Preparation and electrochemical performance of polycrystalline and single crystalline CuO nanorods as anode materials for Li ion battery

Article abstract of DOI:10.1021/jp037075k

A simple and efficient approach is developed for the synthesis of copper oxide nanorods with different morphology and crystallographic structure. Polycrystalline fine rods 10-20 nm thick and several hundred nanometers long and single crystalline thick rods 60-100 nm thick and up to 1 μm long were obtained from the reactions of copper hydrate with caustic soda solution at room temperature and 100°C, respectively. The fine CuO nanorods as anode materials for Li ion battery exhibit a high electrochemical capacity of 766 mA h/g and relatively poor capacity retention as compared to thick nanorods with the single crystalline structure. The correlation between the structural features of the nanorods and their electrode performance is discussed in detail.

Full text of DOI:10.1021/jp037075k

J. Phys. Chem. B 2004, 108, 5547-5551
5547
Preparation and Electrochemical Performance of Polycrystalline and Single Crystalline
CuO Nanorods as Anode Materials for Li Ion Battery
X. P. Gao,* J. L. Bao, G. L. Pan, H. Y. Zhu,* P. X. Huang, F. Wu, and D. Y. Song†
,
†
†
‡
,§
†
†,|
Institute of New Energy Chemistry Material, Nankai UniVersity, Tianjin 300071, China, State Key Laboratory
of Functional Polymer Materials for Adsorption and Separation, Nankai UniVersity, Tianjin 300071, China,
Electron Microscope Unit and School of Chemistry, The UniVersity of Sydney, NSW 2006, Australia, and
School of Chemical Engineering and EnVironmental, Beijing Institute of Technology, Beijing 100081, China
ReceiVed: October 10, 2003; In Final Form: March 5, 2004
A simple and efficient approach is developed for the synthesis of copper oxide nanorods with different
morphology and crystallographic structure. Polycrystalline fine rods 10-20 nm thick and several hundred
nanometers long and single crystalline thick rods 60-100 nm thick and up to 1 µm long were obtained from
the reactions of copper hydrate with caustic soda solution at room temperature and 100 °C, respectively. The
fine CuO nanorods as anode materials for Li ion battery exhibit a high electrochemical capacity of 766 mA
h/g and relatively poor capacity retention as compared to thick nanorods with the single crystalline structure.
The correlation between the structural features of the nanorods and their electrode performance is discussed
in detail.
I. Introduction
structures in large scales. The fine and bulk nanorods of CuO
polycrystalline and single crystalline structure were prepared
in alkaline solution at room temperature and 100 °C, respec-
tively. The electrochemical reaction of lithium with the obtained
CuO nanorods was investigated by cyclic voltammetry and
galvanostatic method.
The pioneering work in carbon nanotubes has stimulated
profound research interest worldwide in synthesizing nanoscale
materials of various metals, polymers, metal oxides, and sulfides
because of superior properties and potential applications of the
new nanostructures.1,2 In particular, the synthesis of inorganic
nanomaterials with a regular morphology on a nanometer scale,
such as nanotubes and nanorods, is of pronounced significance
to material science. Copper oxide with nanometer-scale dimen-
sion and morphological specificity is an advanced material with
potential for diverse applications as semiconductors, organic
II. Experimental Section
Sample Preparation and Characterization. Copper hy-
droxide precipitates were prepared from a 0.5 M solution of
NaOH and a 0.5 M solution of CuCl . The obtained precipitates
were then dispersed in a 15 M solution of NaOH, and the
2
3
4
catalysts, gas sensors, and lithium ion electrode materials.
Various methods for preparing copper oxide thin film, nano-
particles, and nanorods have been reported, such as electro-
dispersion was transferred into an autoclave with a PTFE
container and kept at room temperature and 100 °C for 48 h,
respectively. The precipitates were recovered from the auto-
claved mixture and rinsed with distilled water until the pH
approached about 7. The wet-cake samples were dried at 100
°C in air for 1 day. The morphology and crystal structure surface
of the copper oxide nanorods were characterized using a
transmission electron microscope (TEM, FEI Tecnai 20) with
an accelerating voltage of 200 kV, and an X-ray diffractometer
(XRD, Rigaku D/max-2500) with Cu KR radiation. The infrared
spectra of the nanorods were recorded on a Nicolet 5ZDX
Fourier transform spectrometer in the spectra range of 4000-
5
6
chemical synthesis, solid-state reactions, a template method
7
using carbon nanotubes and thermal oxidation of copper
8
substrates in air or preparation with poly(ethylene glycol) (PEG)
9
surfactant. It is believed that the copper oxide rods formed in
the presence of PEG surfactant through a surfactant assemblage
9,10
because PEG in water can form a chain structure. In addition,
the nanostructured oxides of 3d transition metal elements exhibit
reversible capacities 2-3 times larger than those of graphite as
1
1
negative electrode materials for lithium ion batteries. It has
been identified that the size of the precursor particles has a
crucial effect on the electrochemical behavior of these metal
oxides toward lithium.12 Here we report a novel and simple
approach for the synthesis of copper oxide nanorods, without
any surfactant or template. This approach has a bright perspec-
tive for large-scale and controllable production of copper oxide
nanorods with various morphologies and crystallographic
-
1
400 cm , and CuO powder obtained from the decomposition
of Cu(OH) at 500 °C in air was used as a reference for blank
2
baseline correction. Thermogravimetric analysis (TGA) was
conducted on a NETZSCH TG 209 to monitor the mass loss of
adsorbed water; specimens were heated at a heating rate of 10
deg/min and in a nitrogen gas flow.
Electrochemical Reaction of Li with CuO Nanorods. The
working electrode was prepared by pressing a mixture of the
active materials, conductive material (acetylene black), and
binder (PTFE) in a weight ratio of 85/10/5. Lithium metal was
used as the counter and reference electrodes. The electrolyte
was 1 M LiPF6 in a mixture of ethylene carbonate (EC),
propylene carbonate (PC), and dimethyl carbonate (DMC). The
*
Corresponding authors. X.P.G.: telephone, +86-22-23500876; fax,
+
7
86-22-23505371; e-mail, xpgao@nankai.edu.cn. H.Y.Z.: fax, 61 2 9351
682; e-mail, h.zhu@emu.usyd.edu.au.
†
Institute of New Energy Chemistry Material, Nankai University.
State Key Laboratory of Functional Polymer Materials for Adsorption
‡
and Separation, Nankai University.
§
The University of Sydney.
Beijing Institute of Technology.
|
1
0.1021/jp037075k CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/10/2004

5548 J. Phys. Chem. B, Vol. 108, No. 18, 2004
Gao et al.
Figure 1. XRD patterns of as-prepared nanorods at room temperature
a) and 100 °C (b).
(
volume ratio of EC:PC:DMC in the mixture was 6:3:1. The
galvanostatic method at the discharge-charge current density
of 50 mA/g was used to measure the electrochemical capacity
of the electrode at room temperature. The cyclic voltammetry
(CV) experiment was carried out at the scan rate of 0.1 mV/s
using a CHI 600A potentiostat at room temperature.
III. Results and Discussion
The XRD patterns of the samples as prepared at room
temperature and 100 °C are shown in Figure 1. All peaks in
the patterns of two samples are consistent with the JCPDS (5-
1
3
0
661) data of the copper oxide with a monoclinic phase. This
indicates that CuO can form even at room temperature when
copper hydroxide precipitates are treated with a concentrated
NaOH solution. However, the diffraction peaks of the sample
obtained at room temperature are obviously broader than those
for the sample prepared at 100 °C, suggesting that the former
consists of much finer crystallites. The average crystalline sizes
in the copper oxide samples at room temperature and 100 °C
are estimated, using the Sherrer equation from the XRD patterns,
to be 17 and 73 nm, respectively.
Hydrothermal treatments were shown to influence greatly the
form of the copper oxide nanorods. The morphological features
of the nanorods are highly dependent upon the temperature of
the treatment with caustic soda in autoclaves during the
synthesis. The product from treatment at room temperature was
uniform fine nanorods about 10-20 nm thick and several
hundred nanometers long (as shown in Figure 2), which are
even thinner than the precursor of Cu(OH)2 precipitates that
are irregular particles of 40-60 nm in size. These fine nanorods
tend to aggregate in spindle-like bundles because of their high
surface energy or van der Waals forces, and this is similar to
the bundles of single wall carbon nanotubes.14 In contrast, bulk
copper oxide nanorods were obtained at 100 °C, which have
diameters of ca. 60-100 nm and lengths of up to 1.0 µm.
Obviously the temperature of the treatment in autoclaves has a
profound influence on the morphology and size of copper oxide
nanomaterials. The selected area electron diffraction (SAED)
patterns inserted in Figure 2 indicate an interesting fact that
the bulk nanorods obtained at 100 °C consist of single crystals
while the fine nanorods obtained at room temperature have a
polycrystalline structure, although they both have monoclinic
phase structure according to their XRD patterns. Raising the
temperature of the hydrothermal treatment results in structure
transformation of CuO nanorods from polycrystalline to single
crystalline. This transformation may undergo the Ostwald
ripening process in which larger crystallites grow at the expense
Figure 2. TEM images of Cu(OH)
and as-prepared nanorods at room temperature (b) and 100 °C (c). The
selected area electron diffraction of the as-prepared nanorods is inserted.
2
precipitates dried at 100 °C (a)

CuO Nanorods as Anode Materials for Li Ion Battery
J. Phys. Chem. B, Vol. 108, No. 18, 2004 5549
of smaller crystallites being dissolved.15 Meanwhile, the spindle-
like bundles of copper oxide nanorods grow into bulk nanorods
or nanoplatelets as the temperature is raised.
It has been reported that TiO2 or titanate nanotubes, nanorods,
and nanoribbons can be synthesized hydrothermally in NaOH
solution.1
6-19
For both CuO and TiO2 or titanate, one-
dimensional nanostructures can be formed in concentrated
NaOH solution under hydrothermal conditions without any
surfactant and template. This suggests that these syntheses may
follow a common mechanism, which could also be applied to
other transition metal oxides. The difference is that TiO2 or
titanate nanotube and nanorods prepared from concentrated
1
6,19
NaOH solution contain a large amount of crystal water,
while CuO nanorods prepared in NaOH solution contain no
crystal water due to dehydration in the same solution. It was
-
also found that, in a reaction solution of a high OH concentra-
tion, Zn(OH)42 precursor plays an important role in determining
-
-
the morphologies of ZnO crystallites. The OH concentration
in the reaction solution should be a key factor for controlling
the growth rate of different crystal faces and thus lead to the
formation of an anisotropic particle such as rod-like morphology
as found in the preparation of ZnO nanorods.20 However, Cu-
(
OH)2 precipitates cannot dissolve in concentrated NaOH
2
-
solution to form dissoluble complex ions such as Zn(OH)4
.
Therefore, we believe that CuO crystalline nuclei of nanoscale
size form during the dehydration process of Cu(OH)2 precipitates
in concentrated NaOH solution and grow gradually in a one-
dimensional direction to result in the rod-like structure. The
hydrothermal synthesis in NaOH solution is a simple and
efficient method for producing various inorganic metallic oxide
nanomaterials with a regular geometry and crystal structure.
Cyclic voltammograms (CVs) of electrodes made by fine and
bulk nanorods at a scan rate of 0.1 mV/s are depicted in Figure
Figure 3. Cyclic voltammograms of as-prepared nanorods at room
temperature (a) and 100 °C (b) at a scan rate of 0.1 mV/s.
3. In the first cycle, the three cathodic peaks of lithium in the
fine nanorod electrode are observed at 1.86, 1.05, and 0.74 V
+
vs Li /Li in the voltammograms, respectively, corresponding
The initial discharge-charge curves of the electrode made
by the fine and bulk nanorods at the discharging and charging
current density of 50 mA/g at room temperature are shown in
to the multistep electrochemical lithium reaction process or
additional sites for lithium intercalation. Meanwhile, only one
main anodic peak is located near 2.51 V vs Li /Li, and a small
+
Figure 4. There are three obvious sloping potential ranges
shoulder is observed, corresponding to Li extraction from the
crystal lattice of CuO.21 In the second cycle, decreases of
individual peak intensity and integral area, resulting in reversible
capacity losses, are observed with shifts to 2.25, 1.23, and 0.78
V of the peak potentials in the cathodic direction. The change
of peak potentials in CVs is in agreement with potential range
in discharge curves as found in the next section. In the following
cycles, there is no substantial change in the peak potentials and
curve shape, which remain similar to those in the second cycle.
In contrast, the main potential peaked at 1.08 V with two small
shoulder peak potentials of 1.64 and 0.80 V for the bulk nanorod
electrode in the first cycle, different from that of the fine nanorod
electrode. This is obviously different from the voltammograms
of the fine nanorod electrode. The substantial changes in main
peak intensity and integral area of the bulk nanorod electrode
in the second cycle indicate larger reversible capacity losses as
compared with those of the fine nanorod electrode. The
structural defects of the nanorods are favorable for the diffusion
of Li ions into the CuO crystallites with the formation of the
intercalated compound LixCuO.22 The multipeaks are associated
with differences in the lithium reaction process and can be
attributed to the formation of the imperfection of highly
organized nanotextured CuO lattice with polycrystalline struc-
ture, which facilitates the transport of lithium in surface defects
and in bulk materials.
+
(
2.25-1.8, 1.25-1.0, and 1.0-0.02V vs Li /Li) for the lithium
reaction with the fine CuO nanorods during the first discharge
in accordance with three cathodic peaks in the above CVs. The
first potential range (2.25-1.8 V) is narrowed and shifts slightly
upward during the second discharge, accompanied by a decrease
in the discharge capacity. The first and second discharge-charge
curves of the bulk nanorods are similar to those of CuO spherical
particles with 0.15 and 1 µm diameters as anode materials.12
The wider (1.5-1.0 V) and narrower (1.0-0.02 V) potential
ranges appear for lithium reaction with bulk CuO nanorods, but
are different from that of the fine CuO nanorods. Moreover,
the irreversible capacity loss of the bulk CuO nanorods after
the first discharge increases, compared with that of the fine CuO
nanorods. The change in potential range and irreversible capacity
during the first and second discharges are consistent with the
above observation in CVs. The interaction of nanorods with
lithium ions takes place mostly at the surface rather than in the
2
2
bulk as the crystal size decreases. This means that the
electrochemical reaction of lithium could be readily achieved
in the fine CuO nanorods with polycrystalline structure because
of their large specific surface area and numerous structural
defects (grain boundary, oxygen vacancies, or the presence of
-
23
OH groups ).
Cycle performance of the electrodes made by the fine and
bulk nanorods at the discharging and charging current density

5550 J. Phys. Chem. B, Vol. 108, No. 18, 2004
Gao et al.
Figure 6. Infrared spectra of as-prepared nanorods at room temperature
(a) and 100 °C (b).
for such high specific capacities.26 The large excesses observed
in capacity can also originate from the decomposition of the
electrolyte and subsequent formation of an organic layer
deposited on the surface of the particles that occurs in the low
26
potential region for transition metal oxides. On the other hand,
the peak potential near 1.8 V (the actual peak potential depends
on the scan rate) in the voltammetric curve is attributed to the
-
presence of some residual OH groups in the active form of
2
5
CuO. From the infrared spectra of the nanorods prepared at
room temperature and 100 °C (Figure 6), two stronger absorp-
-
1
tion bands at 1560 and 3470 cm are observed, which are due
to the bending vibration and stretching vibration, respectively,
of adsorbed water. The stronger infrared absorption of the two
bands for the fine nanorods obtained at room temperature
suggests that this sample contains much more adsorbed water.
The thermal gravimetric analysis (not shown) indicates a mass
loss of about 1 wt % below 200 °C, which is completely due to
Figure 4. Discharge-charge curves of as-prepared nanorods at room
temperature (a) and 100 °C (b) at discharging and charging current
density of 50 mA/g at room temperature.
-
the loss of the adsorbed water from the sample. Therefore, OH
groups are mainly responsible for the higher discharge capacity
of the CuO fine nanorods. The discharge capacity of the bulk
nanorod electrode appeared to increase slightly during the first
eight cycles due to crystal perfection, which usually proceeds
with the activation process for electrochemical reaction of
lithium. After 100 cycles, the CuO bulk nanorod electrode is
able to keep its capacity at 470 mA h/g, while the electrochemi-
cal discharge capacity of the fine nanorod electrode decreases
gradually to 550 mA h/g after 100 cycles.
Other researchers have identified that a change in the particle
size of the precursor can strongly affect the capacity retention
in transition metal oxides and LiMn2O4 for electrochemical
4
,12,26,27
21
reaction of lithium.
Novak suggested that the discharge
+
reaction involves intercalation of Li ions into the CuO lattice
+
-
12
according to CuO + xLi + xe S LixCuO. Grugeon proposed
a model involving the formation of metal nanograins during
the first discharge and then the formation-decomposition of
Li2O concomitant with the reduction-oxidation of CuO on
subsequent cycles. In general, the small particles are able to
expand much more easily and can more easily accommodate
the structural strains for the electrochemical reaction of lithium.26
However, lithium interaction in a larger amount can trigger a
phase transition and structure change. Therefore, the electro-
chemical Li-driven irreversible structural, morphological, and
textural changes in the polycrystalline fine CuO nanorods could
be more serious during cycling as compared to the single
crystalline bulk CuO nanorods due to electrochemical reaction
in a large amount. Therefore, the single crystalline structure of
Figure 5. Cycle life of the electrode made by as-prepared nanorods
at room temperature (a) and 100 °C (b) at discharging and charging
current density of 50 mA/g at room temperature.
of 50 mA/g at room temperature is illustrated in Figure 5. The
irreversible discharge capacity after the first cycle is due to the
initial formation of lithium oxide.24 The electrochemical revers-
ible capacities of 766 and 416 mA h/g are achieved during the
second discharge. Based on a maximum uptake of 2Li/CuO,
the theoretical capacity of 674 mA h/g 25 is attained. The
contribution of the reaction of lithium with carbon additive is
at a magnitude of 60 mA h/g, and is thus not taken into account

CuO Nanorods as Anode Materials for Li Ion Battery
J. Phys. Chem. B, Vol. 108, No. 18, 2004 5551
copper oxide is more favorable for stable capacity of the
electrochemical reaction of lithium. Nevertheless, its discharge
capacity is lower, compared with that of the polycrystalline
structure. A similar capacity decay in the electrochemical
reaction of lithium was also reported in nanosized Cu2O, CuO,
and LiMn2O4, which is more rapid and serious than that in the
(3) Yamamoto, K.; Kasuga, T.; Nogami, M. Electrochem. Solid-State
Lett. 1999, 2, 595-596.
(
4) Poizot, P.; Laruelle, S.; Grugeon, S.; Dupont, L.; Tarascon, J. M.
Nature 2000, 407, 496-499.
(5) Poizot, P.; Hung, C. J.; Nikiforov, M. P.; Bohannan, E. W.; Switzer,
J. A. Electrochem. Solid-State Lett. 2003, 6, C21-C25.
(
(
6) Jia, D. Z.; Yu, J. Q.; Xia, X. Chin. Sci. Bull. 1998, 43, 172-174.
7) Wu, H. Q.; Wei, X. W.; Shao, M. W.; Gu, J. S.; Qu, M. Z. Chem.
12,27
corresponding microsized materials.
Phys. Lett. 2002, 364, 152-156.
8) Jiang, X. C.; Herricks, T.; Xia, Y. N. Nano Lett. 2002, 2, 1333-
1338.
(
IV. Conclusion
(9) Wang, W. Z.; Zhan, Y. J.; Wang, G. H. Chem. Commun. 2001,
In summary, copper oxide nanorods of different crystalline
structures and morphologies were synthesized successfully by
a simple and efficient method in NaOH solution. The crystalline
structure and morphology of the products are highly dependent
upon the temperature of the hydrothermal treatment in the
synthesis. The synthesis strategy could also be applied for the
preparation of one-dimensional nanostructures of other transition
metal oxides without any surfactant or template. The fine and
polycrystalline CuO nanorods as anode materials for Li ion
battery exhibit a high electrochemical capacity of 766 mA h/g,
compared to 416 mA h/g for the single crystalline bulk nanorods
due to their large surface area and numerous structural defects.
However, the capacity retention of the fine CuO nanorods of
polycrystals is not as good as that of bulk nanorods of a single
crystal due to the Li-driven irreversible morphological and
crystal changes during the cycles of the electrochemical reaction
of lithium in a large amount.
727-728.
(10) Sun, J. H.; Gong, Y. J.; Fan, W. H.; Wu, D.; Sun, Y. H. Chem. J.
Chin. UniV. 2000, 21, 95-98.
(
11) Tarascon, J. M.; Armand, M. Nature 2001, 414, 359-367.
(12) Grugeon, S.; Laruelle, S.; Herrera-Urbina, R.; Dupont, L.; Poizot,
P.; Tarascona, J. M. J. Electrochem. Soc. 2002, 148 (4), A285-A292.
(13) Joint Committee on Powder Diffraction Standards. Diffraction Data
File No. 5-661; JCPDS International Center for Diffraction Data, Pennsyl-
vania, 1991.
(14) Saito, Y.; Tani, Y.; Miyagawa, N.; Mitsushima, K.; Kasuya, A.;
Nishina, Y. Chem. Phys. Lett. 1998, 294, 593-598.
(
(
15) Kabalnov, A. J. Dispersion Sci. Technol. 2001, 22 (1), 1-12.
16) Kasuga, T.; Hiramatsu, M.; Hoson, A.; Sekino, T.; Niihara, K.
Langmuir 1998, 14, 3160-3163.
17) Wang, Y. Q.; Hu, G. Q.; Duan, X. F.; Sun, H. L.; Xue, Q. K. Chem.
Phys. Lett. 2002, 365, 427-431.
18) Yuan, Z. Y.; Colomer, J. F.; Su, B. L. Chem. Phys. Lett. 2002,
63, 362-366.
19) Gao, X. P.; Zhu, H. Y.; Pan, G. L.; Ye, S. H.; Lan, Y.; Wu, F.;
(
(
3
(
Song, D. Y. J. Phys. Chem. B 2004, 108, 2868-2872.
(20) Zhang, J.; Sun, L. D.; Yin, J. L.; Su, H. L.; Liao, C S.; Yan, C. H.
Chem. Mater. 2002, 14, 4172-4177.
(
21) Novak, P. Electrochim. Acta 1985, 30, 1687-1692.
Acknowledgment. This work is supported by 973 Programs
(22) Kavan, L.; Rathousky, J.; Gratzel, M.; Shklover, V.; Zukal, A. J.
(
(
2002CB211800), the National Key Program for Basic Research
2001CCA05000), and the NSFC (90206043) of China. Finan-
Phys. Chem. B 2000, 104, 12012-12020.
(23) Podhajecky, P.; Zabransky, Z.; Novak, P.; Dobiasova, Z.; Cerny,
R.; Valvoda, V. Electrochim. Acta 1990, 35, 245-249.
cial support from the Australian Research Council (ARC) is
also gratefully acknowledged, and H.Y.Z. is indebted to ARC
for the QE II fellowship.
(24) Wu, G. T.; Wang, C. S.; Zhang, X. B.; Yang, H. S.; Qi, Z. F.; Li,
W. Z. J. Power Sources 1998, 75, 175-179.
(25) Novak, P.; Klapste, B.; Podhajecky, P. J. Power Sources 1985,
1
5, 101-109.
(26) Larcher, D.; Masquelier, C.; Bonnin, D.; Chabre, Y.; Masson, V.;
References and Notes
Leriche, J. B.; Tarascona, J. M. J. Electrochem. Soc. 2003, 150 (1), A133-
A139.
(27) Ye, S. H.; Lv, J. Y.; Gao, X. P.; Wu, F.; Song, D. Y. Electrochim.
Acta 2004, 49 (9-10), 1623-1628.
(
1) Iijima, S.; Ichihashi, T. Nature 1993, 363, 603-605.
(2) Wen, J. G.; Lao, J. Y.; Wang, D. Z.; Kyaw, T. M.; Foo, Y. L.;
Ren, Z. F. Chem. Phys. Lett. 2003, 372, 717-722.