Chiang Mai J. Sci. 2007; 34(1)
29
Chiang Mai J. Sci. 2007; 34(1) : 29-33
www.science.cmu.ac.th/journal-science/josci.html
Contributed Paper
Electron Microscopy Study of the Formation of
Dendrite Cu6Sn5 Powders Synthesized by Solution
Route Method
Siwat Thungprasert* [a], Thapanee Sarakonsri [a], and Tawee Tunkasiri [b]
[a] Department of Chemistry, Faculty of Science, Chiang Mai University, Chiang Mai 50200, Thailand.
[b] Department of Physics, Faculty of Science, Chiang Mai University, Chiang Mai 50200, Thailand.
* Author for correspondence; e-mail : [email protected]
Received : 30 August 2006
Accepted : 27 October 2006.
ABSTRACT
Dendrite Cu6Sn5 powders, a potential anode material for lithium-ion battery, can be
synthesized by performing a redox reaction between stochiometric amount of metals chlorides
and zinc metal in ethylene glycol solvent. The formation of dendrite morphology was
investigated by considering two processing parameters. Two forms of reducing agent; zinc
powders (45 mm) and zinc plate, were the first parameter. Reaction time varying from 1-5
minutes was the second one. From the characterization of the product powders by using
SEM and TEM techniques, Cu6Sn5 compound was discovered to nucleate on reducing agent
particles and subsequently grown into dendrite structure with long branches. These results
have confirmed the proposed mechanism by “point effect of the electric field” for dendrite
formation in this solution system.
Keywords: Cu6Sn5, lithium-ion battery, dendrite structure, electron microscopy.
INTRODUCTION
A lithium-ion battery is an important
portable power source for the current human
society because of its high energy density, high
voltage, long cycle life, flexible design and
considerably safe to use [1]. Lithium-ion
batteries were developed to replace Ni-Cd
batteries which contain toxic materials and
encounter a problem on memory effect [2].
Lithium-ion batteries are commercially made
from graphite negative electrode, which gave
excellent cycle ability without memory effect
problem [3]. However, there is a safety
concern due to lithiated graphite, LixC6, reach
the lithium potential when cells approach a
fully charged state as the positive electrode
(Li1-xCoO2) was a highly oxidizing agent [4].
1.
Moreover, it provides volumetric capacity
only about 750 mAh/cm3 [3]. The Cu6Sn5
compound then was proposed as an
intermetallic insertion negative electrode due
to their high volumetric capacity
(approximately 1760 mAh/cm3) over graphite
electrode and it reacted with lithium at the
voltage above 100 mA, which satisfy the safety
issue [5-7].
The Cu6Sn5, which initially has NiAs type
structure, can rearrange itself into zinc-blende
type structure after the first discharge [8]. The
previons-research reported that Cu 6Sn 5
compound could be synthesized at lowtemperature by the solution route method
[7,9], which also can be used to synthesize
30
other intermetallic compounds. The
morphology of the intermetallic powders was
studied and reported to have dendrite
structure [9]. The dendrites, which had
branches composed of nanometer size
particles, were considered to have relatively
high surface area compared with the material
synthesized by mechanical alloying. The
dendrite structure also could be synthesized
by other methods such as solvothermal or
hydrothermal [10-11]. These two methods
produce nanoscale dendrite structure similar
to solution route method but rewire longer
time to synthesize. Solution route method then
was a convenient and easily controlled method,
which was applied to follow the formation
of dendrite in a short period of time.
As reported initially, the dendrite
morphology was proposed to occur by the
“point effect of the electric field” [7,9]. This
research therefore has applied electron
microscopy techniques which included
scanning electron microscopy (SEM) and
transmission electron microscopy (TEM) to
investigate the dendrite formation and to
prove the above hypothesis. Different forms
of zinc reducing agent, which were zinc
powders and zinc plates, were studied in this
experiment. The plate shape reducing agent
was used because the dendrite development
will be easily observed due to larger particle,
which associated with high electron density,
Chiang Mai J. Sci. 2007; 34(1)
compared with powder reducing agent.
Reaction time was an important parameter
for the study of dendrite for mation
mechanism. Products examination after the
reaction occurred for 1, 2, 3, 4, and 5 minutes
allowed the dendrite development to be
observed.
EXPERIMENTAL PROCEDURE
The experimental procedure was
conducted as followed. The experimental
condition was described by dissolving
stoichiometric amounts (equation (1)) of
CuCl2 (Sigma-Aldrich, purity 99.0%) and SnCl2
(Sigma-Aldrich, purity 99.0%) in ethyl glycol
(JT. Baker, purity 99.0%) at room temperature.
The metal ions were reduced by adding zinc
powder (Fluka, 45 mm particle size) and the
reaction was stirred for 1 minute. Then, the
product powder was filtered and washed by
methanol (Merck, commercial grade). Finally,
it was dried in the oven at 65 ðC for 30
minutes. The subsequence experiments were
performed analogous to the above
experiment except varying reaction times
from 2 to 5 minutes. The other reaction was
carried out using the same procedure as all
the above experiments except using zinc plate
(Riedel-deHaën, purity 99.9%) as reducing
agent and varying the reduction times by 1, 2,
3, 4, and 5 minutes.
2.
Figure 1. SEM micrographs
showing the as-received zinc
powders (a) Cu6Sn5 dendrites on
zinc powder at 1 minute (b),
2 minutes (c), and 5 minutes
reaction time (d).
Chiang Mai J. Sci. 2007; 34(1)
31
3.
RESULTS AND DISCUSSION
Initially, the SEM technique was
employed for morphology study. The asreceived zinc powders were observed in figure
1a to have a particle size about 30-50 mm. At
1 minute reaction, the SEM micrograph in
figure 1b shows that the zinc particle was partly
covered by a large number of small dendrites.
The small size of dendrite particles were
observed due to only a few electrons were
transferred to copper and tin ions for copper
tin compound formation. There was no
temperature effect due to room temperature
reaction. At 2 minutes reaction, the dendrite
particles were found to fully cover the zinc
particle (figure 1c). The mechanism of dendrite
formation can possibly be considered to occur
initially by the Cu6Sn5 intermetallic compound
on the zinc surface had formed in group of
lump and subsequently grown out into
dendrites by the prefer transferring path of
electrons from zinc through intermetallic
compound rather than transferring to
distributed ions in the solvent (figure 2). The
possible reason for dendrite formation was
that there was the electrochemical
overpotential at the solution precipitate
interface that produces an electric field and
act as electrodes for the reduction of tin and
copper ions at the dendrite tip. Intermetallic
compound therefore, has grown out as
branches from lump to dendrite structure by
the “point effect of the electric field”
mechanism [9]. The electrons from zinc
particles had transferred through intermetallic
compound to metal ions in solvent at the top
of branches. At 5 minutes reaction (figure 1d),
the dendrites grown into bigger dendrite than
dendrites at 2 minutes and were dispersed in
the solution due to the disappearance of zinc
metal. It was observed at the end that the
groups of dendrite particle were dispersed in
the solution by the stirring process and some
dendrites had broken into small branches.
The characterization by TEM technique
has shown in figure 3. The particle at the
bottom left corner was confirmed by selected
area diffraction pattern (SAD) to be zinc metal.
These TEM micrographs has proved the
electrons path of migration from zinc particle
to tin and copper ions to grown in to Cu6Sn5
dendrites structure by electrons transferring
through the tip of dendrite branches as shown
in the schematic figure 2. Therefore, the TEM
results were consistent with SEM results and
verified the proposed mechanism of dendrite
formation on zinc particle.
Figure 2. Schematic diagram of the nucleation
mechanism of dendrite Cu6Sn5 on zinc particle
[9].
Figure 3. TEM micrographs of Cu 6 Sn 5
dendrite particles attached to zinc powder and
the corresponding zinc diffraction patterns
from 1 minute reaction time in the inserted
figure.
32
Chiang Mai J. Sci. 2007; 34(1)
Figure 4. SEM micrographs showing the as-received zinc plate surface (a), Cu6Sn5 dendrites
on zinc plate at 1 minute (b), 2 minutes (c), 3 minutes (d), 4 minutes (e), and 5 minutes reaction
time (f).
For the reaction using zinc plate reducing
agent, the SEM image of the as-received zinc
plate is shown in figure 4a. The SEM image
of the product from 1 minute reaction was
shown in figure 4b along with the SEM image
of the products from 2, 3, 4, and 5 minute
reaction time in figure 4c-4f, respectively. At
1 minute reaction, zinc plate bonded the group
of lump on its surface. Similar results as for
zinc powder at the same condition were
observed. Then, the dendrites were observed
to grow into longer branches at 2 minutes
reaction time and continue to grow extensively
as the time increase. Compared with the
previous zinc powder condition, larger
dendrites were formed by using zinc plate
because of a large number of electrons
provided for the reduction process.
This research also discovered that the
number of dendrites at 1 minute reaction time
from the zinc powder condition were relatively
higher than those from the zinc plate
condition. This phenomenon can be explained
by the possibility of the impurity such as oxide
film on the zinc plate surface hindrance the
dendrite nucleation. On the other hand, at 5
minutes reaction, the dendrites from the zinc
plate condition had relatively longer branches
than those from the zinc powders condition.
The reason was due to the insufficient electrons
from zinc powder for extended grown of
dendrites and the reduction process
completed or no more zinc metal powder
before it reached 5 minutes. Therefore,
dendrites particles were found floating around
in the solution.
CONCLUSION
The dendrite Cu 6Sn 5 powder can be
successively synthesized by solution route
method. The formation of Cu6Sn5 dendrite
was effectively monitored by examining the
products at different reaction times using SEM
and TEM techniques. Different forms of zinc
reducing agent were employed to reveal the
dendrite formation behavior. The proposed
nucleation mechanism by “point effect of the
electric field” was confirmed to occur in this
solution route method.
4.
Chiang Mai J. Sci. 2007; 34(1)
ACKNOWLEDGEMENT
CMU electron microscopy research and
service center is thanked for sample testing.
Supported for this work from the Thailand
Research Fund (TRF) and The Commission
on Higher Education, contract no
MRG4680166 is gratefully acknowledged.
5.
REFERENCE
[1] Fransson L.M.L., Vaughey J.T., Bendek
R., Edström K., Thomas J.O., and
Thackeray M.M., Electrochemistr y
Communications, 2001; 3 : 317-323.
[2] Reddy B.R., Priya D.N., Rao S.V., and
Radhika P., Hydrometallurgy, 2005; 77 :
253-261.
[3] Yang S., Zavalij P.Y., Whittingham M.S.S.,
Electrochemistry Communications, 2003; 5 :
587–590.
[4] Kepler D., Vaughey J.T., and Thackeray
M.M., Electrochemical and Soild-State Ioincs,
1996; 90 : 281.
33
[5] Thackeray M.M., Johnson C.S., Kahaian
A.J., Kepler K.D., Vaughey J.T., ShaoHorn Y., and Hackney S.A., ITE Battery
Letters, 1(1).
[6] Thackeray M.M., Vaughey J.T., Kahaian
A.J., Kepler K.D., and Benedek R.,
Electrochemistry Commusication, 1999; 1 :
111-115.
[7] Sarakonsri T., Johnson C.S., Hackney S.A.,
and Thackeray M.M., J. Power Sources,
2005; 153 : 349-327.
[8] Vaughey J.T., Kepler 1 K.D., Benedek R.,
and Thackeray M.M., Electrochemistry
Communications, 1999; 1 : 517–521.
[9] Sarakonsri T., Apirattanawan T.,
Tungprasert S., and Tunkasiri T., J.
Materials Science, 2005; 41 : 4749-4754.
[10] Hu H., Yang B., Li Q., Liu X., Yu W., and
Qian Y., J. Crystal Growth, 2004; 261 :
485–489.
[11] Zhu Y., Zheng H., Li Y., Gao L., Yang
Z., and Qian Y., Materials Research Bulletin,
2003; 38 : 1829–1834.
34
Chiang Mai J. Sci. 2007; 34(1)
หนาวางครับ