Electrochemical Properties of Carbonaceous Thin Films Prepared by Plasma Chemical Vapor Deposition

Article abstract of DOI:10.1149/1.1409542

Carbonaceous thin films were prepared from acetylene and argon by plasma-assisted chemical vapor deposition (plasma CVD). The carbonaceous thin films were characterized mainly by scanning electron microscope (SEM) and Raman spectroscopy. Then their electr

Full text of DOI:10.1149/1.1409542

A1260
Journal of The Electrochemical Society, 148 ͑11͒ A1260-A1265 ͑2001͒
0
013-4651/2001/148͑11͒/A1260/6/$7.00 © The Electrochemical Society, Inc.
Electrochemical Properties of Carbonaceous Thin Films
Prepared by Plasma Chemical Vapor Deposition
Tomokazu Fukutsuka, Takeshi Abe,* Minoru Inaba,* and Zempachi Ogumi*,z
Graduate School of Engineering, Kyoto University, Sakyo-ku Kyoto 606-8501, Japan
Carbonaceous thin films were prepared from acetylene and argon by plasma-assisted chemical vapor deposition ͑plasma CVD͒.
The carbonaceous thin films were characterized mainly by scanning electron microscope ͑SEM͒ and Raman spectroscopy. Then
their electrochemical properties were studied by cyclic voltammetry, charge-discharge measurements, and linear sweep voltam-
metry. From the SEM image carbonaceous thin films appeared flat and pinhole free. Crystallinity of carbonaceous thin films are
affected by the applied rf power from Raman spectra. The difference of the applied rf power also affected the results of cyclic
voltammetry and charge-discharge measurements. The lithium ion storage mechanism of carbonaceous thin film is discussed from
the results of electrochemical measurements.
©
2001 The Electrochemical Society. ͓DOI: 10.1149/1.1409542͔ All rights reserved.
Manuscript submitted August 7, 2000; revised manuscript received June 4, 2001. Available electronically October 8, 2001.
Lithium-ion batteries have been extensively studied because of
their high performance and potentialities. Carbonaceous materials
with sp -type structure have attracted much attention for use as
plasma assist gas, respectively. Substrates were placed on a ground
electrode whose temperature was kept at 873 K. Carbonaceous thin
films were deposited on substrates of nickel and Pyrex glass sheets.
2
®
negative electrodes of lithium ion batteries. Among them, highly
crystallized graphite has been employed as the negative electrode
for lithium-ion batteries in the commercial market. However, vari-
ous carbonaceous materials have been extensively investigated so
far as negative electrode materials for improving the performance of
lithium-ion batteries.1-9
Glow discharge plasma was generated between rf and ground elec-
trodes by an rf power supply of 13.56 MHz, and the applied rf
power was changed from 10 to 90 W. Flow rates of argon and
acetylene were set at 20 and 10 sccm, respectively. The total pres-
sure of the reaction chamber was kept at 133 Pa.
The resultant carbonaceous films were characterized by scanning
electron microscopy ͑SEM, Hitachi S-510͒ and Raman spectros-
copy. The Raman spectra were excited by using a 514.5 nm line ͑50
mW͒ of an argon ion laser ͑NEC, GLG3260͒, and the scattered light
was collected in a backscattering geometry. All spectra were re-
corded using a spectrometer ͑Jobin-Yvon, T64000͒ equipped with a
multichannel charge coupled device ͑CCD͒ detector. Each measure-
ment was carried out at room temperature with an integration time
of 300 s.
Since carbonaceous materials are generally powders, binders are
essential for the preparation of electrodes, which makes it difficult to
elucidate the precise electrochemical properties of the carbonaceous
materials themselves. For the study of carbonaceous negative elec-
trodes, thin films are very useful, because the films can be used as
electrodes without binders and furthermore, it is easy for us to
evaluate their electrode surface area. However, few studies employ-
ing carbonaceous thin films as negative electrodes of lithium-ion
batteries have been made.10 This is because there are some difficul-
Electrochemical measurements were performed by using a three-
electrode electrochemical cell. Lithium metal was used as counter
and reference electrodes, and the electrolyte solution was a mixture
2
ties in preparing thin-film electrodes of the sp -type carbonaceous
materials due to low adherence on a substrate, etc.
͑1:1 by volume͒ of ethylene carbonate ͑EC͒ and diethyl carbonate
A plasma is an ionized gas containing equal numbers of positive
and negative charges, and a number of nonionized neutral excited
species. Glow discharge plasma is unique in that it can generate
chemically very reactive species at low temperatures due to the non-
equilibrium nature of the plasma state. Plasma-assisted chemical
vapor deposition ͑CVD͒ has been used for the preparation of a va-
riety of inorganic materials.1
1,12
The chemical reactions are acceler-
ated in plasma at low temperatures, and plasma CVD can easily give
dense and pinhole free films.
Synthesis of carbonaceous thin films by use of plasma CVD has
been also carried out by many workers. However, most of these
studies have focused on the preparation of diamond films and/or
3
13,14
diamond-like carbon films of sp -type structure,
and little atten-
2
tion has been paid to a synthesis of sp -type carbonaceous thin films.
2
Preparation and characterization of sp -type carbonaceous thin
films by plasma CVD have been already reported by the present
1
5-17
authors.
In this work, electrochemical properties of carbon-
aceous thin-film electrodes prepared by plasma CVD were studied
by cyclic voltammetry ͑CV͒, charge-discharge measurements, and
linear sweep voltammetry.
Experimental
Figure 1 shows a schematic diagram of the plasma CVD appa-
ratus used in this study. Starting materials were acetylene and argon.
Acetylene and argon were selected as the carbon source and the
Figure 1. Schematic of rf reactor for the plasma CVD apparatus. 1, Ar inlet:
2, C2H2 inlet; 3, rf electrode; 4, substrate; 5, thermocouple; 6, ground elec-
trode; 7, tungsten heater; 8, viewport.
*
Electrochemical Society Active Member.
E-mail: ogumi@scl.kyoto-u.ac.jp
z
Downloaded on 2015-01-19 to IP 129.81.226.78 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 148 ͑11͒ A1260-A1265 ͑2001͒
A1261
Figure 2. SEM image of carbonaceous thin film prepared by plasma CVD.
®
Reaction time, 6 h; substrate, Pyrex glass, applied rf power, 10 W.
͑
DEC͒ containing 1 mol dmϪ3 LiClO ͑battery grade by Mitsubishi
4
Petrochemical Company, Limited͒. The cell was assembled in an
argon-filled glove box. Electrochemical properties were examined
by cyclic voltammetry and linear sweep voltammetry ͑Radiometer,
VoltaLab 32͒ and charge-discharge measurements using a battery
test system ͑Hokuto Denko, HJ101SM6͒. Unless otherwise stated,
ϩ
the potential is described vs. Li/Li .
Results and Discussion
Surface morphology of carbonaceous thin films.—Figure 2
shows a typical SEM image of carbonaceous thin film prepared by
plasma CVD at 10 W. From the image, the surface of the carbon-
aceous thin film was very flat and free from pinholes. The film
thickness was determined to be less than 1 ␮m by cross section of
SEM image. Carbonaceous thin films prepared using other rf powers
also gave morphologies similar to those using an rf power of 10 W.
Raman spectroscopy of carbonaceous thin films.—Raman spec-
tra of carbonaceous thin films are shown in Fig. 3. Three main peaks
Ϫ1
around 1360, 1580, and 1620 cm were observed. The peak around
Ϫ1
1
580 cm is well known to be related to the crystallinity of car-
bonaceous materials, and is assigned to the Raman active E mode
Figure 3. Raman spectra of carbonaceous thin films prepared by plasma
CVD. Reaction time, 6 h; substrate, Ni; applied rf power, 10, 50, and 90 W.
2
g
frequency ͑G band͒.18 The peak around 1360 cm is ascribed to the
Ϫ1
Raman inactive A mode frequency.18 Peaks around 1360 and 1620
1
g
Ϫ1
cm appear in the case of finite crystal size and imperfections of
carbonaceous materials, and the former is called a D band and the
latter is a DЈ band. As is given in Fig. 3, the peak around 1580 cm
19
Cyclic voltammogram of carbonaceous thin films.—Figure 4
shows cyclic voltammograms of carbonaceous thin film prepared at
Ϫ1
of the G band became sharper, and its position shifted toward 1580
9
1
0 W. The cyclic voltammogram was measured with a sweep rate of
mV/s in the potential range of 0 to 3 V. For the first sweep, a large
Ϫ1
cm with increasing the applied rf power. Full width at half maxi-
Ϫ1
mum ͑fwhm͒ of the peak around 1580 cm was reported to be
irreversible reduction was observed from the potential around 1.5 to
.5 V. However, it almost disappeared after the second sweep as is
1
9
correlated with the crystallinity of carbonaceous materials, and
0
Ϫ1
single crystals of graphite gave only one peak around 1580 cm
.
evident from Fig. 4. These results indicate that the decomposition of
the solvent and formation of the solid electrolyte interface ͑SEI͒ on
the surface of carbonaceous thin films should occur effectively.2
For all the carbonaceous films prepared at any rf powers, large re-
duction currents at around 1.5-0.5 V appeared at the first cycle of
cyclic voltammograms.
Hence, the result in Fig. 3 indicates that crystallinity of the samples
increased with increasing the applied rf power, which was also con-
firmed by the conductivity measurements.15 The important point of
the spectra was that crystallinity of carbonaceous thin film can be
easily changed by the applied rf power at a fixed temperature as low
as 873 K. In other words, high rf power plasma results in the high
electron temperature, leading to large energy transfer to carbon-
aceous thin film at a fixed temperature. In Raman spectra, peak
1,22
Figure 5 shows cyclic voltammograms at the second sweep for
carbonaceous thin films prepared by rf powers of 10 and 90 W.
Electrochemical properties were found to be different between car-
bonaceous thin films prepared at 10 and 90 W as shown in Fig. 5. In
the oxidation process, the peak of carbonaceous thin film prepared at
90 W was sharp and had a clear shoulder around 0.1-0.2 V as shown
by the arrow in the insertion, while the peak was broad and did not
have such a clear shoulder for the film prepared at 10 W. In the
Ϫ1
intensities around 1360 cm are higher than those around 1580
Ϫ1
cm . This feature indicates that the films contained nongraphitized
2
0
carbon to some extent. From this result, carbonaceous thin films in
this study were composed of the graphitized structures with partly
nongraphitized structures.
Downloaded on 2015-01-19 to IP 129.81.226.78 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1262
Journal of The Electrochemical Society, 148 ͑11͒ A1260-A1265 ͑2001͒
Figure 4. Cyclic voltammograms of carbonaceous thin film in 1 mol dmϪ3
LiClO /EC ϩ DEC (1:1). Sweep rate, 1 mV/s; reaction time, 6 h; substrate,
4
Ni; applied rf power, 90 W.
Figure 6. Charge and discharge characteristics ͑2nd cycle͒ of carbonaceous
Ϫ3
thin films in 1 mol dm LiClO /EC ϩ DEC (1:1). Applied rf power ͑a͒ 10
4
and ͑b͒ 90 W. The inset figure is charge and discharge characteristics of 1st
cycle.
curves, a comparatively large potential plateau appeared at approxi-
mately 1 V. These results indicated that the electrochemical proper-
ties of carbonaceous thin film prepared at 10 W was very similar to
those for graphitizable carbon heat-treated at lower temperatures.25
Figure 6b shows charge and discharge characteristics of a carbon-
aceous thin film prepared at 90 W. At the first cycle ͑inset figure͒, a
very large irreversible capacity above 0.5 V also appeared as is
observed for the film prepared at 10 W. The insertion curve of the
sample prepared at 90 W showed about 350 mAh/g of capacity at
the second cycle, and the extraction curve showed about 200 mAh/g.
On the extraction curves, a very small potential plateau appeared at
approximately 1 V. A small plateau appeared below the potential of
Figure 5. Cyclic voltammograms ͑2nd cycle͒ of carbonaceous thin films in
Ϫ3
1
mol dm
LiClO /EC ϩ DEC (1:1). Sweep rate, 1 mV/s, applied rf
4
power, 10 and 90 W.
reduction process, the peak at 90 W had a clear shoulder around 0.1
V as shown by an arrow in the insertion, but the peak near 0 V for
the film prepared at 10 W was very smooth. These results were
2
3
0
.25 V. In the case of graphite, deintercalation of lithium ion from
correlated with the formation of stage structures of Li-GICs. The
difference in the shape of cyclic voltammograms were also related
with Raman spectra shown in Fig. 3, that is, crystallinity of the film
prepared at 90 W was higher than that of the film prepared at 10 W.
In other words, the degree of crystallinity of the carbonaceous thin
films affected the shapes of cyclic voltammograms. From the above
results, the intercalation and deintercalation behavior of carbon-
aceous thin films was found to be dependent on the applied rf pow-
ers.
2
6
Li-GIC takes place below 0.25 V, and therefore this result should
be correlated with the formation of stage structures of Li-GIC. From
these results, charge and discharge characteristics showed tendencies
similar to the results of cyclic voltammograms for carbonaceous thin
films in this study. The ratio of the capacity in the potential range 0
to 0.25 V to that in the range 0.25 to 3 V was evaluated to be 0.185
for the sample at 10 W and 0.498 for sample at 90 W. Tatsumi et al.
reported that this ratio of capacity became large with increasing
2
7
crystallinity of the mesocarbon microbeads ͑MCMBs͒. Hence, the
present results of charge-discharge measurements support the idea
that the high rf power gave a film of high crystallinity.
Charge and discharge characteristics of carbonaceous thin
films.—Figure 6a shows charge and discharge characteristics of car-
bonaceous thin film prepared at 10 W. At the first cycle ͑inset fig-
ure͒, a very large irreversible capacity above 0.5 V appeared. The
large irreversible capacity is also reported for other carbonaceous
materials.24 This irreversible capacity decreased dramatically after
the second cycle. The insertion curve of the sample prepared at 10
W showed about 1100 mAh/g of capacity at the second cycle and
the extraction curve showed about 600 mAh/g. In the extraction
Mechanism of insertion and extraction of lithium ion for
carbonaceous thin films.—As is mentioned above, the electrochemi-
cal behaviors for the insertion and extraction of lithium ions for the
present carbonaceous thin films are different from those for graphite,
but are similar to those for graphitizable carbon heat-treated at lower
2
5
temperatures. To examine the details of insertion and extraction of
Downloaded on 2015-01-19 to IP 129.81.226.78 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 148 ͑11͒ A1260-A1265 ͑2001͒
A1263
Figure 7. Cyclic voltammograms ͑2nd cycle͒ of carbonaceous thin films prepared by plasma CVD. Applied rf power, 10 W; sweep rate, ͑a͒ 10, ͑b͒ 1, ͑c͒ 0.1,
and ͑d͒ 0.01 mV/s.
lithium ion of the carbonaceous thin films, cyclic voltammetry was
conducted at various sweep rates. Figures 7a-d show cyclic voltam-
mograms at the second sweep of the sample prepared at 10 W.
Sweep rates are ͑a͒ 10, ͑b͒ 1, ͑c͒ 0.1, and ͑d͒ 0.01 mV/s. As given in
Fig. 7, a change of sweep rate influences not only values of magni-
tudes of current but also shapes of cyclic voltammograms; the peak
of the oxidation current around 0.1-0.2 V is relatively large and that
around 0.9-1.0 V is very small at higher sweep rates, while the peak
around 0.1-0.2 V is small and the peak around 0.9-1.0 V is large at
lower sweep rates. By considering the previous literature, oxida-
tion peaks around 0.1-0.2 V, denoted as A in Fig. 7, are due to
deintercalation of the lithium ion stored in graphene layers and that
peaks around 0.9-1.0 V, denoted as B, are ascribed to the extraction
of lithium ion stored in some sites except for graphene layers. In
Fig. 7, the ratio of peak B to peak A increases with decreasing the
sweep rates. A similar tendency was obtained for the sample pre-
pared at 90 W. These results mean that extraction of a potential
around 0.9-1.0 V is kinetically influenced. This is also obvious by
charge-discharge measurements obtained at different current densi-
ties as shown in Fig. 8. Figure 8 shows charge and discharge char-
acteristics at the second cycle of carbonaceous thin films prepared at
8b, indicating that no capacity around 1 V is available for the large
current densities, which is in good agreement with results as given
in Fig. 7a-d.
Next, the lithium ion entities, which can be extracted at around 1
V, are focused by linear sweep voltammetry.
The lithium ion was inserted by sweeping the electrode poten-
tials from 3.0 to 0.3, 0.1, and 0 V, followed by keeping the above
potentials for 10 h, and then the potential was reversibly swept
linearly to 3.0 V. Figure 9 shows positive linear sweep voltammo-
grams of carbonaceous thin films prepared at 10 W with a sweep
rate of 0.1 mV/s. As shown in Fig. 9, the electrodes kept at 0.1 V
͑dashed line͒ and 0.3 V ͑dotted line͒ did not give any peak at around
1 V, while the electrode kept at 0 V ͑solid line͒ exhibited a clear
peak around 1 V.
The above results indicated that lithium ion inserted into the
carbonaceous thin films at 0 V can be extracted both at about 0 and
1 V. If the lithium ion insertion and extraction was a simple redox
reaction, the reduction peak at about 1 V should appear even at the
higher sweep rate ͑Fig. 7a͒, but that was not the case. Thus the
present lithium ion insertion and extraction are not a simple reac-
tion, which led us to consider another lithium ion insertion and
extraction mechanism.
2
5
10 W by two current densities ͑a͒ 263 and ͑b͒ 26.0 mA/g. In Fig. 8,
the capacity of ͑a͒ is very different from that of ͑b͒. This is because
the current density for ͑a͒ is so large that utilization efficiency of
carbonaceous thin film becomes small, leading to a smaller capacity
of ͑a͒ than that of ͑b͒. There is no plateau at around 1 V for the
extraction curve in Fig. 8a, while a clear plateau can be seen in Fig.
The above discussion leads to the fact that transfer of lithium ion
from site A to site B is necessary for the lithium ion to insert into
site B. Otherwise the lithium ion must directly insert into site B.
However, linear sweep voltammetry clearly shows that no insertion
of lithium ion at around 1 V occurred. Figures 7 and 8 clearly show
Downloaded on 2015-01-19 to IP 129.81.226.78 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1264
Journal of The Electrochemical Society, 148 ͑11͒ A1260-A1265 ͑2001͒
Figure 10. Linear sweep voltammograms of carbonaceous thin film pre-
Ϫ3
pared by plasma CVD in 1 mol dm LiClO /EC ϩ DEC (1:1). Applied rf
4
power, 10 W; sweep rate, 0.1 mV/s for 2nd cycle, and 1 mV/s for 3rd cycle.
Each voltammograms is normalized by maximum peak height.
To further clarify that extraction of lithium ion from site B is a
slow process, linear sweep voltammetry with different sweep rates
was conducted. Prior to linear sweep voltammetry, the carbonaceous
thin film was scanned from 3 to 0 V, and held at these potential of 0
V for 10 h. Figure 10 shows linear sweep voltammograms of a
carbonaceous thin film electrode prepared at 10 W at two sweep
rates of 0.1 mV/s ͑second cycle, solid line͒ and 1 mV/s ͑third cycle,
dotted line͒. Each voltammogram is normalized by the maximum
peak height. A broad peak appeared near 1 V at the sweep rate of 0.1
mV/s, while no peak appeared at the rate of 1 mV/s. This result
indicates that no extraction of lithium ion from site B occurs at a
high sweep rate, leading to the conclusion again that extraction of
lithium ion from site B is a slow process.
Figure 8. Charge and discharge characteristics of carbonaceous thin films at
the 2nd charge-discharge cycle in 1 mol dmϪ3 LiClO /EC ϩ DEC (1:1).
4
Applied rf power, 10 W; current density, ͑a͒ 263 and ͑b͒ 26.0 mA/g.
The above results suggest that there should be an activation pro-
cess of lithium ion transfer from site A to site B. The insertion and
extraction mechanisms in this work can be summarized by Fig. 11.
Here, vertical axis represents the total energy against reaction coor-
dinates. At first, lithium ion intercalates into site A as shown by Fig.
11͑2͒. After site A is mostly filled by lithium ion, the electrode
potential should reach about 0 V and the transfer of lithium ion from
site A to site B occurs, probably accompanied by heat generation
Figure 9. Linear sweep voltammograms ͑2nd sweep͒ of carbonaceous thin
Ϫ3
films prepared by plasma CVD in 1 mol dm LiClO /EC ϩ DEC (1:1).
4
Applied rf power: 10 W. Sweep rate 0.1 mV/s.
that the extraction of lithium ion from site B only occurs at the
lower sweep rate and the smaller current density, indicating that the
activation energy should be large for the lithium ion transfer from
site A to site B and that the transfer is a slow reaction. In other
words, site B occupied by the lithium ion is thermodynamically
more stable than site A, which is in excellent agreement with the
Figure 11. An energy state model to explain lithium ion insertion and ex-
traction of carbonaceous thin films.
2
8,29
literature.
Downloaded on 2015-01-19 to IP 129.81.226.78 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 148 ͑11͒ A1260-A1265 ͑2001͒
A1265
due to the energy gap between site A and site B ͓Fig. 11͑3͒ and
Kyoto University assisted in meeting the publication costs of this article.
3
0
͑
4͔͒. During lithium ion transfer from site A to site B, insertion of
References
the lithium ion into site A through the electrolyte occurs simulta-
neously. Finally, site A and site B, are filled with lithium ions. As is
shown in Fig. 11͑5͒, site A filled with lithium ion is thermodynami-
cally more unstable than site B filled with lithium ion, and hence
deintercalation of lithium ion proceeds from site A. When the
deintercalation of lithium ion from site A occurs, the electrode po-
tential for site A becomes more positive, and then, site A becomes
thermodynamically more stable than site B filled with lithium ion.
Then lithium ion is extracted from the B sites ͓Fig. 11͑6͔͒. There are
two possible routes for extraction of lithium ion from site B. One
route is that lithium ions extract directly from site B with a slow
process. Another route is that lithium ions transfer again from site B
to site A accompanied by heat generation again and are extracted
1. J. R. Dahn, T. Zheng, Y. Liu, and J. S. Xue, Science, 270, 590 ͑1995͒.
2. J. R. Dahn, A. K. Sleigh, H. Shi, J. N. Reimers, Q. Zhong, and B. M. Way,
Electrochim. Acta, 38, 1179 ͑1993͒.
. N. Imanishi, H. Kashiwagi, T. Ichikawa, Y. Takeda, O. Yamamoto, and M. Inagaki,
J. Electrochem. Soc., 140, 315 ͑1993͒.
. I. Mochida, C.-H. Ku, S.-H. Yoon, and Y. Korai, J. Power Sources, 75, 214 ͑1998͒.
. T. Zheng, J. S. Xue, and J. R. Dahn, Chem. Mater., 8, 389 ͑1996͒.
. M. Inaba, Z. Shiroma, A. Funabiki, and Z. Ogumi, Langmuir, 12, 1535 ͑1996͒.
. M. Inaba, H. Yoshida, Z. Ogumi, T. Abe, Y. Mizutani, and M. Asano, J. Electro-
chem. Soc., 142, 20 ͑1995͒.
8. M. Inaba, H. Yoshida, and Z. Ogumi, J. Electrochem. Soc., 143, 2572 ͑1996͒.
9. Y.-S. Han, J.-S. Yu, G.-S. Park, and J.-Y. Lee, J. Electrochem. Soc., 146, 3999
͑1999͒.
3
4
5
6
7
3
0
10. M. Mohri, N. Yanagisawa, Y. Tajima, T. Tanaka, T. Mitate, S. Nakajima, M.
Yoshida, M. Yoshimoto, T. Suzuki, and H. Wada, J. Power Sources, 26, 545 ͑1989͒.
through site A. It has recently been shown by Inaba et al. the
reasons for hysteresis in the charge-discharge profiles of mesocarbon
microbeads heat-treated at lower temperatures by calorimetric
study.30 They observed heat generation during lithium extraction.
Hence, the latter route is more probable.
1
1. N. Awaya and Y. Arita, Jpn. J. Appl. Phys., Part 1, 30, 1813 ͑1991͒.
2. E. Kny, L. L. Levenson, W. J. James, and R. A. Auerbach, Thin Solid Films, 85, 23
1981͒.
3. S. Matsumoto, J. Mater. Sci. Lett., 4, 600 ͑1985͒.
1
1
30
͑
The above mechanism for lithium ion insertion and extraction is
in good agreement with that reported by Zheng et al.31 In the case of
large current densities, lithium insertion and extraction proceeds by
14. J. C. Angus and C. C. Hayman, Science, 241, 913 ͑1988͒.
15. T. Abe, T. Fukutsuka, M. Inaba, and Z. Ogumi, Carbon, 37, 1165 ͑1999͒.
16. T. Fukutsuka, T. Abe, M. Inaba, and Z. Ogumi, Tanso, 190, 252 ͑1999͒.
1
7. T. Fukutsuka, T. Abe, M. Inaba, and Z. Ogumi, Mol. Cryst. Liq. Cryst. Sect. A, 340,
17 ͑2000͒.
1
→ 2 → 3 → 7 → 8. For small current densities and keeping 0 V
5
for enough time, the reaction of lithium and carbonaceous thin films
1
1
2
2
8. F. Tuinstra and J. L. Koenig, J. Chem. Phys., 53, 1126 ͑1970͒.
9. G. Katagiri, Tanso, 175, 304 ͑1996͒.
0. D. S. Knight and W. B. White, J. Mater. Res., 4, 385 ͑1989͒.
1. E. Peled, J. Electrochem. Soc., 126, 2047 ͑1979͒.
occurs by 1 → 2 → 3 → 4 → 5 → 6 → 7 → 8.
Conclusions
Carbonaceous thin films were prepared by C H /Ar glow dis-
22. J. O. Besenhard, M. Winter, J. Yang, and W. Biberacher, J. Power Sources, 51, 228
1995͒.
2
2
͑
charge plasma. Carbonaceous thin films in this study were homoge-
neous and pinhole free. RF power of the plasma was found to influ-
ence the crystallinity of carbonaceous thin films. The lithium ion
storage in different sites for the resultant carbonaceous thin films
was clarified by using cyclic voltammetry, charge-discharge mea-
surements, and linear sweep voltammetry. Carbonaceous materials
give various electrochemical properties for use in lithium-ion batter-
ies, and our present carbonaceous thin films can be regarded as a
model of carbonaceous materials heat-treated at lower temperatures.
2
2
2
3. R. Takagi, T. Okubo, K. Sekine, and T. Takamura, Denki Kagaku oyobi Kogyo
Butsuri Kagaku, 65, 333 ͑1997͒.
4. I. Mochida, C.-H. Ku, M. Egashira, and M. Kimura, Denki Kagaku oyobi Kogyo
Butsuri Kagaku, 66, 1281 ͑1998͒.
5. A. Mabuchi, K. Tokumitsu, H. Fujimoto, and T. Kasuh, J. Electrochem. Soc., 142,
1041 ͑1995͒.
26. T. Ohzuku, Y. Iwakoshi, and K. Sawai, J. Electrochem. Soc., 140, 2490 ͑1993͒.
27. K. Tatsumi, N. Iwashita, H. Sakaebe, H. Shioyama, S. Higuchi, A. Mabuchi, and H.
Fujimoto, J. Electrochem. Soc., 142, 716 ͑1995͒.
2
8. S.-J. Lee, T. Itoh, M. Nishizawa, K. Yamada, and I. Uchida, Denki Kagaku oyobi
Kogyo Butsuri Kagaku, 66, 1276 ͑1998͒.
Acknowledgments
2
3
9. S.-J. Lee, M. Nishizawa, and I. Uchida, Electrochim. Acta, 44, 2379 ͑1999͒.
0. M. Inaba, M. Fujikawa, T. Abe, and Z. Ogumi, J. Electrochem. Soc., 147, 4008
This work was financially supported by a Grant-in-Aid for Sci-
entific Research ͑no. 11555238͒ from the Ministry of Education,
Science, Sports and Culture, Japan, and CREST of JST ͑Japan Sci-
ence and Technology͒.
͑
2000͒.
31. T. Zheng, W. R. McKinnon, and J. R. Dahn, J. Electrochem. Soc., 143, 2137
͑1996͒.
Downloaded on 2015-01-19 to IP 129.81.226.78 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).