Facile synthesis and electrochemical properties of Co-S composites as negative materials for alkaline rechargeable batteries

Article abstract of DOI:10.1016/j.electacta.2010.10.029

A series of novel Co-S composites composed of metallic Co and Co ₉S₈ were prepared via a facile hydrothermal method and investigated as negative electrodes for secondary alkaline batteries. Instrumental analyses reveal that the incorporation of Co₉S ₈ nanoflakes leads to better dispersion of Co particles and increases the interspacing between Co particles, greatly increasing the Brunauer-Emmett-Teller (BET) surface area. Thus, the electrochemical performance of the Co electrode is significantly improved. The maximum discharge capacity of the Co-S electrode reaches 420 mAh g^-1 and remains at 410 mAh g^-1 after 200 cycles, which is much higher than the capacity of a pure Co electrode. The shift of the redox peaks in the CV curves and the negative movement of the discharge-potential plateau are attributed to the dissolution of sulfur in the composite, which also favors the capacity of Co. The measurements reveal that the reversible faradic reaction between highly dispersed Co and Co(OH)₂ is dominant for the Co-S composites.

Full text of DOI:10.1016/j.electacta.2010.10.029

Electrochimica Acta 56 (2011) 1106–1110
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Facile synthesis and electrochemical properties of Co–S composites as negative
materials for alkaline rechargeable batteries
Qinghong Wang, Lifang Jiao∗, Hongmei Du, Wenxiu Peng, Dawei Song, Yijing Wang, Huatang Yuan
Institute of New Energy Material Chemistry, Key Laboratory of Advanced Energy Materials Chemistry (MOE), MOE (IRT-0927), Nankai University, Tianjin 300071, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of novel Co–S composites composed of metallic Co and Co9S8 were prepared via a facile
hydrothermal method and investigated as negative electrodes for secondary alkaline batteries. Instru-
mental analyses reveal that the incorporation of Co9S8 nanoflakes leads to better dispersion of Co particles
and increases the interspacing between Co particles, greatly increasing the Brunauer–Emmett–Teller
(BET) surface area. Thus, the electrochemical performance of the Co electrode is significantly improved.
The maximum discharge capacity of the Co–S electrode reaches 420 mAh g−1 and remains at 410 mAh g−1
after 200 cycles, which is much higher than the capacity of a pure Co electrode. The shift of the redox
peaks in the CV curves and the negative movement of the discharge–potential plateau are attributed to
the dissolution of sulfur in the composite, which also favors the capacity of Co. The measurements reveal
that the reversible faradic reaction between highly dispersed Co and Co(OH)2 is dominant for the Co–S
composites.
Received 10 September 2010
Received in revised form 11 October 2010
Accepted 11 October 2010
Available online 12 November 2010
Keywords:
Negative electrode
Co–S composite
Hydrothermal method
Electrochemical performance
Faradaic redox
© 2010 Elsevier Ltd. All rights reserved.
1. Introduction
bility [18]. Sulfur seems to be an ideal metalloid element for the
improvement of the electrochemical performances of Co–X and
Alkaline rechargeable batteries are one of the most impor-
tant power sources in use today. Among the developed alkaline
rechargeable Ni-based batteries, such as nickel/cadmium (Ni/Cd),
nickel/metal hydride (Ni/MH), nickel/iron (Ni/Fe), and nickel/zinc
(Ni/Zn) batteries, the Ni/MH battery is the most promising because
of its high energy density, environmental friendly properties and
high power. However, traditional hydrogen storage alloys, such as
AB5- [1,2], AB2- [3], and AB-type [4] and Mg-based alloys [5–7]
struggle to meet the ever-increasing demands of home appliances,
electric vehicles, computers and other power tools. Therefore, it
is imperative to develop new rechargeable battery materials with
higher energy density, better cycling stability and higher rate capa-
bility.
In recent years, new types of Co–X (X = B, Si, P, S, BN, Si3N4)
materials have been reported to have a highly reversible discharge
capacity and a good cycle lifetime when acting as negative materi-
als for alkaline rechargeable batteries [8–15]. Song et al. prepared
S–Co(OH)2 and used it as a negative material for alkaline recharge-
able batteries, obtaining a high discharge capacity [16]. Lu et al.
synthesized Co–B–S composites and reported good electrochem-
ical properties [17]. In our previous work, we prepared a Co–S
alloy and obtained high discharge capacities and good cycling sta-
Co(OH)2 materials because it is effective, cheap and easily pre-
pared.
Furthermore, a new alkaline rechargeable Ni/Co battery system
using ␣-Ni(OH)2 as the positive electrode material and ␤-Co(OH)2
as the negative electrode material has been proposed by Gao et al.,
who reported that the conversion between ␤-Co(OH)2 and metallic
Co is dominant in the faradaic reaction process of the ␤-Co(OH)2
electrode [19]. The operating principle is similar to that of Ni/Cd
batteries. A 6-M KOH aqueous solution is used as the electrolyte
solution. In hermetically sealed starved electrolyte cells, water
formation during charging at the positive electrodes may cause
electrolyte dilution. In practice, little change occurs in the amount
of water, especially in vented/flooded electrolyte cells [20].
In this paper, a series of Co–S composites are prepared via a facile
hydrothermal method and greatly improve the discharge capacity
and cycle life of a cobalt electrode. Further investigations reveal
that the reaction mechanism of the Co–S electrode is the same as
that of the Ni/Co battery system (Fig. 1).
2. Experimental
2.1. Material synthesis
All chemicals were of analytical grade and were used as
purchased without further purification. Co–S composites with dif-
ferent sulfur contents were prepared via a facile hydrothermal
∗
Corresponding author. Tel.: +86 022 23504527; fax: +86 022 23502604.
E-mail address: jiaolf@nankai.edu.cn (L. Jiao).
0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.10.029

Q. Wang et al. / Electrochimica Acta 56 (2011) 1106–1110
1107
and reference electrodes, respectively. The electrolyte solution was
a 6-M KOH aqueous solution.
The cycle life and charge–discharge curves were tested by
a LAND battery-test instrument (CT2001A). The electrodes were
charged at 200 mA g−1 for 3 h and discharged at 100 mA g−1 up to
the cut-off voltage, which was set at −0.5 V (vs. Hg/HgO). The inter-
val between the charge and discharge was 5 min. The weights of
the Co–S composites were calculated as the active materials. Cyclic
voltammetry (CV) was conducted by a Zahner IM6e electrochem-
ical workstation. The scan rate was 0.2 mV s−1 and the potential
interval was −1.2 V to −0.4 V (vs. Hg/HgO). All the tests were per-
formed at room temperature.
3. Results and discussion
3.1. Chemical composition, specific surface area and structure
The ICP-AES results shown in Table 1 reveal that the chemi-
cal compositions (atomic ratios) of the S1, S2, S3 and S4 samples
are Co10.05S, Co6.08S, Co3.68S and Co2.10S, respectively. The cor-
Fig. 1. XRD patterns of the Co–S samples.
responding specific surface areas are 21.65 m2 g−1, 38.47 m2 g−1
,
40.85 m2 g−1 and 45.01 m2 g−1, which are much higher than that of
pure Co (12.32 m2 g−1). The specific surface area is clearly improved
by the introduction of sulfur and is dependent upon the sulfur con-
tent.
method. In a typical synthesis, 0.9 g of CoCl2·6H2O was dissolved in
20 mL of distilled water, then 0.57 g NaOH and the desired amount
of Na2S2O3·2H2O were added. The resulting mixture was stirred
for 10 min. Next, 15 mL of hydrazine monohydrate (80% v/v) was
added to the well-stirred mixture by simultaneous agitation. The
resulting solution was transferred into a 50-mL Teflon-lined auto-
clave. The autoclave was sealed, maintained at 140 ◦C for 12 h and
then allowed to cool to room temperature. The black precipitate
was collected and washed three times each with water and abso-
lute ethanol. The sample was dried in a vacuum at 50 ◦C for 6 h.
The masses of the Na2S2O3·2H2O additives were 0.1 g, 0.15 g, 0.25 g
and 0.5 g, and the corresponding samples were designated as S1,
S2, S3 and S4, respectively. For comparison, the sample S0 was
prepared under the same conditions but without the addition of
Na2S2O3·2H2O.
a
Co 2p
2.2. Characterization
The crystal structure and micromorphology of the samples were
characterized by X-ray diffraction (XRD, Rigaku D/Max-2500, Cu K␣
radiation), scanning electron microscopy (SEM, Hitachi X-650) and
transmission electron microscopy (TEM, Tecnai 20). The elemen-
tal composition was determined by inductively coupled plasma
atomic emission spectroscopy (ICP-AES) using a USA Themo Jarrel-
Ash Corp. instrument. The specific surface area was tested by a BET
method using a ST-08A specific-surface-area analyzer. The elemen-
tal composition of the product was detected by energy dispersive
spectroscopy (ISIS300, Oxford Instrument). The electronic states of
the elements of the prepared samples were analyzed by X-ray pho-
toelectron spectroscopy (XPS, Kratos Axis Ultra DLD using Al Ka
radiation). All the binding energy values were calibrated using C 1s
as a reference.
775
780
785
790
795
800
805
Binding energy (eV)
b
S 2p
2.3. Electrochemical measurements
The Co–S electrode was constructed by mixing the prepared
composite with carbonyl nickel powders in a weight ratio of 1:3.
The powder mixture was pressed under 30 MPa of pressure into a
small pellet of 10-mm diameter and 1.5-mm thickness. The pellet
was then pressed under 20 MPa of pressure between two pieces of
nickel foam.
158
160
162
164
166
168
170
172
Electrochemical measurements were conducted in a three-
compartment cell using the Co–S electrode as the working
electrode. NiOOH/Ni(OH)2 and Hg/HgO were used as the counter
Binding energy (eV)
Fig. 2. XPS spectra of sample S2: (a) Co 2p, (b) S 2p.

1108
Q. Wang et al. / Electrochimica Acta 56 (2011) 1106–1110
Table 1
The S 2p peak appears near 162 eV, corresponding to the binding
ICP-AES results of the Co–S composites.
energies of Co–S [22]. In accordance with these analyses, it can be
concluded that the prepared Co–S samples are a mixture of Co and
Co9S8.
Samples
Co% (mass%)
S% (mass%)
Composition
Co10.05
S1
S2
S3
S4
94.87
91.80
87.15
79.45
5.13
8.20
12.85
20.55
S
Co6.08
S
Co3.68S
Co2.10S
3.3. Morphology
Typical SEM and TEM images of pure Co and sample S2 are
illustrated in Fig. 3. The pure Co has dendritic shapes with ran-
domly distributed irregular particles. As shown in Fig. 3b, sample
S2 is composed of ball-like particles surrounded by nanoflakes. The
size and microstructure of the sample were further examined with
TEM. As shown in Fig. 3c and d, the samples dispersed on the TEM
grids have ball-like particles with a diameter of 200 nm, coated by
nanoflakes, which increase the BET surface area.
To further confirm the composition of the product, EDX analysis
of S2 was performed. As shown in Fig. 4, the sample has coexisting
Co and S, but the ratio of Co and S differs between the bulk and
the ball-like particles. The results reveal that the composition of
the bulk is Co6.01S, which is consistent with the results obtained by
ICP. The composition of the ball-like particles is Co19S, demonstrat-
ing that the main composition of the ball-like particles is metallic
Co. Therefore, the sample is composed of metallic Co particles and
Co9S8 nanoflakes. The small amount of S detected in the particles
can be attributed to the Co9S8 nanoflakes attached to the surface.
Based on the above analysis, it is clear that the incorporation of
Co9S8 nanoflakes destroys the dendritic structure of Co but leads
to better dispersion of Co particles and increases the interspacing
between the Co particles, greatly increasing the BET surface area.
The phase structures of the samples were examined by XRD.
The XRD pattern of S0 shows that all sharp diffraction peaks can
be indexed to the hcp phase of Co, indicating high purity and good
crystallinity of the product. It is clear that the intensity of the Co
diffraction peaks gradually weakens with increasing sulfur content.
Meanwhile, the diffraction peaks of Co9S8 can be clearly observed
for S2, S3 and S4, illustrating that the incorporation of sulfur can
decrease the crystallinity of Co and introduce a Co9S8 phase. It
seems that the (1 0 1) peaks of Co gradually broaden from S1 to
S4. That is because Co9S8 also displays a peak at about 47.5◦, which
is very close to the (1 0 1) peak of Co. The crystallizability of Co9S8 is
not so well and the peaks of Co weaken from S1 to S4. So the peaks
present at 47.5◦ become weak and broad from S1 to S4.
3.2. Chemical state
To investigate the valence state of the elements, the XPS spec-
tra of the Co 2p and S 2p levels of S2 were recorded. As shown in
Fig. 2a, the Co 2p peak appearing at 778.0 eV can be attributed to Co
metallic state. Additionally, the Co 2p peak at 780.9 eV is present,
indicating that a portion of Co is in the Co2+ oxidized state [21].
Fig. 3. (a) SEM image of pure Co, (b) SEM image and (c,d) TEM images of sample S2.

Q. Wang et al. / Electrochimica Acta 56 (2011) 1106–1110
1109
1.1
1.0
0.9
0.8
0.7
0.6
0.5
1st
3rd
4th
5th
0
100
200
300
400
500
Discharge capacity (mAh g-1)
Fig. 6. Charge–discharge curves of S2 electrode.
best cycle stability. The discharge capacity reaches the maximum of
420.2 mAh g−1 at the fifth cycle and is 410 mAh g−1 after 200 cycles,
which is much higher than that of the pure Co electrode. Sam-
ples S1, S3 and S4 all display better electrochemical performance
than pure Co. There are two reasons for the improvement in the
electrochemical properties. The contact area between Co and the
alkaline solution is a key factor influencing the discharge capacity.
The increase in the BET surface increases the contact area between
Co and the alkaline solution, which improves the electrochemical
activity and use of Co. On the other hand, the morphology change
also improves the performance. For the pure Co electrode, the parti-
cle aggregation is not favorable for the electrochemical properties,
while forthe Co–Scomposites, the enhanceddispersion and smaller
particle size of Co particles improve the performance. However,
introducing too much sulfur may decrease the discharge capac-
ity because it causes too much interspacing between Co particles,
leading to poor electrical conductivity.
Fig. 4. EDX patterns of sample S2 taken from different areas: (a) the macrocosm,
(b) the ball-like particles.
Fig. 6 shows the charge–discharge curves of S2 at charge and dis-
charge currents of 100 mA g−1. Two charge–potential plateaus are
observed. The plateau at −0.92 V is attributed to the reduction of
Co(OH)2 to Co. The plateau at −1.0 V is the hydrogen evolution from
water electrolysis [17,23]. The electrode shows a lower discharge
plateau in the initial four-discharge process. In the following cycle,
the long and flat discharge–potential plateau is between −0.74
and −0.77 V, lower than that of the commercial hydrogen storage
alloy (−0.90 V). The length of the charge and discharge plateaus is
nearly the same, showing high coulomb efficiency and excellent
reversibility.
To further confirm the electrochemical reaction process of the
Co–S electrode, cyclic voltammetry (CV) of the S2 electrode was
performed. In the first five cycles, the oxidation and reduction
peaks shift slightly from −0.55 V and −1.02 V to −0.65 V and
−1.05 V, respectively, corresponding to the negative movement of
the discharge–potential plateau. Song et al. reported that the shift
of the redox peaks in the S–Co(OH)2 system is attributed to the
electrochemical dissolution of amorphous S [16]. To investigate
the reaction in this system, an ICP measurement was performed.
The results show that the S elemental content increases gradually
during the initial four cycles in KOH aqueous solution. Therefore,
the shift of the discharge–potential plateau and redox peaks may
be attributed to sulfur dissolution in the Co–S system. Moreover,
the dissolution of sulfur in the composite may produce new inter-
spacing between the Co particles, leading to a larger contact area
between Co and the alkaline solution. This phenomenon causes the
electrodes to need an activation process (Fig. 7).
3.4. Electrochemical properties
Fig. 5 demonstrates the cycle behavior of pure Co and Co–S
electrodes at a current of 100 mA g−1. Compared to the pure Co
electrode, the prepared Co–S electrodes require an activation pro-
cess but have higher discharge capacities and much better cycle
performance. Sample S2 has the highest discharge capacity and the
500
400
300
200
Co
S1
S2
S3
S4
100
0
0
50
100
150
200
Cycle number (n)
Fig. 5. Cycle life of the Co–S electrodes.

1110
Q. Wang et al. / Electrochimica Acta 56 (2011) 1106–1110
0.10
0.05
The phase transformation between the charged and discharged
state illustrates that the discharge capacity of the electrode is
mainly attributed to the electrochemical oxidation of metallic
Co. Co(OH)2 still coexists with Co in the charged state, indicating
that a partially irreversible conversion between the metallic Co
and Co(OH)2 is involved in the faradic reaction, resulting in the
low use of the metallic Co nanoparticles. Additionally, the peaks
of Co9S8 gradually weaken during the cycling, indicating the
dissolution of sulfur in the composite. Moreover, no peaks of other
cobalt chalcogenides are detected, indicating that the conversion
between Co and Co(OH)2 is the only source of the discharge
capacity.
1st
2nd
3rd
5th
0.00
-0.05
-0.10
-0.15
4. Conclusion
-1.2
-1.0
-0.8
-0.6
-0.4
In summary, a series of novel Co–S composites were pre-
pared via a facile hydrothermal method. The introduction of sulfur
enhances the dispersion of Co particles and increases the BET
surface area. The Co–S composites show high discharge capaci-
ties of 420 mAh g−1 and excellent cycle stability. The shift of the
discharge–potential plateau and redox peaks in the CV curves dur-
ing the initial five cycles may be attributed to the dissolution of
sulfur in the Co–S system, which increases the capacity of Co. The
reversible faradic reaction between Co and Co(OH)2 is dominant
for the prepared Co–S composites.
Potential (V vs. Hg/HgO)
Fig. 7. CV curves of S2 electrode.
Acknowledgments
This work was financially supported by the NSFC
(50631020, 50701025, 50971071, 51071087), the 863 pro-
gram (2007AA05Z149, 2007AA05Z108), the 973 program
(2010CB631303) and the Doctoral Foundation of the Ministry
of Education (20070055064).
References
[1] B. Liao, Y.Q. Lei, L.X. Chen, G.L. Lu, H.G. Pan, Q.D. Wang, J. Power Sources 129
(2004) 358.
[2] T. Kohno, H. Yoshida, F. Kawashima, T. Inaba, I. Sakai, M. Yamamoto, M. Kanda,
J. Alloys Compd. 311 (2000) 5.
[3] M.Y. Song, D. Ahn, I.H. Kwon, S.H. Chough, J. Electrochem. Soc. 148 (2001)
A1041.
[4] J.L. Soubeyroux, D. Fruchart, G. Lorthioir, P. Ochin, D. Colin, J. Alloys Compd. 196
(1993) 127.
Fig. 8. XRD patterns of S2 electrode at different cycles.
[5] N.H. Goo, J.H. Woo, K.S. Lee, J. Alloys Compd. 288 (1999) 286.
[6] T. Abe, T. Tachikawa, Y. Hatano, K. Watanabe, J. Alloys Compd. 330–332 (2002)
792.
[7] J.W. Liu, H.T. Yuan, J.S. Cao, Y.J. Wang, J. Alloys Compd. 392 (2005) 300.
[8] Y.D. Wang, X.P. Ai, H.X. Yang, Chem. Mater. 16 (2004) 5194.
[9] D.W. Song, Y.J. Wang, Y.P. Wang, L.F. Jiao, H.T. Yuan, Electrochem. Commun. 10
(2008) 1486.
[10] Y. Han, Y.J. Wang, Y.P. Wang, L.F. Jiao, H.T. Yuan, Int. J. Hydrogen Energy 35
(2010) 8177.
[11] G. He, L.F. Jiao, H.T. Yuan, Y.Y. Zhang, Y.H. Zhang, Y.J. Wang, Int. J. Hydrogen
Energy 32 (2007) 3416.
A pair of remarkable cathodic and anodic current peaks are
present, suggesting a reversible electrochemical reaction occurring
on the Co–S electrode. The curve shape and peak voltage are very
similar to those of the Co–Si3N4 powder electrode [15], confirming
that the same reversible reaction occurs on the electrode. There-
fore, it is likely that the reversible faradic reaction between Co and
Co(OH)2 is dominant for the Co–S system. From the above analysis,
the reaction of the Co–S electrode can be described as follows:
[12] G. He, L.F. Jiao, H.T. Yuan, Y.Y. Zhang, Y.J. Wang, Electrochem. Commun. 8 (2006)
1633.
[13] Y.L. Cao, W.C. Zhou, X.Y. Li, X.P. Ai, X.P. Gao, H.X. Yang, Electrochim. Acta 51
(2006) 4285.
Co + 2OH−
ꢀ
dischargeCo(OH)2 + 2e
(1)
charge
To validate our inference, the structure change in
charge–discharge process was investigated using XRD patterns. As
shown in Fig. 8, the XRD pattern at the fully charged state in the
first cycle is consistent with the original sample, demonstrating
that the metallic Co is dominant in the Co–S electrode before
discharge. At the fully discharged state in the first cycle, most of
the diffraction peaks can be indexed to Co(OH)2 and the diffraction
peaks of metallic Co become very weak, indicating that the Co
transferred into Co(OH)2 in the discharge process. In the seventh
cycle, the charged and discharged phases are metallic Co and
Co(OH)2, respectively, which are similar to those of the first cycle.
[14] Z.W. Lu, S.M. Yao, G.R. Li, T.Y. Yan, X.P. Gao, Electrochim. Acta 53 (2008) 2369.
[15] S.M. Yao, K. Xi, G.R. Li, X.P. Gao, J. Power Sources 184 (2008) 657.
[16] D.W. Song, Y.J. Wang, Q.H. Wang, Y.P. Wang, L.F. Jiao, H.T. Yuan, J. Power Sources
195 (2010) 7115.
[17] D.S. Lu, W.S. Li, C.L. Tan, R.H. Zeng, Electrochim. Acta 55 (2009) 171.
[18] Q.H. Wang, L.F. Jiao, H.M. Du, W.X. Peng, S.C. Liu, Y.J. Wang, H.T. Yuan, Int. J.
Hydrogen Energy 35 (2010) 8357.
[19] X.P. Gao, S.M. Yao, T.Y. Yan, Z. Zhou, Energy Environ. Sci. 2 (2009) 502.
[20] A.K. Shukla, S. Venugopalan, B. Hariprakash, J. Power Sources 100 (2001) 125.
[21] J. Shen, Z. Li, Q. Yan, Y. Chen, J. Phys. Chem. 97 (1993) 8504.
[22] B. Liu, S. Wei, Y. Xing, D. Liu, Z. Shi, X. Liu, X. Sun, S. Hou, Z. Su, Chem. Eur. J.
(2010), doi:10.1002/chem.200903384.
[23] Y. Liu, Y.J. Wang, L.L. Xiao, D.W. Song, Y.P. Wang, L.F. Jiao, H.T. Yuan, Electrochim.
Acta 53 (2008) 2265.