Synthesis and control of high-performance MnO2/carbon nanotubes nanocomposites for supercapacitors

Article abstract of DOI:10.1016/j.jallcom.2016.07.005

The rational combination of transition-metal oxides and carbon materials has been regarded as an efficient way of assembling the next-generation of electrode materials for supercapacitors. Here, a unique nanomaterial consisting of carbon nanotubes (CNTs) grafted with MnO₂was synthesized for use in supercapacitors with high electrochemical performance. The synthesis was performed in three steps: surface functionalization of the CNTs with carboxyl groups, grafting of Mn²⁺with the carboxyl groups, and transformation of the Mn²⁺into MnO₂nanoparticles by KMnO₄oxidation. The bonds between the CNTs and the MnO₂particles are believed to form between the carbon of the carboxyl groups of the CNTs and the oxygen atoms of the MnO₂lattice. In addition, varying the acid-treatment time enabled control of the growth density of MnO₂and affected the structural integrity of the CNTs, endowing the novel hybrid material with an adjustable electrochemical performance. As a consequence, the MnO₂/CNTs composite with a 36?h acid treatment exhibited a maximum specific capacitance of up to 348?F?g^?1at a cyclic voltammetry scanning rate of 5?mV?s^?1. A symmetric supercapacitor assembled with the electrodes made from this MnO₂/CNTs composite showed a capacitance decay rate of only 8.7% after 5000 cycles.

Full text of DOI:10.1016/j.jallcom.2016.07.005

Accepted Manuscript
Synthesis and control of high-performance MnO /carbon nanotubes nanocomposites
2
for supercapacitors
Jia-Wei Wang, Ya Chen, Bai-Zhen Chen
PII:
S0925-8388(16)32037-0
10.1016/j.jallcom.2016.07.005
JALCOM 38173
DOI:
Reference:
To appear in: Journal of Alloys and Compounds
Received Date: 11 May 2016
Revised Date: 30 June 2016
Accepted Date: 1 July 2016
Please cite this article as: J.-W. Wang, Y. Chen, B.-Z. Chen, Synthesis and control of high-performance
MnO /carbon nanotubes nanocomposites for supercapacitors, Journal of Alloys and Compounds
2
(2016), doi: 10.1016/j.jallcom.2016.07.005.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
1
2
3
Synthesis and control of high-performance
MnO2/carbon nanotubes nanocomposites for
supercapacitors
4
5
Jia-Wei Wang, Ya Chen*, Bai-Zhen Chen
School of Metallurgy and Environment, Central South University, Changsha 410083, China
6
7
*Corresponding author: Ya Chen
8
Tel.: +86 731 88877352; fax: +86 731 88877352.
9
E-mail: chenya1973a@sina.com
10
11
12
13
14
15
16
Author: Jiawei Wang
E-mail: wjwhappygirl@126.com
Coauthor: Baizhen Chen
E-mail: yhscbz@yahoo.com.cn
17
18
1

ACCEPTED MANUSCRIPT
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Abstract: The rational combination of transition-metal oxides and carbon materials has been
regarded as an efficient way of assembling the next-generation of electrode materials for
supercapacitors. Here, a unique nanomaterial consisting of carbon nanotubes (CNTs) grafted with
MnO2 was synthesized for use in supercapacitors with high electrochemical performance. The
synthesis was performed in three steps: surface functionalization of the CNTs with carboxyl groups,
grafting of Mn2+ with the carboxyl groups, and transformation of the Mn2+ into MnO2 nanoparticles
by KMnO4 oxidation. The bonds between the CNTs and the MnO2 particles are believed to form
between the carbon of the carboxyl groups of the CNTs and the oxygen atoms of the MnO2 lattice. In
addition, varying the acid-treatment time enabled control of the growth density of MnO2 and affected
the structural integrity of the CNTs, endowing the novel hybrid material with an adjustable
electrochemical performance. As a consequence, the MnO2/CNTs composite with a 36 h acid
treatment exhibited a maximum specific capacitance of up to 348 F g-1 at a cyclic voltammetry
scanning rate of 5 mV s-1. A symmetric supercapacitor assembled with the electrodes made from this
MnO2/CNTs composite showed a capacitance decay rate of only 8.7% after 5000 cycles.
Keywords ꢀGrafting oxidation; Manganese dioxide; Supercapacitor; Carbon nanotubes;
Nanocomposites;
2

ACCEPTED MANUSCRIPT
37 1. Introduction
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Supercapacitors, also known as electrochemical capacitors or ultracapacitors, have been
regarded as critical complementary charge storage devices to conventional batteries in various
applications because of their higher power density, fast charge/discharge rate and long cycle life [1,
2]. On the basis of the fundamental mechanism of charge storage, supercapacitors are divided into
two categories: electrochemical double-layer capacitors and pseudocapacitors. The electrochemical
double-layer capacitive process is based on charge separation at the electrode/electrolyte interface,
whereas the pseudocapacitive process relies on redox reactions that occur in the electrode materials
[1, 3].
The most critical factors governing the electrochemical performance of supercapacitors are
related to the electrode materials. The electrode-active materials that are widely used for
supercapacitors include carbon, conducting polymers and transition-metal oxides. Among the
investigated electrode materials, transition-metal oxides are considered good candidates because of
their high capacitance. Among the previously investigated transition-metal oxides, RuO2 exhibits
excellent electrochemical properties [4]; however, its high cost and environmental toxicity have
limited its commercial application. Thus, other metal oxides, such as cobalt oxide [5], titanium oxide
[6], vanadium oxide [7], molybdenum oxide [8] and manganese oxide [9] have been investigated as
possible substitutes for RuO2. In addition, two dimensional nanostructure materials are increasingly
used as electrode materials in flexible supercapacitors recently, such as MoS2 [10], MXenes [11],
g-C3N4 [12], graphdiyne [13] and 2D transition metal oxides [14]. In particular, manganese oxides,
which are characterized by a high specific capacitance, low cost, high abundance and
3

ACCEPTED MANUSCRIPT
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
environmentally friendly nature, have attracted increasing interest as electrode-active materials for
supercapacitors [9]. Furthermore, a MnO2 electrode operated in conjunction with a neutral aqueous
electrolyte is considerably safer than most other metal oxides (such as RuO2 and NiO), which usually
operate in conjunction with strongly acidic or alkaline electrolytes. However, given the poor
electrical conductivity and low accessible active area of MnO
materials with a satisfactory specific capacitance is difficult in practice, thus hampering the extensive
commercial application of MnO in energy storage devices [15].
2
, the achievement of MnO electrode
2
2
To circumvent the aforementioned problems, one efficient strategy is to choose an appropriate
matrix with a high surface area to support MnO2 nanostructures, which can not only accelerate
charge transport during the electrochemical redox process but also prevent MnO2 agglomeration [16].
Thus, the incorporation of nanostructured MnO2 into carbonaceous materials would be effective and
represent a significant advancement; specifically, it would provide a large electroactive surface area
and a reliable electrical connection for the maximum utilization of MnO2. In this respect, various
carbonaceous materials, including activated carbon [9, 17], carbon nanotubes (CNTs) [18, 19],
carbon nanofibers (CNFs) [20, 21], graphene [22, 23] and carbon aerogels [24], have been integrated
with MnO2 for fabricating hybrid electrode materials. As one of the most promising carbon
nanomaterials, CNTs have one-dimensional tubular structures and exhibit outstanding
physicochemical properties, including high electrical conductivity, high mechanical strength, high
chemical stability and high activated surface area. The combination of MnO2 and CNTs may
integrate both the excellent electrical conductivity and chemical stability of CNTs and the high
specific capacitance of MnO2. So far researchers have reported several meaningful results concerned
4

ACCEPTED MANUSCRIPT
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
with MnO2/CNTs composite. Li et al. [25] synthesized a MnO2/CNTs composite with long cycle
stability by a modified one-pot technique. The obtained MnO2/CNTs composite showed a specific
capacitance of 201 F g−1 and exhibited an excellent cycle stability (no obvious capacitance decay
after 10 000 cycles at 1 A g−1). Zhu et al. [26] successfully prepared and used 3D CNT-graphene-Ni
foam composite as substrates to react with KMnO4, forming three-dimensional (3D)
MnO2-CNT-graphene-Ni hybrid foam. In aqueous electrolytes, the 3D MnO2-CNT-graphene-Ni
based prototype supercapacitors shown specific capacitances of 251 F g-1 at a current density of 1 A
g-1 with good cycling stability. Mai et al. [27] first developed the flexible hybrid electrode based on
MnO2 nanotubes (NTs) and CNTs composite. The as-synthesized freestanding CNT/MnO2 NT
hybrid electrodes were assembled in the form of solid-state SC devices using polyvinyl alcohol
(PVA)/LiCl as gel electrolyte. The device exhibited a volumetric capacitance of 5.1 F cm-3 and a high
energy density of 0.45 mW h cm-3 for the entire supercapacitor volume.
Within this context, extensive efforts have been dedicated to develop effective synthesis
protocols, such as chemical co-precipitation, thermal decomposition, and electrophoretic and
electrochemical deposition. Notably, the deposition of uniform MnO2 onto hierarchical carbonaceous
substrates, especially monolithic substrates, is challenging. Therefore, the development of a synthetic
technique to prepare a MnO2-carbonaceous material composite with high electrochemical
performance would represent an important achievement. Recently, a grafting oxidation method was
demonstrated to be an efficient technique for the preparation of MnO2-transition metal oxide and
MnO2-organic composites. Muralidharan et al. [28] investigated a material consisting of MnO2
grafted with V2O5 nanostructures that exhibited a specific capacitance, rate capacity and stability
5

ACCEPTED MANUSCRIPT
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
considerably better than those of pristine V2O5; however, the low electronic and ionic conductivity of
MnO2 were not improved. Downard et al. [29] modified the surface of MnO2 nanorods with aryl
groups using a grafting method. The stable attachment of aryl groups opened many opportunities for
enhancing the performance of MnO2 materials through tuning their surface properties. However, the
organic-based systems suffered drawbacks associated with high fabrications costs and the use of
highly toxic solvents. Ramirez-Castro et al. [30] recently prepared a carbon/MnO2 nanocomposite
via molecular bridging. The capacitance of the obtained carbon/MnO2 nanocomposite electrode was
double that of a simple mixture of the two components, highlighting the intimate coupling of carbon
and MnO2.
In our previous study, we successfully grafted uniform MnO2 particles onto the surface of
activated carbon (AC) through a grafting oxidation method [17]. The results preliminarily showed
that the obtained MnO2/AC composite exhibited excellent electrochemical properties. However,
whether the grafting oxidation method could be applied to a carbonaceous material of better
conductivity to prepare a MnO2/carbon composite with a higher capacitance and better
electrochemical performance was unknown. In this paper, we adopted a MnO2/CNTs composite
synthesized through the grafting oxidation method as an electrode material for electrochemical
supercapacitors. The MnO2 uniformly covered the CNTs, which provided a highly effective electron
pathway. In addition, the effect of the CNTs acid-treatment time on the Mn2+ grafting and MnO2
synthesis was studied. The formation mechanism and the effect of microstructure on the
electrochemical capacitive behavior of the binary composites were discussed.
120 2. Experimental
6

ACCEPTED MANUSCRIPT
121
2.1 Materials
122
123
124
125
126
127
Mn(Ac)2·4H2O, KMnO4, and HNO3 (70%) (Sinopharm Chemical Reagent Co., China) were
analytical grade and used without further purification. The multi-walled carbon nanotubes
(MWCNTs, >97% purity) used in this work were purchased from Shenzhen Nanotech Port Co. Ltd.
(Shenzhen, China), where the materials were prepared through chemical vapor deposition (CVD).
All of the aqueous solutions were prepared using double-distilled water. The KMnO4/Mn(Ac)2 molar
ratio was maintained at 2:3 for all of the syntheses.
128
129
2.2 Synthesis of MnO2-CNTs composites
2.2.1 Acid treatment of CNTs
130
131
132
133
134
135
136
137
138
139
140
Multi-walled carbon nanotubes (MWCNTs) (5 g) and 50 mL HNO3 were blended and subjected
to ultrasonication for 1 h. After the resulting mixture was enclosed in a PTFE-lined autoclave, the
autoclave was placed in an oven that was heated to 120 °C at a rate of 2 °C min-1 and maintained at
120 °C for X h (where X was 12 h, 24 h, 36 h, or 48 h); it was then naturally cooled to room
temperature. The obtained acid and MWCNTs dispersion was slowly poured into 10 L deionized
water for dilution. The diluted solution was then filtered and washed with deionized water until the
filtrate was neutral. The filtered MWCNTs powder was dried in an oven at 80 °C for 12 h and then
ground and labeled as C-X (where X was 12 h, 24 h, 36 h, or 48 h).
2.2.2 Grafting process
The acid-treated carbon nanotubes (CNTs) (1 g) were placed into a 100 mL 0.3 mol L-1
Mn(Ac)2 solution and ultrasonicated for 30 min. The obtained mixture was then placed in a water
7

ACCEPTED MANUSCRIPT
141
142
143
144
145
146
147
148
149
150
151
152
bath and maintained at 60 °C. Meanwhile, to prevent the agglomeration of the CNTs, the mixture
was continuously stirred for 24 h and then naturally cooled to room temperature. Afterwards, it was
filtered and washed with deionized water until the filtrate was neutral. The filter cake was dried in an
oven at 100 °C for 12 h, then ground and labeled as C-X+Mn(Ac)2 (where X was the duration of
acid-treatment, 12 h, 24 h, 36 h, or 48 h).
2.2.3 Oxidation process
The C-X+Mn(Ac)2 (1 g) was placed in a 100 mL of a pre-prepared 0.2 mol L-1 KMnO4 solution
and ultrasonicated for 30 min. The resulting mixture was maintained at 60 °C for 24 h in a water bath
under continuous stirring and then naturally cooled to room temperature. Afterwards, the mixture
was filtered and washed with deionized water until the filtrate was neutral. The filtered powders were
dried in an oven at 100 °C for 12 h and then ground and labeled as MnO2/C-X (where X was the
acid-treatment time, 12 h, 24 h, 36 h, and 48 h).
153
2.3 Synthesis of pure MnO2
154
155
156
157
158
159
160
161
MnO2 powder was synthesized through a facile direct redox reaction between KMnO4 and
Mn(AC)2·4H2O. Firstly, 0.02 mol KMnO4 and 0.03 mol Mn(AC)2·4H2O were separately dissolved
in two beakers that contained 100 mL distilled water. Then the Mn(AC)2·4H2O solution was moved
into water-bath and maintained at 60 °C under continuously stirring. Kept the solution stirring and
dropped the prepared KMnO4 solution. After the mixture solution maintained at 60 °C under
continuously stirring for 24 h, the pH of solution was adjusted to 7 and naturally cooled to room
temperature. Afterwards the solution was filtered and washed with deionized water until the filtrate
was colorless. The filtered brown powder was dried in an oven at 100 °C for 12 h, then grinded and
8

ACCEPTED MANUSCRIPT
162
163
marked as pure MnO2.
2.4 Materials characterization
164
165
166
167
168
169
170
171
172
173
174
175
X-ray diffraction (XRD) analysis was performed using a Rigaku-TTRIII X-ray diffractometer
(Japan) equipped with a graphite-monochromated high-intensity Cu Kα radiation source. Fourier
transmission infrared (FTIR) spectra were recorded on a Nicolet 6700 spectrometer over the
scanning range from 400 to 4000 cm−1 at a resolution 4 cm−1; the samples were compressed into KBr
discs. The morphologies of the products were investigated by field-emission scanning electron
microscopy (FESEM, FEI Nova NanoSEM 450) and high-resolution transmission electron
microscopy (HRTEM, JEOL JEM-2100F). Energy-dispersive X-ray (EDX) spectroscopy was
performed on a JEOL JEM-2100 microscope operated at 200 kV. X-ray photoelectron spectroscopy
(XPS) was performed on a RBD-upgraded PHI-5000C ESCA system (Perkin-Elmer) equipped with
an Al-Kα radiation source (hν = 1486.6 eV). The Brunauer-Emmett-Teller (BET) surface areas were
tested by nitrogen sorption experiments conducted at 77 K with a Micromeritics TriStar II 3020
automated gas adsorption analyzer.
176
2.5 Electrochemical measurements
177
178
179
180
181
The electrochemical properties of the as-synthesized samples were measured in both three- and
two-electrode cell systems with a 0.5 mol L-1 Na2SO4 aqueous solution as the electrolyte. To
fabricate the working electrodes, the prepared materials (70 wt.%) were mixed with acetylene black
(20 wt.%) and binder (polyvinylidene fluoride, PVDF, 10 wt.%) in an agate mortar. A slurry of the
aforementioned mixture was produced using N-methyl-2-pyrrolidone (NMP) as a solvent.
9

ACCEPTED MANUSCRIPT
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
In the three-electrode system, the slurry was brushed onto Ni foam with an effective area of ~1
cm2. The foam was then dried in air at 100 °C for 60 min to remove the solvent and uniaxially
pressed into a thin electrode approximately 0.5 mm thick under 10 MPa. After the prepared electrode
was completely dried, the active material mass of each electrode was typically 5 ± 0.5 mg.
For the two electrode system of a symmetric capacitor, the slurry, which was the same as that
previously described, was brushed onto a large piece of Ni foam current collector, then dried for 1 h,
and roll-pressed to a thickness of approximately 0.5 mm. After the Ni foam was completely dried,
the brushed Ni foam current collector was cut into small, round pieces with a diameter of 8 mm and
used as the electrodes. The mass of the active material of each electrode was typically 2.5 ± 0.5 mg.
Before the electrochemical tests, the prepared electrodes were soaked overnight in a 0.5 mol L-1
Na2SO4 solution. The electrochemical properties of the obtained materials were recorded at room
temperature using an electrochemical workstation (CHI 660B, Shanghai CH Instrument Company,
China) using a three-electrode electrochemical system with a 0.5 mol L-1 Na2SO4 aqueous solution as
the electrolyte. The fabricated MnO2 electrode was used as the working electrode. A platinum foil (2
× 2 cm2) and a saturated calomel electrode (SCE) were used as the counter and reference electrodes,
respectively. The ac-impedance spectra were measured with a 5 mV ac signal superimposed on the
open-circuit potential (approximately 0.4 V vs. SCE) over the frequency range from 10−2 to 105 Hz.
Cyclic voltammetry (CV) measurements were performed over the potential range from 0 to 0.8
V (vs. SCE) at a scan rate ranging from 2 to 200 mV s−1 in a three-electrode system. The specific
capacitance was calculated from the CV curves according to eq. (1):
ꢁ
ꢀ =
ꢆ
ꢇ ꢉ dꢊ
(1)
ꢂ∙ꢃ∙∆ꢄ ꢅ
ꢈ
10

ACCEPTED MANUSCRIPT
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
where m (g) is the mass of the active materials, v (mV s-1) is the potential scan rate, ꢀV (V) is the
sweep potential window, and I(V) (A) is the voltammetric current of the curves.
The galvanostatic charge/discharge tests were performed using a NEWARE auto-cycler
(Shenzhen Neware Electronics Co. Ltd., China) in a two-electrode configuration of symmetric
capacitors. The cell was constructed using a filter-paper separator and two prepared electrodes; it was
constructed as a sandwich-type cell (electrode/separator/electrode) in a 0.5 mol L-1 Na2SO4 solution
as the electrolyte. The specific capacitance for the cell could be calculated from the discharge curves
on the basis of eq. (2):
ꢋ∙∆ꢌ
ꢀ = ꢂ∙∆ꢄ
(2)
where I (A) is the applied current, ꢀt (s) is the discharge time, m (g) is the mass of the total active
material of the two electrodes, and ꢀV (V) is the discharging potential window.
The power density (P, kW kg−1) and energy density (E, Wh kg−1) of the symmetric capacitors in
the two-electrode configuration are calculated from the galvanostatic charge-discharge curves by
using the eq. (3) and eq. (4):
ꢑ
ꢍ = ꢎ.ꢏꢐꢇ∆ꢄꢉ
(3)
(4)
ꢒ.ꢓ
ꢔ = ꢕ
ꢌ
where ꢀV (V) is the applied potential window, t (s) is the discharge time.
220 3. Results and discussion
221
222
223
The crystal structure of the carbon nanotubes (CNTs) and the as-prepared MnO2/CNTs
composites (MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48) were analyzed via XRD. As
displayed in Fig. 1, for the CNTs, the prominent diffraction peak at approximately 25.0° is assigned
11

ACCEPTED MANUSCRIPT
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
to the (002) plane reflection of the conjugated graphitic system, indicating that the hexagonal
graphite structure of the CNTs is approximately intact. This diffraction peak is clearly detected in the
composite patterns as well. However, with increasing acid-treatment times of the CNTs, the intensity
of the (002) diffraction peak decreases, revealing a sustained etching of the CNT framework. In the
patterns of the as-prepared MnO2/CNTs composites, the broadened characteristic peaks appear at 2
θ
= 12.8°, 28.8°, 36.7°, 37.5° and 65.1°. On the basis of a comparison with the crystal data of different
MnO2 polymorphs reported in the literature [31], the MnO2/CNTs composites are a poorly
crystallized α-MnO2 material. This poor crystallinity originates mainly from the smaller particle size
of the MnO2 in the composite.
To further elucidate the structure and formation mechanism of the MnO2/CNTs, FTIR
spectroscopy was performed on the CNTs and composites, as shown in Fig. 2. The FTIR spectra of
pure CNTs without any treatment and the CNTs acid-treated at 120 °C for different times (C-12,
C-24, C-36 and C-48) are shown in Fig. 2a. The absorption peak at approximately 3400 cm−1 is
observed n in all of the spectra and is assigned to the stretching vibration of physically adsorbed
water molecules. As evident in Fig. 2a, the amount of structural water in the acid-treated CNTs
decreases with increasing treatment time, reaching a minimum in the case of a treatment time of 48 h.
In the wavenumber region from 400 to 2000 cm−1, three distinct absorption peaks are observed at
1627, 1561 and 1086 cm−1; these peaks are attributed to the carbonyl (C=O) stretching, the carboxyl
group (COO) anti-symmetric stretching, and the C-OH stretching [32], respectively, which indicates
that carboxyl groups exist on the CNTs’ surface after the acid treatment. Compared with the
intensities of these three absorption peaks in the spectrum of the CNTs without any treatment, those
12

ACCEPTED MANUSCRIPT
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
in the spectra of the acid-treated CNTs are more intense, indicating that the content of the carboxyl
groups on the CNTs’ surface increases dramatically after the hydrothermal acid treatment.
Furthermore, the amount of carboxyl groups on the CNTs’ surface increases when the acid-treatment
time increases from 12 h to 36 h. However, the intensities of the two peaks related to C=O and COO
decrease after 36 h of acid treatment, which we attributed to the fact that the carboxyl groups on the
CNTs’ surface are oxidized to CO2. We confirmed this interpretation by monitoring the variation of
the intensity of the absorption peak at 1086 cm−1. The peak intensity increases with increasing
acid-treatment time and reaches a maximum when the time is 48 h. This result is explained by the
carboxyl groups being further oxidized to CO2 when the acid-treatment time exceeded 36 h, where
the carbon layer underneath the carboxyl groups was re-exposed to the acid and formed newly
oxidized C-OH. Furthermore, the absorption peaks in the range from 500 to 800 cm−1 are assigned to
C-S stretching vibrations, which may be caused by a small amount of S impurities in the CNTs.
The CNTs acid treated for different durations were used for grafting; the FTIR spectra of the
resulting grafted CNTs are shown in Fig. 2b. The intensities of absorption peaks associated with the
carbonyl (C=O) stretching at 1629 cm-1 and the carboxyl group (COO) anti-symmetric stretching at
1569 cm-1 significantly increase when going from the C-12+Mn(Ac)2 to the C-36+Mn(Ac)2 material;
however, the peaks related to C=O and COO are less intense in the case of the C-48+Mn(Ac)2
material. Meanwhile, the peak intensity related to C-OH stretching at 1086 cm-1 increases from the
C-12+Mn(Ac)2 to the C-48+Mn(Ac)2 material. These trends are consistent with the variation in the
acid-treatment process. After grafting, two bands at 501 cm−1 and 598 cm−1 are evident in all of the
cases and are assigned to the Mn-O stretching vibrations [33, 34]. These results indicate that Mn2+ is
13

ACCEPTED MANUSCRIPT
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
successfully grafted onto the CNTs surface. Compared with the other grafted CNTs samples,
C-36+Mn(Ac)2 shows two high-intensity Mn-O peaks, indicating that the COO group plays an
important role in Mn2+ grafting onto the CNTs surface. In addition, the absorption peak at 1147 cm−1
is associated with C-O-C stretching. According to our previous discussion, the grafting process is
mainly related to the COO groups of the acid-treated CNTs surface. The grafting process was
maximized when the CNTs acid-treatment time reached 36 h; subsequent extension of the
acid-treatment time did not enhance the grafting process.
Fig. 2c shows the FTIR spectra of the MnO2/CNTs (MnO2/C-12, MnO2/C-24, MnO2/C-36 and
MnO2/C-48) prepared through oxidation of the grafted materials. After oxidation, the absorption
peaks associated with the carbonyl (C=O) stretching at 1629 cm-1 and the carboxyl group (COO)
anti-symmetric stretching at 1569 cm-1 decrease substantially, which contrasts the evident increase in
the absorption peak at 519 cm−1, which is attributed to the Mn-O stretching vibration of octahedral
[MnO6]. Compared with the spectra of the other composite samples, the spectrum of MnO2/C-36
exhibits a high absorption intensity, indicating a high MnO2 content in this sample. Clearly, this
results is attributable to the high content of Mn2+ in C-36+Mn(Ac)2 obtained through the grafting
process. The inset in Fig. 2c contrasts the bands from 400 to 800 cm−1 for the CNTs without any
treatment with the bands of the acid-treated CNTs, the Mn2+ grafted CNTs and the MnO2/CNTs. The
spectra clearly reveal the intensity changes in the Mn-O absorption peaks throughout the whole
grafting-oxidation process. When Mn2+ is grafted onto the CNTs and then oxidized to MnO2, the
intensity of the Mn-O absorption peak gradually increases. Furthermore, bonding between the CNTs
and MnO2 particles are believed to occur between the carbon in the carboxyl groups of the CNTs and
14

ACCEPTED MANUSCRIPT
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
the oxygen atoms from the MnO2 lattice. Hence, we inferred that the bonds between MnO2 and
CNTs exist in the form of C-O-Mn [56]. And he bonds between the CNTs and the MnO2 particles are
believed to form between the carbon of the carboxyl groups of the CNTs and the oxygen atoms of the
MnO2 lattice.
In summary, the amount of grafted Mn2+ is associated with amount of COO on the surface of the
CNTs, whereas the growth density of MnO2 is closely related to the content of Mn2+. Therefore, we
conclude that the amount and distribution of carboxyl groups determine the growth state of the MnO2
coating.
The XPS spectra obtained from the as-prepared MnO2/CNTs composites (MnO2/C-12,
MnO2/C-24, MnO2/C-36 and MnO2/C-48) are shown in Fig. 3. The XPS spectra were analyzed and
fitted using the Avantage XPS software. The C 1s region was adopted as a reference for surface
charging and was set as 284.9 eV. The manganese oxidation state was determined from the Mn 3s
and O 1s core level spectra. The procedure for analyzing the Mn 3s spectrum has been described
previously [34-66]. From the O 1s data, the average manganese oxidation state of the four composite
samples can be computed from the intensities of the Mn-O-Mn and Mn-OH components according to
eq. (3):
ꢇ
ꢉ
ꢇ
ꢉ
ꢛꢜꢝꢞꢠ
ꢖꢗ · State = ꢘꢙꢈ × ꢚ
(3)
ꢟꢚ
ꢡꢢ ꢙꢙꢙ × ꢚ
ꢛꢜꢝꢞꢝꢛꢜ
ꢛꢜꢝꢞꢠ
ꢚ
ꢛꢜꢝꢞꢝꢛꢜ
where S represents the signal of the different components of the O 1s spectrum.
Fig. 3a shows the XPS survey spectra of the MnO2/CNTs composites in which the Mn 2p (~642
eV), Mn 3s (~84 eV), and O 1s (~530 eV) peaks are attributed to MnO2. No Mn 2p3/2 signal
associated with KMnO4 (∼647eV) is observed [37], implying the absence of KMnO4 in these
15

ACCEPTED MANUSCRIPT
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
composites. The Mn 2p, Mn 3s, and O 1s core-level spectra were used to assess the oxidation state of
manganese in the MnO2/CNTs composites. Fig. 3b shows the Mn 2p core-level spectra of the
as-prepared MnO2/CNTs composites (MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48). Such
Mn 2p core-level spectra have been utilized to determine the various Mn species present in a sample
[37]. The binding energies of Mn 2p3/2 and Mn 2p1/2 are centered at 642.49 eV and 654.29 eV,
respectively, in good agreement with the binding energies of Mn 2p3/2 and Mn 2p1/2, as well as the
spin-orbit splitting energy value (∼11.8 eV) in MnO2 [40].
Fig. 3c depicts the Mn 3s core-level spectra for the as-prepared MnO2/CNTs composites
(MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48). Because of the parallel spin coupling of the
3s electron with the 3d electron during the photoelectron ejection, the Mn 3s core-level spectra
generally show peak splitting as a doublet [34, 38]; thus, the energy separation between the two
peaks is related to the mean manganese oxidation state. Because a lower valence implies more
electrons in the 3d orbital, additional interaction can occur upon photoelectron ejection.
Consequently, the energy separation of the Mn 3s spectrum will enlarge as the Mn valence is reduced,
and the inverse trend will be observed when the manganese valence increases. As listed in Table 1,
the relevant data of the Mn 3s spectra for the MnO2/CNTs reveal that the peak splitting of the doublet
of the Mn 3s core-level spectra differs slightly among all of the composites. The separation of peak
energies (∆Eb) of the Mn 3s components for MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48
are 4.75, 4.73, 4.69 and 4.71 eV, respectively. The values of the energy separation between the two
peaks have been reported to be 5.79, 5.50, 5.41, and 4.78 eV for the reference samples of MnO,
Mn3O4, Mn2O3, and MnO2, respectively [34]. Therefore, we conclude that the Mn in MnO2/C-36 has
16

ACCEPTED MANUSCRIPT
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
the oxidation state of manganese proposed in the literature [34-36], the mean oxidation state of
manganese in the composites is estimated to be approximately 4.0.
The O 1s core-level spectrum was also used to confirm the oxidation state of manganese in the
composites. Fig. 3d shows the slight difference in the O 1s envelope among MnO2/C-12, MnO2/C-24,
MnO2/C-36 and MnO2/C-48. To obtain the chemical composition, we analyzed the O 1s spectra
through curve fitting, as shown in Fig. 3d (relevant data in Table 1). The spectra were fitted with
three components related to the Mn-O-Mn bond for the tetravalent oxide, the Mn-OH bond for a
hydrated trivalent oxide and the H-O-H bond for residual structure water [37, 39, 40]. Table 1
presents the mean manganese oxidation states calculated according to eq. (3). With respect to all of
the as-prepared composites, the intermediate oxidation states of 3.56, 3.52, 3.66 and 3.56 are
determined for MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48, respectively; furthermore,
MnO2/C-36 again shows the highest oxidation state. These values are roughly in agreement with
those oxidation states computed from the Mn 3s data. This result indicates an intermediate oxidation
state between Mn4+ and Mn3+, which may be helpful for charge storage in the Faradaic reactions of
MnO2 [41].
Fig. 4 shows FESEM images of the CNTs and the as-prepared MnO2/CNTs composites
(MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48). As presented in Fig. 4a, the CNTs have an
average diameter of 50 nm and lengths that reach hundreds of microns. The CNTs exhibit a curled
morphology with rough and uneven surfaces. The FESEM images of the MnO2/CNTs in Fig. 4b-4e
reveal that the surfaces of the CNTs are coated with MnO2 in all of the cases and that the tubular
morphology of the CNTs is essentially maintained. As evident in the case of MnO2/C-12 in Fig. 4b, a
17

ACCEPTED MANUSCRIPT
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
thin layer of MnO2 was grown on the CNTs’ surface; however, the MnO2 layer is loose, and its
thickness is unevenly distributed. Fig. 4c shows that in MnO2/C-24 the layer of MnO2 becomes fine
and tightly concentrated on the long nanotubes; however, the MnO2 on the short nanotubes remain
loose. Comparatively speaking, the MnO2 layer of MnO2/C-36 in Fig. 4d appears to be more
homogenous. As evident from MnO2/C-48, a piece of the loose MnO2 layer whose thickness is
approximately 25 nm fell off, and the bare CNTs are observable. Carbonaceous materials are usually
added to MnO2 to improve its electrical conductivity [42]. Therefore, the contact degree between the
MnO2 and the added CNTs determines the electrochemical performance of the composites. Among
all of the samples, the MnO2/C-36 composite is expected to exhibit the greatest improvement in
electrochemical performance.
Fig. 5 shows the HRTEM images of the as-prepared MnO2/CNTs composites (MnO2/C-12,
MnO2/C-24, MnO2/C-36 and MnO2/C-48). The CNTs are coated with a MnO2 nanoparticle layer in
all cases. In the case of MnO2/C-12 in Fig. 5a, a layer of relatively loose and uneven MnO2
agglomerations wrap around the CNTs surface, and a clear dividing line between the coating layer
and the CNTs is observed. In the case of the MnO2/C-24 (Fig. 5b) and MnO2/C-36 (Fig. 5c) samples,
the MnO2 layers interact more closely with the CNTs. The MnO2 layer of MnO2/C-36 is the most
uniform and fine among the four composites. The insets of the HRTEM images show that the
interplanar distance is approximately 0.25 nm, which agrees well with the separation between the
[211] planes of α-MnO2.
From previous studies, KMnO4 is known to react with exposed carbon and grafted Mn2+
simultaneously. Compared with the particle size of MnO2 prepared through the reaction between
18

ACCEPTED MANUSCRIPT
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
KMnO4 and carbon, the MnO2 synthesized through the reaction between KMnO4 and the grafted
Mn2+ is much finer. Moreover, the tightly combined CNTs and MnO2 nanoparticle layer contribute to
the improvement in electrochemical performance. The difference in the morphologies of the
composites can be explained by the mechanism of the acid-treatment and grafting process. The FTIR
results indicate that the amount of grafted Mn2+ reaches a maximum when the acid-treatment time is
36 h and that the grafting degree of Mn2+ is closely related to the carboxyl-group content on the
CNTs’ surface. However, in the acid-treatment process, the carboxyl groups may mainly grow at
highly active sites (e.g., the edges of holes and the tips of the CNTs) during the initial stage. As the
acid-treatment time increases, the carboxyl groups gradually and uniformly grow on the surface of
the CNTs. However, when the acid-treatment time exceeds 36 h, some of the carboxyl groups are
further oxidized to CO or CO2, resulting in the CNTs surface being re-exposed and the uniform
coverage of the carboxyl groups being destroyed. The distribution of the carboxyl groups must
influence the distribution of Mn2+ on the surface of the CNTs during the grafting process, as well as
the reactions with KMnO4, the exposed carbon and the grafted Mn2+. This distribution leads to
different morphologies and thicknesses of the MnO2 coating on the surface of the CNTs in the four
composites. Accordingly, the closely packed MnO2 layers of MnO2/C-36 are the most uniform and
fine, which is mainly due to the uniform distribution of carboxyl groups and the major reaction
between KMnO4 and the grafted Mn2+.
The four samples (MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48) were analyzed by
EDX to determine their Mn and C contents; the results are shown in Fig. 6. Clearly resolved peaks
associated with Mn, C, O and K are observed in the spectra of the prepared materials. The elemental
19

ACCEPTED MANUSCRIPT
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
analysis results (wt.%) for the ratio of Mn and C are presented to the right of each spectrum. As
evident in Fig. 6, the content of Mn generally increases when going from MnO2/C-12 to MnO2/C-36
but decreases in the case of the MnO2/C-48 composite. These observations are consistent with the
variations in the Mn-O content observed from the FTIR spectra. The weight percentage ratio of
MnO2 in the MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48 composites are 64.85 wt.%,
69.79 wt.%, 73.56 wt.% and 67.76 wt.%, respectively. And the weight percentage ratio of CNTs in
the four composites are 35.15 wt.%, 30.21 wt.%, 26.44 wt.% and 32.24 wt.%, respectively.
Fig. 7 shows the nitrogen adsorption isotherms for the CNTs and the as-prepared MnO2/CNTs
composites (MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48). The Brunauer-Emmett-Teller
(BET) surface areas of the CNTs, MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48 are 145,
119, 114, 107 and 117 m2 g-1, respectively. The BET surface areas of the MnO2/CNTs composites
are substantially smaller than that of the CNTs. The BET surface area decreases going from
MnO2/C-12 to MnO2/C-36 but increases for the MnO2/C-48 composite, which are contrary to the
trend of the Mn content determined by EDX. This difference is explained by the fact that the grafting
occurs primarily at the entrance of ultramicropores and micropores [61]. Consequently, access to
these pores is substantially reduced for nitrogen and carbon dioxide molecules during the gas
adsorption measurements. Therefore, a further increase in the concentration of grafted Mn2+ leads to
a monotonic decrease of the BET surface area, which is contrary to the variation in the FTIR results
of the Mn2+ grafting. In addition, we note that the surface area of the MnO2 obtained by the
traditional co-precipitation method and the hydrothermal method are 17 m2 g-1 [31] and 45 m2 g-1
[43], respectively. Thus, the surface areas of the MnO2/CNTs composites via the grafting method are
20

ACCEPTED MANUSCRIPT
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
substantially higher.
Fig. 8 shows representative CV curves of the CNTs, the pure MnO2 and the as-prepared
MnO2/CNTs composites (MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48) at a scan rate of 5
mV s-1 in the potential window of 0-0.8 V, using a three-electrode electrochemical system and a 0.5
mol L-1 Na2SO4 solution as the electrolyte. The CV curves are quasi-rectangular and symmetric in
shape, indicating a rapidly reversible Faradaic reaction and ideal capacitive behavior of the
composite electrodes. At a scan rate of 5 mV s-1, the current densities of the as-prepared MnO2/CNTs
composite electrodes are substantially larger than that of the CNTs and pure MnO2 electrode. The
electrode specific capacitance was calculated from the area enclosed by the CV curves through eq.
(1). On the basis of the CV curves, the calculated specific capacitance of the CNTs is no greater than
3 F g-1. And the specific capacitance of the CNTs is 166 F g-1. Furthermore, the CNTs surface of the
MnO2/CNTs composites is uniformly coated with a MnO2 layer; thus, it rarely participated in the
charge-storage process. As a result, we concluded that the capacitance of the CNTs is negligible for
all of the composite electrodes. The calculated results show that the specific capacitances of
MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48 are 195, 292, 348 and 324 F g-1, respectively.
The higher capacitance of MnO2/C-36 can be attributed to the fact that the particle size of the MnO2
active materials in the MnO2/C-36 composites is uniform and fine, which could increase the
probability of conducting pathways being formed between the electrode and the electrolyte at their
interface. These pathways facilitate the electrochemical reaction and enhance the capacitance.
Compared with the composites reported in the previous studies [25-27], the obtained MnO2/C-36
also exhibits a much higher capacitance, which may be ascribed to the tightly contact between the
21

ACCEPTED MANUSCRIPT
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
CNTs and MnO2 in the composite.
Fig. 9 shows the specific capacitances of the CNTs, the pure MnO2 and the as-prepared
MnO2/CNTs composites (MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48) at different CV
scan rates (2, 5, 10, 25, 50, 100, 150, and 200 mV s-1) in the potential window of 0-0.8 V, using a
three-electrode electrochemical system and a 0.5 mol L-1 Na2SO4 solution as the electrolyte. The
specific capacitances of the pure MnO2 and the four composites decrease with increasing scan rate.
And there is no change in the specific capacitances of the CNTs when the scan rate increases from 2
to 200 mV s-1. Notably, the MnO2/C-36 composite electrode exhibits both the best rate performance
and a specific capacitance significantly greater than the other materials for the same scan rate. In
contrast to the MnO2/C-24 electrode, the MnO2/C-48 electrode presents a better rate capability at
scan rates of 2 mV s-1 and 5 mV s-1. However, the MnO2/C-24 electrode presents a better rate
capability when the scan rate is increased from 10 mV s-1 to 200 mV s-1. And the rate performance
and specific capacitance of the MnO2/CNTs composites are obviously better than those of the pure
MnO2.
The charge-storage mechanism of MnO2 is known to be mainly based on a surface process that
includes the adsorption/desorption and insertion/extraction of proton and alkali cations [3, 44, 45].
Clearly, the diffusion of the cations (e.g., Na+) on the MnO2 surface and in its bulk and the transfer of
electrons strongly influence the rate capability. A higher scan rate results in a smaller available
capacitance because of the reduced diffusion time and the low conductivity of MnO2 [44]. Therefore,
the contact between MnO2 and the CNTs dramatically influences the electrochemical properties of
the composites. The fine MnO2 particles synthesized by the grafting oxidation method encase the
22

ACCEPTED MANUSCRIPT
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
CNTs closely, which enhances the electrical conductivity of the MnO2/CNTs composites and
enhances the charge-transfer rate. Remarkably, the MnO2/C-36 composite exhibits the best rate
performance and the highest capacitance, which is attributed to the most uniform distribution and
tight coating of the nanostructured MnO2.
The electrochemical performance of the as-prepared MnO2/CNTs composites (MnO2/C-12,
MnO2/C-24, MnO2/C-36 and MnO2/C-48) was further evaluated by electrochemical impedance
spectroscopy (EIS). The impedance spectra were measured using a three-electrode electrochemical
system with a 0.5 mol L-1 Na2SO4 aqueous solution as the electrolyte. The amplitude of the ac signal
superimposed on the open-circuit potential (approximately 0.4 V vs. SCE) was 5 mV and the
frequency range was from 10−2 to 105 Hz, the Nyquist plots are shown in Fig. 10. It can be seen that
all of the collected impedance spectra consist of depressed arcs at high frequencies, straight lines
inclined at a constant phase angle to the real axis (Warburg region) at intermediate frequencies, and
approximately vertical capacitive lines at low frequencies. The arc is mainly caused by the
charge-transfer reaction at the electrode/electrolyte interface [46]. During the charging/discharging
process, the cations including Na+ and H+ participate in the redox reaction [47]. In addition, the
Warburg impedance is associated with the semi-infinite diffusion of cations in the electrode, and the
capacitive line is related to the accumulation of cations at the impermeable interface under the finite
diffusion condition.
To determine the values of the resistance and capacitance of the oxide electrodes, the measured
impedance spectra were fitted by the complex nonlinear least square (CNLS) method [48] on the
basis of the equivalent circuit shown in Fig. 10. Here, Rs is the solution resistance, Rct corresponds to
23

ACCEPTED MANUSCRIPT
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
the charge-transfer resistance in the pseudocapacitive process, and CPE1 arises from the interfacial
double-layer charging. The Warburg impedance, Zw, is associated with the cation diffusion in the
electrode, and CPE2 represents the non-ideal capacitive behavior at the impermeable interface. The
fitting results for the as-prepared MnO2/CNTs composites (MnO2/C-12, MnO2/C-24, MnO2/C-36 and
MnO2/C-48), including the solution resistance (Rs), the charge-transfer resistance (Rct), the diffusion
resistance (W-R), the diffusion time constant (W-T), the double-layer capacitance (Cdl) and the
pseudocapacitance (Cps), are listed in Table 2.
Remarkably, the charge-transfer resistances of the MnO2/C-12, MnO2/C-24, MnO2/C-36 and
MnO2/C-48 electrodes are only 1.82, 0.73, 0.59 and 1.33 ꢁ, respectively, which are much lower than
that of the MnO2 (4.5 ꢁ) synthesized by the traditional liquid-phase method. Apparently, the rapid
charge-transfer rate of the MnO /CNTs electrode is ascribed to the CNTs electronic structure, which
2
plays a critical role in promoting the energy storage capacity. MnO2/C-36 exhibits a much lower
reaction resistance (Rct), which is attributed to the intimate coating of the CNTs surface by MnO2 to
improve the electrical conductivity. By contrast, the W-R and W-T of MnO2/C-36 are the lowest. The
W-T for finite diffusion is described as W-T = L2 / D [49] (where L is the diffusion depth of cations in
the electrode and D corresponds to the chemical diffusivity). The MnO2 layer of the MnO2/C-36
composite is thinner than the others; therefore, the cations take less time to diffuse through the oxide
layer. According to the Cps values, MnO2/C-36 exhibits a much higher specific capacitance, which is
consistent with the results of the CV tests as well.
To evaluate the cyclic stability of the as-prepared materials, galvanostatic charge-discharge tests
were performed on a two-electrode symmetric supercapacitor system, which were assembled using
24

ACCEPTED MANUSCRIPT
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
the as-prepared MnO2/CNTs composites as the electrode material and a 0.5 mol L-1 Na2SO4 solution
as the electrolyte. The charge-discharge current density was 1 A g-1 with the potential window range
from 0 to 0.8 V. As presented in Fig. 11a, the MnO2/CNTs electrodes show well-shaped triangular
charge/discharge curves with a slight curvature, which originates from the co-contribution of the
pseudocapacitance and double-layer capacitance of MnO2 material. From the discharge curves, the
specific capacitances of MnO2/C-12, MnO2/C-24, MnO2/C-36 and MnO2/C-48 are calculated using
eq. (2) as 192, 283, 331 and 305 F g-1, respectively. MnO2/C-36, with the highest specific
capacitance, is considered an effective material for high-performance electrochemical capacitors.
Figure 11b shows the galvanostatic charge-discharge curves of the MnO2/C-36 at different current
density (1, 2, 5, 10, and 20 A g−1). According to the discharging time, the specific capacitances are
calculated to be 331, 309, 279, 247, and 216 F g−1 at the current density of 1, 2, 5, 10, and 20 A g−1,
respectively. The specific capacitance at 20 A g−1 is about 65% of that at 1 A g−1, which demonstrates
a very high rate capability of the MnO2/C-36 composite.
We also performed long-term charge/discharge cycling tests over 5000 cycles at a current
density of 1 A g-1 (Fig. 12) for these two-electrodes symmetric supercapacitors. In the first 500 cycles,
the capacitances of the four electrodes decrease quickly; however, after 500 cycles, the capacitances
tend to be stable, without any obvious decay. The MnO2/C-36 electrode exhibits only an 8.7% decay
after 5000 cycles, evidently surpassing the performances of electrodes based on MnO
(49.0% decay after 5000 cycles), MnO -active carbon (8.7% decay after 1000 cycles) [35],
MnO -polyaniline (8.6% decay after 1000 cycles) [50], MnO -CNTs textile (20% decay after 200
cycles) [51], MnO -graphene oxide (15.9% decay after 1000 cycles) [52] and MnO -graphene (17.6%
2
nanowires
2
2
2
2
2
25

ACCEPTED MANUSCRIPT
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
decay after 1000 cycles) [47]. Over 5000 charge-discharge cycles, the decay rates of MnO2/C-12,
MnO2/C-24 and MnO2/C-48 were 20.9%, 13.4%, and 29.6%, respectively. In conclusion, the
excellent cyclic performance of MnO2/C-36 suggests that the compact layer of MnO2 in this
composite material could effectively prevent the degradation of active MnO2 nanoparticles.
Impedance measurements were also performed on the two-electrode symmetric supercapacitor
system assembled with the MnO2/C-36 composite electrodes before and after 5000 charge-discharge
cycles. As the Nyquist plots displayed in Fig.13a, only a marginal increase in the Rct value (0.76 ꢁ)
of the electrode after cycling demonstrates a very high electrochemical stability of the MnO2/C-36
composite. The Ragone plot of the MnO2/C-36 composite material is shown in Fig. 13b. According,
the energy density of 19.2 Wh kg−1 with the max power density of 6.4 kW kg−1 are achieved using
the two-electrode symmetric systems at operating voltages of 0.8 V, which are calculated by eqs 2-4.
As compared with the reported previously using the two-electrode symmetric systems of AC//AC [53,
54], CNTs//CNTs [55] and MnO2//MnO2 [53], this symmetric systems exhibits high energy density.
These results indicate that the MnO2/C-36 composite is a very promising electrode material with
both high energy and good electrochemical stability.
533 4. Conclusion
534
535
536
537
538
A unique nanomaterial composed of CNTs grafted with amorphous MnO2 was synthesized for
use in supercapacitors. The bond between the CNTs and MnO2 particles is believed to occur between
the carbon of the carboxyl groups of CNTs and the oxygen atoms of the MnO2 lattice. The high
capacitance of MnO2 as well as the excellent cycling stability and superior electrical conductivity of
the CNTs were exploited in the composite. In addition, MnO2 layers with different thicknesses
26

ACCEPTED MANUSCRIPT
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
coating the surface of the CNTs were prepared through controlling the hydrothermal acid-treatment
time (12, 24, 36 or 48 h). We observed that the MnO2 layer thickness substantially affected the
capacitance, cycling performance and conductivity of the obtained composites. As a consequence,
the MnO2/CNTs composite subjected to acid treatment for 36 h (MnO2/C-36) exhibited a maximum
specific capacitance of 348 F g-1 at a scanning rate of 5 mV s-1. Additionally, the quasi-rectangular
and symmetric CV curves of MnO2/C-36 reflected its ability to undergo a rapidly reversible Faraday
reaction and its good electrochemical behavior. A two-electrode symmetric laboratory supercapacitor,
which was assembled with the MnO2/C-36 composite electrodes and tested through galvanostatic
charge-discharge cycling, showed a specific capacitance of 331 F g-1 at a current density of 1 A g-1,
with an attenuation rate of just 8.7% after 5000 charge-discharge cycles. Furthermore, the impedance
test results revealed that the charge-transfer resistance of MnO2/C-36 composite is only 0.59 ꢁ. The
as-assembled symmetric MnO2/C-36//MnO2/C-36 electrodes exhibit a high energy densities of 29.4
Wh kg−1 with the max power densities of 6.4 kW kg−1, indicating that such a material has great
potential for use in high-performance energy-storage systems. These results are encouraging for
furthering the application of MnO2-based electrodes in high-power-capability devices.
554 Acknowledgements
555
556
557
The authors acknowledge financial support from the National Natural Science Foundation of
China (Grant No. 51374252) and the Natural Science Foundation of Hunan province, China (Grant
No. 12jj3014).
27

ACCEPTED MANUSCRIPT
558 References
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
[1]
B.E. Conway, Electrochemical supercapacitors: scientific fundamentals and technological
applications (POD), Kluwer Academic/Plenum, New York, 1999.
[2] J. R. Miller, P. Simon, Electrochemical capacitors for energy management, Science 321 (2008)
651-652.
[3]
[4]
[5]
W. Wei, X. Cui, W. Chen, D.G. Ivey, Manganese oxide-based materials as electrochemical
supercapacitor electrodes, Chem. Soc. Rev. 40 (2011) 1697–1721.
J.P. Zheng, Ruthenium oxide-carbon composite electrodes for electrochemical capacitors,
Electrochem. Solid–State Lett. 2 (1999) 359-361.
K. Jang, S. Yu, S.H. Park, H.S. Kim, H. Ahn, Intense pulsed light-assisted facile and agile
fabrication of cobalt oxide/nickel cobaltite nanoflakes on nickel-foam for high performance
supercapacitor applications, J. Alloys Compd. 618 (2015) 227–232.
[6]
[7]
A. Ramadoss, S.J. Kim, Vertically aligned TiO2 nanorod arrays for electrochemical
supercapacitor, J. Alloys Compd. 561 (2013) 262-267.
Z.J. Lao, K. Konstantinov, Y. Tournaire, S.H. Ng, G.X. Wang, H.K. Liu, Synthesis of
vanadium pentoxide powders with enhanced surface-area for electrochemical capacitors, J.
Power Sources 162 (2006) 1451-1454.
[8]
[9]
M. Sarfraz, M.F.A. Aboud, I. Shakir, Molybdenum oxide nanowires based supercapacitors
with enhanced capacitance and energy density in ethylammonium nitrate electrolyte, J.
Alloys Compd. 650 (2015) 123-126.
T. Brousse, P.L. Taberna, O. Crosnier, R. Dugas, P. Guillemet, Y. Scudeller, Y. Zhou, F. Favier,
28