Journal of The Electrochemical

Society

You may also like

- Lithium/Polyaniline Secondary Battery
Kinetic Characterization of the Electrochemical Composed of TransportNumberAdjusted
Aluminate Solid Polymer Electrolytes
Intercalation of Lithium Ions into Graphite Ken Onishi, Michiko Matsumoto and
Kiyotaka Shigehara

Electrodes - Use of Organic Esters as Cosolvents in

Electrolytes for Lithium-Ion Batteries with
Improved Low Temperature Performance
To cite this article: Yu‐Chi Chang et al 2000 J. Electrochem. Soc. 147 2033 M. C. Smart, B. V. Ratnakumar and S.
Surampudi

- Electrochemical Kinetic Studies of Li-Ion in

O2-Structured Li2 / 3 ( Ni1 / 3Mn2 / 3 ) O 2
and Li ( 2 / 3 ) + x ( Ni1 / 3Mn2 / 3 ) O 2 by EIS
View the article online for updates and enhancements. and GITT
K. M. Shaju, G. V. Subba Rao and B. V. R.
Chowdari

This content was downloaded from IP address 42.119.81.67 on 08/04/2024 at 16:14

 Journal of The Electrochemical Society, 147 (6) 2033-2038 (2000) 2033
S0013-4651(99)11-022-X CCC: $7.00 © The Electrochemical Society, Inc.

Kinetic Characterization of the Electrochemical Intercalation of Lithium

Ions into Graphite Electrodes
Yu-Chi Chang,a,*,z Jyh-Hwa Jong,a and George Ting-Kuo Feyb,*
aDepartment of Chemical Engineering, Tamkang University, Taipei Hsien, Taiwan 251
bDepartment of Chemical Engineering, National Central University, Chung-Li, Taiwan 320

Lithium intercalated graphites have taken the place of metallic lithium as anodes for secondary lithium batteries. Controlling the
anode-electrolyte interface has been a major technical challenge in the development of lithium-ion battery technologies. The inter-
facial characteristics can be greatly affected by the kinetics of intercalation. However, the kinetics of the electrochemical interca-
lation of lithium into graphites has not been well analyzed yet. Very few kinetic and interfacial parameters have been reported. In
this work, the electrochemical impedance spectroscopy, constant charge step, and galvanostatic pulse polarization techniques were
applied to study the kinetics of the intercalation and deintercalation processes of graphite electrodes in a few important lithium bat-
tery electrolyte solutions. Based on the proposed equivalent circuit model, we determined the kinetic and interfacial parameters of
the intercalation and deintercalation processes. The measured intercalation charge-transfer resistance, exchange current densities,
and intercalation capacitance range between 11 and 28 V cm2, 1.0 and 2.3 mA/cm2, and 1.0 and 2.0 mF/cm2, respectively, depend-
ing on the electrolyte solution compositions. The dependence of these kinetic and interfacial parameters on solvent composition,
electrolyte concentration, storage time, and intercalated state is discussed. In addition, the transfer coefficients have been deter-
mined. The results suggest that the intercalation/deintercalation process is electrochemically reversible.
© 2000 The Electrochemical Society. S0013-4651(99)11-022-X. All rights reserved.

Manuscript submitted November 8, 1999; revised manuscript received January 13, 2000.

Rechargeable lithium-ion batteries have been commercialized order to reveal the intrinsic kinetic characteristics. In this work, we
recently.1 The typical lithium-ion battery is made up of a carbon- attempted such a study using nonstationary methods. The duration of
based negative electrode, a Li1-transition metal-oxide positive elec- each measurement is sufficiently short (a few milliseconds). Piao
trode, and a nonaqueous electrolyte which serves as an ionic path et al.12 determined the exchange current densities of lithium ions
between electrodes and separates the two materials. The electrolyte intercalating into graphite electrodes in PC-EC mixed solutions with
may be a single organic solvent or a mixture of organic solvents.2 A LiClO4 as an electrolyte by ac impedance techniques. It should be
lithium salt added to the solution provides Li1 ions that diffuse back noted that these researchers11,12 did not publish kinetic and interfa-
and forth between electrodes during charge and discharge. Among cial data on the deintercalation processes which are relevant to the
the many types of carbon anodes reported, graphite appears to be the study of irreversible capacities of graphite electrodes.
most desirable candidate due to its high host capacity and low elec- In contrast to lithium metal anode rechargeable batteries, the
trode potential relative to Li metal,3,4 hence we concentrate on the lithium-ion batteries are limited by anode capacity. Therefore, con-
study of well-ordered graphite electrodes. trolling the anode-electrolyte interface has been a major technical
The formation and characterization of the lithium-graphite inter- challenge in the development of lithium-ion battery technologies.
calation compounds have been studied by Ohzuku et al.5 The mech- The interfacial characteristics can be greatly affected by the kinetics
anisms of capacity fading in lithiated graphite electrodes have been of intercalation. It is now commonly understood that in lithium-ion
intensively investigated in recent years.1,4,6-9 Several investigations batteries containing organic electrolytes, the stabilization of lithiat-
dealing with the relation between the electrochemical intercalation ed graphite electrodes is insured by the formation of stable passivat-
of lithium and the crystal structure of carbonaceous materials have ing surface layers.4,8,9 The actual morphology, as well as the com-
been reported by Dahn et al.,10 Jiang et al.,3 and Marquez et al.2 position, of the solid electrolyte interphase is very complex and
However, the kinetics of the electrochemical intercalation of lithium changes with time and with electrolyte composition. Understanding
into graphites has not been well analyzed yet. Despite the impor- the influence of solvents, electrolytes, and the composition of the
tance of electrochemical kinetics in determining the power densities graphite anodes on the kinetics of intercalation is essential to further
of lithium batteries, very few researchers addressed the reaction characterizing the fundamental physical and chemical features that
kinetics.11,12 Takami et al.11 reported the polarization resistance of control interfacial phenomena. In the present work, the ac imped-
lithium ions intercalation into various carbon fibers (heated at dif- ance, galvanostatic pulse polarization,13 and constant charge step14
ferent temperatures) and artificial graphite (Lonza, SFG44) in 1 M methods have been used to study the kinetics of the electrochemical
LiPF6-ethylene carbonate/propylene carbonate (EC/PC, 1:1) and intercalation and deintercalation processes of lithium into well-
1 M LiPF6-ethylene carbonate-diethylene carbonate(EC/DEC, 1:1) ordered graphite electrodes in a few selected electrolyte solutions.
electrolyte solutions, respectively, using an ac impedance technique. The experimentally determined kinetics and interfacial parameters
A few words deserve to be mentioned. Since Takami et al.11 mod- such as charge-transfer resistance, exchange current density, transfer
eled the film-covered carbon/electrolyte interface simply with one coefficient, and intercalation capacitance are presented. The effects
equivalent circuit containing the polarization resistance in parallel of the electrolyte composition, stored time, and intercalation level on
with the corresponding capacitance and derived the value of the these measured kinetic and interfacial parameters are also discussed.
polarization resistance from the diameter of the semicircle in the
impedance spectrum, the reported polarization resistance thus con- Experimental
sisted of the charge-transfer resistance and the film resistance. The
The electrochemical intercalation-deintercalation investigation
kinetics of the intercalation of lithium into graphite anodes is affect-
for the synthetic graphite (Aldrich) was carried out using a special-
ed by the passivating film, and the film may grow during the meas-
ly designed three-electrode hermetically sealed glass cell. The cell
urements. Therefore, the polarization resistance is not equal to the
consisted of a porous graphite electrode supported by a copper foil
charge-transfer resistance, unless the film resistance is negligible. It
(with a geometric area of 1 cm2 and a weight of 32 mg graphite), a
is necessary to reduce the film effect to the lowest degree possible in
lithium rod on a nickel wire reference electrode, and a lithium foil
* Electrochemical Society Active Member. on a nickel wire counter electrode. EC, DEC, and PC were obtained
2034 Journal of The Electrochemical Society, 147 (6) 2033-2038 (2000)
S0013-4651(99)11-022-X CCC: $7.00 © The Electrochemical Society, Inc.

from Riedel-deHae. These solvents were stored over 4 Å molecular (Fig. 1b). Therein, Rp represents the charge-transfer resistance. The
sieves. LiPF6, LiClO4, and LiBF4 were obtained from Aldrich. parameters Cp is the capacitance corresponding to Rp. Rp is in series
Graphite electrodes were made by mixing a slurry containing 10 with the Warburg impedance W. Rf and Cf are the film resistance and
weight percent (wt %) polyvinylidene fluoride binder (Aldrich) and film capacitance, respectively. The overall interfacial resistance Rt
90 wt % graphite powder in a N-methyl-2-pyrrolidinone solution can be estimated by means of the galvanostatic pulse polarization
(Aldrich). The slurry was then coated on a copper foil current col- technique based on the relationship Rt 5 (∂h/∂i)h50.13 The overall
lector. Finally, the porous structures obtained were dried in a vacu- interfacial resistance (Rt) consists of the charge-transfer resistance
um oven at 1208C for 24 h. All cells were assembled in a glove box (Rp) and the resistance associated with the film (Rf). It is essentially
(Mbraun master 130). The moisture and oxygen levels were less than equal to Rp because Rf is proportional to the thickness of film that is
5 and 20 ppm, respectively. extremely thin for each experimental run (time duration a few mil-
The galvanostatic pulse polarization,13 constant charge step,14 liseconds only).
and ac impedance techniques were used to investigate the kinetic Figures 2 and 3 show the variation of Rp with concentration of var-
characteristics of lithium intercalation and deintercalation reactions. ious electrolytes in EC/DEC (1:1) and PC, respectively, during the
For the charge-transfer resistance measurements by means of the charge of graphite anodes at 268C. The EC/DEC ratios appearing in
galvanostatic pulse polarization technique, pulses were typically of this paper are all volume ratios. The measured values of the charge-
1.0-6.0 mA/cm2 amplitudes and 8 ms duration. For the capacitance transfer resistance in various solutions are as follows: 13.0 V cm2 (1
measurements using the constant charge-step method, amplitudes M LiPF6), 14.5 V cm2 (1 M LiClO4), and 18.0 V cm2 (1 M LiBF4),
typically were 0.10-0.25 mA/cm2 and duration 10 ms. All experi- all in the EC/DEC (1:1) mixed solvents; 17.0 V cm2 (1 M LiPF6),
ments were carried out at 26 6 18C. 20.0 V cm2 (1 M LiClO4), and 23.0 V cm2 (1 M LiBF4), all in the PC
solvent. The reported values are as follows: 8 V cm2 for the graphi-
Results and Discussion tized carbon fiber electrodes (heated at 30008C) in 1 M LiPF6-EC/PC
The charge-transfer resistance.—A typical complex impedance (1:1) solutions,11 6 V cm2 (independent of cycle number) for the
plot at a graphite electrode in its intercalated state is shown in Fig. 1a graphitized carbon fiber anodes (heated at 30008C) in 1 M LiPF6-
in which a depressed semicircle in the high-frequency region and a EC/DEC (1:1) solutions,15 6-13 V cm2 (depending on the lithium con-
straight line with an angle close to 458 to the real axis at low fre- tent of the graphite electrode) for the artificial graphite (Lonza,
quencies dominate the plot. The depressed semicircle can be approx- SFG44) in 1 M LiPF6-EC/DEC (1:1) solutions,11 6-80 V cm2 (de-
imately characterized by two partially overlapped semicircles pending on cycle number) for the graphite electrode (Lonza, SFG15)
(Fig. 1a), indicating the existence of two time constants in the high- in 1 M LiPF6-EC/DEC (1:1) solutions,15 and 11-19 V cm2 (depend-
frequency region. The depressed semicircle can be interpreted as ing on the level of preintercalation) for the graphite electrode in 1 M
resulting from the charge-transfer process and the passivating film LiClO4-PC/EC (1:1) solutions.12 Our results are comparable with
formed on the graphite.9,11 The small semicircle is considered to be those of the reported data which were determined using ac impedance
related to the formation of a passive film on the graphite surface. The measurements. As can be seen from Fig. 2 and 3 the charge-transfer
large semicircle is assigned to the charge-transfer reaction of Li resistance decreases monotonical with electrolyte concentration. Fur-
intercalation into graphite. The process of lithium ions in graphite is thermore, the decrease slows down beyond about 1.0 M concentration.
considerably slower than in an electrolyte solution.11 Thus the 458 We also see that the solutions with a LiPF6 electrolyte always have the
slope linear portion at low frequencies is attributed to semi-infinite lowest charge-transfer resistance at each of the concentrations studied
diffusion conditions for the diffusion of lithium ions in graphite. The compared to LiClO4- and LiBF4-based solutions, with the LiBF4-con-
semicircles in the impedance spectrum were not clearly separated; taining solutions having the highest charge-transfer resistance. The
therefore, the elementary processes could not be properly analyzed results that the charge-transfer resistance of graphite electrodes in the
individually. As a first-order approximation of the real situation, we
modeled the film-covered graphite/electrolyte interface as a simple
equivalent circuit consisting of two parallel circuits in series

Figure 1. (a) Impedance diagram of the graphite electrode in a 1 M LiPF6- Figure 2. Variation of Rp with the concentration of various electrolytes for
EC/DEC (1:1) system at an X value of 0.1 in LixC6 at an open-circuit poten- the graphite electrode in EC/DEC (1:1) solutions: (r) LiPF6, (j) LiClO4,
tial of 0.2 V at 268C; (b) an equivalent circuit for our graphite electrodes. (m) LiBF4.
 Journal of The Electrochemical Society, 147 (6) 2033-2038 (2000) 2035
S0013-4651(99)11-022-X CCC: $7.00 © The Electrochemical Society, Inc.

Figure 3. Variation of Rp with the concentration of various electrolytes for

the graphite electrode in PC: (r) LiPF6, (j) LiClO4, (m) LiBF4. Figure 4. The polarization resistance of lithium intercalation into the
graphite electrode in LiPF6-containing solutions of various solvents as
indicated.
LiPF6 solutions is lower than that in the LiClO4 solutions are consis-
tent with the available data for relevant systems reported in the litera-
ture.11,12,15 The present results clearly show that the charge-transfer Figure 6 compares the charge-transfer resistance of graphite elec-
resistance is dependent on the Li salt used. trodes which were intercalated or deintercalated galvanostatically
The performance of the LiXC6 anodes in organic liquid elec- during experiments. It is clear that the charge-transfer resistance of
trolyte systems depends strongly on the surface chemistry developed graphite electrodes during intercalation is higher than that of the
on the carbon.16 It is interesting to investigate to what extent the sol- same electrodes during deintercalation. Experimentally, the interlay-
vent systems affect the kinetics, which is closely related to the sur-
face chemistry developed on the carbon during lithium intercalation
processes. Figure 4 presents the charge-transfer resistance of lithium
intercalation into graphite electrodes in LiPF6-containing solutions
with different solvents. A consistent, monotonic impact of the sol-
vent on the charge-transfer resistance was observed. Solutions with
a PC solvent have the highest polarization resistance. In all EC/DEC
mixture solutions, at a high EC content, the charge-transfer resis-
tance is the lowest and at a lower EC content, the charge-transfer re-
sistance has higher values. The difference in the behavior of graphite
electrodes in the three salt solutions reflects the effect of solvents on
the surface films of the electrodes.
Exploring the interfacial resistance of graphite electrodes during
prolonged storage in nonaqueous electrolyte solutions may help to
provide guidelines for commercialization of lithium-ion batteries
with graphite anodes. Figure 5 compares the charge-transfer resis-
tance of graphite electrodes prepared and stored in LiPF6 1 M solu-
tions of different solvent compositions. The charge-transfer resis-
tance is highly dependent on the solvent used. The charge-transfer
resistance of graphite in PC solutions is much higher than that meas-
ured in EC-DEC solutions and increases severely after storage. In
contrast, the charge-transfer resistance shows no pronounced change
for graphite electrodes in different EC-DEC mixture solvents during
prolonged experiments, but in the EC-DEC mixtures, as the EC con-
tent was higher, the charge-transfer resistance was slightly lower.
The behavior observed in PC solutions is in agreement with those
reported in the literature: PC reacts with graphite causing structural
degradation. The results of these experiments (Fig. 4 and 5) indicate
that mixtures of ethylene and diethyl carbonates are suitable solvent
systems for lithium-ion batteries in which the anode is a graphite- Figure 5. The polarization resistance for the graphite electrode prepared and
lithium intercalation compound. stored in LiPF6 1 M solutions in single PC and EC-DEC mixtures.
2036 Journal of The Electrochemical Society, 147 (6) 2033-2038 (2000)
S0013-4651(99)11-022-X CCC: $7.00 © The Electrochemical Society, Inc.

Figure 6. Polarization resistance vs. LiPF6 electrolyte concentration for the

graphite electrode intercalated/deintercalated with Li in EC/DEC (1:1)
solutions.

er spacing of graphite is found to increase about 10% due to the elec- Figure 7. Variation of the polarization resistance with X in LixC6 for the
trochemical intercalation of lithium into graphite.3 The expanded graphite electrode in 1 M LiPF6-EC/DEC (1:1) solutions.
structure thus facilitates the deintercalation of lithium out of graphite
electrodes. The behavior of graphite electrodes in LiClO4 solutions
is similar to that found in LiPF6 solutions. scopy analysis,18 and the carbons originally constituting the host site
Judging from the experimental results reported by Jiang et al.3 take on a negative charge.
bare lithium ions appear to be the only species involved in the inter- Experimentally, the exchange current density i0 for Reaction 2 is
calation and deintercalation processes, as shown in Eq. 1 given by the relationship i0 5 RT/nFRp. The determined exchange cur-
rent densities range between 1.1 and 2.1 mA/cm2 depending on the
C6 1 xLi1 1 xe r LixC6 [1] electrolyte types and concentrations. The exchange current density
data for lithium ions intercalation into the graphite electrode in 1 M
Figure 7 shows the charge-transfer resistance of graphite electrodes in LiClO4-EC/DEC (1:1) solutions are close to those reported in the lit-
the 1 M LiPF6-EC/DEC (1:1) electrolyte during intercalation/ dein- erature (1 M LiClO4-PC/EC (1:1) solutions).12 The measured i0 values
tercalation as a function of x in LixC6. The charge-transfer resistance, of the present work are based on the apparent geometrical area. There-
as shown in Fig. 7, shows a relatively small variation with x. The com- fore, our exchange current densities are higher than those reported by
mon trend is that the charge-transfer resistance (Rp) decreased slight- Verbrugge and Koch19 who take into account the specific surface and
ly with increasing lithium content of the graphite electrode. The fea- electrode thickness. The transfer coefficient b (5 1 2 a) with respect
tures and values shown in Fig. 7 are in agreement with those reported to Li1 can be determined from the dependence of i0 on CLi1. Figure 8
for the intercalation of lithium ions into the artificial graphite electrode shows the typical results in EC/DEC (3:7) mixture systems. The
in 1 M LiPF6-EC/DEC (1:1) solutions.11 It should be noted that dif- straight lines represent the data according to least squares analysis. For
ferent features have been reported in the literature.11,12 The graphi- the majority of cases, an average value for the charge-transfer coeffi-
tized carbon fibers heated at 2300 and 30008C showed a relatively cient of 0.35 is obtained for the intercalation and deintercalation
small decrease with x in 1 M LiPF6-EC/PC (1:1) solutions.11 The val- processes. The present results suggest that the electron transfer is rea-
ues of Rp were reported to increase with an increase in the x value with sonably reversibly in the electrolyte systems studied. Similar results
some scattered points for the graphite sheet in a 1 M LiClO4-PC/EC have been obtained for electrolyte systems with different EC/DEC
(1:1) electrolyte.12 The present results and those reported in the litera- ratios. One thing merits mention. Similar conclusion that the intercala-
ture suggest that the relation between Rp and x depends on the graphite tion/deintercalation process is electrochemically reversible has been
types and the electrolyte composition. This deserves further study in reported by Fong et al.6 They studied the electrochemical behavior of
order to identify the exact mechanism. Li/graphite cells with 1 M LiAsF6-PC/EC and 1 M LiAsF6-PC sys-
tems at 218C and reported that though the PC decomposition reaction
The exchange current density.—The most significant feature of results in the formation of passive films on the graphite electrode sur-
graphite as a host is its amphoteric character: either electron donors face, reversible Li intercalation still takes place on the film-covered
or acceptors can be accommodated within the carbon layers, the graphite surface even after the surface is passivated.
charge balance being realized through the formation of macroanions The values of the exchange current densities determined at vari-
C2n or macrocations C n .
1 17 An electrochemical approach to the
ous states of charging are presented in Fig. 9. It can be seen that
charge-transfer phenomenon which describes the formation of C 2 6 increasing the intercalation level caused the exchange current densi-
considers C2 6 as a result of the carbon reduction from the intercala- ty to increase slightly because it is weakly dependent on lithium con-
tion of a lithium cation as represented by tent during charge and discharge of the graphite electrode. A similar
trend was reported for the case of PAN-based carbon in 0.5 M
6C 1 Li1 1 e r C2
6 Li
1 [2] LiPF6-PC solutions.19
This indicates that after insertion, the lithium still retains a signifi- The capacitance.—In the constant charge-step measurement the
cant positive charge as evidenced by X-ray photoelectron spectro- injected charge causes the potential of the electrode to deviate from
 Journal of The Electrochemical Society, 147 (6) 2033-2038 (2000) 2037
S0013-4651(99)11-022-X CCC: $7.00 © The Electrochemical Society, Inc.

its original value Ei to a value Ef. The response DE(t) of the circuit
of Fig. 1b is given by

Dq  t  Dq  t 
DE(t ) 5 exp2  1 C exp2 R C  [3]
Cp  Rp p
C f  f f

Since the time duration is very short (a few microseconds) and Rf is

also very small (much smaller than Rp), the capacitance Cp can be
determined experimentally via the relationship Cp 5 Dq/DE, where
Dq 5 IDt, DE 5 Ef-Ei, and I is the applied current during a sufficient
short-duration Dt. Figure 10 shows the measured Cp as a function of
electrolyte concentration. While we found no reported capacitance
data for lithium ions intercalation into the graphite electrode in the lit-
erature, there are reports about the ac impedance diagrams of the arti-
ficial graphite electrodes in 1 M LiPF6-EC/DEC (1:1) solutions at var-
ious intercalation levels determined by ac impedance methods.11 Our

Figure 9. Exchange current density for the graphite electrode as a function

of X in 1 M LiPF6-EC/DEC (1:1) solutions.

measured capacitance values in 1 M LiPF6-EC/DEC (1:1) solutions

are comparable with the data estimated from the ac impedance dia-
grams of Takami and co-workers.11 Figure 10 also shows that the
capacitance increased very slightly with increases in electrolyte con-
centration. Figure 11 represents the capacitance of graphite electrodes
in LiPF6-containing solutions of different solvent compositions. The
data indicate clearly the dependence of the capacitance on the solvent

Figure 8. Exchange current density-electrolyte concentration plot for the

graphite electrode in EC/DEC (3:7) mixture with different electrolytes. (a)
Intercalation: LiPF6 (r), y 5 0.3503x 1 0.2546; LiClO4 (j), y 5 0.3512x
1 0.2345; LiBF4 (m), y 5 0.315x 1 0.1186. (b) Deintercalation: LiPF6 (r),
y 5 3568x 1 0.4447; LiClO4 (j), y 5 0.3345x 1 0.4059; LiBF4 (m), y 5 Figure 10. Variation of Cp with electrolyte concentration for the graphite
0.3404x 1 0.2581. y 5 log i0, x 5 log C. electrode in EC/DEC (3:7) solutions: (r) LiPF6, (j) LiClO4, (m) LiBF4.
2038 Journal of The Electrochemical Society, 147 (6) 2033-2038 (2000)
S0013-4651(99)11-022-X CCC: $7.00 © The Electrochemical Society, Inc.

Figure 12. Capacitance dependence on Li intercalation level in the graphite

electrode in 1 M LiPF6-EC/DEC (1:1) solutions at room temperature.

Figure 11. The capacitance of the graphite electrode in LiPF6-containing calation processes was about 0.35, indicating that the electron trans-
solutions of various solvent mixtures. fer process is electrochemically reversible. The experimental results
also indicate that ethylene and diethyl carbonates mixtures are suit-
composition. These results imply that the electrolyte, as well as the able solvent systems for Li-ion batteries with graphite electrodes.
solvent, does influence the interfacial structure of the systems. Acknowledgment
In order to gain insight into changes at the surfaces of graphite
electrodes during the intercalation and deintercalation processes, a The authors would like to acknowledge the financial support from
series of measurements were performed to determine the depen- the National Science Council of the Republic of China for this project
dence of the capacitance on the intercalation level in graphite elec- under grants NSC 85-2214-E-032-002 and NSC 86-2214-E-032-004.
trode. The typical corresponding Cp /x dependence is shown in The authors also thank Professor W. J. Wang of Tamkang University,
Fig. 12. The capacitance increased slightly with increasing interca- Taiwan, for providing the glove box to carry out the experiments.
lation level during charge and discharge and, at any intercalated References
state, the capacitance was larger during deintercalation processes. 1. P. Arora, R. E. White, and M. Doyle, J. Electrochem. Soc., 145, 3647 (1998).
2. A. Marquez, A. Vargas, and P. B. Balbuena, J. Electrochem. Soc., 145, 3328 (1998).
Conclusions 3. Z. Jiang, M. Alamgir, and K. M. Abraham, J. Electrochem. Soc., 142, 333 (1995).
The kinetic characteristics of graphite electrodes in LiPF6-, 4. D. Aurbach, M. D. Levi, E. Levi, H. Teller, B. Markovsky, G. Salitra, U. Heider,
and L. Heider, J. Electrochem. Soc., 145, 3024 (1998).
LiBF4-, and LiClO4-containing solutions have been studied using 5. T. Ohzuku, Y. Iwakoshi, and K. Sawai, J. Electrochem. Soc., 140, 2940 (1993).
electrochemical impedance spectroscopy in conjunction with the 6. R. Fong, U. von Sacken, and J. R. Dahn, J. Electrochem. Soc., 137, 2009 (1990).
constant charge-step and galvanostatic pulse polarization tech- 7. Z. X. Shu, R. S. McMillan, and J. J. Murray, J. Electrochem. Soc., 140, 922 (1993).
niques. The solvents used were single PC and EC-DEC mixtures. 8. E. Peled, D. Golodnitsky, C. Menachem, and D. Bar-Tow, J. Electrochem. Soc.,
145, 3482 (1998).
The Cole-Cole plot shows that two partially overlapped semicircles 9. E. Barsoukov, J. H. Kim, C. O. Yoon, and H. Lee, J. Electrochem. Soc., 145, 2711
in the high-frequency region and a straight line in the low-frequency (1998).
region dominate the plot. The charge-transfer resistance, exchange 10. J. R. Dahn, K. Sleigh, H. Shi, J. N. Reimers, Q. Zhong, and B. M. Way, Electro-
current density, transfer coefficient, and intercalation capacitance chim. Acta, 38, 1179 (1993).
11. N. Takami, A. Satoh, M. Hara, and T. Ohsaki, J. Electrochem. Soc., 142, 371 (1995).
have been determined experimentally from the galvanostatic pulse 12. T. Piao, S-M. Park, C-H. Doh, and S-I. Moon, J. Electrochem. Soc., 146, 2794
polarization and constant charge-step measurements. The charge- (1999).
transfer resistance and capacitance depend on the types of salt, as 13. Y. Geronov, F. Schwager, and R. H. Muller, J. Electrochem. Soc., 129, 1422 (1982).
well as solvent, used. The intercalation level and the solution com- 14. A. Bard and L. R. Faulkner, Electrochemical Methods, p. 270, John Wiley & Sons,
New York (1980).
position also affect the kinetic and interfacial parameters. The 15. N. Takami, A. Satoh, M. Hara, and T. Ohsaki, J. Electrochem. Soc., 142, 2565
charge-transfer resistance of graphite in LiBF4 is much higher than (1995).
that measured in LiPF6 or LiClO4 solutions in the same solvent sys- 16. D. Aurbach, A. Zaban, A. Schechter, Y. Ein-Eli, E. Zinigrad, and B. Markovsky,
tem. The charge-transfer resistance of graphite in PC solvent in- J. Electrochem. Soc., 142, 2873 (1995).
17. R. Yazami, in Lithium Batteries: New Materials, Developments and Perspectives,
creases remarkably after storage. The charge-transfer resistance de- G. Pistoia, Editor, Chap. 2, Elsevier Science B.V., Amsterdam (1994).
creases slightly with intercalation level as opposed to the capaci- 18. M. W. Verbrugge and B. J. Koch, J. Electrochem. Soc., 143, 600 (1996).
tance case. The transfer coefficient for the intercalation and deinter- 19. M. W. Verbrugge and B. J. Koch, J. Electrochem. Soc., 146, 833 (1999).