Journal of The Electrochemical Society, 147 (4) 1291-1294 (2000) 1291

S0013-4651(99)09-083-7 CCC: $7.00 © The Electrochemical Society, Inc.

A LiCoO2 Cathode Modified by Plasma Chemical Vapor Deposition for

Higher Voltage Performance
Eishi Endo,*,z Toshikazu Yasuda,a Akinori Kita,b Kiyoshi Yamaura,a and Koji Sekaia
Sony Corporation, Frontier Science Laboratories, Yokohama 240-0036, Japan

Surface modification of the electrode dramatically improved the performance of a LiCoO2 electrode charged to 4.6 V. A diamond-
like carbon thin film was deposited uniformly on the electrode surface using dc plasma chemical vapor deposition. This surface
treatment suppressed electrolyte decomposition during the charge process. When a coin-type cell using this electrode was charged
to 4.6 V, the discharge capacity in the first cycle was 212 mAh/g and the charge-discharge efficiency was over 90% through five
cycles. This treatment suggests a new approach to the development of cathodes for lithium-ion secondary batteries.
© 2000 The Electrochemical Society. S0013-4651(99)09-083-7. All rights reserved.

Manuscript submitted September 22, 1999; revised manuscript received December 6, 1999.

LiCoO2 is the cathode (positive electrode) active material used in
commercial lithium-ion secondary batteries.1,2 Since the earliest in-
vestigations, however, practical limitations of the electrode perfor-
mance have been known.1 Specifically, good cyclability can be ob-
tained only between 0.5 < x < 1 in LixCoO2 below 4.2 V (vs.
Li/Li1).1,3 The structural changes occurring during the delithia-
tion/lithiation processes4,5 and the oxidative decomposition of the
electrolyte solution on the electrode surface6-14 have been studied
extensively, and it is recognized that these factors affect the electrode
performance.15 Various efforts have been made to improve electrode
performance, for example, the composition of the electrolyte solu-
tion has been investigated.16-19 Further, it has been reported that the
complete delithiation of LiCoO2 up to 5.2 V is possible using a dry
plastic cell.20 If the decomposition of the electrolyte solution is sup-
pressed and the LiCoO2 cathode is cycled above 4.2 V in solution
cells, the cell performance will be improved dramatically.
We have been studying the surface modification of electrodes21
for the purpose of suppressing electrolyte decomposition and obtain-
ing better electrode performance. In this study, plasma chemical
vapor deposition (CVD) was applied to the LiCoO2 electrode for
solution cells, and the effect of the surface modification on the elec-
trode performance was evaluated.
When hydrocarbon materials are introduced into a dc glow dis-
charge, a diamond-like carbon (DLC) film is usually formed on the
cathode surface.22-25 This DLC film is attracting attention because of
its properties such as high electrical resistivity, mechanical hardness,
and chemical stability. In this study we tried to deposit a DLC film
on the LiCoO2 electrode surface using dc plasma CVD of ethylene.
The modified LiCoO2 electrode was cycled between 3.0 and 4.6 V
in a coin-type cell.
Experimental
Electrode preparation.—For the synthesis of LiCoO2, LiOH,
H2O, and Co(OH)x (Wako Pure Chemicals, Industry, Ltd.) were mix-
ed in a stoichiometric ratio and heated to 5008C for 4 h in air. The
sample was then cooled, crushed, and heated to 9008C for 12 h in air.
The as-prepared LiCoO2 was mixed with 6% synthetic graphite (KS-
15, Lonza) as the conducting material and 3% of poly(vinylidene 1,1-
difluoride) (PVDF, Aldrich) as the binder, and was pressed with an
aluminum mesh to form a pellet electrode.
Plasma CVD.—The as-formed pellet was placed in a vacuum
chamber and positioned on the cathode for dc glow discharge. The
chamber was then evacuated to less than 1023 Torr and C2H4 was
introduced at a flow rate of 10 mL/min standard temperature and
pressure. The system pressure was maintained at 0.08 Torr by con-
* Electrochemical Society Active Member.
a Present address: Sony Corporation, CNC E Company.
b Present address: Sony Corporation, Technical Support Center.
z E-mail: Eishi.Endoh@jp.sony.com
trolling the needle bulb between the chamber and pumping system.
Plasma CVD was carried out by applying dc power of 1.2 kV be-
tween the parallel capacity-coupled electrodes for 1 min.
Characterization.—Scanning electron microscopy (SEM) was
carried out using a Hitachi S-4000 microscope. X-ray photoelectron
spectroscopy (XPS) was performed using an SSI S-Probe spectrom-
eter. X-ray diffraction (XRD) patterns were measured using an RINT
2500V diffractometer (Rigaku Co.) equipped with a Cu Ka radiation
source and a diffracted-beam monochrometer. Raman spectroscopy
was performed using an Ar-ion laser (514.5 nm, 40 mW at sample)
as the excitation source, a monochrometer (JOBIN YVON T64000),
and an LN-cooled 1 in. back-thinned CCD as the detector.
Coin-type cell test.—The performance of the pellet electrode
was evaluated using a coin-type cell. The test cell was assembled in
a 2025 cell with the electrode pellet, a lithium metal counter elec-
trode, a polypropylene separator, and the electrolyte solution. The
electrolyte solution was 1 M LiPF6 in 1:1 propylene carbonate (PC)
and dimethylcarbonate (DMC) (Tomiyama Pure Chem. Industry,
Ltd.). The coin-type cell was assembled in an Argon-filled glove
box (Mecaplex, GB82) at oxygen and water concentrations below 5
and 1 ppm, respectively.
The cell was charged galvanostatically up to 4.2 or 4.6 V (vs.
Li/Li1) at 0.25 mA/cm2, then charged potentiostatically until the
current density decreased below 5 mA/cm2. For safety reasons, the
total charge time was limited to 36 h, regardless of the potential and
current. Discharge was carried out galvanostatically to 3.0 V (vs.
Li/Li1) at 0.25 mA/cm2.
Results and Discussion
Characterization of plasma CVD film.—The deposit on the elec-
trode surface obtained by plasma CVD was characterized first. Fig-
ure 1 shows SEMs of the electrode pellets before and after plasma
CVD. The synthesized LiCoO2 consisted of 10-20 mm particles. No
distinctive morphological change was observed before and after
plasma CVD in the micrographs, suggesting uniform deposition on
the electrode surface by plasma CVD.
The electrode surface was then characterized by XPS. Before
plasma CVD (Fig. 2a), XPS peaks of Li, Co, and O from LiCoO2, C,
and F from the binder, and C from the conducting material were
detected at the LiCoO2 electrode surface. With the exception of C
and O, these elements are not found on the electrode surface after
plasma CVD (Fig. 2b). This result means that the LiCoO2 electrode
surface was completely covered with the plasma CVD film, which
consists mostly of C and a small amount of O.
Figure 3 shows C 1s photoelectron spectra for the LiCoO2 elec-
trodes. XPS peaks at 284.5 and 286 eV in Fig. 3a are attributed to
C–C bonds. The former peak arises from the conducting material
and the latter from the binder. Another XPS peak at 290 eV is attrib-
uted to C–F bond from the binder. After plasma CVD, the peak at

Downloaded on 2014-10-05 to IP 128.119.168.112 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 1292 Journal of The Electrochemical Society, 147 (4) 1291-1294 (2000)
S0013-4651(99)09-083-7 CCC: $7.00 © The Electrochemical Society, Inc.

Figure 2. XPS survey spectra for LiCoO2 electrodes after (a, top) 0 and (b,
bottom) 1 min of plasma CVD.
Figure 1. SEMs of LiCoO2 electrodes after (a, top) 0 and (b, bottom) 1 min
of plasma CVD.
the modified electrode. In dc plasma CVD, the source materials are
290 eV disappeared, which is consistent with the disappearance of ionized or decomposed in the glow discharge, and the CVD film is
the fluorine content in Fig. 2b. Further, the peak at 285 eV became formed when the gas species reach the electrode surface. Due to the
broader, suggesting the deposition of a disordered carbon. long mean-free path in the vacuum system, the CVD film is deposit-
Figure 4 shows the XRD pattern of the CVD carbon film. In order ed on the surface electrode and even on the inside of the porous elec-
to eliminate the peaks from the conducting material of the electrode, trode. In the present plasma CVD, the source material is ethylene,
a nickel plate was used as the substrate. In this dc plasma CVD, the which is the smallest unsaturated hydrocarbon. We found that the
structure of the deposit was not significantly affected by the substrate,
as is discussed below. None of the characteristic peaks of the graphite
(002) plane nor of any other periodic structure was observed, there-
fore, the plasma CVD film must be an amorphous carbon film.
Raman spectroscopy was used to further characterize the CVD
carbon film. The resulting spectrum is shown in Fig. 5. The spectrum
consists of two broad bands at 1349 and 1575 cm21. The first band
indicates an A1g mode arising from the (disordered) turbostratic
structure in carbon (D-band), and the second shows an E2g2 vibration
mode in the graphitic region of carbon materials (G-band).26,27 The
ratio of the D-band to the G-band intensities, R(ID/IG )26,27 expresses
the degree of surface disordering of the graphite powder. In this car-
bon film, the R value is 1.8, which is in the range of amorphous car-
bon. Further, the overall spectrum is quite similar to that of DLC pre-
pared using dc plasma CVD.29 In the dc plasma CVD of ethylene,
the deposit structure is not significantly affected by the substrate
identity. Thus, the carbon film deposited on LiCoO2 electrode by dc
plasma CVD of ethylene can be attributed to DLC.
The SEM figures and XPS spectra suggest the uniform deposi-
tion on and complete coverage of the electrode by the plasma CVD
film. Figure 6 shows a schematic illustration of the cross section of Figure 3. C 1s photoelectron spectra for LiCoO2 electrodes.

Downloaded on 2014-10-05 to IP 128.119.168.112 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, 147 (4) 1291-1294 (2000)
S0013-4651(99)09-083-7 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 4. XRD pattern of plasma CVD film.
LiCoO2 electrode was uniformly covered even inside the pores with
this plasma CVD film. The film thickness on the electrode surface
was estimated to be about 0.1 mm.
There are two ways of modifying the electrode surface. One is
modifying the particles of the electrode active material before pre-
paring the electrode pellet, and the other is modifying the formed
electrode pellet after preparation. In the first way the entire particle
surface can easily be modified, but the surface modification effects
the particle contact, the ease of preparation of the electrode, and the
electronic conductivity of the electrode. In the second way, these
problems can be avoided but uniform treatment of all of the elec-
trode particles is difficult. In this study, plasma CVD was used to
treat the entire electrode surface in a facile manner.
Electrode performance.—The effect of the DLC film on the
LiCoO2 electrode performance was evaluated using coin-type cells.
The electrode impedance was not increased by the plasma CVD film.
Table I shows the LiCoO2 electrode performance charged up to 4.2 V.
In spite of its low electronic conductivity, the DLC film deposited by
plasma CVD did not reduce the discharge capacity or the charge-dis-
charge efficiency. Indeed, these performance measures were slightly
increased by plasma CVD surface treatment. This could be the result
of suppression of the electrolyte decomposition by the plasma CVD
film even at the charge potential of 4.2 V, which is consistent with the
slight decrease of charge capacity. The plasma treatment may enable
better charge-discharge performance at an even higher charging
potential than 4.2 V.
Figure 5. Raman spectrum of plasma CVD film.
Downloaded on 2014-10-05 to IP 128.119.168.112 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
1293
Figure 6. Schematic illustration of the cross section of a LiCoO2 electrode
with a plasma CVD film.
Figure 7a shows the charge-discharge curve of the LiCoO2 elec-
trode without plasma CVD, charged up to 4.6 V. In the charge
process over 4.5 V, both the electrode potential and the charge cur-
rent were unstable. The charge process was therefore terminated after
36 h. The total charge capacity in the first cycle was 295 mAh/g,
which is far beyond the theoretical value of 274 mAh/g. The irre-
versible capacity of the order of 100 mAh/g may be attributed to
electrolyte decomposition during the charge process.
The performance of the electrode charged up to 4.6 V was great-
ly improved by plasma CVD, as can be seen in Fig. 7b. The charge
curve is monotonic, and the charge process was terminated by the
threshold current. The charge capacity was 231 mAh/g (x 5 0.16 in
LixCoO2), which is in good agreement with the result obtained by
Amatucci using a dry plastic cell.20 The discharge curve is also
monotonic, and the discharge capacity was 212 mAh/g. The result-
ing charge-discharge efficiency was 92%, which is higher than that
for a 4.2 V charge without plasma CVD.
We conclude that this improvement in electrode performance
charged up to 4.6 V is due to the suppression of electrolyte decom-
position by the DLC film. Thomas has suggested that the electrolyte
decomposition on the charged LiCoO2 surface is induced by the
strong oxidizing power of Co41 cations.6 In the present study, it
appears clear that the reaction between the electrode and electrolyte
is suppressed by the presence of the DLC film.
A further significant improvement in electrode performance after
plasma CVD is seen in the cycle performance, as shown in Fig. 8. In
the electrode without plasma CVD, the charge process was terminat-
ed after 36 h for every cycle. The discharge-capacity cycle perfor-
mance was not monotonic, and the charge-discharge efficiency was
below 40% after the second cycle. With plasma CVD, the cycle per-
formance is monotonic and the charge-discharge efficiency is above
90% for each cycle. The film growth on the surface of the lithium
counter electrode and structural changes in the LiCoO2 electrode
materials could cause a slight decrease of the capacities upon cycling.
Thus far, we have seen that electrolyte decomposition was sup-
pressed by the DLC plasma film, and the performance of the elec-
trode was improved dramatically when the electrode was charged up
to 4.6 V. The charge potential of 4.6 V was chosen arbitrarily for this
study, and an investigation with a higher charge voltage and other
methods of surface modification will be pursued.
Table I. Electrode performance of LiCoO2 electrodes charged
up to 4.2 V.
Discharge capacity Charge capacity Efficiency
Electrode (mAh g21 ) (mAh g21) (%)
LiCoO2 128.5 142.8 90.0
LiCoO2 131.0 140.4 93.0
(plasma-treated)
 1294 Journal of The Electrochemical Society, 147 (4) 1291-1294 (2000)
S0013-4651(99)09-083-7 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 7. Charge-discharge curves of LiCoO2 electrodes with (a, top) 0 and
(b, bottom) 1 min of plasma CVD charged up to 4.6 V.
We believe that controlling the electrode/electrolyte interface is
crucial to the development of lithium-ion secondary batteries, in-
cluding polymer batteries. The irreversible reactions on both the
cathode and anode surfaces impair battery performance. Thus, in
addition to further developing the electrode materials themselves,
controlling the structure of the electrode interface by surface modi-
fication is essential.
Conclusion
The LiCoO2 electrode surface was uniformly covered with a DLC
film using dc plasma CVD of ethylene. The electrode performance was
improved significantly by this surface treatment. The discharge capac-
ity was 212 mAh/g and the charge-discharge efficiency was above
90% through five cycles, when the electrode was charged up to 4.6 V.
This improvement in electrode performance due to the suppression of
electrolyte decomposition by the DLC film suggests a new approach to
the development of cathodes for lithium-ion secondary batteries.
Acknowledgments
The authors gratefully thank Dr. M. Ata, M. Itabashi, S. Hiraka,
Y. Honjo, A. Kawashima, and N. Okano for their helpful discussions
and technical assistance.
Sony Corporation assisted in meeting the publication costs of this article.
References
1. K. Mizushima, P. C. Jones, P. J. Wiseman, and J. B. Goodenough, Mater. Res. Bull.,
15, 783 (1980).
2. K. Sekai, H. Azuma, A. Omaru, S. Fujita, H. Imoto, T. Endo, K. Yamaura, Y. Nishi,
S. Mashiko, and M. Yokogawa, J. Power Sources, 43-44, 241 (1993).
3. G. G. Amatucci, J. M. Tarascon, and L. C. Klein, Solid State Ionics, 83, 167 (1996).
4. J. N. Reimers and J. R. Dahn, J. Electrochem. Soc., 139, 2091 (1992).
5. T. Ohzuku and A. Ueda, J. Electrochem. Soc., 141, 2972 (1994).
6. M. G. S. R. Thomas, P. G. Bruce, and J. B. Goodenough, J. Electrochem. Soc., 132,
1521 (1985).
Downloaded on 2014-10-05 to IP 128.119.168.112 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
Figure 8. Charge-discharge capacity of LiCoO2 electrodes with (a, top) 0 and
(b, bottom) 1 min of plasma CVD charged up to 4.6 V. (s) charge and (d)
discharge.
7. M. G. S. R. Thomas, P. G. Bruce, and J. B. Goodenough, Solid State Ionics, 18&19,
794 (1986).
8. F. Ossola, G. Pistoia, R. Seeber, and P. Ugo, Electrochim. Acta, 33, 47 (1988).
9. P. Novak, P. A. Christensen, T. Iwasita, and W. Vielstich, J. Electroanal. Chem.,
263, 37 (1989).
10. S. A. Campbell, C. Bowes, and R. S. McMillan, J. Electroanal. Chem., 284, 195
(1990).
11. A. Rasch, E. Cattaneo, P. Novak, and W. Vielstich, Electrochim. Acta, 36, 1397
(1991).
12. K. Kanamura, S. Toriyama, S. Shiraishi, M. Ohashi, and Z. Takehara, J. Electroanal.
Chem., 419, 77 (1996).
13. D. Aurbach, M. D. Levi, E. Levi, B. Markovsky, G. Salitra, H. Teller, U. Heider,
and V. Hilarius, in Proceedings of the Symposium on Batteries for Portable Appli-
cations and Electric Vehicles, C. F. Holmes and A. R. Landgrebe, Editors, PV 97-
18, p. 941, The Electrochemical Society Proceedings Series, Pennington, NJ
(1997).
14. R. Imhof and P. Novak, J. Electrochem. Soc., 146, 1702 (1999).
15. M. Winter, J. O. Besenhard, M. E. Spahr, and P. Novak, Adv. Mater., 10, 725
(1998).
16. E. Plichta, M. Salomon, S. Slane, M. Uchiyama, D. Chua, W. B. Ebner, and H. W.
Lin, J. Power Sources, 21, 25 (1987).
17. E. Plichta, S. Slane, M. Uchiyama, M. Salomon, D. Chua, W. B. Ebner, and H. W.
Lin, J. Electrochem. Soc., 136, 1865 (1989).
18. G. T. K. Fey, M. C. Hsieh, H. K. Jaw, and T. J. Lee, J. Power Sources, 43-44, 673
(1993).
19. J. Dreher, B. Haas, and G. Hambitzer, J. Power Sources, 43-44, 583 (1993).
20. G. G. Amatucci, J. M. Tarascon, and L. C. Klein, J. Electrochem. Soc., 143, 1114
(1996).
21. E. Endo, T. Kihira, A. Kita, S. Yamada, H. Imoto, and K. Sekai, J. Power Sources,
Submitted for publication.
22. K. Fujii, N. Shohata, M. Mikami, and M. Yonezawa, Appl. Phys. Lett., 47, 370
(1985).
23. H. Tsai and D. B. Bogy, J. Vac. Sci. Technol. A, 5, 3287 (1987).
24. M. Malhotra, T. Som, V. N. Kulkarni, and S. Kumar, Vacuum, 47, 1265 (1996).
25. S. Zhang, B. Wang, and J. Y. Tang, Surf. Eng., 13, 303 (1997).
26. T. Tunistra and L. Koenig, J. Chem. Phys., 53, 1126 (1970).
27. D. S. Knight and B. White, J. Mater. Res., 4, 385 (1989).
28. R. Schlesinger, M. Bruns, and H.-J. Ache, J. Electrochem. Soc., 144, 6 (1997).
29. K. Kawada and Y. Sekiya, Netsushori, 34, 279 (1994).