acted as reductants. The crystal
phase of the prepared hybrid was determined by powder XRD pattern.
In Figure 1a, it can be clearly found that all diffraction peaks can be
well-indexed to birnessite-type MnO2 (JCPDS 42-1317), except the
diffraction peaks dated from CNTs. The peak diffraction peaks of CNTs
are looking like noise and weak because their surfaces are covered with
MnO2 sheets. Most importantly, an evident peak, around 2θ=12º,
represents the basic characterization of the layered structure. [13, 14] To
precisely confirm the composition of the as-prepared composite, TG
analysis was performed. As shown in Figure S1, the mass ratio of crystal

composite shows that the MnO2 petal-like sheets with a thickness of 50
nm perfectly wrap on the overall ektexines of CNTs, and the interface
between MnO2 nanosheets and CNTs can be distinctly identified. The
presence of CNTs can effectively inhibit the agglomeration of MnO 2
nanosheets and provide well-defined electron pathways. The HRTEM
pattern shown in Figure 1b (inset) depicts that an interlayer spacing is
about 0.72 nm, which can be well indexed to the (001) crystal plane in
the XRD pattern. The average interlayer spacing is measured to be ≈0.74
nm that is much larger than the radius of Na+ (1.02 Å) (Figure 1c). In
addition, the specific surface area of MnO2/CNTs was investigated,
accompanied by a large value of 81.15 m2 g-1, which profitably provides
many reaction sites to electrochemical reaction. Most importantly, the
presence of mesoporous facilitates the diffusion of electrolyte ion
(Figure S2b). However, it is known to all that MnO2, as cathode material
with a high valence of +4, is poor electron donor in neutral electrolyte,
thus a pre-activation should be carried out before its electrochemical
investigation. The activation process could be easily realized by the first
discharge process in the three-electrode configuration, and the obtained
hybrid is named activated MnO2/CNTs. To confirm the feasibility of
reduction reaction, physical characterizations for the activated
MnO2/CNTs were also performed. Overall XPS spectra of MnO2/CNTs
and activated MnO2/CNTs are shown in Figure S3. By comparison,
except Mn, O, C elements, Na 1s signal appears but without Cl element
in the latter spectrum, demonstrating that Na element comes from
activated MnO2/CNTs instead of NaClO4 electrolyte. To achieve a better
understanding of “how does Na element exist in the activated
MnO2/CNTs?”, the binding energy gaps of Mn 3s before and after
activation have been analyzed (Figure 1d). The expansion of energy
separation effectively confirms an obvious decrease of the Mn oxidation
state with an average valence of 3.75, which is indicative of a veritable
Na+ intercalation with the content about 0.25. Additionally, EDS
patterns and mappings of MnO2/CNTs electrodes before and after
reduction were carried out. Figure S4a and b both reflect the distribution
of C, O, and Mn in MnO2/CNTs. However, the Na element is identified
in both the EDS pattern and mapping of activated MnO2/CNTs, but
without Cl element. Therefore, the intercalation of Na+ into MnO2 layer
is further proved. After electrochemical reduction, TEM investigation
was operated again. As clearly displayed in Figure 1e, the petal-like
sheets are well kept, while the materials are agglomerated because of the
addition of binder. What is surprising is that the interlayer spacing is

Figure 1 (a) XRD pattern, (b) TEM image (SAED pattern in inset) and (c) the corresponding SAED pattern of MnO 2/CNTs, (d) Mn 3s core-level XPS spectra of
MnO2/CNTs and activated MnO2/CNTs, (e) TEM image (SAED pattern in inset) and (f) the corresponding SAED pattern of activated MnO2/CNTs.

water is about 3.4%, indicating a composition of ~0.12 H2O molecular
in per formula of MnO2/CNTs, which contributes to a stable layered
structure.[8a] As shown in Figure 1b, the TEM image of MnO2/CNTs

exaggeratively expanded from 0.72 to 1.52 nm by electrochemical
activation, as shown in HRTEM pattern (inset in Figure1e). Importantly,
the average interlayer spacing is expanded to be ≈1.49 nm. This

This article is protected by copyright. All rights reserved.

Accepted Manuscript

FULL PAPER

10.1002/ente.201800157

Energy Technology

FULL PAPER
off potential is enlarged to 1.0 V, the CV curve still delivers a rectangular
shape but with larger area, symbolizing a typical capacitive behavior
with the adsorption/desorption of Na+. When the upper cut-off potential
is raised to 1.2 V, there is a pair of weak redox peaks at high potential
position, corresponding to the reversible redox reaction of Mn3+/Mn4+ as
a consequence of Na+ (de)intercalation. In addition, the rectangular
shape can be retained in low potential region. The above results indicate

Figure 2 Schematic diagram for the electrochemical process of activated MnO 2/CNTs electrode.

In order to confirm this idea, the electrochemical performance of
activated MnO2/CNTs electrode was investigated in a three-electrode
configuration. As for the activated MnO2/CNTs electrode, CNTs would
not only efficiently prevent the MnO2 sheets from agglomerating that
benefits the unhindered diffusion of electrolyte ions, but also act as
electron carriers to enhance conductivity. Figure S5a displays the EIS
Nyquist plots of pure MnO2 and activated MnO2/CNTs electrodes. What
can be seen that the charge transfer resistance evidently decreases after
the introduction of CNTs. To enlarge the capacitance, the incorporation
of battery-like behaviour into electrochemical process is an efficient way.
Hence, we attempt to expand the capacitance by adjusting the working
potential window. As shown in Figure S5b, with enlarging potential

that the activated MnO2/CNTs electrode not only has capacitive
behavior, but battery-like behavior in a wide potential window, that is,
the Na+ adsorption/desorption and (de)intercalation can occur together,
which can availably enhance charge capability. Figure 3a shows CV
curves of activated MnO2/CNTs electrode at different scan rates with a
potential window of 0-1.2 V. All curves with scan rates ranging from 1
to 20 mV s-1 retain similar shape without great change, meaning a good
rate capability. Figure 3b presents the charge and discharge curves of
activated MnO2/CNTs electrode at different current densities. The
activated MnO2/CNTs electrode can exhibit a high capacitance of 322.5
F g-1 at a current density of 0.5 A g-1, and retain 210 F g-1 at 10 A g-1. To
verify the charge storage mechanism of the activated MnO2/CNTs

Figure 3 (a) CV curves of activated MnO2/CNTs electrode at different scan rates, (b) charge and discharge curves of activated MnO2/CNTs electrode at
different current densities, (c) CV curve of activated MnO2/CNTs electrode with shadowed area representing the surface capacitive contribution, (d)
separations of diffusion-controlled and capacitive charge at different scan rates for activated MnO2/CNTs electrode and (e) cycle stability of activated
MnO2/CNTs electrode at a current density of 1 A g-1 (inset of SEM image of electrode after cycling).

windows, the CV areas of activated MnO2/CNTs are also increasing,
implying an enlarged capacitance. Specifically, when the potential is 00.8 V, the CV curve displays a rectangular shape. When the upper cut-

electrode mentioned above, current contributions of surface controlled
process and diffusion controlled process are distinguished by using
Dunn’s method:[15]

This article is protected by copyright. All rights reserved.

Accepted Manuscript

difference is another powerful evidence to proof that Na+ successfully
intercalates into the interlayer of MnO2 and is possibly accompanied
with low content of crystal water.
By reason of the large interlayer spacing, we guess that the
charge/discharge process of activated MnO2/CNTs might involve
reversible Na+ adsorption/desorption onto the surface of each layer and
fast Na+ (de)intercalation into the 2D interlayer, the process is
symbolically depicted in Figure 2.

10.1002/ente.201800157

Energy Technology

i(V)  k1v  k2v

(1)

(2)
i (V) / v  k1v  k2
where v is the scan rate. The current response (i) at a fixed potential (V)
can be defined as the combination of two separate mechanisms, namely
capacitive effects (k1v) and diffusion-controlled intercalation (k2v1/2). By
determining both k1 and k2, it is thus possible to distinguish the fraction
of the current arising from Na+ intercalation and that from capacitive
processes at specific potentials.
The surface-controlled capacitance contributions are represented
by the shaded region in the CV curve in Figure 3c and Figure S6. As
seen from Figure S7, at low scan rates, especially 0.1 mV s-1, the
diffusion-controlled contribution accounts for 64.4% of the total charge
storage, which is much higher than the surface capacitive contribution,
highlighting that the charge storage mainly depends on the Na+
(de)intercalation. However, when increasing the scan rate to 20 mV s-1,
the diffusion-controlled contribution is only 18.5% of the total charge
storage, illustrating a dominant role of Na+ adsorption/desorption on the
surface of electrode material (Figure 3d). The cycle performance of
activated MnO2/CNTs electrode is shown in Figure 3e. After cycling 5
000 cycles, the capacitance retention is approximately 90% of the initial
capacitance. Additionally, the SEM image of electrode after cycling is
posted in Figure 3e (inset), it is found that the layered structure can be
well retained even if the layers are slightly expanded. The above results
synthetically illustrate the activated MnO2/CNTs electrode has an
excellent electrochemical performance that may be explained by
following factors. Firstly, the co-existence of Na+ and low content
crystal water in the interlayer of MnO2 that not only benefits the
lightsome Na+ (de)intercalation, but resists the structure destruction
during long cycle. Secondly, the co-existence of Na+
adsorption/desorption and (de)intercalation guarantees the capacitance
of the activated MnO2/CNTs electrode at various current density. Last
but not least, there is no phase transition taken place in the
charge/discharge process, which is different from spinel structure
manganese oxide.
1/ 2

1/ 2

polyimide, FT-IR spectrum was performed. As presented in Figure 4b,
in detail, a peak at 3434 cm-1 demonstrates the presence of -OH. The
peaks at 1704, 1670 and 770 cm-1 are ascribed to the vibrations of the
imide C=O bond. A peak at 1348 cm-1 is attributed to the stretching
vibration of the C-N bond, and a peak at 1581 cm-1 is assigned to
stretching vibration of naphthalene. This analysis result is agreement
with previous reports, confirming the successful synthesis of
PNTCDA.[17] In order to gain more proofs, XPS spectrum further
verifies the presences of C, N and O elements with an atom ration of
about 15.3:3.7:2, largely coinciding with the polymer unit formula
C16O4N2H8 (Figure 4c).
The electrochemical properties of PNTCDA were also simply
detected. Figure 4d displays the CV curves of PNTCDA electrode, in
which there are two obvious redox peaks, corresponding to a two-step
combination of Na+ and C=O bond in PNTCDA and the formation of
sodium enolization, namely C-O-Na. This is in keeping with the
previously reported paper.[18] Meanwhile, a reverse reaction can be taken
place in the opposite process. The charge and discharge curves in Figure
4e give discharge capacities at various current densities. The PNTCDA
electrode can deliver a high capacity of 156.3 mAh g-1 at 0.25 A g-1 as
well as a high Coulombic efficiency approaching to 100%. In addition,
the capacity could reserve 128.5 mAh g-1 at high current density of 5 A
g-1, indicating a good rate capability. Moreover, EIS investigation could
provide in-depth insights of kinetics parameters. As displayed in Figure
4f, according to the intercept of the semicircle at the highest frequency
with the real axis, the detected uncompensated ohmic resistance is much
smaller than that in organic system reported in other papers.[19] The
amazed result mainly benefits from the superiority of aqueous
electrolyte, such as high ion conductivity, low viscosity. Expectantly,
the interface resistance defined by the semicircle in the mediumfrequency region is also very small, declaring a fast kinetics process,
which is ascribed to high ionic activity for high-concentration electrolyte,
as well as an insensitive structure of PNTCDA to Na + radius differing
from inorganic bulk materials.

Figure 4 (a) XRD pattern, (b) FT-IR, (c) overall XPS spectrum of PNTCDA, (d) typical CV curves at different scan rates, (e) charge and discharge curves at
various current densities, and (f) EIS plot of PNTCDA electrode.

Figure 4a shows XRD pattern of as-prepared polyimide, in which
there are two broad and weak diffraction peaks at 2θ=16.8º and 26.38º,
suggesting that this polymer is pure polyimide with an amorphous
structure.[16] In order to further confirm the structure of the as-obtained

On account of the reliable electrochemical performances of the
above materials, novel ASICs were fabricated by employing activated
MnO2/CNTs as cathode, PNTCDA as anode and NaClO4-based WiSE,
respectively. Figure 5a symbolically depicts the energy storage

This article is protected by copyright. All rights reserved.

Accepted Manuscript

FULL PAPER

10.1002/ente.201800157

Energy Technology

FULL PAPER

Figure 6 (a) Regone plot of SIC in this work comparing to representative
energy storage devices in other reports, and (b) cycle performance of
ASIC with a current density of 1 A g-1.

electrochemical energy storage devices including aqueous (aq.) and
organic (or.) systems, such as PTCD//PANI (or.), [20] LiMn2O4//graphene
(aq.),[21] TiO2/graphene//AC (or.),[22] CoHCF//CMS (aq.),[12] NTO//PSC
(or.)[24] and Na0.5MnO2//Fe3O4@C (aq.).[11] Cycle stability is another
parameter to evaluate the feasibility of ASICs. Figure 6b displays the
cycling stability and the corresponding Coulombic efficiency of ASIC.
What's exciting is that the device can undergo 10 000 cycles with no
noticeable capacitance fading after initial 300 cycles with a Coulombic
efficiency about 100%, which can be mainly attributed to two factors.
On one hand, the co-existence of Na+ and crystal water in the layer-bylayer MnO2 makes a large space for Na+ storage without phase transition.
On the other hand, the anode material, PNTCDA, has a sensitive and
stable structure for Na+ de/association, as a consequence of an
outstanding stability.

Conclusions
In summary, layered MnO2/CNTs composite was successfully
synthesized and subsequently reduced by electrochemical method along
with the Na+ intercalation. The large layer-by-layer space not only
facilitates the Na+ adsorption/desorption but also realizes fast Na+
(de)intercalation in a wide potential window of 0-1.2 V, leading to a
large capacitance and good rate capability. Based on the fast and
reversible enolation of anode, large capacitance ascribed to capacitive
and battery-like behavior of cathode and wide electrochemical window
of NaClO4-based WiSE, the as-assembled ASIC delivers a maximum
energy density of 78.5 Wh kg-1 at a power density of 550 W kg-1. In
addition, the device can exhibit an ultralong cycle lifetime
(approximately 77% capacity retention after 10 000 cycles), which is no
less than to the SICs reported previously. Apart from the outstanding
electrochemical performance, low cost and high safety make the ASIC
a promising candidate for the next generation energy storage devices.

Experimental Section
Figure 5 (a) Design and charge-storage mechanism of the present ASIC,
(b) CV curves of the PNTCDA and the activated MnO2/CNTs electrodes in

Synthesis of MnO2/CNTs

separate potential windows, (c) charge and discharge curves of PNTCDA
electrode, the activated MnO2/CNTs electrode and the as-assembled ASIC,
and (d) charge and discharge curves of the as-assembled ASIC at different
current densities.

The Ragone plots is shown in Figure 6a. A high energy density of
78.5 Wh kg-1 can be obtained at a power density of 550 W kg-1, which
are calculated from the total mass of both electrodes. It even retains 30.2
Wh kg-1 when the power density increases to 11.0 kW kg-1, which is
comparable and even better than those of the state-of-the-art reported

The layered MnO2/CNTs composite was synthesized as previously
reported.[13] 100 mg of commercial multi walled CNTs (MWCNTs,
Nanjing XFNANO Materials Tech Co., Ltd, China) and 2.5 g of KMnO4
were mixed and ground for 30 min in an agate mortar. Then, the mixed
powder was excessively ultrasonic dispersed in 100 mL of water, and
0.5 mL of concentrated H2SO4 was added subsequently. After that, the
mixed solution was refluxed in an oil bath at 80 °C for 1 h. The
precipitate was collected by filtrating and washed thoroughly with water.

This article is protected by copyright. All rights reserved.

Accepted Manuscript

mechanism for ASIC. In the charging process, Na + breaks away from
cathode and passes through electrolyte to take part in enolation in the
anode. During discharging process, Na+ goes back to cathode along with
reversible oxidation-reduction reactions in the anode and cathode,
respectively. It is worthwhile mentioning that the reasonable mass ratio
between anode and cathode should be exacted before assembling the
device so that the charge stored in cathode and anode can be well kept
balance. In this work, the mass ratio between anode and cathode is 1:3
(namely, 1 mg:3 mg). Figure 5b shows the individual CV curves for
PNTCDA electrode and the activated MnO2/CNTs electrode. According
to the lowest potential of PNTCDA electrode and the highest potential
of activated MnO2/CNTs electrode, a theoretical cell voltage of 2.2 V is
expected. The CV curves of ASIC at different cell voltages are shown in
Figure S8, when the cell voltage is expanded to 2.3 V, water
decomposition slightly occurs with a small current leap at high potential.
Compared to the charge/discharge curves of PNTCDA electrode and
activated MnO2/CNTs electrode, the charge/discharge curve of ASIC at
a cell voltage of 2.2 V are also symmetry along with a better Coulombic
efficiency, displayed in Figure 5c, which demonstrates a reasonable cell
voltage of 2.2 V, as expected. Figure 5d shows galvanostatic
charge/discharge curves of ASIC at different current densities
investigated at a cell voltage of 2.2 V. All inflection points in the
galvanostatic charge/discharge curves are well correspond to the redox
peaks in CV curves displayed in Figure S8a. EIS measurement of the
ASIC was carried out in the 105-0.1 Hz frequency range to study the
conductibility. As shown in Figure S8b, the impedance curve consists
of a semicircle at high frequencies and a line in the low frequency range.
The EIS data were fitted using equivalent circuit model and displayed in
the inset of Figure S8b. It was found that the ASIC has a low solution
resistance (Rs) of 0.91 Ω and charge transfer resistance (Rct) of 3.06 Ω.
This result may be attributed to the high ion conductivity of aqueous
electrolyte and fast kinetics processes for cathode and anode materials.

10.1002/ente.201800157

Energy Technology

FULL PAPER

Synthesis
of
PNTCDA
dianhydride derived polyimide)

(1,4,5,8-naphthalenetetracarboxylic

PNTCDA was prepared according to our previous report.[18b] Equimolar
1,4,5,8-naphthalenetetracarboxylic dianhydride (NTCDA) and ethylene
diamine (EDA) were added in N-methylpyrrolidone (NMP) at room
temperature. After stirring for 4 h, the mixture was refluxed at 200 ºC
under a flow of inert gas for 8 h, and then cooled down to room
temperature. The as-prepared product was collected and rinsed with
methanol and NMP for several times, dried under vacuum at 120 ºC for
12 h. Finally, the as-prepared resultant was heated in N2 atmosphere for
8 h at 300 ºC to remove the residual solvents, then marked as PNTCDA.
Physicochemical characterization
The microscopic morphologies of specimens before and after activation
were recorded using field-emission scanning electron microscopy
(FESEM, HITACHI S-4800), transmission electron microscopy (TEM,
FEI, Tecnai-20), and high-resolution transmission electron microscopy
(HRTEM, JEOL JEM-2010). To estimate the contents of water and
CNTs, thermogravimetric (TG) analysis was carried out from 30 to 800
ºC by the usage of a Perkin-Elmer TG/DTG-6300, with a heating rate of
5 °C min-1. The crystal structure was investigated with a Bruker D8
Advance X-ray diffraction (XRD) using Cu kα radiation. The N2
adsorption/desorption isotherms were carried out by Brunauer-EmmettTeller (BET) measurements using an ASAP-2010 surface area analyzer.
The X-ray photoelectron spectroscopy (XPS) analysis was performed on
a PerkinElmer PHI 550 spectrometer with Al kα (1486.6 eV). Fourier
transform infrared spectra were operated on a Shimadzu IR Prestige-21
FT-IR spectrometer with KBr pellets.
Electrochemical measurement
The MnO2/CNTs cathode was fabricated by pressing slurry containing
active material, Ketjen Black (KB) and polytetrafluoroethylene (PTFE)
with a weight ratio of 8:1:1 on stainless steel grid. The PNTCDA anode
was prepared by pressing 60wt% active material, 30wt% KB, and 10wt%
PTFE on stainless steel grid. The electrochemical performance of
individual electrode was characterized by employing a three-electrode
system, where Ag/AgCl was used as reference electrode and platinum
plate was acted as counter electrode, NaClO4-based WiSE (17 mol
NaClO4/1 L H2O, 17 m) was used as electrolyte with continuous
bubbling nitrogen to remove oxygen in all the electrochemical
measurements. And the masses loading of the active materials on the
electrode are all about 2 mg cm-2. The ASICs were assembled in
CR2032-type coin cell by using PNTCDA as anode, MnO2/CNTs as
cathode and non-woven fabrics as separator in the atmosphere. Cyclic
voltammetry (CV) and electrochemical impedance spectroscopy (EIS)
were carried out using CHI 760e electrochemical work station for both
three-electrode and full ASIC systems. Galvanostatically chargedischarge experiment and cycle performance were tested using a
CT20001A cell test instrument (LAND Electronic Co.).

Acknowledgements

This work was supported by the National Program on Key Basic
Research Project of China (973 Program, no. 2014CB239701), Natural
Science Foundation of China (no. 51372116, 51672128 and 21773118),
and a Project Funded by the Priority Academic Program Development
of Jiangsu Higher Education Institutions (PAPD).

Conflict of interest
The authors declare no conflict of interest.
Keywords: manganese dioxide ·pseudocapacitance·polyimide·salt-inwater ·sodium-ion capacitor
[1] a) B. E. Conway, J. Electrochem. Soc. 1991, 138, 1539; b) G. Wang, L. Zhang, J.
Zhang, Chem. Soc. Rev. 2012, 41, 797-828; c) J. Jiang, Y. Zhang, P. Nie, G. Xu, M.
Shi, J. Wang, Y. Wu, R. Fu, H. Dou, X. Zhang, Adv. Sustainable Syst. 2017,
1700110; d) D. P. Dubal, J. G. Kim, Y. Kim, R. Holze, C. D. Lokhande, W. B. Kim,
Energy Technology 2014, 2, 325-341.
[2] a) W. Zuo, R. Li, C. Zhou, Y. Li, J. Xia, J. Liu, Adv. Sci. 2017, 4, 1600539; b) V.
Aravindan, M. Ulaganathan, S. Madhavi, J. Mater. Chem. A 2016, 4, 7538-7548; c)
H. Wang, C. Zhu, D. Chao, Q. Yan, H. J. Fan, Adv. Mater. 2017, 29.
[3] a) D. P. Dubal, O. Ayyad, V. Ruiz, P. Gómezromero, Chem. Soc. Rev. 2015, 46,
1777-1790; b) Y. D. Zhang, Z. A. Hu, Y. R. Liang, Y. Y. Yang, N. An, Z. Li, H.
Wu, J. Mater. Chem. A 2015, 3, 15057-15067; c) J. Jiang, P. Nie, B. Ding, Y. D.
Zhang, G. Xu, L. Wu, H. Dou, X. Zhang, J. Mater. Chem. A 2017, 5, 23283-23291.
[4] a) L. Suo, O. Borodin, T. Gao, M. Olguin, J. Ho, X. Fan, C. Luo, C. Wang, K. Xu,
Science 2015, 350, 938-943; b) C. Yang, J. Chen, T. Qing, X. Fan, W. Sun, A. von
Cresce, M. S. Ding, O. Borodin, J. Vatamanu, M. A. Schroeder, N. Eidson, C. Wang,
K. Xu, Joule 2017, 1, 122-132; c) J. W. Morales, H. c. R. Galleguillos, F.
Hernández-Luis, R. Rodríguez-Raposo, J. Chem. Eng. Data 2011, 56, 3449-3453.
[5] a) Z. Tian, W. Deng, X. Wang, C. Liu, C. Li, J. Chen, M. Xue, R. Li, F. Pan, Funct.
Mater. Lett., 2017, 10, 1750081; b) K. Nakamoto, R. Sakamoto, M. Ito, A. Kitajou,
S. Okada, Electrochemistry 2017, 85, 179-185; c) L. Suo, O. Borodin, Y. Wang, X.
Rong, W. Sun, X. Fan, S. Xu, M. A. Schroeder, A. V. Cresce, F. Wang, C. Yang, Y.
S. Hu, K. Xu, C. Wang, Adv. Energy Mater. 2017, 7, 1701189; d) D. P. Leonard, Z.
Wei, G. Chen, F. Du, X. Ji, ACS Energy Lett. 2018.
[6] a) Y. Dall'Agnese, P. L. Taberna, Y. Gogotsi, P. Simon, J. Phys. Chem. Lett. 2015,
6, 2305-2309; b) X. Wang, S. Kajiyama, H. Iinuma, E. Hosono, S. Oro, I. Moriguchi,
M. Okubo, A. Yamada, Nat. Commun. 2015, 6, 6544.
[7] a) Y. Chen, Y. Zhang, D. Geng, R. Li, H. Hong, J. Chen, X. Sun, Carbon 2011, 49,
4434-4442; b) Z. Li, Y. An, Z. Hu, N. An, Y. Zhang, B. Guo, Z. Zhang, Y. Yang,
H. Wu, J. Mater. Chem. A 2016, 4, 10618-10626; c) D. P. Dubal, J. G. Kim, Y. Kim,
R. Holze, W. B. Kim, Energy Technology 2013, 1, 125-130; d) K. Wang, S. Gao, Z.
Du, A. Yuan, W. Lu, L. Chen, J. Power Sources 2016, 305, 30-36.
[8] a) M. Toupin, T. Brousse, D. Bélanger, Chem. Mater. 2004, 16, 3184-3190; b) Z.
Lei, J. Zhang, X. S. Zhao, J. Mater. Chem. 2011, 22, 153-160; c) S. Devaraj, N.
Munichandraiah, J. Phys. Chem. C 2008, 112, 4406-4417; d) O. Ghodbane, J. L.
Pascal, F. Favier, ACS Appl. Mater. Interf. 2009, 1, 1130-1139.
[9] a) K. Jeyasubramanian, S. Purushothaman, M. V. Kumar, I. Sushmitha,
Electrochimica Acta 2017, 227, 401-409; b) J. Yan, Z. Fan, T. Wei, J. Cheng, B.
Shao, K. Wang, L. Song, M. Zhang, J. Power Sources 2009, 194, 1202-1207.
[10] N. Zhang, F. Cheng, J. Liu, L. Wang, X. Long, X. Liu, F. Li, J. Chen, Nat. Commun.
2017, 8, 405.
[11] N. Jabeen, A. Hussain, Q. Xia, S. Sun, J. Zhu, H. Xia, Adv. Mater. 2017, 29.
[12] a) E. Moazzen, E. V. Timofeeva, C. U. Segre, J. Mater. Sci. 2017, 52, 8107-8118;
b) S. Kim, S. Lee, K. W. Nam, J. Shin, S. Y. Lim, W. Cho, K. Suzuki, Y. Oshima,
M. Hirayama, R. Kanno, J. W. Choi, Chem. Mater. 2016, 28, 5488-5494; c) L.
Athouël, F. Moser, R. Dugas, O. Crosnier, D. Bélanger, T. Brousse, J. Phys. Chem.
C 2008, 112, 7270-7277.
[13] L. Li, Z. A. Hu, N. An, Y. Y. Yang, Z. M. Li, H. Y. Wu, J. Phys. Chem. C 2014,
118, 22865-22872.
[14] C. Julien, Solid State Ionics 2003, 159, 345-356.
[15] a) S. Dong, L. Shen, H. Li, P. Nie, Y. Zhu, Q. Sheng, X. Zhang, J. Mater. Chem.
A, 2015, 3, 21277-21283; b) H. Wang, Y. Zhang, H. Ang, Y. Zhang, H. T. Tan, Y.

This article is protected by copyright. All rights reserved.

Accepted Manuscript

The resultant was then dried in a vacuum oven at 60 °C for 12 h to obtain
MnO2/CNTs composite. The pure MnO2 was prepared by a similar
procedure with alcohol instead of CNTs.

10.1002/ente.201800157

Energy Technology

[16]
[17]

[18]

[19]
[20]
[21]
[22]
[23]
[24]

Zhang, Y. Guo, J. B. Franklin, X. L. Wu, M. Srinivasan, H. J. Fan, Q. Yan, Adv.
Funct. Mater. 2016, 26, 3082-3093.
Y. Huang, K. Li, J. Liu, X. Zhong, X. Duan, I. Shakir, Y. Xu, J. Mater. Chem. A
2017, 5.
a) Z. Guo, L. Chen, Y. Wang, C. Wang, Y. Xia, ACS Sustainable Chem. Eng. 2017,
5, 1503-1508; b) X. Dong, L. Chen, J. Liu, S. Haller, Y. Wang, Y. Xia, Sci. Adv.
2016, 2, e1501038; c) X. Dong, H. Yu, Y. Ma, J. L. Bao, D. G. Truhlar, Y. Wang,
Y. Xia, Chem. Eur. J. 2017, 23, 2560-2565.
a) L. Chen, W. Li, Y. Wang, C. Wang, Y. Xia, RSC Adv. 2014, 4, 25369-25373; b)
Y. Zhang, P. Nie, C. Xu, G. Xu, B. Ding, H. Dou, X. Zhang, Electrochimica Acta
2018, 268, 512-519.
Y. Meng, H. Wu, Y. Zhang, Z. Wei, J. Mater. Chem. A 2014, 2, 10842-10846.
R. Thangavel, K. Kaliyappan, D. U. Kim, X. Sun, Y. S. Lee, Chem. Mater. 2017,
29, 7122-7130.
P. Pazhamalai, K. Krishnamoorthy, M. S. P. Sudhakaran, S. J. Kim,
ChemElectroChem 2017, 4, 396-403.
Z. Le, F. Liu, P. Nie, X. Li, X. Liu, Z. Bian, G. Chen, H. B. Wu, Y. Lu, ACS Nano
2017, 11, 2952-2960.
K. Lu, B. Song, X. Gao, H. Dai, J. Zhang, H. Ma, J. Power Sources 2016, 303,
347-353.
H. Li, L. Peng, Y. Zhu, X. Zhang, G. Yu, Nano Lett. 2016, 16, 5938-5943.

This article is protected by copyright. All rights reserved.

Accepted Manuscript

FULL PAPER