Interfacial synthesis of porous MnO2 and its application in electrochemical capacitor

Article abstract of DOI:10.1016/j.electacta.2007.07.043

In this paper, the porous manganese dioxide (MnO₂₎ was prepared by an interfacial reaction of potassium permanganate in water/ferrocene in chloroform. The surface area and pore distribution of MnO₂ can be controlled by simply adjusting the reaction time and the content of surfactant in the aqueous phase. The electrochemical performance of the prepared MnO₂ was evaluated as an electrode material for supercapacitors by the means of cyclic voltammetry and galvanostatic charge-discharge tests. Electrochemical tests results indicated that the pore size plays an important role at high charge-discharge rate, the sample with a large pore size shows a better rate capability, while the sample with a small pore size but large surface area delivers a large capacitance at low current rate.

Full text of DOI:10.1016/j.electacta.2007.07.043

Electrochimica Acta 53 (2007) 752–757
Interfacial synthesis of porous MnO and its application
2
in electrochemical capacitor
∗
Xin-hui Yang, Yong-gang Wang, Huan-ming Xiong, Yong-yao Xia
Chemistry Department and Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials,
Fudan University, Shanghai 200433, China
Received 30 March 2007; received in revised form 21 July 2007; accepted 22 July 2007
Available online 26 July 2007
Abstract
In this paper, the porous manganese dioxide (MnO2) was prepared by an interfacial reaction of potassium permanganate in water/ferrocene in
chloroform. The surface area and pore distribution of MnO can be controlled by simply adjusting the reaction time and the content of surfactant
in the aqueous phase. The electrochemical performance of the prepared MnO was evaluated as an electrode material for supercapacitors by the
2
2
means of cyclic voltammetry and galvanostatic charge–discharge tests. Electrochemical tests results indicated that the pore size plays an important
role at high charge–discharge rate, the sample with a large pore size shows a better rate capability, while the sample with a small pore size but large
surface area delivers a large capacitance at low current rate.
©
2007 Elsevier Ltd. All rights reserved.
Keywords: Porous MnO2; Interfacial synthesis; Supercapacitor
1
. Introduction
electrolyte and electrode. Various noble and transition metal
oxides such as RuO2, IrO2, NiO2, CoOx, SnO2, and MnO2
were used as electrode materials [8–12]. Among these materials,
amorphous hydrated ruthenium oxide exhibits ideal pseudoca-
pacitive behavior, remarkably high specific capacitance as high
as 700–1000 F/g [8], and excellent reversibility. However, the
high cost and the toxicity to environment have limited its prac-
tical application. Therefore, some cheap and environmentally
friendly materials have got more and more attention. MnO2
is the most promising candidate because of its low cost and
environmentally benign nature [13].
Manganese oxides as pseudocapacitor electrode materials
were synthesized using different technique such as simple
reduction, coprecipitation, thermal decomposition and sol–gel
processes, etc. Also, various thin film electrodes of MnO2
were synthesized via electrochemical and chemical routes
[14–16]. Organic-aqueous interfacial synthesis is an alterna-
tive useful approach to conventional homogeneous synthesis. A
liquid/liquid interface offers potential for the chemical manipu-
lation of nanoparticles. At a liquid/liquid interface, the particles
are highly mobile and rapidly achieve an equilibrium assembly.
Interfacesbetweenimmisciblefluidshavebeenshowntobeideal
for the assembly of nanometer-sized colloidal particles [17].
For nanoparticles, thermal energy, which causes spatial fluctu-
In recent years, the growing interest in supercapacitors has
been stimulated by their potential application as electrical stor-
age devices operating in parallel with the battery in an electric
vehicle to transiently provide high power [1,2]. Electrochem-
ical supercapacitors (ECs) fill the gap between batteries and
conventional capacitors in terms of their specific energy and spe-
cific power [3–5]. Supercapacitors have been used as small-scale
energy storage devices in stationary electronics, such as mem-
ory back-up devices and solar batteries with semi-permanent
charge–discharge cycle life [6,7].
Depending on the charge storage mechanism, ECs are cat-
egorized as electrochemical double layer capacitor (EDLC)
and pseudocapacitors. In electrochemical double layer capac-
itors, the charge is separated across the interface between the
electrode and the electrolyte. Various carbonaceous materials
ranging from amorphous carbons to carbon nanotubes have
been used as electrode materials in EDLCs. In the case of
pseudocapacitors, the Faradic charge transfer occurs between
∗
Corresponding author. Tel.: +86 21 55664177; fax: +86 21 55664177.
E-mail address: yyxia@fudan.edu.cn (Y.-y. Xia).
0
013-4686/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2007.07.043

X.-h. Yang et al. / Electrochimica Acta 53 (2007) 752–757
753
ations of the particles, is comparable to the interfacial energy.
This energy balance results in a weak interfacial aggregation of
nanoparticles. The liquid/liquid interface provide easy access to
the nanoparticles for their self-assembly by the reaction of the
attached ligands. It has been employed to obtain nanoparticles
such as Au, Ag, Pd, Cu, CuS, CuSe, CuO and Cu(OH)2 in the
literatures [18–24].
X-ray diffraction (XRD, Bruker D8), scanning electronic micro-
scope (SEM, Philip XL30) and BET surface area measurement
(Micromeritics Tristar).
2.2. Electrochemical tests
The electrode of MnO2 was prepared according to the follow-
In this paper, we for the first time report the synthesis of
porous MnO2 employed organic/aqueous interface. The size of
MnO2 was achieved by simply tuning the concentration of the
surfactant in the aqueous phase. The effects of the particle size,
pore size, and the surface area on the capacitance performance
of the MnO2 were studied extensively.
ing steps. The mixture containing 80 wt.% MnO2 and 15 wt.%
acetylene black and 5 wt.% polytetrafluoroethylene (PTFE)
was well mixed, and then was pressed onto nickel foam grid
7
(
1
1.2 × 10 Pa) that served as a current collector (surface is
2
cm ). The typical mass load of electrode material is 5 mg. The
used electrolyte was 0.5 M potassium sulfate. The electrochem-
ical behavior of MnO2 was characterized by cyclic voltammetry
and charge–discharge tests. The experiments were carried out in
a three-electrode glass cell. Platinum foil was used as a counter
electrode, and SCE as a reference electrode. The electrochemical
measurements were performed using a Solartron Instru-
ment Model 1287 electrochemical interface controlled by a
computer.
2
. Experimental
2
.1. Synthesis and characterization of MnO2
MnO2 were prepared at the interface between organic phase
and aqueous phase at room temperature. The aqueous phase
contains a mixture of potassium permanganate (KMnO4) and
sodium dodecylsulfate (SDS) which serves as surfactant in
water. The concentration KMnO4 was 0.01 M and the SDS
concentration is 3, 8, 16, 40, 80, 120, 160 and 320 mM, respec-
tively. 25 ml above solution was added on the 25 ml, 0.03 M
ferrocene/CHCl3 solution. A static organic/aqueous interface
was formed. After a short introduction period, some puce prod-
ucts appeared at the interface. As the puce products were
produced continuously, the color of the aqueous phase turned
lighter and lighter until a puce precipitate layer was formed at
the liquid/liquid interface. The organic phase was removed by
the pipette in a given time varying from 3.5 to 24 h. The puce pre-
cipitates and the aqueous phase were collected and centrifuged
at 6000 rpm for 6 min. The precipitates were settled down to the
bottom of the container under centrifugation then washed with
distilled water and ethanol repeatedly until the liquid phase was
3. Results and discussion
3.1. Characterization of materials
A series of MnO2 were prepared by the interfacial reaction
of potassium permanganate (KMnO4)/sodium dodecylsulfate
(SDS)/ferrocene containing various concentration of SDS vary-
ing from 3, 8, 16, 40, 80, 120 and 160 to 320 mM. The schematic
graph of the formation of MnO2 can be described in Fig. 1. At
the present of SDS, the micelles were formed and acted as col-
loid nanoreactors. The micelles containing permanganate salt
arranged at the aqueous/organic interface, ferrocene molecule
contacted the permanganate at the core of the micelles, redox
took place at the interface and the MnO2 particles were formed in
the micelles. All samples show the very similar XRD patterns.
Fig. 2 shows the selected XRD patterns of MnO2 prepared at
different concentrations of SDS, from 3, 80 to 320 mM (sam-
ples A, E and H). All three samples present broad XRD peaks,
indicating the amorphous nature.
◦
colorless after centrifugation. The product was dried at 70 C
in a vacuum for 10 h. The product prepared from the different
SDS contents mentioned above are denoted as sample A, B, C,
D, E, F, G and H. The prepared MnO2 was characterized by the
Fig. 1. Schematic graph of the formation of MnO2 at interface of potassium permanganate in water/ferrocene in chloroform.

7
54
X.-h. Yang et al. / Electrochimica Acta 53 (2007) 752–757
compact morphologies. N2 adsorption–desorption studies were
performed to determine the specific surface area of the MnO2
prepared at different concentrations of SDS. The BET surface
2
area (Fig. 5) decreases gradually from 264 m /g of sample A to
2
1
73 m /g of sample H. The adsorption average pore diameter is
.4, 13.1 and 16.7 nm for the samples A, E and H, respectively.
7
This is the reason why sample A has the largest surface area
even it has the largest particle size as the SEM images shown in
Fig. 4. It is clear that the surface area mainly contributes from
the pore in the MnO2 particle. The detailed pore size distribution
is presented in Fig. 6. Sample A has a narrow pore size distri-
bution around 5 nm; sample E shows the pore size distribution
with an average pore size of 5 and 15 nm, while sample H has a
pore size around 5 and 30 nm. The amount of micelles increases
and micelle size decreases when the concentration of surfactant
SDS increases, thus leads to smaller particle size. On the other
hand, the distance between the micelles increases when the con-
centrations of SDS increase because of the increasing charge
repulsion. Therefore, the particle size depends on the micelle
size while the pore size is associated with the distance between
the micelles.
Fig. 2. X-ray diffraction patterns of MnO2 synthesized at the interface between
organic/aqueous interface for different concentrations of SDS in aqueous phase.
(
a) 3 mM (sample A); (b) 80 mM (sample E) and (c) 320 mM (sample H).
In order to optimize the reaction parameters, the reaction time
is investigated. Fig. 3 gives the SEM image of the sample E pre-
pared at the different times, 3.5, 4.5, 6.5 and 24 h, respectively.
The particle size of MnO2 increases with increasing of the reac-
tion time. As demonstrated later, the sample prepared at 4.5 h
delivers the highest specific capacitance of 260 F/g. The reaction
time is fixed at 4.5 h. The effect of the added SDS concentration
on the morphology was investigated. Fig. 4 presents the SEM
image of the MnO2. The images reveal the MnO2 is aggregated
spherical particle. The particle sizes decrease from 300 to 30 nm
when the concentrations of SDS increase from 3 to 320 mM. The
samples prepared in low concentration of SDS have much more
3.2. Electrochemical tests
Fig. 7 shows the CV curves of sample E at different scan
rates from 2, 5, to 10 mV/s over the voltage range of 0.0–0.8 V
versus SCE in 0.5 M K2SO4 solution. The Y-axis of capacitance
was calculated by the formula C = i/v where C is the differen-
tial capacitance, i is the current density, v is the scan rate. As
shown in Fig. 6, the typical quasi-rectangular curves indicate
good EDLC behaviors. Moreover, the good rectangular sym-
Fig. 3. SEM images of MnO2 prepared at different reaction times: (a) 3.5 h; (b) 4.5 h; (c) 6.5 h; and (d) 24 h.

X.-h. Yang et al. / Electrochimica Acta 53 (2007) 752–757
755
Fig. 4. SEM of images of the MnO2 materials obtained at room temperature and 4.5 h. Panels a, b, c, d, e, f correspond to the samples synthesized from 3, 8, 40, 80,
60 and 320 mM SDS in aqueous phase, respectively.
1
metry of CV curves can be surely maintained in the 0–0.8 V
region when the scan rate is increased from 2 to 10 mV/s, which
indicates the excellent capacitive ability of MnO2 even at high
current density.
The galvanostatic charge–discharge behavior of MnO2 was
investigated within potential range 0.0–0.8 V versus SCE in
0
.5 M K2SO4 solution at different currents. The current density
2
varied from 40 to 0.5 mA/cm . The specific capacitance (Cm)
can be calculated as follows:
C
It
Cm =
=
m
ꢀVm
where I is the current of charge–discharge, t the time of dis-
charge, ꢀV is 0.8 V, and m the mass of active materials in the
working electrode. Table 1 shows the specific capacitance of the
eight samples at different charge–discharge current densities.
Broadly speaking, the specific capacitance increases with
the decrease of current density for all eight samples, while the
increase tendency varies from sample to sample. The difference
can be more clearly observed in Fig. 8 for the three selected
samples, samples A, E and H. For sample A, the capacitance
Fig. 5. Surface areas of MnO2 materials synthesized at the presence of 3, 8, 16,
0, 80, 120, 160 and 320 mM SDS in the aqueous phase.
4

7
56
X.-h. Yang et al. / Electrochimica Acta 53 (2007) 752–757
Table 1
Specific capacitance of MnO2 prepared at different concentrations of SDS in
different current densities
Current density Specific capacitance (F/g)
2
(
mA/cm )
A
B
C
D
E
F
G
H
4
3
2
1
0
0
0
0
8
5
2
1
0
98
122
132
134
139
146
157
178
222
245
125
135
148
150
151
159
184
238
241
146
150
163
184
198
201
216
238
240
155
165
184
196
225
240
252
256
261
155
165
176
193
205
215
236
255
256
158
168
182
191
195
202
221
226
228
176
184
184
189
193
200
215
216
219
108
129
134
141
151
182
210
229
.5
Sample H behaves the best rate capability among eight
samples.
It is easy to be envisaged that difference in the capacitance
behavior is associated with the pore size and its distribution.
There have been two mechanisms proposed for the charge stor-
age in MnO2-based electrodes. The first one is base on the
+
concept of intercalation of H or alkali metal cations such as
+
K in the electrode during reduction and deintercalation upon
oxidation [16,25].
+
−
MnO2 + H + e ꢀ MnOOH
Fig. 6. Pore size distribution (a) for sample A, (b) for sample E and (c) for
sample H.
or
2
+
−
MnO + C + e ꢀ MnOOC
rapidly decrease from 229 to 151 F/g when the current density
2
+
−
− +
increase from 0.5 to 5 mA/cm , then decrease slowly when the
(MnO2)surface + C + e ꢀ (MnO2 C )surface
2
current rate further increases. The capacitance at 40 mA/cm is
The second mechanism is based on the surface adsorption of
2
only 98 F/g, only 40% of the capacitance at 0.5 mA/cm . Sam-
+
electrolyte cations (C ) on MnO2 [14].
ple E shows the highest capacitance among all eight samples.
The capacitance decreases slowly from 261 to 151 F/g when
+
+
+
+
Where C = Na , K , Li . Despite of the inconsistent pro-
posed mechanism, it has been demonstrated that the capacitance
oftheMnO2 resultsfromitssurfacereaction. Onlyalimitedfrac-
tion of the MnO2 is electrochemically active, the charge storage
might only involve the surface atoms of the MnO2 crystallites or
a very thin layer. Therefore, the pseudocapacitance is critically
2
the current rate increase from 0.5 to 40 mA/cm ; for sample
H, the capacitance decrease very slowly from 219 to 200 F/g
2
when the current rate increases from 0.5 to 5 mA/cm . It keeps
2
2
8
0% of the capacitance at 0.5 mA/cm , even at 40 mA/cm .
Fig. 7. CV curves of MnO2 in 0.5 M K2SO4 electrolyte at (a) 2 mV/s, (b) 5 mV/s
and (c) 10 mV/s scan rate.
Fig. 8. Specific capacitance of samples A, E and H in different current densities.

X.-h. Yang et al. / Electrochimica Acta 53 (2007) 752–757
757
Table 2
4. Conclusion
Pore volume proportions at the range of 7 nm below, 7–20 and 20 nm above, for
samples A, E and H
In this paper, we successfully synthesized the porous MnO2
with various pore sizes, varying from 5 to 30 nm by organic-
aqueous interfacial method. The surface area and pore diameter
of MnO2 can be easily tuned by the content of SDS at aque-
ous phase. The pore diameter is the primary factor to affect the
capacitance at high charge–discharge rate, while surface area
becomes the primary factor at low charge–discharge rate. The
MnO2 synthesized under optimum conditions shows a capaci-
tance of 261 F/g, and exhibits good cycle profile, keeping the
97% of initial capacity over 1300 cycles with a coulomb effi-
ciency approximately 100%.
Pore volume proportion (%)
Pore size range (nm)
Sample
A
E
H
<
7
70
20
10
20
55
35
10
30
40
7
–20
>
20
dependent on the effective surface area which is associated the
particle size and pore distribution.
Generally, thespecificcapacitanceincreaseswiththeincrease
of surface area. Sample A with the largest surface area has lower
capacitance than sample H at high charge–discharge rate. Sim-
ilar results of systems showing lower capacitance for a larger
surfaceareamaterialhavebeenreportedintheliterature[26–29].
This indicates the specific capacitance does not depend on the
surface area. Pore size distribution may be a deciding parameter
in determining the specific capacitance. The pore volume pro-
portions of samples A, E and H at different ranges are shown
in Table 2. Sample A has 70% of the pore volume under 7 nm,
compared to sample H with 10%. It leads to the sharp perfor-
mance difference at high rate charge–discharge process because
the pores under 7 nm are adverse for the fast intercalation and
Acknowledgments
This work was partially supported by the National Natural
Science Foundation of China (No. 20633040), the 863 pro-
gram of China (No. 2006AA05Z218), and the State Key Basic
Research Program of PRC (2007CB9700).
References
[
[
1] B.E. Conway, J. Electrochem. Soc. 138 (1991) 1539.
2] C. Arbizzani, M. Mastragostino, F. Soavi, J. Power Sources 100 (2001)
164.
+
[3] S.T. Mayer, R.W. Pekala, J.L. Kaschmitter, J. Electrochem. Soc. 140 (1993)
446.
deintercalation of K . As the charge–discharge rate decreases
2
to 0.5 mA/cm , sample A has larger capacitance than sample H
[
4] A. Yoshida, S. Nonaka, I. Aoki, A. Nishino, J. Power Sources 60 (1996)
13.
because of larger surface area. Sample E has average distribu-
tion at the range of 3–40 nm. It keeps the active surface area
and performs well at different charge–discharge rates. Sample
H has the best rate capability for the contribution of 40% pore
volume in the form of mesopores with diameter > 20 nm but
poor performance for at low rate for the limitation of surface
area.
2
[
5] Y. Kibi, T. Saito, M. Kurata, J. Tabuchi, A. Ochi, J. Power Sources 60
(1996) 225.
[
[
6] Y.G. Wang, Y.Y. Xia, Electrochem. Commun. 7 (2005) 1138.
7] M. Endo, T. Maeda, T. Takeda, Y.J. Kim, K. Koshiba, H. Hara, M.S.
Dresselhaus, J. Electrochem. Soc. 148 (2001) A910.
[
[
8] J.P. Zheng, P.J. Cygan, T.R. Jow, J. Electrochem. Soc. 142 (2001) 2699.
9] K.C. Liu, M.A. Anderson, J. Electrochem. Soc. 143 (1996) 124.
Fig. 9 gives the charge–discharge cycling test of sample E
[10] Y.S. Yoon, W.I. Cho, J.H. Lim, D.J. Choi, J. Power Sources 101 (2001)
26.
[11] B.E. Conway, V. Briss, J. Wojtowicz, J. Power Sources 66 (1997) 1.
1
synthesized at the present of 80 mM SDS at current density of
2
4
0 mA/cm between 0 and 0.8 V versus SCE. As shown in Fig. 9,
[
[
[
12] C. Lin, J. Ritter, B.N. Popov, J. Electrochem. Soc. 145 (1998) 4097.
13] H.Y. Lee, S.W. King, Electrochem. Solid-State Lett. 4 (2001) A19.
14] H.Y. Lee, J.B. Goodenough, J. Solid State Chem. 144 (1999) 220.
the MnO2 exhibits good cycle profile, the capacitance slightly
decreases 3% of initial capacity over 1300 cycles with a coulomb
efficiency approximately 100%.
[15] H.Y. Lee, V. Manivannan, J.B. Goodenough, C. R. Chim. 2 (1999) 565.
[
16] S. Pang, M.A. Anderson, T.W. Chapman, J. Electrochem. Soc. 147 (2000)
44.
17] Y. Lin, H. Skaff, T. Emrick, A.D. Dinsmore, T.P. Russell, Nature 299 (2003)
26.
4
[
2
[
[
18] L.L. Dai, R. Sharma, C.Y. Wu, Langmuir. 21 (2005) 2641.
19] J.K. Sakata, A.D. Dwoskin, J.L. Vigorita, E.M. Spain, J. Phys. Chem. B
1
09 (2005) 138.
20] F. Reincke, S.G. Hickey, W.K. Kegel, D. Vanmaekelbergh, Angew. Chem.
16 (2004) 464.
[
1
[21] A.W. Dryfe, A.O. Simm, B. Kralj, J. Am. Chem. Soc. 125 (2003) 13014.
[22] U.K. Gautam, M. Ghosh, C.N.R. Rao, Chem. Phys. Lett. 381 (2003) 1.
[23] U.K. Gautam, M. Ghosh, C.N.R. Rao, Langmuir 20 (2004) 10776.
[24] X.Y. Song, S.X. Sun, W.M. Zhang, H.Y. Yu, W.L. Fan, J. Phys. Chem. B
108 (2004) 5200.
[25] M. Toupin, T. Brousse, D. Belanger, Chem. Mater. 16 (2004) 3184.
[26] Y.U. Jeong, A. Manthiram, J. Electrochem. Soc. 149 (2002) A1419.
[27] R.N. Reddy, R.G. Reddy, J. Power Sources 124 (2003) 330.
[28] R.N. Reddy, R.G. Reddy, J. Power Sources 132 (2004) 315.
[29] V. Subramanian, H.W. Zhu, R. Vajtai, P.M. Ajayan, B.Q. Wei, J. Phys.
Chem. B 109 (2005) 20207.
Fig. 9. Cycle performance of MnO2 sample prepared in 80 mM SDS in aqueous
solution, (a) ratio of retained capacitance and (b) coulomb efficiency.